Quality Control of PCR-Based Diagnostic Assays

19.6.1 IAC, or no IAC, that is the question...

It is clear that, given the challenging sample types that need to be tested, the potential for PCR inhibition is a major consideration in MAP PCR-based diagnostic tests to avoid false nega­tives.

The goal of an internal amplification con­trol (IAC) is to signal a potential false-negative result. Thus, delayed (higher Cq value) or lack of amplification of the IAC is an indicator of failure of the PCR. An internal extraction control is similar to an IAC but is added to the initial step of the extraction protocol and un­dergoes all processing and amplification steps. However, there is an assumption implicit with such controls, which is that both the PCR am­plification of the IAC and the target sequence are equally impacted by any PCR inhibitors present in the DNA extract. This assumption has been challenged with findings of differ­ential impact on the IAC and target-specific amplification (Schrader et al., 2012). This may depend on a number of factors, such as the primers used, the GC content of the target regions or competition between the reactions.

Furthermore, a repeat extraction of a sample with evidence of inhibition does not ensure that the issue will not be repeated, leading to challenges in a diagnostic setting with high sample throughout (Acharya et al., 2017c).

Alternative approaches include the dilution of the DNA extract, which can both identify and rectify the inhibition (Acharya et al., 2017c). This is an effective approach, however consider­ation must be given to possible loss of detection of low positive samples due to this effect of the dilution.

Any approach to circumvent potential inhi­bition and the risk of false negatives in a MAP PCR detection assay, be that an IAC, DNA extract dilution and/or amplification curve efficiency assessment, requires validation in the appropri­ate setting.

19.6.2 Validation of assays

For any molecular diagnostic assay, validation of the analytical and diagnostic performance is critical prior to the application of the test. This is described in detail in the World Organisation for Animal Health (OIE) manual (OIE, 2019). PCR assays to detect MAP are generally compared with MAP culture as the ‘gold standard' test to determine the diagnostic accuracy. Though cul­ture is considered the test for benchmarking, it is not perfect; rather it is the best option available. However, with increased analytical and diagnos­tic sensitivity, molecular-based tests can perform better than the gold standard, making determi­nation of diagnostic specificity difficult in the absence of an unexposed animal population.

Considerations in relation to diagnostic test validation have been reviewed in detail (Nielsen et al., 2011); briefly these include the purpose and target condition that the diagnostic test aims to identify (e.g. herd- vs individual-level diagnosis, infected vs infectious), validation sample population selection to avoid bias (e.g. in­cluding randomly selected animals rather than a targeted group at a particular disease stage) and the classification of disease/infection status in the sample population (case definition).

19.6.3 Confirming PCR identification

OIE now states it ‘would be good practice to' in­clude a subsequent sequencing step to confirm PCR positive results, especially when index cases from a previously unaffected location or low Cq positives are detected. This can be done by sequencing the PCR amplicon from a con­ventional PCR, which can be challenging for samples with low initial template quantities. Sequencing of IS 900 PCR products to confirm MAP identity has been performed in a num­ber of studies, to confirm the sequence iden­tity to MAP K10 (Lee et al., 2011; Garg et al., 2015).

19.6.4 International benchmarking

Standardization and harmonization of diagnos­tic testing (Ahmed et al., 2008, 2011) is gaining momentum at many levels, from local to inter­national, for a number of microbial pathogens (Mcgiven et al., 2006; Mengoli et al., 2009; White et al., 2010; Nielsen et al., 2011; Xing etal., 2011; Makela etal., 2012; He et al., 2012; Johnson et al., 2012; McNerney et al., 2014; Henaux et al., 2018). Meaningful comparisons between results, such that inter-laboratory re­sults can be more easily and reliably compared, are becoming increasingly important to sup­port our understanding of the epidemiology of disease, disease outbreaks and to underpin trade agreements between countries. Improved epidemiological knowledge based on accurate prevalence data, supported by consistent, stand­ardized tests, will also benefit policy decisions that support the aims and objectives of control programmes.

An earlier review of the scientific litera­ture on direct faecal PCR for MAP in cattle dem­onstrated very little if any standardization of conventional or qPCR (Marsh et al., 2014). The majority of these assays failed to meet the min­imum information for publication of quantita­tive real-time PCR experiments (MIQE) (Bustin et al., 2009) or standards for reporting of di­agnostic accuracy studies (STARD) (Bossuyt etal., 2003; Bossuyt and Reitsma, 2003; Cohen et al., 2016) guidelines, particularly in terms of the number of animals tested and both ana­lytical and diagnostic sensitivity and specificity data.

A similar international initiative would di­rectly support the level of precision and accura­cy required for control programmes and as new technologies such as whole genome sequencing become a desktop reality in laboratories. This could encompass medical research, to enable cross-validation of diagnostic protocols for MAP in Crohn's disease patients and animals with paratuberculosis through sharing of sample panels. It is critical now that new diagnostic as­says and tests meet guidelines such as OIE, MIQE and STARD. An initiative such as this has been made easier for the paratuberculosis research community through the efforts of a group of well-established paratuberculosis researchers who have published STARD-based guidelines specifically for this disease (Gardner et al., 2011).

19.7