A Phase 3 Study to Compare Between CT-P41 and US-licensed Prolia in Postmenopausal Women With Osteoporosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 3 Study to Compare Between CT-P41 and US-licensed Prolia in Postmenopausal Women With Osteoporosis (NCT04757376), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Lumbar Spine Bone Mineral Density (BMD) at Week 52 - Full Analysis Set (FAS).

2. Clinical Question

The primary endpoint concerns change in BMD at the lumbar spine, specifically vertebrae L1 to L4. The registry record lists this as a Phase 3, randomized, active-controlled study. Its stated scope also includes pharmacokinetics, pharmacodynamics, and safety, but the primary question is the between-group comparison of lumbar spine BMD change at Week 52.

The FAS definition is based on receiving at least 1 full dose of study drug. The endpoint result lists 222 participants for CT-P41 and 212 for US-licensed Prolia, while the FAS contains 239 and 238, respectively. The endpoint description says missing data were handled using multiple imputation under a missing-at-random assumption. Other details about missing observations and the analysis population used for each individual result are not reported in the registry record.

3. Trial Design

The registry identifies NCT04757376 as a Phase 3 study with randomized allocation, quadruple masking, and a parallel-group model. It is described as active-controlled and double-blind in the study title and summary. The study is listed as completed. Enrollment was 479; the recorded start date was 2021-06-17 and the primary completion date was 2023-05-18.

The two biological interventions were CT-P41 and US-licensed Prolia. Each arm description specifies a 60 mg/mL single dose in a solution for injection in a prefilled syringe. The record does not report the allocation ratio, the randomization procedure, the roles covered by quadruple masking, or further details of treatment administration. Those details should not be inferred from the design labels alone.

The primary endpoint was percent change from baseline in lumbar spine BMD at Week 52, assessed by DXA. The measurement was taken at the lumbar spine from L1 to L4, with assessments performed at a central imaging vendor. The stated primary timeframe is baseline at screening to the Week 52 predose assessment.

Secondary outcome domains listed in the registry include BMD at additional sites and time points, fracture incidence, denosumab serum concentrations, and serum carboxy-terminal cross-linking telopeptide of type I collagen (s-CTX). The record also describes a treatment period and maintenance and switching groups in some safety and secondary results. Further descriptions of those phases and the trial’s complete treatment schedule are not reported in the registry record.

The original design’s sample-size assumptions, including any target difference or allowance for missing data, are not reported in the registry record. Interim analyses and any stopping rules 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: The screening period began after the participants had signed the written informed consent form and completed it within 28 days before the 1st randomization (Week 0 Day 1). On Week 0 Day 1, participants were randomized in a 1:1 ratio to receive CT-P41 or US-Prolia during Treatment Period (TP) I. At Week 52 predose, participants in the US-Prolia group were re-randomized 1:1 to either continue US-Prolia or transition to CT-P41. The participants in the CT-P41 group continued with CT-P41 for TP II.

TP I - Week 0, Day 1 to Week 52 Predose

Reasons for not completing this period:

TP II - Week 52 to Week 78

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The FAS was defined as all participants who received at least 1 full dose of the study drug (CT-P41 or US-licensed Prolia). The total number of participants in FAS was 239 and 238 in the CT-P41 and US-licensed Prolia groups, respectively. The overall number of participants analyzed represents the number of participants in FAS who had a BMD assessment result for the lumbar spine by DXA at Week 52.

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 overall number of participants analyzed represents the Intent-to-treat (ITT) Set. The ITT set was defined as all participants randomly assigned to receive study drug (CT-P41 or US-Prolia), regardless of whether or not any study drug was administered.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Lumbar Spine Bone Mineral Density (BMD) at Week 52 - Full Analysis Set (FAS), measured over baseline (screening), Week 52 predose.

The primary endpoint is the percent change from baseline in lumbar spine BMD at Week 52. BMD is a continuous measurement, and the endpoint expresses the change relative to the participant’s baseline measurement. DXA was used to assess BMD, with the lumbar spine measurement covering L1 to L4 and performed at a central imaging vendor.

