A Study to Examine the Efficacy and Safety of REGN5069 in Patients With Pain Due to Osteoarthritis of the Knee: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Examine the Efficacy and Safety of REGN5069 in Patients With Pain Due to Osteoarthritis of the Knee (NCT03956550), a randomized phase 2 study. The primary outcome is Change From Baseline to Week 12 in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Score.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, parallel-group Phase 2 study comparing REGN5069 with matching placebo in patients with knee osteoarthritis pain. The registry describes quadruple masking. The three groups were placebo, REGN5069 100 mg Q4W, and REGN5069 1000 mg Q4W, with a stated 1:1:1 allocation. The registry records enrollment of 259 participants. The start date was 2019-05-21, and the primary completion date was 2020-05-01. The study status is terminated; the reason for termination is not reported in the registry record.

The primary objective was to evaluate REGN5069 compared with placebo in patients with radiographically confirmed knee osteoarthritis and a history of inadequate pain relief or intolerance to current analgesic therapy. The secondary objectives included serum concentrations of functional REGN5069, safety and tolerability during treatment for up to 12 weeks, and anti-drug antibody assessment following multiple IV administrations. The planned treatment duration described in the objectives is up to 12 weeks.

The primary analysis population is described as the full analysis set, including all randomized participants, while the stated n for an outcome is the number evaluable at that specific time point. For the primary endpoint, the registry reports evaluable counts of 84 for placebo, 80 for REGN5069 100 mg Q4W, and 82 for REGN5069 1000 mg Q4W. These time-point counts are not the same concept as the overall enrollment figure. Details about the randomization procedure, concealment, masking roles, and reasons for any differences between enrollment and endpoint evaluability are not reported in the registry record.

The registry identifies mixed models analysis for the sponsor’s primary comparisons. Further details of the model specification and its assumptions are not reported in the registry record. The trial’s own sample-size justification, planned interim analyses, and full statistical analysis plan are also 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.

Before assignment: A total of 259 participants were randomized in 1:1:1 ratio to receive REGN5069 at 100 mg IV Q4W, REGN5069 at 1000 mg IV Q4W, or matching placebo Q4W. Participants were to receive 3 fixed-dose IV infusions at baseline, week 4, and week 8. 171 participants were randomized to receive REGN5069 (86 participants in the 100 mg Q4W group and 85 in the 1000 mg Q4W group) and 88 were randomized to receive placebo.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set population (FAS): included all randomized participants; Here 'n' = number of evaluable participants at the specific time point.

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 Change From Baseline to Week 12 in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Score, measured over Baseline to Week 12.

The primary endpoint was change from baseline to Week 12 in the WOMAC Pain subscale score. WOMAC is a questionnaire-based measure with pain, physical function, and stiffness subscales. The pain subscale consists of five questions; each question is graded from 0 to 10, and the subscale score is calculated as the sum divided by the number of questions. The resulting score is reported on a 0-to-10 Numerical Rating Scale. Higher scores indicate worse pain.

Change from baseline describes how the score at follow-up differs from the participant’s baseline score. A negative change therefore indicates a decrease in the reported pain score. The endpoint is useful for comparing groups because it applies the same defined symptom measure at baseline and follow-up, allowing the change in each randomized group to be described and contrasted with placebo. It captures reported pain rather than a direct measure of joint structure or a broader measure of health.

The registry reports least-squares mean changes and standard errors for each group. A least-squares mean is a model-based group estimate; the registry does not provide enough detail to explain the exact model adjustments behind these estimates. The posted descriptions do not identify a threshold for a clinically meaningful difference, so the numerical size of a score change should not be interpreted against such a threshold. The primary time frame is baseline to Week 12.

7. Results Posted in the Registry

• REGN5069 100 mg Q4W: mean -2.37 versus -2.09 on control; difference -0.28 (95% CI -0.81 to 0.25), standardized effect d = -0.16; z-test p = 0.296.
• REGN5069 1000 mg Q4W: mean -1.98 versus -2.09 on control; difference 0.11 (95% CI -0.41 to 0.63), standardized effect d = 0.06; z-test p = 0.681.

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.

