NATIONAL HARBOR, MD. – The immune checkpoint inhibitor nivolumab has promising activity in mismatch repair–deficient noncolorectal cancers, according to preliminary findings from the first sub-arm of the National Cancer Institute’s landmark Molecular Analysis for Therapy Choice (NCI-MATCH) trial.
NCI-MATCH is a 1,173-site precision medicine trial launched in 2015 to study targeted therapies for patients with relapsed/refractory solid tumors, lymphomas, and myelomas. In the first substudy (arm Z1D), the investigators identified 4,900 subjects with samples that could be tested for “actionable molecular abnormalities,” and from among those, they identified 77 with loss of mismatch repair proteins MLH1 or MSH2. Ultimately 47 patients were treated with nivolumab in the substudy.
Sharon Worcester/Frontline Medical News
Dr. Nilofer Azad
The confirmed overall response rate was 24%, and an additional 27% of patients had stable disease, said Dr. Azad of Johns Hopkins University, Baltimore.
The patients had a median age of 60 years and were heavily pretreated with a median of three prior therapies. The most common histologies among them were endometrioid endometrial cancer (10 patients), prostate cancer (6 patients), and breast cancer (3 patients).
The safety and tolerability of treatment was as expected for single-agent nivolumab treatment. Toxicity was predominantly low-grade fatigue. Anemia was the most common grade 3 toxicity.
“DNA repair defects due to mismatch repair–deficiency are most commonly caused by silencing of mismatch repair proteins MLH1 or MSH2 and, a little less commonly, MSH6 or PMS2. This can happen through DNA mutation, as well as promoter methylation,” Dr. Azad explained. “In fact, nivolumab has already been tested in patients with mismatch repair–deficient colorectal cancer, both alone and in combination with anti-CTLA-4 ipilimumab ... in addition, pembrolizumab was approved earlier this year for pretreated mismatch repair–deficient cancer.”
“So this formed the nidus for our interest and hypothesis that nivolumab would also have activity in mismatch repair–deficient noncolorectal cancer,” she said.
Study subjects had relapsed/refractory cancers, good end-organ function, and good performance status. They were screened for molecular alterations by centralized testing on fresh biopsy tissue, and mismatch repair deficiency was defined through immunohistochemistry as loss of nuclear expression of MLH1 or MSH2. Patients with mismatch repair–deficient colorectal cancer were excluded.
Those in the nivolumab arm received 3 mg/kg every 2 weeks, and after cycle 4, they could be switched to receive treatment every 4 weeks. Imaging was performed every 2 weeks, and patients were allowed to remain in the study as long as their disease had not progressed. A caveat was that patients with progression within the first 24 weeks, but with no more than four new lesions or 40% increase in tumor index lesions, could remain in the study as long as they were clinically stable.
The overall response rate was compared against a null value of 5%.
“We enrolled 35 patients so that we could have 31 evaluable patients, looking for a signal of 5 or greater responses in that patient group to conclude that the arm was promising and worth further testing,” Dr. Azad said. “This gave us 91.8% power to conclude that an agent was promising if the overall response was truly 25%.”
The study met its primary endpoint, with 8 responses out of 34 evaluable patients, she reported.
“Of note, we had five more patients that had unconfirmed responses. Two of those remained on study at the time of data cutoff, so these response numbers may change as the study matures,” she said.
The disease control rate was 56%, and benefit was seen across tumor histologies, she noted.
“The duration of benefit was compelling for these patients,” she said. “The median time to response was 2.1 cycles, and the 6-month progression-free survival was 49%.”
The median duration of response has not been reached.
Follow-up is ongoing, and 12 patients are enrolled in an expansion cohort; results should be reported within the next year.
“Future work includes interrogating tumor tissue and blood to identify possible predictive markers of response and resistance,” Dr. Azad concluded.
Dr. Azad reported having no disclosures.