A Study of Guselkumab and Golimumab Combination Therapy in Participants With Active Psoriatic Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Guselkumab and Golimumab Combination Therapy in Participants With Active Psoriatic Arthritis (NCT05071664), a randomized phase 2 study. The primary outcome is Percentage of Participants Who Achieved Minimal Disease Activity (MDA) at Week 24.

2. Clinical Question

The registry reports MDA achievement of 28.8% in the guselkumab-plus-golimumab group and 21.9% in the guselkumab-plus-placebo group. The posted unadjusted comparison estimates a risk difference of 0.069, with an interval from -0.115 to 0.253 and p=0.473. The estimate points toward a higher observed response with the combination, but the interval includes values on either side of no difference. The result therefore does not establish a clear between-group difference in this endpoint.

The study also assessed other outcomes, including clinical response, disease activity, skin response among participants meeting baseline skin criteria, physical function, enthesitis and dactylitis. These outcomes provide additional perspectives on disease status, but they do not change the definition of the primary question. Their interpretation depends on their own estimates, uncertainty and analysis methods.

3. Trial Design

This was a Phase 2a, multicenter, randomized, double-blind, parallel-group study. The registry reports enrollment of 91 and a completed study. The recorded start date was 2021-10-25, and the primary completion date was 2024-05-14. The sponsor was Janssen Research & Development, LLC.

Random assignment is intended to make comparison groups similar, on average, with respect to both measured and unmeasured characteristics. Double blinding is intended to limit the influence of treatment knowledge on participant experience, clinical assessments and study conduct. These design features support a comparative interpretation, although they cannot eliminate uncertainty in results or guarantee that all sources of bias were prevented. The registry record does not report which specific people were masked.

The active treatment group received guselkumab with golimumab, while the comparator group received guselkumab with placebo. Because guselkumab was common to both groups, the randomized comparison addresses the addition of golimumab in this study context, not guselkumab versus no guselkumab. The record describes subcutaneous administration and lists the treatment schedules as guselkumab 100 mg q4w, with golimumab 50 mg q4w in the combination group.

The primary outcome was the percentage achieving MDA at Week 24, assessed in the full analysis set. The registry defines that set as all participants randomized. Other reported outcomes include ACR 50 at Week 24 and MDA at Week 16, as well as several skin, function and disease-domain outcomes. The registry record does not report the original sample-size justification, detailed allocation procedures, treatment adherence, or the prespecified statistical analysis plan. Those details should not be inferred from the posted outcome comparisons.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) included all participants who were randomized in this study.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Who Achieved Minimal Disease Activity (MDA) at Week 24, measured over Week 24.

The primary endpoint was the percentage of participants who achieved minimal disease activity at Week 24. MDA is a composite state intended to represent satisfactory disease activity across multiple PsA domains: joint symptoms, skin psoriasis, the participant's perspective on pain and disease activity, physical function and enthesitis. A composite can capture improvement across different aspects of a condition, although it also means that a single percentage does not show which component criteria were met by each participant.

The registry classifies a participant as achieving MDA when at least five of seven criteria are fulfilled at the visit. These criteria include tender joint count (68 joints) <=1, swollen joint count (66 joints) <=1, PASI <=1, pain assessment <=15 on a 100-unit VAS, global assessment of disease activity <=20 on a 100-unit VAS, HAQ-DI <=0.5, and tender entheseal points <=1 using the Leeds Enthesitis Index. The outcome is reported as a percentage, so it summarizes the proportion reaching the composite threshold rather than the average level of each individual component.

MDA is useful for a comparative trial question because it applies a defined response criterion to both randomized groups at the same assessment time. It should not be read as evidence that every domain improved, or as a direct measure of change from baseline. The registry record does not report component-level primary outcome results.

7. Results Posted in the Registry

• Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w: 28.8% (17/59) versus 21.9% (7/32) on control; risk difference 6.9 percentage points (95% CI -11.5 to 25.3), risk ratio 1.32 (0.61–2.84); unadjusted two-proportion z-test p = 0.473.

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.

