PF-06651600 for the Treatment of Alopecia Areata: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PF-06651600 for the Treatment of Alopecia Areata (NCT03732807), a randomized phase 2, phase 3 study. The primary outcome is Percentage of Participants With an Absolute Severity of Alopecia Tool (SALT) Score of Less Than or Equal to 20 at Week 24.

2. Clinical Question

The primary endpoint represents the proportion of participants who met the stated SALT threshold at the specified assessment time. It is a binary outcome: each participant either met the threshold or did not, and the group result is reported as a percentage. The registry record reports results for five active regimens and placebo, with confidence intervals for the group percentages.

The clinical question is narrower than whether hair loss changed in every participant. It focuses on how often participants reached a particular scalp hair-loss score by week 24. Other listed outcomes address a more stringent SALT threshold, patient-reported change, and SALT-based measures at additional visits. Those endpoints add context, but they do not replace the primary comparison.

3. Trial Design

This was a randomized, parallel-group, placebo-controlled, dose-ranging study of PF-06651600 in adults and adolescents with alopecia areata and 50% or greater scalp hair loss. The registry lists the study as Phase 2 and Phase 3, with 718 participants enrolled. The study status is completed. Pfizer is listed as the sponsor.

Seven sequences are described. Four active regimens began with induction treatment for 4 weeks followed by a maintenance dose for 44 weeks, or used maintenance treatment for 48 weeks. The placebo sequences received placebo for 24 weeks; one then received induction treatment for 4 weeks followed by maintenance treatment for 20 weeks, while another received maintenance treatment for 24 weeks. The primary endpoint was assessed at week 24, when these placebo sequences had not yet entered their subsequent active-treatment period according to the described schedules.

The masking field identifies quadruple masking. The brief study summary describes the trial as double-blinded and explains that the sponsor, study doctors, staff, and patients would not know the assigned treatment or dose. The registry record does not explain the difference between these masking descriptions.

The primary endpoint is reported for five active dose regimens and placebo. The main comparisons concern the percentage with an absolute SALT score of 20 or less at week 24. The registry identifies the Full Analysis Set as the primary analysis population, including randomized participants regardless of whether they received study medication, with a specific exception for missing SALT scores attributed to COVID-19-related reasons. Further detail on the handling of other missing values is not reported in the registry record.

The registry record does not report the original sample-size justification, planned statistical analysis plan, or details of interim analyses. The enrollment total should therefore be read as the number enrolled, not as a description of the assumptions used to design the trial or of the number contributing to every endpoint analysis.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Total 1097 participants signed the informed consent form. Out of which 379 participants were screen failures, 718 actually enrolled into the study and were assigned to study treatments.

Only the first 6 of 7 groups are shown.

Treatment Period (up to Week 24)

Reasons for not completing this period:

Treatment Extension (Week 25 up to 48)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: FAS included all participants who were randomized, regardless of whether they received study medication. Participants with missing SALT scores due to coronavirus disease-19 related reasons were excluded from this analysis, while participants with missing data due to other reasons were considered as non-responders. Here, "Overall Number of Participants Analyzed" signifies participants evaluable for this outcome measure.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Full analysis set (FAS) included all participants who were randomized, regardless of whether they received study medication.

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With an Absolute Severity of Alopecia Tool (SALT) Score of Less Than or Equal to 20 at Week 24, measured over Week 24.

The primary endpoint is the percentage of participants with an absolute SALT score of 20 or less at week 24. SALT is a quantitative assessment of alopecia areata severity based on scalp hair loss. The registry describes scores from 0, representing normal hair, to 100, representing severe disease, with higher scores indicating greater severity. The endpoint turns that score into a threshold outcome: participants are classified according to whether their score falls at or below 20 at the specified visit.

A threshold endpoint is useful for communicating how many participants reached a defined state, rather than only describing average score changes. It is nevertheless a simplified view of the underlying measure. Participants just on opposite sides of the threshold are counted in different categories, while participants with different scores on the same side are grouped together. The percentage therefore does not describe the full distribution of SALT scores or the amount of change for each participant.

The week-24 endpoint also gives a time-specific comparison. It does not establish whether the same results occurred earlier or later. The registry lists additional assessments and other SALT thresholds as secondary outcomes, but those measures address distinct questions. The primary endpoint should be interpreted as the proportion meeting the SALT threshold at week 24, with its accompanying uncertainty and the registry’s stated analysis population.

