A Study of the Safety, Efficacy, and Biomarker Response of BMS-986165 in Participants With Moderate to Severe Ulcerative Colitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Safety, Efficacy, and Biomarker Response of BMS-986165 in Participants With Moderate to Severe Ulcerative Colitis (NCT04613518), a randomized phase 2 study. The primary outcome is Percentage of Participants in Clinical Response at Week 12.

2. Clinical Question

3. Trial Design

NCT04613518 was a phase 2 randomized, parallel-group study of BMS-986165 in participants with moderate to severe ulcerative colitis. The registry lists quadruple masking and a placebo comparator. Bristol-Myers Squibb was the sponsor. The recorded enrollment was 38, and the study status is completed. The start date was 2021-03-15 and the primary completion date was 2023-05-31.

The study was originally designed to compare two BMS-986165 doses with placebo during a 12-week double-blind period. The registry summary says that, after this period, all subjects receive active therapy in an open-label extension. A protocol amendment removed one dose-treatment arm from the 12-week double-blind period without changing the open-label extension. The endpoint population description identifies the BMS-986165 12 mg BID and placebo BID PO arms as the prespecified population for collection of the primary outcome; it also notes removal of the BMS-986165 6 mg BID arm under Protocol Amendment 02.

The primary endpoint was the percentage of participants with clinical response at week 12. The posted counts are 26 for BMS-986165 12 mg BID and 8 for placebo BID PO. The safety record also includes a double-blind 6 mg BID group with 4 participants at risk, as well as an open-label 6 mg BID group. Those safety groups should not be confused with the two-arm primary endpoint comparison.

The registry record does not report the rationale for the enrollment target, the planned allocation ratio, detailed eligibility criteria, or the operational details of masking. It also does not report the precise timing or rules governing the amendment beyond the stated removal of a dose arm. Therefore, the documented design supports a description of the randomized, masked, placebo-controlled comparison, but not a reconstruction of every design decision.

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: BMS-986165 6 mg BID arm was removed under Protocol Amendment 02.

Double-Blind Treatment Period

Reasons for not completing this period:

Open-Label Treatment Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Prespecified to be collected for all randomized participants in Arms "BMS-986165 12 mg BID" and "Placebo BID PO". "BMS-986165 6 mg BID" was removed under Protocol Amendment 02.

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.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants in Clinical Response at Week 12, measured over At week 12.

The primary endpoint is the percentage of participants in clinical response at week 12. It is a binary outcome summarized as a percentage: each participant either meets the stated response definition or does not. A participant qualifies through a reduction in modified Mayo score from baseline of at least 2 points and at least 30%, along with improvement in rectal bleeding, defined as a decrease of at least 1 point in its subscore or an absolute subscore no greater than 1.

The modified Mayo score used for this endpoint excludes the physicians’ global assessment. It combines stool frequency, rectal bleeding, and endoscopic subscores, each on a 0-to-3 scale. The overall modified score has a 9-point scale, with scores from 5 to 9 denoting moderate to severe disease. The registry says the score used to determine eligibility also served as the baseline disease activity score.

In general, a response endpoint translates several clinical changes into a single outcome that can be counted across participants and compared between randomized groups. The definition matters: changing the score threshold or the rectal bleeding condition would change who qualifies as a responder. A percentage is readily interpretable as the share meeting the prespecified criteria, while a confidence interval communicates uncertainty around that estimate. The endpoint is a specific measure at week 12; it does not summarize every aspect of disease activity, safety, biomarker response, or outcomes during the open-label extension.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is binary at the participant level: clinical response is either met or not met. The arm result is reported as a percentage of participants with response. For this type of outcome, the observed percentage gives a direct summary of the proportion meeting the endpoint definition, while a confidence interval describes its sampling uncertainty. The reported intervals are 95% confidence intervals.

What does the confidence interval say?

In general, a confidence interval is constructed by a statistical method that, across repeated samples under its assumptions, would cover the underlying quantity at the stated confidence level at the intended rate. It shows a range of values consistent with the estimate and the method’s uncertainty. It is not a guarantee that the true value lies in the interval, and it does not indicate the probability that a specific value is true. The BMS-986165 12 mg BID response estimate is 53.8%, with interval limits of 33.4% and 73.4%. The placebo response estimate is 50.0%, with limits of 15.7% and 84.3%. The width of the placebo interval signals limited precision in that arm’s percentage estimate.

