Residency · Residency · Pathology

Next-Generation Sequencing in Oncology

Introduction

Next-generation sequencing (NGS) has become an indispensable tool in oncologic pathology, enabling comprehensive genomic profiling of tumors for diagnosis, prognosis, and therapy selection. Understanding the principles, platforms, and clinical applications of NGS is essential for the modern pathologist.

Principles of NGS

Overview

NGS performs massively parallel sequencing of millions of DNA fragments simultaneously, also known as high-throughput or second-generation sequencing. It enables detection of single nucleotide variants (SNVs), insertions and deletions (indels), copy number alterations (CNAs), and structural rearrangements or fusions. The key advantage over Sanger sequencing is the simultaneous analysis of hundreds to thousands of genes in a single assay.

General Workflow

The NGS workflow begins with nucleic acid extraction of DNA and/or RNA from FFPE tissue, fresh tissue, or liquid biopsy. Library preparation involves fragmentation, adapter ligation, and amplification. Target enrichment for targeted panels uses hybrid capture or amplicon-based methods. Sequencing consists of clonal amplification on a flow cell followed by sequencing by synthesis or other chemistry. Finally, bioinformatics performs alignment to the reference genome, variant calling, annotation, filtering, and interpretation.

Sequencing Platforms

Illumina platforms (MiSeq, NextSeq, NovaSeq) use sequencing by synthesis with short reads of 150-300 bp and dominate clinical use. Thermo Fisher Ion Torrent (Ion GeneStudio S5) is semiconductor-based and detects pH change from nucleotide incorporation. Oxford Nanopore (MinION, PromethION) provides long-read, real-time sequencing that enables structural variant and methylation detection. PacBio (Revio) offers long-read, high-fidelity (HiFi) sequencing for research and emerging clinical applications.

Approaches to Oncology Sequencing

Targeted Gene Panels

Targeted panels focus on 50-500 or more cancer-relevant genes with known clinical significance. They achieve high depth of coverage (500-1000x or more), providing sensitivity for low-frequency variants. Examples include MSK-IMPACT, FoundationOne CDx, Tempus xT, and institutional custom panels. These panels are suitable for FFPE tissue with limited DNA input and are preferred for clinical decision-making due to fast turnaround and established pipelines.

Whole Exome Sequencing (WES)

WES sequences all protein-coding regions, covering approximately 1-2% of the genome and about 20,000 genes. Coverage is typically 100-300x. It is used in research and some comprehensive tumor profiling programs and can identify novel variants not captured by targeted panels.

Whole Genome Sequencing (WGS)

WGS sequences the entire genome including non-coding regions, detecting structural variants, copy number changes, and regulatory mutations. Coverage per base is lower (30-60x) but the approach is comprehensive. It is increasingly used in hematologic malignancies and pediatric oncology.

RNA Sequencing

RNA sequencing detects gene fusions, gene expression profiles, and splice variants. It is critical for fusion-driven tumors such as EML4-ALK in lung cancer, EWSR1 fusions in sarcoma, and BCR-ABL1 in CML. Anchored multiplex PCR methods (such as Archer/ArcherDx) enable detection of fusions with unknown partners. RNA sequencing is complementary to DNA-based sequencing.

Bioinformatics Pipeline

Key Steps

Alignment maps reads to the human reference genome (hg19/GRCh37 or hg38/GRCh38). Variant calling identifies differences from the reference using tools such as GATK, MuTect2, and VarDict. Annotation predicts functional impact (missense, nonsense, frameshift) and cross-references databases including ClinVar, COSMIC, and OncoKB. Filtering removes artifacts, germline variants (using paired normal or population databases), and variants of uncertain significance. Interpretation and reporting classifies variants by clinical actionability.

Variant Classification in Somatic Oncology

The AMP/ASCO/CAP joint consensus guidelines provide a tiered framework. Tier I includes variants with strong clinical significance such as FDA-approved therapy targets or established diagnostic and prognostic markers. Tier II includes variants with potential clinical significance supported by clinical trials or emerging evidence. Tier III covers variants of unknown clinical significance. Tier IV designates benign or likely benign variants.

