Comparing clinical benefit of trastuzumab deruxtecan (T-DXd) and sacituzumab govitecan (SG) in a large cohort of HER2-negative metastatic breast cancer (MBC).

Abstract

1076 Background: T-DXd and SG are antibody-drug conjugates (ADCs) increasingly used in HER2-negative BC, however, there are insufficient data to guide ADC sequencing and use in tumors of various HER2 expression levels. Methods: A total of 4033 HER2 negative MBC treated with SG or T-DXd that underwent tumor profiling at Caris Life Sciences (Phoenix, AZ) were studied. HER2 low (Her2-L), ultra low (Her2-UL) and null (Her2-N) were tested by IHC and CISH. Hormone receptor status (HR+/-) was tested by ER and PR IHC. Real-world clinical data were obtained from insurance claims. Time on treatment (TOT) was determined as the interval from start to end of the ADCs. Cox proportional hazards model was used for hazard ratio (HR) and log-rank for p values. Results: Overall, 1444 (36%) were treated with T-DXd but not SG (T-only) while 1808 (45%) with SG but not T-DXd (S-only). HR+ cases comprise 64% of T-only and 24% of S-only cohorts; 75% and 68% of T-only and S-only tumors were taken from metastatic sites. As expected, HER2-L, HER2-UL and HER2-N cohorts treated with T-DXd had decreased TOT (4.8 months (m), 4.1m and 3.5m, p< .001) while HER2 status had no impact on SG TOT (3.0m, 2.8m and 3.4m). Interestingly, even in HER2-N group, T-only showed a borderline better TOT than S-only (Table, p=.053); this effect was significant in HR+ HER2-N subset but not significant in HR-HER2-N. Similarly, in HER2-UL and L, T-DXd TOT was significantly longer than SG TOT; when further stratified by HR status, the effect was highly significant in HR+ and not seen in HR- cohorts. In cohorts crossed over from one ADC to another, patients treated with T-DXd first (N=420) or SG first (N=361) showed no TOT difference (10.4m vs. 10.8m, p= .4); although the HER2-N subset had moderate preference of SG first (11.5m vs. 8.5m, HR=0.66 [0.52-0.84], p< .001, while TOT were similar in HER2-UL (HR=0.93, p=.7) and HER2-L (HR=1.04, p=.7) groups. Conclusions: We report outcome from a large real world dataset and demonstrate that T-DXd shows statically significant improved outcome in HR+ tumors across HER2 subgroups while in TNBC, both agents exhibit comparable benefit. In patients treated with both ADC’s, SG first showed preferred outcome in HER2-N group but not in HER2-UL or L. We provide important insight on clinical benefit of the two widely used ADCs in breast cancer and warrants further validation in independent cohorts. T-DXd (T-only) TOT and SG TOT (S-only) in cohorts stratified by HER2 and HR status. All HR+ HR- TOT (T;S, months) N (T;S) HR [95% CI] p TOT N HR p TOT N HR p HER2-N 4.7; 3.4 262; 1116 1.1 [1-1.3] 0.053 4.8; 3.0 209; 277 1.5 [1.2-1.8] <0.001* 4.6; 3.5 48; 822 1.1 [0.8-1.4] 0.6 HER2-UL 4.8; 3.0 295; 289 1.4 [1.2-1.7] <0.001* 5.1; 2.5 245; 99 1.8 [1.4-2.2] <0.001* 3.2; 3.0 46; 188 1.1 [0.8-1.6] 0.4 HER2-L 4.9; 3.5 707; 244 1.2 [1-1.4] 0.011* 5.1; 3.1 595; 77 1.5 [1.2-1.9] <0.001* 4.2; 3.9 100, 164 1.1 [0.8-1.3] 0.7 *:Significant.