An Efficacy and Safety Study of Palovarotene to Treat Preosseous Flare-ups in FOP Subjects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Efficacy and Safety Study of Palovarotene to Treat Preosseous Flare-ups in FOP Subjects (NCT02190747), a randomized phase 2 study. The primary outcome is Percentage of Responders at Week 6.

2. Clinical Question

3. Trial Design

This was a Phase 2 randomized, double-blind, placebo-controlled, parallel-group study of palovarotene for preosseous flare-ups in subjects with FOP. The registry identifies the masking as triple. Enrollment was 40. The study is listed as completed, with a start date of 2014-07-14 and a primary completion date of 2016-05-23. Clementia Pharmaceuticals Inc. is identified as the sponsor.

The intervention descriptions include two palovarotene dose-level groups and a placebo group. The 10/5 mg group received 10 mg once daily for 14 days, followed by 5 mg once daily for 28 days. The 5/2.5 mg group received weight-adjusted doses of 5 mg once daily, followed by 2.5 mg once daily for 28 days. A further dose-level description specifies weight-adjusted 10 mg once daily followed by 5 mg once daily for 28 days. Placebo was taken once daily for the same duration as the palovarotene dose groups in both cohorts. Details of how participants were assigned among dose levels and cohorts are not reported in the registry record.

The primary outcome was the percentage of responders at Week 6, using the Per Protocol population and only interpretable outcomes. The reported arm sizes for that outcome were 20 in the Palovarotene 10/5 mg group, 9 in the Palovarotene 5/2.5 mg group, and 9 in the placebo group. Those arm sizes describe the outcome analysis, not necessarily every participant enrolled or exposed to study drug. The registry lists 24 outcome measures and 8 analyses. It also lists 23 secondary outcomes, including new HO at Weeks 6 and 12, changes in HO area, biomarker changes, and response at Week 12. The registry record does not report the full statistical analysis plan or the rationale for the enrollment target.

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: The study included clinically diagnosed FOP subjects at least 6 years of age (Cohort 2) or 15 years of age and older (Cohort 1) with symptomatic onset of a flare-up within 7 days of treatment start and accessible for treatment and follow-up.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The Per Protocol (PP) population included all subjects who were eligible for full analysis set (FAS) population, completed Week 6 study visit with no major protocol deviations with at least 80% compliance with study drug, and had an evaluable radiograph or computed tomography (CT) at Week 6 sufficient to allow determination of HO at flare-up site.

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: The safety population included all subjects who received at least 1 dose of study drug (placebo or palovarotene).

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Responders at Week 6, measured over Baseline (Day 1) and Week 6 (Day 42).

The primary endpoint was the percentage of subjects classified as responders at Week 6, with the time frame described as baseline (Day 1) and Week 6 (Day 42). A responder had no or minimal new heterotopic ossification at the flare-up site compared with baseline, assessed by plain radiographs. Minimal new HO meant new HO with an HO score <=3 in both anterior/posterior and lateral projections. If one view was not interpretable or not evaluable, the remaining evaluable view was used. The score ranges from 0, meaning no HO, to 6, meaning a single contiguous HO with a longest dimension >2 diameters of reference normotopic bone in any projection. The highest score from the two projections determined the classification.

The endpoint turns a radiographic assessment into a binary outcome: each evaluated subject either met the responder definition or did not. Reporting the percentage of responders makes the proportion meeting that definition visible in each group. The endpoint was based on primary-read radiology reviews, using a double-read process with consensus adjudication. Radiography and CT scans were examined independently by scan type, flare-up region, and imaging time point to assess whether radiography was sufficient for measuring new HO formation. Only subjects with interpretable outcomes were evaluated.

This endpoint describes the Week 6 radiographic classification; it is not a direct measure of all symptoms, function, or longer-term outcomes. The registry lists several secondary outcomes at Weeks 6 and 12, but their results are not included in the primary result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a percentage of responders represent?

General statistical explanation: a binary outcome places each evaluated participant into one of two categories, such as meeting or not meeting a prespecified responder definition. The percentage in a group describes the share of evaluated participants assigned to the responder category. It does not describe how far an individual outcome changed, and it depends on the precise rule used to classify a response.

In this study, response was defined using the Week 6 radiographic HO score relative to baseline. The reported percentages are 100.0% for Palovarotene 10/5 mg, 88.9% for Palovarotene 5/2.5 mg, and 88.9% for placebo. The corresponding outcome-analysis group sizes are 20, 9, and 9. The sample sizes matter because percentages based on smaller groups can be less stable: a small change in the number of participants classified as responders can materially change the percentage. The registry does not report confidence intervals for these percentages.

