Chronic Pain Master Protocol (CPMP): A Study of LY3526318 in Participants With Osteoarthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Chronic Pain Master Protocol (CPMP): A Study of LY3526318 in Participants With Osteoarthritis (NCT05080660), a randomized phase 2 study. The primary outcome is Change From Baseline in Average Pain Intensity as Measured by the Numeric Rating Scale (NRS).

2. Clinical Question

3. Trial Design

NCT05080660 is a completed phase 2 randomized, placebo-controlled trial sponsored by Eli Lilly and Company. Its official title describes a trial evaluating LY3526318 for osteoarthritis pain. Enrollment was 160. The study began on 2021-10-12, and primary completion was recorded as 2022-06-06.

The registry lists a double-masked, parallel design. Participants assigned to 250 mg of LY3526318 took it orally once daily for the first 4 weeks and then switched to placebo once daily for the next 4 weeks. Participants assigned to placebo took it once daily for the 8-week treatment period. The primary outcome was assessed at Week 4, before the active-treatment group’s scheduled switch.

The study is described as part of the Chronic Pain Master Protocol, identified in the record as H0P-MC-CPMP (NCT05986292). The registry summary says the master protocol was intended to accelerate development of new treatments for chronic pain. Further details of the relationship between the master protocol and this trial, including any shared procedures or design features, are not reported in the registry record.

The registry lists osteoarthritis and osteoarthritis of the knee as conditions. It also lists secondary outcomes covering pain, WOMAC pain, stiffness and physical-function measures, and patient-reported overall improvement. The outcome record contains repeated WOMAC outcome titles, so the exact organization of the planned outcome family is not clear from the listed titles alone. Details of allocation procedures, stratification, and the rationale for the enrollment target are not reported in the registry record.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All enrolled participants who received at least one dose of study drug. Here, the overall number of participants analyzed includes number of participants with non-missing values at Week 4.

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: All enrolled or randomized participants who received at least one dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Average Pain Intensity as Measured by the Numeric Rating Scale (NRS), measured over Baseline, Week 4.

The primary endpoint is change from baseline to Week 4 in average pain intensity, measured using the Numeric Rating Scale. Participants described their average pain over the past 24 hours on a scale from 0, meaning no pain, to 10, meaning pain as bad as they could imagine. The record says the scale was used during the preliminary data-entry period and daily throughout the study.

A change score focuses on how a participant’s pain rating differs from that participant’s baseline rating. A negative change indicates a lower score at follow-up than at baseline; the scale’s direction gives those changes their pain-intensity meaning. The registry reports posterior mean changes rather than only a single observation, with a 95% credible interval for each group.

Daily pain ratings can capture variation over time that a single rating may miss. The registry’s repeated-measures approach uses the available measurements across the assessment period in estimating the Week 4 result. The primary time frame is Baseline and Week 4. The registry does not report a separate clinically important change threshold, so the statistical estimates cannot be classified against such a threshold from this record alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a numerical pain score, and the registry summarizes it using means. The listed method is a Bayesian mixed model for repeated measures. This approach is designed for outcome measurements collected repeatedly over time and estimates average change while accounting for the structure of measurements within participants. The registry does not report the complete model specification, including its covariates, covariance structure, or prior distributions.

What does a posterior mean represent?

A posterior mean is a model-based estimate of the average value of a parameter after combining the model assumptions with the observed data. In this trial, the registry reports posterior mean change from baseline for each group and a posterior mean difference between groups. These are estimates, not direct counts of participants whose pain improved. The posterior mean also does not show how much individual participants varied; participant-level distributions are not reported in the registry record.

How should the credible intervals be read?

The reported intervals are 95% credible intervals. In general, a Bayesian credible interval describes uncertainty about a parameter conditional on the model and its assumptions. The interval for the between-group difference extends from -0.26 to 0.73. Because it includes values on both sides of the null difference, the interval does not settle the direction of the contrast. Its width also indicates that the estimate remains uncertain. The registry does not report a p-value for the primary comparison.

A credible interval is not interchangeable with a frequentist confidence interval, even when both are labeled with the same coverage percentage. The primary outcome description calls the results posterior means with 95% credible intervals, while a general dispersion label also refers to a 95% confidence interval. The endpoint-specific description identifies the Bayesian credible interval as the reported measure. The registry does not explain the difference in those labels.

Why use repeated-measures modeling?

Repeated pain ratings can be correlated because measurements from the same participant are related. A mixed model for repeated measures can account for that structure while estimating changes over time. In general, such a model can use repeated observations without treating every rating as independent. Its estimates still depend on the model specification and on assumptions about the observed data. The registry does not describe the assumptions used for missing observations or report sensitivity analyses for departures from those assumptions.

The primary endpoint description says the Numeric Rating Scale was collected daily throughout the study, and the reported primary time frame is Baseline and Week 4. The record does not specify how daily observations were reduced or weighted to produce each analysis value, or how visits outside the endpoint time frame entered the model. Those details are not reported in the registry record.