Baseline-to-follow-up change is useful because it describes how a measured quantity differs from its starting value during the study. Comparing that change between randomized groups focuses the primary analysis on the contrast between CT-P41 and US-licensed Prolia rather than on either group’s follow-up value alone. The Week 52 predose assessment is the specified endpoint time.

For the primary result, the registry reports least-squares means and standard errors. Least-squares means are model-based group estimates; they are not necessarily identical to simple arithmetic averages. The analysis description specifies an analysis of covariance (ANCOVA) coupled with multiple imputation for missing data, assuming data were missing at random. The registry does not report all model details, such as the covariates included.

In general, a continuous endpoint can show the size and precision of a group contrast, while a confidence interval communicates a range of values compatible with the analysis. BMD is one of several outcome domains in the study. The primary endpoint alone does not summarize fracture outcomes, pharmacokinetics, pharmacodynamics, or safety.

7. Results Posted in the Registry

• US-licensed Prolia: mean 5.15 versus 4.96 on control; difference 0.19 (95% CI -0.67 to 1.05), standardized effect d = 0.04; z-test p = 0.660.

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.

Ctrough of Denosumab at Week 52 - PK-TP II Subset

• 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 deviations.
• Result:
  • US-licensed Prolia Maintenance: mean 66.60 versus 70.24 on control; difference -3.64 (95% CI -34.63 to 27.35), standardized effect d = -0.03; z-test p = 0.818.
  • Switched to CT-P41: mean 126.15 versus 70.24 on control; difference 55.91 (95% CI -46.23 to 158.05), standardized effect d = 0.18; z-test p = 0.283.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between US-licensed Prolia Maintenance 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 Switched to CT-P41 and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcome list extends beyond the primary lumbar spine BMD comparison. It includes BMD at other skeletal sites or time points, fracture incidence, denosumab concentrations, and s-CTX. These outcomes address related but distinct questions. BMD and s-CTX are measurements of bone-related outcomes, fracture incidence is an event outcome, and serum concentration measures concern pharmacokinetics.

The detailed Week 52 trough concentration results show substantial variability, and the reported comparison intervals include zero. This is a less precise pattern than a point estimate alone might imply. These concentration comparisons do not establish a corresponding difference in the primary BMD endpoint, because concentration and BMD are different measures. Nor can they be used as a substitute for the sponsor’s primary ANCOVA result.

Numerical results for the other listed secondary outcomes are not reported in the registry record. As a result, the overall pattern across secondary BMD measurements, fractures, and s-CTX cannot be assessed from the available numerical information. Their listed presence indicates the study’s breadth of outcomes, not a finding about their results.

In general, secondary analyses often involve multiple outcomes, time points, or analysis populations. If many hypotheses are tested, some apparently notable results may occur by chance unless multiplicity is addressed. The registry does not report a multiplicity adjustment or a hierarchy for secondary outcomes. Accordingly, unadjusted secondary results carry less weight than a prespecified primary analysis and should be interpreted as supporting or exploratory evidence rather than as independent confirmation.

The registry lists 13 secondary outcomes as posted and 14 outcome measures overall. The record does not provide enough numerical detail in the reported secondary results to synthesize a complete endpoint-by-endpoint pattern.

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

• Percent Change From Baseline in Lumbar Spine, Total Hip, and Femoral Neck BMD at Week 52 - FAS
• Percent Change From Baseline in Lumbar Spine, Total Hip, and Femoral Neck BMD at Week 78 - FAS-TP II Subset
• Incidence of New Vertebral, Nonvertebral, and Hip Fractures During TP I - FAS
• Incidence of New Vertebral, Nonvertebral, and Hip Fractures During TP II - FAS-TP II Subset
• Trough Serum Concentration (Ctrough) of Denosumab at Weeks 0 and 26 - Pharmacokinetic (PK) Set
• Ctrough of Denosumab at Week 52 - PK-TP II Subset
• Maximum Serum Concentration (Cmax) of Denosumab After First Dose - PK Set
• Percent Change From Baseline for Serum Concentration of Serum Carboxy-terminal Cross-linking Telopeptide of Type I Collagen (s-CTX) at Weeks 26 and 52 - Pharmacodynamic (PD) Set