What does the Cochran-Armitage test assess?

General statistical explanation: the Cochran-Armitage test is a test for trend in proportions across groups that have an ordered structure. Rather than making only a single pairwise comparison, it assesses whether the proportions follow a systematic pattern across the specified order. Its result depends on the group ordering and the test's assumptions. The registry names this test for the three groups and reports p = 0.166.

The reported test is not a direct estimate of the difference between either palovarotene group and placebo. A p-value is a measure of how unusual data at least as incompatible with a null model would be if that model and the test assumptions held. It is not the probability that the null model is true, and it does not measure the size of an effect. Without a reported effect estimate and confidence interval for each pairwise comparison, readers cannot use the trend-test result to determine the precision of those contrasts. The analysis significance threshold is not reported in the registry record.

Why does the analysis population matter?

General statistical explanation: a Per Protocol analysis focuses on participants who meet protocol-related conditions for inclusion in the analysis. That can be useful for describing outcomes among participants with adequate adherence and evaluable measurements, but it does not answer exactly the same question as an analysis of all randomized participants. Excluding participants because of adherence, protocol deviations, incomplete visits, or uninterpretable scans can affect which participants contribute to the estimate. The direction and size of any resulting effect cannot be inferred without additional information.

The registry describes the primary population as including subjects eligible for the full analysis set who completed the Week 6 visit, had no major protocol deviations, met at least 80% study-drug compliance, and had an evaluable radiograph or CT at Week 6. Only interpretable outcomes were evaluated. The registry record does not report the full population definition, a participant flow accounting for exclusions, or a sensitivity analysis using another analysis population.

What can be learned from missing confidence intervals?

General statistical explanation: a confidence interval describes the range of estimates reasonably compatible with the data and a specified interval procedure. It communicates precision as well as the estimated effect. An interval around a percentage or a difference can show how much uncertainty remains, whereas a point estimate alone cannot. No confidence intervals for the response percentages or pairwise contrasts are reported in the registry result. Their absence means precision cannot be judged from interval estimates in this result.

How should multiplicity and secondary outcomes be considered?

General statistical explanation: when a study examines several outcomes, time points, or comparisons, the chance of obtaining at least one apparently unusual result can increase if every test is interpreted separately without accounting for the set of tests. Methods for handling this issue depend on the prespecified testing strategy. The registry lists 24 outcome measures, 8 analyses, and 23 secondary outcomes. The multiplicity strategy, any hierarchy of tests, and any adjustment procedure are not reported in the registry record. The primary trend-test result should therefore be interpreted as the reported analysis, not as proof that a broader family of outcomes was controlled for multiplicity.

Why does over-dispersion not apply in the same way?

General statistical explanation: over-dispersion is a concern for count or rate models when observed variation exceeds what a model assumes. The primary endpoint is a percentage based on a binary responder classification, not a count of recurrent events per participant. The registry does not report an over-dispersion parameter or a model for recurrent counts for this primary endpoint. Other listed outcomes, including biomarker changes and HO area, have different measurement structures and require their own methods; results and methods for those outcomes are not reported in this primary result.

What do planning calculations mean?

General statistical explanation: sample-size calculations are planning illustrations based on assumptions about a target effect, statistical threshold, allocation, and method. They are not observed trial outcomes and do not indicate the probability that a completed study detected an effect. The registry record does not report the trial's own sample-size justification. No trial-specific planning calculation can be inferred from the enrollment figure or the Week 6 response percentages.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Palovarotene 10/5 mg", "Palovarotene 5/2.5 mg", "Placebo")
n      <- c(20, 9, 9)
events <- c(20, 8, 8)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Palovarotene 10/5 mg  20  20  100.0  83.2  100.0  83.9  100.0  83.9  99.9
ARM  Palovarotene 5/2.5 mg  9  8  88.9  51.8  99.7  56.5  98.0  55.5  97.5
ARM  Placebo  9  8  88.9  51.8  99.7  56.5  98.0  55.5  97.5

11. Analyses Reported by the Sponsor

The registry's reported primary analysis is a Cochran-Armitage test of trend across the three groups, with p = 0.166. The arm-level responder percentages are also reported: 100.0% for Palovarotene 10/5 mg, 88.9% for Palovarotene 5/2.5 mg, and 88.9% for placebo. The test evaluates a trend across groups and is not a pairwise palovarotene-versus-placebo effect estimate.