7. Results Posted in the Registry

• Ritlecitinib (PF-06651600) 200 mg Then 50 mg: 30.6% (38/124) versus 1.5% (2/130) on control; risk difference 29.1 percentage points (95% CI 20.7 to 37.5), risk ratio 19.92 (4.91–80.81); unadjusted two-proportion z-test p = <0.001.
• Ritlecitinib (PF-06651600) 200 mg Then 30 mg: 22.3% (27/121) versus 1.5% (2/130) on control; risk difference 20.8 percentage points (95% CI 13.1 to 28.5), risk ratio 14.50 (3.52–59.70); unadjusted two-proportion z-test p = <0.001.
• Ritlecitinib (PF-06651600) 50 mg: 23.4% (29/124) versus 1.5% (2/130) on control; risk difference 21.8 percentage points (95% CI 14.1 to 29.6), risk ratio 15.20 (3.71–62.37); unadjusted two-proportion z-test p = <0.001.
• Ritlecitinib (PF-06651600) 30 mg: 14.3% (17/119) versus 1.5% (2/130) on control; risk difference 12.7 percentage points (95% CI 6.1 to 19.4), risk ratio 9.29 (2.19–39.35); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Participants With an Absolute SALT Score of Less Than or Equal to 10 at Week 24: Analysis 4

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Ritlecitinib (PF-06651600) 200 mg Then 50 mg: 21.2% (25/118) versus 1.6% (2/125) on control; risk difference 19.6 percentage points (95% CI 11.9 to 27.3), risk ratio 13.24 (3.21–54.68); unadjusted two-proportion z-test p = <0.001.
  • Ritlecitinib (PF-06651600) 200 mg Then 30 mg: 12.6% (15/119) versus 1.6% (2/125) on control; risk difference 11.0 percentage points (95% CI 4.6 to 17.4), risk ratio 7.88 (1.84–33.72); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ritlecitinib (PF-06651600) 200 mg Then 50 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ritlecitinib (PF-06651600) 200 mg Then 30 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With an Absolute SALT Score of Less Than or Equal to 10 at Week 24: Analysis 1

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Ritlecitinib (PF-06651600) 200 mg Then 50 mg: 21.8% (27/124) versus 1.5% (2/130) on control; risk difference 20.2 percentage points (95% CI 12.7 to 27.8), risk ratio 14.15 (3.44–58.27); unadjusted two-proportion z-test p = <0.001.
  • Ritlecitinib (PF-06651600) 200 mg Then 30 mg: 13.2% (16/121) versus 1.5% (2/130) on control; risk difference 11.7 percentage points (95% CI 5.3 to 18.1), risk ratio 8.60 (2.02–36.60); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ritlecitinib (PF-06651600) 200 mg Then 50 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ritlecitinib (PF-06651600) 200 mg Then 30 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With Patient Global Impression of Change (PGI-C) Score of Moderately Improved or Greatly Improved at Week 24

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Ritlecitinib (PF-06651600) 200 mg Then 50 mg: 52.5% (63/120) versus 9.6% (12/125) on control; risk difference 42.9 percentage points (95% CI 32.6 to 53.2), risk ratio 5.47 (3.11–9.61); unadjusted two-proportion z-test p = <0.001.
  • Ritlecitinib (PF-06651600) 200 mg Then 30 mg: 45.4% (54/119) versus 9.6% (12/125) on control; risk difference 35.8 percentage points (95% CI 25.4 to 46.1), risk ratio 4.73 (2.67–8.38); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ritlecitinib (PF-06651600) 200 mg Then 50 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ritlecitinib (PF-06651600) 200 mg Then 30 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The listed secondary outcomes broaden the week-24 question beyond the primary SALT threshold. They include a more stringent SALT threshold, patient-reported global change, exposure-response analyses, other assessment visits, and improvement from baseline. Across the posted secondary results that include arm percentages and reported comparisons, the pattern is generally consistent with the primary finding: some active regimens have higher percentages meeting favorable SALT thresholds or reporting improvement than placebo. These measures are related to the primary endpoint but are not interchangeable with it.

The secondary outcomes differ in definition, assessment time, and analysis population. A stricter SALT cutoff classifies fewer participants as meeting the endpoint, while a patient global impression captures participants’ reported perception of change rather than the SALT scalp-hair assessment. Repeated assessments also address change over time, unlike the single week-24 primary endpoint. The exposure-response outcomes address a different modeling question.