Change From Baseline to Week 12 in WOMAC Total Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • REGN5069 100 mg Q4W: mean -2.15 versus -1.83 on control; difference -0.32 (95% CI -0.81 to 0.17), standardized effect d = -0.20; z-test p = 0.204.
  • REGN5069 1000 mg Q4W: mean -1.81 versus -1.83 on control; difference 0.02 (95% CI -0.48 to 0.52), standardized effect d = 0.01; z-test p = 0.937.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between REGN5069 100 mg Q4W and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between REGN5069 1000 mg Q4W and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 12 in WOMAC Physical Function Subscale Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • REGN5069 100 mg Q4W: mean -2.09 versus -1.74 on control; difference -0.35 (95% CI -0.85 to 0.15), standardized effect d = -0.21; z-test p = 0.173.
  • REGN5069 1000 mg Q4W: mean -1.75 versus -1.74 on control; difference -0.01 (95% CI -0.51 to 0.49), standardized effect d = -0.01; z-test p = 0.969.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between REGN5069 100 mg Q4W and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between REGN5069 1000 mg Q4W and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 12 in Patient Global Assessment (PGA) Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • REGN5069 100 mg Q4W: mean -0.75 versus -0.59 on control; difference -0.16 (95% CI -0.37 to 0.05), standardized effect d = -0.23; z-test p = 0.142.
  • REGN5069 1000 mg Q4W: mean -0.80 versus -0.59 on control; difference -0.21 (95% CI -0.42 to 0.00), standardized effect d = -0.30; z-test p = 0.054.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between REGN5069 100 mg Q4W and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between REGN5069 1000 mg Q4W and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary continuous outcomes cover WOMAC total score, WOMAC physical function, and Patient Global Assessment, alongside the primary pain measure. Across these measures, the lower-dose estimates generally point toward more negative changes than placebo, while the higher-dose estimates are often close to placebo. Patient Global Assessment estimates are more negative in both dose groups than in placebo, but their comparisons remain uncertain. Taken together, the secondary results do not supply a clear, consistent pattern of separation across doses and outcome domains.

These outcomes add context to the primary result because they address related aspects of symptoms and function. They do not remove the uncertainty in the primary comparison, and they should not be treated as independent confirmation without considering how many outcomes were examined and how the analyses were specified. The registry does not report a multiplicity adjustment for secondary comparisons. When many outcomes or comparisons are examined, some apparently notable results may occur by chance even when there is no underlying difference. Unadjusted secondary findings therefore carry less weight than a prespecified primary analysis, particularly when their role in a testing hierarchy is not reported.

The registry also lists secondary measures concerning WOMAC stiffness, improvement in WOMAC pain, adverse events, imaging abnormalities, and anti-drug antibodies. Detailed numerical findings for those measures are not reported in the registry record in the results summarized for the continuous secondary outcomes. The registry lists eight secondary outcomes, but the posted information does not establish a complete inferential framework for interpreting them together. These limitations make the secondary outcomes useful for describing the scope of evaluation, but not a basis for an unqualified overall conclusion.

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

• Change From Baseline to Week 12 in WOMAC Total Score
• Change From Baseline to Week 12 in WOMAC Physical Function Subscale Score
• Change From Baseline to Week 12 in Patient Global Assessment (PGA) Score
• Change From Baseline to Week 12 in WOMAC Stiffness Subscale Score
• Percentage of Participants With ≥30% Improvement in WOMAC Pain Subscale Score
• Number of Non-Serious and Serious Treatment-Emergent Adverse Events (TEAEs) Through End of Study
• Number of Imaging Abnormalities Consistent With Adjudicated Arthropathies Through End of Study
• Number of Participants With Presence of Anti-REGN5049 Antibody Development Through End of Study

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a score on a scale, rather than a count or a yes/no event. The registry reports least-squares mean change and standard error within each group. A mean describes the group’s average observed change, while a standard error describes the precision of that estimated mean. It is not the same as the spread of individual participants’ scores.

What does the group comparison estimate?

The mean difference compares the estimated change in a REGN5069 group with the estimated change in the placebo group. A negative difference on this endpoint means the REGN5069 group’s estimated change was more negative than placebo’s; a positive difference means it was less negative. This direction is specific to the score’s interpretation: higher WOMAC scores indicate worse pain. The comparison is an estimate, not a direct statement about what happened to every participant.

The recalculated comparisons are unadjusted comparisons from the posted group summary statistics. Their confidence intervals and p-values are not the same as the sponsor’s mixed models results. The registry identifies mixed models analysis for the sponsor’s comparisons, but the model specification, covariates, handling of repeated observations, and other technical details are not reported in the registry record. It is therefore not possible to reproduce the sponsor’s model from the posted information.

How should a confidence interval be understood?

