Prolactin and the Hair Follicle

The deal

What just happened

On 2026-06-24, Absci priced a $100M offering, led by Eli Lilly (about $40M for equity), to fund ABS-201, an AI-designed anti-prolactin-receptor (anti-PRLR) antibody, in androgenetic alopecia and endometriosis.

That is a large bet on a hair-loss mechanism that has nothing to do with DHT. The pathway it targets is real. The question this page works through is what the actual human trials behind it do, and do not, show.

The science

Prolactin is a hair-cycle brake

Human scalp follicles both make prolactin and carry its receptor, and both are up-regulated at the anagen-to-catagen switch. Adding prolactin to organ-cultured human follicles significantly inhibits hair-shaft elongation and forces premature catagen (Foitzik 2006).

The same holds in mice: prolactin induces catagen and delays hair regrowth, and prolactin-receptor-knockout mice have an altered hair-cycle clock (Craven 2001). So removing the brake, by blocking PRLR, could in principle prolong the growth phase, on a pathway entirely separate from DHT. The diagram below shows the logic: prolactin signals through PRLR to push the follicle into catagen; block PRLR and the follicle should stay in anagen longer.

The precedent

HMI-115: this is not AbSci's idea

Hope Medicine built the first anti-PRLR antibody for hair, HMI-115 (originally Bayer's BAY-1158061, licensed to Bayer in 2019). Its Phase 1b in Australia enrolled 16 patients (12 male) over 24 weeks. It was open-label, with no placebo group.

In the men, mean non-vellus target-area hair count rose about 14 hairs/cm2 vs baseline. The company reported this as statistically significant, with no p-value released. On paper that looks like minoxidil-range, but it comes from an uncontrolled study with no placebo group, so a chunk of that 14 is the placebo effect a real trial would subtract out. Adjust for that and the true effect is probably smaller than what a generic minoxidil delivers in a controlled trial.

The evidence

The evidence, by the actual trials

Three trials define what is actually known. One open-label study is the only efficacy signal. One rigorous placebo-controlled Phase 2 finished over a year ago and has never been released. The lead AI program has not measured hair on a single person. The ladder and table below lay out exactly where the case stands.

The tell

Published one trial, buried the other

HMI-115's endometriosis Phase 2 was published in The Lancet Obstetrics, Gynaecology and Women's Health in November 2025, and it was positive: dysmenorrhea pain fell roughly 27 to 42% across doses.

The hair Phase 2, which finished a full year earlier, has stayed silent. When a company publishes one indication and sits on the other, the silent one is rarely the winner. That is an inference, labeled as such, not an assertion that the hair trial failed. But it is the single most informative fact on this page: the rigorous hair test exists, it is finished, and no one has shown the result.

The economics

The value-prop reality

PRLR drugs are monoclonal antibodies: grown in mammalian cell culture, dosed in hundreds of milligrams, and priced like biologics, which means thousands to tens of thousands of dollars per year. For a lifelong, cosmetic, out-of-pocket hair market competing with a roughly $20-per-month generic, that is a steep bar.

ABS-201's roughly 65-day half-life (two to three injections over six months) is partly designed to ease that bar. For context, other anti-PRLR antibodies exist (LFA102, Novartis), but they were oncology programs that did not pan out, and none has been tested for hair.

The verdict

The honest verdict

The prolactin pathway is real and the safety looks clean, so this is not a scam. But $100M is riding on a target whose only rigorous efficacy test is unreleased, and whose lead AI program has not measured hair on a single person.

Sit with that unreleased trial, because it is the most important fact on this page. Companies publish the studies that help them and stay quiet on the ones that do not. HMI-115's sponsor did exactly that: the endometriosis Phase 2 went into The Lancet, while the hair Phase 2, which finished a full year earlier and was the rigorous, placebo-controlled test everyone wanted, has never been shown. The most likely reason a completed, controlled hair trial never sees daylight is that the result was not worth showing. That is an inference, not a proven fact, and the data could still surface. But a finished trial that goes silent almost never went well.

And that silence is not just HMI-115's problem, it is the warning sign hanging over ABS-201. ABS-201 blocks the very same receptor, on the very same pathway, for the very same indication. The one rigorous test of whether blocking PRLR grows hair has, in all likelihood, already been run, and the people who ran it chose not to share the answer. A second antibody hitting the same target does not get to start from hope; it starts a step behind, carrying the burden of proving in a controlled trial what the first one apparently could not. Until ABS-201 actually measures hair against placebo, the most informative data point we have is the trial nobody will let us see.

Real target. Clean safety. Hype ahead of the evidence. Not a scam, and not a breakthrough. The pathway is real. The hype is the product.

Caveats

Limitations

This page summarizes published literature, company disclosures, and public trial registrations, and is for informational and educational purposes only. It is not medical advice and not investment advice. ABS-201 and HMI-115 are investigational; neither is approved for hair loss, and both are discussed off-label.