The registry lists 23 outcome measures and 16 analyses, and 16 secondary outcomes are posted. Not every listed secondary outcome has a complete result and comparison in the record summarized by these results; additional outcome findings are not reported in the registry record. No inference should be made about unposted outcomes from the available pattern.

Multiple secondary outcomes and treatment comparisons raise the possibility of chance-positive findings when each is considered separately. The registry record does not report a multiplicity adjustment plan for these secondary tests. Consequently, unadjusted secondary p-values carry less weight than prespecified, multiplicity-controlled evidence would. They can help describe whether different measures point in a similar direction, but they should be treated as supportive rather than as independent confirmation of the primary result.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Percentage of Participants With an Absolute SALT Score of Less Than or Equal to 10 at Week 24: Analysis 4
• Percentage of Participants With an Absolute SALT Score of Less Than or Equal to 10 at Week 24: Analysis 1
• Percentage of Participants With Patient Global Impression of Change (PGI-C) Score of Moderately Improved or Greatly Improved at Week 24
• Exposure Response of PF-06651600 on Regrowth of Lost Hair Based on Absolute SALT Score of Less Than or Equal to 20 at Week 24: Maximum Effect (Emax) Model
• Exposure Response of PF-06651600 on Regrowth of Lost Hair Based on Absolute SALT Score of Less Than or Equal to 10 at Week 24: Maximum Effect (Emax) Model
• Percentage of Participants With an Absolute SALT Score of Less Than or Equal to 20 at Week 4, 8, 12, 18, 28, 34, 40, and 48
• Percentage of Participants With an Absolute SALT Score of Less Than or Equal to 10 at Week 4, 8, 12, 18, 28, 34, 40, and 48
• Percentage of Participants With at Least 75% Improvement in SALT Score (SALT75) From Baseline at Week 4, 8, 12, 18, 24, 28, 34, 40, and 48

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: a person either had a SALT score of 20 or less at week 24 or did not. The group percentage expresses the proportion meeting that criterion. The posted confidence intervals describe uncertainty around those group percentages; a confidence interval is a range of values compatible with the data and the statistical method used, rather than a range containing a known fixed percentage of individual outcomes.

The primary comparisons are between each active regimen and placebo. A risk difference expresses the difference between the two group proportions. When expressed in percentage points, it directly describes the separation between the percentages meeting the endpoint. The registry’s reported analyses give differences and confidence intervals using the Miettinen and Nurminen method. This is the sponsor-reported analysis and is the relevant result to prioritize when it differs from the simple recalculation based on posted group summaries.

The recalculated comparisons are unadjusted, meaning they are simple comparisons of the posted summary data and do not reproduce every detail of a sponsor analysis. They are useful as a check on the scale and direction of the group contrast, but they should not be treated as replacements for the reported analysis. The registry record does not provide enough detail to establish whether all analysis choices, missing-data handling, or any planned adjustments are identical between the recalculation and sponsor analysis. The discrepancy flag makes that distinction especially important.

A p-value describes how incompatible the observed data are with a null comparison under the assumptions of the test. A small p-value is not the probability that the null hypothesis is true, and it does not describe the size or clinical importance of a difference. The reported p-values for four active-regimen comparisons are paired with estimated differences and confidence intervals, which provide information about magnitude and precision. The record does not report a multiplicity adjustment for these several comparisons, so the p-values should not be read as if they necessarily account for testing multiple regimens and outcomes.

Confidence intervals for a group percentage and for a between-group difference answer different questions. The interval around a group percentage describes uncertainty in that percentage; the interval around the risk difference describes uncertainty in the contrast. Overlap or non-overlap of separate group-percentage intervals is not the formal comparison reported by the sponsor. For a direct treatment-versus-placebo interpretation, the sponsor’s interval for the difference is the more relevant interval.

The sample-size illustrations use the observed effect as a starting point and consider smaller fractions of that effect. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Their purpose is to show how a planning calculation changes under stated assumptions, not to recover the original design calculation. The registry record does not report the trial’s own sample-size justification or the assumptions used when the study was designed.