A confidence interval describes uncertainty around an estimated comparison. A narrower interval indicates greater precision than a wider interval, all else being equal. An interval that includes zero does not rule out a difference in either direction, and it also includes the possibility of little or no group difference. It does not mean that every value in the interval is equally plausible, nor does it show the range of individual participant responses. The confidence level used for the posted intervals is not reported in the registry record.

What do the p-values communicate?

A p-value measures how compatible the observed data are with a specified null comparison under the analysis assumptions. It does not measure the probability that the null hypothesis is true, the size of a difference, or the clinical importance of an estimate. The recalculated p-values are unadjusted. The registry’s reported p-values come from mixed models analysis. The threshold and any multiplicity procedure used in the sponsor’s formal analysis are not reported in the registry record, so the p-values should not be treated as a complete account of the analysis plan.

Why are recalculated results approximate?

The simple recalculation uses posted group means and standard errors to construct an unadjusted comparison. The pooled standard deviations used for standardized-effect and planning illustrations are approximated from those standard errors. This is not the same as having participant-level data or the sponsor’s model output. Recalculation can be useful as a transparent check on the direction and approximate precision of the posted summaries, but it cannot recover unreported modeling choices, missing-data handling, or adjustment procedures.

What does the analysis population mean?

The primary endpoint description defines the full analysis set as including all randomized participants. It also clarifies that the stated n is the number of evaluable participants at the specific time point. The endpoint counts therefore describe participants contributing evaluable information at Week 12, rather than automatically establishing how missing observations were handled. The registry does not report the amount or pattern of missing outcome data, the method used to address missing data, or sensitivity analyses assessing its impact.

General statistical context: repeated-measures models can use information from measurements collected over time and can account for correlations among observations, depending on their specification. That general description should not be taken as a claim about the exact model used in this trial. The registry reports mixed models analysis but does not report enough detail to determine its exact structure or assumptions.

11. Confidence Intervals and Precision

The primary comparison estimates are accompanied by intervals that express uncertainty. For REGN5069 100 mg Q4W versus placebo, the recalculated mean difference is -0.28, with interval limits of -0.805 and 0.245. For REGN5069 1000 mg Q4W versus placebo, the mean difference is 0.11, with interval limits of -0.414 and 0.634. Each interval includes zero, so neither comparison precisely identifies the direction of the difference. The endpoints of an interval are not a range of individual responses; they describe uncertainty in the estimated group contrast.

The group standard errors for the primary endpoint are 0.188 for placebo, 0.191 for REGN5069 100 mg Q4W, and 0.19 for REGN5069 1000 mg Q4W. Standard errors reflect uncertainty in estimated means, not the variability among all individual WOMAC scores. The registry does not give the confidence level for the recalculated intervals or the participant-level distribution of scores.

The intervals are wide enough to include values on both sides of zero. This limits how firmly the observed differences can be distinguished from no difference or from a difference in the opposite direction. Precision is separate from whether an estimate is important to patients: a clinically meaningful threshold is not reported in the registry record. The model-based sponsor results do not include an estimate or interval in the posted analysis fields, so the recalculated intervals should not be represented as sponsor-model intervals.

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 figures are planning illustrations expressed as participants per arm for this continuous outcome, not total event counts. They are based on the observed effect and the stated planning assumptions: a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect defined as the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column describes the planned number of participants per arm under those assumptions for a design with an 80% chance of detecting the specified effect if that effect is the true one and the assumptions hold. The 90% power column describes the corresponding planning target for a design with a 90% chance. These are design calculations, not guarantees about results in a future study. The effect fractions are planning scenarios based on the observed estimate; they are not established clinical targets.

Smaller effects require more participants to distinguish them from random variation under otherwise unchanged assumptions. This is why the figures increase when the planned effect is a smaller fraction of the observed effect. The very large values associated with especially small observed comparisons reflect the instability of planning directly from a small estimated effect; they should be read as illustrations, not as a recommended trial size. These figures are not observed power: observed power is not a useful substitute for interpreting the estimated effect and its uncertainty after a study has been completed.

The calculations do not incorporate details that are not reported, such as the sponsor’s original design rationale, any planned allowance for missing data, or the analysis adjustments used in the trial. They should not be treated as a reconstruction of the study’s protocol or as a sample-size recommendation.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.280; pooled standard deviation σ = 1.716.
2. n = 2 × (1.96 + 0.842)² × 1.716² / -0.280² = 589.5, rounded up to 590 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
delta <- -0.28                      # observed difference in means
sigma <- 1.71589930350038                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3956550)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 590 (80% power), 790 (90% power)
 75% of observed effect: n per arm = 1,049 (80% power), 1,404 (90% power)
 50% of observed effect: n per arm = 2,359 (80% power), 3,157 (90% power)
power.t.test: 590.5 
Simulated power at n = 590 per arm: 0.800 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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 identifies mixed models analysis as the sponsor’s method for both primary comparisons. Its reported p-values are 0.298 for placebo versus REGN5069 100 mg Q4W and 0.689 for placebo versus REGN5069 1000 mg Q4W. The posted sponsor analyses do not include the corresponding model-based estimates or confidence intervals in the available results.

