Immunotherapy-chemotherapy combination may reduce recurrence of triple-negative breast cancer
Click Here to Manage Email Alerts
BARCELONA, Spain — The addition of pembrolizumab to neoadjuvant chemotherapy significantly increased the pathologic complete response rate among patients with early-stage triple-negative breast cancer, according to results of the randomized phase 3 KEYNOTE-522 study presented at European Society for Medical Oncology Congress.
In addition, an interim analysis of EFS showed a benefit among patients who received pembrolizumab (Keytruda, Merck). This preliminary finding suggests the addition of immune therapy to neoadjuvant chemotherapy prevents breast cancer recurrence, according to Peter Schmid, MD, PhD, FRCP, clinical director of St. Bartholomew Breast Cancer Centre in London and lead of the Centre for Experimental Cancer Medicine at Barts Cancer Institute.
“The [13.6 percentage-point] difference we see in pathologic complete response is highly statistically significant but also hugely meaningful clinically, because pathologic complete response is linked to better longer-term outcomes,” Schmid told HemOnc Today. “I’m very optimistic that this will translate into a significant benefit in EFS over time.”
Triple-negative disease is the most aggressive breast cancer subtype. It accounts for about 15% of breast cancer cases and disproportionately affects young women.
Neoadjuvant chemotherapy followed by surgery offers the greatest chance for pathologic complete response, which leads to cure in 85% to 90% of cases. By comparison, up to half of women who have residual viable tumor tissue after surgery experience recurrence, typically within 3 years.
Results of the KEYNOTE-173 and I-SPY 2 trials showed the addition of neoadjuvant pembrolizumab — an anti-PD-1 monoclonal antibody — to chemotherapy resulted in promising antitumor activity and a manageable safety profile for patients with early triple-negative breast cancer.
In the placebo-controlled KEYNOTE-522 trial — the first phase 3 trial to evaluate immunotherapy for early breast cancer — researchers evaluated the neoadjuvant pembrolizumab-chemotherapy combination followed by adjuvant pembrolizumab.
The analysis included 1,174 patients with previously untreated, nonmetastatic, centrally confirmed triple-negative breast cancer. All patients had stage T1c N1-2 or T2-4 N0-2 disease.
Researchers assigned 784 patients to 200 mg pembrolizumab every 3 weeks in combination with neoadjuvant chemotherapy (four cycles of paclitaxel and carboplatin, followed by four cycles of doxorubicin or epirubicin plus cyclophosphamide). After definitive surgery, these patients received adjuvant pembrolizumab for up to nine cycles, or until unacceptable toxicity or disease recurrence.
The other 390 patients chemotherapy plus placebo.
Pathologic complete response — defined as ypT0/Tis ypN0, meaning clearance of the cancer from the breast after chemotherapy — and EFS served as dual primary endpoints.
“At the time we designed the trial, we weren’t sure if we’d see a difference in pathologic complete response,” Schmid said in an interview. “We hoped we would, but a lot of immunotherapy trials showed more benefit in time endpoints than in actual response. That’s why we had two endpoints: pathologic complete response and the equally important — or possibly more important — long-term endpoint of EFS.”
Secondary endpoints included OS, efficacy in the subset of patients with PD-L1-positive disease and additional measurements of pathologic complete response.
Median follow-up was 15.5 months (range, 2.7-25).
The pathologic complete response analysis included 602 evaluable patients.
Results revealed a statistically significant improvement in pathologic complete response rate with pembrolizumab plus chemotherapy in the entire population (64.8% vs. 51.2%; P = .00055), the subgroup of patients with PD-L1-positive disease (68.9% vs. 54.9%) and the subgroup of those with PD-L1-negative disease (45.3% vs. 30.3%).
Interim EFS analysis showed a favorable trend for pembrolizumab-treated patients (HR = 0.63; 95% CI, 0.43-0.93).
“It’s almost unheard of in adjuvant trials to see such an early separation of the curves,” Schmid said. “That is hugely encouraging because it shows this is likely to translate into the long-term benefit we hoped we’d see. It is early and many things can happen, but seeing that signal that early gives us all the confidence that the pathologic complete responses we’re observing with the addition of immune therapy are similar to the pathologic complete responses we see with chemotherapy alone and, therefore, likely will translate to long term benefit — which, in this disease setting, means cure.”
Investigators reported no new safety signals in KEYNOTE-522. They reported no significant differences between the pembrolizumab-chemotherapy and placebo-chemotherapy groups with regard to rates of grade 3 or higher treatment-related adverse events (78% vs. 73%) or death (0.4% vs. 0.3%).
Many of the adverse events were the result of intensive chemotherapy, Schmid said. Adverse events potentially associated with immune therapy occurred among 42% of patients assigned pembrolizumab and 21% of those assigned placebo; however, no new safety signals emerged, Schmid said.
“We don’t yet have 3 years, 5 years or 10 years of safety follow-up data and, although in other disease entities we have seen very few late toxicities, we do need to obviously follow patients for them,” Schmid said.
Translational research will continue to try to determine why certain patients did not achieve pathologic complete response.
