A Study of the Efficacy and Safety of MEDI-546 in Systemic Lupus Erythematosus: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Efficacy and Safety of MEDI-546 in Systemic Lupus Erythematosus (NCT01438489), a randomized phase 2 study. The primary outcome is Percentage of Participants Achieving an Systemic Lupus Erythematosus (SLE) Responder Index [SRI (4)] Response With Oral Corticosteroids (OCS) Tapering at Day 169.

2. Clinical Question

The endpoint combines change in SLE disease activity with constraints intended to rule out specified worsening and to require a corticosteroid reduction. It therefore asks about a joint clinical response under a tapering condition, rather than disease activity alone or steroid reduction alone. The registry record describes the eligibility population and the endpoint definition, but details such as how missing endpoint data were handled are not reported in the registry record.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase 2 study of MEDI-546 in participants with systemic lupus erythematosus. The registry lists 626 participants as enrolled. The listed treatment groups were anifrolumab 300 mg, anifrolumab 1000 mg, and matching placebo. Each regimen was described as an intravenous infusion every 4 weeks for 48 weeks.

The registry lists the study start as 2012-01 and primary completion as 2015-04, with overall status recorded as completed. These dates describe the study record; they do not by themselves give the timing of each participant’s follow-up or establish when every analysis was conducted.

The primary endpoint was the percentage achieving an SRI (4) response with oral corticosteroid tapering at Day 169. The mITT population was defined as randomized participants who received any investigational product and had a baseline primary efficacy measurement. For this outcome, the record says N denotes evaluable participants. The posted arm counts for the primary outcome are 102 for placebo, 99 for anifrolumab 300 mg, and 104 for anifrolumab 1000 mg. Those evaluable counts are not the same description as the overall enrollment count.

The registry lists 14 outcome measures and 12 posted secondary outcomes. These include a later SRI response with tapering, a separate steroid-taper outcome among participants taking at least 10 mg/day at baseline, safety outcomes, anti-drug antibodies, and a pharmacodynamic measure. The original sample-size rationale, planned allocation proportions, any interim analyses, and the full statistical analysis plan are not reported in the registry record. The brief description of randomization and masking does not establish further details of the procedures used.

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 626 participants were screened out of which 319 participants did not meet eligibility criteria and were considered screen failures, and 307 participants were randomized into the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The modified Intent-To-Treat (mITT) population included all randomized participants who received any investigational product and had a baseline primary efficacy measurement. Here, "N" signifies evaluable participants for this outcome measure.

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: Intent-to-treat (ITT) population included all participants who were randomized into the study.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving an Systemic Lupus Erythematosus (SLE) Responder Index [SRI (4)] Response With Oral Corticosteroids (OCS) Tapering at Day 169, measured over Day 169.

The primary endpoint is the percentage of participants achieving an SRI (4) response with oral corticosteroid (OCS) tapering at Day 169. The registry defines an SRI (4) responder as a participant with a reduction in baseline SLEDAI-2K score of >= 4 points, no worsening measured by the Physician Global Assessment, and no new BILAG-2004 Index A organ system score and no more than one new or worsening BILAG-2004 Index B organ system score. The registry defines worsening on the Physician Global Assessment as an increase of >= 0.3 from baseline on a 0 to 3.0 visual analog scale.

The taper condition requires a sustained OCS reduction from Day 85 through Day 169: less than 10 mg/day and less than or equal to the dose received on Day 1. This composite endpoint is used to represent response that satisfies both disease-activity criteria and a steroid-reduction condition. In general, a composite endpoint can capture more than one relevant dimension in a single outcome, but its interpretation depends on understanding every component and how they are combined.

The registry says the primary SRI outcome was analyzed with logistic regression. It does not report further detail about the model specification, covariates, or missing-data handling in the primary endpoint description.

7. Results Posted in the Registry

• Anifrolumab 300 mg: 34.3% (34/99) versus 17.6% (18/102) on control; risk difference 16.7 percentage points (95% CI 4.8 to 28.6), risk ratio 1.95 (1.18–3.21); unadjusted two-proportion z-test p = 0.007.
• Anifrolumab 1000 mg: 28.8% (30/104) versus 17.6% (18/102) on control; risk difference 11.2 percentage points (95% CI -0.2 to 22.6), risk ratio 1.63 (0.98–2.74); unadjusted two-proportion z-test p = 0.057.

