Optimizing Perioperative Treatment Strategies in TNBC

Transcript:

Tiffany A. Traina, MD: In early stage triple-negative breast cancer, there has been quite a shift into the utilization of neoadjuvant therapy, even for patients with node-negative disease. In my practice, for patients with a tumor of about 2 cm and node-negative or node-positive disease, we often would consider the use of preoperative chemotherapy—often anthracycline and taxane-based therapy—prior to surgery. There have been a couple of studies now that have reported a compelling pathCR [pathologic complete response] advantage with the addition of a checkpoint inhibitor to that preoperative therapy. There are data from the I-SPY study and others that have suggested improvement in pathCR.

We do know that failure to achieve a pathCR is associated with a poor prognosis, and we also have the ability to incorporate postsurgical therapy, such as adjuvant capecitabine for 6 months, in those highest-risk patients with a subsequent improvement in overall survival. I think where we need to be a bit cautious is whether improvements in pathologic complete response translate to improvement in event-free survival at a later date.

We've just begun to see some preliminary data with the use of pembrolizumab as well as atezolizumab at the most recent San Antonio Breast Cancer Symposium in 2019. For example, there are some clear differences between the KEYNOTE-522 study and the Michelangelo trial. The Michelangelo trial offered the addition of atezolizumab to nab-paclitaxel and carboplatin as compared to nab-paclitaxel and carboplatin alone.

Patients went on to have surgery and then they received anthracycline on the back end. The primary end point of that study is event-free survival. That's an important study that was powered to show survival, although those data have not yet been presented. A secondary end point was pathologic complete response, and it was shown that atezolizumab did not appear to increase pathologic complete response in the overall intention-to-treat population.

When you look at the subset of patients who had PD-L1 [programmed death-ligand 1]—positive disease based on the SP142 assay, the pathologic complete response rate was a bit higher, about 52% compared to 48%. Overall, those patients with PD-L1–positive disease performed better than those who were PD-L1 negative. Again, we are uncertain about the incremental addition of the checkpoint inhibitor in terms of pathCR.

Really, my takeaway message is that while these data are encouraging for using pathCR as a primary end point, we really need to see the impact on event-free survival and balance risk-benefit when we're analyzing what could potentially be long-term immune-related adverse events, such as endocrinopathies, balanced by the ultimate long-term survival impact.

I think in medical oncology, our paradigm has often been investigating novel agents in advanced disease and then gradually moving those agents up as they show promise. Checkpoint inhibitors are following that same pathway. We have compelling data from metastatic triple-negative breast cancer with the incorporation of a checkpoint inhibitor to backbone chemotherapy.

We've now started to see neoadjuvant data suggesting benefit in terms of pathCR and perhaps event-free survival, but we await final data. It naturally follows to begin to explore using these agents in the adjuvant setting to see if we can cure more women of this disease. One trial going on right now is IMpassion131, which is a randomized study looking at a backbone of anthracycline and taxane-based chemotherapy with randomization to a year of atezolizumab or not.

These studies are now powered for long-term survival, and that is ultimately what we're trying to help achieve for our patients. Those studies are important, and we look forward to seeing those results. I am certainly hopeful that we can be able to refine which patients will benefit from these therapies and know with whom we should escalate treatment. Perhaps we can find a group of women who do not need escalation of therapy and can use a simpler approach.

Transcript Edited for Clarity