Abstract Background: Keynote-522 administers four cycles of Carboplatin + Paclitaxel (Carbo/P) together with pembrolizumab (Pembro), followed by Doxorubicin/Epirubicin + Cyclophosphamide (EC). In routine care, some centres invert the order, initiating therapy with EC. Whether sequencing influences delivery of the complete regimen or the likelihood of achieving a pathologic complete response (pCR) is uncertain. MethodsElectronic drug-administration logs from all patients with stage I-III TNBC treated with Pembro-EC-Carbo/P between January 2022 and July 2024 were analysed. Each EC or Carboplatin infusion was counted as a macro-cycle (maximum four each), and every Paclitaxel dose was grouped into twelve weekly micro-cycles, yielding a theoretical maximum of twenty cycles. Carboplatin administrations were classified as high-dose (area under the curve, AUC ≈ 5) or low-dose (AUC ≈ 1.5). pCR was defined as ypT0 or documented “pCR”. Pathogenic germline variants (pV) and clinical stage at diagnosis were retrieved from the hospital registry. Group comparisons used the Mann-Whitney U and Fisher’s exact tests (α = 0.05). ResultsSixty-five patients met inclusion criteria (median age 50 years; stage I 10%, II 46%, III 44%). Forty-four began with EC (EC-first) and twenty-one with Carbo/P (Carboplatin-first). EC-first patients completed a median of 18 cycles compared with 15 cycles in Carbo-first (p = 0.004); 73% versus 38% finished phase-2 Carboplatin, and 46% versus 24% received at least 18 cycles. High-AUC Carboplatin exposure (median four infusions) and cumulative Paclitaxel dose (2,300 mg versus 2,250 mg) were similar. The pCR rate was 48% (21/44) for EC-first and 38% (8/21) for Carboplatin-first (p = 0.46). Documented BRCA1/2 (pV) or other pVs were present in 8% of patients; their pCR rate was 57% compared to 43% in patients without a pV. Therapy discontinuation occurred in 14% of both cohorts, most commonly due to hematologic toxicity or disease progression. ConclusionsCommencing pembrolizumab-based neoadjuvant therapy with EC was associated with superior regimen completion and a numerically higher, though not statistically significant, pCR rate compared with starting with Carbo/P, without an increase in early termination. Of note, there were considerably more stage III patients than in the trial population—potentially harder to sterilise—which may account for the lower pCR rates seen. These real-world findings support prospective evaluation of an EC-first sequence and emphasise the importance of maintaining dose intensity in everyday practice. Citation Format: J. Wagner, I. Vasconcelos, J. Blohmer. Sequencing Matters: Impact of EC-first versus Carboplatin-first Ordering on pCR in Triple-Negative Breast Cancer Real-World-Data abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS5-05-15.
Wagner et al. (Tue,) studied this question.