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 Achieving an Systemic Lupus Erythematosus (SLE) Responder Index [SRI (4)] Response With Oral Corticosteroids (OCS) Tapering at Day 365

• 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:
  • Anifrolumab 300 mg: 51.5% (51/99) versus 25.5% (26/102) on control; risk difference 26.0 percentage points (95% CI 13.0 to 39.0), risk ratio 2.02 (1.38–2.96); unadjusted two-proportion z-test p = <0.001.
  • Anifrolumab 1000 mg: 38.5% (40/104) versus 25.5% (26/102) on control; risk difference 13.0 percentage points (95% CI 0.4 to 25.6), risk ratio 1.51 (1.00–2.28); unadjusted two-proportion z-test p = 0.046.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Anifrolumab 300 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Anifrolumab 1000 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants on Oral Corticosteroids (OCS) >=10 mg/Day of Prednisone or Equivalent at Baseline Who Were Able to Taper to Less Than or Equal to (<=) 7.5 mg/Day at Day 365

• 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:
  • Anifrolumab 300 mg: 56.4% (31/55) versus 26.6% (17/64) on control; risk difference 29.8 percentage points (95% CI 12.8 to 46.8), risk ratio 2.12 (1.33–3.39); unadjusted two-proportion z-test p = <0.001.
  • Anifrolumab 1000 mg: 31.7% (20/63) versus 26.6% (17/64) on control; risk difference 5.2 percentage points (95% CI -10.6 to 21.0), risk ratio 1.20 (0.69–2.06); unadjusted two-proportion z-test p = 0.520.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Anifrolumab 300 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Anifrolumab 1000 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend the study’s questions beyond the Day 169 primary endpoint. They include a later SRI (4) response with OCS tapering, tapering among participants taking at least 10 mg/day of prednisone or equivalent at baseline, treatment-emergent adverse events, laboratory and vital-sign abnormalities, electrocardiogram abnormalities, anti-drug antibodies, and a pharmacodynamic gene-signature measure. Together, these outcomes address later response, steroid reduction in a defined subgroup, safety observations, and biological measures.

Across the posted binary efficacy outcomes, the pattern is not identical for the two anifrolumab doses or across endpoints. The 300 mg group has higher posted percentages than placebo for the listed responder and taper outcomes, while the size and certainty of the contrasts vary. The 1000 mg group also has higher posted percentages than placebo for some of these outcomes, but the evidence varies by endpoint. This broad pattern provides context for the primary endpoint without changing its definition or replacing its analysis.

Secondary outcomes are useful for understanding whether findings appear across different time points and measures, but they generally carry less confirmatory weight when many outcomes are examined. Each additional comparison creates more opportunity for a result to appear notable by chance. The registry record does not report a multiplicity adjustment strategy or a hierarchy defining which secondary outcomes were confirmatory. The recalculated secondary p-values are unadjusted, and sponsor analyses are not posted for those secondary outcomes in the record.

Secondary findings should therefore be interpreted as supportive or exploratory descriptions rather than as independent confirmation of the primary result. Different endpoint definitions and populations also matter: the baseline steroid-taper outcome applies to a subgroup, while the Day 365 responder outcome uses a later assessment. The secondary outcomes do not establish that every dimension of disease activity, steroid use, safety, or biological response followed the same pattern.

