Corey J. Langer, MD, FACP: Personally, I believe there’s still a role for bevacizumab in select patients despite the astounding survival advantages seen with KEYNOTE-189. There are patients who aren’t necessarily candidates for immunotherapy. They may have underlying autoimmune disease or have had an organ transplant—kidneys, liver, heart. They’re ill-suited. There are some patients with less significant immune-related disorders, like Graves thyroid disease, in whom I would certainly consider immunotherapy. For individuals with serious collagen vascular disease, I would much sooner give bevacizumab in combination with pemetrexed/carboplatin or paclitaxel and carboplatin. In addition, there are individuals with a more marginal performance status, 1.5 or 2, in whom I would much sooner give bevacizumab with chemotherapy. Those individuals were explicitly excluded from KEYNOTE-189 and virtually all other immunotherapy trials.
Finally, KEYNOTE-189, as well as KEYNOTE-024 and KEYNOTE-042, explicitly excluded individuals with molecular aberrations. As a result, bevacizumab, which has actually shown some benefit in combination with TKIs in, for instance, patients with EGFR mutations, would be my drug of choice. I would far sooner give that in combination with chemotherapy when those individuals become TKI refractory. So, bevacizumab, although I’m clearly using it less, still has a major role up front in treatment-naïve advanced non—small cell lung cancer.
ECOG 4599 stipulated maintenance treatment with bevacizumab beyond the initial 6 cycles with paclitaxel/carboplatin. It was given during chemotherapy and was continued beyond. In truth, we’ve never really isolated the benefit of maintenance therapy with bevacizumab, but there was a very interesting landmark analysis from that trial that looked at folks who had gone through all 6 cycles, both in the control arm and in the bevacizumab arm, and compared the outcomes in those who started maintenance. There may be confounding variables, but in that maintenance group, beyond the first 6 cycles (that 18- or 20-week cut point), there was an additional survival advantage. In my mind, that has validated the use of bevacizumab maintenance. For many years, that was the standard of the single-agent bevacizumab. But with the introduction of pemetrexed/carboplatin and pemetrexed/carboplatin/bevacizumab, we actually started doing combination maintenance—pemetrexed with bevacizumab.
There is a clinical trial that has not reported yet that is asking this question. I believe it finished accrual almost 2 years ago. Individuals who are candidates for angiogenesis inhibition get a baseline regimen of paclitaxel/carboplatin and bevacizumab—4 cycles. Then they are randomized to maintenance therapy with bevacizumab alone, switch maintenance with pemetrexed, or the combination. I think that will be the definitive trial in terms of addressing whether we truly see a survival advantage for the combination. In my practice, in individuals who have been started on bevacizumab with pemetrexed, I’ve generally continued both, together, as maintenance. For pemetrexed, we have clear-cut prospective randomized phase III studies that show a survival advantage, so I’m not ready to abandon that. Again, pemetrexed has been used as maintenance treatment in the context of KEYNOTE-189.