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it records percent change in BMD rather than a yes-or-no event. The registry reports least-squares means for each group and a standard error for each estimate. These summaries describe the estimated group values and their sampling uncertainty. The posted values are 4.96 with a standard error of 0.300 for CT-P41, and 5.153 with a standard error of 0.321 for US-licensed Prolia.

What does ANCOVA contribute?

ANCOVA, or analysis of covariance, estimates a group contrast while accounting for covariates included in the model. In general, covariate adjustment can improve the precision of a comparison when the selected covariates are associated with the outcome. The registry says that ANCOVA was used for the primary efficacy endpoint, but the specific covariates, model terms, diagnostics, and handling of any departures from model assumptions are not reported in the registry record. The reported ANCOVA estimate is therefore the appropriate reference for the sponsor’s stated primary analysis, while the recalculation from summary statistics is a simpler, unadjusted comparison.

How should multiple imputation be understood?

The endpoint description says multiple imputation was used for missing data under a missing-at-random (MAR) assumption. In general, multiple imputation represents uncertainty about missing values by creating completed datasets, analyzing them, and combining the estimates. MAR means that, conditional on information included in the imputation process, the probability that a value is missing does not depend on its unobserved value. This is an assumption, not something established simply by using the method. The registry does not report the number of missing primary endpoint values, variables used in the imputation, the number of imputed datasets, or sensitivity analyses for departures from MAR.

What is the unadjusted recalculation?

The recalculated contrast uses the posted group summaries rather than the trial’s ANCOVA model. For the primary endpoint, it gives a mean difference of 0.193 for US-licensed Prolia relative to CT-P41, with a 95% confidence interval from -0.668 to 1.054 and an unadjusted p-value of 0.66. The pooled standard deviation used in the calculation is approximate, as indicated by the registry-derived summary information. The calculation is useful as a transparent check on the scale and uncertainty of the posted group results, but it does not reproduce the ANCOVA with multiple imputation.

In general, an unadjusted p-value describes how compatible the data are with a null contrast under the assumptions of that particular test. It does not address all sources of uncertainty, prove equivalence, or account for multiple outcomes unless a multiplicity procedure is applied. The primary reported analysis has no p-value recorded. The registry record does not report whether additional inferential procedures were prespecified.

Why can standard errors and standard deviations differ?

A standard deviation describes variation among observed measurements, whereas a standard error describes uncertainty in an estimated mean. In general, the standard error depends on both the data’s variability and the amount of information contributing to the estimate. They are not interchangeable. The primary group results report standard errors; the Week 52 trough concentration results report standard deviations. The large variation in those concentration values is relevant when interpreting their mean comparisons, because considerable dispersion can make estimates less precise.

What does an equivalence analysis require?

In general, equivalence is assessed by defining acceptable bounds for the group difference in advance and checking whether the confidence interval lies entirely within those bounds. A finding that a conventional difference test is not statistically significant is not, on its own, evidence of equivalence. The registry labels the primary ANCOVA analysis as equivalence and reports a 90% interval, but it does not report the equivalence bounds or decision rule. The formal equivalence conclusion cannot be assessed from the reported label and interval alone.

The registry reports a Full Analysis Set definition based on receiving at least 1 full dose. For the primary result, the group counts analyzed are 222 and 212, while the FAS counts are 239 and 238. The reason for the difference between these counts is not reported in the registry record. Analysis populations for every secondary endpoint and details of any additional per-protocol analyses are not reported in the registry record.

11. Confidence Intervals and Precision

The sponsor-reported primary ANCOVA estimate is -0.19, with a 90% confidence interval from -0.76 to 0.38. The interval includes zero, so it is compatible with values on either side of no difference under that analysis. Its endpoints also indicate that the estimate is not exact: the data leave a range of plausible contrasts rather than identifying a single value with certainty.