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

• Percentage of Participants Achieving an Systemic Lupus Erythematosus (SLE) Responder Index [SRI (4)] Response With Oral Corticosteroids (OCS) Tapering at Day 365
• Percentage of Participants on Oral Corticosteroids (OCS) >=10 mg/Day of Prednisone or Equivalent at Baseline Who Were Able to Taper to Less Than or Equal to (<=) 7.5 mg/Day at Day 365
• Number of Participants With Treatment-Emergent Adverse Events (TEAEs), Adverse Events of Special Interest (AESIs) and Treatment-Emergent Serious Adverse Events (TESAEs)
• Number of Participants With Clinically Significant Laboratory Abnormalities in Investigations Reported as Treatment-Emergent Adverse Events
• Number of Participants With Vital Signs Abnormalities Reported as Treatment-Emergent Adverse Events (TEAEs)
• Number of Participants With Electrocardiogram (ECG) Abnormalities Reported as Treatment-Emergent Adverse Events (TEAEs)
• Percentage of SLE Participants With Positive Anti-drug Antibody (ADA)
• Neutralization Ratio of 21-Gene Type I Interferon (IFN) Signature for Participants With Positive Baseline Pharmacodynamic (PD) Gene Signature

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: a participant either meets the registry’s SRI (4) response definition with the required OCS taper at Day 169 or does not. The arm percentages describe the proportion meeting that definition. A comparison of percentages can be expressed as a risk difference, which is the difference between the event proportions in two groups. In general, a risk difference is on the probability scale and is directly read as an absolute contrast in event frequency. The recalculated risk differences are unadjusted and based on posted event counts and evaluable denominators.

What does logistic regression estimate?

Logistic regression is a method for modeling the probability of a binary outcome. Its coefficient is commonly expressed as an odds ratio: a contrast in odds between groups, potentially conditional on model terms. Odds and probabilities are related but are not the same quantity, so an odds ratio should not be read as a risk difference or as a ratio of response percentages. The registry reports logistic regression and odds ratios for its own primary analyses, but does not report the full model specification or covariates in the listed analysis information.

The sponsor-reported odds ratios are 2.38 for anifrolumab 300 mg versus placebo and 1.94 for anifrolumab 1000 mg versus placebo. The registry gives 90% confidence intervals for those estimates and reports p-values of 0.014 and 0.063, respectively. The independently recalculated risk differences are on a different scale and use simple unadjusted summary-statistic calculations. Because the registry analyses and recalculations differ, the sponsor’s own analysis takes precedence in describing the sponsor-reported inferential result. The recalculations remain a separate descriptive check rather than a replacement for the reported analysis.

What does a p-value tell the reader?

In general, a p-value describes how compatible data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor does it quantify the size or practical importance of an effect. The recalculated p-values are explicitly unadjusted. The registry-reported p-values are associated with its logistic-regression analyses; the record does not state whether they account for multiplicity across the two dose comparisons or other outcomes.

The recalculated estimates include intervals around the risk differences. In general, an interval conveys uncertainty about an estimate under the method used to construct it. It is not a range guaranteed to contain the true value. The registry’s listed sponsor analyses instead provide 90% confidence intervals for odds ratios. The confidence level and methods for the recalculated risk-difference intervals are not specified in the registry record, so they should not be conflated with the sponsor’s 90% intervals.

Why distinguish adjusted and unadjusted results?

An adjusted model can account for selected variables or design features included in its specification; an unadjusted comparison does not make those adjustments. The primary outcome description says that SRI was analyzed by logistic regression, while the registry’s reported analyses identify logistic regression and odds ratios. However, the covariates and detailed model structure are not reported in the registry record. The risk-difference recalculations use only the posted summary statistics and are not a recreation of an undocumented model.

General statistical interpretation also depends on how participants were assigned, followed, and analyzed. Randomization and masking are recorded, and the mITT population is defined, but detailed information about allocation procedures, deviations, missing outcomes, and the handling of incomplete follow-up is not reported in the registry record. Those omissions limit independent assessment of whether the analysis population and assumptions align with the design.

What does the endpoint definition contribute?

The SRI (4) measure requires improvement in SLEDAI-2K, no specified physician-assessed worsening, and limits on new or worsening BILAG-2004 organ-system scores. The Day 169 primary endpoint adds a sustained OCS taper requirement. Combining criteria can make the outcome more demanding and can connect disease-activity response with a treatment-burden measure. At the same time, a composite result does not show the contribution of each component unless component-level results are reported. Those component-level primary results are not reported in the registry record.

