Alopecia Areata Approvals: SALT Endpoints, Scalp-Coverage Thresholds and Patient-Reported Hair Outcomes at FDA and EMA

Alopecia areata approvals: SALT endpoints, scalp-coverage thresholds and how FDA and EMA treated patient-reported hair outcomes

Three oral JAK inhibitors support the current US alopecia areata (AA) market: baricitinib (Olumiant), ritlecitinib (Litfulo) and deuruxolitinib (Leqselvi). Baricitinib and ritlecitinib are also centrally authorised in the EU. Every pivotal program enrolled patients with at least 50% scalp hair loss (baseline SALT ≥50) and measured success as an absolute residual-hair-loss responder threshold on the Severity of Alopecia Tool (SALT), not as a percent change from baseline. The regulators diverged on which threshold should be primary and on how much weight patient-reported hair outcomes could carry in labeling.

Key takeaways

SALT ≤20 (at least 80% scalp coverage) is the de facto US primary endpoint. It was primary for baricitinib at Week 36 and for ritlecitinib and deuruxolitinib at Week 24 ¹¹⁷¹⁰⁰¹⁴⁷.
The EU accepted a stricter primary for ritlecitinib. Following CHMP scientific advice, the EU primary endpoint was SALT ≤10 at Week 24, with SALT ≤20 as a supportive secondary ⁷⁹²¹²⁶⁶. The FDA/PMDA protocol used SALT ≤20 as primary for the same trial ¹⁸³.
FDA accepted SALT ≤20 as clinically meaningful mainly on regulatory judgment. The reviews do not contain a patient-anchored derivation of the exact cutoff ¹⁹⁴¹⁶⁷¹⁵⁶.
EMA pointed to patient evidence for the threshold. This came from patient panels (baricitinib) and a quantitative patient-preference study (ritlecitinib) ⁸⁷²²⁰⁶⁸.
Patient-reported hair outcomes reached labeling unevenly. The Scalp Hair Assessment PRO appears in the US Olumiant label ¹¹⁸. A satisfaction-with-scalp-hair PRO (SPRO) appears in the Leqselvi label ¹⁶². FDA kept ritlecitinib PGI-C and eyebrow/eyelash results out of the US label, while the EU SmPC includes them ¹⁸⁷¹⁰⁵⁷⁰.

Pivotal design and SALT thresholds by product

Product	Pivotal trials	Entry criterion	Primary endpoint	Key SALT secondaries
Olumiant (baricitinib), FDA	AA-1 and AA-2, randomized, placebo-controlled, 1,200 patients ¹²³	≥50% scalp hair loss by SALT for >6 months ¹²³	SALT ≤20 at Week 36 ¹¹⁷	SALT ≤10 at Week 36 ¹¹⁷
Olumiant (baricitinib), EMA	JAHO (Phase 3 portion) and JAIR ⁸⁵⁹⁶	SALT ≥50 ⁸⁵⁸⁶	SALT ≤20 at Week 36 ⁸³⁸⁴⁸⁵	SALT ≤20 at Weeks 16 and 24; SALT50 at Week 12; SALT90 at Week 36; SALT ≤10 at Weeks 24 and 36 ⁸⁴⁸⁵
Litfulo (ritlecitinib), FDA	B7981015 (ALLEGRO-2b/3), Phase 2b/3 dose-ranging, N=718, ages ≥12 ⁹⁹¹⁰⁵	SALT ≥50, no terminal regrowth in prior 6 months ⁹⁹¹⁰⁰	SALT ≤20 at Week 24 ⁹⁹¹⁰⁰¹⁰⁵	SALT ≤10 at Week 24 ¹⁰⁰¹⁰¹
Litfulo (ritlecitinib), EMA	Same single pivotal study, B7981015 ⁷⁶¹¹¹	SALT ≥50 ⁷⁶⁷¹⁷²	SALT ≤10 at Week 24 ⁶⁶⁷¹⁷⁵¹¹¹	SALT ≤20 at Week 24; SALT ≤10 and ≤20 through Week 48; PGI-C as key secondary ⁶⁶⁷¹⁷⁵
Leqselvi (deuruxolitinib), FDA	CP543.3001 (N=706) and CP543.3002 (N=517), 24 weeks ¹⁴³¹⁵⁰	SALT ≥50 at screening and baseline ¹⁴⁷	SALT ≤20 at Week 24 ¹⁴⁷¹⁵²	SALT ≤20 at Weeks 8 to 20; SALT ≤10 at Week 24 (exploratory); SPRO responder at Week 24 ¹³⁸¹⁴⁰

