Germline Sequencing Improves Tumor-Only Sequencing Interpretation in a Precision Genomic Study of Patients With Pediatric Solid Tumor

Abstract

Purpose: Molecular tumor profiling is becoming a routine part of clinical cancer care, typically involving tumor-only panel testing without matched germline. We hypothesized that integrated germline sequencing could improve clinical interpretation and enhance the identification of germline variants with significant hereditary risks.

Materials and methods: Tumors from pediatric patients with high-risk, extracranial solid malignancies were sequenced with a targeted panel of cancer-associated genes. Later, germline DNA was analyzed for a subset of these genes. We performed a post hoc analysis to identify how an integrated analysis of tumor and germline data would improve clinical interpretation.

Results: One hundred sixty participants with both tumor-only and germline sequencing reports were eligible for this analysis. Germline sequencing identified 38 pathogenic or likely pathogenic variants among 35 (22%) patients. Twenty-five (66%) of these were included in the tumor sequencing report. The remaining germline pathogenic or likely pathogenic variants were single-nucleotide variants filtered out of tumor-only analysis because of population frequency or copy-number variation masked by additional copy-number changes in the tumor. In tumor-only sequencing, 308 of 434 (71%) single-nucleotide variants reported were present in the germline, including 31% with suggested clinical utility. Finally, we provide further evidence that the variant allele fraction from tumor-only sequencing is insufficient to differentiate somatic from germline events.

Conclusion: A paired approach to analyzing tumor and germline sequencing data would be expected to improve the efficiency and accuracy of distinguishing somatic mutations and germline variants, thereby facilitating the process of variant curation and therapeutic interpretation for somatic reports, as well as the identification of variants associated with germline cancer predisposition.

FIG 1.

Post hoc comparison of tumor and matched germline sequencing. Of the 400 patients enrolled in GAIN at the time of this study, 160 had tumor sequencing reports and separate germline sequencing reports available. Each tumor-only sequencing report also had a separate clinical interpretation report (GAIN report) that conveyed therapeutic recommendations on the basis of the variants present in tumor-only sequencing. GAIN, Genomic Assessment Improves Novel Therapy.

FIG 2.

Germline pathogenic and likely pathogenic variants. (A) There were 18 heterozygous P/LP variants detected across 12 different autosomal dominant cancer risk genes. Nine of the tumors had a second mutation within the same gene (*). (B) There were 20 heterozygous P/LP variants detected across 10 different autosomal recessive cancer risk genes. None of the tumors had a second mutation within the same gene. P/LP, pathogenic or likely pathogenic.

FIG 3.

Comparison of germline variants within tumor-only sequencing reports. (A) Tumor-only sequencing SNVs. Pie chart demonstrating that only a minority (29%) of variants identified by tumor-only sequencing are somatic events with the rest being verified as germline variants by review of germline sequencing data. (B) Tumor-only sequencing SNVs by clinical tier. Germline P/LP variants were identified in all four clinical actionability tiers within the tumor-only sequencing report, including tiers 1 and 2. Although there were germline VUS present within tier 2, most of the VUS were within tiers 3 and 4. (C) SNVs annotated as potentially germline in tumor-only sequencing report. SNVs were flagged as possibly germline in all four clinical actionability tiers within the tumor-only sequencing report. More than half of the flagged variants within tiers 1 and 2 were somatic events, whereas the majority of flagged variants in tiers 3 and 4 were verified as germline. B/LB, benign or likely benign; P/LP, pathogenic or likely pathogenic; SNV, single-nucleotide variant; VUS, variant of uncertain significance.

FIG 4.

Comparison of somatic and germline VAFs in tumor and normal samples. Violin plot demonstrating the distribution of AF of variants identified from tumor-only sequencing (from tumor) and from germline sequencing (from normal). The height of the violin plot indicates the abundance of variants having AFs indicated along the x-axis. Vertical dashed lines embedded within the violin plots indicate the median AFs in the indicated cohort, and vertical dotted lines indicate the upper and lower quartiles of AFs within the data. Vertical dashes that run across the entire graph indicate AFs of 40% and 60%. Variants detected in tumor sequencing with AFs between 40%-60% are often interpreted to be likely germline when matching germline sequencing data are unavailable. AF, allelic fraction VAF, variant allele fraction.

FIG A1.

Pairwise comparison of germline VAFs in tumor versus normal tissue. Pairwise comparison of the AF of germline variants identified from tumor-only sequencing (tumor tissue) to the AF from germline sequencing (normal tissue). Each variant is color-coded to indicate the copy-number status (amplification, gain, single copy deletion, or neutral) of the corresponding gene. Copy-number gains and loss could not explain many germline VAFs outside of 40%-60% in tumor tissue. AF, allelic fraction; CN, copy number; VAF, variant allele fraction.