Immune Intervention With Rituximab to Preserve Beta Cell Function in Early Onset Type 1 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Immune Intervention With Rituximab to Preserve Beta Cell Function in Early Onset Type 1 Diabetes (NCT01280682), a randomized phase 4 study. The primary outcome is Change of 3-hour Mean Area Under the Curve (AUC) of C-peptide.

2. Clinical Question

3. Trial Design

The registry record describes a completed, randomized, single-masked, parallel-group phase 4 trial. It lists enrollment of 120 and reports a start date of 2010-07 and primary completion of 2016-12. The record does not specify which people were masked, how randomization was implemented, or whether allocation was concealed.

The intervention arm received rituximab in addition to insulin; the control arm used insulin alone. The registry descriptions identify the participant groups as newly diagnosed with type 1 diabetes. Further eligibility criteria, the setting, and details of background insulin management are not reported in the registry record. The record lists 77 participants for the rituximab primary outcome and 30 for the parallel control primary outcome. These outcome-group counts should not be interpreted as a complete account of enrollment or participant flow.

The main outcome was change in three-hour mean AUC of C-peptide. The registry describes its assessment as six months from baseline, while the timeframe label says six months after participants completed the injection. The exact relationship between these timing descriptions is not clarified in the registry record. Secondary outcomes include fasting C-peptide, peak C-peptide, and HbA1c levels, with the posted C-peptide changes assessed at six months after participants completed the injection.

The registry record does not report the original sample-size justification, the planned primary statistical model, an intention-to-treat definition, or procedures for handling missing outcome data. Although it lists the trial as completed, completion status by itself does not establish how many participants had outcome measurements or how all randomized participants were analyzed.

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: patients whose C-peptide were too low

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Change of 3-hour Mean Area Under the Curve (AUC) of C-peptide, measured over 6 months after participants completed the injection.

The primary endpoint is change in three-hour mean C-peptide AUC. During a mixed meal tolerance test, C-peptide is measured at timed intervals over three hours. The registry says that the AUC was calculated from those measurements using the trapezoidal rule, then divided by the three-hour interval, expressed as AUC/180, to give a mean value. The change is the follow-up value relative to baseline.

C-peptide is released alongside endogenous insulin and is used in research as an indicator of the body's own insulin secretion. In a trial focused on preserving beta-cell function, a change in C-peptide offers a way to compare the groups' measured secretory function over follow-up. It is a biological outcome, not a direct measure of symptoms, complications, treatment burden, or quality of life.

The registry states that all mean C-peptide AUC data were transformed as log(mean AUC) for analysis. That matters when interpreting a difference: a model operating on transformed values does not necessarily have the same straightforward interpretation as a difference on the original measurement scale. The posted endpoint nevertheless labels the unit as pmol/L and gives group means and a mean difference. The registry record does not explain how the transformed analysis relates to those displayed units or specify a back-transformation.

The timing description is not fully consistent: the endpoint description identifies six months from baseline, while the timeframe label refers to six months after participants completed the injection. The record does not clarify which anchor governed the posted assessment.

7. Results Posted in the Registry

• Rituximab: mean -0.01 versus -0.35 on control; difference 0.34 (95% CI 0.23 to 0.45), standardized effect d = 1.28; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change of Fasting C-peptide

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Rituximab: mean -0.03 versus -0.34 on control; difference 0.31 (95% CI 0.15 to 0.47), standardized effect d = 0.86; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Rituximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change of Peak C-peptide

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Rituximab: mean -0.01 versus -0.29 on control; difference 0.28 (95% CI 0.15 to 0.41), standardized effect d = 0.99; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Rituximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary C-peptide outcomes follow the same broad direction as the primary outcome: the average changes are less negative in the rituximab group than in the parallel control group. Their listed unadjusted comparisons also have p < 0.001. This consistency offers supporting context because fasting and peak C-peptide capture related aspects of endogenous secretion, while the primary endpoint summarizes C-peptide response during a mixed meal test.

The secondary outcomes are not interchangeable with the primary endpoint. They use different C-peptide summaries, and HbA1c is another listed outcome that does not appear among the posted secondary results included in these comparisons. Although the registry lists HbA1c levels as a secondary outcome, numerical HbA1c results are not reported in the registry record. The secondary findings therefore do not establish a pattern across every planned or listed secondary measure.

Unadjusted secondary results generally carry less weight than a prespecified primary analysis. When several outcomes are examined, the chance of seeing at least one apparently strong result by chance can rise unless multiplicity is addressed. The registry record does not report a multiplicity procedure, a hierarchy among secondary outcomes, or a plan specifying which comparisons would be confirmatory. It also lists no reported analyses. The secondary p-values should therefore be treated as descriptive supporting evidence rather than independent confirmation.

The secondary outcomes relate biologically to the primary C-peptide endpoint, but similarity among correlated measures does not remove uncertainty about model choices, missing data, or outcome selection. The registry does not report whether the secondary analyses used the same participants, a common model, or the same missing-data handling as the primary outcome. Those details are needed to assess how directly the results reinforce one another.

