Checkpoints Show Different Outcomes in Breast Cancer

SAN ANTONIO – The PD-L1 checkpoint inhibitor atezolizumab (Tecentriq) plus chemotherapy came up short in getting women with early stage triple-negative breast cancer to a pathologic complete response in the neoadjuvant setting when compared with chemotherapy alone, researchers reported here.

In the preliminary results of the NeoTRIPaPDL1 Michelangelo trial, 43.5% of patients achieved a pathologic complete response when they were taking atezolizumab and the chemotherapy regimen of carboplatin and nab-paclitaxel compared with 40.8% of those taking only the chemotherapy (P=0.66), reported Luca Gianni, MD, of Ospedale San Raffaele, Milan, Italy.

The Italian group recruited 280 patients into their study of patients diagnosed with triple-negative breast cancer (HER-2 negative, estrogen receptor negative, and progesterone receptor negative) with early stage, high-risk disease, or locally advance disease. Triple-negative disease is particularly difficult to treat because of its aggressiveness and a paucity of good drugs for it. Most current therapies for breast cancer are aimed at one of the targets lacking in triple-negative disease, leaving broadly cytotoxic agents as the principal options.

Gianni did show in his study that patients who had PD-L1 positive tumors fared better with atezolizumab therapy than women who had PD-L1 negative tumors. Women with a high level of PD-L1 were twice as likely to achieve a pathologic complete response if they were in the atezolizumab group of patients.

"Continuous follow-up for the primary endpoint of event-free survival and other efficacy endpoints is ongoing, and molecular studies are under way," Gianni said.

At a press conference at the annual San Antonio Breast Cancer Symposium, Gianni noted that the results were preliminary and did not include the primary endpoint of event-free survival. But the failure to differentiate from placebo for pathologic complete response stood in contrast to another study at the session in which the PD-1 checkpoint inhibitor pembrolizumab (Keytruda) clearly showed superiority in getting triple-negative breast cancer patients to pathologic complete response status in the neoadjuvant setting – considered a major step in eliminating disease entirely.

In the KEYNOTE-522 trial, presented by Peter Schmid, MD, PhD, of Barts Cancer Institute, London, 64.8% of the triple-negative breast cancer patients achieved pathologic complete response with pembrolizumab plus two chemotherapy regimens compared with 51.2% of patients treated with chemotherapy alone. Main results of the trial were presented at the European Society for Medical Oncology earlier this year.

Schmid said that in the placebo-controlled study, pembrolizumab was associated with event-free survival of 91.3% at 18 months compared with 85.3% with chemotherapy alone. In the trial, 784 patients were assigned to receive pembrolizumab and 390 were in the chemotherapy-only arm of the trial.

Schmid also said that treatment with pembrolizumab achieved numerically better results when the patients were stratified by disease stage, lymph node involvement, and PD-L1 expression level. He noted that even in patients who were PD-L1 negative, there was still a trend towards higher rates of pathologic complete response with pembrolizumab.

Both Gianni and Schmid suggested the differences in outcomes could have been due to the type of chemotherapy selected, differences in patients, or perhaps the difference between the two agents (their mechanisms are not exactly the same) in dealing with earlier treatment stages.

But Edith Perez, MD, of the Mayo Clinic in Jacksonville, Florida, told MedPage Today that the failure of atezolizumab to do better in getting patients to pathologic complete response "was very surprising. I would have expected to at least have seen a trend toward an advantage with atezolizumab that would have been consistent with the data we saw for pembrolizumab. Based on this surrogate endpoint, this really has to be considered a negative trial." Perez did not participate in either trial.

She dismissed the argument that the chemotherapy regimen was at fault. "I think there may be a problem with the [immunotherapy] agent. PD-1 agents such as pembrolizumab may work better in this population with chemotherapy while the PD-L1 agent atezolizumab may not do as well in triple-negative breast cancer along with chemotherapy," Perez suggested.

Although she acknowledged that pathologic complete response wasn't the primary endpoint for the atezolizumab trial, Perez suggested that the pathologic response rates in both studies should at least lean in the same direction.

She said that atezolizumab appeared to show activity in the metastatic setting, and was surprised that it did not show similar activity in the early stage setting.