The second day of the American Society of Clinical Oncology (ASCO) conference kicked off with a massive, standing-room-only session devoted to the expanding role of checkpoint inhibitors in the treatment of numerous cancer types. Checkpoint inhibitors are drugs that “release the brakes” on the immune system, allowing it to mount a stronger attack on cancer. Three separate clinical trials, for three different cancers, were discussed in this well-attended event, aptly named, “Immunotherapy for Every Patient: Check Your Enthusiasm.”
First up was Dung T. Le, M.D., of Johns Hopkins, who gave an exciting talk on the role of a tumor’s DNA mutations in determining response to checkpoint blockade therapy. There is increasing evidence that tumors with more DNA mutations respond better to this form of therapy. New mutations create novel molecular flags, called neo-antigens, which can be detected by the body’s patrolling immune cells. The research that Dr. Le presented lends further support to this idea by showing that tumors with a genetic defect in their DNA repair machinery, called mismatch repair (MMR) deficiency, are more responsive to treatment with anti-PD-1 therapy than tumors without this defect. MMR-deficiency leads to the buildup of DNA mutations in cancer cells, and this defect is found in several types of cancer, including one prominent form of colorectal cancer.
As Dr. Le discussed, the response rate of patients with MMR-deficient colorectal cancer to the PD-1-blocking antibody pembrolizumab (Keytruda®) was 62% compared to 0% for patients with MMR-proficient colorectal cancer. This large difference in response rate can be explained by the fact that tumors with MMR-deficiency have on average 1,700 mutations in their DNA, versus just 70 mutations for cancers without this defect. Based on these findings, a new clinical trial, called Keynote-164, is opening soon for patients with MMR-deficient colorectal cancer. According to Le, this is the “first study to use a tumor’s genetics to guide immunotherapy.” A paper describing the study was published today in the New England Journal of Medicine, and CRI Scientific Advisory Council member and former CRI investigator Drew Pardoll, M.D., Ph.D., is a co-author.
Next up was a talk by Anthony El-Khoueiry, M.D., of the University of Southern California, on the results of a clinical trial of the PD-1 checkpoint inhibitor nivolumab (Opdivo®) in liver cancer. The results of this phase I/II trial showed that 19% (8 of 42) patients with liver cancer responded to nivolumab, including 2 patients who had a complete response. Perhaps even more interesting, 48% of patients on study had stable disease, implying immune control—though not elimination—of their tumors. Many of these responses were also durable, lasting more than 1 year. And the toxicities associated with treatment were less than what is commonly seen with the standard of care therapy, a drug called sorafenib (Nexavar®).
There was good reason to suspect that liver cancer might respond to anti-PD-1 therapy. First, liver cancer is often associated with inflammation, which means that immune cells are already present in the tumor. Second, many cases of liver cancer are caused by infection with either the hepatitis B or C virus, which means they contain viral antigens that are recognizable by the immune system as foreign. And virally infected liver cells also express PD-L1, the binding partner for the PD-1 “brake” on T cells. The encouraging results of this phase I/II trial led Lawrence Fong, M.D., of UCSF and CRI’s clinical trials network, to conclude that liver cancer is “yet another difficult-to-treat cancer” where PD-1 therapy is proving effective.
Immunotherapy for Lung Cancer
By far, the biggest news of the day was the results of a large phase III trial of nivolumab (Odpivo®) versus the chemotherapy drug docetaxel in non-squamous non-small cell lung cancer (NSCLC). This is the most common form of lung cancer, and the biggest cancer killer both in the U.S. and around the world. The FDA approved nivolumab earlier this year for the treatment of squamous NSCLC, a rarer form of the disease. A phase III trial testing the drug in the more common, non-squamous type was stopped early, in April of this year, because the trial’s endpoints were met. That trial’s senior investigator was Julie Brahmer, M.D., of Johns Hopkins University School of Medicine, who is also member of the CRI-SU2C Cancer Immunology Dream Team, and today we learned the complete study results.
As presented by Luis Paz-Ares, M.D., Ph.D., of Hospital Universitario 12 de Octubre in Spain, patients receiving nivolumab showed improved survival compared to docetaxel—the standard therapy for advanced non-squamous NSCLC patients for whom first-line treatments have failed. Patients on nivolumab lived, on average, 3 months longer than those on docetaxel. That may not sound like a lot, but keep in mind that is an average across all 582 patients, and these are patients who saw their disease progress on prior therapy.
There were important lessons to glean in the details of the study. Most impressively, nivolumab nearly doubled overall survival for patients whose tumors were PD-L1 positive (more than 17 months compared to 9 months for docetaxel). This finding echoes the results of a phase I trial with another PD-1 blocking antibody, pembrolizumab (Keytruda®), in lung cancer that were presented at the annual meeting of the American Association for Cancer Research in April.
PD-L1 is a protein that binds to the PD-1 receptor on immune cells. Cancers often make PD-L1 as a way to ward off an immune attack. It makes intuitive sense that tumors making PD-L1 would respond better to anti-PD-1 therapy, since this drug acts to release the same brake that the cancer has engaged. But a big question in the field is whether having PD-L1 in your tumors (or not) is a useful “predictive biomarker” for whether you are a good candidate for anti-PD-1 therapy.
As Roy Herbst, M.D., Ph.D., of Yale School of Medicine, pointed out in his discussion of Paz-Ares’s talk, patients who were PD-L1-negative responded to nivolumab about as well as they did to docetaxel—but with far fewer side effects than this conventional chemotherapy. So clearly the PD-L1-negative patients are benefiting from anti-PD-1 therapy too. And there is no way yet to know if these patients will go on to have durable, long-term control of their disease, as a subset of patients on checkpoint blockade therapy do.
To the question running through everyone’s mind in the session—“Is nivolumab the new standard of care for previously treated, non-squamous NSCLC?”—Dr. Herbst answered decisively: Yes! This is obviously very exciting news for patients. We could be looking at FDA approval of this drug soon for this indication.
However, to the other main question—“Should we use the PD-L1 biomarker to determine patient selection?”—Herbst said, “No, not yet.” Though patients with more PD-L1 expression respond better, even patients without PD-L1 expression can respond to nivolumab. And determining which patients are PD-L1-positive can also be tricky, since tumors are intrinsically heterogeneous: one part of the tumor might be negative for PD-L1 and other might be positive.
Dr. Herbst ended the session by addressing where we go from here. He pointed to the need for more research to understand the biology of these tumors and their biomarkers. He also looked forward to studies that will bring PD-1 therapy into use for patients earlier in their treatment journey. That would be a fine day indeed, when we might be able to do away with chemotherapy altogether for lung cancer.
Check back tomorrow for updates from day 3 of ASCO, including results of an important clinical trial in melanoma.