Remember, you can license the asset — not the homework. Go break some eggs.
👀 The Read
This month, Memento Medicines launched out of RA Capital's Sera accelerator with a $93M Series A — and on the same day, licensed its entire lead program, a Tie2/VEGF bispecific for wet AMD and diabetic macular edema, from two Korean biotechs. The cash funds IND-enabling studies. First-in-human is slated for 2027.
The read: notice what the CEO led with — a "robust preclinical data package established by our partners." That's the whole story, and the whole risk. When you in-license, you inherit a nonclinical package built for someone else's plan — their species, their endpoints, their region's expectations, their route of administration. "Robust" for the originator is not the same as "IND-ready" for your FDA filing, your indication, your intravitreal injection. The gap between those two is exactly what the $93M and the next 18 months are quietly paying for. Founders who win a deal like this run a nonclinical gap analysis on day one: what transfers, what has to be repeated under GLP, what FDA will expect that a Korean originator never generated. Founders who lose assume the package travels — and find the holes during IND review, when a hold costs a year they didn't budget. The asset is only as fundable as the package underneath it.
❓ Just licensed an asset and not sure the data travels? Reply — I read every one.
💭 The Decision
Which inherited studies do you trust — and which do you re-run?
Every licensed asset comes with data. There's always data. The decision isn't whether there's a package — it's which pieces of it actually count toward your IND. And the expensive mistake is rarely the missing study. It's the study that exists but won't count: run in a non-relevant species, outside GLP, by the wrong route, or to a standard FDA doesn't recognize. It looks like progress in the data room. It's a zero on your IND.
So the call you own: study by study, sort the inherited package into accept as-is, bridge, and repeat under GLP. Get it wrong in the generous direction and you file on data that won't hold — refuse-to-file, or a hold. Get it wrong in the cautious direction and you burn months re-running studies that would have transferred fine. Either error has a number attached. You make this call in diligence, before the term sheet — not after the wire clears.
🔧 The Move
This week, if you're looking at any in-licensed or acquired asset: ask for the actual GLP study reports — not the summary deck — for the pivotal tox studies. Then check three things against the IND you intend to file:
GLP status — were the pivotal studies run under GLP — the FDA's quality standard for tox data — or just 'GLP-like,' which FDA won't accept?
Species relevance — for a biologic, was the tox species pharmacologically relevant to your target (ICH S6)? An antibody tested in a species it doesn't bind to tells FDA nothing.
Route + Frequency — does the dosing route match your clinical plan, and does the dosing regimen meet or exceed it?
Any answer that comes back as "we'll get to it" isn't a footnote. It's your timeline and your budget line. Find it now, while it's a diligence question — not later, when it's a hold letter.
🎓 Nonclinical Academy
The Academy covers the full inherited-package gap analysis — how to read a licensed dossier the way FDA will. But if you've got a live deal on the desk right now, skip the course and just reply. Faster fix this week.