Chapter 14 Benign, premalignant, and malignant tumours in gynaecology

Cancer screening for women

Cervical cancer

Cervical dysplasia and human papillomavirus

Neoplastic conditions of the endometrium

Gestational trophoblastic neoplasia

Ovarian and fallopian tube cancer

Palliative care

Vulval cancer

Vulval intraepithelial neoplasia

Cancer screening for women

Screening for cancer in women must include both those cancers with disease presentations that are different in women and those that are unique for women.

Wilson and Junger (1968) have outlined the criteria for the desirable characteristics of a successful screening programme. These are

• the condition should be important

• an accepted treatment must be available for the condition

• the facilities for diagnosis and treatment must be available

• a latent or early symptomatic stage must exist

• a sensitive and specific screening test must be available

• the test must be acceptable to the population

• the natural history of the condition must be understood

• an agreed treatment policy must exist

• the cost must be acceptable

• case finding must be a continuous process.

It is estimated that there will be increased incidence of gynaecological cancers over the next 5 years. The incidence of endometrial cancer will increase overall as a consequence of an ageing population, and an increasing problem of obesity. The incidence of cases of ovarian cancer, despite the use of the combined pill and hormone replacement therapy will increase due to the ageing population and more accurate assessment through the use of multidisciplinary working.

Lung cancer screening

• No primary lung cancer screening is recommended, although low-dose computed tomography may benefit high-risk patients (Bach et al. 2007).

• Lung cancer remains the number one cause of cancer death overall worldwide for men and second or third for women. Primary prevention of lung cancer with smoking cessation is the ultimate goal. Women have trailed men in decreasing the rate of smoking, but are beginning to show an overall decline. In addition, women may be at risk because of secondary inhalation of smoke. At this point, routine yearly exams or computed tomography has not been shown to make a difference in early diagnosis or overall survival. Therefore, there is no screening recommendation for lung cancer in women.

Colon cancer screening

• The UK Bowel Screening Programme is currently being rolled out throughout the UK and will achieve nationwide coverage by 2009. The NHS Bowel Cancer Screening

Programme offers screening every 2 years to all men and women aged 60–69.

• Colon cancer has wide geographic and ethnic variation in incidence. Family history has a major impact on the frequency of colon cancer screening. For women for whom there is a family history of familial polyposis or multiple colon cancers, with or without associated uterine and breast cancer, screening begins earlier than the recommended age of 50, and is likely to be more frequent. Screening in these patients should be full colonoscopy and managed in the context of overall screening for genetic predisposition to these cancers.

• Colonoscopy is preferred for women, given the higher incidence of right-sided lesions (Schoenfeld et al. 2005). The issues are the preparation required, the risks, and the lack of adequate access to trained health professionals. Faecal occult blood testing may have a role, but is both less sensitive and less specific than colonoscopy as a screening tool. Adequate use requires changing diet before testing to avoid false positives. It should be used with an additional evaluation. Sigmoidoscopy will miss right-sided lesions and has similar risks to colonoscopy, although technically easier.

• The use of computed tomography colonoscopy is still in the developmental stages and not appropriate at this time for standard screening.

Breast cancer screening

The NHS Breast Screening Programme provides breast screening every 3 years for all women in the UK between 50 and 70 years. Because the programme is a rolling one, every woman will receive an invitation for her first screening between her fiftieth and fifty-third birthday.

Breast cancer is the most common cause of cancer deaths for women worldwide (21% of cancer deaths for women). Access to and frequency of screening directly impact this statistic.

Family history also has a major impact on the frequency and type of breast cancer screening. In particular, women with a family history of >1 first-degree relative with breast cancer; or breast cancer in a first-degree relative before age 40; breast and ovarian cancer or independent bilateral breast cancer in the same woman; or multiple breast and ovarian cancers in the family all are at higher risk for the disease. In addition, an increased incidence of early age diagnosis of cancers in the family, such as colon cancer, may put women at higher risk. This risk requires assessment, often with genetic counselling, to determine the level of risk and the role that screening (versus other forms of cancer risk reduction such as medication or surgery) will play.

For women who are at higher risk for breast cancer, evaluation with magnetic resonance imaging (MRI), with or without concurrent mammogram, may be more sensitive and specific in the detection of early breast cancers.

Women who have had prior breast cancer should have more frequent screening depending on their overall risk for recurrence. For many of these patients, screening every 6 months may be recommended and for those at higher family risk, screening with MRI may be recommended for one of those intervals (DeMartini et al. 2008; Warner et al. 2008).

The use of clinical breast examinations has been widely promoted, but at present there is no evidence for efficacy as a sole method of screening, nor its use as an adjunct for screening. However, given the low cost and ability to pick up some lesions chronologically distant from mammographic screening, it remains generally encouraged.

Encouraging screening and assuring access to screening both make a substantial difference in early detection and overall survival with this disease. Any strategy to achieve a high participation rate is likely to result in improvements in participation and is encouraged (Bonfill et al. 2008). Gynaecological cancers

Vulval cancer screening

In the UK there are no recommendations for vulval cancer screening for the general population.

Vulvar cancer is generally situated on the external labia majora and minora. As the most common forms of cancer in this region are squamous, close visual inspection forms the primary means of screening as with other epidermal cancers. There is no evidence that magnification of the skin with colposcopic equipment improves the overall screening outcomes.

Women at higher risk for development of these cancers, and therefore appropriate for annual screening include women with a history of cervical or vaginal dysplasia or cancer, women with high exposure to nicotine, or women with prior vulvar dysplasia or cancer. In addition, women with a history of chronic immune suppression are at higher risk. The invasive form of these cancers peaks in the 70s and 80s, therefore inspection of the vulvar areas in the 60s and beyond is warranted.

Vaginal cancer screening

There is no UK recommendation for vaginal cancer screening.

Screening may be indicated because of a past history of cervical dysplasia, cancer or prior exposure to DES (diethylstilbestrol).

Cervical cancer screening

• The NHS Cervical Screening Programme in England invites all women between the ages of 25 and 64 for a cervical screening test every 3–5 years. The new guidelines are

• Under 25s: no screening

• 25 years: first invitation

• 25–49: 3 yearly

• 50–64: 5 yearly

• 65+: only screen those who have not been screened since aged 50 or have had recent abnormal tests.

• Cervical cancer is rare in women under the age of 25 years.

• The National Cervical Screening Programme in Scotland offers women aged 20–60 years a cervical smear every 3 years.

• In the UK, all cervical screening services have moved to liquid-based cytology (LBC) so reducing the number of inadequate smears.

• Worldwide, cervical cancer remains the number one cause of death from gynaecological cancer in underdeveloped countries with a quarter of a million deaths per year. Eighty-three per cent of the incidence and 86% of the mortality occurs in this population and the World Health Organization estimates that this will continue to rise by at least 10% if no actions are taken to prevent or screen and treat for this cancer. Given the development of both a prevention vaccine and effective screening schemes, this is a cancer that has the potential for major reduction and virtual elimination in the future (Wright et al. 2007).

• The general recommendations for cervical cancer screening have changed based on the tight linkage to the presence of oncogenic subtypes of HPV. Recommendations for screening of women who received the HPV vaccine have not been formalized and should follow the standard guidelines until there is more clarity on the long-term success and correlates for disease with prior vaccination.

• Special efforts to recruit women from populations with limited screening and high cervical cancer incidence rates should be national concerns, as the incidence directly correlates with access to screening systems.

• HPV high risk subtype testing has become a standard component of triage of abnormal cytological results. Its use as a primary method of screening is in active research. Advantages include the rapidity of the testing, the potential for self-administration, and potentially a broader application to both resource rich and resource poor environments.

Uterine cancer screening

• There is no standard screening for uterine/endometrial cancer.

• The highest prevalence is in North America with age-standardized rates of around 15–18 per 100 000 and lowest in most of Africa and southeast Asia.

• Uterine cancer is generally a disease of women after the menopause for which less than 50% is detected on standard cervical cytology. There are potential pre-invasive forms of cancer that can be identified with endometrial sampling. An ultrasound scan and/or sampling of the endometrium is indicated in most circumstances of post-menopausal uterine bleeding as well as in select premenopausal patients with heavy, irregular bleeding to make a diagnosis of hyperplasia or cancer. Up to 25% of endometrial cancers occur in premenopausal women.

• The role of a ‘screening’ endometrial biopsy has been evaluated for tamoxifen, which is associated with an increased risk for endometrial cancer. To date, there is no evidence that routine screening endometrial biopsies results in greater detection or outcome differences in this population. Transvaginal uterine sonography also has a high false-positive rate and is not recommended as a screening methodology with tamoxifen.

Fallopian tube and ovarian cancer screening

• There is no evidence that screening with present modalities of ultrasound and CA125 impacts either diagnosis or outcome of ovarian or fallopian tube cancer in normal risk patients.

• The issue of whether a screening intervention might provide some protection for women at high risk for fallopian tube/ovarian/peritoneal cancer is as yet undetermined. The conventional standard of examination, ultrasound, and CA125 every 6 months for these patients who have not had risk reduction surgery has not been shown to make a difference in overall detection or outcome. The development of new serum markers, in combination with CA125 may offer an advance for the high-risk groups.

• The results of the ovarian cancer screening trial (UKCTOCS) will help to inform decisions regarding the implementation of an ovarian cancer screening programme. Public awareness of ovarian cancer is low and symptom screening may contribute to earlier diagnosis and treatment. Further reading

Bach PB, Jett Jr, Pastorino U, et al. Computed tomography screening and lung cancer outcomes. JAMA 2007: 279; 953–61.

Bonfill X, Marzo M, Pladevall M, et al. Strategies for increasing the participation of women in community breast cancer screening. Cochrane Database Syst Rev 2008; 2: CD00075320.

DeMartiniW, Lehman C, Patridge S. Breast MRI for cancer detection and characterization: a review of evidence-based clinical applications. Acad Radiol 2008;15:408–16.

Ferley J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002: cancer incidence, mortality and prevalence worldwide. IARC CancerBAse No. 5, version 2.0. Lyon: IARC Press 2004.

Schoenfeld P, Cash B, Flood A, et al. Colonoscopic screening of average risk women for colorectal neoplasia. CONCCeRN Study Investigators. N Engl J Med 2005: 352: 2061–8.

Warner E, Messersmith H, Causer P, et al. Systematic Review: using magnetic resonance imaging to screen women at high risk for breast cancer. Ann Intern Med 2008;148:671–9.

Wilson JMG, Junger G. Principles and practice of screening for disease. Public Health Papers 34. Geneva: WHO 1968.

Wright TC, Massad SL, Dunton CJ, et al. 2006 consensus guidelines for the management of women with cervical intraepithelial neoplasia or adenocarcinoma in situ. Am J Obstet Gynecol 2007;197:340–5. Internet resources

www.cancerscreening.nhs.uk/bowel/finalreport.pdf

www.cancerbackup.org.uk

www.isdscotland.org/isd/cancer

http://info.cancerresearchuk.org/cancerstats

www.eveappeal.org.uk/consensus

Cervical cancer Introduction

• Carcinoma of the cervix uteri remains the second most common cancer in women under the age of 35 despite a well-established UK-wide screening programme for the detection of pre-invasive disease.

• There are over 2700 new cases of cervical cancer diagnosed per annum in the UK, and regretfully, almost 1000 women still die from this disease each year (Cancer Research UK).

• Screening will only detect 30% of women with cervical cancer with the majority of cases occurring in those out-with the current screening programme. Risk factors

• Sexually active women are at risk of exposure to human papillomavirus (HPV). The aetiological association of HPV with genital disease is related most commonly to four subtypes which are 6, 11, 16, and 18. HPV subtypes 16 and 18 are attributed to 70% of cervical cancer cases.

• Early age at first intercourse, multiple sexual partners, oral contraceptives (some studies), cigarette smoking, low socioeconomic status, increased parity, family history, associated genital infections, and no circumcision in the male partner have also been documented as increasing the risk of developing cervical cancer. Clinical staging

• Cervical cancer is clinically staged using the FIGO (International Federation of Gynaecology and Obstetrics) criteria and was updated in 2009. This is solely based on clinical examination and does not include findings obtained from imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), or positron emission tomography (PET). Table 14.2.1 shows the complete FIGO staging system in detail.

• Current recommendations suggest that there is not sufficient evidence to advocate the use of sentinel node surgery in preference to pelvic lymphadenectomy to predict the tumour status of lymph nodes which are left in situ (SIGN 2008). Additionally, there is conflicting evidence regarding the average number of lymph nodes that should be attained during a surgical lymphadenectomy, and the exact relevance of this to long-term outcome in patients. Clinical presentation

• Patients can present with common, non-specific symptoms and signs which must always be investigated appropriately (SIGN 2008). These can include intermenstrual bleeding (IMB), post-coital bleeding (PCB), postmenopausal bleeding (PMB), a clinically suspicious looking cervix, blood-stained vaginal discharge, or pelvic pain.

• Women with symptoms or signs such as those above are more likely to be diagnosed with an advanced stage cervical tumour.

• Premenopausal women with abnormal vaginal bleeding, and some postmenopausal women, when considered appropriate by clinicians, should be tested for Chlamydia trachomatis infection (SIGN 2008).

• There is no evidence that performing an unscheduled cervical screening smear outwith the programme is of any use (SIGN 2008).

Table 14.2.1 FIGO staging criteria

Pathological classification

• Cervical cancer is diagnosed by detailed histopathological examination of tumour biopsies. The World Health Organization (WHO) histological classification system is summarized in Table 14.2.2.

• Histopathology reports should feature essential criteria that can be used by multidisciplinary teams to categorize women into higher risk groups for metastatic disease which will require adjuvant treatment (Van de Putte et al. 2005). These criteria should include tumour type, size, extent of disease, which includes involvement of vaginal wall or parametrium, depth of invasion, pattern of invasion (infiltrative or cohesive invasive front), lymphovascular space invasion (LVSI), status of resected margins (presence of tumour and distance from margin), status of lymph nodes (including site and number of nodes involved), and presence of pre-invasive disease if present (SIGN 2008).

Table 14.2.2 WHO histological classification, subtypes and examples of tumours of the cervix uteri

Radiological staging

• It is essential that all patients with clinically visible, histologically proven cervical cancer undergo accurate radiological imaging. This is essential in planning further management of patients at time of initial diagnosis, relapse, and in the event of treatment complications. It has been reliably shown that MRI is more exact in this context than computed tomography imaging (accuracies 40–97%) (Hricak et al. 2005). MRI (including urinary tract and para-aortic nodes) is therefore the first choice of imaging modality in most patients. If MRI is contraindicated or in clinically evident FIGO Stage IV disease, a post-contrast spiral or multislice CT scan of chest, abdomen and pelvis is indicated (SIGN 2008).

• Patients who are considered inoperable (i.e. those more likely to have nodal/metastatic disease) should undergo a PET scan (SIGN 2008), as they may still be curable with chemoradiotherapy. There is evidence that PET scanning can accurately detect metastatic pelvic/para-aortic lymphadenopathy (sensitivities 75–100%) (Loft et al. 2007), which can importantly alter management and survival of patients.

• Investigation of bladder and bowel using cystoscopy and sigmoidoscopy is used if clinical staging or radiological imaging (CT/MRI) cannot omit metastatic spread to these organs. Surgical management (non-fertility sparing)

• Surgical management of cervical cancer is becoming increasingly managed on an individual basis and should be performed by an appropriately trained gynaecological surgeon.

• Surgery for early stage disease can allow preservation of ovarian function, avoiding early menopause. In addition less radical surgery will avoid shortening/fibrosis of the vagina which conserves sexual function in women.

• Standard treatment for FIGO IA1 disease has been simple hysterectomy, although fertility-sparing surgery should now be considered, as discussed below.

• Radical hysterectomy (RH) is recommended for FIGO IB1 disease and involves concurrent removal of uterus, cervix, parametrial tissues, and upper vagina (SIGN 2008). This is normally combined with a pelvic lymphadenectomy.

• RH is not recommended for those patients with tumour measuring >4 cm (i.e. >FIGO IB2), where the incidence of nodal metastasis is 36% compared with 6% (2 cm. Fertility-sparing surgery

• Over the last decade, results from fertility-sparing surgery in younger women with cervical cancer have been encouraging. However, no RCTs have compared different methods of surgery for this (Dursun et al. 2007).

• In those women with early stage disease and no LVSI (FIGO IA1, IA2 or microscopic IB1), knife cone biopsy or large loop excision of the transformation zone (LLETZ) can be offered (SIGN 2008).

• A further uterine-sparing alternative to simple hysterectomy (FIGO IA2) or RH (FIGO IB1) is radical vaginal trachelectomy (RVT). This involves vaginal resection of the cervix, the upper 1–2 cm of vaginal cuff and the medial portions of the cardinal and uterosacral ligaments. The cervix is transected at the lower uterine segment and a prophylactic cerclage is placed at the time of surgery. If the tumour diameter is cervix is varied, being composed of non-keratinizing squamous epithelium and mucin-secreting columnar epithelium lining the ectocervix and endocervix respectively.

• The transformation zone (TZ), which evolves as columnar epithelium undergoes metaplasia to form squamous epithelium, is the area which exists between the original squamocolumnar junction (SCJ) and the new SCJ. This is the most common area for dysplastic lesions to develop. The TZ undergoes physiological metaplasia at different stages of development. The location of the TZ varies depending on age. In young women, the TZ is located on the outer surface of the immature cervix, whereas in menopausal women the TZ may recede into the endocervical canal, resulting in an unsatisfactory colposcopic examination. Human papillomavirus and precursors of invasive cervical carcinoma

• Over the last 30 years, the association between human papillomavirus (HPV) and risk of anogenital cancers has been indisputably established (zur Hausen 2002). Cervical cancer resulting from exposure to the HPV remains a principal cause of death from cervical cancer worldwide.

• Invasive cervical carcinoma is preceded by a spectrum of well-defined precursor lesions (dysplastic changes) within the squamous epithelium, described as cervical intraepithelial neoplasia (CIN).

• Precursor lesions are graded CIN I–III and reflect a continuum of change. These can be generally grouped into low grade CIN I and high grade CIN II/III. Whether or when a lesion will become an invasive cancer is unknown; however, progression is more likely with higher grade CIN, which can be considered the requisite precursor to cervical cancer. Analysis has shown that approximately 30–40% of CIN III progress to invasive carcinoma if untreated (McIndoe et al. 1984). Table 14.3.1 illustrates the pathological differences between HPV and the varying degrees of cervical dysplasia.

• HPVs infect humans principally through sexual contact and complete an infectious cycle only in a fully differentiated keratinised squamous epithelium. Infection occurs via microtrauma to the epithelium. Approximately 100 types of HPV have been isolated which are grouped into low and high risk (i.e. oncogenic) categories associated with benign lesions (warts) and precancerous/cancerous lesions, respectively.

• Ninety per cent of genital warts are due to predominantly HPV 6 and 11, which are considered low risk. Oncogenic high-risk HPVs isolated in the genital tract most commonly are HPV 16, which along with HPV 18, and close relatives including HPV 31 and HPV 33, are the cause of cervical cancer. In fact, in approximately 80% of CIN II/III biopsies and 99% of biopsies taken from invasive cervical carcinoma (Smith et al. 2007), high-risk HPV DNA sequences can be isolated.

• The natural history of CIN and HPV, in effect, mirror one another. Young sexually active people are at high risk of acquiring HPV infection. The lifetime risk of acquiring a genital infection with HPV is approximately 80%, with most infections resolving without clinical manifestations or a serological antibody response (Walboomers et al. 1999).

