Abstract
Molecular diagnostics can and should only be applied in patient care when an extensive package of quality criteria is met. This applies to both the technical quality (an assay which identifies the appropriate nucleic acid sequence) and the ability to discriminate between affected or not in a clinical setting. This chapter describes all criteria that have to be fulfilled to validate such molecular diagnostic assays if developed “in house” or have been established in a commercial setting and are available as CE/IVD-marked or FDA-cleared diagnostic kits. Many tools for trouble shooting during such validation steps are presented, as are the criteria to implement such tests by verification. The last part of the chapter focuses on the quality control in routine laboratory applications of molecular diagnostic assays by the first-, second- and third-line controls.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
The usually used distinction between health and disease is not frequently used in this chapter. On the contrary, “affected and unaffected” are discriminated. Various situations pertain to this, e.g. carrying the BRCA mutation means “affected” whereas the person might not be diseased yet. Another example is colonising potential pathogens not causing disease in healthy persons but positive in a molecular diagnostic assay.
- 2.
In-house methods and laboratory-developed tests (LDTs) are both used to describe non-commercial available molecular assays, mostly on the basis of PCR technology. The FDA defines a laboratory developed test (LDT) as an in vitro diagnostic test that is manufactured by and used within a single-laboratory LDTs which are also sometimes called “in-house” developed tests or “home brew” tests.
- 3.
Instead of technical validation, the term analytical validation is used. If a commercial kit is used, one speaks of technical verification whilst the manufacturer has performed the technical validation.
- 4.
Instead of diagnostic validation, some sub-disciplines use the term clinical validation. If a commercial kit is used, the manufacturer has performed diagnostic validation.
- 5.
A transition to ISO 15189 as worldwide quality standard for diagnostic laboratories is an ongoing process (https://www.horizondiscovery.com/reference-standards/what-are-reference-standards/quality-controlled/iso-15189-accreditation-and-implementation). It is expected that many laboratories will have been certified according to this standard in 2019.
- 6.
CE-marking (Conformité Européen) of an in vitro diagnostic (IVD) test in Europe or a US Food and Drug Administration (FDA)-approved or FDA-cleared test in the USA.
- 7.
If the new reagent(s)/kit is obtained, together with a certificate of analysis by the manufacturer, the laboratory has to prove that its performance fulfils the manufacturer’s criteria. Then quarantine implies the period to establish this.
- 8.
It is well known that aerosols are formed during pipetting liquids into the pipette at micro-scale and adhere to pipette’s solid shaft. In the next pipetting step, these compounds can be detached and introduced into the fluid whilst pipetting with a new clean tip. In this way, aerosols containing targets can contaminate PCR reagents. Positive displacement pipettes in combination with filter tips prevent this type of contamination.
- 9.
Next to contamination risk during analytical procedures, aerosols form pathogens or (cultured) tumour cells comprise a health risk obliging to work under Good Microbiological Techniques (GMT) conditions.
- 10.
Gloves also prevent contamination with RNases and DNases (see also Sect. 6.3.2.5).
- 11.
SYBR®Green I is a considerably less mutagen than ethidium bromide when applied in the Ames test. Due to a much better intercalation into dsDNA and a five times larger quantum yield, SYBR®Green I is 103 times more sensitive. This pertains also to agarose electrophoresis. Using SYBR®Green I for agarose electrophoresis reveals more by-products of PCR. In this respect, EvaGreen® developed for HRMA performs even better than SYBR®Green I.
- 12.
In molecular pathology, and especially when analysing tumours, it is important to select the areas within the section containing the tumour cells. These are indicated on haematoxylin–eosin-stained sections by the pathologist.
- 13.
- 14.
Checkerboard and chessboard are synonyms used interchangeably.
- 15.
Commercial systems, capable of recording temperature profiles in individual reaction vessels, are available for various platforms.
- 16.
IU and copies per ml are not coupled in a 1:1 relationship. Moreover, 1 IU of HIV-1 does not equal to 1 IU of HBV. Within an international standard, exact numbers of virus are coupled. This allows inter-laboratory standardisation.
- 17.
MIQE advises not to use the term house-kee** gene. The correct term is reference gene, or internal control gene; being a gene that shows constitutive expression in those cells/tissues that are investigated in the study. Examples are HPRT (often very low expression level), actin and GAPDH (often high expression), albumin (extreme high expression in the liver), 18S rRNA (extreme high expression). These genes show, dependent of the tissue/cell type, developmental stage, age, at least 6-log units difference in expression. It is important to choose reference genes with a comparable expression as the gene under study. A balance in expression (i.e. comparable number of initial targets) is necessary to amplify both targets in the same reaction vessel with comparable efficiency.
