6011 Background: Trastuzumab deruxtecan (T-DXd) is a HER2-directed antibody–drug conjugate with established efficacy across multiple HER2-expressing malignancies. Salivary gland carcinoma (SGC) is a rare disease with limited treatment options, particularly for tumors with HER2-low expression. The MYTHOS trial is a multicenter, investigator-initiated phase II study evaluating T-DXd efficacy in recurrent or metastatic (RM) SGC patients with HER2 overexpression or HER2-low expression. Here, we report the efficacy and safety results of the HER2-low cohort (Cohort 2). Methods: Eligible patients had histologically confirmed RM SGC with HER2-low expression (IHC 1+ or IHC 2+/ISH−), as determined by central assessment according to the ASCO/CAP 2018 breast cancer guidelines, and no indication for curative treatment. Patients received T-DXd at 5.4 mg/kg intravenously every 3 weeks. The primary endpoint was confirmed objective response rate (ORR) assessed by independent central review (ICR) according to RECIST v1.1. Secondary endpoints included disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and safety. The required sample size for Cohort 2 was 33 patients, based on a threshold ORR of 25%, an expected ORR of 50%, 83% power, and a two-sided alpha of 0.05 using an exact binomial test. The primary efficacy decision was based on the prespecified Simon’s two-stage minimax design in the first 33 patients; efficacy in the full analysis set (FAS) was summarized descriptively. Results: A total of 36 patients were included in the FAS, including 30 with salivary duct carcinoma (SDC). HER2-low status comprised IHC 2+/ISH− in 14 and IHC 1+ in 22; 25 had received prior systemic therapy for RM disease. Median follow-up was 25.1 months. The ORR by ICR was 38.9% (14/36; 95% CI, 23.1–56.5%), and the DCR was 94.4% (95% CI, 81.3–99.3%). Median PFS was 8.7 months (95% CI, 6.5–13.3), and median OS was 24.8 months (95% CI, 18.7–NE). In a prespecified subgroup analysis by histology, the ORR by ICR was 46.7% (14/30; 95% CI, 28.3–65.7%) in SDC and 0% (0/6; 95% CI, 0–45.9%) in other SGC subtypes. Common grade ≥3 adverse events (>10%) were neutrophil count decreased (36.1%), lymphocyte count decreased (19.4%), white blood cell count decreased (11.1%), and decreased appetite (11.1%). Drug-related interstitial lung disease (ILD)/pneumonitis occurred in 9 (25.0%; grade 1/2 in 7, grade 3 in 1, and grade 5 in 1). There was one drug-related death due to ILD/pneumonitis. Conclusions: Although the prespecified primary efficacy endpoint was not met, T-DXd demonstrated clinically meaningful antitumor activity in patients with HER2-low RM SGC, particularly in those with SDC. The safety profile was generally consistent with the known profile of T-DXd in the Japanese population, with ILD/pneumonitis remaining an important identified risk requiring careful monitoring. Clinical trial information: jRCT2011210017.
Kinoshita et al. (2026) studied this question.