A Study to Characterize Diclofenac's Plasma and Knee Exposure After Application of Diclofenac Gel to the Knee of Subjects With Osteoarthritis That Have a Scheduled Arthroplasty: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Characterize Diclofenac's Plasma and Knee Exposure After Application of Diclofenac Gel to the Knee of Subjects With Osteoarthritis That Have a Scheduled Arthroplasty (NCT03497039), a randomized phase 1 study. The primary outcome is Diclofenac Concentration in Treated Knee Synovial Tissue After 7 Days Topical Application of Study Treatment to the Knee (12 Hours After Last Administration of Study Treatment).

2. Clinical Question

The study also registered secondary outcomes concerning the ratio between diclofenac concentration in treated-knee synovial tissue and plasma, and the ratio between concentration in synovial fluid and plasma. These outcomes describe distribution across sampled compartments. The record identifies 2 secondary outcomes, but does not report analyses for them.

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-arm Phase 1 study. The registry record identifies quadruple masking. Participants were assigned to topical DDEA gel or placebo gel, with treatment applied to the knee scheduled for arthroplasty. Both groups had the same application schedule: 4 gram (g) over 400 square centimeter (cm^2) twice a day for 7 days.

The study's stated purpose was to characterize diclofenac exposure in plasma and in the knee joint after topical administration. Surgery made it possible to collect synovial tissue, and the registered primary endpoint was the diclofenac concentration in treated-knee synovial tissue. The collection was scheduled for Day 8, up to a maximum of Day 15 in case of surgery delay. The endpoint description specifies collection during surgery, 12 hours after the last administration of study treatment.

The registry record reports enrollment of 47 participants and a completed study. It gives the start date as 2018-07-19 and primary completion date as 2019-03-01. For the primary endpoint, the record lists 29 participants in the DDEA gel group and 16 in the placebo gel group. Safety counts use different at-risk totals: 29 in the active-treatment group and 17 in the placebo group. These are endpoint- and analysis-specific figures; they should not be treated as interchangeable denominators.

The primary endpoint is an exposure measure, not a comparison of pain or other clinical outcomes. Although the study was randomized and placebo-controlled, the posted primary result contains an active-group concentration estimate but no placebo concentration estimate or reported treatment-comparison analysis. A detailed randomization procedure, allocation concealment procedure, and protocol rationale for the study design 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.

Before assignment: A total of 47 participants with osteoarthritis were enrolled out of which 46 were randomly assigned in 2 arms either to receive Diclofenac Diethylamine (DDEA) 2.32 percent (%) gel treatment or placebo gel. 1 participant withdrew consent before starting treatment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analyzable population consists of all participants who were randomized, received at least 1 dose of investigational product, completed surgery and had evaluable synovial tissue or synovial fluid sample. Here, number analyzed indicates participants with diclofenac tissue concentration values greater than LOQ.

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: Safety population included all participants who were randomized and received at least one dose of investigational product.

6. Primary Endpoint

The registry lists the primary outcome as Diclofenac Concentration in Treated Knee Synovial Tissue After 7 Days Topical Application of Study Treatment to the Knee (12 Hours After Last Administration of Study Treatment), measured over At Day 8 (up to a maximum of Day 15 in case of surgery delay).

The primary endpoint is the concentration of diclofenac in synovial tissue from the treated knee, measured in nanograms per gram (ng/g). Tissue was collected during the participant's scheduled arthroplasty of the target knee. The registered collection time was Day 8, with collection allowed up to a maximum of Day 15 in case of surgery delay; the endpoint description places sampling 12 hours after the last study-treatment administration.

This measurement was used to characterize local drug exposure in tissue from the treated joint. A tissue concentration can describe how much diclofenac was measured in the sampled material at the specified time. It does not, by itself, describe the concentration throughout the joint, the duration of exposure, or a clinical response. Concentration in tissue is also distinct from concentration in plasma and from concentration in synovial fluid. The record registered separate secondary outcomes for ratios of tissue or fluid concentration to plasma concentration.

