Tarsus to Buy Alkeus for Up to $800M, Betting on a Drug for a Rare Blinding Disease

Tarsus Pharmaceuticals is acquiring Alkeus Pharmaceuticals in a deal worth up to $800 million, adding an experimental drug for a rare, blinding eye disease.

Tarsus will pay $270 million in cash plus $180 million in stock up front, with up to $350 million more tied to regulatory approval and first sale, plus sales royalties to Alkeus’s backers (Bain Capital Life Sciences and TCGX). Expected to close later in 2026.

What Stargardt disease does

The prize is gildeuretinol (ALK-001), being developed for Stargardt disease, an inherited condition causing progressive vision loss.

Stargardt is the most common inherited macular degeneration, usually caused by mutations in a gene encoding a transporter that clears a byproduct of the visual cycle from photoreceptor cells.

When that clearance fails the byproduct accumulates and forms toxic deposits — lipofuscin — in the retinal pigment epithelium, the supporting layer photoreceptors depend on. As that layer degenerates, the photoreceptors above it die.

It typically begins in childhood or adolescence and destroys central vision progressively, leaving peripheral vision. Patients lose the ability to read, recognise faces and drive, generally while young.

The elegance of the mechanism

Gildeuretinol is a modified form of vitamin A designed to prevent the formation of damaging deposits.

Vitamin A is essential to vision — it is the light-absorbing component of the visual pigment, and blocking it would cause blindness rather than prevent it. The problem is not vitamin A itself but a side reaction in which two molecules combine to form the precursor of the toxic deposit.

The modification slows specifically that reaction. Substituting deuterium, a heavier isotope of hydrogen, at the position involved makes the relevant bond harder to break, so the dimerisation proceeds more slowly while the molecule’s normal role in vision is unaffected.

Vertex founder and veteran chemist Josh Boger, Alkeus’s executive chairman, called it “the most elegant possible way for a drug” to work — potentially “the most perfect drug” if it succeeds.

The evidence so far

More than 400 patients have received it — some for seven years — and one placebo-controlled study showed a 29.5% slowing of the yearly growth of those deposits. The FDA granted breakthrough therapy designation in 2021.

Seven years of exposure is unusual for an unapproved drug and is genuinely valuable here. A therapy intended for lifelong use from childhood must be safe over decades, and long open-label follow-up is the only way to see slow-emerging problems.

The 29.5% figure needs reading carefully. Slowing lesion growth by roughly a third is meaningful for a disease with no treatment; it is not stopping the disease, and patients would still lose vision, more slowly.

The endpoint problem

Measuring deposit growth rather than vision reflects a practical constraint that shapes the whole programme.

Visual acuity in Stargardt declines in steps rather than smoothly, and varies between patients depending on which retinal regions are affected when. Detecting a treatment effect on acuity requires very large numbers or very long follow-up.

Lesion area is measurable precisely by retinal imaging, changes continuously, and correlates with functional loss — making it a far more tractable endpoint. It is a surrogate, and regulators accept it in this disease for want of anything better.

What Tarsus is buying

A Phase 3 trial of roughly 230 patients is ongoing, with results expected in 2029.

The structure reflects that distance. Only $450 million is paid up front, with $350 million contingent on approval and first sale — so the buyer defers roughly 44% of the price until the risk has resolved.

Where this sits among Stargardt approaches

Several strategies have been attempted against this disease, and their relative logic explains why a small-molecule approach attracts this price.

Gene therapy is conceptually the most direct fix, since Stargardt is usually caused by a single defective gene. The obstacle is size: the gene involved is too large for the viral vectors most commonly used to deliver genetic material to the retina, and workarounds involving split delivery add complexity and reduce efficiency.

Approaches targeting the visual cycle more bluntly — reducing vitamin A delivery to the retina to slow deposit formation — have been tried and carry an obvious cost, since the same pathway is required for vision, particularly in low light.

A deuterated vitamin A avoids both problems. It needs no vector, works as an oral drug, and slows the harmful reaction specifically rather than throttling the whole pathway. That combination — conceptually clean, deliverable, and applicable regardless of which mutation a patient carries — is what makes it valuable despite an uncertain readout.

For Tarsus, an eye-care company with commercial infrastructure, the fit is in distribution: rare retinal disease is treated at a limited number of specialist centres, which is a narrow channel to serve. The deal is a bet on a rare-disease programme with long follow-up but a readout still years away. Business news, not investment or medical advice.