How should the arm comparison be read?

The registry reports a mean difference in final values of 3.8, with a 95% confidence interval from -35.7 to 43.4. The outcome itself is a percentage, so this contrast is reported on the scale of the percentage outcome. The interval includes zero and extends in both directions. In general, an interval spanning zero does not provide clear evidence of a nonzero difference at the corresponding two-sided level, but it does not prove the absence of a difference. Its broad range leaves considerable uncertainty about the size and direction of the contrast. No p-value is reported for this analysis.

For a binary endpoint, a difference in percentages is one possible comparison measure. Other measures, such as a relative contrast, would answer a different question and are not reported as the sponsor’s analysis. No additional comparison should be inferred from the arm percentages. In particular, this explanation does not calculate a new contrast from the posted summaries.

Why can precision be limited?

In general, the precision of an estimated percentage depends in part on how many participants contribute and on the distribution of outcomes. Smaller groups usually yield less precise estimates, all else being equal. The registry reports 26 participants for BMS-986165 12 mg BID and 8 for placebo in the primary outcome. These counts provide context for the interval widths, but they do not explain missing endpoint observations or establish the analysis denominator beyond the posted result. The record does not report whether any participants were excluded from the calculation, how missing outcomes were handled, or whether the displayed denominator reflects every randomized participant in each arm.

Binary data can also be affected by over-dispersion in some settings. In general, over-dispersion means that the observed variability exceeds what a chosen model would expect, potentially making model-based uncertainty too small if it is not accounted for. The registry does not report a model or an over-dispersion assessment for the primary endpoint, so no conclusion about either applies to this trial’s analysis.

What can be inferred from the reported method?

The registry labels the endpoint as a percentage and gives confidence intervals for each arm, along with a separate reported between-arm analysis. It does not name the method used to calculate the intervals or the contrast. The analysis-method field is not populated in the posted record. Without the specific method and analysis details, the calculation cannot be independently reproduced from the registry summary alone. General statistical interpretation can describe what percentages and confidence intervals represent, but it cannot identify unreported modeling choices.

The record lists five outcome measures and one analysis, and it posts no secondary outcome analyses. That information does not establish whether multiple outcomes were formally tested, whether a hierarchy was prespecified, or whether any adjustment for multiple comparisons was used. Such procedures affect how inferential results should be interpreted and are not reported in the registry record.

How do planning illustrations differ from trial results?

In general, sample-size planning uses assumptions about the effect to detect, variability, significance level, power, allocation, and analysis method. A planning illustration based on an observed effect is not an observed power calculation and does not show the probability that the trial’s result is correct. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These assumptions describe an illustration, not the trial’s original design justification. The registry record does not report the trial’s own sample-size justification or the planning assumptions used by the investigators.

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("BMS-986165 12 mg BID", "Placebo BID PO")
n      <- c(26, 8)
events <- c(14, 4)
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  BMS-986165 12 mg BID  26  14  53.8  33.4  73.4  35.5  71.2  35.3  71.3
ARM  Placebo BID PO  8  4  50.0  15.7  84.3  21.5  78.5  21.2  78.8

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s between-arm analysis as a mean difference in final values of 3.8, with a 95% confidence interval from -35.7 to 43.4. It does not report a p-value or name the statistical method used. The analysis is posted alongside the arm-specific response percentages and their confidence intervals.

No recalculated two-arm estimate is available from the posted summary statistics for this binary percentage endpoint. The registry record identifies values as missing or not estimable for a two-arm comparison of this endpoint type. Accordingly, there is no independent recalculation to compare with the sponsor’s reported contrast. The displayed arm percentages should not be used to create a new comparison in this interpretation.

Because the discrepancy flag is false, the record does not identify a conflict between a recalculation and the sponsor’s analysis. The sponsor’s reported estimate remains the available contrast, while its broad confidence interval should be considered when judging precision. The record does not provide enough information to reconstruct the analysis method or determine how the contrast was calculated.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and one analysis, while the number of posted secondary outcome analyses is zero. Those counts describe what is recorded, but they do not establish the full prespecified statistical plan. The registry record does not report whether the primary endpoint was part of a formal testing hierarchy, whether other endpoints were tested, or whether any multiplicity adjustment was used. It also does not report interim analyses or stopping rules. These details matter because interpreting several statistical tests as though only one had been conducted can overstate the strength of evidence; in general, multiplicity procedures address the increased chance of false-positive findings across multiple tests.