• When a low-risk lesion develops, i.e. CIN I, this represents the infectious cycle of HPV. Ten per cent of women will not clear the virus and develop persistent HPV infection with a strong cell-mediated immune response. Those with high-risk HPVs will carry a risk of progression to CIN II/III, and thereafter a significant risk of development of invasive carcinoma. However, many women who acquire HPV infection will not progress to cervical cancer, and therefore other cofactors may affect the risk (Muñoz et al. 2006), as shown in Table 14.3.2.

Table 14.3.1 Colposcopic and pathological features of low and high-grade lesions

Cervical screening

• The UK National Health Service Cervical Screening call and recall Programme (NHSCSP) was established in the late 1980s and has successfully aided prevention of cervical cancer, saving approximately 5000 lives per year in the UK (Peto et al. 2004).

• The NHSCSP permits early detection, treatment and follow-up of precancerous changes of the cervix. The success behind the programme relies on the efficient call, recall and tracking of women, the use of quality assurance tools, a multidisciplinary approach to diagnosis and treatment, accredited training of clinicians and the widespread use of national clinical guidelines.

Table 14.3.2 Definite and potential cofactors affecting risk of cervical HPV carcinogenesis

• Limitations to the programme do exist, including the inability to prevent or treat HPV by the programme, and the fact that some women will develop a malignant lesion of the cervix despite regular screening. False-positive/negative results are foreseeable due to the subjective nature of identification of abnormal cells and potential poor quality of samples.

• Quality has however improved since the introduction of liquid-based cytology (LBC), with a clear reduction in reported inadequate smears. Other suggested advantages of LBC versus conventional Pap smear tests include a more homogeneous sample collection, increased sensitivity and specificity and more efficient laboratory handling. LBC produces an enriched cellular sample with excellent preservation which is the ideal platform for ancillary tests and future molecular profiling studies.

• Cervical screening intervals have been standardized. In England and Wales, invitations are sent on a 3 yearly basis to women aged 25–49 years and 5 yearly to those aged 50–64 years (NHS Cervical Screening Programme). The lower age for first invitation to screening in Scotland is 20 years.

• The current geographical disparity between the lower ages to commence screening has raised several issues. The advantage of not screening women and characteristic proliferation of the microvasculature (punctuation or mosaicism). Colposcopic features of low and high grade CIN are shown in Table 14.3.1.

• Cervical dysplasia should be treated when there is a reasonable expectation that, if untreated, the patient will run the risk of subsequent development of cervical carcinoma. The likelihood of resolution of that lesion and the risk associated with treating the patient must be taken into account.

• The majority of low-grade lesions (CIN I) will regress spontaneously and do not require any treatment following a period of colposcopic observation.

• High grade lesions (CIN II/III) are at risk of progression to invasive disease if untreated (McIndoe et al. 1984), and should therefore be managed with ablative or excisional methods of treatment. However, high-grade lesions diagnosed in pregnancy may be followed up conservatively until delivery. Ablative treatment is only feasible if the lesion is visualized in its entirety at colposcopy and there is biopsy-confirmed pre-invasive disease with no glandular abnormality or microinvasive element. Ablation includes the use of cryotherapy, cold coagulation, electrocautery and carbon dioxide laser vaporization. The most commonly used excisional technique is LLETZ (large loop excision of the transformation zone). These procedures can be performed in an outpatient setting under local anaesthetic.

• A policy of ‘see and treat’, i.e. look and LLETZ, has become more common in recent years. This is performed on those lesions considered high grade, but without actual histological confirmation of this.

• Patients should undergo cytological evaluation at 6 and 12 months following surgical treatment of CIN. Annual smears should continue thereafter for a minimum of 5 years before a return to the normal screening programme. HPV testing: when, how and why?

• A Health Technology Assessment review has concluded that HPV testing cannot currently be recommended for primary screening without further research.

• Current evidence does, however, support limited introduction of the test in order to improve the management of women with cytological samples showing borderline nuclear abnormality (BNA) or mild dyskaryosis.

• There remains uncertainty about the negative predictive value of an HPV test in the presence of persistent mild dyskaryosis and the safety associated with reduced surveillance of patients. This aspect has been evaluated in pilot studies.

• TOMBOLA (Trial of Management of Borderline and Other Low grade Abnormal smears) is a randomized controlled trial that determines both the most effective and efficient management options for women whose cervical smear tests indicate low grade CIN, incorporating the role of HPV testing for triaging these women. Clinical, economic and psychosocial outcomes have been determined in early studies. In contrast, most previous studies had looked at high grade abnormalities. Initial results from TOMBOLA suggest that women with a negative HPV test following a BNA smear result may not require referral for colposcopy. HPV risk was increased in non-white women, smokers and those using hormonal contraceptive measures.

• ARTISTIC (A Randomised Trial of HPV Testing in Primary Cervical Screening) aims to examine whether HPV testing will add effectiveness to the current cervical screening programme. Preliminary results (Kitchener et al. 2006) suggested that HPV testing has potential as a primary screening test in future practice but this would lead to a significant increase in retesting and referral rates.

• It has recently been shown that HPV testing may be a more accurate assessment of ‘test of cure’ following treatment for CIN than current colposcopy and/or cytology methods (Kitchener et al. 2008). Data suggest that those women who are both cytology negative and HPV negative at 6 months after treatment for CIN can be returned to routine 3 year follow-up.

• Overall, the introduction of HPV testing to triage women could facilitate a reduction in referral to colposcopy, length of follow-up following treatment of CIN, overall cost to the NHS and, importantly, anxiety for women. Psychological sequelae of cervical screening and HPV testing

• Screening, abnormal smears and treatment of dysplastic changes of the cervix are well documented causes of anxiety and psychosexual dysfunction amongst affected women, with no correlation shown between grade of abnormality and levels of anxiety (Gray et al. 2006).

• More recently, a number of studies using well validated psychological tools have addressed reactions to being informed of a positive HPV infection (Maissi et al. 2004). The highest level of anxiety has been shown to be amongst those women with abnormal cytology and a positive HPV test compared to those with a negative HPV result. However, women with a positive HPV test but negative cytology show similar scores.

• It is imperative to continue to increase awareness of the possibility and consequences of HPV infection through educational resources in younger people. HPV vaccines

• Screening tests are a means of detecting abnormalities within cells at an early stage, but the solution to any viral disease is a vaccine.

• Developing an attenuated HPV vaccine has been hampered by the lack of an effective culture system, but this has been solved by the production of virus-like particles (VLPs) which are structurally like a virus but essentially harmless due to a lack of DNA (Schiller et al. 2000).

• Two VLP-based prophylactic vaccines have proven to be extremely effective in clinical trials. Gardasil is a quadrivalent vaccine containing all four HPV types and is protective against genital warts (HPV 6 and 11) as well as cervical cancer (HPV 16 and 18). Cervarix is a bivalent vaccine which contains HPV 16 and 18, targeting cervical cancer. Clinical trials have shown that Gardasil and Cervarix are 100% effective in preventing infection with HPV 16 and 18 for 5 years and 4.5 years respectively. Both vaccines result in significantly higher antibody titres than those produced by natural infection (Harper et al. 2006; Villa et al. 2006). There is also preliminary evidence of cross-protection from vaccination against HPV 31 and 45 (Harper et al. 2006).

• The UK NHS HPV vaccination programme commenced in September 2008. The bivalent vaccine (Cervarix) will be used. This involves the offer of immunisation of all girls 12–13 years in a phased programme involving schools and local NHS services, with a 2 years catch-up programme commencing in 2009 for girls up to, but not including, 18 years old. There are no plans to immunise boys against HPV.

• Future research must focus on the exact degree of cross-protection against other HPV types and the effect of this vaccine on preventing high grade CIN. It may be envisaged that eventually the vaccination programme may affect the delivery of the current screening programme. However, as the vaccination programme currently will only be protective against 70% of cervical cancers, neither will replace one another and both will remain vital for the foreseeable future. Further reading

Gray NM, Sharp L, Cotton SC, et al. Psychological effects of a low-grade abnormal cervical smear test result: anxiety and associated factors. Br J Cancer 2006;94:1253–62.

Harper DM, Franco EL, Wheeler CM, et al. Sustained efficacy up to 4.5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from randomised control trial. Lancet 2006;367:1247–55.

Holowaty P, Miller AB, Rohan T, To T. Natural history of dysplasia of the uterine cervix. J Natl Cancer Inst 1999;91:252–8.

Kitchener HC, Almonte M, Wheeler P, et al. HPV testing in routine cervical screening: cross sectional data from the ARTISTIC trial. Br J Cancer 2006;95:56–61.

Kitchener HC, Walker PG, Nelson L, et al. HPV testing as an adjunct to cytology in the follow up of women treated for cervical intraepithelial neoplasia. Br J Obstet Gynaecol 2008;115:1001–7.

McIndoe WA, McLean MR, Jones RW, Mullins PR. The invasive potential of carcinoma in situ of the cervix. Obstet Gynecol 1984;64:451–8.

Maissi E, Marteau TM, Hankins M, et al. Psychological impact of human papillomavirus testing in women with borderline or mildly dyskaryotic cervical smear test results: cross sectional questionnaire study. BMJ 2004; 328: 1293.

Muñoz N, Castellsague X, Barrington de Gonzalez A, Grissmann L. HPV in the etiology of human cancer. Vaccine 2006; 24 (Suppl 3): S1–10.

National Health Service. Cervical screening programme, England, 2006–7. NHS Health and Social Care Information Centre, Statistical Bulletin: wwwicnhsuk/statistics-and-data-collections/screening/cervical-cancer/cervical-screening-programme-2006–07-% 5Bns% 5D

Peto J, Gilham C, Fletcher O, Matthews FE. The cervical cancer epidemic that screening has prevented in the UK. Lancet 2004;364:249–56.

Schiller JT, Lowy DR. Papillomavirus-like particle vaccines. J Natl Cancer Inst Monogr 2000;28:50–4.

Smith JS, Lindsay L, Hoots B, et al. Human papillomavirus type distribution in invasive cervical cancer and high-grade cervical lesions: a meta-analysis update. Int J Cancer 2007;121:621–32.

Villa LL, Costa RLR, Petta CA, Andrade RP, et al. High sustained efficacy of a prophylactic quadrivalent human papillomavirus types 6/11/16/18 L1 virus-like particle vaccine through 5 years of follow up. Br J Cancer 2006;95:1459–66.

Walboomers JMM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:12–19.

Wellings K, Nanchahal K, Macdowell W, et al. Sexual behaviour in Britain: early heterosexual experience. Lancet 2001;358:1843–50.

zur Hausen H. Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2002;2:342–50. Patient resources

www.fightcervicalcancer.org.uk

Jo’s Trust, Weedon Villa, Everdon, Northamptonshire, NN11 3BQ: www.jotrust.co.uk

British Society for Colposcopy and Cervical Pathology www.bsccp. org.uk

Cancer Research UK: www.cancerhelp.org.uk

Tell Her: An Educational Website on Cervical Cancer and HPV: www.tellher.com

Neoplastic conditions of the endometrium

Endometrial cancer is rising in incidence in the developed world, in association with rising obesity levels. Oestrogens are powerful stimulants of endometrial proliferation, and a ‘premalignant’ condition, endometrial hyperplasia, is well described. Endometrial hyperplasia may precede, or coexist with, endometrial cancer. Endometrial hyperplasia is further subdivided by WHO classification into simple and complex hyperplasia, and the management of this condition is discussed below.

Although more common in postmenopausal women, up to 25% of cases of endometrial cancer occur in premenopausal women. The majority of cases are endometrioid adenocarcinomas, and a common presenting symptom is postmenopausal bleeding. Further investigation with ultrasound/endometrial sampling is indicated in patients with postmenopausal bleeding to exclude endometrial hyperplasia/cancer. The majority of endometrial cancers are diagnosed at an early stage following presentation and investigation.

The mainstay of treatment for endometrial cancer remains surgery in the form of total abdominal hysterectomy and bilateral salpingo-ophorectomy. FIGO staging for endometrial cancer was revised in 1988 to become a surgical staging, including pelvic lymph node status. In recent years, the value of routine lymphadenectomy has been questioned.

Overall 5 year survival for all stages of endometrial carcinoma is 86%, with 97% for disease confined to the uterus.

Following surgery, adjuvant radiotherapy may be considered for patients at high risk of local relapse.

The role of chemotherapy in endometrial cancer has yet to be adequately defined. Epidemiology

There is considerable geographic variation in the rates of endometrial cancer (Table 14.4.1).

• Endometrial cancer occurs most commonly in postmenopausal women, with 90% of cases occurring in women over the age of 50

• Women under 50 are an important group who may present with irregular menstrual bleeding, and investigation to rule out endometrial cancer in this group should be considered.

• A further important 5% of cases occur in women under 40. In these patients polycystic ovarian syndrome and genetic cancer predisposition due to Hereditary Non Polyposis Colon Cancer (HNPCC) may result in the development of endometrial cancer.

Table 14.4.1 Age-standardized incidence rate (per 100 000) for endometrial cancer

A number of risk factors for endometrial cancer have been identified:

• Obesity: a body mass index of greater than 29 is associated with a threefold increased risk of endometrial cancer.

• Polycystic ovarian syndrome: the anovulatory cycles lead to unopposed oestrogen stimulation of the endometrium.

• Parity: epidemiological evidence suggests that parous women have a decreased risk of developing endometrial cancer, with the first pregnancy associated with the maximum protective effect.

• Unopposed oestrogen hormone replacement therapy (HRT): the risk of endometrial hyperplasia and endometrial cancer in patients taking unopposed oestrogen HRT is well described, with an increased relative risk of around 2.7 for unopposed oestrogen HRT compared with HRT with progestogen added for at least 10 days per month.

• Anti-oestrogens: the selective oestrogen receptor modulator Tamoxifen is widely used in the treatment of breast cancer. Tamoxifen blocks the effect of oestrogen on breast tissue but has a stimulatory effect on the endometrium. The relative risk of endometrial cancer for patients using Tamoxifen for 5 years is 2.0.

• Feminizing ovarian tumours: although very rare, oestrogen secretion from ovarian tumours such as granulosa cell tumours can cause endometrial hyperplasia and cancer.

• Family history: HNPCC is a cancer susceptibility syndrome caused by a germline mutation in one of the DNA mismatch repair genes. Those with mutations have an increased risk of endometrial and ovarian cancer in addition to their risk of nonpolyposis colon cancer (Sonada et al. 2006). Consideration should be given to this syndrome in young patients presenting with either endometrial or non-polyposis colon cancer. Aetiology

Bohkman proposed a new classification for endometrial cancer into Types I and II in 1983.

Type I endometrial cancer:

• predominates, accounting for more than 80% of cases

• is associated with oestrogenic stimulation of the endometrium

• is often associated with endometrial hyperplasia

• histologically is usually Grade 1 to 2 endometrioid adenocarcinoma

• is associated with mutations in the PTEN tumour suppression gene

Type II endometrial cancer:

• includes serous, clear cell, carcinosarcoma, small cell and other rare subtypes

• is associated with a higher risk of metastatic disease

• is not associated with oestrogenic stimulation of the endometrium

• is associated with mutations in the p53 tumour suppression gene

The risk factors for endometrial hyperplasia, and therefore Type 1 endometrial cancer, are linked to oestrogenic stimulation of the endometrium and have been outlined above. Pathology

The pathology of endometrial hyperplasia, endometrial carcinoma and smooth muscle tumours is considered below:

Endometrial hyperplasia

• Defined as an increase in the ratio of endometrial glands to stroma greater than one to one.

• Endometrial hyperplasia is divided into simple and complex hyperplasia.

• Hyperplasia is further subdivided depending upon the presence or absence of cellular atypia.

• Although simple hyperplasia can often be differentiated from complex in view of the gland to stroma ratio (may be up to three to one in complex hyperplasia), the distinction between complex hyperplasia with atypia and G1 endometrial cancer can be difficult.

Type 1 endometrial carcinoma (80% of cases)

• Endometrioid adenocarcinoma.

• Usually G1 or G2, arising on a background of endometrial hyperplasia.

Characterized histologically (according to FIGO system) as

• G1: well-formed glands with less than 5% solid non-squamous areas.

• G2: well-formed glands with between 6% and 50% solid non-squamous areas.

• G3: well-formed glands with more than 50% solid nonsquamous areas.

Other features such as degree of cytological atypia and infiltrative growth pattern may increase the grading to G3 (high grade).

Type 2 endometrial carcinoma (10% of cases)

• These tumours are all designated high grade.

• Serous, clear cell, mixed (adenosquamous) and small cell carcinomas form the majority of Type 2 carcinomas.

• Serous and clear cell subtypes are the most common.

• These subtypes may be found in association with high-grade endometrioid elements in a tumour.

• Ten per cent of the volume of the tumour should consist of serous or clear cell elements to allow classification of the tumour as serous or clear cell.

• Serous tumours frequently metastasize to the omentum and/or paraaortic lymph nodes.

Smooth muscle tumours

Smooth muscle tumours of the uterus range from the very common benign leiomyoma (fibroid) to sarcomas such as leiomyosarcoma or carcinosarcoma.

• Known also as malignant mixed Müllerian tumours (MMMT), carcinosarcoma, together with leiomyosarcoma, account for the majority of malignant smooth muscle tumours.

• Malignant uterine tumours account for under 5% of the total.

• The aetiology of these malignant tumours remains controversial.

• Of all patients with fibroids, less than 1% will have a leiomyosarcoma. Clinical suspicion of the diagnosis should be raised if a patient has a rapidly enlarging fibroid.

• Also included in this group of tumours are variants with an uncertain malignant potential (smooth muscle tumour of uncertain malignant potential (STUMP)).

• Histological categorisation of these intermediate risk groups of smooth muscle tumours may require specialist histology review.

• It has been suggested that factors such as obesity, oestrogen stimulation and nulliparity increase the risk of MMMT.

• There is a risk of both local and distant recurrence with leiomyosarcoma and MMMT, with lung metastases occurring frequently. Staging

The staging of endometrial cancer is surgical following the 1988 FIGO classification, and updated in 2009 (Tables 14.4.2 and 14.4.3).

In cases unsuitable for surgery the FIGO clinical staging system of 1971 can be used. Clinical approach

Endometrial hyperplasia

Presentation

• Endometrial hyperplasia may be asymptomatic and be detected incidentally following either a pelvic ultrasound or cervical smear (reported as glandular neoplasia).

• Postmenopausal bleeding is associated with endometrial hyperplasia in 15% of cases.

• Premenopausal bleeding irregularities leading to a diagnosis of endometrial hyperplasia may occur in patients with risk factors such as obesity, polycystic ovarian syndrome or Tamoxifen use as discussed above.

Table 14.4.2 FIGO staging for carcinoma of the corpusuteri (2009)

Table 14.4.3 FIGO staging for uterine sarcomas (2009)

The investigation of patients with postmenopausal bleeding or irregular bleeding in patients with risk factors for endometrial hyperplasia is described below.

The management of endometrial hyperplasia is primarily determined by the histological finding of cellular atypia in the endometrial biopsy.

• Endometrial hyperplasia without cellular atypia is associated with a low risk of progression to malignancy and treatment with progestogens is appropriate.

• Following initial treatment it is recommended that histological proof of regression to normal endometrial histology is obtained at 3–6 months.

• If this has not been achieved, consideration may be given to increasing the dose of progestogen or to hysterectomy.

• There is controversy about the optimum follow up regimen for patients treated in this way, but some authors advocate an annual assessment of the endometrium.

• The risk of coexisting malignancy in patients with endometrial hyperplasia with cellular atypia is commonly quoted as 20–25%, but at least one study found the risk to be as high as 50%.

