ESMO Congress 2026

The evidence base for the B7-H3 ADC class currently centers on small-cell lung cancer: in separate phase III trials, the two anti-B7-H3 ADCs tambotatug pelitecan and risvutatug rezetecan each significantly improved overall survival, progression-free survival, and objective response rates versus topotecan ^[1]. A third trial, also presented at the 2026 World Congress on Lung Cancer, showed encouraging early efficacy for the PD-L1×VEGF-A bispecific antibody pumitamig plus the anti-B7-H3 ADC elfetabart drozuntecan in SCLC across lines of therapy and doses ^[1]. This is the starting point for interpreting the current readout: the combination's previously disclosed clinical evidence was limited to lung cancer.

What is being reported at ESMO26 is the Elfe-D + pumitamig data across solid tumors, described in the on-site post as the first clinical data beyond lung cancer. In the efficacy-evaluable population of N=139, uORR was 48.9%, cORR 39.6%, and DCR 87.8%; in melanoma (n=26), uORR was 30.8%, cORR 26.9%, and DCR 80.8%. The post also emphasized that activity was broadly maintained despite prior IO and/or bevacizumab exposure.

Relative to prior knowledge, this readout extends the combination's evidence from lung cancer to multiple solid tumors and suggests that activity is retained in previously treated patients, including those with prior antiangiogenic exposure, providing an early signal for this combination strategy beyond lung cancer. On safety, the post summarized it as manageable: treatment-related adverse events were mainly grade 1–2 GI and hematologic TRAEs, with no DLTs, no treatment-related deaths, and 2 investigator-reported ILD cases. The open question raised by the poster—whether the activity is driven by the IO effect, the antiangiogenic effect, or synergy between the two—has no answer in the current materials.

Several limitations should be noted: all figures come from on-site conference social media posts; tumor-type composition, lines of therapy, dosing, and design details were not disclosed (to be confirmed); the melanoma subgroup included only 26 cases, limiting the precision of this subgroup readout; and comparison with the phase III SCLC data in the background literature can only stay at the level that the class has phase III validation in SCLC while this is early data beyond lung cancer, without supporting cross-tumor efficacy inference.