The US Food and Drug Administration's use of pathologic complete response as regulatory endpoint: Did it pay off?

In 2013, the US FDA granted pertuzumab (Perjeta, Genentech/Roche) accelerated approval (AA) for the neo-adjuvant treatment of HER2 positive breast cancer. Approval was later extrapolated to the adjuvant setting by the NCCN and many providers. Pertuzumab remains the first and only drug to utilize an improvement in the rate of pathologic complete response (pCR), as the endpoint for regulatory approval. At the time the FDA utilized pCR as the basis for approval, their own analysis had suggested no correlation at the trial level between improvements in pCR and the more important clinical endpoints of recurrence and death. The phase III trial APHINITY served as the post-market commitment confirmatory trial for pertuzumuab, and recently led to conversion to regular approval in both the neoadjuvant and adjuvant setting. In that trial, pertuzumab drug showed a statistically significant 0.9% (93.2 → 94.1%) improvement in disease free survival (HR 0.81, 95% CI, 0.66–1.0, p = 0.045) without any improvement in overall survival (HR 0.89, 95% CI, 0.66–1.21, p = 0.47). Cardiac safety was prioritized in APHINITY as HER2 directed therapies are known for cardiac side effects. A primary cardiac event, defined as NYHA class III or IV heart failure and a drop of 10% in LVEF to below 50% or cardiac death, occurred in 17 patients (0.7%) in the pertuzumab group and 8 patients (0.3%) in the placebo group. Cardiotoxicity of pertuzumab will likely be higher in real world practice, as the drug becomes used is patients with greater comorbidity, which has been shown in real world data for traztuzumab. Therefore, because of the small benefit and potential harms of the drug, and the lack of survival benefit, pertuzumab may have a risk-benefit balance that will ultimately tip towards greater harm in practice. We believe that an impartial consideration of the evidence should give us pause before granting future AA based on improvements in pCR.