Deuruxolitinib has no EPAR in the EMA register. A separate EU application for Cinainu, a botanical-extract product for AA in children and adolescents, is listed as "Application withdrawn" ⁴.

Primary and key secondary results

Baricitinib (US label, Week 36). SALT ≤20 was achieved by 22% (2 mg) and 35% (4 mg) versus 5% on placebo in AA-1, and by 17% and 32% versus 3% in AA-2 ¹¹⁷. SALT ≤10 results were 13% and 26% versus 4% in AA-1, and 11% and 24% versus 1% in AA-2. The 2 mg SALT ≤10 result in AA-2 was not statistically significant under the multiplicity plan ¹¹⁷. Response depended strongly on baseline severity. Pooled SALT ≤20 rates with 4 mg were 48% in patients with 50% to 94% baseline loss, but 21% in those with 95% to 100% loss ¹²⁰. The EPAR states that 4 mg was superior to both 2 mg and placebo ⁹³. In JAIR, several stringent secondaries (SALT ≤10 at Week 24, SALT50 at Week 12) failed for one or both doses within the testing hierarchy ⁹⁶.

Ritlecitinib (Week 24). For the marketed 50 mg regimen, FDA's recommended analysis showed SALT ≤20 in 23.0% versus 1.6% on placebo, a difference of 21.4 percentage points (95% CI 13.4 to 29.5) ¹⁰¹. SALT ≤10 was 13.4% versus 1.6% ¹⁰¹. The EU primary analysis gave SALT ≤10 rates of 13.42% for 50 mg and 21.29% for 200/50 mg versus 1.54% on placebo. The 10 mg arm did not separate from placebo ¹¹².

Deuruxolitinib (Week 24). SALT ≤20 was achieved by 29.2% (8 mg BID) and 39.8% (12 mg BID) versus 0.8% on placebo in CP543.3001, and by 32.1% and 36.7% versus 0.8% in CP543.3002 (all p<0.0001) ¹⁴⁵. The Leqselvi label reports SALT ≤10 for 8 mg BID as 20% (Trial AA-1) and 24% (Trial AA-2), versus 0% on placebo. These estimates were not adjusted for multiplicity ¹⁶². Only 8 mg twice daily is approved; the label states that deuruxolitinib 12 mg is not approved.

How regulators justified the SALT threshold

FDA: accepted on regulatory judgment, not patient-derived validation

Ritlecitinib. The FDA review called SALT ≤20 "clinically meaningful hair regrowth" and a treatment-success endpoint ¹⁰³¹⁹⁴. The reviews read SALT ≤10 as a stricter dichotomization of the same measure that could still be useful to patients and prescribers ¹⁰¹. At an earlier meeting FDA had agreed to SALT ≤10 at Week 24 as the Phase 2b/3 primary endpoint and asked the sponsor to clarify how SALT90 relates to absolute SALT ≤10 ¹⁸⁴. The clinical pharmacology review noted that patients with very high baseline SALT need large reductions to reach SALT ≤20 ¹⁹⁰. Patient-Focused Drug Development input is cited for disease burden and unmet need, but not for choosing the cutoff ¹⁹⁴.
Deuruxolitinib. In pre-Phase 3 advice, FDA said the SALT ≤20 at Week 24 primary endpoint "appears reasonable". It called the same threshold at earlier visits "clinically meaningful and supportive" ¹⁶⁷. Cognitive interviews for the SPRO offer some indirect patient context. Participants described being "mostly satisfied" as roughly 80% to 90% coverage and "very satisfied" as 90% to 100% ²¹⁸. FDA did not treat that as validation of the SALT cutoff.
Baricitinib. The publicly posted FDA materials for this approval are labeling and approval documents. They contain no reviewer rationale for the SALT ≤20 threshold ¹¹⁷²³⁵.