The recalculated unadjusted comparison has a 95% confidence interval from -0.668 to 1.054. This interval also includes zero. It is based on posted summary statistics and differs from the ANCOVA in both method and contrast orientation, so its limits should not be substituted for the sponsor’s reported interval. The registry’s reported analysis does not include a p-value.

In general, a confidence interval expresses the uncertainty associated with an estimate under the analysis assumptions. A wider interval indicates less precision than a narrower interval on the same scale. Interval width can be affected by outcome variability, the amount of information available, and the analysis model. It does not describe the range of individual participant outcomes.

For the Week 52 denosumab trough concentration comparisons, the reported estimates also have broad intervals, and the comparison with the maintenance Prolia group includes zero. The switched-group comparison likewise has an interval spanning zero. These results are less precise than a point estimate alone might suggest, particularly given the large standard deviations reported for the concentrations. They concern pharmacokinetics, not the primary BMD outcome.

The registry does not report the equivalence margin needed to interpret the primary 90% interval against a formal equivalence criterion. Thus, the interval conveys uncertainty about the estimated contrast, but the available record does not establish whether it meets the trial’s intended equivalence standard.

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 planning figures are participants per arm for comparisons of continuous outcomes. They are not total event counts. The figures use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning scenario uses a stated fraction of the observed effect as the effect to detect. The original trial’s own design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the number of participants per arm estimated to give an 80% chance of detecting the specified effect under the planning assumptions, if that effect is the true effect and the assumptions hold. The 90% power column gives the corresponding planning target for a 90% chance. These are design-planning concepts, not guarantees about what a future study will achieve.

The rows distinguish planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller effects are harder to distinguish from sampling variation, so they require more participants per arm under the same planning assumptions. The estimates for the Week 52 concentration comparisons also differ between comparator groups because their observed effects and variability differ.

These values are planning illustrations based on the observed effect, not observed power. They should not be read as the actual planned sample size for this trial or as evidence that the completed study had a particular probability of detecting an effect. The registry’s enrollment was 479, but the record does not state that enrollment was determined using these illustrative calculations.

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 = 0.193; pooled standard deviation σ = 4.571.
2. n = 2 × (1.96 + 0.842)² × 4.571² / 0.193² = 8,797.5, rounded up to 8,798 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 <- 0.1931                      # observed difference in means
sigma <- 4.57132868383821                     # 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(4757376)
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 = 8,798 (80% power), 11,778 (90% power)
 75% of observed effect: n per arm = 15,640 (80% power), 20,938 (90% power)
 50% of observed effect: n per arm = 35,190 (80% power), 47,110 (90% power)
power.t.test: 8798.4 
Simulated power at n = 8,798 per arm: 0.803 (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’s own primary analysis is an ANCOVA reported as an equivalence analysis. It gives a mean difference of -0.19 and a 90% confidence interval from -0.76 to 0.38. The registry does not report a p-value for this analysis. The equivalence bounds and the decision rule are not reported in the registry record, so the formal equivalence assessment cannot be reconstructed from the posted estimate and interval alone.

The separate recalculation uses the posted group summaries and produces an unadjusted mean difference of 0.193 for US-licensed Prolia relative to CT-P41, with a 95% confidence interval from -0.668 to 1.054 and an unadjusted p-value of 0.66. Its direction is opposite to the sponsor estimate because the group contrast is expressed in the opposite order. It is also unadjusted and does not apply the stated ANCOVA and multiple-imputation approach.

The two results are not marked as discrepant in the registry analysis. The recalculation is best understood as a simple comparison that can help readers understand the posted summaries; it does not replace the trial’s reported analysis. The sponsor result should be interpreted using its own analysis description, while recognizing that important details, including the equivalence margin, are not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The primary endpoint is identified as percent change from baseline in lumbar spine BMD at Week 52, and its analysis description specifies ANCOVA with multiple imputation under a missing-at-random assumption. The registry record does not report a complete statistical analysis plan, the specific ANCOVA covariates, or the equivalence bounds. It also does not report a multiplicity strategy for the secondary outcomes.

