Rational design of a bispecific ADC targeting ITGB4 and TROP2 for co-expressing solid tumors

Integrin beta 4 (ITGB4) was identified as a promising tumor-specific antigen through the GNOCLE™ platform. ITGB4 is highly expressed in head & neck, colorectal, esophageal, and other solid tumors, but shows minimal expression in matched normal tissues by immunohistochemistry, supporting a favorable therapeutic window. ITGB4 promotes epithelial detachment, migration, and invasion, and its high expression is associated with therapy resistance and poor prognosis. Target profiling further revealed the frequent co-expression of ITGB4 and TROP2 in multiple cancers, including pancreatic, lung, bladder, and HER2-negative breast cancers, with minimal overlap in normal tissues, supporting the tumor-selective potential of ITGB4/TROP2 co-targeting. TROP2-targeting antibody-drug conjugates (ADCs) have been actively developed, with several already approved for the treatment of solid tumors. However, relatively high TROP2 expression in normal epithelia can drive on-target/off-tumor toxicity, thereby restricting the therapeutic index of TROP2-targeting ADCs. This highlights an unmet need for strategies that preserve the antitumor benefit of TROP2 targeting while improving tumor selectivity. Leveraging the complementary biology of ITGB4 and TROP2, we designed a bispecific ADC targeting ITGB4 and TROP2 to preferentially deliver cytotoxic payloads to dual-positive cancer cells, thereby increasing therapeutic effect and expanding the therapeutic window. To construct the bispecific antibody, we combined the variable region of sacituzumab with our humanized anti-ITGB4 antibody, GENA-120B17. Guided by target affinity and relative expression, we generated 1+1 and 2+2 valency formats to optimize efficacy and selectivity. The bispecific antibodies showed enhanced internalization in TROP2/ITGB4 double-positive cancer cells compared with sacituzumab or GENA-120B17, without increased uptake in target-negative cancer cells or normal cells. These bispecific antibodies were then conjugated, via a hydrophilic, cleavable linker (LinkerE) designed to reduce retro-Michael elimination, to either a topoisomerase I inhibitor or a microtubule inhibitor. The resulting bispecific ADCs exhibited superior in vitro cytotoxicity in double-positive cancer cell lines relative to the corresponding monospecific ADCs, while maintaining a safer profile in normal cells. In conclusion, this bispecific ADC design broadens the therapeutic index and offers a safer, more effective treatment approach that addresses key limitations of existing TROP2-targeted ADC therapies.