What became immediately apparent was that there were very big differences in how these tumors responded to immunotherapy depending on their common mutations. Tumors that had common mutations in the KL subgroup exhibited primary resistance to immunotherapy. The likelihood of responding was only 7.4%. On the other hand, tumors in the KP subgroup had a 35.7% likelihood of responding to immunotherapy. This is more than 5-fold compared with the KL subgroup. This suggests that we should be looking at what happens with mutations together with KRAS.
In addition to differences of objective response rates (ORR) there were statistically significant differences in progression-free survival and OS. The median OS of the KP subgroup that responded to immunotherapy was more than double that of the median OS of the KL subgroup. This suggests that it adds an extra parameter to what we believe should be a composite predictive biomarker panel to determine benefit from immunotherapy. PD-L1 expression plays a role and will continue to have one.
There are emerging data about tumor mutational burden as a predictive biomarker for response to immunotherapy. We believe that those 2 parameters should be coupled with mutations or genetic alterations. There are 3 individual genes that we are particularly interested in, which are KRAS, p53,
. By adopting an integrative approach to biomarker development, we can achieve greater predictive power.
What is the promise of combinations with immunotherapy in the field of lung cancer overall?
The KEYNOTE-021 data suggested that the combination of pembrolizumab with platinum-doublet chemotherapy of carboplatin/pemetrexed was associated with a 55% ORR. This is almost double that of chemotherapy alone, which is in the region of 25% to 30%.
What was particularly interesting was that the benefit of combining chemotherapy to immunotherapy seemed to extend to PD-L1–negative tumors. These are tumors where we would not use immunotherapy as first-line treatment. Those are the more challenging tumors. By combining chemotherapy with immunotherapy, we are achieving an additive effect. We do not know whether it is synergistic, but at least there is an additive effect. This is a treatment option for patients whom monotherapy with a PD-1 inhibitor is not currently indicated.
What else is important to note about genomic testing?
We would advocate for broad genomic profiling in every patient with lung adenocarcinoma that walks through the clinic. Some of these genetic alterations are already linked to FDA-approved therapies, and those should continue to be tested.
However, there is a broad panel of emerging biomarkers. For example, with MET
exon 14 skipping mutation, the HER2
mutation, or RET
rearrangements, there are no FDA-approved drugs in lung cancer. Yet, there are drugs [that target these mutations] that have been approved in other disease settings. Even if we cannot offer something as standard of care, we can at least direct patients into appropriate clinical trials.
Overall, broad genomic profiling in addition to determining PD-L1 expression status should be the standard of care for patients with stage IV disease that walk through the clinic.
- Skoulidis F, Byers LA, Diao L, et al. Co-occurring genomic alterations define major subsets of KRAS -mutant lung adenocarcinoma with distinct biology, immune profiles, and therapeutic vulnerabilities. Cancer Discov. 2015;5(8): 860–877. doi: 10.1158/2159-8290.CD-14-1236.
- Borghaei H, Paz-Ares, L, Horn L. Nivolumab versus docetaxel in advanced nonsquamous non–small-cell lung cancer. N Engl J Med. 2015;373:1627-1639. doi:10.1056/NEJMoa1507643.