Universal Genetic Testing Reveals Variants Linked With Increased Cancer Risk in CRC

One in 6 patients with colorectal cancer (CRC) harbor genetic changes that substantially increase their risk for cancer, especially in those who are younger at the time of diagnosis, according to the data from a prospective, multicenter study published in JAMA Oncology.^1,2

A total of 361 patients with CRC enrolled on the study and underwent germline genetic testing with a panel that included approximately 80 genes known to increase risk for cancer.2 Results showed that these cancer-linked genetic changes were detected in 15.5% of patients with CRC (n = 56). Moreover, these changes predominantly occurred in high-penetrance cancer susceptibility genes like APC, BRCA2, and TP53. A younger age at diagnosis was also linked with increased likelihood of genetic changes associated with increased susceptibility to cancer.2

“Colorectal cancer is one of the most common types of cancer. Recently, we've seen an unfortunate increase in incidence among people younger than 50-years-old. Improving our understanding of who is at risk and implementing appropriate screening strategies is essential to ensuring these cancers are caught and treated earlier when outcomes are typically better,” Robert Nussbaum, MD, chief medical officer of Invitae Corporation, stated in a press release. “Access to information about inherited risk factors is also important after diagnosis, not only for the patients themselves…but also in alerting other family members of the need to be tested to see if they are also at increased risk and need more intensive surveillance to catch cancer early and prevent more serious disease.”

Hereditary factors are key players in the risk of developing several cancers. The detection of a germline predisposition can impact treatment decisions, risk-reducing interventions, cancer screening, and testing, in patients and their relatives. The impact of broad-based testing for inherited germline variants in patients with cancer vs traditional strategies of selection for genetic testing remains an open question.

In this study, investigators sought to examine the occurrence of pathogenic and likely pathogenic germline variants (PGVs) in inherited cancer susceptibility genes identified by a broad universal testing approach vs a targeted approach based on clinical guidelines. Investigators also wanted to evaluate the link between universal genetic testing and clinical management and the proportion of families who receive cascade family variant testing (FVT) when available at no cost.

For the prospective study, investigators enrolled patients with solid tumors who were receiving care at any of the 3 Mayo Clinic Cancer Centers or the Mayo Clinic Health System from April 2018 to March 2020. Patients were between 18 years and 85 years of age, had a new or active cancer diagnosis, confirmed pathologic diagnosis of carcinoma, and received treatment in medical oncology, radiation oncology, dermatology, or surgical clinics. Those with malignant hematologic cancers and those who were under surveillance following curative cancer, were not included in the study.

Participants were selected for cancer type, disease stage, family history of cancer, race/ethnicity, or age. All had undergone germline testing with a next-generation sequencing platform that tested for 83 genes; as of July 2019, it tested for 84 genes. Test results were analyzed by a certified genetic counselor; those with PGVs were offered genetic counseling. PGVs were classified as either high, intermediate, or low penetrance or recessive actionable variants; relative risk in these subsets was greater than 4, between 2 and 4, and less than 2, respectively.

“A PGV was considered incremental if it was detected based on the universal testing performed in this study and would not have been identified in targeted testing approaches recommended by published guidelines,” the study authors noted.

Additionally, free cascade FVT was offered to all blood relative of participants with PGVs within a 90-day window of the final test results. Patients were educated about the FVT program during genetic counseling; they were also provided assistance in relaying this information to their relatives via a letter and online video that detailed the benefits and risks of genetic testing.

The final analytic cohort included 2984 patients, who had a median age of 61.4 years; 53% of the patients were male. Additionally, at the time of genomic analysis, 18.6% (n = 535) of patients had stage 0/I disease, 16.7% (n = 477) had stage II disease, 20.7% (n = 593) had stage III disease, and 43.9% (n = 1257) had stage IV disease. Just over 5% (5.3%; n = 159) of patients were Hispanic/Latino, 3.7% were African American (n = 110), and 1.8% were Asian (n = 53). Moreover, 34.1% (n = 1019) of patients had a family history of cancer in a first-degree relative.

Of the 2984 total patients, 13.3% (n = 397) were found to harbor 426 PGVs. In 27 patients, more than 1 PGV variant was identified; 2 of these patients had 3 variants detected. Moreover, 149 of the PGVs had high penetrance, 133 had moderate penetrance, and 65 had low penetrance. Fifty patients were noted to harbor variants linked with recessive syndromes.

The 6 most common PGVs were found in BRCA1 and BRCA2, mono-allelic MUTYH, CHEK2, Lynch mismatch repair genes, and ATM. Approximately 47% of patients (n = 1415), variants of uncertain significance. The incidence of PGVs in different cancer types ranged; the incidence was 20.6% for ovarian cancer, 15.9% for pancreatic cancer, 15.3% for colorectal cancer, 13.7% for prostate cancer, 14.7% for lung cancer, 14.5% for cholangiocarcinoma, 13.3% for endometrial cancer, and 14.2% for bladder cancer.

Notably, the variant rate by cancer stage was found to be similar; this rate was 13.7% in stage 0 to II disease and 13.1% in stage III to IV disease (P = .24). Patients who had a family history of cancer in the same organ system were found to have a high rate of PGVs of 21.8%.

Additional data indicated that 192 cases had incremental findings that would not have been identified via phenotype or family history–based testing criteria using established guidelines; this accounted for 6.4% of the overall cases and 48.4% of the 397 patients with PGVs. Of these 192 patients, 18.2% (n = 35) had high-penetrance PGVs, 43.2% (n = 83) had low-penetrance PGVs, and 17.2% (n = 33) had recessive-penetrance PGVs. Of the 149 patients with high-penetrance PGVs, 28.2% (n = 42) experienced clinically actionable management and treatment adjustments indicated by the PGVs that were identified.

Additionally, 17.6% (n = 70) of patients with PGVs had relatives that accepted and underwent FVT within a 3-month window of their test result. Forty-five percent (n = 79) of the 176 family members had positive results.

Data from logistic regression analyses indicated that younger age at the time of cancer diagnosis was linked with a higher likelihood of having a PGV (odds ratio [OR], 1.38; 95% CI, 1.06-1.78). Although patients who had CRC, pancreatic cancer, or prostate cancer were noted to have comparable PGV prevalence vs those with breast cancer, those with melanoma were observed to have a significantly lower likelihood of having PGVs (OR, 0.58; 95% CI, 0.32-1.00).

“In this large, prospective multisite cohort study with a broad mixture of cancer types and stages, a variety of academic and community cancer practices, and patients unselected for family cancer history, universal multigene panel testing was associated with increased detection of clinically actionable, heritable variants over the predicted yield by targeted testing based on clinical guidelines,” the study authors concluded. “This study offers significant insight into the performance of multigene panel testing and has broad implications for its wide clinical implementation and acceptance in oncology practice.”

References

1. Samadder N, Johnson D, Boardman L, et al. Comparison of universal genetic testing vs guideline-directed targeted testing for patients with hereditary cancer syndrome. JAMA Oncol. 2021;7(2):230-237. doi:10.1001/jamaoncol.2020.6252
2. Study finds risk-causing genetic changes in nearly 1 in 6 patients with colorectal cancer. News release. Invitae Corporation. April 15, 2021. Accessed April 23, 2021. https://prn.to/3eGBCkA