Abstract PD8-01: Phase 3 SOPHIA study of margetuximab (M) + chemotherapy (CTX) vs trastuzumab (T) + CTX in patients (pts) with HER2+ metastatic breast cancer (MBC) after prior anti-HER2 therapies: Final overall survival (OS) analysis

Background: CTX + dual HER2-targeting monoclonal antibodies (mAb) remains a standard of care for treatment (Tx) of both HER2+ early-stage and MBC. However, when the SOPHIA trial was launched, limited Tx options existed after progression on T, pertuzumab (P), and ado-trastuzumab emtansine. M, an Fc-engineered anti-HER2 mAb, targets the same epitope as T and exerts similar antiproliferative effects. M enhances CD16A-mediated ADCC compared to T. Furthermore, M treatment is associated with increased HER2-specific T- and B-cell responses and increased T-cell clonality compared to baseline. The phase 3 SOPHIA (NCT02492711) study demonstrated PFS benefit of M vs T, both + CTX, in HER2+ MBC pts. M improved PFS over T, with a 24% relative risk reduction (HR .76; 95% CI .59-.98; P=.033; median, 5.8 [95% CI 5.5-7.0] months (mo) vs 4.9 [95% CI 4.2-5.6] mo (Rugo HS, et al. JAMA Oncol 2021), resulting in FDA approval. Median OS after 270 mortality events (2nd interim analysis) was 21.6 mo with M vs 19.8 mo with T (HR .89; 95% CI .69-1.13; P=.33). Here we report final OS after 385 events, as well as updated safety.Methods: Pts with disease progression after ≥2 lines of anti-HER2 Tx, including P, and 1-3 lines of Tx for HER2+ MBC were randomized 1:1 to CTX + either M (15 mg/kg) or T (8 mg/kg loading dose, then 6 mg/kg), both given IV every 3 weeks. Randomization was stratified by number of metastatic sites (≤2, >2), lines of Tx for MBC (≤2, >2), and CTX choice (capecitabine, eribulin, gemcitabine, or vinorelbine). Sequential primary end points were central-blinded review of PFS and OS.Results: The intent-to-treat (ITT) population comprised 536 pts (M, 266; T, 270). At the median follow-up of 20.2 mo among all ITT pts, pts received a median of 7 cycles of M + CTX vs 6 cycles of T + CTX. Median OS after 385 events in the ITT population was 21.6 mo with M vs 21.9 mo with T (HR .95; 95% CI .77-1.17; P=.62; Table). Based on a prespecified, non-α-allocated exploratory analysis, a numerical OS advantage in favor of the M arm was observed in the subgroup of pts homozygous for the CD16A-158F low-affinity allele (median OS, 23.6 vs 19.2 mo; HR .72; 95% CI .52-1.00; nominal P=.05). In contrast, in the small subgroup of CD16A-158V homozygotes, median OS was longer for T vs M (31.1 mo vs 22.0 mo; HR 1.77; 95% CI 1.01-3.12; nominal P=.04). Grade ≥3 adverse events (AE) occurred in 146 pts (55.3%) receiving M vs 141 pts (53.0%) receiving T. Serious AEs were seen in 47 pts (17.8%) receiving M vs 51 pts (19.2%) receiving T. Incidence of infusion-related reactions was higher with M (36 [13.6%]) vs T (9 [3.4%]). Left ventricular dysfunction requiring delay or cessation of M/T administration occurred in 4 pts (1.5%) receiving M and in 7 pts (2.6%) receiving T. Conclusions: The median OS in the ITT population was not statistically different between the 2 arms. An exploratory analysis of CD16A genotyping indicates a numerical OS advantage in favor of M in F homozygous pts, along with a numerical OS advantage in favor of T in V homozygous pts. Safety of M + CTX was similar to previous reports and consistent with M FDA-approved label. Studies of M in HER2+ breast cancer pts with different CD16A allelic variants are warranted, including MARGOT, the neoadjuvant investigator-initiated study on the efficacy of M vs T in pts carrying F-allelic variants of CD16A. Median OSITT analysisPrespecified, non-α-allocated exploratory analysis (n=506)N=536CD16A-158F carriers (F/F and F/V)(n=437)CD16A-158F homozygotes (F/F)(n=192)CD16A-158F/V heterozygotes(n=245)CD16A-158V homozygotes (V/V)(n=69)aM + CTX, mo21.6 (n=266)23.3 (n=221)23.6 (n=102)21.3 (n=119)22.0 (n=37)T + CTX, mo21.9 (n=270)20.8 (n=216)19.2 (n=90)22.0 (n=126)31.1 (n=32)HR (95% CI)0.95 (0.77-1.17)0.86 (0.69-1.08)0.72 (0.52-1.00)0.96 (0.71-1.30)1.77 (1.01-3.12)P value0.620.19b0.05b0.78b0.04bAbbreviations: CTX, chemotherapy; HR, hazard ratio; ITT, intent to treat; M, margetuximab; mo, months; OS, overall survival; T, trastuzumab. Cutoff date: June 14, 2021. aThis subgroup was characterized by an imbalance in poor prognostic features. bNominal P value. Citation Format: Hope Rugo, Seock-Ah Im, Fatima Cardoso, Javier Cortes, Giuseppe Curigliano, Antonino Musolino, Mark D. Pegram, Thomas Bachelot, Gail S. Wright, Cristina Saura, Santiago Escrivá-de-Romaní, Michelino De Laurentiis, Gary N. Schwartz, Timothy Pluard, Francesco Ricci, William Gwin, III, Christelle Levy, Ursa Brown-Glaberman, Jean-Marc Ferrero, Maaike de Boer, Sung-Bae Kim, Katarína Petráková, Denise A. Yardley, Orit Freedman, Erik H. Jakobsen, Einav Nili Gal-Yam, Rinat Yerushalmi, Peter A. Fasching, Emily Ashley, Shengyan Hong, Minori Rosales, William J. Gradishar. Phase 3 SOPHIA study of margetuximab (M) + chemotherapy (CTX) vs trastuzumab (T) + CTX in patients (pts) with HER2+ metastatic breast cancer (MBC) after prior anti-HER2 therapies: Final overall survival (OS) analysis [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD8-01.