- 18.
Reference [11] evaluates various RNA-protecting fluids. It is important to optimise for each individual target.
- 19.
Use of reverse transcriptases without RNase H domain is advised.
- 20.
cDNA synthesis of bacterial and viral RNA can be performed with random hexamers and gene-specific primers. The latter are not the best choice in routine molecular diagnostics.
- 21.
A well-known example is the screening for high-risk HPV types in the healthy population. The chance to find a high-risk HPV type is not inconsiderable. However, most of the virus will be cleared by the immune system, so the diagnostic threshold must be rather high and the diagnostic assay not so sensitive. On the contrary, patients with an indication for careful follow up in a gynaecological clinic must care genotyped for HPV with high diagnostic sensitivity and diagnostic specificity. Of course, different study sets were needed when those assays were developed.
- 22.
If no plasmids, or purified micro-organisms, are available, phases 2A and 2B have to be necessarily combined. In that case, matrix- and target-dependent issues cannot be separated.
- 23.
If the development of a LDT is not possible with purified samples containing single targets, steps C2 and C3 are combined with available samples. Mostly, archival patient samples certainly containing a high number of targets are used.
- 24.
Two sources were used as control for this paragraph about technical validation; sometimes the text corresponds. The reason is that the scientific world does not need more confusion about terms describing the quality of an assay than already present in the literature. Clin Biochem Rev Vol 29 Suppl (i) August 2008 and the website of Westgard http://www.westgard.com/glossary.htm.
- 25.
Definition provided and modified from CAP (college of American pathologists), Clinical Laboratory Improvement Amendments (CLIA) and Westgard (www.westgard.com).
- 26.
The description of the variation in the outcome data of a specific assay within and between laboratories is often used in a statistical context. In the laboratories, the terms intra- and inter-laboratory variation are also frequently applied. Both sets of terms designate the same principle.
- 27.
QCMD proficiency study on CMV (Schuurman et al. 2010) Final report. QCMD 2010 Human Cytomegalovirus DNA (CMVDNA10) EQA Programme www.qcmd.org/QCMD_CMVDNA_Report_2010.pdf.
- 28.
The description of the variation in the outcome data of a specific assay within and between laboratories is often used in a statistical context. In the laboratories, the terms intra- and inter-laboratory variation are also frequently applied. Both sets of terms designate the same principle.
- 29.
The gold standard method is the validated diagnostic assay that is considered as the current best to address a given diagnostic question by professionals with respect to reliability and accuracy. In bacteriology, e.g. this can be microscopy, bacterial or viral culture, or serology, or transmission electron microscopy. In pathology, this is often immunohistochemistry or in situ hybridisation. Not every laboratory, however, uses the same gold standard method for comparable diagnostic questions.
- 30.
A limitation of PCR and other amplification assays is the fact that only positive results can be found that hybridise to primers and probes. When a mutation changes the primer- or probe-binding site of a pathogen no longer allowing their annealing, the assay will become false negative. An unknown genotype not binding primers or probes can also be missed. Therefore, PCR results always need to be evaluated with a clinical view. This holds as well for the detection a fragment or genomic DNA or RNA of pathogens just cleared by the immune system. Such fragment can result in a positive PCR.
- 31.
Up to now, a uniform golden standard for a molecular diagnostic test detecting a given standard is lacking for most cases. It can occur that at different laboratories, when using different platforms, and even different primers and probes can arrive to a same diagnosis. Each laboratory performs its diagnostic validation with its own study set and only the answers matters. Of course, laboratories use accepted standards as much as possible. Proficiency studies (see Sect. 6.4.5.) indicate whether at given laboratory performs proficient with respect to the reference laboratory. Some exceptions exist, e.g. in phytopathology platforms, controls and procedures are strictly laid down by EU regulations. Other examples are being introduced in sub-disciplines, e.g. HPV screening in the Netherlands by the FDA-cleared Roche test.
- 32.
The College of American Pathologists (CAP) prefers the use of the terms “true positive ratio” instead of diagnostic sensitivity and “true negative ratio” instead of diagnostic specificity.
- 33.
