A big part of this is agreeing on definitions and standardization of procedures so we can speak among different institutions and countries to compare trial outcomes.
—Thomas Herzog, MD
The immune checkpoints are really not a home run in ovarian cancer, based on some of these early results, suggesting that we’re going to need combinations, and we’re going to need to define patients who are [most likely] to benefit.
—Kunle Odunsi, MD, PhD
Ovarian cancer clinical trialists are forming a working group to develop a standard definition of pathologic complete response in ovarian cancer treated with neoadjuvant chemotherapy. Such agreement within the field potentially would enable the U.S. Food and Drug Administration (FDA) to consider accepting pathologic complete response as an endpoint to support accelerated approval of new therapies.
The effort is an outgrowth of a September 3 workshop on ovarian cancer endpoints sponsored by the FDA, ASCO, the American Association for Cancer Research, and the Society of Gynecologic Oncology.1
“A big part of this is agreeing on definitions and standardization of procedures so we can speak among different institutions and countries to compare trial outcomes,” Thomas Herzog, MD, Clinical Director of the University of Cincinnati Cancer Institute and one of the workshop organizers, said in an interview.
The field also has yet to define which ovarian cancer patients are most likely to benefit from neoadjuvant chemotherapy, rather than initial surgery followed by chemotherapy, Dr. Herzog said. “That determination would be best answered in a clinical trial, but a working group could formulate recommendations based on current evidence,” he said.
The working group would be led by some of the organizations involved in the workshop and would be open to participation by investigators, therapeutics sponsors, and patient advocates, Dr. Herzog said.
Low Bar to Overcome
The interest in standardizing pathologic complete response as an endpoint for ovarian cancer studies stems from its acceptance by the FDA in 2014 as an endpoint to support accelerated approval for neoadjuvant therapies for high-risk, early-stage breast cancer. In an October 2014 Guidance for Industry, the agency provided two acceptable definitions of pathologic complete response in early-stage breast cancer.
A new therapy for ovarian cancer wouldn’t have to improve pathologic complete response much to demonstrate an increase in activity. Most studies that have tracked it reported a pathologic complete response of less than 10%, Geoffrey Kim, MD, Director of the Division of Oncology Products 1 in the FDA Office of Hematology and Oncology Products, said in an interview.
“Due to the apparently low rates of [pathologic complete response] with standard regimens, it is a potentially low bar to overcome at this time in terms of assessing activity,” Dr. Kim said. “It was accepted at the workshop that in patients for whom neoadjuvant chemotherapy is indicated—and it would be very important to understand exactly who that population would be—this would be a potential platform to assess drug activity. Whether this activity (or how much of an increase in activity) is needed to predict long-term outcome is the key question.”
Recommendations for Development
Gwynn Ison, MD, Clinical Reviewer in gynecologic malignancies in the FDA Office of Hematology and Oncology Products, said a link between pathologic complete response in neoadjuvant treatment and long-term outcome in ovarian cancer has yet to be definitively established. She noted that a 2010 randomized trial in Europe suggested that neoadjuvant chemotherapy isn’t inferior to primary debulking surgery in ovarian cancer.2
Dr. Ison’s advice to sponsors: “Talk to the FDA early and often during the development process.”
The FDA doesn’t accept the CA125 test as a sole independent endpoint for approval, Dr. Kim said. He recommended that trials collect and report both CA125 and radiographic information. Other workshop participants suggested that every ovarian cancer trial collect data on a variety of potential biomarkers.
Susanna I. Lee, MD, PhD, Chief of Women’s Imaging at Massachusetts General Hospital and Associate Professor of Radiology at Harvard Medical School, Boston, described the extensive development of imaging biomarkers to predict response or patient outcome response in some ovarian cancer trials.
David M. Gershenson, MD, Professor, Department of Gynecologic Oncology & Reproductive Medicine, The University of Texas MD Anderson Cancer Center, Houston, advocated for trials of low-grade serous and clear cell ovarian cancer—a notoriously hard-to-treat form of the disease. Rare tumors present an opportunity to achieve larger benefits in smaller sample sizes of patients, he said.
Phenotypes Will Point the Way for Immunotherapy
Ovarian cancer can be stratified by the presence or absence of tumor-infiltrating lymphocytes and expression of programmed cell death ligand (PD-L1), said Kunle Odunsi, MD, PhD, Cancer Center Deputy Director, Chair of the Department of Gynecologic Oncology, and Executive Director of the Center for Immunotherapy at Roswell Park Cancer Institute, Buffalo, New York.
“Going forward with immunotherapy trials, it’s going to be important to stratify patients based on some of these phenotypes,” Dr. Odunsi said. “For example, if your patient has a tumor that is rich in tumor-infiltrating lymphocytes, the type of immunotherapy that you will select for that patient is likely to be a checkpoint blockade, because you want to rescue these T cells that are probably dysfunctional. So by using simple immunohistochemistry, you can begin to select patients and tailor therapy.”
Most studies exploring blockade of programmed cell death protein 1 (PD‑1) and PD-L1 in ovarian cancer have been very small, Dr. Odunsi said. The largest so far was a phase Ib study of avelumab in 75 patients, presented at the 2015 ASCO Annual Meeting by Mary L. (Nora) Disis, MD, medical oncologist at the Fred Hutchinson Cancer Research Center and the Seattle Cancer Care Alliance.3
PD-L1 positivity was not required for entry into the study. The objective response rate was 10.7% (95% confidence interval = 4.7%–19.9%) after a median follow-up duration of 5 months. The stable disease rate was 44%, and the disease control rate was 54.7%. The median duration of response was 21 weeks, with 62.5% of responses ongoing at the time data were reported.
“The immune checkpoints are really not a home run in ovarian cancer, based on some of these early results, suggesting that we’re going to need combinations, and we’re going to need to define patients who are [most likely] to benefit,” Dr. Odunsi said.
However, Sanjeeve Bala, MD, MPH, Medical Officer in the FDA Office of Hematology and Oncology Products, noted that in immunotherapy approvals in other cancers, the FDA considered the prolonged duration of response an important hallmark of clinical benefit.
Dr. Kim said a similar scenario is likely to play out in ovarian cancer. “People focus on how high that response rate number is, but with the immune checkpoint inhibitors, it’s really the duration of the response that’s impressive,” he said. “When patients respond, they respond for a long time. That’s what makes immunotherapy an exciting modality of treatment right now.” ■
Disclosure: Drs. Herzog, Kim, Ison, Lee, Gershenson, Odunsi, and Bala reported no potential conflicts of interest. Dr. Disis has received grant funding from EMD Serono, VentiRx, and Seattle Genetics and has stock in VentiRx and Epithany.
1. FDA Public Workshop: Ovarian Cancer Endpoints Workshop. Presentations available at http://www.fda.gov/Drugs/NewsEvents/ucm463933.htm. Accessed October 15, 2015.
2. Vergote I, Trope CG, Amant F, et al: Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer. N Engl J Med 363:943-953, 2010.
3. Disis ML, Patel MR, Pant S, et al: Avelumab (MSB0010718C), an anti-PD-L1 antibody, in patients with previously treated, recurrent or refractory ovarian cancer. 2015 ASCO Annual Meeting. Abstract 5509. Presented June 1, 2015.