FROM JAMA ONCOLOGY
after treatment with PD-1 and PD-L1 inhibitors. The findings could supplement other biomarkers, and suggest that chemotherapy could be avoided in some patients.
“We found a TMB value … of 19 mutations per megabase was a strong discriminator of response and nonresponse, and that corresponds to approximately the 90th percentile for TMB in our dataset. That is a higher threshold than has been previously proposed to be used for a TMB cutoff across different datasets or in lung cancer, but it did seem to be a strong discriminator of response, and that also translated into an improvement in progression free survival and overall survival in patients treated with immunotherapy,” said study coauthor Mark Awad, MD, PhD, in a podcast hosted by JAMA. He is a cancer researcher at Harvard Medical School, Boston. The research was published online in JAMA Oncology.
The value was reinforced when the team looked at deciles of TMB, from the lowest 10%, 20%, up to 90%. “It did seem like there was an inflection point, but only at the really higher levels of TMB – above the 80th, or especially the 90th percentile for TMB. That’s where it seemed to make a big difference in terms of improvements in response rate, progression-free, and overall survival,” Dr. Awad said.
The values of TMB and levels of PD-L1 expression also interacted in a useful way. “If you’re looking at PD-L1 on one axis and TMB on the other, it does seem that higher PD-L1 and the higher TMB can really identify patients with strong and great outcomes to immune checkpoint inhibitors. By contrast, low PD-L1 and low TMB really identifies patients that are not likely to benefit from immunotherapy alone and obviously might need to escalate it or more intensified therapy,” he said.
The results could help inform clinical decisions, though Dr. Awad included a caveat that the study was retrospective. In particular, patients with high TMB levels who might not tolerate chemotherapy well could be candidates for immunotherapy alone, “if you feel like there would be time to try immunotherapy alone rather than chemoimmunotherapy, and hopefully spare or avoid some of the chemotherapy toxicities, with the understanding that you wouldn’t want a patient’s disease to rapidly progress. You have to choose these cases carefully,” he said.
Dr. Awad suggested that TMB can be used alongside other factors such as PD-L1 mutations, KRAS mutation status, STK-11, and KEAP1 mutations. “I think all of these features will start to tip the scales one way or the other in terms of using immunotherapy alone or immunotherapy in combination with chemotherapy, and hopefully as new trials are developed, TMB and other predictive biomarkers can be used to stratify populations within a trial to hopefully ensure balance between treatment arms, and also to identify cancers that are less likely to respond to immune checkpoint inhibitors, such that we can really develop more tailored regimens for patients that will or won’t be as likely to respond to immunotherapy.”
The study included 1,552 patients with advanced NSCLC, with a median age of 66; 53.5% were women. The median TMB was 9.82 mutations per megabase. The researchers categorized patients as low TMB (fewer than 19 mutations per megabase) or high TMB (19 or more mutations). The high TMB group associated with better outcomes after treatment with PD-1/PD-L1 inhibitors, including overall response rate, progression-free survival, and overall survival. The associations occurred in the discovery cohort as well as two other independent cohorts. The same relationship occurred in all PD-L1 tumor proportion score subgroups.
Patients with NSCLCs with high TMB as well as PD-L1 expression of 50% or higher had an overall response rate of 57% and had the longest PFS and OS with ICI treatment (18.1 months and 47.7 months, respectively). On the other hand, patients with low TMB and PD-L1–negative NSCLC had the lowest ORR at 8.7% and the shortest PFS and OS (2.1 months and 10.4 months, respectively).
Dr. Awad has consulted for Achilles, AbbVie, Neon, Maverick, Nektar, and Hegrui. He has received grants and personal fees from Genentech, Bristol-Myers Squibb, Merck, AstraZeneca, and Lilly. He has received personal fees from Maverick, Blueprint Medicine, Syndax, Ariad, Nektar, Gritstone, ArcherDx, Mirati, NextCure, Novartis, EMD Serono, and NovaRx.