EMA: patient evidence cited in support

Baricitinib. CHMP accepted SALT ≤20 at Week 36 because SALT is used in clinical practice and research. The threshold had also been identified in physician and patient panels as a meaningful treatment-success target ⁸⁷¹⁹⁹. In one panel, 34 of 35 patients (97%) regarded moving from ≥50% missing scalp hair to ≤20% as treatment success ⁸⁷²²⁰. No EMA scientific advice was sought for the adult program. The applicant built its AA-IGA, scalp PRO and eyebrow/eyelash ClinROs on FDA recommendations ⁸⁷. For adolescents, PDCO agreed the same Week 36 SALT ≤20 primary, with SALT ≤10 as a key secondary ²⁰⁰. The EU indication now covers severe AA in adults and adolescents 12 years and older. In the adolescent study (257 patients), scalp hair loss improved from over 50% to under 20% at Week 36 in 42% on 4 mg and 27% on 2 mg, versus about 5% on placebo ²³⁶.
Ritlecitinib. CHMP scientific advice endorsed SALT ≤10 at Week 24 ("near remission") as primary. CHMP warned that success in severe AA may take longer than 24 weeks and recommended considering a longer placebo-controlled period ⁷⁹²¹². The assessors used the larger SALT ≤20 effects, together with PGI-C and eyebrow/eyelash outcomes, to support the clinical relevance of the modest SALT ≤10 result ²¹¹¹¹¹. In the adult patient-preference study, "most or all" scalp regrowth (explicitly mapped to SALT ≤20) was the most important attribute, with 42% relative importance. Adolescents gave it 61.6% ⁶⁸⁷⁹. The effects table states that SALT ≤20 was perceived by patients with severe AA as clinically relevant ⁷⁷.

Patient-reported hair outcomes: what FDA reviewers said

Baricitinib: PRO in the label, no COA rationale in the record

The US label reports a Scalp Hair Assessment PRO responder endpoint. Responders moved from a baseline score of ≥3 on a 5-point scale to 0 or 1 with at least a 2-point reduction. Rates were 16% (2 mg) and 33% (4 mg) versus 5% on placebo in AA-1, and 16% and 34% versus 4% in AA-2 ¹¹⁸. Score 1 represents 1% to 20% scalp hair loss, which lines up with SALT ≤20 ¹¹⁷. Eyebrow and eyelash benefit appears only as a qualitative statement of improvement with 4 mg in patients with substantial baseline loss, with no numbers ¹¹⁸. No FDA clinical or COA review of this supplement's content validity or meaningful-change threshold is in the posted approval record ¹¹⁸²³⁵.

Ritlecitinib: eyebrow, eyelash and PGI-C kept out of the US label

The FDA review found favorable Eyebrow Assessment (EBA) and Eyelash Assessment (ELA) results. However, "issues with the measurement properties of the instruments ... limited interpretability of the data, as well as their statistical analyses". FDA stated these results "will not be included in the product labeling" ¹⁰⁵.
PGI-C (responder = "moderately improved" or "greatly improved") and Patient Satisfaction with Hair Growth (P-Sat) were submitted PROs ¹⁷⁸¹⁸⁰. The review states that secondary endpoints not under multiplicity control "will not be included in labeling" ¹⁷⁹.
Section 14 of the US prescribing information presents only SALT ≤20 and SALT ≤10 at Week 24 ¹⁸⁷.
FDA's development-phase advice on generic PROs set a demanding standard. It said EQ-5D-5L "lacks evidence of content validity for use in estimating clinical benefit for labeling claims" and asked for content-validity evidence for HADS and SF-36v2 ¹⁷⁷.