The registry does not report pairwise estimates or confidence intervals for either palovarotene group compared with placebo. A two-arm recalculation from posted summary statistics is not available for these comparisons in the registry result. The trend-test p-value should not be substituted for a pairwise effect estimate, and it does not reveal the precision of either contrast. The registry record does not report additional details about the sponsor's analysis beyond the named test and its p-value. No discrepancy between a sponsor-reported estimate and a recalculated estimate is reported.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists 24 outcome measures, 8 analyses, and 23 secondary outcomes. The secondary outcomes cover new HO at Weeks 6 and 12, change in the area of new HO, response at Week 12, and changes in several biomarkers at specified follow-up times. The primary result is the Week 6 responder outcome. Results for the secondary outcomes are not reported in this primary result.

When a trial evaluates multiple outcomes, time points, or comparisons, interpretation depends in part on whether the analyses were planned as a family and whether any multiplicity adjustment or testing hierarchy was used. The multiplicity strategy and the statistical analysis plan are not reported in the registry record. It is therefore not possible to determine from this record whether the reported trend test was part of a prespecified sequence or whether results across outcomes were adjusted for multiple testing.

The primary endpoint used a Per Protocol population. The stated criteria include eligibility for the full analysis set, completion of the Week 6 visit, no major protocol deviations, at least 80% study-drug compliance, and an evaluable radiograph or CT at Week 6. Only subjects with interpretable outcomes were evaluated. The full definition of the analysis population, participant flow into the analysis, and numbers excluded for each reason are not reported in the registry record.

The record does not describe a missing-data method or sensitivity analyses addressing missing visits or uninterpretable imaging. Because inclusion required an evaluable scan, the reported percentages characterize the evaluated subjects rather than necessarily all randomized participants. The available information does not establish how excluded outcomes would have affected the response percentages or the trend test.

13. 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: haemorrhagic ovarian cyst (11.1%), asthmatic crisis (10.0%), condition aggravated (9.5%), myoclonus (4.8%).

Serious adverse event figures use safety-group denominators, which differ from the Week 6 endpoint-analysis group sizes. The registry reports 2 serious adverse events among 21 at risk in the Palovarotene 10/5 mg group, 1 among 9 at risk in the Palovarotene 5/2.5 mg group, and 1 among 10 at risk in the placebo group. Deaths are reported as 0 in each group. These counts describe the registry's listed serious adverse events; they are not the same outcome as the Week 6 radiographic response.

The listed top serious adverse event terms and percentages are haemorrhagic ovarian cyst, 11.1%; asthmatic crisis, 10.0%; condition aggravated, 9.5%; and myoclonus, 4.8%. The registry also reports other adverse events affecting all participants at risk in each safety group: 21 of 21, 9 of 9, and 10 of 10, respectively. These summaries do not provide a complete account of each event's timing, severity, duration, or relationship to treatment.

Group counts and event-term percentages should be read in the context of the safety denominators and the study's limited enrollment. The registry record does not report further detail that would support causal conclusions from these figures. They should be interpreted as descriptive safety reporting, separately from the efficacy endpoint.

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

The result describes the proportion of evaluated Per Protocol subjects who met a specific radiographic responder definition at Week 6. The reported percentages were 100.0% for Palovarotene 10/5 mg, 88.9% for Palovarotene 5/2.5 mg, and 88.9% for placebo. The registry's reported test of trend across the three groups had p = 0.166. These findings are limited to this endpoint, time point, population, and method of radiographic classification.

The figures do not establish that the palovarotene groups differed from placebo, and they do not establish that the groups were equivalent. The trend test is not a pairwise comparison, and pairwise effect estimates and confidence intervals are not reported. The p-value is not a measure of the probability that treatment has no effect and does not describe the size or precision of a treatment contrast.

The result also does not answer every clinical question raised by the study. It does not directly describe later outcomes, all aspects of flare-up experience, or changes in function. The registry lists secondary outcomes at Weeks 6 and 12, including HO area and biomarker changes, but their findings are not part of this primary result. The Week 6 outcome depends on interpretable imaging and a prespecified threshold for the HO score; it should not be generalized to outcomes that use different definitions.