The primary endpoint population is described as prespecified to include all randomized participants in the BMS-986165 12 mg BID and placebo BID PO arms. The posted endpoint counts are 26 and 8, respectively. The record does not explain how those counts relate to missing observations, withdrawals, exclusions, or the number randomized to each arm. It does not state whether the primary analysis followed an intention-to-treat, modified intention-to-treat, per-protocol, or other population definition.

The missing-data approach is not reported in the registry record. In general, missing outcomes can affect a binary response estimate if participants without an observed endpoint are handled differently from those with observed outcomes. Possible approaches make different assumptions; none should be attributed to this analysis without documentation. The record also does not report sensitivity analyses for missing data, imputation procedures, or whether response status was assigned for participants who discontinued before week 12.

The 6 mg BID arm was removed from the double-blind period under Protocol Amendment 02, and the primary endpoint population description names the other two arms. The registry does not report the timing of the amendment relative to enrollment or provide enough information to assess its effect on the analysis population or testing plan. These omissions limit assessment of the endpoint’s prespecification and the robustness of the reported comparison.

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: colitis ulcerative (12.5%), bacteroides bacteraemia (3.8%), nephrolithiasis (3.8%), deep vein thrombosis (3.8%), anal fissure (3.4%).

During the double-blind treatment period, the record reports serious adverse events affecting 4 of 26 participants at risk in the BMS-986165 12 mg BID group, 0 of 4 in the BMS-986165 6 mg BID group, and 1 of 8 in the placebo BID PO group. The record reports no deaths in these groups. These are event counts and denominators from the safety record and should be read separately from the week 12 clinical-response outcome.

For the open-label BMS-986165 6 mg BID treatment period, 2 of 29 participants at risk were reported as affected by serious adverse events, with 0 deaths. The registry also lists other adverse events affecting 19 of 26 in the 12 mg BID double-blind group, 4 of 4 in the 6 mg BID double-blind group, 6 of 8 in the placebo double-blind group, and 20 of 29 in the open-label group. The record does not provide enough detail in these summary counts to assess event duration, severity beyond the serious classification, relatedness to treatment, or timing.

The listed serious-event terms include colitis ulcerative at 12.5%, Bacteroides bacteraemia at 3.8%, nephrolithiasis at 3.8%, deep vein thrombosis at 3.8%, and anal fissure at 3.4%. The registry excerpt does not specify the denominators or further context for each term. These figures describe reported events; they do not by themselves establish causation or comparative risk. The small groups and distinct treatment periods also limit direct comparisons across the safety categories.

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

The posted result answers a narrow question: what proportion of participants in the two prespecified arms met the clinical-response definition at week 12, and what between-arm contrast was reported? The observed percentages were 53.8% for BMS-986165 12 mg BID and 50.0% for placebo BID PO. The reported contrast was 3.8, with a 95% confidence interval from -35.7 to 43.4. That interval is broad and spans zero, so the estimate is imprecise and does not establish a clear difference between groups.

This does not mean the treatments are equivalent. A finding that does not clearly distinguish a difference is not the same as evidence that no difference exists. The interval remains compatible with contrasts in either direction, and the limited endpoint counts constrain precision. The record does not report a prespecified equivalence or non-inferiority margin, so the results cannot be interpreted as demonstrating equivalence or non-inferiority.

The result also does not answer whether BMS-986165 changes biomarkers, how outcomes compare after the double-blind period, or how long-term safety differs between groups. The trial summary describes an open-label extension after the initial 12-week period, but the primary endpoint result is specifically at week 12. The open-label period does not supply a randomized placebo comparison for the primary endpoint.

Nor do the posted results explain every aspect of the clinical-response outcome. The response definition combines a change in the modified Mayo score with a rectal bleeding condition. A participant who meets the criteria is counted as a responder, but the binary endpoint does not show the distribution of score changes or the separate contribution of each component. Those details are not reported in the primary outcome description.

