XtalPi (2228.HK) today announced that its incubated portfolio company, Signet Therapeutics, has completed preclinical studies for SIGX2649, a potential first-in-class (FIC) or best-in-class (BIC) pan-TEAD inhibitor. Signet plans to submit simultaneous Investigational New Drug (IND) applications to regulatory authorities in both China and the U.S. Detailed preclinical data will be showcased at the 2026 American Association for Cancer Research (AACR) Annual Meeting.
SIGX2649 represents the second innovative pipeline asset co-developed by XtalPi and Signet to reach the clinical stage, following the clinical-stage gastric cancer drug SIGX1094. Under the partnership agreement, XtalPi is entitled to a significant share of future commercialization proceeds from the pipeline, reaching as high as double-digit percentages. This milestone further validates the efficiency of the “AI + Organoids” drug discovery paradigm and underscores XtalPi’s ability to consistently generate high-value clinical assets.
Addressing an Unmet Need in the Hippo Pathway
TEAD proteins are core components of the Hippo signaling pathway, a critical driver in the progression of solid tumors such as mesothelioma, liver cancer, and lung cancer. While the global solid tumor market is projected to exceed $532 billion by 2032, targeting the Hippo pathway remains a significant challenge. Current TEAD inhibitors face a delicate balance between potency and safety; currently, no TEAD inhibitors are approved globally, with the most advanced candidates only in Phase I trials.
SIGX2649 is a pan-TEAD inhibitor designed to overcome these hurdles. Preclinical data reveals a “dual-inhibition” mechanism: it effectively blocks all four TEAD isoforms while promoting the recruitment of the transcriptional co-repressor VGLL4. This differentiated approach results in:
Superior Antitumor Activity: Stronger anti-proliferative effects across multiple in vitro and in vivo models.
Enhanced Safety: Significantly lower renal toxicity compared to rival compounds.
Synergistic Potential: Remarkable efficacy when combined with RAS pathway inhibitors in KRAS-mutant solid tumors.
The Power of “AI + Organoids”
The discovery of SIGX2649 was powered by XtalPi’s AI + Robotics platform in synergy with Signet’s patient-derived organoid models. This integration allows for rational drug design alongside efficacy evaluations that closely mimic the human physiological environment.
To identify the candidate, XtalPi’s AI platform generated a library of millions of molecules. Using high-precision computational tools, the platform screened for activity, isoform selectivity, and developability, narrowing millions of candidates down to hundreds. These molecules were then validated using Signet’s organoid models. Through iterative optimization of ADMET properties using combined physical and AI models, SIGX2649 was ultimately selected as the Preclinical Candidate (PCC).
“The success of SIGX1094—which received a Prix Galien nomination—and now SIGX2649, demonstrates the clinical translatability and commercial scalability of our platform,” said a representative from XtalPi. “We remain committed to driving ‘source innovation’ and helping our partners accelerate the development of differentiated therapies for patients with high unmet needs.”
About Signet Therapeutics
Signet Therapeutics is a pioneer in the “Organoid + AI” drug discovery model. Founded at Harvard University and headquartered in Shenzhen, Signet focuses on developing first-in-class targeted therapies. Its lead asset, SIGX1094, the world’s first targeted therapy for diffuse gastric cancer, has received FDA Orphan Drug and Fast Track designations and is currently in Phase I clinical trials. Signet leverages disease models that mirror patient genomics to bridge the gap between laboratory research and clinical success.