Deuruxolitinib: the most detailed FDA COA analysis

The Satisfaction of Hair PRO (SPRO) is a single-item, 5-category rating of satisfaction with scalp hair. It was a multiplicity-controlled key secondary endpoint, with responders defined as "satisfied" or "very satisfied" at Week 24 ¹⁶⁰¹⁶¹¹³⁹. It succeeded for both doses in both trials ¹⁴⁰:

Trial	Placebo	8 mg BID	12 mg BID
CP543.3001	4.7%	42.1%	53.0%
CP543.3002	1.7%	46.5%	51.7%

Observed responder rates, nonmissing data ¹⁶⁸.

FDA's conclusions:

Content validity. The review found that SPRO "assesses a relevant and important aspect of AA". It also found that the applicant "in general, established content validity and the other measurement properties" ¹³⁹. FDA had earlier required a balanced response scale ¹⁶⁰¹⁶⁴. It noted that the qualitative sample was not fully representative of the US AA population ¹⁷¹.
Reliability. Some test-retest estimates fell below conventional thresholds, including a Phase 2 ICC of 0.65 ¹⁵⁹.
Meaningful change. This was the main unresolved issue: "There is insufficient evidence to support what represents a meaningful score change in the SPRO score" ¹³⁹. Anchor-based estimates ranged from 0.54 to 3.06 points ¹⁵⁶¹⁶⁵. The reviewer said SALT and ClinRO anchors "do not take into account the patient perspective". The reviewer added that "in the absence of qualitative data, it is difficult to determine what constitutes a meaningful change on the anchor scale" ¹⁵⁶.
Labeling. FDA had steered the sponsor away from mean change toward a satisfied/very-satisfied responder analysis ¹⁶⁹. It said SPRO "could potentially support a labeling claim". It recommended that any claim be worded narrowly as satisfaction "with hair on scalp" and that the full response distribution be shown ¹⁶¹¹³⁹. The label presents SPRO as a key secondary outcome in a table on patient satisfaction with scalp hair coverage ¹⁶².
Other instruments. FDA disagreed that the Hair Quality PRO (QPRO) item content, including eyebrow and eyelash items, was supported by qualitative work ¹⁶⁴. Without the instruments and their development history, FDA could not "fully comment on the adequacy" of the Brigham eyebrow and eyelash tools (BETA/BELA). Those endpoints were not multiplicity-adjusted ¹⁶⁹¹⁷⁶.

Patient-reported hair outcomes: what EMA assessors said

Baricitinib

Supportive but not independent. CHMP considered the Scalp Hair Assessment PRO supportive of SALT's clinical relevance. It noted, however, that SALT and the PRO "essentially measure the same construct", namely the amount of scalp hair loss ⁹³⁸⁷.
Satisfaction not measured. Satisfaction also depends on where residual hair loss sits, and it was not measured directly. Skindex-16 and HADS supported it only indirectly ⁸⁷⁹³.
Validation. Interviews with 10 dermatologists and 45 patients supported content validity, and test-retest ICCs were 0.70 to 0.85 ⁸⁵. The PRO had been validated only internally in the pivotal trials, with no external validation. CHMP accepted the missing sensitivity-to-change evidence in context ⁸⁷²²⁰.
Eyebrow and eyelash measures. The ClinROs showed good reliability (ICC 0.89 to 0.90) and strong correlation with the matching PROs ²¹⁹²²⁰. They were not suitable for distinguishing Skindex-16 quality-of-life differences ⁸⁷²²⁰. CHMP called the comparison of new PROs against sponsor-developed new ClinROs "scientifically suboptimal" ⁸⁷.
Results. Pooled Week 36 scalp PRO response was 33.7% (4 mg) versus 4.5% on placebo ²³⁰. With 4 mg, Skindex-16 Emotions and Functioning and HADS Anxiety improved significantly. EMA read these changes as a reduced psychological burden ⁹³.
SmPC. At CHMP's request, the indication points prescribers to the definition of severe AA in SmPC section 5.1 ⁸⁷.

