Remember, everyone reaches for the same wrench — but not every bolt is the same size. Go break some eggs.
👀 The Read
The NOAEL isn’t the answer.
It’s actually 1 of 4.
There's a reflex to treat the NOAEL as the output of a tox program — the number the whole safety case rests on, the one that sets your first-in-human dose. It usually is, for a conventional small molecule. But "usually" isn't "always," and reaching for the NOAEL out of habit means anchoring your starting dose to the wrong thing.
The NOAEL exists to answer one question: what's the highest dose with no adverse effect, so you can work back from it to a safe human dose. This matters if your drug’s safety is driven by toxicity or exposure — but what if it’s not?
→ Oncology throws the NOAEL out on purpose. The NOAEL tells you which dose isn’t toxic, but in advanced cancers we want to know which dose is toxic but still manageable. So the starting dose comes off the STD10 (the dose severely toxic to 10% of rodents) or the HNSTD (highest non-severely-toxic dose in non-rodents), not a NOAEL. Anchoring an oncology start to a clean NOAEL would leave patients badly underdosed.
→ Biologics often don't use the NOAEL either. When safety is pharmacologically driven, the relevant anchor is the MABEL — the minimum dose that produces the intended biological effect. Because the danger is the target being hit too hard or too long, and a NOAEL can sit well above the dose that's already pharmacologically active.
→ And oligonucleotides, siRNA, and similar modalities can break the framework entirely where conventional plasma exposure doesn't drive the tissue effect, you may be setting doses off administered dose and target-organ findings, not a plasma-exposure NOAEL at all.
So the NOAEL is a tool, not the answer. The real question is which anchor your molecule and indication demand — and defaulting to the NOAEL because it's what you know is how a starting dose ends up indefensible.
The Misconception: The NOAEL is the number your tox program produces, and your first-in-human dose comes off it.
The Correction: The NOAEL is the default, but not a universal rule. Oncology sets the starting dose off the STD10 or HNSTD because toxicity is expected, biologics driven by pharmacology anchor to the MABEL, and oligonucleotides/siRNAs that don’t have exposure may be set off dose and target-organ findings.
🎓️ More of this in my course:
Knowing which anchor your program needs — NOAEL, STD10/HNSTD, MABEL, or dose-driven — is exactly what the course teaches. The full starting-dose framework is in The Complete Guide to Nonclinical Development.
