Trial Of Immune Drug In Breast Cancer Points To Need For Biomarkers

In In The News by Barbara Jacoby

By: Elaine Schattner


At first glance, results with a powerful immune agent in breast cancer appear negative. In an early-phase clinical trial, only 10% of patients with metastatic disease responded to atezolizumab (Tecentriq). Yet for those few women who did respond, the benefit was dramatic. Among 112 women with evaluable, advanced disease, 11 have experienced durable responses lasting at least one year. All of those (“100%”) who responded at one year remained in remission at two years; so far, the median duration of objective responses is 21 months.

Atezolizumab, an antibody to PD-L1 manufactured by Genentech, belongs to a growing class of medicines called immune checkpoint inhibitors. This drug has been approved by the FDA for use in lung and bladder tumors, but its use in breast cancer is experimental. This week, Dr. Peter Schmid of London presented preliminary results from the Genentech-sponsored study at the annual meeting of the American Association of Cancer Researchers (AACR) in Washington, D.C.

The finding is noteworthy because unlike some malignancies, breast cancer is not usually sensitive to immune treatment. Years ago, doctors recognized that melanoma, kidney cancer and some types of lymphoma occasionally wane in response to immune triggers. Approximately a third lung cancers may respond to immune-activating drugs like Opdivo and Keytruda. But trials of these agents in breast cancer have been disappointing. Last December, a study of avelumab (Bavencio) presented at the San Antonio Breast Cancer Symposium showed a response rate below 5%. A year earlier, I wrote on Keytruda in triple-negative breast cancer, as reported at the 2014 meeting. Less than 20% of patients responded in that study, and it took doctors months to tell if the patients were benefiting from the drug.

What struck me about this trial of atezolizumab is that a few breast cancer patients really do benefit. The problem is, there’s no way to know in advance who those patients will be. The need for a biomarker is urgent, because trials of immune checkpoint inhibitors, including this study and the Keynote trials of Keytruda, continue to enroll desperate patients with few or no treatment options. A woman with metastatic cancer considering an investigational agent might prefer to try a different drug, or opt for palliation alone, if she knew her chances were only 1 in 10 of responding.

The study presented by Dr. Schmid includes 115 women with a hard-to-treat form of breast cancer called triple-negative disease. Triple-negative refers to a type of breast cancer that lacks estrogen receptors, progesterone receptors or elevated levels of Her2. Because this breast cancer form is not sensitive to estrogen-blocking agents, or to drugs like Herceptin or Perjeta that work through the Her2 molecule, women with this kind of breast cancer have few non-investigational treatment options apart from old chemotherapy drugs.

“This is an aggressive form of breast cancer,” Schmid emphasized. “There is no targeted therapy at the moment. Chemotherapy is the mainstay of treatment.” Most women with this form of metastatic breast cancer live only 9 to 12 months, he said. “That is the unfortunate reality.”

There is an unmet need for better treatments of this form of breast cancer. Most women in this non-randomized trial received numerous prior treatments that didn’t work; the median number of chemotherapy agents patients tried before joining this study was seven. Among patients enrolled, 65% of had metastases involving visceral organs such as the liver or lungs; 30% had bone metastases. The median age of the study participants was 53 years, with a wide range: ages 29 to 82.

Schmid directs the Breast Centre at St. Bartholomew’s Hospital and Barts Cancer Institute in London. His research includes development and application of experimental drugs and biomarkers in breast and lung cancer. In the presentation at AACR, he indicated that examining tumors for PD-L1, a protein that might predict responsiveness to a drug like atezolizumab, was inconclusive; levels did not predict responses to the drug. He also showed exploratory data for tumor-infiltrating lymphocytes and CD8+ T-cells in the tumor specimens; results for these biomarkers to predict immune responsiveness are encouraging, he suggested. But he showed no definitive data for a biomarker.