• In this group, hysterectomy is recommended. In some cases, for example young patients wishing to retain fertility, treatment with progestogen has been used. Careful assessment of response is essential in this group. Patients should be warned of:

• Low pregnancy rates in studies of conservative management of endometrial hyperplasia/early endometrial cancer (15–30%)

• the significant risk of persistence or progression of the endometrial changes (20–50%)

Endometrial cancer

Symptoms of endometrial cancer

Postmenopausal bleeding is the most common presenting symptom in patients with endometrial cancer.

The clinical approach to diagnosis is based on the fact that 90% of patients with endometrial cancer are diagnosed over the age of 50, and up to 10% of patients with post-menopausal bleeding have an underlying carcinoma.

In premenopausal women, irregular bleeding may lead to endometrial sampling resulting in the diagnosis.

Investigation of postmenopausal bleeding

Increasingly, units have established ‘postmenopausal bleeding clinics’ in order to exclude endometrial cancer as a cause for postmenopausal bleeding.

• Using transvaginal ultrasound measurement of the endometrium, the pretest risk of endometrial cancer can be reduced from around 10% to 1% if the endometrial thickness is less than 5 mm.

• Endometrial sampling can be performed as an outpatient procedure to further investigate patients with thickened or irregular endometrial scan findings.

• Hysteroscopy can also be performed as an outpatient procedure to obtain histological samples of the endometrium.

• Hysteroscopy is viewed as the ‘gold standard’ investigation, either outpatient or inpatient.

Positive peritoneal cytology following hysteroscopy

Following hysteroscopy, the significance of intraperitoneal tumour cells is controversial. According to FIGO Staging, positive peritoneal cytology upstages to Stage 3a.

In one review of 17 studies, the positive cytology rate increased with grade of tumour and depth of invasion from 8% to 16% (G1 compared with G3), 8% to 17% (superficial versus deep myometrial invasion). Although there was a strong association between positive washings and recurrence, the conclusion of the review was that other adverse features predominated over the positive washings to explain these findings (Slomovitz et al. 2005).

In the review, however, only patients with G1, inner half myometrial invasion with no LVSI had unaffected survival when the washings were positive.

New developments in diagnostic investigation of PMB

With the advent of liquid based cytology, and the addition of immunocytochemistry it is possible to obtain cytological samples from the endometrial cavity. Promising results from early studies have been published, with cytohistological concordance rates of up to 98% reported.

Assessment of patients with endometrial cancer

Once the diagnosis of endometrial cancer is established, surgery is the mainstay of treatment for apparent early stage disease.

Surgery is also the mainstay for patients with smooth muscle cell tumours of the uterus, without pelvic lymphadenectomy.

Preoperative assessment with imaging modalities such as ultrasound, CT and MRI has been used to detect metastatic disease in patients with high-grade disease, and spread to the cervix or beyond the uterus.

Chest X-ray, and thorough preoperative assessment must also be performed. Many patients with endometrial cancer have a high BMI, and are at increased risk of:

• thromboembolic complications

• wound infection/breakdown

• chest infection.

Consideration is given to performing an extended hysterectomy if there is evidence of cervical stromal involvement. MRI has been used to assess this but has limitations in terms of both false-positive and false-negative results.

Treatment planning, in the primary, adjuvant, or recurrent disease setting, takes place in formal multidisciplinary team meetings. With members drawn from gynaecological oncology, medical, and clinical oncology, histo- and cytopathology, radiology, and often palliative care services, individual treatment plans are drawn up. The treatment plans are formulated with reference to agreed treatment protocols which are based on regional guidelines.

Surgery may still be appropriate in patients with evidence of metastatic disease in order to control symptoms of bleeding.

Surgery in endometrial cancer

The recommendation for surgical staging of endometrial cancer is for a midline laparotomy to be performed to allow adequate assessment of the peritoneal cavity/contents.

Peritoneal washings should be taken prior to any other procedure, followed by extrafascial total abdominal hysterectomy and bilateral salpingooophorectomy.

In patients with cervical stromal involvement, a radical hysterectomy may be considered as above.

For patients with a preoperative diagnosis of serous carcinoma, additional staging in the form of omental and paraaortic node biopsy should be considered.

Although part of the FIGO staging, the role of routine pelvic and/or paraaortic lymphadenectomy is controversial.

• In a survey of practice, 54% of American Gynaecological Oncologists reported that they routinely performed lymphadenectomy compared to 24% in the UK.

• Some studies have suggested a therapeutic benefit of lymphadenectomy, but the recent ASTEC trial (to date the only prospective, randomised trial to determine the answer to this question) did not demonstrate a therapeutic advantage to the lymphadenectomy group. At 3 years of follow up there were identical survival rates (89% in the lymphadenectomy group versus 88% in the no lymphadenectomy group).

In recent years the role of laparoscopic hysterectomy has been evaluated in endometrial cancer.

• In published series the recurrence rates are similar for those treated by open or laparoscopic hysterectomy.

• Hospital stay is reduced in the laparoscopic group, although the risk of complications due to injury to intra-abdominal structures was higher in the laparoscopic groups in early studies. Radiotherapy

For patients with endometrial cancer radiotherapy is delivered as external beam radiotherapy (EBRT) or by brachytherapy.

• EBRT involves the use of radiation whose source is distant from the patient.

• brachytherapy involves the use of sealed radiotherapy sources placed in close proximity to the treated tissue (e.g. intracavitary treatment).

Radiotherapy is used in three situations in endometrial cancer: as primary treatment, adjuvant treatment following surgery, or for palliation.

As primary treatment in patients not fit for surgery

Although inferior to surgery, primary radiotherapy has a 60% 5-year survival in apparent FIGO Stage 1 disease.

As adjuvant treatment following surgery

Adjuvant radiotherapy to the vaginal vault and/or pelvis has been used extensively in patients with adverse prognostic features such as:

• high-grade disease

• deep myometrial invasion

• lymphvascular space invasion (LVSI).

In this situation many studies have shown a reduction in vault recurrence in patients treated with radiotherapy, but no increase in overall survival.

Adjuvant radiotherapy can be associated with side-effects which are described as ‘early’ or ‘late’.

• Early side-effects of pelvic radiotherapy include diarrhoea, cystitis and skin reactions, and in some cases bone marrow suppression in the pelvis.

• Late effects include bowel obstruction/fistula formation, changes in bladder capacity, and vaginal atrophy and possible narrowing of the vagina. In order to prevent vaginal narrowing it is recommended to use vaginal dilators during and following treatment.

Recent research has tried to identify subgroups of patients who might expect a survival advantage following adjuvant radiotherapy, and in whom the risk of side effects might be more justified following adjuvant radiotherapy. The PORTEC study in patients with FIGO stage 1 disease confirmed the findings of earlier studies with lower rates of vaginal/pelvic recurrence in the radiotherapy group with no overall survival advantage.

In addition, the PORTEC study showed that salvage treatment with radiotherapy in patients who develop a vault recurrence can result in long term survival in patients who have not previously received radiotherapy (Creutzberg et al. 2003).

• Only 4% of patients who had received radiotherapy relapsed locally compared with 14% of those who did not receive radiotherapy but

• There was an 89% complete response rate to radiotherapy following local relapse in patients who had not received radiotherapy, with a 65% 5year survival.

• Overall 8-year survival was 71% in the group who received adjuvant radiotherapy compared to 77% in the group who did not receive adjuvant radiotherapy.

• A recent systematic review concluded that

• adjuvant EBRT should not be used for low-risk (Ia, IbG1) or intermediate-risk (1bG2) endometrial cancer but

• is associated with a 10% survival advantage for high-risk (1cG3) tumours.

In this review the authors suggest that these findings challenge the role of staging lymphadenectomy (Johnson et al. 2007). The review interprets the data that show the reduction in local relapse following adjuvant radiotherapy without improved overall survival, and considers the risk of severe complications following radiotherapy of 3%, and the risk of minor complications at 20%.

For palliation

In addition to radiotherapy for relapsed disease at the vaginal vault which is associated with a good long-term prognosis as discussed above, radiotherapy may also be used to control bleeding in a palliative setting. In this situation one or two fractions are given. It may even be possible to use radiotherapy in this situation in patients who have previously undergone radiotherapy treatment previously.

Follow-up

Currently, the role of routine follow-up for patients with endometrial cancer confined to the uterus is being debated. Following the publication of small studies which do not show a difference in outcome for patients followed up in clinic routinely versus telephone consultations, the results of larger studies are awaited to confirm these findings. Treatment of advanced disease

The adjuvant treatment of Stage 2 and more advanced disease remains controversial.

As discussed, for patients identified with possible Stage 2 disease on preoperative assessment, consideration of extended hysterectomy should be given.

Endometrial cancer can be classified as FIGO stage 3a with one or more of the following features:

• positive washings (see section above)

• serosal tumour involvement

• direct extension to the adenexa.

The role of adjuvant therapy in this situation remains controversial.

Adjuvant radiotherapy may be used to try and control disease in the pelvis as described above.

Chemotherapy

Chemotherapy regimens including cisplatin, carboplatin, doxorubicin and paclitaxel have been used.

Patients considered for chemotherapy treatment for endometrial cancer are often elderly, have comorbidities and may have undergone previous radiotherapy. The risk of chemotherapy in this population may be higher than for other patient groups, and this must be taken into account when considering chemotherapy treatment in this situation.

To date there is little evidence to show an overall survival advantage following chemotherapy treatment in endometrial cancer. Treatment of recurrent disease

A common site of recurrence following surgery alone for endometrial cancer is the pelvis, and the vaginal vault in particular.

• Salvage treatment with radiotherapy in this situation can result in 5-year survival of more than 50% in published studies.

• Any role for exenterative surgery in recurrent disease must be considered in the context of the considerable morbidity reported in such patients.

• For patients with recurrent disease systemic treatment with progestogens is often used. The response rates of 15–30% have been reported, with median overall survival times of up to 13 months. The side-effects include weight gain and an increased risk of thromboembolic events.

• Chemotherapy regimens including platinum and taxane combinations have shown activity in recurrent endometrial cancer, with response rates of up to 30–40% reported, with median overall survival times of 13 months. Further reading

Creutzberg CL, van Putten WLJ, Koper PC, et al. Survival after relapse in patients with endometrial cancer: results from a randomised trial. Gynecol Oncol 2003;89:201–9.

Johnson N, Cornes P. Survival and recurrent disease after postoperative radiotherapy for early endometrial cancer: systematic review and meta-analysis. Br J Obstet Gynaecol 2007;114:1313–20.

Slomovitz BM, Ramondetta, LM, Lee CM, et al. Heterogeneity of Stage IIIa endometrial carcinomas: implications for adjuvant therapy. Int J Gynecol Cancer 2005;15:510–16.

Sonoda Y, Barakat RR. Screening and the prevention of gynaecologic cancer: endometrial cancer. Best Practice Res Clin Obstet Gynaecol 2006;20:363–77.

Brinton LA, Lacey Jr JA, Devesa SS, Sherman ME. Epidemiology of uterine corpus cancers. In: Gynecologic cancer: controversies in management. Philadelphia: Elsevier Churchill Livingstone; 2004 189–207.

Creasman WT. Adenocarcinoma of the uterus. In: Clinical gynecologic oncology. China: Mosby Elsevier 2007: 147–84. Internet resources

www.cancerresearchuk.org

www.cancer.gov/cancertopics/types/endometrial

www.iarc.fr/ Patient resources

www.cancerbacup.org.uk

www.cancerhelp.org.uk

www.NHS.uk/illness

Gestational trophoblastic neoplasia Definition

Gestational trophoblastic neoplasia (GTN) encompasses a diverse group of lesions that originate from the fetal trophoblast whose behaviour may be benign or malignant. The lack of histological confirmation in its malignant form often poses much difficulty and controversy in its diagnosis, investigations, and management. Epidemiology

• There are no reliable figures on the true incidence of GTN due to the difficulties mentioned above.

• It is considered to be a rare disease in the UK, with a calculated incidence of 1 in 714 live births (RCOG 2004). However, with the problems associated with reporting, due to lack of histological diagnosis, especially with regards to partial moles, the true incidence is considered to be higher.

• There also appears to be ethnic variations in the incidence of GTN in the UK, with women from Asia having a higher incidence compared with non-Asian women (1 in 387 versus 1 in 752 live births)

• GTN may occur at any age but appears to be more frequent in teenagers and women above 40. The risk increases with previous history of GTN. Pathology

• GTN is a heterogeneous group of diseases that arise from one or more of the three main types of trophoblasts found in the placenta: cytotrophoblast, syncytiotrophoblast, or intermediate trophoblast. They can be benign or malignant. Histologically, it is classified into complete hydatidiform mole (CHM), partial hydatidiform mole (PHM); invasive mole (IM), choriocarcinoma (CC), and placental site trophoblast tumour (PSTT).

• In recent years, new entities, including epithelioid trophoblastic tumour, have been added to this family.

• Hydatidiform mole (both CHM and PHM) is the most common form of GTN and can behave in a malignant or benign fashion. However, IM, and CC are considered to be malignant. PSTT has uncertain biological behaviour.

• CHMs are diploid and androgenetic (paternal) in origin, with no evidence of fetal tissue. They arise either as a consequence of duplication of the haploid sperm following the fertilization of an ‘empty’ ovum or after dispermic fertilization of an ‘empty’ ovum. CHMs carry a 20% risk of malignant sequelae.

• PHMs are triploid in origin with two sets of paternal haploid genes and one set of maternal haploid genes. They occur after dispermic fertilization of an ovum. There is evidence of fetal tissue or fetal red blood cells. PHMs rarely are followed by GTN but require the same follow up for potential malignant sequelae as a CHM. Diagnosis and management of GTN

Clinical presentation

• The classic presentation of molar pregnancy include vaginal bleeding, anaemia, excessive uterine enlargement, hyperemesis gravidarum, early pre-eclampsia and hyperthyroidism, associated with remarkably elevated human chorionic gonadotrophin (hCG) titres. However, with the availability of high-resolution vaginal ultrasound and the practice of early dating scan, currently almost all patients with molar pregnancy are diagnosed and treated before they develop the classical symptoms. In almost all cases of PHM, the initial diagnosis is one of incomplete or missed abortion and the diagnosis is only made after the histological examination of the products of conception. Hence, the importance of sending for histology the products of conception in every case of pregnancy failure.

• Women who present with persistent abnormal vaginal bleeding after any pregnancy (normal, abortion, ectopic, or molar) should have a pregnancy test to exclude GTN.

• A high index of suspicion is necessary in women of reproductive age who present with acute respiratory or neurological symptoms, as these may be evidence of GTN in the lungs or brain. The finding of pulmonary nodules on a chest radiograph after normal pregnancy suggests GTN. These events may occur remotely from the antecedent pregnancy. A simple pregnancy test may help clinch the diagnosis.

• A young woman with an unknown primary neoplasm or poorly explained hyperthyroidism should have her serum hCG tested.

• Approximately half the cases of GTN follow molar pregnancy, one-fourth follow normal pregnancy, and one-fourth follow abortion or ectopic pregnancy.

• The major risk factors for molar pregnancy include maternal age (older than 40 and younger than 20 years) and a history of molar pregnancy.

• The risk of developing a second molar pregnancy after a primary mole is approximately 20–40 times greater than the initial risk.

Investigations

• Ultrasound remains the most important diagnostic modality in GTN. In CHM, the uterus shows the absence of a gestational sac but the cavity is filled with cystic spaces: the typical ‘snowstorm’ appearance. Its value is more limited in PHM with no definite diagnostic features. More recently, high-resolution vaginal ultrasound, with colour Doppler, has been used to diagnose GTN involving the uterine myometrium (IM and CC).

• Measurement of serum levels of hCG has been the traditional method of diagnosing GTN. Although very high levels are suspicious, there is no pathognomonic value of hCG to make the diagnosis. However, it is the tumour marker par excellence in the follow-up of patients with GTN to assess response to treatment and to diagnose recurrence.

• Chest X-rays are mandatory in all cases of GTN as the lungs are the commonest site for metastases.

• CT scans and MRI may be used judiciously in metastatic survey of other organs, especially the liver and brain.

Staging and prognostic scoring

GTN is unusual in gynaecological oncology for several reasons. The most common manifestation of GTN, the hydatidiform mole remains benign in most patients. In malignant GTN, the histology is not often available and treatment is based on serum levels of hCG and radiological images. Also, the anatomic stage is less important than other risk factors in making treatment decisions. For several years, various staging and prognostic scoring systems have been used by clinicians to assess the prognosis and selection of the appropriate treatment of GTN, especially, metastatic GTN. However, this did not allow a meaningful comparison of the results and assess the efficacy of the treatment between different centres. The recognition that a universally accepted, precise system of classification would be needed to standardize clinical care has led to the amalgamation of two widely used systems, namely those of the Federation Internationale Gynecologie et Obstetrique (FIGO) staging and the World Health Organization (WHO) prognostic scoring systems (Table 14.5.1). A discussion on the merits of the individual system is beyond the scope of this section (Kohorn 2001). Suffice it to say that the current FIGO classification includes both the stage and scoring, for example Stage III:8, i.e. stage III disease with a prognostic score of 8. In the scoring system, scores 0–6 are considered low risk and scores 7 and above are considered high risk. Generally, in most centres, the prognostic scoring is used more often than stage to determine the treatment: low-risk patients would be treated with single-agent chemotherapy whereas high-risk patients would be treated with multi-agent chemotherapy (see below). Management

• Patients treated for GTN should not become pregnant for approximately 6–12 months after treatment to allow accurate assessment of hCG levels.

• Fertility rates and pregnancy outcomes are similar in patients treated for GTN compared with the general population.

Complete hydatidiform mole

• The standard of care in the treatment of CHM is surgical evacuation of the uterus by suction curettage. This can be performed in any uterus regardless of its size. It is advisable that the suction be performed with a concurrent oxytocin drip running in, to reduce the risks of haemorrhage, perforation, and trophoblastic embolization. There is no need for a routine second curettage in the absence of persistent bleeding. In older patients (>40 years), total hysterectomy, with the mole in situ, may be an option.

Table 14.5.1 FIGO 2000 classification for GTN

• The use of prostaglandins for medical termination or precurettage cervical dilatation is not recommended, as it has been shown to increase the risk of trophoblastic embolization to the lungs and lung metastases (Tidy et al. 2000).

• In the rare event of a twin pregnancy with one viable fetus and a molar pregnancy, the mother needs to be counselled on the risks. These include early pre-eclampsia and pulmonary embolism. But, if the mother wishes, she may be allowed to carry on the pregnancy. The chance of delivering a viable baby is about 40%. There is no increase in the risk of malignant GTN following delivery (Sebire et al. 2002).

• After evacuation of the uterus, all patients should be followed up with weekly serum HCG till they become negative; then, monthly for 6 months. Subsequent hCG measurements are required under the following circumstances: irregular vaginal bleeding; amenorrhea; evidence of metastatic disease. In the serial monitoring of HCG, three patterns that raise suspicion of malignant GTN include: (1) persistent rise in the values; (2) plateauing of the values; and(3) a secondary rise after an initial fall.

• Reliable contraception should be advised after evacuation of the mole, mainly to avoid the confusion that a new pregnancy can cause in the serial monitoring of hCG. Although barrier contraception may be the safest, they are also the least reliable. Intrauterine devices or implantable hormone devices may cause irregular vaginal bleeding which might be mistaken for persistent GTN. Oral contraceptives are the most preferred. There seems to be differing opinion as to the timing of starting the pills. In the UK, it is recommended that the contraceptive pills be started after biochemical remission (i.e. hCG becomes negative). However, in the USA and many other centres (including Singapore), the pills are prescribed soon after evacuation.