Overall, the record documents a randomized, masked phase 2 comparison and gives arm-specific response estimates with confidence intervals. It supports a cautious description of the observed week 12 results and their uncertainty. It does not support a definitive conclusion about treatment benefit, lack of benefit, equivalence, or longer-term outcomes.

15. Why This Trial Matters Statistically

This trial illustrates why a point estimate alone is an incomplete account of a clinical comparison. The reported arm percentages, 53.8% and 50.0%, appear close, but the confidence intervals and the between-arm interval show substantial uncertainty. The placebo arm’s interval, from 15.7% to 84.3%, is broad. Reporting interval limits alongside percentages helps readers distinguish an estimate from the precision with which it is known.

The record also shows how a protocol amendment can affect the scope of a result. The study was originally designed with two active dose arms and placebo during the double-blind period, but Protocol Amendment 02 removed one dose arm from that period. The primary endpoint population description specifies the 12 mg BID and placebo arms. The results therefore address that stated comparison and should not be extended to the removed dose arm.

There is also an important distinction between the trial’s design and a planning illustration based on an observed effect. The registry record does not report the trial’s sample-size justification. Any illustration using an observed effect is a planning aid, not evidence that the trial had a particular probability of detecting a true effect. This distinction helps prevent retrospective calculations from being mistaken for the original design rationale.

Finally, the record demonstrates that registry summaries may leave important statistical details unreported. The analysis method, missing-data handling, multiplicity strategy, and analysis-population rules are not specified. Those omissions do not erase the reported estimates, but they limit how fully readers can evaluate their derivation and robustness. For clinical researchers and biostatisticians, separating the directly reported findings from unreported analysis details is essential to a calibrated interpretation.

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 posted primary outcome has limited precision, particularly for the placebo arm, which has 8 participants in the endpoint result. The 95% confidence interval for that arm ranges from 15.7% to 84.3%. The reported between-arm interval is also broad, extending from -35.7 to 43.4. These features constrain how specifically the result can characterize the difference between groups.

The registry does not identify the statistical method used for the arm confidence intervals or the reported contrast. It does not provide a p-value for the contrast, a detailed statistical analysis plan, or enough information to reproduce the analysis from the summary. It also does not describe how the reported endpoint denominators relate to randomization, discontinuation, or missing week 12 outcomes.

Although the endpoint population is described as prespecified for all randomized participants in the 12 mg BID and placebo arms, the record does not name the analysis population or state whether all randomized participants had observed outcomes. Missing-data handling, imputation, and sensitivity analyses are not reported. The absence of those details makes it difficult to assess whether different reasonable assumptions about unavailable outcomes would change the result.

The study’s amendment removed the 6 mg BID arm from the double-blind period, but the record does not give the timing of the amendment or explain its effect on the analysis plan. There is no posted secondary analysis to clarify other outcomes, and the record does not report multiplicity procedures or interim analyses. The primary result is confined to week 12 and does not provide a randomized comparison for the open-label extension. Finally, the serious adverse event summaries do not provide event-level context, treatment relatedness, or enough detail to support causal interpretation.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: A Study of the Safety, Efficacy, and Biomarker Response of BMS-986165 in Participants With Moderate to Severe Ulcerative Colitis — NCT04613518.
• Registry record last updated 2024-07-12; status last verified 2024-07.
• Results first posted 2024-07-12.
• 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

NCT04613518 is a completed phase 2 randomized, quadruple-masked, parallel-group study sponsored by Bristol-Myers Squibb. It enrolled 38 participants with moderate to severe ulcerative colitis. The recorded start date was 2021-03-15, and the primary completion date was 2023-05-31. The study assessed BMS-986165 against placebo during a 12-week double-blind period, followed by an open-label extension according to the registry summary.

The primary outcome was clinical response at week 12, defined by changes in the modified Mayo score and rectal bleeding. The posted response percentages were 53.8% for BMS-986165 12 mg BID and 50.0% for placebo BID PO. The reported contrast was 3.8, with a 95% confidence interval from -35.7 to 43.4. The interval indicates substantial uncertainty around the comparison.

The registry record does not report the statistical method, missing-data approach, multiplicity plan, or sample-size justification. The result should therefore be interpreted as the posted week 12 comparison, not as a complete account of all trial outcomes or the open-label extension.