What do the planning illustrations mean?

The study’s sample-size illustrations are planning exercises based on the observed effect, not the trial’s original design justification. Their stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation; the target effect is a stated fraction of the observed effect. Such calculations describe the participant count that an assumed future design might require under those assumptions. They are not observed power and do not show the probability that this completed trial detected an effect. The sample-size justification used when the trial was designed is not reported in the registry record.

Planning calculations based on an observed effect also inherit the uncertainty of that estimate. When the estimated contrast is imprecise, a planning figure should not be treated as a dependable prediction of the sample size needed for a later study. The calculations are illustrations conditional on their assumptions, not a substitute for a design rationale based on a prespecified target effect.

What is not established by the analysis description?

The registry does not report whether there were interim analyses, how randomization was generated or concealed, or whether analyses were adjusted for multiple outcomes. It does not give enough detail to reproduce the full Bayesian model. The record’s description supports identifying the general analysis framework and interpreting the posted estimates, but not reconstructing every modeling choice or evaluating its sensitivity.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a posterior mean difference of 0.23, with a 95% credible interval from -0.26 to 0.73. This is the result of the Bayesian mixed model for repeated measures described for the primary endpoint. No p-value is reported for the comparison.

A simple unadjusted recalculation from group summary statistics would not necessarily reproduce this model-based result. The model uses repeated measurements and may incorporate features not recoverable from the posted summary values alone. The registry does not report a separate numerical recalculated contrast. No discrepancy is flagged between the sponsor analysis and the comparison assessment in the record; the sponsor’s stated analysis remains the reported result.

The interval spans values on both sides of the null difference, so the sponsor estimate remains uncertain in direction and magnitude. It should be interpreted as a model-based estimate for the Week 4 comparison, not as a direct measurement of individual benefit or as a statement about later outcomes. Details needed to reproduce the sponsor’s complete analysis, including prior distributions and full model specification, are not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 20 outcome measures and 20 analyses, with 19 secondary outcomes posted. The secondary outcomes include pain intensity, WOMAC pain, stiffness and physical-function subscales, and overall improvement as measured by the patient’s global impression of change. Several WOMAC titles appear more than once in the listed outcomes. The record does not explain whether these repetitions represent distinct time points, analysis populations, or duplicated labels.

When a study examines multiple outcomes or analyses, the chance of finding an apparently notable result somewhere among them can increase. Statistical approaches can account for this multiplicity, but the registry record does not report whether any adjustment was planned or applied. It also does not identify a hierarchy among the secondary outcomes or describe whether they were confirmatory or exploratory. The primary endpoint is identified, but the handling of the wider outcome family is not reported in the registry record.

The primary analysis population is described as enrolled participants who received at least one dose of study drug. The overall number analyzed includes participants with non-missing values at Week 4. The group counts reported for the primary analysis are 97 and 52. The record does not provide a complete participant flow, reasons for missing Week 4 values, or a comparison of participants with and without observed outcome data.

The repeated-measures model is named, but the method used to handle missing data is not described in enough detail to assess its assumptions. The registry does not report sensitivity analyses, imputation details, or an alternative analysis population. It also does not state whether all randomized participants were included in any analysis. These omissions limit assessment of how missingness and outcome selection affect interpretation.

12. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: atrial fibrillation (1.9%), diplopia (0.9%), dizziness (0.9%), dysarthria (0.9%).

The registry reports serious adverse event counts by treatment period. During Week 1 to 4, the LY3526318 group had 1 serious adverse event among 107 participants at risk, and the placebo group had 1 among 53 at risk. During Week 5 to 8, the LY3526318/placebo group had 0 among 99 at risk, and the placebo group had 0 among 51 at risk. No deaths were reported in any of these groups.

These are period-specific counts, with different numbers of participants at risk across periods. The Week 5 to 8 LY3526318/placebo group reflects the scheduled switch from LY3526318 to placebo. The counts should therefore be read in light of the treatment sequence rather than as if each period represented a separate, unchanged treatment assignment.

The registry lists atrial fibrillation at 1.9%, and diplopia, dizziness, and dysarthria at 0.9% each among the top serious adverse event terms. The record does not specify the group allocation for these listed terms in the information reported alongside the summary. These serious adverse event figures do not describe all adverse events, and the counts alone do not establish whether an event was caused by study treatment.

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

The primary result is an estimate of the difference between randomized groups in average pain-score change at Week 4. The reported posterior mean difference is 0.23, with a 95% credible interval from -0.26 to 0.73. This interval spans both negative and positive values, so the result does not give a precise direction for the between-group contrast. The registry does not report a p-value.

