Insilico Medicine just nominated their 30th preclinical candidate.

Thirty. That’s genuinely insane.

The new molecule, ISM0387, is an MTA-cooperative PRMT5 inhibitor with brain penetration. It came out of Insilico’s UAE-based team — fun fact: it’s actually their first asset designed there — and was discovered in roughly six months after their Pharma.AI/Chemistry42 platform generated and screened 90 candidate molecules.

PRMT5 might sound random to most people, but it’s actually a hot oncology target right now.

The MTA-cooperative angle is the differentiated piece: these inhibitors selectively kill MTAP-deleted tumors by exploiting the elevated MTA levels those cells already have.

Translation: less collateral damage in healthy tissue.

Companies like Mirati (now BMS) and Tango have been chasing MTAP synthetic lethality for years. Insilico’s pitch with ISM0387 is brain penetration on top of MTA-cooperativity, which opens up CNS oncology indications and primary brain tumors.

This is cool and all but I think the bigger story is the platform.

Insilico has 31 programs in active development across fibrosis, oncology, immunology, and CNS. Ten of those have IND clearance.

Lead asset rentosertib is a TNIK inhibitor for idiopathic pulmonary fibrosis — another fun fact: this is the first drug where both the target and the molecule were AI-designed — and Phase 2a results were published in Nature Medicine showing dose-dependent improvement in forced vital capacity.

The company IPO’d on the Hong Kong Stock Exchange in 2025 with a market cap around HK$19.5 billion.

Pharma partnerships followed. Sanofi signed a collaboration with up to $1.2 billion in milestones. Eli Lilly committed over $100 million. Exelixis licensed Insilico’s USP1 inhibitor for BRCA-mutated tumors. Menarini took the KAT6A inhibitor for $500+ million in deal value.

The other thing worth noting about ISM0387: the location. The UAE has been pouring capital into building a biotech ecosystem from scratch. Mubadala-backed funds. The Emirates Drug Establishment. Free zones for life sciences. Insilico opened the UAE office and now — six months later — they’re nominating a candidate out of it. Computational biotech moves geographically untethered. The lab can be wherever the talent and capital are.

For context, traditional pharma quotes around 4 to 5 years from target identification to IND. Insilico claims roughly 30 months at one-tenth the cost.

Evaxion has been showing similar speed advantages in a different therapeutic context, and even Novo Nordisk now has a public partnership with OpenAI.

The era when generative AI in drug design was a pitch deck flourish is APPARENTLY over – that was fast.

It’s a tool people are using to actually nominate molecules. Thirty of them, in Insilico’s case.

The next test?? Approval…this will be a WILDDD day.

None of the 30 has cleared Phase 3 yet. But rentosertib is moving and the platform is generating candidates faster than most fully-staffed pharma chemistry departments could.

If even three of these 30 candidates make it to market, the AI drug discovery argument is done.