Power is the probability, under a specified statistical model and effect assumption, of detecting an effect at the chosen significance threshold. The sample-size illustrations show participant counts per arm for the specified power targets. They are conditional on the assumed effect size and other planning assumptions; they are not measures of how likely the completed trial was to detect its true effect. Smaller assumed differences generally require more participants to achieve the same planned power, particularly when the endpoint is uncommon and the contrast is correspondingly difficult to estimate precisely.

The Full Analysis Set is described as including randomized participants regardless of whether they received study medication. The registry states that participants with missing SALT scores for COVID-19-related reasons were excluded from the primary analysis, while participants with missing data for other reasons were considered as an additional category whose description is incomplete in the registry record. The detailed rule for those other missing values is not reported in the registry record. This matters because an endpoint percentage can depend on how missing outcomes are handled.

General statistical context: a randomized design supports comparison of assigned groups, but interpretation still depends on the analysis population, endpoint definition, missing-data rules, and prespecified methods. Those details are not all available in the registry record. In particular, it does not report whether interim analyses were planned or whether multiplicity procedures were used for the family of outcomes and treatment comparisons.

11. Confidence Intervals and Precision

The reported confidence intervals show that the week-24 percentages are estimated with varying precision across groups. For example, the interval for the 200 mg then 50 mg group runs from 22.53% to 38.76%, while the interval for placebo runs from 0.0% to 3.65%. The interval for the 10 mg group runs from 0.0% to 4.99%. These ranges communicate uncertainty around each group’s estimated percentage, rather than certainty about an individual participant’s outcome.

Between-group confidence intervals are more directly informative about the treatment-placebo contrasts. In the sponsor-reported analyses, the intervals for the four reported active comparisons have lower limits above zero: 21.17 to 37.91 for 200 mg then 50 mg; 13.65 to 29.18 for 200 mg then 30 mg; 14.65 to 30.23 for 50 mg; and 6.69 to 20.36 for 30 mg. The registry record does not report a sponsor confidence interval for the 10 mg comparison.

Precision depends on both the amount of data and the frequency of the endpoint. The 10 mg and placebo groups have relatively low observed percentages, and their group intervals extend to zero. A low event proportion can leave limited information for estimating a contrast, even when the overall enrollment is substantial. Enrollment across a trial is not the same as the number contributing to a specific endpoint or comparison; the primary endpoint group sizes range from 59 to 130.

Confidence intervals do not address every uncertainty. They do not account automatically for unreported multiplicity adjustments, uncertainty about missing-data assumptions, or differences between the registry’s sponsor analysis and simple recalculations. The sponsor-reported interval should be used for the comparisons it reports; the record does not provide the same sponsor comparison details for every regimen.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size display is a planning illustration based on the observed effect, not a reconstruction of the trial’s original design. The figures are participants per arm for the binary outcomes in this analysis; they are not total event counts. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set at the listed fraction of the observed effect.

The 80% power column gives the estimated participants per arm for a design that would have an 80% probability of detecting the assumed effect under those planning assumptions. The 90% power column gives the corresponding per-arm count for a 90% probability. Higher planned power generally requires more participants for the same effect assumption. Rows planning for 75% or 50% of the observed effect assume a smaller difference to detect than a row planning for 100% of the observed effect; smaller effects need more participants to distinguish them from random variation with the same power target.

For example, the comparison based on the 200 mg then 50 mg regimen lists 24 participants per arm for 80% power when planning for 100% of the observed effect, and 59 participants per arm when planning for 50% of the observed effect. These are hypothetical planning counts, not the enrollment or endpoint-analysis counts in the trial.

The original trial’s sample-size justification and design assumptions are not reported in the registry record. These illustrations are not observed power: they do not measure the probability that the completed study detected its true effect. They show how a hypothetical sample-size calculation changes when the assumed effect and target power change.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.015; treatment proportion p₁ = 0.306; average p̄ = 0.161.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5197 = 1.0185.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4772 = 0.4016.
4. n = (1.0185 + 0.4016)² / (0.2911)² = 23.8, rounded up to 24 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.0153846153846154                          # control proportion
p1 <- 0.306451612903226                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(3732807)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 24 (80% power), 32 (90% power)
 75% of observed effect: n per arm = 35 (80% power), 46 (90% power)
 50% of observed effect: n per arm = 59 (80% power), 78 (90% power)
power.prop.test: 23.8 
Simulated power at n = 24 per arm: 0.893 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry’s sponsor-reported primary analyses compare four active regimens with placebo using the Miettinen and Nurminen method. Those results include estimated differences, 95% confidence intervals, and p-values. For those comparisons, the sponsor’s own analysis takes precedence over simple recalculations from the posted group percentages because a discrepancy is flagged.

