Pembrolizumab monotherapy is associated with antitumor activity in patients with advanced recurrent ovarian cancer, interim results from the phase 2 KEYNOTE-100 study suggest.
Notably, objective response rates among study subjects increased in tandem with increased programmed death-ligand 1 (PD-L1) expression, which helps define the population most likely to benefit from single agent pembrolizumab (Keytruda), Ursula A. Matulonisreported during an oral abstractsession at the annual meeting of the American Society of Clinical Oncology.
Further, no new safety signals were identified, said Dr. Matulonis, medical director and program leader of the Medical Gynecologic Oncology Program at of Dana-Farber Cancer Institute and professor of medicine at Harvard Medical School, both in Boston.
All patients received intravenous pembrolizumab at 200 mg every 3 weeks for 2 years or until progression, death, unacceptable toxicity, or consent withdrawal, and tumor imaging was performed every 9 weeks for a year, then every 12 weeks thereafter until progressive disease, death, or study completion.
The overall response rate (ORR) among 285 patients in Cohort A, who had one to three prior chemotherapy lines for recurrent advanced ovarian cancer and a platinum-free or treatment-free interval of 3-12 months, was 7.4%, with mean duration of response of 8.2 months. The ORR among 91 patients in Cohort B, who had four to six prior chemotherapy lines and a platinum-free or treatment-free interval of at least 3 months, was 9.9%; the mean duration of response was not reached in Cohort B.
Among all-comers, the ORR was 8.0%, including 7 complete responses and 23 partial responses. Mean duration of response was 8.2 months, and 65.5% of responses lasted at least 6 months. Further, responses were observed across all subgroups, Dr. Matulonis said, noting that responses were seen regardless of age, prior lines of treatment, progression-free/treatment-free interval duration, platinum sensitivity, and histology.
“The one factor that did predict response was a [combined positive score] of 10 or higher, where there were more responses,” she said.
The ORRs among those with PD-L1 expression as measured using the combined positive score (CPS), which is defined as the number of PD-L1–positive cells out of the total number of tumor cells x 100, was 5.0% in those with CPS less than 1, 10.2% in those with CPS of 1 or greater, and 17.1% in those with CPS of 10 or greater (vs. the 8.0% ORR in the study), she explained, noting that all complete responses occurred in those with CPS of 10 or higher.
Grade 3-4 treatment-related adverse events occurred in 19.7% of patients, and included fatigue in 2.7%, and anemia, colitis, increased amylase, increased blood alkaline phosphatase, ascites, and diarrhea in 0.8-1.3%. One treatment-related death occurred in a patient with Stevens-Johnson syndrome, and another occurred in a patient with hypoaldosteronism. Immune-mediated adverse events and infusion reactions were most commonly hyperthyroidism and hypothyroidism, and most cases were grade 1-2, she said.
KEYNOTE-100 is an ongoing study that followed KEYNOTE-028, which demonstrated the clinical activity of pembrolizumab in patients with advanced ovarian cancer. To date, KEYNOTE-100 has enrolled 376 patients with epithelial ovarian, fallopian tube, or primary peritoneal cancer and confirmed recurrence after frontline platinum-based therapy. All had a tumor sample available for biomarker analysis.
The patients had a mean age of 61 years, 64% and 35% had performance status scores of 0 and 1, respectively, and 75% had high-grade serous disease.
Median follow-up in Cohort A at the time of the current analysis was 16.7 months, and in Cohort B, the median follow-up was 17.3 months. Treatment was ongoing in 15 and 6 patients in the cohorts, respectively. Reasons for discontinuation included radiographic progression (204 and 62 patients), clinical progression (24 and 17 patients), adverse events (22 and 3 patients), and patient withdrawal (9 and 3 patients). Complete responses occurred in 1 and 0 patients in the groups, respectively.
Median progression-free survival in both cohorts was 2.1 months, and overall survival was not reached in Cohort A, while it was 17.6 months in the more heavily pretreated Cohort B.
“Recurrent ovarian cancer is the leading cause of death from gynecologic cancer. The majority of our patients relapse after first-line platinum and taxane-based chemotherapy, and the degree of platinum sensitivity will predict the tumor response rates with platinum, as well as survival time,” she said, noting that subsequent recurrences become increasingly platinum and treatment resistant.
Current treatment options in these patients include chemotherapy with or without bevacizumab; the ORRs with single-agent immune checkpoint blockade are about 10%, but in KEYNOTE-028, patients with PD-L1–positive advanced recurrent ovarian cancer had an ORR of 11.5% with pembrolizumab treatment, she said.
“With 16.9 months median follow-up, the results confirm that pembrolizumab monotherapy in recurrent ovarian cancer elicits modest antitumor efficacy,” Dr. Matulonis concluded, noting that further analysis for biomarkers predictive of pembrolizumab response are ongoing.
Invited discussant Janos Laszlo Tanyi, MD, of the University of Pennsylvania, Philadelphia, said the findings underscore the overall modest ORRs of 5.9%-15% seen with anti-PD-1 or PD-L1 monotherapy in patients with advanced recurrent ovarian cancer, but noted the importance of the finding that the subpopulation of patients with increased PD-L1 expression may experience greater benefit.
Dr. Matulonis reported consulting or advisory roles with 2X Oncology, Clovis Oncology, Fujifilm, Geneos Therapeutics, Lilly, Merck, and Myriad Genetics, and research funding from Merck and Novartis. Dr .Tanyi reported having no disclosures.
To read this entire article by Oncology Practice in The Clearity Portal, please click here to login.