Background: TAK-164 is a second-generation ADC comprising a human IgG1 monoclonal antibody targeting GCC conjugated to a DNA-damaging alkylating agent by a peptide linker. TAK-164 demonstrated cytotoxic and antitumor activity in GCC-expressing cells and xenograft mouse models. This first-in-human study investigated the safety, pharmacokinetics (PK), and preliminary efficacy of TAK-164. Methods: Adult pts with GCC-positive, advanced/metastatic GI cancers received TAK-164 intravenously on day 1 of a 21-day cycle (Q3W). Dose escalation proceeded based on cycle 1 safety data via a Bayesian model of modified toxicity probability interval starting at 0.004 mg/kg. Results: Thirty-one pts were enrolled. Median age was 58 years (range 32–72), 58.1% of pts were female and 64.5% had colon carcinoma. The median number of prior lines of therapy was 4 (range 2–9). TAK-164 was given at 0.004 (n = 1), 0.008 (n = 1), 0.016 (n = 1), 0.032 (n = 5), 0.064 (n = 7), 0.12 (n = 7), 0.16 (n = 2), 0.19 (n = 3), 0.25 (n = 3) and 0.32 mg/kg (n = 1). No pts had dose-limiting toxicities (DLT) in cycle 1 up to 0.32 mg/kg. Three pts had adverse events (AEs) after cycle 1 considered to be DLTs: 1 pt receiving 0.19 mg/kg (grade 3 pyrexia and grade 5 hepatic failure) and 2 pts receiving 0.25 mg/kg (1 pt had grade 3 nausea, and grade 4 platelet count decrease and neutrophil count decrease; 1 pt had grade 4 hepatic failure and grade 4 platelet count decrease). Dosing was capped at 0.19 mg/kg due to hepatic toxicity and the recommended phase 2 dose (RP2D) was determined as 0.064 mg/kg based on safety and tolerability beyond cycle 1. Overall, pts received a median of 2 (range 1–8) treatment cycles. TAK-164-related treatment-emergent AEs (TEAEs) reported in 77.4% of pts included platelet count decrease (58.1%), fatigue (38.7%), and anemia (32.3%). TAK-164-related grade ≥3 TEAEs reported in 32.3% of pts included platelet count decrease (12.9%), alanine aminotransferase increase, aspartate aminotransferase increase, fatigue, and anemia (all 9.7%). Three pts discontinued due to TAK-164-related TEAEs. There was a dose-dependent increase in TAK-164 maximum plasma concentration and exposure over the range 0.016–0.32 mg/kg, with no meaningful accumulation in PK with repeat Q3W dosing. One pt receiving TAK-164 0.19 mg/kg showed γH2AX induction via immunohistochemistry in a post-treatment biopsy, demonstrating target engagement. One pt with low baseline GCC expression who received 5 cycles of TAK-164 0.008 mg/kg had an unconfirmed partial response at cycle 4; 11 of 25 (44.0%) evaluable pts had a best overall response of stable disease. Conclusions: TAK-164 appeared to have a manageable safety profile up to 0.064 mg/kg in pts with advanced GI cancers; hepatic toxicity was identified as a potential risk. The RP2D was determined as 0.064 mg/kg but was considered insufficient to derive significant clinical benefit. Clinical trial information: NCT03449030