The recalculated results use the posted group means and standard errors to make simple unadjusted comparisons. They give p-values of 0.296 and 0.681, respectively, and mean differences of -0.28 and 0.11. Small differences between these p-values and the mixed models p-values can arise because the calculations are not the same analysis. The registry does not report enough model detail to identify the source of each difference or reproduce the sponsor’s results.

The recalculation is best treated as a descriptive check on the posted summary statistics, not as a replacement for the sponsor’s model-based analysis. The sponsor analysis is the trial’s reported inferential method. Neither set of p-values, on its own, describes the size or clinical importance of a difference; the estimated direction and uncertainty also matter. A detailed account of the prespecified analysis plan and any multiplicity handling is not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, dose groups, or comparisons. The more opportunities there are to find a result that appears unusual by chance, the more important it is to know whether the analysis plan controlled the overall error rate or specified a hierarchy of tests. This study had a primary pain endpoint, two dose-versus-placebo comparisons, and several secondary outcomes. The registry does not report whether multiplicity adjustment was used, whether comparisons were tested in a prespecified order, or which secondary analyses were intended to be confirmatory.

The recalculated p-values are explicitly unadjusted and should be interpreted as such. The reported sponsor analyses are mixed models analyses, but their p-values alone do not explain whether multiplicity was addressed. The formal statistical analysis plan and the sponsor’s multiplicity strategy are not reported in the registry record. As a result, the nominal p-values cannot be read as a complete account of the probability of false-positive findings across all outcomes and comparisons.

Missing-data handling is also not reported in enough detail to assess its effect on the estimates. The primary endpoint description states that the full analysis set included all randomized participants and that the number shown is the number evaluable at the specific time point. The registry does not report the number or reasons for unavailable Week 12 measurements, the assumptions used for missing observations, or sensitivity analyses. Differences between randomized and evaluable counts do not by themselves identify how missing data were handled.

Interim analyses are not reported in the registry record. Nor are detailed model assumptions, planned covariate adjustments, or a complete description of the analysis populations for each secondary outcome. These omissions do not establish that particular procedures were absent; they mean the procedures cannot be verified from the registry information. Interpretation should therefore distinguish clearly between the posted outcome summaries, the sponsor’s stated mixed models method, and information that remains unreported.

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: covid-19 (1.2%), clostridium difficile colitis (1.2%), dizziness (1.2%), transient ischaemic attack (1.2%), myocardial infarction (1.2%).

The registry reports serious adverse events for placebo, REGN5069 100 mg IV Q4W, and REGN5069 1000 mg IV Q4W. The respective numbers at risk were 88, 85, and 84; the numbers reported with serious adverse events were 3, 3, and 0. The registry records zero deaths in each group. These are group-specific counts as recorded, not rates recalculated from the available information.

Serious adverse events and other adverse events are separate entries in the posted results. The numbers with other adverse events were 48 for placebo, 33 for REGN5069 100 mg IV Q4W, and 36 for REGN5069 1000 mg IV Q4W. The registry also lists serious-event terms with frequencies including COVID-19, Clostridium difficile colitis, dizziness, transient ischaemic attack, myocardial infarction, and pneumonia. The listed percentages are 1.2% for the first five terms and 1.1% for pneumonia; the record does not identify an arm alongside each term in the summarized information.

These counts describe events recorded during the study and do not establish that a treatment caused an event or that groups had equivalent safety. Exposure duration, event timing, detailed event narratives, and statistical comparisons of safety are not reported in the registry record. Small group-specific counts should be read descriptively.

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

The primary findings describe changes in WOMAC Pain scores through Week 12 in the randomized groups. The estimates indicate score decreases in placebo and both REGN5069 groups, with the lower-dose group’s estimated decrease more negative than placebo’s and the higher-dose group’s estimate less negative than placebo’s. The recalculated comparisons are uncertain: each interval includes zero and extends to values in the opposite direction from the point estimate.

