Impact of carboplatin schedule on pCR in a modified KEYNOTE-522 regimen with dose-dense AC for triple negative breast cancer.

[BACKGROUND] As triple-negative breast cancer (TNBC) has a poorer prognosis when compared to hormone- and human epidermal growth factor 2-positive disease, it is vital to find treatments and schedules to improve outcomes inpatients with TNBC. The purpose of this study is to compare pathologic complete response (pCR) rates in early-stage TNBC between weekly versus every three-week carboplatin dosing with paclitaxel and pembrolizumab followed by dose-dense doxorubicin and cyclophosphamide (ddAC) and pembrolizumab in a modified KEYNOTE-522 regimen.

[METHODS] A retrospective, single-center review was conducted on patients who received both treatment and surgery at the James Cancer Hospital at The Ohio State University Medical Center (The James) for TNBC with carboplatin, paclitaxel, and pembrolizumab followed by ddAC and pembrolizumab.

[RESULTS] A total of 92 patients were included in this study; 51 patients received weekly carboplatin and 41 patients received every three-week carboplatin. The pCR rate at time of surgery was 47.1% in the weekly group and 70.1% (P = 0.03) for the every three-week group. Dose reductions of chemotherapy (35.3% vs. 26.8%) and dose delays of greater than 7 days due to immunotherapy toxicities (23.5% vs. 14.6%) were greater in the weekly cohort. There were no differences in grade 3 or higher neutropenia between groups nor the use of granulocyte colony stimulating factor support. Infusion hypersensitivity reactions (iHSR) occurred in 19.6% of weekly patients and 4.9% of every three-week patients (P = 0.06).

[CONCLUSION] In this single-center analysis, every three-week carboplatin dosing followed by ddAC in a modified KEYNOTE-522 regimen provides higher pCR rates at the time of surgery. Patients who received weekly carboplatin dosing experienced more iHSR. These findings strongly warrant additional studies to determine the relationship of carboplatin dosing to TNBC outcomes.