• When the patient does get pregnant after a previous molar pregnancy, it is recommended that an early transvaginal ultrasound scan is done to ensure a normal viable pregnancy. Following one molar pregnancy, the risk of a repeat molar pregnancy in a subsequent conception is 1% (Rice et al. 1989). It increases to about 15% after 2 molar pregnancies (Bagshawe et al. 1986).

Partial hydatidiform mole

• The management of PHM is the same as for CHM.

Medical complications of hydatidiform mole

• Anaemia

• Toxaemia

• Hyperthyroidism

• Hyperemesis gravidarum

• Cardiac failure

• Pulmonary insufficient (rare).

Invasive mole

• A diagnosis of IM is seldom made clinically without surgery. Most of the non-metastatic (low risk) GTN following a molar pregnancy is believed to be IM. The presence of a vascular nodule in the myometrium, following evacuation of a mole and persistent high levels or rising levels of hCG may be suggestive of an IM.

• Chemotherapy, usually with a single agent, methotrexate (MTX) or actinomycin D (ACT-D) is the treatment of choice.

• In chemoresistant cases, resection of the nodule or hysterectomy may be indicated.

Choriocarcinoma

• CC develops in about 3–5% of the patients with CHM. Of all CC, 50% are preceded by a hydatidiform mole, 25% by an abortion, and the other 25% by a full term pregnancy. A clinical diagnosis of CC is rare. Most of the metastatic (high risk) GTN, especially those outside the lungs, are believed to be CC.

• Chemotherapy with multiple agents (see below) is the treatment of choice.

• Surgery has an important role in selected cases (see below). Further management

A diagnosis of persistent GTN is made when any of the following criteria are fulfilled:

There is a rise of HCG on three consecutive weekly measurements.

There is plateauing of the HCG levels for 3 or more weeks.

HCG levels remain elevated for 6 months or more.

There is radiological evidence of metastatic disease in the presence of positive hCG.

There is histological diagnosis of CC.

The disease is then assessed using the FIGO classification to determine the stage and risk score. It is highly recommended that patients with persistent GTN be referred to tertiary referral centres experienced in the management of GTN, so that optimal results may be obtained.

Low-risk patients (score up to 6)

• Chemotherapy with single agent is the treatment of choice in this group. Various protocols are used (see box) but they all give equally good results with reported remission rates reaching 100% in most centres.

• The role of surgery will be discussed below

Chemotherapy protocols for low risk GTN

1. Methotrexate 100 mg/m2 i.v. over 30 minutes followed by methotrexate 200 mg/m2 i.v. over 12 hours. Further therapy withheld as long as serum hCG regression pattern is satisfactory

2. Methotrexate 0.4 mg/kg i.v. or i.m. every day ? 5 days; repeat every 12–14 days (7–9-day window)

3. Methotrexate 1 mg/kg i.m. on days 1, 3, 5, 7; folinic acid 0.1 mg/kg i.m on days 2, 4, 6, 8; repeat every 15–18 days (7–10-day window)

4. Methotrexate 40 mg/m2 weekly

5. Actinomycin D 10–13 μg/kg i.v. every day ? 5 days; repeat every 12–14 days (7–9-day window)

6. Actinomycin D 1.25 mg/m2 i.v. every 14 days Cycles are repeated until hCG becomes negative and a further two or three cycles are given

High-risk patients (score of 7 or more)

• In this group, the mainstay of treatment is with multiagent chemotherapy. Again various regimens are used (see box), but there is no strong evidence to determine the best combination chemotherapy (Xue et al. 2006). Familiarity with a particular regime seems to be the most important factor in the choice of a regime. The most common regime is the EMA/CO (etoposide, methotrexate/folinic acid rescue, dactinomycin/cyclophosphamide and vincristine). However, in many centres, including the author’s, the regime of choice is EMA (without CO). Remission rate of about 86% have been reported (Newlands 2003).

• Salvage chemotherapy with novel combination of presently used drugs and use of newer drugs such as paclitaxel and gemcitabine may be required in patients with multiple metastases, especially those with liver and brain metastases and not responding to the regular combination therapy.

• Surgery has been utilized mainly in patients with chemoresistant tumour foci (see below).

• Fertility rates following chemotherapy is not reduced compared to patients who have not had chemotherapy. Outcomes of pregnancies are comparable to normal population. It is conventional to advise patients to avoid getting pregnant for at least 6–12 months after chemotherapy. Surgery in GTN

• The traditional role of surgery in GTN has been in a salvage setting, where despite multiple cycles of chemotherapy, resistant tumour foci exist. However, primary chemotherapy is not without danger in certain sites, especially the brain, where chemotherapy can cause necrosis and haemorrhage of the tumour, which can have serious or even fatal consequence.

• There has been a recent trend to do primary surgery, in selected patients, to avoid prolonged chemotherapy or its complications. Examples of such surgery would include hysterectomy in low risk, non-metastatic disease; thoracotomy to remove large, solitary lung metastasis; and craniotomy in solitary brain metastasis at accessible sites (Ilancheran et al. 1980).

Chemotherapy protocols in high risk GTN

EMA/CO

Week 1, Day 1

Actinomycin D 0.5 mg i.v. bolus

Etoposide 100 mg/m2 i.v. in 500 mL N saline over 30 minutes

Methotrexate 300 mg/m2 i.v. in 1 L N saline over 12 hours

Week 1, Day 2

Actinomycin D 0.5 mg i.v. bolus

Etoposide 100 mg/m2 i.v. in 500 mL N saline over 30 minutes

Folinic acid 15 mg oral/i.m 12 hourly ? 4 doses, starting 24 hours after commencing methotrexate

Week 2, Day 1

Vincristine (Oncovin) 1.4 mg/m2 i.v. bolus (maximum 2 mg)

Cyclophosphamide 600 mg/m2 i.v. in 500 mL N saline over 30 minutes

EP/EMA (regime for patients with resistance to EMA/CO)

Etoposide and cisplatin alternating weekly with methotrexate, actinomycin D and etoposide

Week 1, Day 1 (EP)

Etoposide 150 mg/m2 i.v. in 500 mL N saline over 30 minutes

Cisplatin 25 mg/m2 i.v. over 4 hours

Cisplatin 25 mg/m2 i.v. over 4 hours

Cisplatin 25 mg/m2 i.v. over 4 hours

Week 2, Day 1 (EMA)

Etoposide 100 mg/m2 i.v. over 30 minutes

Methotrexate 300 mg/m2 i.v. over 24 hours

Actinomycin D 0.5 mg i.v. bolus

Week 2, Day 2

Folinic acid 15 mg po 12 hourly ? 4 doses to start 24 hours after starting methotrexate

• Surgery may also be used when complications occur; for example, hysterectomy for a perforating invasive mole or choriocarcinoma with uncontrolled bleeding. Craniotomy may be necessary to stem cerebral haemorrhage from metastases.

• A notable exception to avoid surgery is vaginal metastasis. Typically, vaginal metastasis in GTN occurs in the anterior vaginal wall, just below the urethra, as a haemorrhagic nodule (Fig. 14.5.1). Any attempt to biopsy or excise the nodule can result in a torrential haemorrhage, as the nodule is often the ‘tip of the iceberg’ of a large vascular mass. These nodule are extremely chemosensitive and chemotherapy should be first choice; otherwise embolisation of the feeder vessels should be considered. Radiation therapy in GTN

With the availability of effective chemotherapy, radiation always had a limited role in the treatment of GTN. It has been employed most frequently to treat patients with brain or liver metastases, in an effort to minimize haemorrhagic complications from disease at these sites (Soper 2003).

Fig. 14.5.1 Vaginal metastasis in GTN. Placental site trophopblastic tumour (PSTT)

• PSTT is among the rarest of GTN, accounting for about 1-2% of GTN. It is a very unique form of GTN, arising from the intermediate trophoblast of the placenta. It can occur after any pregnancy including molar pregnancy, abortion, term delivery, or ectopic pregnancy. It displays a wide clinical spectrum of behaviour, from benign to an extremely aggressive, metastatic tumour, often unresponsive to chemotherapy.

• Unlike other GTN, it is characterized by a low level of hCG. However, human placental lactogen (HPL) levels are raised in the serum and the hormone is expressed in immunostaining of the histological sections. Final diagnosis is only on histology of the curetting or hysterectomy specimen.

• Fortunately, most cases are confined to the uterus and most (85–90%) behave in a benign fashion. The most common presenting symptom is vaginal bleeding. When diagnosis is confirmed on histology, meticulous effort must be taken to identify metastatic disease. The WHO prognostic scoring is of little use in PSTT, as its behaviour is unpredictable.

• Unless child bearing is very important, current recommendation for uterus-confined PSTT is a hysterectomy with perioperative single agent chemotherapy. For metastatic disease, aggressive chemotherapy with multi-agent chemotherapy regimes like EP/EMA is preferred. Further reading

Bagshawe KD, Dent J, Webb J. Hydatidiform mole in England and Wales, 1973-1983. Lancet 1986; 673–7.

Ilancheran A, Ratnam SS. The role of surgery in gestational trophoblastic disease. Int Journal Gynecol Obstet 1980;18:237–9.

Kohorn, E. the new FIGO 2000 staging and risk factor scoring systems for gestational trophoblastic disease: description and critical assessment. Int J Gynecol Cancer 2001;11:73–7.

Newlands ES. The management of recurrent and drug-resistant gestational trophoblastic neoplasia (GTN). Best Practice Res Clin Obstet Gynaecol 2003;17:905–23.

Rice LW, Lage JM, Berkowitz RS, et al. Repetitive complete and partial hydatidiform moles. Obstet Gynecol 1989;74:217–9.

Royal College of Obstetricians and Gynaecologists. The management of gestational trophoblastic neoplasia. Clinical Green Top Guidelines 38. 2004.

Sebire NJ, Foskett M, Paradinas FJ, et al. Outcome of twin pregnancies with complete hydatidiform mole and healthy co-twin. Lancet 2002: 359; 2165–6.

Soper JT. Role of surgery and radiation therapy in GTD. Best Practice Res Obstet Gynaecol 2003;17:943–55.

Tidy J, Gillespie AM, Bright N, et al. Gestational trophoblastic disease: a study of mode of evacuation and subsequent need for chemotherapy. Gynecol Oncol 2003;78:9–12.

Xue Y, Zhang J, Wu TX, et al. Combination chemotherapy for high risk gestational trophoblastic tumour. Cochrane Database Syst Rev 2006; 3: CD005196.

Ovarian and fallopian tube cancer

The incidence of ovarian cancer is low, being the fourth most common cancer in women in the UK following breast, bowel, and lung cancer. For each individual woman, the lifetime risk of ovarian cancer is estimated at 1 in 70 (1.5%). The poor 5-year survival rate in the UK of about 25% is because approximately 70% of women are diagnosed with stage 3 or stage 4 disease. Nevertheless, around 90% of patients who are found to have stage 1a and 1b ovarian cancer will have a 5-year survival.

The rates of ovarian cancer vary considerably between countries, with the highest rates in Scandinavia, North America, and the UK and are lowest in Africa, India, and the Far East. Over 6000 new cases of ovarian cancer are diagnosed each year in the UK and there are about 4500 deaths from the disease.

About 5–10% of cases are hereditary and when child-bearing is complete, a hysterectomy and bilateral salpingooophorectomy should be considered. Although the oral contraceptive pill decreases the risk of ovarian cancer by up to 50% in women in the general population, improved diagnosis and an ageing population are likely to result in an increasing incidence of reported cases of ovarian cancer.

Over 85% of ovarian cancers are of epithelial origin. About 10% are undifferentiated and the remainder are non-epithelial cancers and are rare. Pathology

• Ovarian tumours are classified on the basis of their cell or tissue of origin (Decruze et al. 2006).

• Neoplasms derived from coelomic epithelium (epithelial cancers)

• serous tumour

• mucinous tumour

• endometroid tumour

• mesonephroid (clear cell) tumour

• Brenner tumour undifferentiated carcinoma

• carcinosarcoma and mixed mesodermal tumour

• serous cystadenomas account for approximately 50% of epithelial ovarian cancers.

• Neoplasms derived from germ cells

• dysgerminoma

• endodermal sinus tumour (yolk sac tumours)

• embryonal carcinoma

• polyembryonal

• choriocarcinoma

• teratoma (immature, mature or monodermal)

• mixed forms (tumours composed of the above in any possible combination)

• tumours composed of germ cell and sex cord stromal derivative (gonadoblastoma)

• Neoplasms derived from specialist ovarian stroma

• granulosa-theca cell tumours (granulosa tumour and thecoma)

• Sertoli–Leydig tumours (arrhenoblastoma and Sertoli tumour)

• gynandroblastoma

• lipid cell tumours

• Neoplasms derived from non-specific mesenchyme

• fibroma, haemangioma, leiomyoma

• lymphoma

• sarcoma

• Metastatic ovarian tumours

• gastrointestinal tract (Krukenberg)

• breast

• endometrium

• lymphoma.

The non-epithelial cancers have an incidence of 6 per million women per year. They account for around 10–15% of all ovarian cancers and little is known about risk factors, although they do not appear to be similar to those found in patients with epithelial cancers. The malignant germ cell tumours are most common in teenagers and have been associated with the maternal use of hormones in early pregnancy. The granulosa cell tumour, which tends to occur in the older woman, is the most common malignant sex cord stromal tumour. Aetiology

• The strongest risk factor for epithelial ovarian cancer is related to reproductive factors such as the number of lifetime ovulatory cycles. High parity, use of the combined oral contraceptive, late menarche, early menopause, and early age at first pregnancy all reduce the risk.

• 5% to 10% of all epithelial cancers are hereditary. There are three clinical genetic manifestations.

• ‘site specific’ ovarian-cancer

• hereditary breast-ovarian cancer syndromes

Both these are associated with mutations in the BRCA1 and BRCA2 tumour suppressor genes, which are present in about 90% of hereditary cases.

• hereditary non-polyposis colorectal cancer (HNPCC and Lynch II).

Women with Lynch syndrome have a 40–60% of lifetime risk of endometrial cancer and a 10–12% lifetime risk of ovarian cancer. HNPCC is associated with alterations in the DNA mismatch repair genes.

Ovarian cancers caused by BRCA mutations tend to occur in younger women. The lifetime risk of ovarian cancer is 40–50% and 20–30% for a BRCA1 and BRCA2 mutation respectively. Cancers associated with these mutations are usually poorly differentiated serous carcinomas.

• Epithelial ovarian cancers have been associated with asbestos exposure, the use of talc, endometriosis, and pelvic inflammatory disease. Screening

Women at average risk

• An effective strategy for the early detection of ovarian cancer would significantly decrease mortality from this disease.

• To date, no screening strategies have been proven to decrease mortality from ovarian cancer and therefore screening in women at low risk of developing ovarian cancer cannot be recommended.

• CA125 and transvaginal ultrasonography to detect ovarian cancer have been used in large population-based studies as screening tools.

• Half of women with early-stage ovarian cancer will have a CA125 level within normal limits.

• Currently there is emphasis on educating women to be aware of the symptoms of ovarian cancer, such as bloating, increasing abdominal girth, a change in bowel or bladder habit, and abdominal or pelvic discomfort.

Women with a strong family history or a proven gene mutation

• Annual screening for CA125 and transvaginal ultrasonography can be offered.

• The limitations and implications of these tests should be fully explained.

• Women with BRCA mutations should be offered prophylactic salpingo-oophorectomy once their family is complete (Griffiths et al. 2005). Those who carry the HNPCC mutation should consider prophylactic hysterectomy along with salpingo-oophorectomy. Staging

Ovarian cancer spreads by three primary methods: direct extension to adjacent organs, exfoliation of tumour cells causing intraperitoneal dissemination, and lymphatic/vascular embolization. The omentum is the most frequent site of metastases.

The FIGO staging is based on surgical and pathological findings (Table 14.6.1). Accurate staging is essential to determine prognosis and management strategy. Prognosis

Although staging is mainly surgical the final stage is dependent on the pathology of the surgical specimens and the cytology of peritoneal washings or ascites (Table 14.6.2). Clinical approach

Diagnosis

Ovarian cancer generally presents with an insidious onset of vague non-specific symptoms. Many patients are referred to specialists other than a gynaecologist and clinicians need to be alert to the possibility that vague abdominal symptoms may be due to ovarian cancer.

• Main presenting symptoms in patients with primary ovarian cancer are (in order of frequency):

• abdominal swelling

• abdominal pain/pelvic pressure

• gastrointestinal complaints

• vaginal bleeding

• dysuria

• fatigue

• dyspnoea

• back pain.

Approximately 10% of women with early ovarian cancer may be asymptomatic.

• Pelvic examination will reveal a mass in 40–70% of patients and 20–30% will have clinically detectable ascites.

• Ascites, pleural effusions, enlarged inguinal, or supraclavicular lymph nodes and skin metastases suggest advanced disease.

• Where clinical examination is negative then the most valuable investigation in a woman presenting with nonspecific symptoms is ultrasonography. If a pelvic mass is confirmed then a CA125 should be checked.

• A risk of malignancy index (RMI) is an effective discriminator between cancer and benign lesions (Jacob et al. 1990). The RMI is calculated using the menopausal status (M), ultrasound findings (U), and the serum CA125. The ultrasound findings are scored with 1 point for each of the following criteria.

Table 14.6.1 FIGO staging of ovarian cancer

Table 14.6.2 FIGO Annual report on the results of treatment (2003)

• multilocular cyst

• evidence of solid areas

• evidence of metastases

• presence of ascites

• bilateral lesions

RM = U ? M ? CA125

where U = 0 for ultrasound score 0, U = 1 for ultrasound score of 1, U = 3 for ultrasound score of 2–5, M = 1 if premenopausal, and M = 3 if postmenopausal. When an MRI cut-off level of 200 is used the sensitivity is 85% and the specificity is 97%.

• A chest X-ray and CT scan of the abdomen and pelvis will be helpful in assessing the extent of the disease and aid with the preoperative counselling of the patient and the planning of her surgery. MRI has limitations in detecting upper abdominal disease but is better than CT for the work-up of an indeterminate adnexal/ovarian mass.

Tumour markers

CA125 is a tumour-associated antigen and levels will be increased in up to 80% of women with epithelial tumours. However, increased levels can also be seen in many non-malignant conditions such as menstruation, endometriosis, pelvic inflammatory disease, acute pancreatitis, diverticulitis, and congestive cardiac failure. Other malignant tumours may also give rise to a raised CA125, including endometrial, pancreatic, breast, and lung.

Inhibin levels can be a useful marker for follow-up in patients with granulosa cell tumours, and elevated levels are also seen in postmenopausal women with mucinous tumours.

• Where investigations suggest ovarian cancer, women should be referred to a centre with a gynaecological oncology multidisciplinary team. Evidence suggests that treatment in cancer centres is associated with an improved prognosis. Women with early-stage disease may not be identified by the RMI and therefore not referred to a cancer centre.

Management

Surgery is the cornerstone for the management of patients with presumed ovarian cancer. It is diagnostic, in early-stage disease curative and most importantly allows the cancer to be staged accurately. For the majority of patients with ovarian cancer, optimizing surgery is the best way to improve survival.

Surgery

• Exploratory laparotomy is a standard approach for patients with ovarian cancer with total abdominal hysterectomy, bilateral salpingo-oophorectomy, and tumour reductive surgery.

• Optimal tumour debulking resulting in no residual tumour nodules larger than 1 cm is known to be associated with improved survival, and therefore the aim of surgery should be optimal tumour debulking wherever possible.

• Although there have been few surgical randomized controlled trials in ovarian cancer, currently only patients who are unable to tolerate surgery or who have cancer that is known preoperatively to preclude optimum tumour debulking are considered for neoadjuvant chemotherapy.