“In the neoadjuvant setting, there are some patients you possibly overtreat and some who, unfortunately, are still not cured,” Schmid told HemOnc Today. “It’s possible that some patients may not need the addition of the immune checkpoint inhibitor, but it’s also possible that we can improve on our results. ...
“As it stands, we have taken all patients in a nonselected way, and we have a [13.6 percentage point] difference in pathologic complete response,” Schmid added. “That is a very big step forward, but we will do all of the additional analyses to understand the data even better.”
The results of this trial “could have a major impact on treatment” for patients with triple-negative breast cancer, according to Fabrice André, MD, PhD, professor of medical oncology at Institut Gustave Roussy in Villejuif, France.
“This is a good situation to test whether the FDA will approve a drug for triple-negative breast cancer based on [pathologic] complete response. The combination of ... pembrolizumab plus chemotherapy could become a standard of care if approved,” André, who was not involved with the study, said in an ESMO-issued press release.
Pembrolizumab’s effect on response rates was expected, but the low predictive value of PD-L1 expression proved surprising, André said.
“This is probably because most early-stage triple-negative breast cancers express some degree of PD-L1, sometimes below the predefined cutoff of positivity,” he said. “The role of chemotherapy as a sensitizer for anti-PD-1 in PD-L1-low early-stage triple-negative breast cancer should also be explored.
“The next step will be to define which patients are resistant and prioritize which targets should be hit in this population,” André added. “We also have to determine the impact of this new class of drug on survivorship issues.” – by Mark Leiser
Reference: Schmid P, et al. Abstract LBA8_PR. Presented at: European Society for Medical Oncology Congress; Sept. 27-Oct. 1, 2019; Barcelona, Spain.
Disclosures: Schmid reports honoraria from, consultant/advisory roles with, or research grants or funding to his institution from Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Celgene, Eisai, Genentech, Merck, Novartis, OncoGenex, Novartis, Pfizer, Puma Biotechnology and Roche. He also reports his spouse serves as a consultant to Genentech/Roche. Please see the abstract for all other authors’ relevant financial disclosures.
Perspective
Back to Top
Sherene Loi, MD, PhD
This is the first phase 3 neoadjuvant immunotherapy study to report, so this is very exciting.
Despite breast cancer not being considered an immunogenic solid tumor type, there has been recent approval of atezolizumab (Tecentriq, Genentech), the first checkpoint inhibitor to be approved in breast cancer for a PD-L1-positive population. We also have learned that, in advanced disease, we need to have enrichment of pre-existing immunity such as PD-L1 positivity for efficacy of these agents. In the advanced setting only 41% is PD-L1-positive in the first-line setting.
In KEYNOTE-522, most interesting is the data by PD-L1 status. PD-L1 is prognostic. If you are positive, you have a higher pathologic complete response compared with PD-L1-negative. The delta is 14.2%, so the increase is the largest pathologic complete response rate we have seen in triple-negative breast cancer. Interestingly, in the PD-L1-negative patients, the relative increase in pathologic complete response rate looks slightly better, so this is surprising given in the advanced setting it’s only the PD-L1 positive patients who benefit.
I think it is worth highlighting the difference between weekly vs. three-weekly schedule of carboplatin, where the relative increase in pathologic complete response does seem to be slightly better with weekly schedule. This raises questions in my mind about the chemotherapy regimen and if giving such myelosuppressive chemotherapy to patients with pre-existing immune response may be slightly detrimental for enhancing that pre-existing immune response. That is just a hypothesis, but it’s one I think we need to work out in the next generation of trials.
The EFS results are encouraging but it is still early and the data are too unstable. It is unclear whether a pathologic complete response rate increase with [an immuno-oncology agent] will translate to an EFS improvement.
I think it is important to highlight the rates of immune-related adverse events, including hypothyroidism (14.9%), hyperthyroidism (5.1%), adrenal insufficiency (2.7%), hypophysitis (1.8%), pneumonitis (1.9%), colitis (1.8%) and hepatitis (1.4%). These can be lifelong and require ongoing endocrine replacement. This is even more reason to wait for mature EFS data and to try to improve our selection of patients for this kind of treatment.
There is no doubt in my mind that in my mind that checkpoint blockade will help us better treat patients with early-stage triple-negative breast cancer; however, we will need to wait for longer and more mature EFS data and the other neoadjuvant studies coming in this space to understand the benefit, as well as the rate of lifelong adverse events.
We absolutely need to figure out dose and scheduling, particularly with the chemotherapy backbone. What chemotherapies are additive, synergistic or even detrimental? In the setting of high PD-L1 expression, this may allow us room for chemotherapy de-escalation with the addition of immunotherapy agents.
With regard to pembrolizumab, what is the best scheduling and duration. Is this window effect real, and do we really need one year, or perhaps we need more and we hope that biomarkers will help guide us in this way.
We await the biomarker analysis that will give us insights with regard to tumor composition, tumor mutational burden, germline BRCA status and triple-negative breast cancer subtypes. Also, do immune adverse events correlate with better outcomes? Who absolutely does not benefit, and who benefits the most?
Read more about
Click Here to Manage Email Alerts