作者
Jens Huober,C.H. Barrios,Naoki Niikura,Michał Jarząb,Y-C Chang,Shannon Leigh Huggins-Puhalla,J.L. Pedrini,L. Zhukova,Michael A. Curran,Daniel Eiger,Veronica Craine,Chiara Lambertini,E. Restuccia,H. Zhang
摘要
IMpassion050 (NCT03726879) showed no increase in pathological complete response (pCR; ypT0/is ypN0) rates for neoadjuvant A + PH + CT vs. placebo (pbo) + PH + CT in the intention-to-treat (ITT) or PD-L1+ populations in patients (pts) with high-risk HER2+ eBC. Safety was consistent with that of A in other combination studies. We report descriptive long-term efficacy and safety. Pts with a primary tumour >2 cm and histologically confirmed positive lymph node status were randomised 1:1 to A/pbo + PH + CT. After surgery, pts continued A/pbo + PH up to 1 yr. Pts with residual disease could switch to A/pbo + trastuzumab emtansine (T-DM1) for 14 cycles. Secondary endpoints: disease-free survival (DFS; time from surgery to disease recurrence or death); event-free survival (EFS; time from randomisation to disease recurrence, tumour progression or death); safety. Stratification: stage at diagnosis; hormone receptor (HR) status; PD-L1 status.Table: 127PDFS (N = 434 [217 per arm]; pts who had surgery)EFS (N = 454 [226 A, 228 pbo]; ITT population)3-yr event-free rate, % (95% confidence interval [CI])A: 92.9 (89.5, 96.4) Pbo: 88.5 (84.1, 92.8)A: 91.4 (87.7, 95.1)Pbo: 89.0 (84.8, 93.2)Hazard ratio (95% CI)0.71 (0.38, 1.32)Unstratified0.90 (0.50, 1.59)StratifiedCovariates, hazard ratio (95% CI)Stage at diagnosisT20.38 (0.13, 1.06)0.56 (0.24, 1.34)T3–T41.13 (0.50, 2.56)1.33 (0.60, 2.94)HR status+0.43 (0.17, 1.13)0.57 (0.24, 1.36)–1.07 (0.46, 2.53)1.28 (0.58, 2.82)PD-L1 status+1.38 (0.51, 3.69)1.67 (0.65, 4.32)–0.44 (0.19, 1.02)0.58 (0.27, 1.22)pCR statusNon-pCR0.71 (0.31, 1.63)0.95 (0.45, 1.98)pCR0.66 (0.26, 1.70)0.76 (0.30, 1.88) Open table in a new tab At data cut-off (24 Aug 2023), 411/454 pts (90.5%) remained on-study (A arm: 206/226; pbo arm: 205/228); median follow-up was 44.2 and 43.4 months, respectively. A median of 14 PH cycles was given in both arms. DFS/EFS, including subgroup analyses, are shown in the table. There were 24 deaths (A arm: 11; pbo arm: 13). In the adjuvant phase (A vs. pbo), more pts had adverse events of special interest (AESIs; 59.4% vs. 47.0%); there were no grade 5 AESIs; and the most common AEs (in ≥10% pts in either arm) were radiation skin injury (24.0% vs. 19.4%), arthralgia (20.7% vs. 17.1%) and diarrhoea (20.7% vs. 13.4%). Although there was no increase in pCR rates, 3-yr DFS/EFS rates were numerically improved with A in the ITT population, including in pts with PD-L1–negative, HR+ or stage T2 disease. Safety was consistent with the known profiles of PH, T-DM1 and A. These results highlight the value of long-term clinical endpoints when investigating early-stage cancer immunotherapies.