Percentage of Participants Who Achieved American College of Rheumatology (ACR) 50 at Week 24

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w: 44.1% (26/59) versus 21.9% (7/32) on control; risk difference 22.2 percentage points (95% CI 3.1 to 41.3), risk ratio 2.01 (0.99–4.12); unadjusted two-proportion z-test p = 0.035.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieved Minimal Disease Activity (MDA) at Week 16

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w: 32.2% (19/59) versus 12.5% (4/32) on control; risk difference 19.7 percentage points (95% CI 3.2 to 36.2), risk ratio 2.58 (0.96–6.92); unadjusted two-proportion z-test p = 0.039.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieved Psoriasis Area and Severity Index (PASI) 90 Response at Week 24 Among Participants With >= 3% Body Surface Area (BSA) Psoriatic Involvement and an Investigator Global Assessment (IGA) Score of >=2 (Mild) at Baseline

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w: 54.5% (12/22) versus 38.5% (5/13) on control; risk difference 16.1 percentage points (95% CI -17.6 to 49.7), risk ratio 1.42 (0.65–3.11); unadjusted two-proportion z-test p = 0.358.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Group 1: Guselkumab 100 mg q4w Plus Golimumab 50 mg q4w and the control. The interval gives the range of effects that remain compatible with the data.

The reported secondary outcomes cover clinical response, disease activity, skin response in a baseline-defined group, physical function, enthesitis and dactylitis. Among the secondary comparisons with numerical results in the registry record, the response percentages are higher in the combination group than in the guselkumab-plus-placebo group. This descriptive direction is consistent with the primary point estimate, which also favors the combination group, but it does not remove the uncertainty in the primary comparison.

The secondary results do not all answer the same question. They use different outcome definitions and assessment times, and some apply only to participants meeting specified baseline skin or disease-domain criteria. Their estimates therefore should not be treated as interchangeable or as multiple measurements of a single endpoint. Outcomes assessed in a selected baseline subgroup may also be based on less information than the overall primary analysis.

Unadjusted secondary p-values carry less weight when many outcomes are examined. Each additional comparison creates another opportunity for a small p-value to arise by chance, even if there is no corresponding underlying effect. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting these secondary outcomes. Their results are therefore better regarded as supportive or exploratory rather than independent confirmation of the primary finding.

The registry lists a broader set of secondary outcomes than the numerical comparisons. A complete assessment across all outcomes would require their estimates, uncertainty, analysis methods and missing-data handling. The record does not provide a basis for declaring a consistent effect across every listed domain. The most defensible reading is that the reported secondary comparisons generally point in the same direction as the primary estimate, while remaining subject to uncertainty and multiplicity.

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

• Percentage of Participants Who Achieved American College of Rheumatology (ACR) 50 at Week 24
• Percentage of Participants Who Achieved Minimal Disease Activity (MDA) at Week 16
• Percentage of Participants Who Achieved Psoriasis Area and Severity Index (PASI) 90 Response at Week 24 Among Participants With >= 3% Body Surface Area (BSA) Psoriatic Involvement and an Investigator Global Assessment (IGA) Score of >=2 (Mild) at Baseline
• Percentage of Participants Who Achieved PASI 100 Response at Week 24 Among the Participants With >=3% BSA Psoriatic Involvement and an IGA Score of >=2 (Mild) at Baseline
• Percentage of Participants With an IGA-psoriasis Response at Week 24 Among Participants With >=3% BSA Psoriatic Involvement and an IGA Score of >=2 (Mild) at Baseline
• Change From Baseline in Health Assessment Questionnaire-Disability Index (HAQ-DI) at Week 24
• Percentage of Participants Who Had Resolution of Enthesitis at Week 24 Among the Participants With Enthesitis at Baseline
• Percentage of Participants Who Achieved Resolution of Dactylitis Response at Week 24 Among the Participants With Dactylitis at Baseline

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant either met the MDA definition at Week 24 or did not. For a binary outcome, the group response percentage describes the observed proportion meeting the criterion. The primary results report 28.8% in the combination group and 21.9% in the guselkumab-plus-placebo group. These percentages are descriptive summaries of the trial groups; by themselves they do not quantify uncertainty in the comparison.

What does a risk difference express?

A risk difference subtracts the response proportion in the comparator group from the response proportion in the treatment group. It is an absolute measure: it expresses the gap between response proportions on the proportion scale. The unadjusted recalculation is 0.069, with an interval from -0.115 to 0.253. A value of zero corresponds to no difference between group proportions. The interval includes zero, so the posted estimate does not pin down whether the underlying difference is positive or negative.

Risk differences are often straightforward to interpret because they describe absolute separation between group probabilities. They do not adjust for other variables. The recalculation is explicitly unadjusted and is based on the posted group summaries. It is not a replacement for a prespecified adjusted analysis, and the registry record does not report enough of the statistical plan to establish whether this recalculation was the planned primary method.

What does the p-value add?