The primary outcome was summarized by treatment group as a geometric mean with a two-sided 95% confidence interval. The registry record says that values below the limit of quantification were replaced by 0 for the placebo group. The posted result gives an estimate for the DDEA gel group, but no placebo estimate is reported. Thus, the reported primary endpoint provides a description of measured tissue exposure in the active-treatment group rather than a quantified contrast between treatments.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a geometric mean describe?

A geometric mean is a summary used for measurements that are analyzed on a logarithmic scale. It is calculated by transforming observations to the log scale, averaging there, and transforming the average back to the original scale. This makes it suitable for concentration data when the statistical model treats variation as multiplicative rather than additive. It is not the same summary as an arithmetic average, and it does not mean that every participant had a concentration near the reported estimate.

The registry record specifies a geometric mean for the tissue concentration and says that the confidence interval assumes normality on the logarithmic scale. The reported estimate of 1.573 ng/g and interval from 1.124 to 2.2 ng/g should be understood in that framework. The interval quantifies uncertainty in the estimated group geometric mean under the stated model. It does not describe the spread of individual concentrations; a separate measure of participant-level dispersion is not reported.

How should the confidence interval be read?

A confidence interval communicates how precisely a group-level quantity has been estimated, conditional on the sampling and statistical assumptions. A narrower interval would generally indicate more precision for the quantity being estimated, while a wider interval would indicate less. The interval applies to the active-treatment group's geometric mean concentration. Since the placebo estimate is not reported, there is no interval for a difference or ratio between randomized groups to interpret.

A confidence interval is not a guarantee that an individual participant's value lies within its limits. It also does not address every possible source of uncertainty. For example, the interval does not by itself resolve whether the enrolled and analyzable participants represent other populations, whether tissue concentration is a useful surrogate for clinical outcomes, or how concentrations vary across regions of a joint. Those are distinct questions from precision of the estimated geometric mean.

Why use the logarithmic scale?

Concentrations can be positive and can vary in a way where proportional changes are more natural to model than fixed additive changes. A logarithmic transformation can make a skewed distribution more symmetric and can support statistical methods that assume approximate normality. This is a general rationale for log-scale analysis; the registry record specifically states the log-scale normality assumption for the confidence interval but does not describe diagnostic checks or alternative analyses.

Measurements below a laboratory's limit of quantification present a special issue. Such observations indicate that a concentration could not be quantified reliably under the assay's defined conditions; they are not necessarily identical to a true concentration of zero. The registry record says that values below the limit of quantification were replaced by 0 for the placebo group. It does not report the number of such values or describe a sensitivity analysis using another approach. The consequences of the substitution therefore cannot be assessed from the posted information.

What can randomization and masking contribute?

Randomization is intended to reduce systematic differences between treatment groups at assignment, so that a comparison can be less vulnerable to confounding than an observational comparison. Masking is intended to limit influences from awareness of assigned treatment. These are general design principles, not a guarantee that groups remain identical or that all sources of bias are eliminated. The record identifies randomized assignment, a parallel-arm design, and quadruple masking, but does not report the detailed procedures used to generate or conceal assignments or the roles covered by the masking label.

Even when a study is randomized, a treatment contrast requires results for both arms and an analysis that compares them. Here, the primary result includes an active-group estimate but no placebo concentration estimate or reported comparison. The design features alone cannot supply the missing numerical contrast.

What do the planning illustrations represent?

The registry record describes planning illustrations based on a fraction of the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation. These are assumptions for hypothetical sample-size planning. A calculation of this kind asks how many participants per arm might be needed under its assumptions to detect a specified effect with a chosen power. It is not a record of the trial's original sample-size justification, and it is not observed power. The trial's own sample-size justification is not reported in the registry record.

General sample-size calculations depend on the assumed effect, variation, test, allocation, and target power. When an effect is estimated from a small study, planning based on that estimate may be uncertain. Illustrations based on smaller fractions of an observed effect can show how planning changes under different assumptions, but they do not establish that the effect is clinically important or likely to recur. The record's planning description does not supply a basis for treating such illustrations as evidence about treatment efficacy.

