Huntington’s Needs More Than a Better Curve

Skyhawk’s 12-month data for SKY-0515 are encouraging. But an external natural-history comparison is a signal, not a controlled verdict.

July 3, 2026
Editorial
For people living with Huntington’s, hope has to be strong enough to survive a real trial.[Gorodenkoff] / Shutterstock.com

IPM Take

Huntington’s families do not need another reason to wait. They need a reason to believe.

Skyhawk’s new SKY-0515 data offer a credible reason to keep watching. The company reports lower mutant huntingtin in blood and favourable trends across functional, motor and cognitive measures after 12 months.

But the comparison that matters is still missing. A curve that looks better than natural-history expectations is not yet the same as a treatment effect that survives a robust controlled trial.

Executive Summary

Skyhawk Therapeutics reported 12-month interim data from its Phase 1/2 programme for SKY-0515, an investigational oral RNA-splicing modifier for Huntington’s disease.

The company reported favourable mean changes from baseline across four components of the composite Unified Huntington’s Disease Rating Scale: total functional capacity, total motor score, symbol-digit modalities testing and Stroop word reading. These results were compared with expected worsening generated through propensity score-weighted analyses of Enroll-HD natural-history data.

Skyhawk also reported dose-dependent reductions in mutant huntingtin protein in blood of up to 69%, alongside reductions in PMS1 messenger RNA of up to 26%.

The results are company-reported and have not yet been published in a peer-reviewed journal. After the initial 84-day randomised, placebo-controlled period, all participants entered a blinded active-treatment extension. The full 12-month comparison is therefore not against a contemporaneous placebo group.

Why it matters

  • Patients / advocates: Huntington’s has no established disease-modifying treatment. Early data deserve attention, but hope should not be built on claims that the study design cannot yet sustain.
  • Researchers / academia: Natural-history comparisons can help generate hypotheses. They cannot fully remove the risk of selection bias, unmeasured confounding or differences in study populations.
  • Regulators: Biomarker lowering may be important, but any future regulatory case will need to connect biological engagement to outcomes patients recognise in daily life.

A better curve can be powerful.

In Huntington’s disease, where movement, cognition, independence and family life can all be steadily eroded, even the possibility of stabilisation carries enormous emotional force. Skyhawk’s 12-month update therefore deserves serious attention.

The company reported favourable trends across functional, motor and cognitive measures compared with expected decline derived from Enroll-HD natural-history data. It also reported reductions in mutant huntingtin and PMS1, two targets closely tied to the biological rationale behind the programme.

There is real scientific logic here.

SKY-0515 is designed to alter RNA splicing and reduce both mutant huntingtin and PMS1. The first target sits at the centre of Huntington’s pathology. The second is linked to somatic CAG-repeat expansion, a process associated with disease progression. A therapy that can influence both is genuinely interesting.

But interesting is not the same as confirmed.

The 12-month comparison does not place SKY-0515 against a concurrent placebo group. It compares treated participants with an expected disease trajectory generated from external natural-history datasets. That can be informative. It can also mislead when treated groups differ in ways that statistical matching does not fully capture.

This is where the field has to hold its nerve.

Huntington’s cannot afford to dismiss early progress because perfect evidence takes time. It also cannot afford another cycle in which biomarker excitement outruns evidence, only for families to be left carrying the consequences of a result that looked more decisive than it was.

The next trial needs to prove that a better curve is not simply better modelling. It needs to show that the treatment changes the course of disease.

Source & Evidence