In an update of the PREDIX-HER2 trial, trastuzumab emtansine (T-DM1) remained equivalent to standard neoadjuvant chemotherapy plus dual-targeted HER2 therapy in producing pathologic complete remissions (pCRs) among patients with HER2-positive, metastatic breast cancer.
The new data also suggest tumor-infiltrating lymphocytes (TILs) and dramatic improvements in PET-CT scans can predict favorable outcomes in both treatment groups. Though these findings will be useful for research purposes, they likely won’t influence routine clinical practice.
, of Karolinska Institutet in Stockholm, from PREDIX-HER2 during the European Society for Medical Oncology: Breast Cancer virtual meeting.
PREDIX-HER2 included patients with HER2-positive breast cancer and tumor size greater than 20 mm or lymph node metastases.
Patients received neoadjuvant therapy (NAT) with docetaxel and trastuzumab plus pertuzumab (DTP) or T-DM1 every 3 weeks for a planned total of six courses. The protocol permitted switching to the competing treatment for progression, lack of response, or drug-related severe toxicity.
Postoperatively, all patients received triweekly epirubicin plus cyclophosphamide – four courses for the T-DM1 arm and two courses for the DTP arm. All patients then received triweekly adjuvant trastuzumab for 11 courses. The 62% of patients whose tumors were hormone receptor (HR)–positive received standard endocrine therapy postoperatively.
Updated results, predictors of pCR
At the 2019 ASCO annual meeting, PREDIX-HER2 investigators reported that, when compared with DTP, T-DM1 produced the same likelihood of pCR with less toxicity (). Updated data presented at ESMO Breast Cancer 2020 showed similar results.
The pCR rate was 45.5% in the DTP arm and 43.9% in the T-DM1 arm (P = .824). pCR rates were higher for HR-negative tumors – 63.6% in the DTP arm and 59% in the T-DM1 arm – than for HR-positive tumors – 36.4% in the DTP arm and 33.9% in the TDM-1 arm.
Three patients had disease progression with T-DM1, and none progressed with DTP. However, almost twice as many patients switched from DTP to T-DM1, compared with the other sequence.
Dr. Hatschek reported that the presence of at least 10% TILs predicted pCR in both treatment groups. Among patients who achieved a pCR, 52.2% had at least 10% TILs in baseline biopsies, and 30.4% had less than 10% TILs.
In addition, a decrease of FDG maximum standardized uptake value by more than 75% on protocol-required PET-CT scans was highly predictive of pCR. Among patients who achieved a pCR, 70.3% had a maximum standardized uptake value decrease of more than 75%, and 22.5% had a decrease of 75% or less.
At median follow-up of 28.5 months, event-free survival was similar between the treatment arms. Overall, there were 13 cases of progression, relapse, contralateral breast cancer, distant metastases, or death from any cause. There were five such events in the DTP arm and eight in the TDM-1 arm.
Dr. Hatschek concluded that neoadjuvant T-DM1 may be as effective as standard NAT in all clinical subgroups evaluated. Both TILs and PET-CT showed the potential to predict pCR and merit further study in the NAT setting.