Insilico Medicine is heading to the clinic with its seventh pipeline candidate discovered and designed using generative artificial intelligence (AI), a treatment for methylthioadenosine phosphorylase (MTAP) deleted cancers, which account for 15% of all cancers.
The company is planning to launch a Phase I multicenter, open label study of ISM3412 following FDA approval of its IND application. The study will be designed to assess the drug in patients with locally advanced and metastatic solid tumors.
All patients will receive oral daily dosing of ISM3412 as a monotherapy, with expansion arms planned for the study across additional multiple indications. The trial’s primary endpoints will include safety, tolerability, recommended Phase II dose determination and preliminary efficacy.
Insilico has not set a date for data release from the study, but does have an idea about the types of cancers it plans to pursue.
“We are interested in going after MTAP-/- tumors in non-small cell lung cancer (NSCLC), mesothelioma, esophageal, bladder cancers etcetera,” Michelle Chen, PhD, Insilico’s chief business officer, told GEN Edge.
ISM3412 is a small molecule selective methionine adenosyltransferase 2A (MAT2A) inhibitor designed to target cancers with MTAP deletion, one of the most common genetic deletions. A study published February 13 in Journal of Clinical Pathology reported that MTAP deletion is seen in 15 percent of all human cancers, with poor prognosis in several forms of cancer. Earlier studies showed those forms of cancer to include NSCLC (24 percent of patients), bladder cancer (approximately 30 percent), and pancreatic cancer (about 40 percent).
“We have done the bioinformatics analysis to show that in lung and bladder cancers, MTAP deletion is mutually exclusive with other genetic aberrations with approved target therapy such as KRAS/EGFR and FGFR3/BRCA mutations. In addition, there is no target therapy drug approved in mesothelioma and esophageal cancers,” Chen said.
When the trial launches later this year, Insilico will become the latest drug developer focused on treating cancer by inhibiting MAT2A, a rate-limiting enzyme in the methionine cycle that primarily catalyzes the synthesis of S-adenosylmethionine (SAM) from methionine and adenosine triphosphate (ATP).
Ideaya selects Phase II dose
Among drug developers studying MAT2A inhibitors in the clinic is Ideaya Biosciences, whose MAT2A inhibitor candidate IDE397 is in a Phase II IDE397-001 trial (NCT04794699) assessing the drug as monotherapy and in combination with docetaxel or paclitaxel, for MTAP-deletion solid tumors. IDE397 is also in two combination therapy trials:
- An Amgen-sponsored Phase I/II study (NCT05975073) assessing IDE397 with AMG 193 in MTAP-deletion NSCLC.
- An Ideaya-sponsored Phase I study (NCT04794699) evaluating IDE397 with Gilead Sciences’ marketed Trodelvy® (sacituzumab govitecan-hziy) in MTAP-deletion bladder cancer. Ideaya announced the study last December.
On Monday, Ideaya announced its selection of a Phase II expansion dose for IDE397 monotherapy in MTAP-deletion squamous NSCLC, citing a favorable adverse event profile and observed preliminary clinical efficacy, including multiple partial responses by RECIST 1.1, designed to capture disease progression and treatment response.
“We believe IDE397 is well positioned as a potential first-in-class MAT2A inhibitor and encouraged to see preliminary translation of our preclinical activities to the clinic in the MTAP-deletion squamous NSCLC setting,” Michael White, PhD, Ideaya’s chief scientific officer, said in a statement. “Next, through this year we look forward to the potential to clinically validate several important preclinical hypotheses we have generated on several mechanistically high conviction rational combinations in the MTAP-deletion setting.”
Ideaya has developed IDE397 alone since 2022, when GlaxoSmithKline declined to pay Ideaya $50 million to exercise rights to develop the drug or any other MAT2A inhibitor candidates of the company. GSK had paid Ideaya $100 million upfront to launch the collaboration in 2020.
Asked if Ideaya’s clinical progress prompted Insilico to advance ISM3412 into the clinic, Chen said the positive data “is very encouraging to us,” and emphasizing how IDE397 differs from ISM3412.
“The structure of our MAT2A inhibitor is different from Ideaya’s [and] shows great preclinical efficacy, safety, PK, PD and other drug-like properties,” Chen said.
“Aside from the potential to be used as a monotherapy, we have preclinical data to show that it can be used together with chemotherapy agents and a PRMT5 inhibitor to achieve synergy.”
