Introduction

Next generation sequencing (NGS) is increasingly being used to guide care in cancer patients. Yet tumor testing may also identify variants, which, if confirmed to be present in the germline, may have implications for patients and their at-risk family members. While confirmatory germline testing is recommended following detection of suspected germline pathogenic or likely pathogenic (P/LP) variants, numerous barriers can make this challenging. Through the current study, we sought to characterize P/LP germline variants among patients who underwent tumor-normal matched next generation sequencing (TNmNGS), and assess follow-up including documented rates of confirmatory germline testing discussions, referrals, and uptake.

Methods

Procedures developed at the Vanderbilt-Ingram Cancer Center (VICC) for tumor sequencing include centralized oversight for all clinical tumor NGS. After the oncologist orders TNmNGS testing the report is released through a centralized, HIPAA-compliant portal. Each report is reviewed by an oncologist with molecular expertise who then proactively communicates with treating providers when actionable findings are identified, including variants suggestive of an underlying germline cancer predisposition syndrome, and highlights the option for Hereditary Cancer Clinic evaluation. Additional forums for discussion of cases include weekly, multidisciplinary, molecular tumor boards and other case conferences. Testing for solid tumors includes a 648 gene NGS panel with whole transcriptome RNA seq, microsatellite instability (MSI) status, and tumor mutational burden (TMB). The matched normal component allows for the identification of germline alterations and includes 46 cancer-related genes based on well-established guidelines; the platform is not validated as a standalone germline test. For the current analysis, we categorized the association of cancer type with genes in which P/LP variants were identified as established, not established, or unknown, according to the Genetics/Familial National Comprehensive Cancer Network guidelines. Clinical data was collected through chart review and included: tumor type on which TNmNGS was performed, additional malignancy diagnoses, NGS findings including copy number loss or second pathogenic somatic variant in gene(s) of interest, TMB and MSI-high status, if germline testing was offered and pursued per documentation or referrals, and completed confirmatory germline testing.

Results

From May 2020 to September 2021, 916 patient reports of solid malignancy were ordered and reviewed, of which 83 germline variants were identified in 81 (8.84%) individuals which warranted follow-up. The population in which P/LP findings were identified consisted of 42 females (52%) and 39 males (48%). Median age was 61 years. At the time of the current review 23 (28%) patients were deceased. Seventy-two patients (89%) had advanced cancer. Tumor types included breast (10), colorectal (10), ovarian (10), non-small cell lung (9), sarcoma (7), tumor of unknown origin (TUO) (6), melanoma (5), biliary (3), gastric (3), pancreatic (3), prostate (3), thyroid (3), and other (9). Of the 83 P/LP variants, 24 (29%) had an established cancer risk association with the patients’ cancer type on which tumor testing was performed; 54 (66%) had no established risk association between gene and tumor type; and 5 (6%) were identified in TUO. We observed P/LP germline variants in 16 different genes. In select genes (ATM, BRCA1, BRCA2, and CDKN2A), copy number loss or separate pathogenic somatic variant were common. Additional genomic findings are outlined in Table 1.
Among 81 patients in whom a P/LP variant was reported on TNmNGS, 70% had documented evidence of discussions with their provider, referrals, or testing regarding their germline finding. Among these patients, 15 (19%) had completed testing prior to TNmNGS; 20 (25%) had germline testing completed after identification of suspected germline variants on TNmNGS; 22 (27%) had been referred for genetic counseling and testing and/or had a documented conversation regarding this option; the remaining 24 (30%) had no documented evidence of germline testing discussions, referrals, or testing. Proportions of confirmatory germline testing completed in males (11/39; 28%) were lower than in females (24/42; 57%) (p<0.01), driven by 15 females with breast or ovarian cancer. Of the 23 patients deceased at the time of the current review, only 9 (39%) had completed confirmatory testing.

Conclusion

We show that this Precision Oncology review protocol resulted in a genetic evaluation and/or provider communication in 70% of patients with possible germline P/LP on tumor NGS testing. This protocol could be used to expand cancer genetic services across healthcare systems. Barriers to care exist in several areas with more females than males completing confirmatory testing, limited time for patients with advanced cancer patients to complete additional testing, and lack of evidence for cancer association with some of the cancer types in these patients. Additional research is needed to improve access and to better understand the uptake of confirmatory testing for cancer patients undergoing NGS testing.

VIEW THE PUBLICATION