Corey J. Langer, MD, FACP: Unfortunately, in the past 20 years, we have yet to identify a reliable biomarker. There have certainly been a number of studies looking at predictive models for angiogenesis inhibition, both in the context of bevacizumab and ramucirumab, as well as a whole fleet of tyrosine kinase inhibitors that target angiogenesis. We are batting zero. VEGF selection and other markers have all failed to pan out, sadly. I don’t know whether we will ever find one. Remember, we are targeting the blood vessels, not necessarily the tumor. Finding a biomarker in that sort of setting may be far more elusive. There may be special synergy that’s going on between angiogenesis inhibition and chemotherapy that cannot be identified on a biomarker level. We do know that certain patients are destined to do poorly, at least with bevacizumab. This includes patients with squamous cell and folks who have baseline cavitating lesions. This has nothing to do with efficacy. It has everything to do with toxicity and the risk for hemorrhage. We do not have a reliable tissue-based biomarker. We have only some clinical biomarkers.
Not all that much has been identified regarding the mechanism of resistance with angiogenesis inhibitors, particularly with monoclonal antibodies. Intriguingly, we have studies in colorectal carcinoma where we have looked at continuing bevacizumab beyond progression in combination with other agents, compared with second-line treatment alone. We have actually observed a survival advantage. So, it may be wrong to empirically stop the angiogenesis inhibitor. That same question was asked in the context of the AvaALL trial in advanced non—small cell lung cancer. Of course, we were looking at the standard second-line regimens at that time, which were docetaxel, pemetrexed, and gefitinib, or erlotinib, and not immunotherapy. Sadly, that trial was negative. Intriguingly, there was some hint, not so much in the second-line but in the third- or fourth-line setting, that we see a progression-free survival benefit. But it did not translate into a survival advantage.
I would reproach that whole issue in the era of immunotherapy. I would have total equipoise for looking at second-line immunotherapy in combination with either bevacizumab or ramucirumab compared with immunotherapy alone. In fact, those trials are ongoing. There are phase I/phase II trials combining ramucirumab with pembrolizumab. There’s evidence that VEGF inhibitors can upregulate T-cell expression and VEGF, which can lead to immunotherapy resistance. Immunotherapy can reverse that and may potentially work in synergy with angiogenesis inhibition. Clinically, this may be panning out. The response rates in the phase I/phase II data for ramucirumab and pembrolizumab are about 35%. This is nearly double what we’d observe with immunotherapy alone. So, that regimen is now being looked at in the frontline setting, in individuals with 50% or higher expression. Clearly, this is an area of research that we need to consider.
I would strongly suspect that angiogenesis continues to be a major element in tumor growth beyond the initial therapy. We need to devise more creative ways of targeting angiogenesis in individuals whose disease has progressed on a combination of angiogenesis inhibition and chemotherapy. We haven’t figured that out yet. A few trials that have tried to address that have been negative. This is still an important area of research. Tumors go from dormancy to growth in the presence of neovascularization. If we can prune those vessels and deprive the tumors of growth elements that make them proliferate and metastasize, we’ll be way ahead of the curve.