Biotech Industry · global
BigHat raises $75 million as AI-designed cancer antibodies move into human testing
BHB810, which targets gastrointestinal tumors including gastric cancer, has entered a Phase 1 trial, while another candidate is expected to begin human studies in 2027. The new funding will support clinical development and put AI-designed molecules to practical tests of efficacy and safety.
AI can help design antibodies, but whether it can turn them into safe, effective cancer drugs remains a question for human trials. BigHat Biosciences, a biotech company based in San Mateo, California, announced on September 24, 2026, that it had completed a $75 million Series C financing round. The funds will support its protein design platform and advance candidate therapies including BHB810 and BHB299 toward generating clinical data.
The round was co-led by DFJ Growth and Premji Invest, with participation from Catalio Capital Management, LG Technology Ventures, and existing investors. Transaction data from S&P Capital IQ published via MarketScreener also confirmed that the financing closed on September 24. For BigHat, the funding follows the entry of its first internally developed drug into clinical trials and will support the company's transition from molecular design to testing in patients.
The most advanced candidate is BHB810, an antibody-drug conjugate (ADC) targeting CDH17. This type of drug links an antibody that recognizes a target to a cell-killing drug, with the aim of delivering the drug to tumors. BigHat announced on September 1 that the first patient had been dosed. The Phase 1 trial is initially enrolling patients with advanced gastric cancer and gastroesophageal junction tumors to assess safety, tolerability, how the drug behaves in the body, and preliminary antitumor activity.
The company said BHB810 had produced complete or near-complete tumor clearance in nearly 30 patient-derived and cell-derived tumor models, including gastric cancer models with low or heterogeneous CDH17 expression. These results provided a basis for moving into human studies, but they are company-reported preclinical evidence and cannot be translated directly into tumor responses or survival benefits in patients. The financing announcement also provided no human efficacy or safety results.
AI's specific role in this development process is to help design and optimize antibodies, with experimental results then fed back into the models. According to BigHat's description of its Milliner platform, candidate molecules undergo automated experimental testing, and the resulting data are used in the next round of design. This creates an iterative process of computational prediction and experimental validation. However, the financing announcement did not disclose the size of the training dataset or provide comparative data sufficient to quantify AI's advantages over other design methods.
Another candidate therapy, BHB299, targets solid tumors expressing CEACAM6 and uses a T-cell engager design intended to direct immune cells to attack tumors. The company said its preclinical development was nearing completion, with human trials planned to begin in 2027. This remains a development timeline target and should not be interpreted as enrollment having begun.
The new funding gives BigHat the resources to expand its design platform and advance its drugs in parallel, but the outcomes of these efforts need to be assessed separately. Whether experimental data can improve molecular properties is a platform-level question; whether patients can benefit with acceptable side effects must be answered through clinical studies. As BHB810 enters human trials, the company's next most consequential evidence will be the safety and antitumor data it actually generates.