Interpretation is further bounded by the Per Protocol analysis criteria. Participants had to meet adherence, visit-completion, and imaging requirements to contribute to the primary endpoint. The result therefore describes the analyzed participants and does not necessarily represent every enrolled or randomized participant. The registry does not report the full analysis plan, a complete account of missing outcomes, or a trial-specific sample-size justification. Those omissions limit assessment of precision, missing-data impact, and how the analysis fits within the full set of planned outcomes.

15. Why This Trial Matters Statistically

This trial illustrates why a clinical endpoint's definition matters as much as its label. “Responder” was not a general judgment about a participant's overall course; it was a classification based on new HO at a particular flare-up site, assessed from radiographs at Week 6 and compared with baseline. The rule included a specified score threshold, provisions for an uninterpretable view, and use of the higher score when both projections could be assessed. Those details determine which observations count as responses.

The study also highlights the distinction between descriptive percentages and comparative evidence. The arm-level percentages provide a direct description of the analyzed groups, while the reported Cochran-Armitage test assesses a trend across groups. Neither the percentages nor the trend-test p-value supplies a pairwise estimate with a confidence interval. Without those measures, the magnitude and precision of a palovarotene-versus-placebo contrast cannot be read from the registry result.

The Per Protocol criteria make the analysis population another central statistical consideration. Eligibility depended on completing the Week 6 visit, protocol adherence, study-drug compliance, and interpretable imaging. These requirements clarify who contributed to the endpoint, but they also mean the result is not simply a summary of everyone enrolled. The registry does not report full missing-data handling or a complete participant accounting.

Finally, the record lists multiple outcomes and analyses, making transparent reporting of the analysis plan and multiplicity strategy important for interpreting the full set of results. The record gives useful information about the primary endpoint and the reported trend test, but does not report the broader statistical plan. The study is therefore a useful example of how endpoint construction, population selection, comparison method, and reporting completeness shape what a clinical trial result can support.

16. 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.

The primary result is based on the Per Protocol population, not an analysis explicitly described as including all randomized participants. The stated requirements include Week 6 visit completion, no major protocol deviations, at least 80% study-drug compliance, and an evaluable radiograph or CT. Only interpretable outcomes were included. The record does not provide a complete participant flow or counts for each reason for exclusion, so the extent and possible influence of exclusions cannot be assessed.

The registry reports arm-level responder percentages and a Cochran-Armitage trend-test p-value, but not pairwise effect estimates or confidence intervals. This prevents direct assessment of the size and precision of each palovarotene-versus-placebo comparison from the reported primary result. The significance threshold and full statistical analysis plan are not reported in the registry record.

The outcome is specific to radiographic HO classification at Week 6. Its binary response definition does not describe the full range of imaging findings or broader clinical outcomes. The record lists secondary outcomes at other time points and for other measures, but their results are not included in the primary result. The multiplicity strategy is not reported, limiting interpretation across the full outcome set.

Enrollment was 40, while the reported primary analysis group sizes were 20, 9, and 9. The registry does not report the trial's sample-size justification, so it is not possible to determine from this record what effect the study was designed to detect or how the planned enrollment related to statistical precision. Safety-group denominators differ from endpoint-analysis sizes, and the reasons for those differences are not reported in this result. The available safety figures are descriptive and do not supply further detail about event timing or treatment attribution.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: An Efficacy and Safety Study of Palovarotene to Treat Preosseous Flare-ups in FOP Subjects — NCT02190747.
• Registry record last updated 2021-02-16; status last verified 2021-01.
• Results first posted 2020-06-24.
• 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.

20. Record Summary

NCT02190747 is a completed Phase 2 randomized, double-blind, placebo-controlled study of palovarotene for preosseous flare-ups in subjects with FOP. The registry identifies the design as triple-masked and parallel-group, with enrollment of 40. The study start date is 2014-07-14 and the primary completion date is 2016-05-23; Clementia Pharmaceuticals Inc. is listed as sponsor.

The primary outcome was the percentage of responders at Week 6, based on a radiographic definition of no or minimal new HO at the flare-up site. In the Per Protocol analysis, the response percentages were 100.0% for Palovarotene 10/5 mg, 88.9% for Palovarotene 5/2.5 mg, and 88.9% for placebo. The registry reports a Cochran-Armitage test of trend with p = 0.166.

The record lists 24 outcome measures, 8 analyses, and 23 secondary outcomes. Pairwise effect estimates and confidence intervals for the primary comparisons, the full analysis plan, the multiplicity strategy, and the trial-specific sample-size justification are not reported in the registry record.