There are no reported primary analyses or treatment-comparison statistics in the registry record: it lists 0 analyses. Therefore, explanations of confidence intervals, randomization, and planning assumptions clarify how to read the posted information, but cannot replace an unreported comparison or an unreported analytic plan.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes or makes several statistical comparisons. If many tests are conducted, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for the testing structure. The registry record lists 4 outcome measures, including the primary tissue concentration and 2 secondary concentration-ratio outcomes. It does not report analyses for these outcomes, and a multiplicity procedure or hierarchy for interpreting multiple outcomes is not reported in the registry record.

The record identifies a primary endpoint, which distinguishes it from secondary outcomes, but does not describe whether further endpoints or subgroup analyses were planned. An analysis plan specifying adjustment, gatekeeping, or another approach to multiple testing is not reported in the registry record. No inference should therefore be made about how multiplicity was handled beyond what the record explicitly states.

The primary endpoint population is defined as randomized participants who received at least 1 dose, completed surgery, and had an evaluable synovial tissue or synovial fluid sample. This is a post-randomization, evaluability-based definition rather than a statement that every randomized participant contributed to the primary endpoint. The record lists 29 participants for the active group and 16 for placebo for the primary result. It does not report reasons why other participants may not have contributed to those endpoint counts.

The record describes one treatment of below-limit-of-quantification values: values in the placebo group were replaced by 0. The number of affected samples and any other missing-data rules, imputation procedures, or sensitivity analyses are not reported in the registry record. The endpoint-specific analysis population and this laboratory-value handling are important context when interpreting the reported estimate; they do not establish how other missing observations were treated.

11. 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: escherichia urinary tract infection (3.4%), hypotension (3.4%).

The safety record lists serious adverse events among 29 participants at risk in the DDEA gel group and 17 in the placebo group. It reports 1 participant affected in the DDEA gel group and 0 in the placebo group. The record also lists Escherichia urinary tract infection and hypotension among the top serious-event terms, each at 3.4%. It does not identify an arm for those terms in the listed summary, so they should not be assigned to a treatment group on that basis.

Other adverse events are reported for 16 of 29 participants in the DDEA gel group and 10 of 17 in the placebo group. The record lists 0 deaths in each group. These are descriptive counts from the posted safety information. They do not establish that a treatment caused an event, and the counts alone do not provide a comparative safety analysis. The record does not report a statistical test or adjusted comparison of adverse-event frequencies.

Safety denominators differ from the primary endpoint counts, particularly for placebo, for which the primary result lists 16 participants and the safety summary lists 17 at risk. Such differences can arise because endpoints use different analysis populations or data-availability rules; the specific explanation is not reported in the registry record. Serious adverse event counts are also distinct from the count of participants with other adverse events. They should not be added together or treated as equivalent categories.

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

The result establishes that the registry record reports a geometric mean diclofenac concentration in treated-knee synovial tissue for the DDEA gel group: 1.573 ng/g, with a 95% confidence interval from 1.124 to 2.2 ng/g. The measurement was made in surgically collected tissue after the registered topical regimen and at the specified post-dose sampling time. It is a description of tissue exposure for the participants included in that endpoint group.

The result does not establish how that concentration compares with placebo. Although the study had a randomized placebo group and the primary endpoint lists 16 placebo participants, a placebo concentration estimate is not reported in the registry record. A numerical contrast, its uncertainty, and a significance test are therefore not available in the posted primary result. The active-arm estimate should not be treated as an estimate of the treatment effect.

The concentration result also does not establish a change in pain, function, or quality of life. The trial's primary measurement concerns drug concentration in tissue, and the posted record does not present a clinical efficacy result. Tissue exposure can be relevant to understanding where a drug is measured after application, but it is not by itself proof of a clinical benefit or harm.

Nor does a single tissue-sampling time describe the full time course of exposure. The study registered plasma and knee exposure as its purpose and included secondary concentration-ratio outcomes, but their results are not reported in the registry record. The available primary estimate does not characterize plasma exposure, synovial-fluid concentration, exposure at other times, or concentrations in other tissues.