The recalculated unadjusted p-value is 0.473. A p-value describes how unusual results at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It does not give the probability that the null hypothesis is true, and it does not measure the probability that the findings will replicate. A p-value should be considered alongside the effect estimate, its interval and the design, rather than used as a stand-alone decision rule.

The sponsor's reported analysis used logistic regression and gives an odds ratio, a 90% confidence interval and a p-value. Logistic regression models the odds of a binary outcome. Odds are not the same as probabilities, so an odds ratio should not be read as a direct percentage-point change or as the risk difference. The registry record does not report the covariates, link details or other model specifications, so further claims about adjustment cannot be made.

Why can the analyses differ?

The recalculation and the registry-reported analysis use different effect measures and methods: one is a simple unadjusted risk difference, while the other is a logistic-regression odds ratio. Differences in scale and model can produce different numerical estimates and p-values without indicating a data error. The registry does not mark a discrepancy between its results and the recalculation. The reported analyses should therefore be understood according to their separate methods, not blended into a single estimate.

Confidence intervals communicate uncertainty around an estimate. For a risk difference, an interval crossing zero encompasses no difference as well as differences in either direction. For an odds ratio, the corresponding no-association value is one. The sponsor's 90% interval is 0.6 to 3.3, and it includes one. The recalculated interval's confidence level is not stated in the registry record, so comparisons of interval widths across methods should be cautious.

Why is over-dispersion relevant?

Over-dispersion is a general concern when data vary more than a statistical model expects. It can make standard errors too small if the model understates variability, leading to intervals that are too narrow and p-values that are too optimistic. The primary outcome is a participant-level binary response, and the registry record does not report an over-dispersion assessment or a model addressing it. There is no basis in the record to assert that over-dispersion was present or that a particular correction was needed.

The registry also does not report the detailed handling of missing Week 24 outcomes, the treatment of intercurrent events, or sensitivity analyses for the primary endpoint. Those choices can affect a binary response analysis. The full analysis set is defined as all randomized participants, but that population definition alone does not specify how missing endpoint values were treated. These methodological details remain not reported in the registry record.

11. Confidence Intervals and Precision

The primary comparison has a risk-difference estimate of 0.069 and an interval from -0.115 to 0.253. The interval includes zero and extends to values favoring either group. This width indicates that the estimate is not precise enough to rule out a range of possible between-group effects. It is not appropriate to treat the point estimate alone as the definitive size or direction of an underlying treatment effect.

Confidence intervals are a way to describe sampling uncertainty under an analysis method. In repeated applications of the same method under its assumptions, intervals constructed at a stated confidence level would contain the true value at the stated long-run rate. A confidence interval is not a guarantee that the true effect lies within a particular interval, nor does it show the probability that a specific effect is true. The confidence level for the recalculated risk-difference interval is not reported in the registry record.

The sponsor's logistic-regression result is reported as an odds ratio of 1.4, with a 90% confidence interval from 0.6 to 3.3. Because this interval includes one, it is compatible with no odds association as well as effects in either direction. Its scale and model differ from the unadjusted risk-difference recalculation, so the interval endpoints should not be compared as though they measured the same quantity.

Precision is affected by the amount of information available and by the observed distribution of outcomes. The registry record reports enrollment of 91, but does not report an original precision target or sample-size justification. The planned precision of the trial therefore cannot be evaluated against an aim stated in the record. The result is best described as an uncertain estimate, rather than as evidence of no effect or as a precise estimate of benefit.

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 80% power and participants per arm for 90% power. They are illustrations based on the observed effect, not a report of the trial's original design assumptions. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the treatment groups, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column illustrates a planning scenario in which a study has an 80% chance of detecting the specified effect under the planning assumptions, if that effect is the true one and the statistical model is appropriate. The 90% power column applies the same idea at a higher target probability of detection. These are design probabilities for hypothetical studies, not probabilities that the observed result is correct. The calculations are not observed power.

Each planning row considers a fraction of the observed effect: the full observed effect, 75% of it, or 50% of it. A smaller effect is harder to distinguish from random variation, so the planning illustration requires more participants per arm to target the same power. This is a general feature of study planning, not evidence that a future trial would reproduce the observed effect.