In an abstract presented at the American Association for Cancer Research (AACR)’s 2023 Annual Meeting, Insilico researchers reported that ISM3412 “strongly inhibited proliferation in multiple cancer cell lines with natural MTAP deficiency (including pancreatic cancer, bladder cancer, and NSCLC,” and showed improved efficacy when combined with docetaxel, compared to monotherapy.
“ISM3412 exhibited favorable drug-likeness properties, including low molecular weight, good solubility and permeability, and low plasma protein binding. In vivo PK data revealed low clearance and high oral bioavailability. Further, it was well tolerated with no significant hepatobiliary toxicity,” the researchers wrote.
Servier lists 2025 trial
Also disclosing an MAT2A inhibitor in its development pipeline is Servier, whose S95035 is listed as being in Phase I/II development. S95035 is the subject of a planned Phase I trial (NCT06188702) that has yet to recruit patients, according to ClinicalTrials.gov. The trial is described as a first-in-human multicenter, open-label dose escalation study of S095035 in adults with advanced or metastatic solid tumors with homozygous deletion of MTAP who have failed to respond to or have progressed after at least 1 prior treatment regimen, and for whom additional effective standard treatment is not available.
Previously, Servier pursued development of another MAT2A inhibitor, S095033. Servier acquired S95033 when it bought the commercial, clinical, and research-stage oncology portfolio of Agios Therapeutics for up to $2 billion plus potential royalties, in a deal completed in 2021. Agios had developed S95033 under the name AG-270.
Last year Servier disclosed on its website that it had terminated an earlier 83-patient Phase I trial (NCT03435250) of S095033 in June 2022 “because the Sponsor decided to discontinue the AG-270/S95033 clinical development program. This decision was not taken because of concerns regarding the safety of the study drug.”
At the time of the termination, Servier revealed, the maximum tolerated dose of AG-270/S95033 had been reached as a single agent in Treatment Arm 1, studying 40 participants with MTAP-deleted advanced solid tumors or lymphoma; and in combination with docetaxel in Treatment Arm 2, studying 25 participants with MTAP-deleted NSCLC. The trial was terminated during dosing of the approximately 18 patients in Treatment Arm 3, assessing the drug in combination with nab-paclitaxel and gemcitabine in participants with MTAP-deleted pancreatic cancer.
“The Sponsor decided to stop enrollment into the study based on strategy change,” Servier stated.
The company did not say if that change was due to the trial’s scant positive results, which included no complete response in any patient:
- S95033 Monotherapy: Two participants showed a partial response, one with sex cord stromal cell cancer (lasting three years) and the other with non-small cell lung cancer (lasting 48 weeks). The best overall response was progressive disease.
- S95033 plus docetaxel: One participant with esophageal cancer showed partial response (lasting 16 weeks); however, a best overall response was not evaluable for this participant. For most patients, the best overall response was stable disease.
- S95033 plus nab-paclitaxel and gemcitabine: One participant with pancreatic cancer showed partial response (lasting 96 weeks). For most patients, the best overall response was stable disease.
Lesson learned
“The first generation of MAT2A inhibitor from Agios had a very narrow therapeutic window. We learned the lesson from it and utilized generative AI and other approaches to design a second generation MAT2A inhibitor,” Chen said. “This resulted in a significant increase in the therapeutic window up to hundreds of folds in vitro (MTAP deletion vs WT).”
“Our subsequent preclinical testing demonstrated our molecule to be superior in efficacy, and it’s safe with good drug-like properties,” she added.
The termination of S095033 appears to have ended plans for a Phase I/II trial (NCT05312372) disclosed by Servier on ClinicalTrials.gov in a posting last updated in December 2022. That study, reportedly set to start in May 2025, was designed to determine the safety profile, tolerability, pharmacokinetics, and preliminary antineoplastic activity of S095033 in combination with paclitaxel in participants with advanced or metastatic esophageal squamous cell carcinoma (ESCC).
ISM3412 is among the Insilico pipeline candidates developed by its AI platform that treat cancers that have eluded other forms of therapy through synthetic lethality, where mutations in two genes together result in cell death, but a mutation in either gene alone does not.
Last year, Insilico licensed to Exelixis for $80 million upfront another of its synthetic lethality candidates, the small molecule ubiquitin specific protease 1 (USP1) inhibitor ISM3091 for tumors with homologous recombination deficiency (HRD).
