This lecture, the final part of a "Molecular Pathology for Surgical Pathologists" series, provides a comprehensive guide to interpreting genomic variants detected by sequencing. The presentation begins with a review of molecular biology fundamentals, including DNA structure, the central dogma, variant types, and sequencing workflows.
The core of the lecture focuses on two key classification frameworks for variant interpretation. The first is the Clingen/CGC/VIC oncogenicity classification, a rigorous, point-based system used to determine a variant's functional impact and its ability to drive cancer. The second is the AMP/ASCO/CAP framework, which categorizes variants into four tiers based on their clinical significance, considering therapeutic, diagnostic, and prognostic evidence.
The speaker provides a detailed overview of essential public and private databases used in variant analysis, including population databases (Nomad), cancer-specific databases (COSMIC, OncoKB, cBioPortal), and clinical repositories (ClinVar). The practical application of these frameworks and tools is demonstrated through two illustrative cases: a sinonasal tumor where a driving HER2 mutation was identified, and a liver lesion initially diagnosed as a GIST that was reclassified as metastatic uveal melanoma based on a characteristic GNA11 mutation. The session concludes with a summary of best practices for reporting molecular findings.
A Guide to Molecular Variant Interpretation for Pathologists
0:02:08 Review of Molecular Pathology Basics: The session starts with a review of fundamental concepts from the previous lecture, including DNA structure, the central dogma (DNA to RNA to protein), types of genomic alterations (copy number changes, gene fusions, SNVs), specimen selection, and the NGS data pipeline (FASTQ -> BAM -> VCF).
0:10:23 The Core Approach to Variant Interpretation: A two-step approach is outlined for every potential variant: 1) technical assessment to confirm the variant is real and not an artifact, and 2) functional assessment to determine if the variant is biologically impactful.
0:14:08 Germline vs. Somatic Variants: The lecture explains the critical distinction between germline (hereditary) and somatic (tumor-acquired) variants. While the molecular mechanism may be the same, the clinical implications can differ vastly. For example, a high allele frequency (~50%) may suggest a germline origin, but this cannot be confirmed without dedicated germline testing.
0:17:18 Framework 1: Oncogenicity Classification (Clingen/CGC/VIC): This framework provides a rigorous, point-based system to classify a variant's functional role in cancer. Variants are scored and categorized as Oncogenic, Likely Oncogenic, VUS (Variant of Uncertain Significance), Likely Benign, or Benign based on criteria such as being a null variant in a tumor suppressor or occurring in a known cancer hotspot.
0:27:59 Framework 2: Clinical Significance Classification (AMP/ASCO/CAP): This system categorizes variants based on their clinical actionability. It uses a four-tier system (Tier 1-4) supported by levels of evidence (A-D) to define a variant's role in therapy, diagnosis, or prognosis, referencing FDA approvals and professional guidelines like NCCN.
0:33:03 Essential Databases and Tools for Interpretation: An overview of critical resources is provided:
Nomad: A population database used to filter out common, benign polymorphisms.
COSMIC & Cancer Hotspots: Databases that catalog somatic mutations found in cancers, helping identify recurrent and significant variants.
OncoKB & cBioPortal: Expert-curated knowledge bases that link variants to clinical actionability, therapies, and clinical trials.
ClinVar: A public archive of variants and their asserted clinical significance, submitted by labs and expert panels.
0:51:00 Case Study 1: Sinonasal Tumor: A 75-year-old female's tumor sequencing revealed several variants. An ERBB2 (HER2) exon 20 duplication was identified as the oncogenic driver, showing functional similarity to EGFR mutations seen in this tumor type. An ABL1 variant was dismissed as a common sequencing artifact by checking an internal lab database.
1:05:10 Case Study 2: Reclassifying a Liver Lesion: A liver biopsy, initially diagnosed as a GIST, was sequenced. The results showed no typical GIST mutations but revealed a GNA11 mutation, a characteristic hotspot variant for uveal melanoma. This molecular finding prompted re-evaluation with IHC stains (SOX10, HMB45), which confirmed a melanocytic origin and changed the diagnosis to metastatic uveal melanoma.
1:08:18 A Practical Workflow for Variant Analysis: The speaker shares his personal step-by-step approach: 1) Check allele frequency and technical quality. 2) Review automated ClinVar calls for known pathogenic variants. 3) Filter by variant type (e.g., removing synonymous/intronic variants). 4) Filter out common polymorphisms using Nomad. 5) Check against an in-house database for known artifacts. 6) Interpret the remaining relevant genes in the clinical context.
1:09:16 Best Practices for Molecular Reporting: A final summary outlines the key elements of a comprehensive molecular pathology report, including clear patient/sample information, key findings, detailed variant information (transcript ID, allele frequency), classification tier, and technical methodology.