Nemolizumab does not clear the bar in dialysis-associated itch

Design

12-week randomised, placebo-controlled trial of nemolizumab 30 mg or 60 mg every four weeks

Population

258 haemodialysis patients with chronic kidney disease-associated pruritus

Primary outcome

proportion with at least a four-point improvement in Worst Itch Numerical Rating Scale at week 12

Effect

60 mg 47.7% (p=0.0686), 30 mg 38.5% (p=0.37) versus placebo 32.4% — primary endpoint not met

Itch in haemodialysis is common, miserable and badly served by existing options. Nemolizumab targets interleukin-31, the cytokine most closely tied to itch, and works in atopic dermatitis and prurigo nodularis. NIKAIA-1 randomised 258 haemodialysis patients with chronic kidney disease-associated pruritus to nemolizumab 30 mg, 60 mg or placebo every four weeks for twelve weeks.

The primary endpoint — a four-point or greater improvement on the Worst Itch Numerical Rating Scale at week 12 — was not met at either dose: 47.7% on 60 mg (p=0.0686) and 38.5% on 30 mg (p=0.37), against 32.4% on placebo. Itch-related quality of life on Skindex-10 improved with 60 mg at a nominal p<0.01. Onset was quick: at week 4, between 24.3% and 26.6% of nemolizumab patients had reached the four-point improvement against 7.4% on placebo. Drug-related adverse events were similar across groups.

A missed primary endpoint is a missed primary endpoint, and the week-4 and quality-of-life findings are hypothesis-generating rather than evidence of benefit. The pattern is recognisable though: a large early separation that narrows as the placebo group catches up, which is characteristic of itch trials and of a 32.4% placebo response. If a phase 3 follows, an earlier endpoint and a bigger sample are the obvious changes. Nothing here alters what to offer a dialysis patient with itch today.

• Do not offer nemolizumab for uraemic pruritus on the strength of this trial
• Recheck the basics first: dialysis adequacy, phosphate, parathyroid hormone and drug causes
• Emollients, and gabapentinoids where tolerated, remain the accessible mainstays in Indian practice
• A 32.4% placebo response is why uncontrolled reports of itch treatments mislead so reliably
• Twelve weeks may be the wrong endpoint: the separation was widest at week 4

The statistics, in plain English

A p-value of 0.0686 for the 60 mg dose means the difference from placebo — 47.7% against 32.4% — is not distinguishable from chance at the conventional threshold, and calling it a trend does not change what the trial showed. The week-4 comparison was not the primary endpoint and was not adjusted for multiple looks, so it cannot carry the conclusion. The Skindex-10 result is described as nominally significant, meaning it too was uncorrected for testing many outcomes. With 258 patients spread across three arms, power to detect a modest true difference was limited.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.