• Where neoadjuvant chemotherapy is administered, interval debulking or interval cytoreductive surgery is usually considered after three cycles, although the impact on survival is unclear.

• It is not routine to perform second-look surgery after chemotherapy outwith clinical trials as it has not been shown to be of any benefit to patients.

• Secondary tumour debulking surgery may be considered for patients with residual disease detected on second-look surgery or where an isolated late recurrence has been detected. It should not be recommended in patients who will not receive postoperative chemotherapy.

• Palliative surgery is used to alleviate symptoms and 10–40% of women with terminal ovarian cancer develop bowel obstruction. Surgery produces a rapid resolution of symptoms, although this will not be achieved in about 20% of patients. The median survival following surgery for bowel obstruction is about 2 months. The merit of surgery requires to be individualized.

Surgical technique

• In most cases a vertical incision should be used. Ascitic fluid should be drained or peritoneal washings taken for cytology. Any encapsulated tumour masses should be removed intact and all peritoneal and intestinal surfaces, liver, spleen, kidneys, lesser sac, omentum, diaphragms, and pelvic/paraortic lymph nodes examined/palpated.

• Total abdominal hysterectomy, bilateral salpingooophorectomy, and omentectomy should be performed and biopsies taken from any suspicious areas. An appendicectomy should be performed for mucinous tumours.

• In a patient who is medically sufficiently fit, optimal debulking (1 cm tumour diameter or less) should be attempted as this reduction in tumour load is associated with better survival by improving the effect of adjuvant chemotherapy (Bristow et al. 2006). Current evidence suggests that best practice for advanced ovarian cancer is cytoreductive surgery (Pomel et al. 2008). A recent study suggests that delay before initiating chemotherapy after major surgery is not an important factor (Aletti et al. 2007).

• In the UK, demonstrable benefit has been seen where a patient receives her initial surgery by a gynaecological oncologist.

• Lymphadenectomy is not routinely performed in the UK as part of maximal surgical debulking in patients who have advanced ovarian cancer, although there is some evidence that it may be associated with an improved survival. Lymphadenectomy is helpful in patients who are thought to have early stage disease.

• If a suspicious cyst is managed laparoscopically, rupture should be avoided and aspiration performed in a closed bag prior to removal of the cyst through the abdominal wall port. Laparoscopic staging is feasible but should be reserved for early stage disease because of concerns regarding port site metastases.

Management of ovarian cancer in young women

• For young nulliparous women with a stage Ia tumour, more conservative procedures should be considered.

• A unilateral oophorectomy combined with surgical staging would normally be considered optimal primary treatment for borderline or germ cell tumours. A key issue for stage Ia epithelial tumours is the histological type. Mucinous and endometroid lesions have a better prognosis than serous lesions. The tumour should be well differentiated.

• The optimal requirements for conservative management include the following: a desire to maintain fertility, a normal pelvis which is free of adhesions, no invasion of the capsule, negative peritoneal washings, adequate evaluation of the contralateral ovary, an infracolic omentectomy, and selected lympadenectomy. Close follow-up is required with excision of the residual ovary after completion of childbearing.

Chemotherapy

• For optimally staged grade I, stage Ia, and stage 1b epithelial ovarian tumours the prognosis is excellent and chemotherapy is not usually recommended. However information from the International Collaborative Ovarian Neoplasm (ICON 1) (Colombo et al. 2003) and Adjuvant Chemotherapy in Ovarian Neoplasm (ACTION) trials confirms the benefit in terms of disease-free interval and 5-year survival of chemotherapy over observation in the population studied (Trimbos et al. 2003). The implications of these studies are unclear but patients with medium or high risk stage I ovarian cancer should be considered for chemotherapy.

• Large trials have all now confirmed that both progression-free survival and overall survival are improved in patients receiving platinum/paclitaxel chemotherapy (Piccart et al. 2000). Although survival advantages were not confirmed in the ICON 3 study, standard chemotherapy for advanced ovarian cancer in the UK is now a combination of platinum and paclitaxel.

• Chemotherapy is tailored to the individual and there remains a role for single agent carboplatin for frail or elderly patients.

• Carboplatin causes less nephrotoxicity and neurotoxicity than cisplatinum but nausea, allergic hypersensitivity reactions, gastrointestinal. and haematological side-effects can be troublesome.

• Paclitaxel commonly causes alopecia, nausea, diarrhoea, mucositis, hypersensitivity reactions, and neurological side-effects.

• There has recently been renewed interest in intraperitoneal chemotherapy and new data (GOG 172 trial) confirms that intraperitoneal chemotherapy can provide improved outcomes for patients with small volume or no visible residual disease and an intact peritoneal cavity following surgery. In this trial cisplatinum and paclitaxel were administered intraperitoneally (Armstrong et al. 2006). However, toxicity has been a significant problem with only 42% of women being able to complete all planned cycles of intraperitoneal therapy mainly because of toxicity. More studies are required to determine the place of this treatment.

Chemotherapy for relapsed disease

• Decisions about further treatment are usually made on clinical or radiological findings. A raised CA125 is seen in about 85% of cases and may predate clinical evidence of relapse by an average of 4 months.

• The choice of chemotherapy at relapse is usually based on the length of time the women has been off treatment (Ali et al. 2007).

• Women with platinum-sensitive tumours (tumours relapsing 12 months or more after initial platinum therapy) can be treated again with platinum-based chemotherapy. The ICON 4 trial using carboplatin + paclitaxel showed a 7% improvement in survival in the combined therapy group (Palmer et al. 2003). This improvement needs to be balanced against the increased toxicity of combination treatment and take into account the side-effects experienced after first line therapy.

• Women with platinum-resistant ovarian cancer (those relapsing within 6 months of completing treatment) or platinum-refractory disease (those who have progressive disease on platinum therapy) may be considered for salvage chemotherapy, which would include liposomal doxorubicin, topotecan, gemcitabine, and oral etoposide.

Other adjuvant therapies

• Radiotherapy is not currently recommended as adjuvant treatment but may have a palliative role in patients with recurrent disease.

• In view of the significant toxicity associated with chemotherapy, hormonal agents may be appropriate; 10–15% of ovarian tumours demonstrate a response. The most frequently used hormonal agents are gonadotrophin-releasing hormone agonists, tamoxifen, anti-androgens, and aromatase inhibitors such as letrozole.

Borderline malignant epithelial ovarian tumours

• Account for approximately 15% of all epithelial ovarian cancers and 95% are of the serous or mucinous type.

• The characteristics histological findings are stratification of the epithelial lining of the papillae, formation of microscopic papillary projections, pleomorphism, cellular atypia, mitotic activity, and no stromal invasion.

• Approximately 80% of women present with disease confined to the ovaries and less than 15% have advanced disease.

• There is a 10-year survival rate of approximately 95% for stage 1 disease, although recurrences may occur up to 20 years after initial treatment.

• Extra ovarian implants associated with stage III serous borderline tumours may be either invasive or non-invasive. The invasive implants are associated with a poor clinical outcome compared with the non-invasive group.

• The aim of surgery should be to completely remove the tumour. For most patients a total abdominal hysterectomy and bilateral salpingo-oophorectomy with omen-tectomy should be performed; however, if fertility is desired, a unilateral oophorectomy alone (provided the contralateral ovary appears healthy) does not appear to have a detrimental effect on survival.

• Women with poor prognosis borderline tumours may be prescribed chemotherapy as an adjuvant to surgery.

Non-epithelial ovarian cancers: germ cell tumours

• These tumours account for 10% of all ovarian tumours, the most common being a dysgerminoma (48%) followed by endodermal sinus (yolk sac) tumour (20%). They occur mainly in young women and a conservative approach to surgery is appropriate and radiotherapy avoided.

• Approximately 20% of patients with germ cell tumours treated with surgery alone will recur and therefore all patients except those with stage Ia dysgerminoma and stage Ia grade 1 immature teratoma should be treated with adjuvant chemotherapy.

• Bleomycin, etoposide, and cisplatin have been shown to result in a 96% sustained response rate.

• Ninety per cent of patients relapse within 2 years of initial treatment.

• Measurement of the tumour markers (hCG, AFP, and LDH) are useful as part of the preoperative evaluation and follow up monitoring.

Sex cord stromal tumours of the ovary

These constitute about 7% of ovarian malignancies.

Granulosa cell tumours

• These account for 70% of sex cord stromal tumours.

• Most tumours produce oestrogen and a few are androgenic. Serum inhibin may be a useful tumour marker.

• There are two forms: juvenile and adult. The former presents with precocious bleeding and the latter with postmenopausal bleeding. Endometrial hyperplasia or adenocarcinoma may occur as a consequence of the oestrogen production.

• The majority are diagnosed as stage I tumours and treated by hysterectomy, bilateral salpingo-oophorectomy, omentectomy, and cytoreductive surgery. In adolescent girls fertility sparing surgery should be the aim.

• Spread is similar to epithelial ovarian cancer and the stage of diagnosis seems to be the most important prognostic factor.

• The overall 5-year survival rate is around 80% however these tumours are characterised by late recurrences.

• For stage II or III tumours adjuvant chemotherapy is recommended and traditionally the combination of bleomycin, etoposide, and cisplatin has been used. Other combinations include vincristine, actinomycin, and cyclophosphamide or vinblastine, bleomycin, and cisplatin.

Sertoli–Leydig tumours

• These tumours are rare accounting for less than 0.5% of all ovarian tumours.

• The tumours may secrete oestrogen and androgen. Clinical presentation will depend on the hormones secreted, and in young women menstrual disorders are the most common symptom.

• Since over 95% of these tumours are confined to one ovary, unilateral salpingo-oophorectomy may be performed where preservation of fertility is required. In all other cases, total abdominal hysterectomy, bilateral salpingo-oophorectomy and a staging laparotomy is the appropriate management.

• Adjuvant chemotherapy (as for granulose cell tumours) is appropriate where there is a substantial risk of recurrence, although there is no evidence base to support this approach because of the rarity of the tumour.

Other uncommon ovarian tumours

• Sarcoma of the ovary (mixed Müllerian tumours (MMTs) and pure sarcomas) have a poor prognosis. After standard surgery, paclitaxel and carboplatin instead of ifosfamide and cisplatin have been used.

• Small cell carcinoma of the ovary is an aggressive tumour and following surgery the literature suggests that etoposide and cisplatin, or vincristine, actinomycin, and cyclophosphamide are the most useful combinations.

Fallopian Tube Cancer

• This is a rare malignancy.

• It is likely to have a similar aetiology and molecular pathogenesis to ovarian cancer.

• Women undergoing prophylactic bilateral oophorectomy should also have a bilateral salpingectomy performed.

• The diagnosis is rarely made preoperatively. Abnormal vaginal bleeding is the most common symptom. The classic symptoms of pain, which is often colicky, and a watery vaginal discharge are less common.

• Abnormal glandular cells on cervical smear may be seen in up to 25% of cases.

• Eight per cent will have been identified as having a pelvic mass and investigation and treatment is similar to that for ovarian cancer.

• The FIGO staging is similar to that for ovarian carcinoma using a surgical pathological staging system.

• The diagnostic pathological criteria for diagnosing a primary fallopian tube carcinoma include the following: The tumour should arise from the endosalpinx.

The histology should resemble the epithelium of the tubal mucosa.

Transition from benign to malignant epithelium should be present.

The endometrium and ovaries should be normal or contain a tumour but by a cystological appearance, small size and distribution appear to be metastatic.

• The 5-year survival for patients with fallopian cancer is slightly higher than for ovarian cancer, perhaps reflecting the higher proportion with early stage disease. The rate of lymph node metastases is higher for fallopian tube than ovarian cancer and this will adversely affect survival. Further reading

Aletti GD, Long HJ, Podratz KC, Cliby WA. Is time to chemotherapy a determinant of prognosis in advanced-stage ovarian cancer. Gynecol Oncol 2007;104:21–6.

Ali SN, Ledermann JA. Current practice and new developments in ovarian cancer chemotherapy. Obstet Gynaecol 2009;9:265–9.

Armstrong DK, Bundy B, Wenzel L, et al. Gynecologic Oncology Group. Intraperitoneal Cisplatinum and paclitaxel in ovarian cancer. Engl J Med 2006;354:34–43.

Bristow RE, Berek JS. Surgery for Ovarian cancer: how to improve survival. Lancet 2006;367:1558–60.

Colombo N, Guthri D, Chiari S, et al. International collaborative ovarian neoplasm trial 1: a randomised trial of adjuvant chemotherapy in women with early-stage ovarian cancer. J Natl Cancer Inst 2003;95:125–32.

Decruze SB, Kirwan JM. Ovarian cancer. Curr Obstet Gynaecol 2006;16:161–7.

FIGO (International Federation of Gynecology and Obstetrics). Annual report on the results of treatment in gynaecological cancer. Int J Gynecol Obstet 2003 83 (Suppl 1): IX–XXII, 1–229.

Griffiths SE, Lopes T, Edmondson RJ. The role of prolphylactic salpingo-oophorectomy in women who carry mutations of the BRCA genes. Obstet Gynaecol 2005;7:23–7.

Jacobs I, Oram D, Fairbanks J, et al. A risk of malignancy index incorporating CA125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer. Br J Obstet Gynaecol 1990;97:922–9.

Parmar MK, Ledermann JA, Colombo N, et al.; Icon and AGO Collaborators. Placlitaxel plus platinum-based chemotherapy versus conventional platinum-based chemotherapy in women with relapsed ovarian cancer: The ICON four–AGO–OVAR–2.2 Trial. Lancet 2003;361:2099–106.

Piccart MJ, Bertelsen K, James K, et al. Randomised intergroup trial of cisplatin-paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer; three year results. J Natl Cancer Inst 2000;92:699–708.

Pomel C, Barton DPJ, McNeish I, Shepherd J. A statement for extensive primary cytoreductive surgery in advanced ovarian cancer. Br J Obstet Gynaecol 2008;115:808–10.

Trimbos JB, Parmar M, Vergote I, et al. International Collaborative Ovarian Neoplasm Trial 1 and Adjuvant Chemotherapy in Ovarian Neoplasm Trial. Two parallel randomised phase iii trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J Natl Cancer Inst 2003;95:105–12.

Palliative care Definition

The World Health Organization defines palliative care as improving the quality of life of patients and families who face life-threatening illness, by providing pain and symptom relief and spiritual and psychosocial support, from diagnosis through to the end of life and in to bereavement.

Palliative care

• provides relief from pain and other distressing symptoms

• affirms life and regards dying as a normal process

• intends neither to hasten or postpone death

• integrates the psychological and spiritual aspects of patient care

• offers a support system to help patients live as actively as possible until death

• offers a support system to help family cope during the patients illness and in bereavement

• uses a team approach

• aims to enhance quality of life, and may also positively influence the course of the illness

• may be valuable early in the course of the illness, in conjunction with other therapies that are intended to prolong life. These include chemotherapy or radiation therapy, and those investigations needed to better understand and manage distressing clinical complications

• puts an emphasis on open and honest communication

• respects autonomy and choice.

Knowledge of palliative care principles is encouraged as part of the practice of all health professionals caring for patients with life-limiting illness.

In the UK, the addition of gynaecological cancer nurse specialists to cancer multidisciplinary teams has greatly enhanced the care of women during diagnosis and treatment. These nurses are able to play a key role in exchange of information between hospital and community teams as well as liaising with specialist palliative care services.

For patients with complex or unresolved symptoms, or those approaching the end of life, involvement of specialist palliative care services, where available, is recommended. Epidemiology

In the UK, about 7000 of the 17 000 women diagnosed with gynaecological cancer per annum will die from their disease. The incidence of common gynaecological cancers differs in developing and Western countries and is changing in both. The incidence of cervical cancer is falling whereas ovarian cancer is increasing. In the developed world, ovarian cancer deaths now exceed those of cervical and endometrial cancer combined. At diagnosis, two-thirds of ovarian tumours are at an advanced stage because most women with the disease are asymptomatic for a long time. Clinical approach

Breaking bad news

Bad news can be defined as information that alters a patient’s view of their future for the worse. Breaking bad news is not easy and can be daunting and demanding, but it is a skill that can be learnt, and it is an essential part of treatment and care. The following steps can assist clinicians in their approach to discussing difficult news.

• Find time and a private place.

• Try to have a relative or nurse with you.

• Know the facts before the meeting.

• Find out what the patient knows, e.g. ‘What have you been told?’

• Find out what the patient wants to know, e.g. ‘Are you the sort of person who likes to know everything about your illness?’

• Give a warning shot that bad news is coming, e.g. ‘I’m afraid the test results were not very good’.

• Impart the news slowly, and as simply as possible, checking understanding and gauging response along the way and be prepared to stop if necessary.

• Avoid saying ‘There is nothing more that can be done’. Don’t lie, but offer realistic hope, e.g. controlling symptoms, treating pain, etc.

• Listen to concerns, e.g. ‘What are your main concerns at the moment?’

• Encourage venting of feelings.

• Summarize concerns and give an opportunity for questions.

• Leave patients with a plan.and offer availability. Most patients need further explanation and support.

Principles of pain control in cancer

History and examination

Full assessment of the cause of the pain is integral to successful pain relief. Patients often have more than one pain, and different pains have different causes. Pain perception can be affected by psychological and emotional factors and pain can remain difficult to control if these factors are not identified and addressed.

Investigation

In the palliative care setting, investigations should be targeted at finding the cause of pain in order to plan appropriate treatment. For example, an isotope bone scan to look for bony metastases.

Unnecessary and inappropriate investigations that do not change management should be avoided.

Management

Cancer pain management may consist of drug and non-drug measures. With proper assessment and a logical stepwise approach to analgesic prescribing, at least 80% of chronic cancer pain can be controlled. Where pain is not responding to treatment, specialist advice should be sought. Consideration must also be given to treating pain by means of surgery, radiotherapy, and chemotherapy where possible and appropriate (Forbes 1998).

Common cancer-related pains

• Visceral/soft tissue

• opioid sensitive so use ‘analgesic ladder’ (see below).

• Bone pain

• sensitive to non steroidal anti-inflammatory drugs (NSAIDs)

• partly opioid sensitive

• radiotherapy may be helpful

• consider intravenous bisphosphonates.

• Nerve related

• partly opioid sensitive

• adjuvant analgesics may also be needed (see below)

• Non-drug measures such as transcutaneous electrical nerve stimulation (TENS) machine or nerve block may need to be considered in conjunction with drug therapy.

Cancer unrelated pain

1 Treatment related, e.g. constipation, radiotherapy.

2 Coincident illness or condition, e.g. diabetes, arthritis.

Principles of cancer pain control

• ‘By the clock’: cancer pain is continuous and therefore requires regular analgesia at appropriate dose intervals.

• ‘By mouth’: use oral route unless the patient is unable to swallow or there are concerns about absorption.

• ‘By the analgesic ladder’

Analgesic ladder

• Step 1: non-opioid analgesic, e.g. paracetamol

• Step 2: weak opioid plus non-opioid, e.g. dihydrocodeine + paracetamol

• Step 3: strong opioid plus non-opioid, e.g. morphine + paracetamol

Adjuvant analgesics

Drugs primarily used for indications other than pain, but which can provide pain relief in certain situations. They can be used in combination with drugs at all steps of the analgesic ladder.