Multiplicity matters when several outcomes or comparisons are examined. In general, each additional statistical test creates another opportunity for a result to appear notable by chance. A prespecified hierarchy, adjustment, or other strategy can control how such findings are interpreted. No such approach is reported in the registry record, so secondary results should not be treated as though they were adjusted for multiple testing.

The primary endpoint description states that multiple imputation was used and that missingness was assumed to be MAR. The number of missing endpoint values, the imputation model, the number of completed datasets, and sensitivity analyses under alternative missing-data assumptions are not reported in the registry record. The observed difference between the FAS counts and the counts listed for the endpoint is not explained in the record.

The FAS includes participants who received at least 1 full dose. The registry reports 239 participants in the CT-P41 FAS and 238 in the US-licensed Prolia FAS. The primary endpoint results list 222 and 212 participants, respectively. The record does not explain how these analysis counts were determined or whether additional analysis populations were used for the reported ANCOVA.

Interim analyses, stopping rules, prespecified subgroup analyses, and the original sample-size justification are not reported in the registry record. Without those details, readers cannot evaluate those aspects of the analysis plan or determine how the posted analyses relate to any other planned analyses.

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: cataract (1.0%), hormone level abnormal (1.0%), benign neoplasm of adrenal gland (1.0%), epistaxis (1.0%), angina unstable (0.9%).

The registry lists serious adverse events for the initial CT-P41 and US-licensed Prolia groups, as well as separate maintenance and switching groups. In the initial groups, 7 of 239 participants in the CT-P41 group and 10 of 238 in the US-licensed Prolia group are listed as affected by serious adverse events. The registry lists 1 death in the CT-P41 group and 0 in the US-licensed Prolia group.

For the maintenance and switching groups, the serious adverse event counts are 8 of 220 for CT-P41 Maintenance, 3 of 100 for US-licensed Prolia Maintenance, and 0 of 101 for participants who switched to CT-P41. Death counts are 1, 0, and 0, respectively. The record presents these as distinct groups; they should not be added together to create a combined total because the relationship between groups and study periods is not fully described.

The registry also lists serious adverse event terms with percentages, including Cataract, Hormone level abnormal, Benign neoplasm of adrenal gland, and Epistaxis at 1.0% each, Angina unstable at 0.9%, and Gastrointestinal perforation at 0.5%. These figures describe posted safety information, but the record does not provide enough context to compare event timing or exposure across all groups. Safety counts are descriptive and do not establish that an intervention caused an event.

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

The primary result describes estimated mean percent change in lumbar spine BMD at Week 52 for the two randomized groups. The estimates are close in magnitude, and both the registry’s reported interval and the recalculated interval include zero. This indicates uncertainty about the direction and size of the between-group contrast; it does not prove that the groups are identical.

The registry labels the sponsor’s ANCOVA as an equivalence analysis. A formal equivalence interpretation depends on prespecified margins and a decision rule, neither of which is reported in the registry record. Without those details, the reported estimate and 90% interval cannot establish whether the study met its equivalence criterion. Likewise, a non-significant unadjusted p-value is not a substitute for an equivalence assessment.

The primary outcome is BMD change, not fracture incidence, symptoms, quality of life, or another clinical outcome. The study lists fracture outcomes among its secondary measures, but numerical results for those measures are not reported in the registry record. The BMD result therefore should not be presented as a conclusion about fracture outcomes.

The trial’s findings also apply to the population and analysis described in the registry: postmenopausal women with osteoporosis who entered the study and, for the FAS, received at least 1 full dose. They do not by themselves establish results for people outside that population or for other patterns of treatment. The primary endpoint analysis used multiple imputation under a MAR assumption, and the record does not report enough detail to evaluate that assumption or the sensitivity of the result to missing data.

