Nektar Therapeutics (NKTR) released Phase 2b trial results Tuesday that tell one of those "so close, yet so far" stories that make clinical trials equal parts science and nail-biting drama. Their experimental drug rezpegaldesleukin for severe alopecia areata showed promising hair regrowth, but the numbers came up just short of statistical significance. Then came the twist: exclude four patients who shouldn't have been in the trial to begin with, and suddenly the drug hits its target.

The REZOLVE-AA trial enrolled 92 patients with severe-to-very-severe alopecia areata, an autoimmune condition where your immune system essentially declares war on your hair follicles. The result is sudden, patchy hair loss that can progress from coin-sized bald spots to complete scalp hair loss (alopecia totalis) or even total body hair loss (alopecia universalis). It's not just cosmetic—the psychological impact can be significant.

The Numbers Tell an Interesting Story

Nektar measured success using the Severity of Alopecia Tool (SALT) score, tracking the percentage reduction from baseline at Week 36. Both doses of rezpegaldesleukin more than doubled the treatment effect seen with placebo, with most patients experiencing hair growth starting around Week 16 or later.

Here's where it gets interesting. The primary endpoint narrowly missed statistical significance: the 24 µg/kg dose achieved a 28.2% mean SALT reduction, the 18 µg/kg dose hit 30.3%, and placebo came in at 11.2%. The p-values were 0.186 and 0.121 respectively—close, but not quite under the magic 0.05 threshold that makes statisticians happy.

But then Nektar discovered something: four of the 92 patients in the modified intent-to-treat analysis had major study eligibility violations that should have disqualified them from randomization in the first place. When you exclude those four patients, both treatment arms suddenly meet statistical significance. The numbers shift to 29.6% for the higher dose, 30.4% for the lower dose, and just 5.7% for placebo, with p-values of 0.049 and 0.042.

The kicker? One single patient in the placebo arm with an eligibility violation accounted for the 5.5 percentage point difference in placebo performance. The treatment arms themselves showed essentially the same absolute effect whether you included the ineligible patients or not.

Beyond the Primary Endpoint

Both rezpegaldesleukin doses demonstrated dose-dependent clinical effects compared to placebo across key secondary endpoints. These included achieving SALT scores of 30 or less, 20 or less, and 10 or less—basically, measuring how many patients reached different thresholds of hair regrowth. At every timepoint in the study, the treatment arms separated from placebo, showing consistent effects throughout the 36-week period.

Patients who showed hair growth but hadn't reached a SALT score above 20 by Week 36 have the option to continue treatment for an additional 16 weeks through Week 52 in a blinded extension period. That data should arrive in early second quarter of 2026.

What This Means Going Forward

Nektar plans to present the full REZOLVE-AA results at a medical conference in 2026. The question regulators and investors will grapple with is which analysis matters more—the original modified intent-to-treat population that missed significance, or the corrected analysis excluding patients who shouldn't have been enrolled.

This isn't Nektar's first rodeo with rezpegaldesleukin. Back in September, the company reported that the same drug met its primary endpoint in a Phase 2b study for atopic dermatitis (eczema). That trial showed statistically significant improvement in the Eczema Area and Severity Index at week 16 compared with placebo, along with hitting key secondary endpoints including EASI-75, EASI-90, Itch Numerical Rating Scale, vIGA-AD, and Body Surface Area measurements in patients with moderate to severe disease.

The market wasn't thrilled with Tuesday's news—Nektar shares fell 7.86% to $49.11 according to market data. Which makes sense when your headline is "missed primary endpoint" even if the asterisk says "but actually hit it with the right patient population." Clinical trial results live and die by statistical significance, and even a narrow miss creates uncertainty about what regulators will think.

The real test will be whether the FDA and other regulators view the eligibility violation exclusions as legitimate post-hoc corrections or cherry-picking data to get the desired result. If the company can make a convincing case that those four patients genuinely shouldn't have been enrolled and that the corrected analysis represents the true trial population, rezpegaldesleukin could still have a path forward for severe alopecia areata patients who have limited treatment options.