The endpoint also has a specific time point and taper window. A result at Day 169 does not automatically describe response at other times. The registry separately lists an SRI (4) response with tapering at Day 365, and separate secondary outcomes. Such outcomes provide additional perspectives, but they do not change the definition of the primary endpoint.

11. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimated contrast. In general, a confidence interval gives a range of values compatible with the observed data under the method and assumptions used. A narrower interval indicates greater statistical precision than a wider interval, but neither the interval nor its width alone establishes whether a result is clinically important.

For the unadjusted risk-difference recalculation comparing anifrolumab 300 mg with placebo, the estimate is 0.167 and the interval runs from 0.048 to 0.286. For 1000 mg compared with placebo, the estimate is 0.112 and the interval runs from -0.002 to 0.226. The second interval includes zero, while the first does not. These statements concern the posted recalculated intervals, not the sponsor’s regression analyses. The confidence level for these recalculated intervals is not reported in the registry record.

The sponsor’s own analyses report odds ratios with 90% confidence intervals: 2.38, with limits 1.33 and 4.26, for the 300 mg comparison; and 1.94, with limits 1.08 and 3.49, for the 1000 mg comparison. These intervals describe uncertainty on the odds-ratio scale and should not be interpreted as risk-difference intervals. The estimate and interval do not account for all uncertainty in design, measurement, missingness, or analysis choices.

Event counts and evaluable denominators also matter for precision: the primary endpoint counts are 18 of 102 for placebo, 34 of 99 for 300 mg, and 30 of 104 for 1000 mg. The registry does not report a separate precision analysis for other possible populations or assumptions. The confidence intervals therefore support interpretation of the specified comparisons, not certainty about every population or outcome of interest.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the original trial’s design calculations. For these binary outcomes, the figures are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. The original trial’s sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the number of participants per arm required, under those planning assumptions, for an 80% chance of detecting the selected effect if that effect is the true one. The 90% power column gives the corresponding number per arm for a higher chance of detection under the same assumptions. Power is a property of a planned design under stated assumptions; it is not a measure of the probability that a result is true.

The rows for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect illustrate different target effect sizes. A smaller effect is harder to distinguish from random variation, so the planning illustration requires more participants per arm as the target effect becomes smaller. The figures are outcome-specific: the observed effects and baseline event proportions differ among endpoints and comparisons, so the planning requirements also differ.

The accompanying planning curve expresses this relationship visually for the stated scenario; it does not replace the assumptions or turn the calculation into a forecast of a future study. These figures are not observed power. They should not be used to infer that the completed trial was adequately or inadequately sized without the original design rationale, which is not reported in the registry record.

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 = 0.176; treatment proportion p₁ = 0.343; average p̄ = 0.260.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6203 = 1.2157.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6089 = 0.5125.
4. n = (1.2157 + 0.5125)² / (0.1670)² = 107.1, rounded up to 108 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 <- 0.176470588235294                          # control proportion
p1 <- 0.343434343434343                          # 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(1438489)
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 = 108 (80% power), 143 (90% power)
 75% of observed effect: n per arm = 181 (80% power), 242 (90% power)
 50% of observed effect: n per arm = 384 (80% power), 513 (90% power)
power.prop.test: 107.1 
Simulated power at n = 108 per arm: 0.808 (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 reports logistic-regression analyses for the primary endpoint as odds ratios. For anifrolumab 300 mg versus placebo, the sponsor-reported odds ratio is 2.38, with a 90% confidence interval from 1.33 to 4.26 and p = 0.014. For 1000 mg versus placebo, it is 1.94, with a 90% confidence interval from 1.08 to 3.49 and p = 0.063.

The recalculations use the posted summary counts to estimate unadjusted risk differences. They are not the same estimand or necessarily the same analysis as the registry’s logistic regression. The values differ from the sponsor’s reported results, and the record flags a discrepancy. Accordingly, the sponsor’s own analysis takes precedence when stating the registry-reported analysis. The recalculated results can still help readers understand the absolute contrast in the posted percentages, but should not be substituted for the reported model-based odds ratios.