Common adjuvant analgesics for cancer pain

1 Non steroidal anti-inflammatory drug: bone pain, pleuritic pain, hepatomegaly

2 Corticosteroid: raised intracranial pressure, hepatomegaly, and nerve compression

3 Antidepressants and anticonvulsants: neuropathic pain due to nerve damage or compression, and tenesmus.

4 Bisphosphonates: bone pain (Gralow et al. 2007).

Use of strong opioids

• Gain pain control: use immediate release morphine, starting dose 5–10 mg 4 hourly (i.e. six doses daily) and titrate against pain with a 30–50% increase in dose every 2 or 3 days if pain uncontrolled. Reassess pain control each day if possible.

1 Maintenance: change to 12 hourly modified release morphine, e.g. MST, by dividing total 24-hour morphine dose by 2. Breakthrough dose of immediate release morphine should be made available (one-sixth of 24-hour dose of morphine).

2 Other strong opioids may be considered if the patient develops unacceptable side effects on morphine. For example, fentanyl causes less opioid induced constipation than morphine. Seek specialist advice from palliative care team if pain is unrelieved or patient has unacceptable side-effects from opioids.

Common side-effects of opioids (strong and weak)

• Constipation: must be anticipated and prevented in all patients on weak and strong opioids. A regular stimulant laxative must be commenced at the same time (see constipation section)

• Sedation: may occur with the first few doses but then usually lessens.

• Nausea: is a common problem during the first few days of treatment. If it occurs, haloperidol, domperidone, cyclizine or metoclopramide are all useful.

Principles for control of nausea and vomiting

History and examination

A logical approach to the use of antiemetics depends on knowledge of the cause of the symptom. The patient should be assessed with regard to the cause and whenever possible this cause should be addressed. Potential causes for nausea and vomiting include drugs, renal failure, deranged liver function, hypercalcaemia, ascites, bowel obstruction, constipation, brain metastases, sepsis, and humoral tumour effects.

Investigation

This may include full blood count, serum urea, and electrolytes, including corrected calcium, plain abdominal film, abdominal ultrasound scan, and head CT scan.

Management

Consider the following.

• Stop or change drugs which are causing nausea.

• Avoid drugs with anticholinergic effects if gastric stasis is present (hyoscine, antidepressants, cyclizine).

• Antiemetics may be necessary for the first few days when opioid therapy is initiated.

• Correct abnormal biochemistry, particularly malignant hypercalcaemia.

Principles of nausea and vomiting control

• Ensure chosen antiemetic is used regularly and to a maximum dose before changing.

• If first drug is ineffective, change to a drug from another pharmacological group.

• If first drug is partially effective, add a drug from another pharmacological group.

• Cyclizine and other anticholinergic drugs antagonize some of the effects of metoclopramide and domperidone. The combination should be avoided if possible.

• If symptoms persist for more than 24–48 hours in spite of the oral route, change to a non-oral route.

Target antiemetic to probable cause of nausea and vomiting.

• Drug induced and biochemical: metoclopramide, haloperidol, cyclizine or levomepromazine

• Squashed stomach due to ascites or hepatomegaly, and delayed gastric emptying: metoclopramide or domperidone

• Stimulation of gastrointestinal receptors e.g. local tumour, oropharyngeal candida, bowel obstruction: cyclizine, levomepromazine

• Central causes, e.g. raised intracranial pressure, vestibular or motion related: cyclizine

• Chemotherapy induced: ondansetron, granisetron.

• Note that the place of 5HT3 antagonists in non-chemotherapy induced nausea and vomiting is not yet clear. They can cause severe constipation and should not be used first or second line.

Medical management of malignant intestinal obstruction (Mercadante et al. 2007).

• About a quarter of women with advanced ovarian cancer develop bowel obstruction and medical or surgical palliative treatment can be used.

• Clinical features include abdominal pain and distension, vomiting, colic, constipation, and loud bowel sounds.

• Surgery should always be considered but may not be indicated or desired.

• It may be possible to keep a patient tolerably symptom controlled (although some vomiting may still occur) with medication given via a syringe driver, without a nasogastric tube or intravenous infusion.

• Main principles of management are to control nausea, colic, and other abdominal pain.

• Generally, when complete intestinal obstruction occurs, prokinetic agents such as metoclopramide and stimulant laxatives are to be avoided.

• In subacute obstruction, particularly if it is functional, a combination of metoclopramide and dexamethasone may be effective in restoring function.

• Patients may be able to tolerate small amounts of food and fluid if nausea is well controlled.

• If large volume vomits persist after a trial of hyoscine butylbromide (buscopan), a proton pump inhibitor can reduce gastric secretions. The somatostatin analogue octreotide may also be helpful. Seek palliative care advice.

• If thirst is severe, intravenous or subcutaneous fluids can be used, although the patient may be comfortable with good oral care and ice to suck.

• Nasogastric tube or drainage gastrostomy may be considered, in discussion with the patient, to reduce symptoms if all else fails.

Drugs in syringe driver

• Abdominal pain: morphine or diamorphine (starting dose 5–10 mg/24 hours or 50% of total oral 24-hour dose).

• Nausea: haloperidol (2.5–5 mg/24 hours) ±cyclizine (150 mg/24 hours) or levomepromazine alone (starting dose 6.25 mg/24 hours)

• Colic and to reduce secretions: hyoscine butylbromide (60–120 mg/24 hours)

• Second line to reduce secretions: octreotide (range 300–600 μg/24 hours)

Do not mix cyclizine and hyoscine butylbromide in a driver as they may precipitate.

Constipation

History and examination

Constipation (difficulty in defaecation) is common in patients with advanced cancer. Immobility, reduced oral intake, constipating drugs, and raised serum calcium are all contributory factors. Constipation should be anticipated in all patients taking opioids, 5HT3 receptor antagonists, or anticholinergic drugs (e.g. cyclizine, tricyclic antidepressants). Chronic constipation can cause anorexia, vomiting, colic, tenesmus, spurious diarrhoea, urinary retention, and mental confusion.

Abdominal palpation may reveal hard faecal material in the descending and sigmoid colon. Examination should include a rectal examination to guide management. In addition to determining the presence or absence of faeces and its consistency, it will also reveal a palpable tumour mass which may be exacerbating the problem.

Investigation

Serum urea and electrolytes will confirm any dehydration, hypokalaemia due to diuretics, or hypercalcaemia. Plain abdominal film should not be used to diagnose constipation, but may confirm dilated loops of bowel in bowel obstruction.

Management

Constipation is a common cause of distress and is better anticipated and avoided where possible. Laxatives are the main treatment of constipation in palliative care patients. They should be started as soon as strong or weak opioids are prescribed. The laxative dose should be increased if the dose of opioid is increased (Twycross et al. 1991).

A good first choice is the combination of a stimulant laxative with a softening agent, e.g. senna and docusate sodium or a combination laxative such as codanthromer.

Danthron (in codanthromer) stains urine red and should be avoided in the incontinent patient as it can cause perianal skin irritation. Its use is restricted to constipation in terminally ill patients.

Movicol is an osmotic laxative which can be effective.

In general, lactulose alone is not effective for opioid-induced constipation and should be avoided in patients with inadequate fluid intake. Lactulose can cause flatulence and abdominal cramps.

Twenty-five per cent of patients on laxatives may still need rectal measures at times. High fluid intake, fruit, and fruit juice (especially prune juice) can all help.

For patients with existing constipation, the following principles can be helpful:

• Is rectum full?

• Hard faeces: lubricate using glycerine (glycerol) suppositories or soften with an arachis oil enema

• When clear commence stimulant laxative with faecal softener

• Soft faeces: start stimulant laxative

• If no success: trial of Movicol.

• Is rectum empty?

• Exclude obstruction or high constipation

• Is colon full?

• If colic is present: start faecal softener

• If colic is absent: start stimulant laxative ± softener

It is not normally necessary to use strong stimulant laxatives such as Picolax.

Fybogel is best avoided in cancer patients as many are unable to drink adequate volumes of fluid and it can lead to impaction.

Recurrent malignant ascites

History and examination

Ascites results from an imbalance between fluid influx and efflux in the peritoneal cavity.

Ascites can be asymptomatic when mild but distressing when severe. Clinical features include abdominal distension, abdominal discomfort or pain, early satiety, dyspepsia, acid reflux, nausea and vomiting, leg oedema, and breathlessness.

Ovarian cancer is the commonest primary tumour associated with ascites, present in 30% of patients at presentation and 60% in the terminal phase.

Investigation

Blood tests including urea and electrolytes, liver function tests, and INR are helpful prior to attempting an ascitic tap. Abnormal clotting may need correcting prior to drainage. Abdominal ultrasound and marking of drainage point may be advisable to confirm the presence of ascites, to avoid tumour masses, or if loculation of ascites is suspected. Ascitic fluid may be required to help with cancer diagnosis if this is the patient’s presenting symptom.

Management

If appropriate and successful, chemotherapy may control ascites. However, in most cases of malignancy-related ascites, the prognosis is poor and the principle in managing these patients should be one of minimal intervention. Unnecessary investigations should be avoided and drains should be left in for only a short time.

Paracentesis is appropriate for patients with a tense distended abdomen. The aim is to remove as much fluid as possible using a suitable catheter. Individual hospitals will usually have their own protocols but the following principles apply.

• If there is substantial ascites in the form of a tense abdomen and fluid thrill, it is usually safe to proceed without diagnostic imaging. Ultrasound can be helpful to mark an area for drainage if paracentesis has been difficult in the past or if solid tumour is easily palpable. In some centres, the drain is inserted under ultrasound guidance.

• In patients with no cytological proof of malignancy, send all ascitic fluid to cytology for analysis as well as 20 mL to bacteriology for microscopy and culture

• Clamping of drains to reduce the drainage rate is usually unnecessary, particularly for volumes less than 5 L.

• There is no evidence that leaving the drain in for 24–48 hours or more is safer or more effective. Leave the catheter on free drainage for a maximum of 6 hours. and then remove (Stevenson J et al. 2002).

• The limited evidence available suggests that significant hypotension is not usually a problem, the administration of intravenous fluids is rarely needed, and intravenous albumen has no proven role.

• Diuretics: spironolactone antagonizes aldosterone and may slow down reaccumulation of ascites. The starting dose is 100–200 mg mane with a typical maintenance dose of 300 mg per day. Urea and electrolytes need to be monitored. If spironolactone alone is ineffective, frusemide 40 mg once daily can be added.

• Paracentesis may need to be repeated every 3–6 weeks if diuretics do not slow down reaccumulation.

• Peritoneovenous shunts are rarely used for malignant disease.

Breathlessness

History and examination

Cancer-related breathlessness can be caused by the cancer directly, general debility, treatment, or concurrent illness.

Potential causes include

• chest infection

• anaemia

• pleural effusion

• tense ascites

• pulmonary emboli

• lung metastases

• lymphangitis

• cancer cachexia

Investigation

Appropriate investigations to help make the diagnosis will include blood tests and a chest X-ray. Blood gases and ECG may be appropriate in the acutely unwell patient. Pulse oximetry will determine whether the patient is hypoxic.

Management

Management strategies should be targeted, where possible, at the probable cause of the breathlessness. Where this is not possible, general management principles can be applied.

• Reassurance and explanation.

• Reverse the reversible, e.g. drain pleural effusion or ascites unless the patient is in the terminal phase of their illness.

• Consider oxygen if the patient is hypoxic.

• Use of a fan, either fixed or hand held, to promote movement of air across the patients face can be helpful in the absence of hypoxia and avoids oxygen dependence.

• Occupational therapy and physiotherapy expertise can help the patient to manage their shortness of breath by adapting their environment and teaching coping strategies such as pacing and relaxation techniques.

• Pharmacological treatment includes:

• low-dose opioids, e.g. oramorph 2.5–5 mg 6 hourly initially and titrated according to effect, duration of effect, and adverse effects

• benzodiazepines, e.g. diazepam at a starting dose of 2 mg twice or three times daily. Use lowest effective dose as the drug can accumulate and cause memory loss and falls. Lorazepam is long acting but can be used sublingually for rapid onset in panic attacks at a dose of 0.5–1 mg.

Malignant fistulae

History and examination

The definition of a fistula is an abnormal communication between two hollow organs or between a hollow organ and the skin.

Advanced cervical cancer is the most common cause of a fistula among gynaecological malignancies. Fistulae in advanced cancer can develop as a result of postoperative infection, pelvic radiotherapy, or a combination of the two. A few are caused solely by tumour progression and necrosis. The fistula may be vesicovaginal, rectovaginal, or a combination. Patients will usually volunteer a history of pneumaturia or passage of faeces per vagina. In the case of a rectovaginal fistula, examination may reveal tumour or a palpable defect in the vaginal wall.

Investigation

Abdominal imaging such as a CT scan may be required to confirm tumour recurrence or in an attempt to identify the site of the fistula if further treatment is planned. Radiological advice will help to determine the most appropriate investigation.

Management

A multidisciplinary approach to decision-making and symptom management is recommended in these patients. This may include a gynaecological oncologist, palliative care specialist, continence nurse, and gynaecological nurse specialist.

• Patients with a poor prognosis and performance status may be best managed conservatively. Stoma may not be trouble free, and there may be psychological reasons not to want surgery. However, rectovaginal fistulae can be distressing to patients due to discomfort and odour.

• Conservative management will need to concentrate on:

• urine collection: usually with pads, or tampons inserted in to the vagina

• faecal collection: using pads

• skin protection: use of barrier cream to prevent excoriation and soreness

• odour control (see fungating wounds).

• Surgical options will include:

• urinary diversion: usually by formation of an ileal conduit

• bowel diversion: the simplest possible. Widespread intra-abdominal disease can make this difficult. Increasing expertise in colonic stenting may provide new, less invasive options for management of malignant fistulae

• fistula repair: if possible and appropriate.

Tumour fungation

History and examination

A fungating cancer is a primary or secondary tumour which has ulcerated the skin. The tumour may proliferate or cavitate and is often associated with pain, pruritis, exudate, malodour, bleeding, infection, and psychological distress to patient, carers, and family.

Investigation

Microbacterial swabs may be needed to rule out infection such as Staphylococcus aureus and pseudomonas as an additional cause of malodour and exudate. Serum platelet levels and clotting should be checked if heavy bleeding occurs.

Management

• If the tumour is sensitive to radiotherapy or chemotherapy, a significant reduction in tumour size and skin healing can be achieved (Forbes K 1998).

• Pain due to fungating wounds can be multifactorial. The general approach to management will be as outlined above using oral analgesia and drugs for neuropathic pain may need to be included. In addition, there is some evidence that topical morphine may reduce pain (Back et al. 1995; Krajnik et al. 1997). Nociceptive afferent nerve fibres contain peripheral opioid receptors that are silent except in the presence of local inflammation. Generally, morphine is applied as a 0.1% (1 mg/mL) gel in a water-based gel (e.g. Intrasite gel). The amount of gel varies according to the size and site of the tumour but is typically 5–10 mL applied twice to three times daily. It can be kept in place by a non-absorbable pad or dressing such as Opsite.

• Pruritis is probably caused by inflammatory agents such as prostaglandins and may respond to non-steroidal anti-inflammatory drugs.

• Exudate: specialist dressings can be chosen depending on the relative amounts of odour and infection.

• Malodour is particularly problematic. It is caused by a combination of tumour necrosis and deep anaerobic infection. Treatment includes oral or topical metronidazole, charcoal dressings, occlusive dressings such as Opsite or Granuflex, and debridement of devitalized tissue to remove the source of odour-forming bacteria using agents such as topical Manuka honey. The patient may need to be nursed in a single room. Odour in the room may be masked by aromatic oil burners. Another option is to place a cat litter tray under the bed.

• Surface bleeding may be controlled using physical measures such as application of adrenaline-soaked gauze (1 in 1000), silver nitrate sticks, alginate dressings, and diathermy. Oral haemostatic drugs such as tranexamic acid 1 g qds may also help. Radiotherapy and embolization may be needed to control more severe recurrent spontaneous bleeding.

The syringe driver

The syringe driver is a small, portable battery-operated infusion pump, used to give medication subcutaneously over 24 hours. It can be used when other routes (e.g. oral, buccal, rectal, or transdermal) are unsuitable. Local guidelines on use will often exist. They are often used to deliver drugs at the end of life but can be useful at other stages of illness.

Indications

• Inability to swallow

• Persistent nausea and vomiting

• Interference with oral absorption

• Unconscious patient.

Mixing drugs in a syringe driver

Principles:

• Try to avoid mixing more than three drugs in a syringe at one time.

• All of the following drugs can be mixed with morphine or diamorphine in a syringe driver.

• Antiemetic: Metoclopramide, haloperidol, cyclizine

• Sedative and antiemetic: Levomepromazine

• Sedative: Midazolam

• Colic: Hyoscine butylbromide (Buscopan)

• Terminal secretions: Hyoscine hydrobromide (crosses blood–brain barrier and sedating)

• Cyclizine and hyoscine butylbromide should not be mixed in the same syringe as they may precipitate.

• As needed (prn) doses of subcutaneous diamorphine or morphine should be prescribed for breakthrough pain, calculated to be a one-sixth of the total 24-hour dose.

• To covert from oral morphine to subcutaneous diamorphine, divide total 24-hour morphine dose by 3 to obtain total 24-hour diamorphine dose. This is based on 3 mg of oral morphine (any form) being equivalent to 1 mg of diamorphine injection.

• To convert from oral morphine to subcutaneous morphine, divide total 24-hour oral morphine dose by 2 to obtain total 24-hour subcutaneous morphine dose. This is based on 2 mg of oral morphine (any form) being equivalent to 1 mg of morphine injection.

Palliative care emergencies

Malignant hypercalcaemia

• Common in cervical cancer and is a poor prognostic sign.

• Can occur in the absence of bone metastases due to PTHrP (parathyroid hormone-related protein) produced by the tumour.

• Consider treating with intravenous fluids and intravenous bisphosphonates if the patient has a persistently raised serum corrected calcium level greater than 2.8 mmol/L and is symptomatic of hypercalcaemia, unless the patient is in the terminal phase of their illness.

• Renal function will dictate safe fluid volumes and correct dose of bisphosphonate.

Table 14.7.1 Drugs in a syringe driver for the terminal phase

Massive haemorrhage

• Frightening for patient, family, and staff

• If a patient is at risk of a large bleed, consider prescribing crisis medication of midazolam 2.5–5 mg subcutaneously and morphine 5–10 mg subcutaneously (or one-sixth of normal daily oral dose converted to the subcutaneous route) just in case.

• Red or green towels or blankets should be available to soak up blood.

• Patient may become semiconscious quickly

• A reassuring presence is the most important factor; therefore, stay with the patient.

• If a second nurse or doctor is present, ask them to administer the crisis drugs.

Spinal cord compression

Consider in patients that develop problems with urinary retention, faecal incontinence, and sensory changes or loss of motor function in the lower limbs.

• Requires immediate commencement of high dose steroids (dexamethasone 4 mg qds) followed by urgent MRI scan of spine within 24 hours. Discussion with oncology team for advice regarding urgent radiotherapy or neurosurgical opinion is advised.

The terminal phase

Optimal management of the terminal phase depends on recognition that the patient is dying. Consistent communication of this fact within the medical and nursing team to the patient’s family, and sometimes to the patient themselves is essential. Signs that death may be approaching include the patient becoming bed bound, only taking small amounts of food and fluid, becoming semicomatose, and no longer being able to take their tablets. This allows discontinuation of inappropriate drugs, investigations and observations, and prescription of appropriate drugs for symptom control (Ellershaw et al. 2003).

If the patient is unable to swallow, appropriate drugs for pain, nausea, chest secretions, and agitation can prescribed for prn (as required) administration in anticipation of symptoms, or mixed in a syringe driver if symptoms are present (Table 14.7.1).