Finally, the analysis applies to a specific group: participants with osteoarthritis undergoing scheduled arthroplasty who met the endpoint's evaluability definition. The registry result should be interpreted within that setting and regimen. It does not, on its own, answer questions about other populations, application schedules, or clinical outcomes.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of the distinction between measuring exposure and estimating a treatment contrast. A randomized, placebo-controlled design can support a comparison when appropriate outcomes are reported for both groups. In the posted primary result, the active group's geometric mean and confidence interval are available, but the placebo concentration estimate and a between-group analysis are not. The study's design and its available numerical result therefore answer different parts of the research question.

The endpoint also illustrates why the measurement scale and summary statistic matter. Synovial tissue concentration is reported as a geometric mean, with uncertainty estimated under a log-scale normality assumption. Readers need to distinguish that group summary from individual values and from an arithmetic mean. The confidence interval describes precision for the active-group summary, not the range of participant concentrations and not the uncertainty in a treatment difference.

Sample collection during scheduled arthroplasty makes it possible to measure drug concentration in joint tissue directly, but it defines a particular study context and an evaluable population. Participants needed to receive treatment, complete surgery, and have an evaluable sample for the endpoint population. These conditions help explain why the endpoint counts need not match enrollment or safety counts, while also limiting the scope of the result.

The record lists secondary concentration-ratio outcomes and a primary outcome, yet reports no analyses. This underscores the importance of complete outcome reporting and a clearly described analysis plan when multiple measurements are registered. The statistical interpretation depends not only on the endpoint definition and point estimate, but also on knowing which groups were analyzed, how below-quantification observations were handled, and whether a formal comparison was conducted.

14. 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 most important limitation of the posted primary result is that it reports a DDEA gel concentration estimate but no placebo estimate or formal between-group comparison. Randomization does not make an unreported comparison recoverable from the active-arm result. The primary result therefore cannot quantify the difference in tissue concentration between study groups.

The endpoint is based on tissue collected during arthroplasty, at a defined time after the last application. The record specifies Day 8, with collection allowed up to a maximum of Day 15 in case of surgery delay, and 12 hours after the last administration. This is a particular sampling context rather than a description of exposure over time. The record does not report the distribution of actual collection days or the impact of any delay.

The endpoint population required receipt of at least 1 dose, completed surgery, and an evaluable sample. The reasons for not contributing to the listed endpoint counts are not reported in the registry record. A detailed accounting of exclusions, missing samples, or participant flow is also not reported. The endpoint counts and safety denominators differ, and the record does not explain each difference.

The registry specifies replacement of below-limit-of-quantification placebo values by 0, but does not report how many values were affected or whether alternative handling was examined. The result also depends on a log-scale normality assumption for the confidence interval; model checks and sensitivity analyses are not reported. The spread of individual observations is not reported.

Finally, outcome analyses are not reported for the 2 secondary concentration-ratio measures, and the record lists 0 analyses. The statistical analysis plan, multiplicity strategy, and trial-specific sample-size justification are not reported in the registry record. These gaps limit assessment of the completeness and robustness of the intended analyses.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study to Characterize Diclofenac's Plasma and Knee Exposure After Application of Diclofenac Gel to the Knee of Subjects With Osteoarthritis That Have a Scheduled Arthroplasty — NCT03497039.
• Registry record last updated 2020-02-24; status last verified 2020-02.
• Results first posted 2020-02-24.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT03497039 is a completed randomized, double-blind, placebo-controlled, parallel-arm Phase 1 study of topical DDEA gel versus placebo gel in participants with osteoarthritis scheduled for arthroplasty. The registry record reports enrollment of 47 participants. Both treatments were applied to the target knee at 4 gram (g) on 400 square centimeter (cm^2) twice a day for 7 days.

The primary endpoint was diclofenac concentration in treated-knee synovial tissue, collected during surgery after the last administration. The active-treatment group had a reported geometric mean of 1.573 ng/g, with a 95% confidence interval from 1.124 to 2.2 ng/g, among 29 participants. A placebo concentration estimate and a treatment-comparison analysis are not reported in the registry record. The record identifies 2 secondary concentration-ratio outcomes but does not report their analyses. The study started on 2018-07-19 and reached primary completion on 2019-03-01.