The landscape of oncology is undergoing a seismic shift, driven by the convergence of generative artificial intelligence and precision medicine. Mesothelioma—a notoriously aggressive and difficult-to-treat cancer affecting the lining of the chest or abdomen—has long represented a significant "unmet need" for clinicians. Patients who experience disease progression after initial standard-of-care therapies are often left with few viable options. However, recent data from Insilico Medicine, a Cambridge, Massachusetts-based biotech firm, offers a new glimmer of hope.
At the European Society for Medical Oncology (ESMO) Congress held in Madrid on October 25, researchers unveiled the first human clinical trial results for ISM6331, a novel oral TEAD inhibitor designed entirely through generative AI. The drug, which targets the Hippo signaling pathway, demonstrated encouraging tumor-shrinking capabilities in a cohort of heavily pre-treated mesothelioma patients, signaling a potential leap forward for AI-discovered therapeutics.
The Core Data: Clinical Efficacy and Patient Outcomes
The ongoing Phase 1 dose-escalation study provides a vital snapshot of how this AI-designed molecule performs in a real-world clinical setting. Out of 22 evaluable patients with mesothelioma, four achieved a partial response, translating to an objective response rate (ORR) of 18.2%. Perhaps more significant is the disease control rate: 68.2% of patients saw their tumors either shrink or remain stable by the time of their first clinical scan.
The durability of these responses is also a point of interest for researchers. Three of the four observed responses were evident as early as week 8, at the first imaging milestone. Most notably, as of the August 14 data cutoff, three of these responses were ongoing, with one patient maintaining a positive trajectory for approximately one year—an impressive feat for a patient population that had already exhausted multiple previous lines of therapy.
The trial participants were a high-risk group, having received a median of two prior lines of therapy. Some responders had been treated with as many as five or six previous regimens, underscoring the potential for ISM6331 to provide efficacy where conventional drugs have failed. Melina Marmarelis, an oncologist at the University of Pennsylvania’s Abramson Cancer Center, presented these findings in Madrid, highlighting that at the highest dose tested (130 mg once daily), two out of five evaluable patients achieved a partial response.
Chronology: The Path to ESMO 2026
The journey of ISM6331 from a digital concept to a clinical candidate is a testament to the accelerated timelines made possible by generative AI.
- Discovery and Design: Utilizing its proprietary "Chemistry42" generative chemistry software, Insilico Medicine targeted the TEAD family of proteins. By blocking TEAD palmitoylation—a fatty-acid modification essential for protein activity—the company sought to inhibit the downstream signaling that drives tumor growth.
- Preclinical Validation: Before entering the clinic, the drug was rigorously tested for its ability to block all four members of the TEAD family, a design choice intended to prevent compensatory resistance mechanisms within the cancer cells.
- Clinical Enrollment: The Phase 1 dose-escalation trial (NCT06566079) commenced, recruiting patients with various solid tumors, including those with mesothelioma, regardless of their specific Hippo pathway mutation status.
- Data Presentation (October 2026): The findings were presented at the ESMO Congress, establishing the first clinical proof-of-concept for the drug.
- Future Milestones: Following the positive Phase 1 signals, Insilico has set its sights on a randomized Phase 2a dose-optimization trial, tentatively planned for 2027.
The Mechanism: Why Targeting the Hippo Pathway Matters
To understand the significance of ISM6331, one must look at the biological machinery of cancer cells. The Hippo signaling pathway is a critical regulator of organ size and tissue homeostasis. In many cancers, including mesothelioma, this pathway is dysregulated, often due to the loss of the NF2 gene. When the Hippo pathway fails, its downstream effectors—YAP and TAZ—become hyperactive, binding with TEAD proteins to turn on genes that promote uncontrolled cell proliferation and survival.
Insilico’s approach is to use ISM6331 as a "pan-TEAD" inhibitor. Unlike some targeted therapies that focus on a single protein isoform, ISM6331 blocks all four TEAD members. According to Halle Zhang, Insilico’s vice president of clinical development for oncology, this broad-spectrum inhibition is intentional: "They can compensate for one another," she explained, noting that inhibiting the entire family is the most effective way to shut down the oncogenic signaling of the Hippo pathway.
Safety Profile and Official Perspectives
A critical hurdle for any new oncology drug is the balance between efficacy and toxicity. As a class, TEAD inhibitors have been associated with proteinuria—a condition where excess protein leaks into the urine—which serves as an "on-target" effect.
Insilico’s clinical team has been monitoring this closely. Across all 36 patients treated in the trial (which included both mesothelioma and other solid tumor cases), four instances of treatment-related proteinuria were recorded. Importantly, these cases were mild (Grade 1 or 2) and reversible.
"Proteinuria is the on-target effect described for TEAD inhibitors as a class, so we monitor kidney function as dose increases," said Halle Zhang. Beyond renal concerns, three patients at the highest dose level experienced Grade 3 adverse events, including asymptomatic elevations in pancreatic enzymes, fatigue, and drug-related rashes. Investigators reported no dose-limiting toxicities across the 15–130 mg range, allowing for continued dose escalation to determine the optimal therapeutic window.
The Competitive Landscape
ISM6331 enters a highly competitive and evolving field. Vivace Therapeutics, a biotech based in San Mateo, California, has already made significant strides in this space with its own palmitoylation blocker, VT3989. In October 2025, Nature Medicine published results from a Phase 1/2 trial of VT3989, which showed a 26% response rate among 47 mesothelioma patients. Vivace has since moved to advance its candidate into a randomized Phase 3 trial, known as the "sTEADfast" study, to compare its efficacy directly against standard chemotherapy.
The presence of multiple players in the TEAD inhibitor space suggests that this mechanism of action is becoming a cornerstone of emerging oncology research. While Vivace’s data provides a benchmark, Insilico’s use of generative AI to identify and optimize its molecule early in the pipeline may provide unique advantages in terms of future scalability and molecule refinement.
Implications for the Future of AI in Medicine
The emergence of ISM6331 is more than just a win for mesothelioma patients; it is a proof-of-concept for the entire AI-driven drug discovery industry. For years, skeptics questioned whether algorithms could truly produce molecules capable of succeeding in the high-stakes, high-failure environment of human clinical trials. By successfully bringing an AI-designed molecule to the ESMO Congress and demonstrating clear clinical activity, Insilico has moved the conversation from "theory" to "demonstration."
As the company looks toward a 2027 Phase 2a study, the focus will shift from simple dose-escalation to biomarker identification. Insilico is currently analyzing tumor biology to determine which specific patient populations are most likely to respond to TEAD inhibition. This precision approach—matching the drug to the biology—is the hallmark of modern oncology and is further enhanced by the speed and predictive capabilities of generative AI.
The success of these early trials suggests that we are entering an era where drug discovery is no longer a slow, serendipitous process of trial and error, but a data-informed engineering discipline. While the road to regulatory approval remains long, the preliminary success of ISM6331 offers a promising blueprint for how technology can shorten the distance between a scientific hypothesis and a life-saving treatment. For patients battling mesothelioma, this progress represents not just a new drug, but a fundamental change in the tools available to combat their disease.
About the Author