The recalculated risk differences are unadjusted comparisons derived from posted summary statistics. Differences between those calculations and sponsor results can arise when the analyses use different estimation details or analysis populations. The registry record does not provide enough information to identify the reason for the discrepancy, so it should not be attributed to a particular method or data-handling choice.

The sponsor-reported analysis lists results for the 200 mg then 50 mg, 200 mg then 30 mg, 50 mg, and 30 mg regimens. It does not list a sponsor comparison for the 10 mg regimen in the primary results. The recalculated 10 mg comparison has a confidence interval spanning zero and an unadjusted p-value of 0.936, but that calculation is not a substitute for an unreported sponsor analysis. No additional sponsor result for that comparison is reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The study includes several active regimens compared with placebo, multiple outcome measures, and assessments at more than one visit. This creates a multiplicity issue: when many statistical comparisons are made, the chance of observing at least one apparently unusual result can increase even if none of the underlying effects is real. Common approaches include prespecifying a testing sequence or adjusting the threshold for a family of tests, but the registry record does not report whether such procedures were planned or used.

The sponsor-reported primary comparisons include four active-regimen analyses, each with a reported p-value. The registry record does not state whether these comparisons were part of a multiplicity-controlled testing strategy. The posted secondary outcomes add further analyses, but their multiplicity handling is also not reported. Thus, individual p-values should be interpreted as reported, without assuming they control the overall false-positive risk across all endpoints and doses.

Missing-data handling is another important limitation. The primary population is described as the Full Analysis Set, including all randomized participants regardless of whether they received study medication. Participants whose SALT scores were missing for COVID-19-related reasons were excluded from the primary analysis. For participants with missing data for other reasons, the registry description is incomplete: it says they were considered as “”, and the specific rule is not reported in the registry record.

Excluding one category of missing assessments and handling other missing observations under a particular rule can affect the analysis denominator and the resulting percentage. The registry record does not report enough detail to assess how sensitive the primary findings are to alternative assumptions about missing outcomes. It also does not report the full prespecified analysis plan, interim-analysis details, or the original sample-size justification. These omissions limit assessment of whether all reported analyses followed a prespecified hierarchy.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidal behaviour (1.6%), eczema (1.6%), abortion spontaneous (1.5%), conversion disorder (1.5%), heavy menstrual bleeding (1.5%).

The safety counts are descriptive and are listed separately by reported safety group. Serious adverse events were reported for 4 of 131 participants in the 200 mg then 50 mg group, 2 of 129 in the 200 mg then 30 mg group, 2 of 130 in the 50 mg group, 1 of 132 in the 30 mg group, and 2 of 62 in the 10 mg group. The placebo groups listed in the safety results had 0 of 65 and 3 of 66 participants with serious adverse events. Deaths were reported as 0 in every listed group.

The counts for other reported adverse events were 84 of 131, 76 of 129, 81 of 130, 77 of 132, 32 of 62, 39 of 65, and 39 of 66 across the listed groups. The registry also lists serious-event terms and percentages, including suicidal behaviour at 1.6%, eczema at 1.6%, spontaneous abortion at 1.5%, conversion disorder at 1.5%, heavy menstrual bleeding at 1.5%, and appendicitis at 0.8%.

These figures should be read as counts in the registry’s safety summaries, not as adjusted comparisons of risk. The record does not provide enough detail to infer causation or to compare the groups statistically. Several placebo safety groups are identified by later treatment sequences, so their group labels should be read as recorded rather than combined into a single placebo category.

17. What the Result Does, and Does Not, Mean

The primary results describe the proportion of participants in each group who reached a week-24 SALT score of 20 or less. The sponsor-reported comparisons show estimated differences from placebo for four active regimens, with confidence intervals that exclude zero. This is evidence about the recorded endpoint and analysis at that visit; it is not a complete account of all dimensions of alopecia areata or each participant’s experience.

The outcome is a threshold measure. It does not show the full spread of SALT scores, the amount of improvement among participants who did not meet the threshold, or whether participants just above the threshold experienced similar changes to those just below it. It also does not establish how durable a response was after week 24. Other outcomes address some related questions, but the registry record does not report every listed secondary result.