Ritlecitinib

CHMP scientific advice asked for a subjective treatment-benefit measure. As a result, PGI-C became the key secondary endpoint, which CHMP regarded as valid and "easily interpretable" ⁷⁵⁷⁹. PGI-C response was 49.2% versus 9.2% on placebo, a difference of 40.0 percentage points (95% CI 28.9 to 51.1) ⁷⁰.
CHMP treated EBA and ELA as a validated ClinRO set, identical to the scales used in the Olumiant AA registration ⁷⁵. Week 24 responses were 29.0% versus 4.7% (EBA) and 28.9% versus 5.2% (ELA) ⁷⁰.
PGI-C, EBA and ELA all appear in the SmPC efficacy table ⁶⁷⁷⁰. The applicant-developed AAPPO questionnaire showed patient-reported hair improvement, but no active-versus-placebo difference on emotional symptoms. It does not appear in the SmPC efficacy presentation ⁷³.

US and EU labeling of hair outcomes compared

Outcome	US label	EU SmPC / EPAR
Baricitinib Scalp Hair Assessment PRO	Numeric responder results included ¹¹⁸	Accepted as supportive, noted as the same construct as SALT ⁹³⁸⁷
Baricitinib eyebrow/eyelash	Qualitative statement only ¹¹⁸	ClinRO and PRO responder data assessed; pooled ClinRO rates reported ⁸⁶²²⁴
Ritlecitinib PGI-C	Not in section 14 ¹⁸⁷¹⁷⁹	In SmPC, 49.2% vs 9.2% ⁷⁰
Ritlecitinib EBA/ELA	Excluded (measurement properties, multiplicity) ¹⁰⁵	In SmPC efficacy table ⁷⁰
Deuruxolitinib SPRO	Satisfied/very satisfied responder table ¹⁶²	No EPAR

Timing and durability

CHMP judged Week 36 appropriate for baricitinib because responses kept rising through Week 52. The SmPC therefore advises considering discontinuation only if there is no benefit by Week 36 ⁸⁷²⁰¹²⁰². Among Week 52 responders, 96% who stayed on 4 mg and 74% who stepped down to 2 mg maintained SALT ≤20 at Week 76 ²⁰²²⁰⁸. The US label reports similar AA-2 figures: 98% maintenance on 4 mg and 75% after down-titration ¹²⁰. For ritlecitinib, CHMP concluded that benefit increased for up to 24 months on treatment ²¹¹¹¹¹. Neither EPAR provides relapse data after complete withdrawal. CHMP flagged this as an uncertainty, noting that most AA patients relapse months or years after remission ⁸⁷¹⁹⁹²⁰⁴.

Implications for AA development programs

Enroll at SALT ≥50 and prespecify an absolute SALT responder endpoint. SALT ≤20 has regulatory acceptance in both regions. A single-trial EU program may face scientific-advice pressure toward SALT ≤10 and a longer placebo-controlled period ⁷⁹²¹².
Build patient evidence for the threshold early. EMA credited patient panels and preference studies ⁸⁷⁶⁸, while FDA found anchor-based work unpersuasive without qualitative data ¹⁵⁶.
For a hair PRO to reach the US label, it needs multiplicity control, a responder definition patients understand, and narrowly scoped claim wording. The deuruxolitinib SPRO shows the pattern ¹³⁹¹⁶¹¹⁶⁹. Eyebrow and eyelash instruments need documented measurement properties to avoid the ritlecitinib outcome in the US ¹⁰⁵.