Establishing Clinicalgrade Assays Presentation Title for Support of

Establishing Clinical-grade Assays Presentation Title for Support of Drug Trials February 23, 2011 May 3, 2012 Patrick Hurban Expression Analysis

Topics • • • Validated assays in clinical trials Analytical validation Multi-analyte tests An ADME case study The impact on Proficiency Testing Transitioning to NGS

Validated assays in clinical trials • Inclusion/Exclusion versus a treatment decision • Cytochrome P 450 s (CYP) – Ultra-rapid, Extensive, Intermediate, Poor metabolizers • Selecting dosage based on metabolism profile • Exclusion of patients from a clinical trial based on risk of an adverse event • Selection of patients for efficacy

Analytical validation • “The objective of validation of an analytical procedure is to demonstrate that it is suitable for its intended purpose” (ICH Harmonised Tripartite Guideline). • “Confirmation, through the provision of objective evidence, that the requirements for a specific intended use or application have been fulfilled” (ISO 9000: 2005)

Criteria • Criteria differ for quantitative and qualitative tests Category Example Accuracy The genotype is called correctly Reproducibility The genotype is called correctly within and between runs Robustness The genotype is called correctly even when the sample was stored incorrectly Reportable Range The genotypes called by the assay at a given locus Reference Range The genotypes that are known at a given locus Limit of detection Minimal amount of DNA that produces a valid genotype call Sensitivity A genotype is called when it comprises at least 20% of the starting sample Specificity A genotype is not called when it is not present

A simple RFLP test • The reportable range is tested during validation using known reference material. Controls Unknowns • Batch controls used during ongoing conduct of the assay are designed to produce a positive result for each value (genotype) in the reportable range.

Multi-analyte tests • When tests are comprised of multiple analytes, is there a requirement to affirm each reportable analyte? – During validation – For each batch • How does this impact quantitative versus qualitative tests? • Is a pattern of analytes sufficient?

Multi-analyte assays in clinical trials • Identify multiple exclusion/inclusion criteria simultaneously • Use marker combinations to refine criteria • Pre-screen populations of potential candidate to create pools of individuals with known marker sets • Identify individuals within pools to create tailored cohorts • Examples – ADME – Oncology

DMET • Affymetrix array – MIP-based amplification of target DNA • • • 231 genes 1936 SNP, CNV and Indels Core ADME genes plus additional content • Validation – 71 genomic DNA samples fully sequenced (Sanger) across the core set – But only 36% of the known rare alleles were known to be included – Reportable range confirmed for some, but not all of the content of the assay

Impact on proficiency testing • DMET assay utilized for analysis of PGX samples in CAP Proficiency Program • DMET assay reports CYP 2 D 6 *1 A/*65 but CAP expects *1/*10 • CYP Nomenclature database no longer has an entry for *1 – supplanted by *1 A, *1 B, *1 C, *1 D, *1 E, *1 XN • *65 is a more is an extension of *10 Allele Genotypes *10 A: 100 C>T; 1661 G>C; 4180 G>C *10 B: -1426 C>T; -1237_-1236 ins. AA; -1235 A>G; -1000 G>A; 100 C>T; 1039 C>T; 1661 G>C; 4180 G>C *10 D: 100 C>T; 1039 C>T; 1661 G>C; 4180 G>C, CYP 2 D 7 -like 3'-flanking region *65 100 C>T; 310 G>T; 843 T>G; 1661 G>C; 2850 C>T; 3384 A>C; 3584 G>A; 3790 C>T; 4180 G>C; 4481 G>A

Next Generation Sequencing • Each base position can exist in multiple discrete states • We can set criteria for variant calling sensitivity dependent upon factors such as quality and depth of coverage • What happens when we identify mutations of interest for a subject but have not specifically validated for these mutations? • How do we deal with potential complications such as allelic drop -out? Courtesy Promega website

Summary • Validation is contextual and must produce an assay that is suitable for the intended purpose. • Multi-analyte assays complicate validations because it may not be feasible to validate the entire reportable range. • The traditional use of batch control material also becomes complicated with multi-analyte assays. • Understanding the performance of an assay against reference methods can be difficult when the assay has greater resolution.

Recommended reading • • • ICH Harmonised Tripartite Guideline, Validation of Analytical Procedures: Text and Methodology “Recommended Principles and Practices for Validating Clinical Molecular Pathology Tests, ” Jennings, et al, Arch Pathol Lab Med, V 133, p. 743, 2009 “A Standardized Framework for the Validation and Verification of Clinical Molecular Genetic Tests, ” Mattocks, et al, Eur J Hum Gen V 18, p. 1276, 2010. “Multiplex Assay for Comprehensive Genotyping of Genes Involved in Drug Metabolism, Excretion, and Transport, ” Daly, et al, Clin Chem V 53, p. 1222, 2007. “Influence of Cytochrome P 450 Polymorphisms on Drug Therapies: Pharmacogenetic, Pharmacoepigenetic and Clinical Aspects, ” Ingelman-Sundberg, et al, Pharmacology & Therapeutics V 116, P. 496, 2007