Some centres use integrated care pathways, e.g. the Liverpool Care Pathway for the Dying Patient (LCP), as the final multiprofessional documentation. These pathways aim to standardize care for the dying patient by applying evidence-based practice and appropriate guidelines related to care of the dying. Further reading

Back IN, Finlay I. Analgesic effect of topical opioids on painful skin ulcers. J Pain Symptom Manage 1995; 10: 493.

Dickman A, Schneider J, Varga J. The syringe driver: continuous subcutaneous infusions in palliative care. Oxford: Oxford University Press: 2005.

Ellershaw J, Ward C. Care of the dying patient: the last hours or days of life. BMJ 2003;326:30–4.

Ellershaw J, Wilkinson S. Care of the dying: a pathway to excellence. Oxford: Oxford University Press 2003.

Forbes K. Complications of radiotherapy for gynaecological malignancy. Eur J Palliat Care 1998;5:175–8.

Gralow J, Tripathy D. Managing metastatic bone pain: the role of bisphosphonates. J Pain Symptom Manage 2007;33(4):462–72.

Hanks G, Cherny N, Christakis N, Fallon M (eds) Oxford textbook of palliative medicine. Oxford: Oxford University Press 2009.

Krajnik M, Zylicz Z. Topical morphine for cutaneous cancer pain. Palliat Med 1997; 11: 326.

Mercadante S, Casuccio A, Mangione S. Medical treatment for inoperable malignant bowel obstruction: a qualitative systemic review. J Pain Symptom Manage 2007;33:217–23.

Stevenson J, Gilbert J. The development of clinical guidelines on paracentesis for ascites related to malignancy. Palliat Med 2002;16:213–18.

Twycross R, Wilcox A. Symptom management in advanced cancer. Radcliffe Medical Press 2001.

Twycross RG, Harcourt JMV. The use of laxatives at a palliative care centre. Palliat Med 1991;13:159–60. Internet resources

Information for professionals on drugs usage in palliative care: www.palliativedrugs.com

Website for national end of life programme containing resources, information and examples of good practice: www.endoflifecare.nhs.uk.

Scottish Intercollegiate Guidelines Network (SIGN) guidelines on control of pain in adults with cancer: www.sign.ac.uk. Patient resources

Cancer Research UK—cancer statistics: info.cancerresearchuk.org/cancerstats/types

Macmillan Cancer Support website containing help, support and advice for those suffering from cancer and information on common cancers, treatments and side-effects: www.macmillan.org.uk

World Health Organization statements on palliative care: www.who.int/cancer/palliative/en/

Vulval cancer

Carcinoma of the vulva is a rare gynaecological cancer, accounting for 3% of gynaecological tumours, with approximately 1000 new cases registered each year in the UK, a rate of 3 per 100 000 of the population. It is the twentieth most common cancer in women and ranks as the nineteenth most common cause of cancer deaths in women. The disease commonly occurs in older women, with two-thirds of cases occurring over the age of 70 years. Approximately 15% of vulval cancers occur in women who are less than 40 years of age and are increasingly seen in younger women. Ninety per cent of cancers are squamous cell cancers and traditionally treatment has been by radical vulvectomy with groin node dissection. Over the past decade or so, the management of this condition has undergone considerable evolution with an emphasis on individualizing treatment for each case rather than a uniform surgical policy (Burke et al. 1995). These cancers are most effectively treated by a multidisciplinary team with expertise in specialized surgical and radiotherapeutic techniques.

Minimally invasive stage 1a tumours can be treated by wide local excision alone; however, all other tumours are likely to require inguinal node dissection or radiotherapy as well as definitive treatment of the vulval lesion.

Postoperative radiotherapy improves regional disease control and survival in patients with positive inguinal lymph nodes. Pelvic lymph node dissection is no longer performed. In cases of advanced vulval cancer or where there is involvement of the anus, rectovaginal septum, urethra, or vagina, complete responses have been reported using a regime of cisplatin and 5-fluorouracil (5FU) with external beam radiation to the vulva.

The physical and psychosexual impact of vulval cancer and its treatment should be managed in a proactive fashion with early involvement of nurse specialists and experts in this field (Allan 2003). Pathology

• About 95% of malignant vulval neoplasm are squamous cell carcinomas. These tumours can be divided into subtypes. Keratinizing tumours are most common. Basaloid and warty subtypes are associated with human papillomavirus (HPV) infection and occur in younger women.

• Verrucous carcinoma is a well-differentiated squamous carcinoma that is locally invasive but does not metastasize to regional lymph nodes. Radiotherapy can lead to anaplastic change and rapid invasion.

• Malignant melanomas account for approximately 5% of vulval neoplasms. The prognosis for malignant melanomas of the vulva depends on depth of invasion as described by Breslow.

• Basal cell cancers are treated by wide local excision as this tumour rarely involves the lymphatics.

• Adenocarcinoma of the Bartholin’s gland and sarcomas are extremely uncommon, each accounting for 1–2% of vulval neoplasms. Aetiology

• There are two causes of vulval cancer, one related to HPV infection usually types 16, 18, and 31 and occurring in younger women, and the other seen in elderly patients arising in a background of differentiated vulval intraepithelial neoplasia (VIN), often after years of lichen sclerosis.

• VIN: the lifetime risk of developing vulval cancer is approximately 4–7% if a woman has VIN. However, the relationship between squamous cell carcinoma and VIN is not as well defined as in cancer of the cervix, and recent studies suggest that VIN presenting in young women carries a greater risk of recurrence with a progression to invasive cancer of 10–15%. The incidence of VIN has been reported to have increased by up to threefold over the past two decades in women under 35 years of age.

• VIN3 can be subdivided into three types: warty, basaloid and differentiated. Basaloid and warty vulval cancer are associated with adjacent undifferentiated VIN and are often HPV positive. There is often a solitary HPV type, usually type 16.

• The more common keratinizing squamous cell carcinomas, seen in the older woman, is associated with adjacent differentiated VIN which is usually HPV negative and often associated with squamous hyperplasia and lichen sclerosis.

• Lichen sclerosis is associated with squamous cell carcinoma. About 30% of SCCs arise in women who have lichen sclerosis and the lifetime risk of developing cancer with lichen sclerosis is estimated at 3–5%. The chronic pruritus associated with this disease has been implicated in the aetiology of vulval cancer described as the ‘itch scratch cycle’.

• Smoking and immunosuppression also appear to be more common in younger patients who develop vulval neoplasm. These patients are prone to multifocal, multicentric intraepithelial disease with the associated risk of carcinoma.

• P53 mutations are often seen in the HPV-negative cancers in older women. Staging

In 1988 (updated in 2009) the FIGO (Table 14.8.1) staging for vulval cancer was changed to a surgical staging system based on the findings of the vulval excision and inguinal lymph node dissection, based on the tumour–node–metastases (TNM) classification. Prognostic factors

• The 5-year survival for patients is related to the FIGO stage (Table 14.8.2)

• Metastatic disease in the regional lymph nodes is the most important independent prognostic factor influencing survival. Most publications have reported higher rates of recurrence and death in patients whose tumours are deeply invasive are high grade, have a high mitotic rate and/or associated with lymphovascular space invasion (Homesley et al. 1993).

• The depth of invasion is the most important factor predicting lymph node involvement. When the depth of invasion is between 1 and 2 mm the incidence of positive inguinofemoral nodes is 8%, rising to 30% when the depth of invasion is between 3 and 5 mm.

• The status of the inguinofemoral and pelvic nodes are important in predicting survival. The 5-year survival for a patient with negative lymph nodes is 80–90%, falling to 40–50% where there are metastases to lymph nodes. Three or more unilateral positive nodes or two or more bilateral nodes have a 30% 5-year survival.

Table 14.8.1 FIGO staging for vulval carcinoma (2009)

• Both the size and the number of inguinofemoral lymph node metastases affect survival. Intracapsular metastases are associated with a better outcome than cases where there is extra capsular disease. Patients with fewer than three nodes involved have a low risk of pelvic node disease and a good prognosis.

Table 14.8.2 FIGO stages

• Patients with pelvic lymph node involvement are generally considered to have an extremely poor prognosis; however, these data were collected from patients who underwent pelvic lymph node dissections before the mid 1980s. This procedure is rarely performed nowadays and patients with inguinal metastases will have postoperative radiotherapy with fields that include the pelvic lymph nodes. As a consequence, the impact of current treatment on the prognosis of patients with pelvic lymph node metastases is unknown. Clinical approach

Diagnosis

Symptoms: key points

• The vast majority of patients are symptomatic, complaining of itch, vulval pain, burning, bleeding, or swelling.

• A delay in diagnosis is not uncommon because of embarrassment both for the patient as well as the primary care physician, which may postpone examination. Topical treatments may be prescribed before a thorough examination has taken place.

• An invasive cancer may be incidental findings at the time of biopsy of presumed VIN.

• Vulval warts are uncommon in postmenopausal women and there should be a low threshold for performing a biopsy.

• Multiple biopsies may be required in patients who are immunosuppressed or exhibit evidence of HPV infection.

• A possibility of multicentric HPV-related genital intraepithelial neoplasia dictates that a thorough examination of the lower genital tract including a cervical smear and evaluation of the perianal region is required.

Investigations: key points

• Biopsy is mandatory before planning management. Increasingly photographs are used as a diagnostic aid and are particularly useful when planning primary surgical treatment.

• ‘Incisional’ biopsy is preferable to ‘excisional’ biopsy prior to a multidisciplinary team referral. A limited biopsy allows the gynaecological oncologist to fully assess the vulval lesion.

• Punch biopsy under local anaesthetic in the clinic is usually possible. A 4–6 mm punch biopsy can be taken under local anaesthetic using a Keyes biopsy. Ideally, the tumour should be photographed. However, a wedge biopsy may be required in larger vulval tumours or where there is discrepancy between the clinical findings and pathology. A biopsy from the edge with some normal adjacent skin is required to facilitate the diagnosis.

• Pelvic examination remains the most important tool for evaluation of the local extent of vulval tumours. The location, gross morphology, sites of involvement, and dimensions of the tumour should be carefully recorded. The proximity of the tumour to midline structures should also be noted.

• Patients with extensive disease may require examination under anaesthesia to assess the extent of involvement of structures such as the urethra and anal sphincter. Ideally these examinations should be performed by the multi-disciplinary team including a gynaecological oncologist, general/urological/plastic surgeons, and radiation oncologist.

• CT and MRI can be helpful in detecting pelvic or inguinal lymphadenopathy in patients with extensive disease.

• Suspicious groin nodes should be sampled using fine needle aspiration or a similar technique, in the outpatient clinic.

Management

General: key points

• As many of the patients with a vulval cancer will be elderly with coexisting medical conditions, preoperative investigations and assessment by an anaesthetist will be required. In patients who are fit for anaesthesia, radical surgery is treatment of choice.

• The preoperative assessment should also include the likely need for occupational therapy and social services to ensure that there is a prompt supported discharge plan.

• As the treatment of vulval cancer has moved towards a more individualized, multidisciplinary approach, surgery should be undertaken in a cancer centre where there are experienced gynaecological oncology surgeons working in conjunction with other specialties.

• The treatment of vulval cancer is surgical and is carried out with curative intent. There is good evidence that patients who have inguinal metastases from vulval cancer frequently can be cured if their regional disease is addressed with appropriate treatment (radiotherapy). The vulva is richly supplied with lymphatics, and regional spread of vulval cancers is common.

• The benefit of radiotherapy in achieving local regional control and improved survival has been well demonstrated although the role of chemoradiotherapy has not yet been fully evaluated.

• There should be psychological support for patients undergoing treatment for vulval cancer at an early stage (Green 2000).

Local disease: key points

• Stage Ia tumours: many series have demonstrated that the risk of lymph metastases is less than 1% for patients with tumours that invade no more than 1 mm. These are the only cases that can safely be treated without addressing the inguinofemoral lymph nodes. Wide local excision should be performed with a 1 cm margin of normal tissue. Most gynaecologists will use a 1.5-cm margin on the fresh surgical specimen. If the surgical specimen reveals inadequate excision or deeper invasion than anticipated, then re-excision and treatment of the inguinofemoral lymph nodes is required.

• Stage Ib–II tumours: the traditional radical vulvectomy plus en bloc bilateral groin node dissection is rarely undertaken. Nowadays, the aim of the primary surgery is to obtain a wide excision that extends to the perineal fascia with a 2-cm margin of uninvolved tissue. In most cases, this does not require resection of the entire vulva, although it is extremely important that the primary cancer is excised with adequate margins as this directly affects the risk of local recurrence. One study suggested that the margin of clearance of tumour is the best predictor of local recurrence. All of the recurrences on the vulva occurred where the surgically free margin was less than 8 mm. Primary closure of the wound may not be possible and some form of reconstructive procedure may be required.

• When cancers are close to the urethra or anus, it may be impossible to achieve adequate tumour free margins without compromising organ function. In these circumstances collaboration with urology and colorectal surgery will be required. Chemoradiotherapy should be considered either to reduce tumour volume to allow sphincter-sparing surgery or to replace surgery completely.

• All stage Ib and stage II tumours will require treatment to the inguinofemoral lymph nodes because of the risk of regional metastases.

• Stage III–IVa tumours: radical vulvectomy with bilateral groin node dissection should be performed if it is anticipated that the tumour can be removed without compromising urethral or anal function and with negative margins.

• Fixed groin nodes should be biopsied and groin and pelvic radiation considered rather than surgery. Phase II trials of concurrent 5-fluorouracil with or without cisplatin with radiotherapy resulted in complete response rates of 53–89% for advanced disease.

• When these tumours involve the urethra, bladder, anus, rectum, or vagina, although sometimes curable with ultraradical surgical procedures, many institutions are moving towards an organ sparing approach involving radiotherapy, chemoradiotherapy with or without surgery. Multidisciplinary cooperation is required so that the best results from this approach can be achieved (Blake 2003).

• Incomplete resection of a vulva tumour should be avoided as the dose of radiotherapy required to sterilise microscopically positive surgical margins is similar to that required to control gross disease and will add morbidity without clearly improving the likelihood of local control. Dramatic responses have been seen with chemoradiotherapy, although there have been no phase III trials of chemoradiotherapy in women with vulval cancer. The results for chemoradiotherapy treatment in patients with advanced cervical and anal cancers are encouraging and are likely to be similar for vulval disease.

Vulval reconstruction: key points

• The support of the plastic surgeons can improve the cosmetic and functional result following excision of a vulval tumour. There are a wide range of surgical options which will reduce infection and breakdown particularly in the perineal area.

• There is an increasing trend towards greater use of simple rotational flaps and transposition for small and medium sized defects (e.g. lotus petal) (Yii et al. 1996). Where the defect is more extensive it may be necessary to use pedicled flaps from the thigh or abdominal wall.

Regional disease: key points

• It is essential to address the risk of spread to the inguinofemoral lymph nodes at the time of initial management of the vulval cancer. Fewer than 20% of patients survive 5 years after an inguinal recurrence of previously treated vulval cancer.

• Treatment to the regional lymph nodes applies to all vulval cancers apart from stage Ia.

• The overall incidence of positive groin nodes in early disease is about 15%. The incidence of lymph involvement increases with the depth of invasion of the tumour.

• The groin node dissection consists of removing the nodes from the superficial compartment bounded by the inguinal ligament superiorly, the border of sartorius laterally and the border of adductor longus medially. The medial deep femoral lymph nodes should also be removed, as superficial inguinal node dissection alone is associated with a higher risk of groin node recurrence. Preservation of the long saphenous vein has been advocated to reduce postoperative groin wound and subsequent lower limb problems.

• A unilateral groin node dissection is appropriate if the cancer is well lateralized, being more than 1 cm from the midline. In all other situations a bilateral dissection should be performed.

• Where a unilateral dissection has revealed positive nodes then the current practice would be to remove the contralateral nodes. The incidence of positive contralateral nodes with negative ipsilateral groin nodes where the vulval tumour is less than 2 cm in diameter, and not involving the midline, is quoted at between 0 and 0.4%.

• Where the lymph node dissection reveals inguinal metastases, postoperative radiotherapy should be considered. The standard recommendation is that postoperative radiotherapy should be given if two or more groin nodes are involved with microscopic disease or if there is evidence of extracapsular spread or complete replacement in any lymph node. Radiotherapy should be given to the lateral pelvis and inguinal regions to the side or sides with positive nodes (Thomas et al. 1989).

• Studies of patients who have positive inguinal lymph nodes after radical vulvectomy and lympadenectomy have shown that postoperative radiotherapy reduces the rate of groin recurrence and significantly improves the survival rate.

• More selective groin node surgery may be possible in the future through lymphatic mapping and sentinel node biopsy (Acheson 2007). The combination of injecting radiolabelled technetium and isosulfan blue dye looks promising for the future to reduce the morbidity from groin node dissection. Complications/side-effects of treatment

Surgery

• General: as this disease predominantly affects elderly women between the ages of 60 and 75 years, there may be coexisting medical problems increasing the morbidity and mortality of treatment. A perioperative mortality rate of 1–2% is quoted.

• thromboembolic disease

• infection

• psychological and psychosexual sequelae.

• Vulval surgery

• wound problems, infection and breakdown

• cystocele and rectocele

• narrowing of the introitus

• osteomyelitis

• sexual dysfunction.

• Inguinofemoral lymph dissection

• wound problems, infection and breakdown

• groin lymphocysts

• leg lymphoedema

• cellulitis

• nerve damage.

Radiotherapy

• Skin desquamation

• Skin necrosis.

Reducing morbidity: key points

• Management should involve a multidisciplinary team in a cancer centre.

• An individualized approach with separate groin incisions (triple incision approach) has replaced the classical radical surgical resection with lymphadenectomy.

• Preliminary studies indicate that sentinel nodes can be identified in most cases of vulval carcinoma which would avoid formal lymph node dissection for over 60% of patients with stage Ib/II tumours. This would help reduce the 15–20% incidence of chronic leg lymphoedema and groin wound infection and dehiscence which occurs in over 50% of cases.

• Advanced tumours and those involving bladder, urethra, rectum, or anus should be considered for chemoradiotherapy rather than ultraradical surgery with loss of organ function. Concurrent chemoradiation using cisplatin and 5FU has produced encouraging results, ‘downstaging’ tumours so that a surgical approach can be considered. Current chemoradio-therapeutic regimes are not well tolerated by elderly people.

Follow-up

• Although the role of routine follow up on survival has been questioned for gynaecological cancers, patients find follow up visits very reassuring and it is an opportunity to look for long-term complications of treatment and any psychosexual concerns.

• A typical follow up schedule is for all patients to be seen every 3 months for the first 2 years, every 6 months for 3 years and then discharge after 5 years.

Recurrent vulval cancer

• Management and prognosis depend on the site and extent of the recurrent disease (Piura 1993). Wide local excision of a localised recurrence provides a 56% five year survival rate when the inguinofemoral nodes are not involved. Recurrence in the regional nodes is associated a poorer prognosis and a 5-year survival of approximately 20%.

• A multidisciplinary approach involving clinical oncologists, gynaecologists and general/urological/plastic surgeons will result in the most appropriate management for this group of patients.

• Treatment options will include

• wide local excision

• radical vulvectomy with pelvic exenteration

• radiotherapy or chemoradiotherapy sometimes with further surgery. Suitable chemotherapy regimes are platinum and 5FU or cisplatin, methotrexate, and bleomycin.

Bartholin’s gland carcinoma

• This is a rare vulval cancer with a peak age incidence in the mid-60s.

• It has a tendency to spread into the ischiorectal fossa and to the inguinal lymph nodes.

• It may spread directly to the pelvic lymph nodes.

• Treatment is radical vulvectomy and inguinal lymphadenectomy. To achieve adequate margins the rectum and a considerable part of the vagina may need to be removed. Reconstructive surgery with support from the plastic surgeons will often be necessary.

