Tumor mutational burden does not predict response to immunotherapy in all cancer types
Click Here to Manage Email Alerts
Tumor mutational burden predicted immunotherapy response for some, but not all, cancer types, according to study results published in Annals of Oncology.
For this reason, additional tumor type-specific research is needed to determine how tumor mutational burden can be best applied in cancers for which it does not appear to be associated with outcomes, such as breast and prostate cancers and glioma, researchers noted.
“The outstanding clinical responses to immunotherapy in a subset of patients with cancer have rightfully generated a lot of excitement. However, as only some patients will respond to immunotherapy, there is a need for clinical biomarkers to identify which patients may benefit from this powerful treatment,” Daniel J. McGrail, PhD, postdoctoral fellow at The University of Texas MD Anderson Cancer Center, told Healio. “[Because] immune cells can recognize mutant proteins from cancer cells, one prevailing hypothesis is that tumors with more mutations, referred to as ‘tumor mutational burden high,’ will lead to more immune activation and better response to immunotherapy. We had previously observed that high tumor mutational burden was only associated with increased immune activation in certain cancer types and were concerned when the FDA approved an immunotherapy for any cancer with high tumor mutational burden based on a limited number of patients and cancer types.”
Specifically, pembrolizumab (Keytruda, Merck), an anti-PD-1 therapy, received FDA approval in June for patients with unresectable or metastatic tumor mutational burden-high solid tumors that progressed after prior treatment, and for whom no satisfactory alternative treatment options exist. The agency based the approval on results of the phase 2 KEYNOTE-158 trial; however, that trial did not include patients with breast, prostate, brain and certain other cancer types that typically have not responded to immune checkpoint blockade therapy, according to researchers.
McGrail and colleagues gathered data on more than 10,000 patient tumors included in The Cancer Genome Atlas and compared the association between tumor mutational burden status and tumor immunogenicity as measured by CD8 T-cell infiltration. They identified tumors with and without a strong correlation between tumor mutational status and T-cell infiltration and assessed associations of mutational burden with immunotherapy outcomes based on objective response rates (n = 1,551) and OS (n = 1,936).
Results showed an ORR to immune checkpoint blockade of 39.8% (95% CI, 34.9-44.8) for tumor mutational burden-high tumors — significantly higher than for those with tumor mutational burden-low tumors (OR = 4.1; 95% CI, 2.9-5.8) — for cancer types with a correlation between CD8 T-cell responsiveness and neoantigen load, including melanoma, lung and bladder cancers.
Conversely, for certain cancer types without a correlation between CD8 T-cell responsiveness and neoantigen load, including breast and prostate cancers and glioma, the ORR was 15.3% (95% CI, 9.2-23.4) for tumor mutational burden-high tumors, which was significantly lower for tumor mutational burden-low tumors in this category (OR = 0.46; 95% CI, 0.24-0.88).
Retrospective analyses across various DNA sequencing approaches, the use of different immune checkpoint inhibitors and variations in clinical outcomes reported across cohorts served as limitations of the study, researchers noted.
“As expected, in cancers where high tumor mutational burden corresponded with immune activation, such as melanoma and lung cancer, we also found that high tumor mutational burden could accurately predict response to immunotherapy,” McGrail said.
“However, for many common cancers, such as those of the breast, brain or kidney where tumor mutational burden was not associated with immune activation, we found high tumor mutational burden failed to predict response to immunotherapy,” he added. “We hope these analyses will help inform how to best utilize the new FDA immunotherapy approval for high-tumor mutational burden tumors, and prompt further studies to validate and refine our observations.”
The researchers are now assessing biomarkers to predict response to immunotherapy for the cancer types in which tumor mutational burden lacks predictive value.
“At a more fundamental level, we are also seeking to understand why biomarkers for immunotherapy response diverge between cancer types and leverage this knowledge to improve immunotherapy approaches for the treatment of cancer,” McGrail said.
For more information:
Daniel J. McGrail, PhD, can be reached at The University of Texas MD Anderson Cancer Center, 6565 MD Anderson Blvd., Houston, TX 77030; email: djmcgrail@mdanderson.org.
Collapse
Read more about
Click Here to Manage Email Alerts