The planning assumptions use equal allocation, whereas these figures are expressed per arm. The sample-size illustration does not establish that the completed trial was designed using these assumptions, and it should not be used as a retrospective claim about whether the study was adequately powered. The registry record does not report the trial's own original planning target, chosen detectable effect or rationale for enrollment.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.219; treatment proportion p₁ = 0.288; average p̄ = 0.253.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6152 = 1.2057.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6132 = 0.5161.
4. n = (1.2057 + 0.5161)² / (0.0694)² = 615.8, rounded up to 616 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
p0 <- 0.21875                          # control proportion
p1 <- 0.288135593220339                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(5071664)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 616 (80% power), 824 (90% power)
 75% of observed effect: n per arm = 1,071 (80% power), 1,433 (90% power)
 50% of observed effect: n per arm = 2,352 (80% power), 3,147 (90% power)
power.prop.test: 615.8 
Simulated power at n = 616 per arm: 0.799 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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-reported analysis uses logistic regression and reports an odds ratio of 1.4, a 90% confidence interval from 0.6 to 3.3, and p=0.557. This is the sponsor's posted analysis for the primary endpoint. The interval includes one, the value corresponding to no odds association, and the p-value does not indicate a clear difference under that analysis.

The simple recalculation uses the posted group response summaries to estimate an unadjusted risk difference of 0.069, with an interval from -0.115 to 0.253 and p=0.473. This is not the same effect scale or model as the reported odds ratio. A risk difference concerns the gap in response proportions; an odds ratio compares odds. Their numerical values and p-values should not be expected to match.

The registry record does not mark a discrepancy between the posted results and the recalculation. The record also does not provide enough model detail to reconstruct the logistic regression or determine the reason for every numerical difference between methods. For that reason, the reported analysis and unadjusted recalculation should be read as distinct summaries. Neither analysis resolves the broad uncertainty in the primary comparison.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several endpoints, time points, subgroups or statistical comparisons. The more comparisons considered, the greater the chance that at least one may appear persuasive by chance alone. This matters particularly for secondary outcomes, which span several PsA domains and include more than one assessment time. Unadjusted results do not account for that collection of comparisons.

The registry record does not report whether a multiplicity adjustment was used, whether outcomes were organized in a testing hierarchy, or which secondary analyses were intended to support confirmatory claims. It also does not report interim analyses or stopping rules. Those details should not be inferred from the posted p-values or from the study's completed status. Without the prespecified plan, the degree to which any secondary result was protected against multiplicity is not known.

The primary outcome is defined for the full analysis set, which the registry says includes all randomized participants. This identifies the population to which the primary endpoint applies, but it does not specify the missing-data method. The registry record does not report the number of missing Week 24 assessments, the method used to handle missing outcomes, or sensitivity analyses testing other assumptions. Missing outcomes can affect a percentage-based response analysis, so the population definition alone is not enough to assess that issue.

The record also does not provide details on adherence, treatment discontinuation, or handling of events that might affect endpoint interpretation. These items are relevant to understanding the contrast represented by an analysis. In the absence of those details, interpretation should remain tied to the reported randomized-group summaries and the analysis descriptions actually available, without assuming unreported procedures.

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 (1.7%), pneumonia mycoplasmal (1.7%), neuroendocrine tumour (1.7%), chronic obstructive pulmonary disease (1.7%).

The registry reports serious adverse events among 4 participants at risk in the guselkumab-plus-golimumab group and 0 participants at risk in the guselkumab-plus-placebo group. The corresponding groups at risk were 59 and 32. The registry also reports 0 deaths in each group. These are descriptive counts during the period covered by the safety record; they are not, by themselves, evidence that a treatment caused or prevented an event.

The listed serious adverse event terms each have a reported frequency of 1.7%: Covid-19, Pneumonia Mycoplasmal, Neuroendocrine Tumour, and Chronic Obstructive Pulmonary Disease. The registry summary does not connect each listed term to a specific treatment group in the information presented, so no group assignment should be inferred from the terms alone.

Safety counts are difficult to interpret without context on event definitions, follow-up duration, exposure, severity, relatedness and missing safety information. Those details are not reported in the registry record as part of these figures. The study's enrollment and observed counts are limited, and rare events may not be characterized reliably in a study of this scale. The counts should be read as a record of reported events, not as a comprehensive comparison of long-term safety.

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

The primary result describes the proportion of participants who reached the MDA threshold at Week 24 in each randomized group. The observed percentages were 28.8% with guselkumab plus golimumab and 21.9% with guselkumab plus placebo. The point estimate for the unadjusted risk difference favors the combination group, but the interval includes no difference and effects in either direction. The result is therefore uncertain rather than conclusive about the between-group contrast.

This finding does not show that every participant experienced the same response, nor does it indicate which MDA criteria drove the group percentages. MDA is a composite that can be achieved through different combinations of its criteria. The registry record does not provide the primary endpoint's component-level results. The result also does not establish effects beyond the reported assessment time or answer questions about outcomes not measured by the posted endpoint.