Malignant melanoma

• This is the second most common invasive cancer in this area.

• Melanomas are usually pigmented and raised but approximately 25% may be amelanotic and resemble a squamous cell cancer.

• Prognosis is related to the size of the lesion and the depth of invasion. The Clark classification and Breslow thickness are of prognostic benefit and apply to the vulva.

• First-line treatment is radical wide local excision. Routine inguinal lymphadenectomy has little place in the management of this disease as it spreads primarily by blood born metastases. A margin of 1 cm is sufficient for lesions with a depth of less than 2 mm and a 2–3-cm margin for lesions with a depth of more than 2 mm.

• Prognosis is related to the size of the lesion and the depth of invasion, and in general it is poor because of the late stage of presentation. Further reading

Acheson N. New developments sentinel node mapping. Obstet Gynaecol 2007;9:270–5.

Allen J. The clinical nurse specialist in gynaecological oncology–the role in vulval cancer, best practice and research. Clin Obstet Gynaecol 2003;17:591–607.

Blake P. Radiotherapy and chemoradiotherapy for carcinoma of the vulva. Best Pract Res Clin Obstet Gynaecol 2003;17:649–61.

Breslow A. Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg 1970 172: 902.

Burke TW, Levenback C, Coleman RL, et al. Surgical therapy of T1 and T2 vulvar carcinoma: further experience with radical wide excision and selective inguinal lymphadenectomy Gynecol Oncol 1995;57:215–20.

Green MS, Naumann RW, Elliot M, et al. Sexual dysfunction following vulvectomy. Gynecol Oncol 2000;77:363–77.

Homesley HD, Bundy BN, Sealis A, et al. Prognostic factors for groin node metastasis in squamous cell carcinoma of the vulva (a Gynecologic Oncology Group Study). Gynecol Oncol 1993;49:279–83.

Thomas G, Dembo A, De Petrillo A, et al. Concurrent radiation and chemotherapy in vulvar cancer Gynecol Oncol 1989;34:263–7.

Piura B. Recurrent squamous cell carcinoma of the vulva: a study of 73 cases. Gynecol Oncol 1993;48:189–95.

Yell N. Flaps in vulval vaginal reconstruction. Br J Plastic Surg 1996;49:547–54.

Vulval intraepithelial neoplasia Introduction

The concept of ‘squamous cell carcinoma in situ’ was first described in 1922 by French authors Hudelo et al. (1922). Over the years, a bewildering variety of terms, such as leucoplakia, Bowen’s disease, Bowenoid papulosis, Bowenoid dysplasia, Bowenoid atypia, erythroplasia of Queyrat, carcinoma simplex, squamous cell carcinoma in situ, and hyperplastic dystrophy with severe atypia, have been used to describe this condition. The incidence of vulval intraepithelial neoplasia is increasing worldwide, primarily due to the increasing occurrence of this disease in young women, who account for 75% of cases (Judson et al. 2006).

In 1987 the International Society for the Study of Vulvovaginal Disease (ISSVD) and the International Society of Gynecological Pathologists recommended a unifying term: vulval intraepithelial neoplasia (VIN). By definition it consists of neoplastic cells confined to the boundaries of the squamous epithelium. It may regress either spontaneously or, after incomplete surgical removal, persist or progress to invasive carcinoma if untreated.

The ISSVD further recommended (ISSVD 1990) grading VIN into:

• VIN 1 mild dysplasia: only the lower third of the epithelium involved

• VIN 2 moderate dysplasia: lower half of the epithelium involved

• VIN 3 severe dysplasia/carcinoma in situ: lower two-thirds to full thickness involvement

This grading system was extrapolated from that applied to the non-keratinizing squamous epithelium of the cervix. However, the problems of extrapolating the grading of VIN from its cervical counterpart are that

• the vulval skin is different from its cervical counterpart in both health and disease

• cancer pathogenesis in the vulva is different from that of the cervix

• unlike the cervix, a VIN 1 to VIN 3 continuum has not been described.

• The pathological diagnosis of VIN 1 is poorly reproducible, uncommon, and generally represents reactive change or HPV effect.

Accordingly, the ISSVD in 2003 revised the terminology for squamous vulval intraepithelial neoplasia (Sideri et al. 2005), based on morphological manifestation, clinical characteristics, and HPV content.

VIN, usual type (VIN-u)

This is characterized by lack of cell maturation throughout the epithelium, nuclear aneuploidy, and abnormal mitotic figures and is further subdivided into

• VIN, warty type: the lesion comprises highly pleomorphic cells, keratinization, multinucleation, and verrucous surface pattern

• VIN, basaloid type: lesion comprises uniform, undifferentiated basal cell-type population

• VIN, mixed (warty/basaloid) type.

It is no longer necessary to subdivide morphologically HPV-related VIN-u into low-grade (VIN 1) and high-grade (VIN 2, 3) types. VIN 1 is a flat condyloma, and although nearly 50% contain high-risk (oncogenic) HPV types, only a minority contains HPV type 16 as opposed to high-grade VIN which carries HPV 16 in more than 90% of cases. VIN 1 is not considered to precede high-grade VIN but may only indicate risk for developing VIN-u and thus the term VIN 1 has been discarded.

VIN, differentiated type (VIN-d)

This is HPV unrelated, with nuclear atypia and involves only the lower third or sometimes even only the basal cell layer of the squamous epithelium. The degree of nuclear atypia is similar to repair-related atypia.

Note that the occasional example of VIN which cannot be classified into either of the above VIN categories (usual type and differentiated type) may be classified as VIN, unclassified type. The rare VIN of pagetoid type may be classified as such, or placed in this category.

Abandoning the three-grade system simplifies the terminology, which better addresses the clinical need to identify vulval cancer precursors, and will also enhance diagnostic reproducibility and allow better communication among gynaecologists, pathologists, and other healthcare providers dealing with vulval disease. Epidemiology and pathogenesis

VIN-u

• VIN-u is caused by oncogenic HPV types 16 and 18, and less frequently by other oncogenic types 33, 35, 51, 52, and 68.

• It tends to occur in younger women with a mean age of 30 years, and in most cases is associated with the same demographic, behavioural patterns and high-risk HPV-positive rates seen in women with high-grade CIN.

• There is also an associated history of smoking, multiple sex partners, and early onset of sexual activity.

• Immunosuppression and smoking may be greater risk factors for VIN than CIN.

• It is the precursor lesion of HPV-related invasive squamous cell carcinoma of the vulva, which is increasing in frequency among younger women.

• The most common symptoms of VIN-u are pruritus and a burning discomfort. VIN-u tends to be multifocal and multicentric with confluent involvement of the interlabial grooves, posterior fourchette, perineum, perianal skin, and cervical epithelium in up to two-thirds of women with VIN.

• Colposcopy is helpful in defining the limits of the lesion, in detecting foci of possible early invasion, and in the exclusion of high grade CIN with which VIN-u is often associated. The definitive diagnosis is based on biopsy of the vulva.

VIN-d

• VIN-d represents another high-grade lesion, but is of different morphology and behaviour. Comprising less than 5%, it is often recognized adjacent to an existing squamous cell carcinoma.

• It is not related to HPV and is thought to be the precursor lesion to the non-HPV-related invasive squamous cell carcinoma of the vulva observed more often in elderly women, most of whom are older than 70 years and lack the behavioural risk factors of those observed with VIN-u. It tends to be unifocal and unicentric.

• The pathogenic background may be related to chronic itch-scratch cycles that are associated with lichen simplex chronicus and lichen sclerosus (Scurry 1999). The release of unidentified carcinogens together with the local environment of chronically irritated and inflamed skin may play a major role in the development of intraepithelial neoplasia and ultimately invasive well-differentiated keratinizing SCC of the vulva in elderly people. It is believed that this may be mediated by p53 (Yang 2000). Women with symptomatic lichen simplex chronicus or lichen sclerosus who are treated with high potency topical steroids rarely, if ever, progress to invasive squamous cell carcinoma. However, about 80% of the invasive SCC in elderly women are associated with untreated long-standing symptomatic lichen simplex chronicus or lichen sclerosus. Natural history

VIN-u

VIN-u is now regarded as the precursor of non-keratinizing basaloid and warty SCC of the vulva skin. In a large series of 405 women with VIN 2–3 (Jones 2005), 63 women received no treatment. The lesion regressed in 47 patients and progressed to invasion before treatment in 10 patients. Seventeen women (3.8%) developed invasive vulva, perianal, or periurethral cancer more than 1 year after treatment for VIN. This is similar to a systematic review of 97 published papers involving 3322 patients by van Seters et al. (2005) in which they reported a 3.3% incidence of invasive cancer after treatment for VIN (van Seters et al. 2005). With a follow up of up to 40 years, 50% of the 405 women in Jones’s series needed at least one or more treatments for persistent or new VIN. In the same series, the surgical resection margin was significantly related to the risk of requiring retreatment for persistent or recurrent disease. Positive margins were found in 47% of the women treated with excision, and they had a 50% risk of requiring retreatment in the next 5 years. On the other hand, women with negative margins had only a 15% risk of retreatment. The HPV-related invasive vulval cancers tend to be diagnosed in their early stages, and survival rates are excellent.

VIN-d

VIN-d precedes well-differentiated keratinizing SCC of the vulva skin. SCC arising from HPV-unrelated VIN tends to present with deep invasion and advanced clinical stage including lymph node metastasis (40% stage III) and 5-year survival rates of between 25–50% in women with stage III–IV disease. The risk of progression to invasion seems greater in VIN-d than in VIN-u (ISSVD 1990).

Overall the difference in prognosis of the two varieties of vulval carcinomas is due at least in part to the fact that women with VIN-u are much younger and seek medical attention much earlier than their elderly counterparts with VIN-d. Differential diagnosis

Conditions that can mimic VIN include

• lichen sclerosus

• lichen planus

• lichen simplex chronicus

• flat or papular condylomata

• seborrhoeic keratosis

• basal cell carcinoma

• rarely, extramammary Paget’s disease of the vulva.

Classically, it presents as an eczematoid, crusty, erythematous area on the vulva skin. Pruritus is present in more than 70% of patients. Histologically, it is really an intraepithelial adenocarcinoma. Diagnosis is often missed as it is treated as an eczema. It is best established by biopsy, which shows the characteristic malignant cells within the epithelium. Treatment is by wide excision, although obtaining clear margins may be difficult. These patients should also be evaluated for the possibility of synchronous neoplasms (breast, rectum, bladder, urethra, cervix, and ovary).

In most cases and where there is any doubt, or if the lesion does not respond to therapy, diagnosis must be confirmed with a biopsy. Clinical evaluation

• There is immense variation in the clinical appearance of VIN based on colour, surface contour and topography. VIN-u, the premenopausal form, which is multifocal and multicentric is more commonly seen than VIN-d, which tends to be unicentric and occurring in the postmenopausal age group.

• About 50% of women are asymptomatic and are diagnosed either when a lesion is observed during a routine gynaecological examination or at colposcopy for abnormal cervical cytology.

• In patients with symptoms of chronic vulval irritation or pruritus, the outer genitalia must be carefully examined. The use of the colposcope to perform vulvoscopy is an important adjunct to inspection techniques. The value of performing a vulvoscopy includes:

• defining the extent of disease

• directing biopsies of the most severe areas

• excluding overt invasion

• directing laser treatment if necessary, through visualization of anatomical landmarks, thereby allowing depth of vaporization to be determined.

• The vulva must be carefully inspected for hyperkeratosis before the application of 5% acetic acid. This must be applied for a much longer period than the cervix as the vulva epithelium takes it up rather more slowly. A systematic examination must be made of the vaginal introitus, labia minora and majora, perineum, followed by the clitoris, urethral orifice, perianal area, and anal canal. During examination, the following characteristics should be documented:

• topography: uni- or multifocal

• surface contour: flat, raised, or micropapillary

• colour tone: white, red, brown, grey, or dark

• degree of acetowhite change

• blood vessels: absent, punctuate or mosaic.

• VIN lesions are characteristically papular and over half exhibit superficial parakeratosis, which is defined as the retention of nuclear chromatin material in the usually acellular keratin layer of the epithelium.

• The Collins toluidine blue test, in which a 1% aqueous solution of dye is applied to the external genital area, and then washed off with 1% acetic acid after 2 minutes, may be useful to demonstrate this abnormal maturation of the vulva skin. Although vulvoscopy aids in localizing precancerous change, any area that looks suspicious whether it be white, grey, red, dark raised, or flat must be biopsied or excised to assess the severity of change.

• Colposcopic examination of the cervix and vagina should also be carried out.

• Biopsy of the vulva can be easily achieved with local anaesthetic in the clinic. EMLA (Eutectic Mixtures of Local Anaesthetics) cream (lidocaine 2.5% and prilocaine 2.5%) can be applied to the area followed by local infiltration of 1% xylocaine, which will help to elevate the skin and facilitate the biopsy. A punch biopsy, a scalpel, or a proprietary Keys punch biopsy instrument may be utilized to obtain a satisfactory sample of vulva skin with 2–5 mm depth for accurate histological assessment. The small skin defect can be closed with a 3.0 suture or left to granulate after application of Monsel’s (ferric subsulphate) solution, or silver nitrate. Management of VIN

• It is important to correlate clinical symptoms and signs with histology prior to embarking on any form of therapy. Many patients are young and the emotional trauma of any procedure is considerable.

• Determining the management of patients with VIN continues to be challenging and inexperienced clinicians should refer the patient to colleagues with expertise in gynaecological dermatology.

• The rationale for treating any patient with VIN would be:

• to relieve symptoms,

• to avoid recurrences and the risk of invasive cancer.

• There has been a gradual trend towards individualization of treatment and conservative therapy, with preservation of normal anatomy and vulval function. Based on the new pathogenic–morphologic classification, broadly, the HPV-related VIN-u should be treated with ablative and/or excisional techniques, whereas the precursors or VIN-d would be best treated with potent corticosteroid preparations or topical pro-apoptotic, non-steroidal anti-inflammatory products such as 0.1% tacrolimus. Treatment options

Observation

This is appropriate for asymptomatic lesions and patients who will be compliant to long-term follow-up. It is essential to first rule out invasion through multiple or extended biopsies.

Surgery

• For localized lesions, CO2 laser vaporization and wide local excision are most suitable. Wide local excision is associated with recurrence rates of 12-30%, and margins of 8-10 mm are recommended. For CO2 laser vaporization, a recurrence rate of 5–40% has been described. This is as a result of treating to an insufficient depth or lateral extent. A margin of 15 mm has been found to result in lower recurrence rates than a 5 mm or less margin (Ferenczy et al. 1985). In terms of depth, laser vaporization should be carried to depths of 2–3 mm (Reid’s third surgical plane) (Reid 1992) for hair-bearing areas and 1 mm for non-hair-bearing skin.

• For very extensive multifocal lesions extending into hair-bearing areas of the vulva, laser treatment would not be applicable because the necessary depth of treatment would not produce a good cosmetic result especially in young women. In such cases, skinning vulvectomy with or without skin graft would be recommended. Simple vulvectomy in a young patient is mutilating, does not significantly reduce the recurrence rate and is therefore no longer recommended. However, there may be a place for this in older patients with extensive and symptomatic VIN, to relieve symptoms and to exclude occult invasion.

• Recurrence rates range from 12% to 50% irrespective of treatment modality. Major reasons for treatment failure include failure to include all satellite lesions in the treatment field and reactivation of latent HPV DNA in vulva skin adjacent to or at a distance to the lesional epithelium. Negative surgical margins are associated with a recurrence rate of 15% whereas margin positivity is associated with a 50% recurrence rate (Yang B 2000). There appears to be two distinct groups of patients who require retreatment for VIN:

• women diagnosed with VIN within 2 years of the original treatment who probably have persistent disease

• another group of women who are diagnosed at 4–5 years or more who probably have new areas of VIN.

• For invasive cancer, the time intervals are a median of 2.4 years for women who progressed from their original VIN, whereas patients who were apparently originally cured but developed a new focus of disease that progressed to invasive cancer were diagnosed a median of 13.5 years after their original treatment (Yang B 2000). Multifocal lesions have a higher recurrence rate of 40% compared with 16.7% in unifocal lesions most likely due to positive resection margins.

Medical treatment

• These are used mainly in those who decline surgery or wish to preserve the cosmetic appearance of the vulva. Invasive disease must be excluded prior to medical treatment. Many of these medications are off-label use or experimental in nature.

• Imiquimod 5% cream (Aldara) is a topical immune system modulator that induces local proinflammatory cytokine production and cell-mediated immunity. Randomized trials comparing Imiquimod to placebo for treatment of VIN have demonstrated complete response in 35–80% and partial response in 10–46% in the Imiquimod arms. A skin reaction and discomfort at the treatment site is common and this affects compliance. An escalating dose regimen starting with an application once a week for 2 weeks, then twice a week for 2 weeks, then, if tolerated well, three times a week is recommended. A typical treatment course lasts 16 weeks.

• Topical 5% 5-fluorouracil (Efudex) cream is no longer used in the treatment of VIN as the chemical desquamation results in painful erosions with delayed healing and eventual scarring of the vulva and introitus.

• Experimental treatments include photodynamic therapy, ultrasonic surgical aspiration, and cidofovir, a potent antiviral agent.

Role of HPV vaccines in preventing VIN

Since HPV 16 appears to be the dominant HPV type associated with high-grade VIN, followed by HPV 18, it was anticipated that the HPV vaccine, developed primarily to reduce the incidence of cervical cancer, would also reduce the risk of VIN. This indeed was shown in the trials using the quadrivalent vaccine. Data from three placebo controlled trials showed that the immunization was 100% effective against the development of high-grade VIN caused by HPV 16/18, over a follow up period of 3 years, in the ‘according-to-protocol’ population. Further reading

Ferenczy, et al. Latent papillomavirus and recurring genital warts. N Eng J Med 1985; 313: 784.

Hudelo M, Boulanger-Pilet M, Cailliau M. Erythrokeratodermie verruqueuse en nappes symetriques et progressives congentales. Bull Soc Fr Dermatol Syphiligr 1922; 29: 45.

ISSVD. Report of the Committee on Terminology of the International Society for the Study of Vulvar Diseases. New Nomenclature for Vulvar Disease J Reprod Med 1990; 35: 484.

Jones R. Vulvar Intraepithelial Neoplasia: Aspects of the Natural History and Outcome in 405 Women Obstet Gynecol 2005;106:1319–26.

Judson PL, Habermann EB, Baxter NN, et al. Trends in the incidence of invasive and in situ vulvar carcinoma. Obstet Gynecol 2006; 107: 1018.

Reid. Laser surgery in the lower genital tract. In: Coppleson M (ed.) Gynaecological oncology, vol 2, 2nd edn. Edinburgh: Churchill Livingstone 1992: 1094.

Scurry J. Does lichen sclerosus play a central role in the pathogenesis of human papillomavirus negative vulvar squamous cell carcinoma? The itch-scratch-lichen clerosus hypothesis. Int J Gynecol Cancer 1999;9:89–97.

Sideri M, Jones RW, Wilkinson EJ, et al. Squamous vulvar intraepithelial neoplasia 2004 modified terminology, ISSVD Vulvar Oncology Subcommittee. J Reprod Med 2005: 50: 807–10.

Van Seters, et al. Is the assumed natural history of vulvar intraepithelial neoplasia iii based on enough evidence? A systematic review of 3322 published patients. Gynecol Oncol 2005; 907: 645.

Yang B. Vulvar intraepithelial neoplasia of the simplex (differentiated) type a clinicopathologic study including analysis of HPV and p53 expression. Am J Surg Pathol 2000: 24.