Recent recommendations from the ESMO Precision Medicine Working Group and National Comprehensive Cancer Network (NCCN), coupled with US Food and Drug Administration (FDA) review and approval, have highlighted the use of tumor mutational burden (TMB) testing in clinical practice for a number of different tumor types.
As the treatment armamentarium for cancer has expanded rapidly in recent years, primarily due to the dawn of cancer immunotherapy and molecularly guided therapies, there is now an increasing need for the clinical use of comprehensive genomic profiling (CGP) to identify patients with specific genomic alterations and genomic signatures, such as a high TMB (≥10 mutations per megabase, per the ESMO Precision Medicine Working Group), who may benefit from such treatments. TMB has emerged as a clinically relevant biomarker for a number of metastatic solid tumors that are treatable with cancer immunotherapy, and has joined the existing portfolio of available tools for personalization of cancer treatment.
TMB: A tumor-agnostic genomic signature
Detection of actionable genomic alterations and genomic signatures, such as a high TMB, through CGP is becoming increasingly important in cancer care. TMB (the number of somatic protein-coding base substitutions and insertion/deletions per megabase in a tumor specimen) approximates the burden of immunogenic neoantigens produced by a tumor: patients with a higher TMB may be more likely to develop an anti-tumor immune response. Thus, TMB is a genomic signature with the potential to provide diagnostic, prognostic, and predictive insights for clinical decisions, considering the wide uptake of immune checkpoint inhibitor therapies across indications.
Formerly considered to be a “one-size-fits-all” approach, immune checkpoint inhibition via the PD-1/PD-L1 axis has since demonstrated improved response rates in patients with a high TMB across ten solid tumor types: response rates were 29 percent and 6 percent in patients with a high TMB and a low TMB across different cancer types, respectively. These data build on findings from a TMB analysis across 29 tumor types, which found a positive correlation between TMB and objective response rate with anti-PD-(L)1 treatment (see figure below). Such findings have elevated the importance of TMB as a tumor-agnostic genomic signature and mark an important step on the path toward tailored cancer care, with support from regulators and clinical societies.
1Adapted with permission of the American Society for Clinical Investigation, from “PD-L1 expression and tumor mutational burden are independent biomarkers in most cancers,” Yarchoan, M. et al. (2019) JCI Insight 4(6): e126908. pii: 126908; permission conveyed through Copyright Clearance Center, Inc.
ACC: adenoid cystic carcinoma; dMMR: deficient mismatch repair; HCC: hepatocellular carcinoma; H&N: head and neck; NSCLC: non-small cell lung cancer; ORR: objective response rate; PD-1 / PD-L1: programmed cell death protein 1 / ligand 1; TMB: tumor mutational burden; RCC: renal cell carcinoma; SCLC: small cell lung cancer.
Regulatory status and clinical application of TMB
The FDA recently approved the first, and so far only, CGP companion diagnostic assay to measure TMB. This assay allows for identification of patients who may be appropriate for treatment with immune checkpoint inhibitors, following progression on prior treatment and for whom there are no satisfactory alternative treatment options.
A recent report from the ESMO Precision Medicine Working Group also supports TMB testing for a number of solid tumors, including cervical cancer, well- and moderately differentiated neuroendocrine tumors, salivary cancer, thyroid cancer, and vulvar cancer. Further validation of the testing approach comes from NCCN: TMB analysis using an FDA-approved test is recommended ahead of second-line treatment decisions for cervical cancer, uterine sarcoma and endometrial carcinoma, vulvar cancer, bone cancer, and cancer of unknown primary. Where there are no alternative treatment options, NCCN recommends immune checkpoint inhibition for patients with these TMB-high tumors.
While it is common practice in the clinical trial setting, TMB assessment is currently not a standard procedure for community oncologists. The positive outlook shared by the FDA, the ESMO Precision Medicine Working Group, and the NCCN helps to build practitioners’ confidence in biomarker-driven treatment approaches generally, but translating this into practical change demands further education and the wide availability of robust, reliable analytical tools.
Adopting a biomarker-driven approach to treatment decisions
Personalized healthcare aims to optimize patient outcomes with the right treatment at the right time. Adapting existing procedures may be essential to integrate TMB testing (and other biomarker-driven approaches, where possible) into community clinical practice and clinicians should feel empowered to do this on the strength of the new recommendations.
CGP combines a highly sensitive assay and sophisticated algorithm that the accurate measurement of TMB demands. Compared with alternative analytical methods, CGP assessment of TMB also has a lower cost, shorter turnaround time, and lower requirement for DNA: crucial factors when considering its deployment in community clinical practice. Furthermore, CGP can facilitate parallel analysis of other genomic signatures, such as microsatellite instability (MSI), thus allowing for potential identification of multiple actionable biomarkers from a single analysis.
MSI was the first biomarker to lead to tumor-agnostic approval of an immune checkpoint inhibitor, after data showed favorable responses to PD-1 blockade in solid tumors with high levels of MSI (MSI-H). MSI-H tumors generate increased levels of neoantigens that are recognized by the patient’s immune system, allowing T-cells to launch an anti-tumor immune response. Across a variety of tumor types, MSI-H is a partial subset of TMB-high: the majority of MSI-H samples are also TMB-high, but not vice versa. This was demonstrated through MSI-only testing in colorectal cancer, which failed to identify TMB-high tumors that could have responded well to immune checkpoint inhibition. PD-L1 has also demonstrated clinical utility as a predictive biomarker for response to checkpoint inhibition. Current evidence suggests that PD-L1 expression predicts response independently from TMB, with low correlation between the two biomarkers in various advanced solid tumors. A combined biomarker analysis of TMB, MSI and PD-L1 may offer complementary detail to prescribing clinicians, compared with a reliance on one test in isolation, to inform a treatment decision.
Conclusions
The increased use of personalized cancer therapy is now an achievable goal. The FDA approval of the first CGP assay for measurement of TMB and recent recommendations from regulatory and standards bodies ultimately support physicians in their clinical practice, optimize treatment options to provide molecularly guided approaches with immune checkpoint inhibitors, and improve outcomes for oncology patients.