The registry does not provide the complete model specification, covariate list, or a full explanation for the discrepancy. It is therefore not possible to establish from the registry record why the estimates differ. The reported odds ratios and p-values should be attributed to the sponsor analysis, while the risk differences should be identified as unadjusted recalculations from summary statistics.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when statistical analyses include several comparisons or outcomes. In general, the more hypotheses examined, the greater the chance that at least one result will appear statistically unusual even if no underlying effect exists. Methods such as prespecified testing hierarchies or adjustments can address this issue, but the appropriate approach depends on the trial’s design and objectives.

This study has two dose comparisons for the primary endpoint, as well as a number of secondary outcomes. The registry lists 14 outcome measures and reports 12 secondary outcomes. It does not report whether the primary comparisons were adjusted for multiplicity, whether a testing sequence was used, or how the secondary outcomes were prioritized. The reported p-values should not be assumed to have multiplicity control when that information is not reported in the registry record. Secondary recalculated p-values are explicitly unadjusted.

Missing-data handling is another important part of interpreting an efficacy analysis. The primary population is described as randomized participants who received any investigational product and had a baseline primary efficacy measurement, and the outcome description says N represents evaluable participants. However, the rules for missing Day 169 assessments, treatment discontinuation, incomplete taper information, and other missing measurements are not reported in the registry record. Without those rules and the amount and pattern of missingness, readers cannot assess how incomplete observations may have affected the estimates.

The registry also does not report sensitivity analyses showing whether conclusions change under alternative assumptions about missing outcomes. It is not appropriate to infer a particular imputation method or missing-data strategy from the population label alone. A defined mITT population specifies who is included in broad terms; it does not resolve every question about outcome availability or analysis handling.

The endpoint’s composite nature adds an interpretive consideration. The record supplies the response criteria and taper condition, but component-level primary results are not reported in the registry record. Thus, the overall endpoint does not reveal which criterion most often determined whether a participant met the response definition. These gaps do not invalidate the posted analyses, but they limit the degree to which their robustness and confirmatory status can be independently assessed.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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: systemic lupus erythematosus (5.9%), chest pain (2.9%), influenza (2.0%), pneumonia (2.0%), ischaemic stroke (2.0%).

The safety results list participants at risk and participants with serious adverse events in each group. The registry records 19 participants with serious adverse events among 101 at risk in the placebo group, 16 among 99 at risk in the anifrolumab 300 mg group, and 18 among 105 at risk in the anifrolumab 1000 mg group. These are participant counts, not event totals, and the registry’s listed at-risk denominators differ from the primary efficacy denominators.

The serious adverse event terms listed most frequently include systemic lupus erythematosus at 5.9%, chest pain at 2.9%, and influenza, pneumonia, ischaemic stroke, and vasculitis at 2.0% each. These term percentages describe the registry’s posted safety listings; they do not establish that an event was caused by study treatment. The registry record does not report a death count in the adverse-event information.

Safety counts need to be interpreted alongside the study’s follow-up, event definitions, and exposure. The listed summaries alone do not give individual event timing, duration, severity, relationship assessments, or a complete account of every safety outcome. They support description of the reported events, not a broad conclusion about safety beyond the available record.

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

The primary endpoint asks whether participants met a specified disease-activity response definition while also sustaining an OCS taper at Day 169. The posted percentages and comparisons describe that combined outcome in the study’s evaluable mITT population. They do not describe all aspects of SLE, all possible treatment outcomes, or every participant’s clinical course.

The registry-reported logistic-regression results and the unadjusted risk-difference recalculations are distinct analyses. The sponsor’s reported odds ratios and confidence intervals should be used when describing the sponsor analysis, while the recalculations provide an unadjusted contrast on the absolute-risk scale. Because the record flags a discrepancy, readers should not treat the two sets of estimates as interchangeable or combine them into a single conclusion.

In general, statistical evidence does not by itself establish clinical importance, and an estimate does not guarantee that a similar result will occur in another study or population. The interval estimates indicate uncertainty under their respective methods. They do not address all sources of uncertainty, including the unreported details of multiplicity handling, missing-data methods, and model specification.

The study was randomized and quadruple-masked according to the registry, which are design features relevant to interpreting group comparisons. Still, the record does not supply all operational details needed to evaluate how those features were implemented. Nor does it report the original sample-size justification or a complete statistical analysis plan.