The within-group estimates are also important to distinguish from the comparison. The LY3526318 group’s posterior mean change was -0.95, and the placebo group’s was -1.18. These are changes from baseline within each group; neither alone estimates the effect of assignment to LY3526318 relative to placebo. The comparative question is addressed by the between-group estimate and its uncertainty.

The result is limited to the specified Week 4 endpoint. The active-treatment group was scheduled to switch to placebo after the first 4 weeks, so this primary comparison does not establish an effect beyond the period of active treatment described for that group. The record does not provide additional results in this summary that would support conclusions about later pain outcomes.

The trial enrolled 160 participants, while the primary analysis counts were 97 and 52. The endpoint description says the overall analyzed numbers include participants with non-missing Week 4 values. Without a participant-flow account and details of missing-data handling, the extent to which unavailable measurements may affect interpretation cannot be assessed from the registry record.

The estimates should not be treated as proof of clinical importance or lack of clinical importance. The registry does not report a threshold for a meaningful change on this scale. Nor do the posted results establish treatment effects in populations or settings beyond the participants and procedures described in this trial. They are statistical estimates from a particular randomized comparison, with uncertainty and reporting limitations.

14. Why This Trial Matters Statistically

This trial illustrates why interpreting a randomized comparison requires attention to the contrast, not just the change observed within each group. Both groups had negative posterior mean changes from baseline, but the between-group estimate is the quantity that addresses the comparative question. Its interval spans values on both sides of the null difference, emphasizing that a point estimate alone does not convey the uncertainty in the result.

The record also illustrates the importance of matching the analysis to the treatment sequence. The LY3526318 group received active treatment for the first 4 weeks and then switched to placebo, while the primary endpoint was assessed at Week 4. The timing makes the primary comparison distinct from any assessment after the switch. Describing the full sequence helps prevent the Week 4 estimate from being read as a result for the entire 8-week period.

The Bayesian reporting is another statistical feature worth noting. Posterior means and credible intervals summarize model-based estimates and uncertainty; they are not identical to a p-value-based analysis. The registry names a Bayesian mixed model for repeated measures, but does not provide the complete specification. Readers can identify the analysis framework while recognizing that independent reproduction is limited by missing technical detail.

Finally, the difference between enrollment and the reported Week 4 analysis counts makes population and missing-data reporting consequential. A complete account of participant flow, analysis inclusion, and missingness assumptions would help readers understand the scope of the estimate. These features matter when comparing trial results, planning future studies, or judging how much confidence to place in a model-based contrast.

15. 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 comparison is reported with a credible interval that extends from -0.26 to 0.73. That range leaves uncertainty about the direction and magnitude of the contrast. The registry does not report a p-value or a threshold for a clinically important change, so neither statistical nor clinical importance can be established from those measures alone.

The endpoint description identifies a Bayesian mixed model for repeated measures but does not give the full model specification or prior distributions. It also does not report the handling of missing values, sensitivity analyses, or assumptions about why observations may be missing. The analysis counts of 97 for LY3526318 and 52 for placebo are described as including participants with non-missing Week 4 values, but the registry does not explain the participant flow from enrollment of 160 to those counts.

Multiplicity handling is not reported, despite 20 outcome measures and 20 analyses being listed and 19 secondary outcomes posted. The record does not describe adjustment procedures, an outcome hierarchy, or the status of secondary analyses. Repeated WOMAC outcome titles also make the organization of the listed outcome set unclear.

The treatment sequence limits the scope of the primary endpoint: the Week 4 result corresponds to the period when the LY3526318 group was scheduled to receive active treatment, while that group switched to placebo for the next 4 weeks. The registry summary does not provide enough information to draw conclusions about the effect of active treatment beyond that assessment. Details of randomization procedures, stratification, masking implementation, interim analyses, and the design-stage sample-size rationale are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Chronic Pain Master Protocol (CPMP): A Study of LY3526318 in Participants With Osteoarthritis — NCT05080660.
• Registry record last updated 2024-07-17; status last verified 2024-06.
• Results first posted 2023-09-06.
• 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.

19. Record Summary

NCT05080660 is a completed, randomized, double-masked, parallel-group phase 2 trial sponsored by Eli Lilly and Company. It enrolled 160 participants with osteoarthritis, including osteoarthritis of the knee. The active-treatment group received 250 mg of LY3526318 once daily for the first 4 weeks and then placebo for the next 4 weeks; the comparator group received placebo for the 8-week treatment period.

The primary endpoint was change from baseline to Week 4 in average pain intensity on the Numeric Rating Scale. The reported posterior mean changes were -0.95 for LY3526318 and -1.18 for placebo. The posterior mean difference was 0.23, with a 95% credible interval from -0.26 to 0.73. The record describes a Bayesian mixed model for repeated measures and does not report a p-value for the comparison.

Interpretation is limited by uncertainty in the comparison, incomplete reporting of model and missing-data details, and the absence of reported multiplicity procedures. Serious adverse event counts were reported by treatment period, with no deaths recorded.