For example, in clinical chemistry, the population healthy and affected will both give a normal distribution when measuring an analyte in blood. In molecular diagnostics, the whole “healthy population” can be absent; in that case only targets are found in samples from affected persons. This situation is not only present in microbiology, but also in pathology, haematology and genetics when translocations are only present in patients.
- 34.
In qPCR, the initial number of targets is mostly indicated by the Cq and not in weight or genome equivalents. Due to the read-outs from qPCR, a low Cq is indicative for a high number of targets. In Fig. 6.25a, true positives are localised at the left side of the threshold. In case of conventional PCR (and semi-quantitative or density measurements), the more common situation is found, i.e. that a high measured value corresponds to a large number of initial targets. As molecular diagnostics based on PCR technology is mostly based on qPCR, the population affected and healthy have a reverse position on the X-axis.
- 35.
This question was part of a large clinical study (MINDACT) with the hypothesis that chemotherapy is not necessary in lymph node negative low-risk primary tumours.
- 36.
van Drongelen A, et al. Comparison of market authorisation systems of medical devices in USA and Europe RIVM Letter report 2015-0001.
- 37.
- 38.
- 39.
- 40.
- 41.
Assessors perform their tasks on behalf of a licensed body, e.g. the “Raad voor Accreditatie” in the Netherlands.
References
Vandesompele, J et al. Elimination of primer-dimer artifacts and genomic coamplification using a two-stepSYBR green I real-time RT-PCR. Anal Biochem 2002; 303: 95–98.
Von Keyserling, H et al. The use of melting curves as a novel approach for validation of real-time PCR instruments. BioTechniques 2011; 51:179–184. https://doi.org/10.2144/000113735
Gangisetty O, et al. The optimization of TaqMan real-time RT-PCR assay for transcriptional profiling of GABA-A receptor subunit plasticity. J Neurosci Methods 2009; 181: 58–66.
Wilhelm, J et al. Influence of DNA target melting behavior real-time PCR quantification. Clin Chem 2000; 6:1738–1743.
Breslauer KJ, et al. Predicting DNA duplex stability from the base sequence. Proc Natl Acad Sci USA. 1986;83:3746–50.
Santa Lucia J. A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbour thermodynamics. Proc Natl Acad Sci USA. 1998;95:1460.
van Pelt-Verkuil E, van Leeuwen WB, te Witt R, editors. Molecular diagnostics. Part 2: Clinical, veterinary, agrobotanical and food safety applications. https://doi.org/10.1007/978-981-10-4511-0. ISBN: 978-981-10-4510-3 (hard copy) 978-981-10-4511-0 (E-book).
Berg KD, et al. Detection of microsatellite instability by fluorescence multiplex polymerase chain reaction. J Mol Diagn 2000; 2: 20–28. https://doi.org/10.1016/S1525-1578(10)60611-3
Bustin SA. A-Z of Quantitative PCR Fig. 10.34 (IUL Biotechnology, No. 5) ISBN 10; -09636817-8-8; ISBN-13: 978-0963681782.
Green RL, et al. Developmental validation of the quantifiler™ real-time PCR kits for the quantification of human nuclear DNA samples. J Forensic Sci 2005; 50:1–17.
Li Y, et al. Salivary transcriptome diagnostics for oral cancer detection. Clin Cancer Res. 2004; 10: 8442–50
Kleter B, et al. Development and clinical evaluation of a highly sensitive pcr-reverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. J Clin Microbiol. 1999;37:2508–17.
Henegariu O, et al. Multiplex PCR: dritical parameters and step-by-step protocol. BioTechniques 1997; 23:504–511.
Weegink C. Hepatitis C infection. The quest for new treatment strategies. Thesis. 2004. Amsterdam: AMC
Price CP. Evidence-based laboratory medicine: supporting decision-making. Clin Chem 2000; 46:1041–1050.
Bustin SA. Quantification of mRNA using real-time reverse transcription PCR (RT-PCR): trends and problems. J Mol Endocrinol 2002; 29: 23–39.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
van Pelt-Verkuil, E., van Leeuwen, W.B. (2019). Quality Assurance, Management and Control in Molecular Diagnostics. In: van Pelt-Verkuil, E., van Leeuwen, W., te Witt, R. (eds) Molecular Diagnostics. Springer, Singapore. https://doi.org/10.1007/978-981-13-1604-3_6
Download citation
DOI: https://doi.org/10.1007/978-981-13-1604-3_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-1603-6
Online ISBN: 978-981-13-1604-3
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)