In general, statistical uncertainty is not the same as proof of no difference, and an observed estimate is not a guarantee of the true effect. The posted results support a bounded description of the measured endpoint and its uncertainty, rather than a broad claim about all outcomes or all patients.

18. Why This Trial Matters Statistically

This trial offers a randomized, active-controlled comparison of CT-P41 and US-licensed Prolia in postmenopausal women with osteoporosis, with a primary endpoint based on change in lumbar spine BMD. Its design includes masking and a parallel-group structure, which are important features for interpreting a direct treatment comparison. The primary result is accompanied by a reported ANCOVA estimate and confidence interval, allowing readers to consider both the estimated contrast and its precision.

The record is also a useful example of why the analysis label matters. The sponsor reports an equivalence analysis, while a simple recalculation from group summaries is an unadjusted difference comparison. Those approaches answer related but not identical statistical questions. The equivalence margin is not reported, so the formal interpretation cannot be completed from the posted result alone.

More broadly, the trial illustrates how one study can include outcomes with different meanings: BMD, fractures, drug concentrations, and a pharmacodynamic marker. A result in one domain should not be taken as a result in another. Secondary outcomes also require care when multiple comparisons are made, especially when the registry does not describe a multiplicity strategy.

Finally, the record demonstrates the value of reporting analysis populations, missing-data methods, and uncertainty alongside estimates. It identifies the FAS and the MAR assumption but leaves important implementation details unreported. Those limits do not erase the posted results; they define how narrowly and cautiously those results can be interpreted.

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 primary analysis description does not identify the ANCOVA covariates, the full model specification, or the equivalence margin. These omissions limit independent assessment of the reported model and prevent evaluation of the formal equivalence criterion. The registry does not report a p-value for the sponsor’s primary analysis.

The missing-data description names multiple imputation and a MAR assumption, but does not state how many primary endpoint observations were missing, how imputation was implemented, or whether sensitivity analyses examined other assumptions. The endpoint counts are lower than the FAS counts, and the reason for that difference is not reported in the registry record.

The record does not report the original sample-size justification, interim analyses, stopping rules, or a multiplicity strategy. These details would help readers understand the study’s design targets and how the collection of secondary outcomes was intended to be interpreted. The planning illustrations based on observed effects are not a substitute for those original design details.

Numerical results are not reported in the registry record for most of the listed secondary domains. This prevents a complete assessment of the pattern across BMD sites, fracture outcomes, and pharmacodynamic measures. The detailed concentration results have substantial variability and should be interpreted as pharmacokinetic findings, not as direct evidence about the primary endpoint.

The safety record presents several named groups, including maintenance and switching groups, but does not fully explain their relationship to study periods or participant overlap. Their counts should therefore be read as group-specific descriptive information rather than combined into an overall total. The registry information also does not establish causation for individual adverse events.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Phase 3 Study to Compare Between CT-P41 and US-licensed Prolia in Postmenopausal Women With Osteoporosis — NCT04757376.
• Registry record last updated 2024-06-14; status last verified 2024-05.
• Results first posted 2024-05-16.
• 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

NCT04757376 is a completed Phase 3 randomized, quadruple-masked, parallel-group, active-controlled study of CT-P41 and US-licensed Prolia in postmenopausal women with osteoporosis. Enrollment was 479. The recorded study start date was 2021-06-17, and the primary completion date was 2023-05-18.

The primary endpoint was percent change from baseline in lumbar spine BMD at Week 52, assessed by DXA. The registry reports an ANCOVA mean difference of -0.19 with a 90% confidence interval from -0.76 to 0.38 and labels the analysis as equivalence. The equivalence bounds are not reported in the registry record. An unadjusted recalculation gives a mean difference of 0.193 for US-licensed Prolia relative to CT-P41, with a 95% confidence interval from -0.668 to 1.054.

The record lists additional BMD, fracture, pharmacokinetic, pharmacodynamic, and safety outcomes. Numerical results for most secondary outcomes and important details of the analysis plan are not reported in the registry record.