Those results do not establish a precise difference between either dose and placebo. They also do not establish that there is no difference. An estimate near zero with an interval spanning both directions is compatible with a range of possible group contrasts. The registry does not report a threshold for a clinically meaningful change, so the posted score differences cannot be classified against such a threshold from the available information.

The results do not answer every question about treatment. The primary endpoint concerns a self-reported pain subscale at Week 12; it is not a measure of long-term outcomes, structural change in the knee, or every aspect of function and health. The secondary outcomes offer related information but are subject to the same need for uncertainty-aware interpretation, as well as the unreported multiplicity approach.

The safety counts are descriptive and do not establish causation or rule out uncommon harms. The registry marks the study terminated, but the reason for termination is not reported. Neither the status nor the posted endpoint estimates, in isolation, explain the decision to stop the study.

Overall, the posted record supports a cautious statistical description: there are group estimates for the Week 12 pain outcome, but the simple comparisons are not precise enough to resolve the direction and size of the differences. This is an interpretation of the available registry results, not medical advice or a treatment recommendation.

18. Why This Trial Matters Statistically

This trial illustrates why a group mean and its uncertainty need to be considered together. Looking only at the point estimates could emphasize the more negative change in the REGN5069 100 mg Q4W group, while overlooking that its comparison interval includes zero and values in the opposite direction. Looking only at a p-value would also omit the estimated size and direction of the comparison. Precision and effect magnitude answer different questions.

The two dose comparisons also show why a simple assumption of a steadily increasing response with dose should not be imposed on the observed results. The posted primary estimate for the higher dose is not more negative than the placebo estimate, while the lower-dose estimate is. This pattern is descriptive; the intervals do not precisely establish the differences, and the registry does not provide a dose-response analysis.

For biostatisticians and clinical researchers, the record also highlights the distinction between a sponsor’s model-based analysis and an unadjusted recalculation from summary data. The recalculation can provide a transparent sense of the comparison, but it cannot reconstruct model details that are not reported. Similarly, planning calculations based on an observed effect are illustrations, not evidence that the completed trial had a particular probability of success.

Finally, outcome breadth matters. Pain was the primary endpoint, while other symptom and safety domains were also listed. Without a reported multiplicity strategy, secondary findings require caution. The registry is informative about the study question and posted summaries, while important features of the analysis plan and missing-data approach remain unavailable.

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.

The registry marks the study as terminated, but does not report why it was terminated. The status alone should not be used to infer the cause or to explain the outcome estimates. The primary endpoint is measured through Week 12, and the registry does not provide longer-term primary outcome results in the information. This limits conclusions to the stated follow-up period.

The study population was specifically described as patients with radiographically confirmed knee osteoarthritis and a history of inadequate relief or intolerance to current analgesic therapy. The registry record does not report enough eligibility detail to assess how broadly the findings apply beyond that population. It also does not report a clinically meaningful threshold for interpreting WOMAC score differences.

The primary outcome summaries give evaluable counts at the endpoint, but the record does not provide the amount, timing, or reasons for missing measurements or explain how missing data were handled. The sponsor’s method is identified as mixed models analysis, but its specification and assumptions are not reported. A full analysis plan, multiplicity procedure, and interim-analysis details are also not reported in the registry record.

Safety summaries are counts and do not provide enough information to assess exposure-adjusted event rates, event timing, or causality. The listed serious-event terms are not connected to a treatment arm in the summarized information. The study’s planned sample-size rationale is not reported; the available planning illustrations use observed effects and should not be mistaken for the original design basis.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Examine the Efficacy and Safety of REGN5069 in Patients With Pain Due to Osteoarthritis of the Knee — NCT03956550.
• Registry record last updated 2021-07-01; status last verified 2021-06.
• Results first posted 2021-07-01.
• 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

NCT03956550 was a randomized, double-blind, quadruple-masked, parallel-group Phase 2 study of REGN5069 for pain due to knee osteoarthritis. It compared REGN5069 100 mg Q4W and REGN5069 1000 mg Q4W with matching placebo using a stated 1:1:1 allocation. Enrollment was 259, and the registry records the study as terminated.

The primary endpoint was change from baseline to Week 12 in WOMAC Pain score, for which higher scores indicate worse pain. The posted group estimates decreased in all three groups. Simple unadjusted comparisons had intervals including zero, while the registry reports mixed models p-values of 0.298 and 0.689 for the two dose-versus-placebo comparisons. The sponsor’s model details, the termination reason, and important aspects of missing-data and multiplicity handling are not reported in the registry record.