Key Clinical Applications

Therapy Selection (Predictive Biomarkers)

EGFR mutations in lung adenocarcinoma guide use of osimertinib and other TKIs. ALK/ROS1/RET/NTRK fusions are targeted by crizotinib, entrectinib, and selpercatinib. BRAF V600E in melanoma, colorectal, lung, and thyroid cancer is targeted by vemurafenib and dabrafenib plus trametinib. KRAS G12C in lung and colorectal cancer is targeted by sotorasib and adagrasib. BRCA1/2 mutations guide PARP inhibitor use with olaparib. MSI-H and TMB-H status guides immune checkpoint inhibitor therapy with pembrolizumab. ERBB2 (HER2) amplification or mutations guide trastuzumab and trastuzumab deruxtecan therapy. IDH1/2 mutations in AML, cholangiocarcinoma, and glioma are targeted by ivosidenib and enasidenib.

Diagnostic Applications

NGS aids tumor classification by identifying defining molecular alterations (such as IDH-mutant glioma or FLT3-mutated AML). The WHO Classification of Tumours increasingly incorporates molecular criteria. Molecular profiling can also suggest tissue of origin for tumors of unknown primary.

Prognostic Applications

TP53 mutations confer adverse prognosis across many tumor types. 1p/19q co-deletion in oligodendroglioma indicates favorable prognosis and treatment response. Mutational signatures reveal patterns indicative of specific mutagenic processes such as UV exposure, tobacco, APOBEC activity, and mismatch repair deficiency.

Quality Considerations

Pre-Analytical

FFPE tissue quality is a major concern because formalin fixation introduces artifacts (C>T deamination) and cold ischemia time affects DNA and RNA integrity. Minimum tumor content is typically greater than 20% tumor cellularity for reliable variant detection. Pathologist review for tumor content estimation and macrodissection guidance is essential. DNA and RNA quality assessment uses DIN score and fragment size analysis.

Analytical

Depth of coverage must be at least 500x for somatic variant detection at 5% VAF. Variant allele frequency (VAF) represents the proportion of reads supporting the variant; low VAF may reflect subclonal populations or poor tumor content. The typical sensitivity limit is 5% VAF for targeted panels, though this is lower with unique molecular identifiers (UMIs). Run-level quality metrics include Q30 scores, uniformity of coverage, and on-target percentage.

Post-Analytical

Variant interpretation requires multidisciplinary expertise. Molecular tumor boards integrate genomic findings with clinical and pathologic context. Periodic re-analysis is warranted as new evidence and therapies emerge. Incidental pathogenic germline variants require genetic counseling referral.

Clinical Pearls

Targeted gene panels with high depth of coverage are the standard for clinical oncology NGS, balancing comprehensive coverage of actionable genes with practical turnaround time and cost. RNA sequencing is essential for detecting gene fusions, which are therapeutically actionable in many tumor types and may be missed by DNA-only approaches. Pathologist assessment of tumor cellularity and tissue quality is a critical pre-analytical step that directly impacts the reliability of NGS results. The AMP/ASCO/CAP tiered classification system provides a standardized framework for reporting somatic variants by level of clinical evidence.

References

1. Li MM, et al. Standards and guidelines for the interpretation and reporting of sequence variants in cancer: a joint consensus recommendation of the AMP, ASCO, and CAP. J Mol Diagn. 2017;19(1):4-23.
2. Mardis ER. The impact of next-generation sequencing on cancer genomics. Annu Rev Genomics Hum Genet. 2019;20:631-657.
3. Chakravarty D, et al. OncoKB: a precision oncology knowledge base. JCO Precis Oncol. 2017;1:1-16.
4. Jennings LJ, et al. Guidelines for validation of next-generation sequencing-based oncology panels. J Mol Diagn. 2017;19(3):341-365.

Read this lecture as Markdown