Statistical significance and clinical meaning are not identical. A p-value describes compatibility with a null comparison under the analysis assumptions, while the estimated difference and its confidence interval describe the size and precision of the contrast. The registry record does not provide a threshold for a clinically important between-group difference, so the statistical results alone do not establish how meaningful a particular difference is to patients.

The study was randomized and placebo-controlled, but interpretation is still bounded by its analysis population, missing-data handling, multiplicity information, and the discrepancy between recalculations and sponsor analyses. The registry record does not report enough detail to resolve all of these concerns. Nor do the data presented support conclusions about treatment outside the studied population, regimens, and assessment time. The results are statistical descriptions of this trial, not individualized predictions or medical advice.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares several dose regimens with placebo on a clearly defined, binary endpoint at a common assessment time. The dose-ranging structure makes the pattern across regimens relevant, while the placebo group provides a concurrent reference for interpreting the week-24 percentages. Random allocation is an important design feature for making group comparisons, although the validity of any particular estimate still depends on analysis details.

The outcome also illustrates the distinction between a percentage in each arm and a treatment-placebo contrast. The arm percentages describe how often participants met the threshold; the estimated difference and its confidence interval address the comparative question more directly. Reporting both helps readers distinguish the observed response frequency from the uncertainty around the contrast.

The record also demonstrates why statistical interpretation should not stop at p-values. Precision, multiple comparisons, missing-data rules, and analysis-population definitions all affect what can be concluded. The sponsor-reported results are available for four primary contrasts, while a recalculated comparison is available for the 10 mg regimen without a sponsor analysis listed. The flagged discrepancy makes it important to distinguish sponsor estimates from unadjusted recalculations rather than treating them as interchangeable.

Finally, the sample-size illustrations show how planning depends on the effect assumed. They are not a substitute for the original design justification, which is not reported in the registry record. This distinction helps prevent retrospective calculations from being mistaken for evidence about the trial’s original power or planning decisions.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record does not report the original sample-size justification, the full prespecified statistical analysis plan, or whether interim analyses were conducted. It also does not report a multiplicity strategy for the several doses, outcomes, and visit times. These gaps make it difficult to judge how the reported tests fit into the complete planned analysis.

The primary endpoint is a thresholded SALT measure at week 24. It does not capture the complete distribution of scores, individual trajectories, or persistence beyond that assessment. While the registry lists secondary outcomes addressing other thresholds, patient-reported change, and additional visits, not all listed results are reported in the available record.

The primary analysis description excludes participants with missing SALT scores attributed to COVID-19-related reasons, but the rule for handling missing data for other reasons is incomplete. The record therefore does not allow a full assessment of the consequences of missing observations or the robustness of the findings under alternative missing-data assumptions.

A discrepancy is flagged between recalculated and sponsor-reported analyses. The sponsor’s own results take precedence where reported, but the registry record does not explain the source of the discrepancy. Sponsor comparisons are not reported for the 10 mg primary comparison. As a result, that comparison’s recalculated estimate should not be treated as equivalent to a sponsor-reported result.

The registry’s masking descriptions also differ: the masking field says quadruple, while the brief summary describes double blinding and names the parties kept unaware. The record does not explain the difference. These limitations qualify interpretation of the posted statistics; they do not, by themselves, establish that the results are biased.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: PF-06651600 for the Treatment of Alopecia Areata — NCT03732807.
• Registry record last updated 2022-02-24; status last verified 2022-02.
• Results first posted 2022-02-24.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT03732807 is a completed randomized, placebo-controlled, dose-ranging Phase 2/3 study of PF-06651600 in adults and adolescents aged 12 years and older with alopecia areata and 50% or greater scalp hair loss. The registry lists 718 participants enrolled and Pfizer as the sponsor.

The primary outcome was the percentage of participants with a SALT score of 20 or less at week 24. The sponsor reported comparisons with placebo for four regimens using the Miettinen and Nurminen method. A discrepancy with simple recalculations is flagged, so sponsor analyses take precedence where available. The registry record does not report a sponsor comparison for the 10 mg regimen.

Several secondary outcomes and safety summaries are listed. Important information—including the original sample-size justification, multiplicity procedures, complete missing-data handling, and some secondary results—is not reported in the registry record. The posted statistics should therefore be interpreted as endpoint-specific results with the stated precision and limitations, not as a complete account of every planned analysis.