The comparison concerns adding golimumab to guselkumab in the study population described, versus guselkumab with placebo. It is not a comparison of guselkumab with no active treatment. The trial population had active PsA and an inadequate response to prior anti-TNF-alpha therapies, so the findings should be understood in that setting rather than automatically applied to other populations or treatment contexts.

Failure to demonstrate a clear difference is not proof that the treatments are equivalent. Equivalence or noninferiority requires an appropriate design, a prespecified margin and analysis plan; these are not reported in the registry record. Likewise, a point estimate favoring one group does not establish that the treatment works. The posted evidence should be described through its estimates, uncertainty and study limitations, without converting an uncertain comparison into a definitive clinical conclusion.

18. Why This Trial Matters Statistically

This study illustrates why a clinical trial result should be read through both its effect estimate and its uncertainty. The observed primary response percentages favor the combination group, but the unadjusted risk-difference interval spans zero and values on either side. That distinction prevents a directional point estimate from being mistaken for a settled conclusion.

It also demonstrates that the same endpoint can be summarized on different statistical scales. The unadjusted risk difference describes the absolute separation between response proportions, while logistic regression reports an odds ratio. These measures answer related but distinct questions. Their estimates and p-values are not interchangeable, and an odds ratio is not a percentage-point difference.

The endpoint itself matters. MDA combines criteria from several PsA domains, providing a broad response definition but concealing the pattern of individual components when only an overall percentage is reported. Secondary outcomes can add context, but many comparisons also create a multiplicity concern, particularly when adjustment and testing hierarchy are not reported.

Finally, sample-size planning illustrations should not be confused with the completed trial's design rationale or with observed power. The registry record does not report its original sample-size justification. This is a useful reminder that planning calculations based on an observed effect are conditional illustrations, and can be especially unstable when the observed estimate is uncertain. Clear reporting of planned methods, missing-data handling and multiplicity procedures is needed to assess how confidently 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record leaves important design and analysis details unreported. It does not give the original sample-size justification or the study's full statistical analysis plan. Consequently, it is not possible to confirm the planned primary model, target effect, interim-analysis approach or rules for interpreting secondary outcomes from this record alone.

The primary estimate is imprecise: its unadjusted risk-difference interval includes zero and extends in both directions. The registry-reported logistic-regression interval likewise includes the no-association value. The study therefore does not resolve the size or direction of the primary between-group difference with precision.

The primary endpoint is a composite threshold, and component-level primary results are not reported. The reported percentages cannot show which MDA criteria participants met, whether a particular domain accounted for the response pattern, or how individual participants changed from baseline. The record also does not report the missing-data method, the number of missing primary assessments, or sensitivity analyses. These omissions limit assessment of how robust the endpoint comparison is to incomplete outcome information.

Several secondary outcomes involve distinct time points or restricted baseline populations, and the registry record does not state whether multiplicity was controlled. The available comparisons should not be treated as definitive evidence across all listed outcomes. For safety, the posted serious adverse event counts are descriptive and the record does not provide enough detail on duration of follow-up, event relatedness or exposure to support a broad comparative safety conclusion.

These limitations are specific to what is reported in the registry record. They do not establish that particular procedures were absent; where the details are missing, their status is simply not reported in the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Guselkumab and Golimumab Combination Therapy in Participants With Active Psoriatic Arthritis — NCT05071664.
• Registry record last updated 2026-07-23; status last verified 2026-06.
• Results first posted 2026-07-23.
• 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

NCT05071664 is a completed, randomized, double-blind, multicenter Phase 2a parallel-group study of guselkumab plus golimumab compared with guselkumab plus placebo in participants with active psoriatic arthritis and inadequate response to prior anti-TNF-alpha therapies. The registry reports enrollment of 91, with the study starting on 2021-10-25 and primary completion on 2024-05-14.

The primary endpoint was MDA achievement at Week 24 in the full analysis set. The posted response percentages were 28.8% and 21.9%, respectively. The unadjusted recalculation does not establish a clear difference, and the registry-reported logistic-regression analysis also has an interval spanning its no-association value. Secondary outcomes provide additional measures but require cautious interpretation given uncertainty and unreported multiplicity procedures. Serious adverse events were reported in 4 participants in the combination group and 0 in the comparator group; deaths were 0 in both groups. Key details on original sample-size planning, missing-data methods and the full statistical plan are not reported in the registry record.