These findings should not be read as medical advice, as proof that a treatment works or does not work, or as a personalized prediction. They are statistical descriptions of the outcomes recorded for this trial. The distinction between primary and secondary outcomes, the composite endpoint definition, the discrepancy between analyses, and the limits of registry-level reporting all shape what can reasonably be concluded.

18. Why This Trial Matters Statistically

This trial illustrates why a reported percentage, a model-based odds ratio, and an absolute risk difference are not interchangeable summaries. The primary outcome is binary, but the sponsor’s own analysis uses logistic regression and reports odds ratios, while the recalculations use posted summary counts to produce unadjusted risk differences. Seeing these results on their respective scales helps clarify what each analysis describes and why analysis details matter.

The endpoint also combines disease-activity criteria with a corticosteroid-taper requirement. That makes the outcome more specific than a response measure alone, while also making the result dependent on multiple criteria. Statistical interpretation benefits from knowing the endpoint definition and the population used to evaluate it, rather than relying only on the outcome label.

The registry’s discrepancy between its reported analysis and the simple recalculations is a useful reminder that summary data cannot reproduce every model-based result. The sponsor’s own analysis takes precedence for the registry-reported inferential findings, but the differences underscore the value of clear model specifications and transparent reporting. The record does not report enough detail to explain why the estimates differ.

Finally, the study has multiple comparisons and several secondary outcomes, yet multiplicity handling and missing-data methods are not reported in the registry record. This matters because interpretation depends not only on the observed estimates, but also on how analysis choices, uncertainty, and the family of outcomes are addressed. The trial offers a case study in reading registry results carefully without overclaiming what the posted statistics establish.

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.

Several trial-specific limitations arise from the scope of the registry record. The primary analysis population is defined, but details of missing endpoint data, treatment discontinuation, deviations from assigned treatment, and sensitivity analyses are not reported in the registry record. The evaluable denominators therefore describe who contributed to the posted endpoint, but the record does not fully explain how unavailable outcomes were handled.

The registry states that the primary SRI outcome was analyzed with logistic regression, but does not list the full model specification or covariates. It reports odds ratios and 90% confidence intervals, while the recalculated risk differences are unadjusted and differ from the sponsor results. The record flags this discrepancy but does not explain its source. The sponsor analysis takes precedence for reporting the sponsor’s inferential result; the recalculations should be treated as separate summary-statistic checks.

The registry does not report a multiplicity strategy for the two dose comparisons and the multiple secondary outcomes. It also does not report the original sample-size justification, planned effect size, or assumptions behind the trial’s design calculations. The planning illustrations based on observed effects do not resolve those questions.

The primary endpoint is a composite requiring response criteria and sustained tapering. Component-level primary results are not reported in the registry record, so the overall response percentage cannot show which component determined endpoint status. The record also does not provide all details of masking or randomization procedures, or a full account of safety follow-up and exposure. These limitations narrow what can be independently assessed from the posted information without negating the results that are reported.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of the Efficacy and Safety of MEDI-546 in Systemic Lupus Erythematosus — NCT01438489.
• Registry record last updated 2016-10-07; status last verified 2016-08.
• Results first posted 2016-08-15.
• 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

The registry record describes a completed randomized, quadruple-masked, parallel-group phase 2 study of anifrolumab (MEDI-546) in systemic lupus erythematosus. It lists 626 enrolled participants and compares anifrolumab 300 mg and 1000 mg with matching placebo.

The primary outcome is SRI (4) response with OCS tapering at Day 169, analyzed in a defined mITT population. The record posts arm percentages and sponsor-reported logistic-regression odds ratios, as well as summary data used for unadjusted risk-difference recalculations. Because those analyses differ, the sponsor’s own analysis takes precedence for the registry-reported inferential results.

The record also lists secondary efficacy, safety, and biological outcomes. Multiplicity handling, missing-data methods, complete model details, and the original sample-size justification are not reported in the registry record. The posted statistics therefore support a bounded interpretation of the recorded outcomes rather than a complete assessment of every design and analysis decision.