Photo Courtesy © PER/Claudio Papapietro 2013
Naiyer A. Rizvi, MD
The future of PD-1 and PD-L1 inhibition in non-small cell lung cancer (NSCLC) is bright, with ongoing studies suggesting that the strategy will lead to a “new world” in the treatment of the disease, according to a presenter at the 8th
Annual New York Lung Cancer Symposium, held in New York City November 9.
Nivolumab and three other immune checkpoint inhibitors being studied in the clinic are all demonstrating similar promising activity, explained the presenter, Naiyer A. Rizvi, MD, a thoracic oncologist at Memorial Sloan-Kettering Cancer Center (MSKCC) in New York. Furthermore, Rizvi said, researchers are beginning to gather data indicating that PD-L1 expression is a biomarker for success with the drugs.
“For these patients, the single-agent response rate to second-line chemotherapy with docetaxel, for example, is 8%, and the duration of response is a few months, so we don’t have a lot of good options for advanced-stage lung cancer,” Rizvi said. “The initial data with nivolumab and other compounds is that patients who have been heavily pretreated have a response rate of over 20%, with 24% alive 2 years later, which is pretty remarkable. The responses we’ve seen with our patients have been pretty dramatic, like nothing we’ve seen before. There’s no question it’s going to change the landscape of how we treat lung cancer.”
Rizvi said excitement arose about the use of PD-1 and PD-L1 agents in the treatment of NSCLC when a study of nivolumab demonstrated durable responses in half of 129 patients, with higher doses of the drug bringing better results.
Nivolumab and another IgG4 antibody, MK-3475, are still being explored in the clinic, as are MPDL3280A and MEDI4736, both IgG1 antibodies, Rizvi said. All have reached phase III development except for MEDI4736, which is being investigated in a phase I trial.
The drugs target PD-1, an inhibitory T-cell co-receptor, or PD-L1, a ligand expressed on the tumor. It is thought that interactions between PD-1 and the ligands PD-L1 or PD-L2 can lead to antitumor immune suppression; the compounds in development are designed to interrupt those interactions, allowing T cells to fight cancer.
Updated results from the expansion cohort of the phase I trial of PD-1 antibody nivolumab (CA209-003) included results for 129 patients with previously treated, advanced NSCLC. For those who took 3 mg/kg of nivolumab, the overall response rate (ORR) was 24.3 % and median survival was 14.9 months, Rizvi said. Many responses occurred within the first 8 to 16 weeks, he said.
According to an analysis presented at the 2013 World Conference on Lung Cancer in October, at a median 20.3-month follow-up, overall survival (OS) rates were 42% at 1 year and 24% at 2 years across all cohorts of NSCLC patients in the study, regardless of nivolumab dose. In that same population, median OS was 9.9 months, based on Kaplan-Meier estimates.
Rizvi said that responses continue in 10 of 22 remaining patients. He added that, among patients who discontinued the trial for reasons other than disease progression, six of the remaining seven continued to respond. These results raise questions about how long patients need to take the drug, he said.
“If you reset your immune system, if you turn your T cells back on, do you need to keep dosing with this drug?” he asked. “We don’t really know the answer to that.”
Additional study participants benefited from nivolumab, but were not included in data as positive responders because their benefit was delayed or they experienced initial inflammation of their tumors that mimicked progression, Rizvi pointed out.
About 5% to 10% of patients can be expected to exhibit that reaction, but an indication of eventual outcome is that those who will eventually respond tend to feel better while on nivolumab, he said.
A clue about biomarkers came in a separate analysis presented at the World Conference on Lung Cancer, which looked at PD-L1 expression and its association with clinical activity in the 129 patients from the phase I trial. The patients who tested positive for PD-L1 expression—about half the group—benefited more from nivolumab, although PD-L1-negative patients also responded.
There were few grade 3 or 4 adverse events, with pneumonitis occurring at a rate of 6%, and diarrhea, rash, and hepatic dysfunction also occurring, Rizvi said.
Currently, the drug is being tested in phase I-IV trials of patients with advanced NSCLC, either as monotherapy; in combination with drugs including ipilimumab and erlotinib; or in comparison with docetaxel, Rizvi explained. The trials are in populations unselected for biomarkers, although tissue samples are being collected in the phase II trial and evaluated for PD-L1 expression and drug response, he said.
Future trials will test the combination of nivolumab and ipilimumab in lung cancer, the doctor said.
“We learned from melanoma data with ipilimumab and nivolumab that the response rate was nothing short of dramatic—a 13% complete response rate,” he said. “PD-L1-negative melanoma patients have a worse response rate to nivolumab than PD-L1-positive patients, but when you give both nivolumab and ipilimumab, PD-L1 status doesn’t matter.”
MK-3475, another PD-1 antibody, has demonstrated similar efficacy to nivolumab in an ongoing phase I trial (NCT01295827) of 38 patients with squamous or non-squamous NSCLC, Rizvi said. In the trial, the drug has demonstrated an ORR of 24% (9 patients) by investigator review and 21% (7 patients) by independent review.
Patients with high levels of PD-L1 expression had an ORR of nearly 70%, while patients with lower rates of expression experienced a lower response rate, the doctor said.
Rizvi noted that patient performance status didn’t seem to affect response, meaning that the drug may be appropriate even for frail or elderly patients. There have been no grade 3-5 adverse events in the trial, he said.
The trial has now expanded to include more than 1000 patients, with some cohorts containing strictly PD-L1-positive patients with pretreated NSCLC.
Meanwhile, another phase I trial is testing the drug in combination with chemotherapy in advanced NSCLC, and a phase III trial is comparing MK-3475 against docetaxel in previously treated patients with PD-L1-positive NSCLC, Rizvi said.
In a phase Ia trial the PD-L1 antibody MPDL3280A demonstrated an ORR of 23% in all participants; response improved with higher PD-L1 positivity, Rizvi said.
The drug had little toxicity, and those results hinted that PD-L1 compounds might cause less pneumonitis than PD-1 compounds, the doctor said.
Investigators also found that smokers or former smokers, who had a 26% ORR to the drug, fared better than never-smokers, who had a 10% ORR, Rizvi said.
“Smoking status is an interesting sub-analysis of the trial, and may play a role in how we select patients for this drug and how we think about lung cancer,” he said. “If patients have a large mutational burden, there is a lot of antigen release, and they are more likely, over time, to turn on an adaptive immune response that will allow for tumors to escape. Patients who are smokers or ex-smokers fall into that category, and their tumors have a more inflammatory component. Those patients may be more likely to respond to these agents.”
PD-1/PD-L1 compounds are being investigated in small-cell lung cancer, as well, Rizvi noted, and a number of trials of other novel immune targets and combinations are being conducted at MSKCC.
<<< View coverage from the New York Lung Cancer Symposium