The overall direction is coherent across the posted C-peptide results, but the evidence remains bounded by the information available in the registry record. It does not show results for every listed secondary measure, establish a multiplicity-adjusted conclusion, or demonstrate longer-term effects.

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

• Change of Fasting C-peptide
• Change of Peak C-peptide
• HbA1c Levels

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a measured quantity, and the registry reports a mean for each group. A mean difference compares those group averages by subtracting the control mean from the intervention mean. The posted primary contrast is 0.34. A positive value means the intervention-group mean is higher on the comparison scale. Because both arm means are negative changes, this contrast reflects a less negative average change in the rituximab group; it does not imply that every participant's C-peptide increased.

The arm-level spread is reported as a standard error. A standard error describes the precision of an estimated mean, rather than the spread of measurements among individual participants. It generally becomes smaller when a mean is estimated more precisely. It should not be read as a range containing most participant results. The registry does not post standard deviations or individual-level values for the primary outcome.

The endpoint is described as log-transformed for analysis. In general, a logarithmic transformation can make a skewed measurement more suitable for a statistical model and can change the scale on which differences are interpreted. A difference on a logged scale is not automatically a difference in original units. The record does not specify the precise model, how estimates were returned to the displayed scale, or how standard errors were derived. The reported result is therefore best interpreted using the registry's labels while recognizing that the scale is not fully documented.

What does an unadjusted comparison mean?

An unadjusted comparison uses the group summaries without accounting for other covariates. It can be a useful description of the contrast between randomized groups, but it is not a substitute for knowing the prespecified model or analysis population. Adjustment may be part of a trial's planned analysis, especially when the protocol identifies baseline factors or stratification variables. Whether that occurred is not reported in the registry record. The comparison listed for the primary result has p < 0.001, but the record contains no reported analyses describing a sponsor model or its assumptions.

The displayed confidence interval expresses uncertainty around the estimated mean difference under the calculation's assumptions. In general, a confidence interval gives a range of values compatible with the data and method at the stated confidence level. Narrower intervals indicate greater precision, while wider intervals indicate more uncertainty. An interval that does not include zero is consistent with a nonzero difference under the corresponding two-sided test, but it does not establish clinical importance or eliminate possible bias and missing-data concerns.

Why can over-dispersion matter?

Over-dispersion is a general issue when observed variability is greater than the statistical model expects. It is particularly relevant for outcomes that are counts or rates, where a simple model may understate uncertainty if events cluster or vary more than assumed. The posted primary endpoint is continuous, not a count outcome, and the registry does not report a model assessment for over-dispersion. It is therefore not possible to say that over-dispersion affected this result. More generally, checking whether a model's assumptions fit the data helps ensure that standard errors, intervals, and tests are not too optimistic.

The registry describes the C-peptide AUC calculation and states that a log transformation was used, but it does not identify a particular statistical test or model for the reported contrast. No adjustment strategy, handling of repeated measures, missing-data method, or sensitivity analysis is described. These omissions make it difficult to reproduce the inferential result from the registry record alone.

In randomized trials, randomization is intended to balance prognostic factors on average, supporting causal interpretation of a between-group contrast when assignment, follow-up, and analysis are appropriately handled. The record confirms random allocation but does not provide its method or enough information about attrition and analysis populations to assess those conditions. The result should thus be understood as a posted between-group association within the trial, with the limitations of the reported methods.

11. Confidence Intervals and Precision

The primary mean difference is 0.34, and the reported confidence interval lies entirely above zero. This indicates that the posted estimate is not merely centered on zero with uncertainty extending across both directions. The interval represents uncertainty in the group contrast under the calculation's assumptions; it does not capture every source of uncertainty, such as possible bias from missing data, deviations from the planned analysis, or ambiguity in the measurement scale.

The interval's bounds are reported to more decimal places than are appropriate to reproduce in explanatory text, so the important interpretation is its location relative to zero and the size of the estimate. The arm means are -0.01 and -0.35, with standard errors of 0.03 and 0.05, respectively. These standard errors describe uncertainty in the estimated arm means, not the individual-level variability in C-peptide change.

The p-value is reported as p < 0.001. A small p-value is evidence against a null model under the test's assumptions, but it is not a measure of precision by itself. The confidence interval provides more information about the range of differences compatible with the data. Neither measure answers whether the observed contrast is clinically important, and the registry record does not identify a threshold for clinical importance.

The record's account of log-transformed analysis complicates interpretation of the displayed mean difference, which is presented with a pmol/L unit. The exact model scale and any conversion to the displayed scale are not reported. Precision on the analysis scale may not translate directly to precision on the original scale.

The interval and p-value are based on the posted summaries and unadjusted comparison. The registry does not report a complete statistical analysis plan, participant flow for this endpoint, or sensitivity analyses. Those omissions limit how comprehensively the reported precision can be assessed.

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, not a total across the trial. They illustrate sample sizes under a stated planning calculation rather than documenting the trial's original design assumptions. The assumptions listed are a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. The target effect is set as a stated fraction of the observed effect. The original trial's sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the illustrative participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect were the true effect. The 90% power column makes the same kind of calculation with a higher target chance of detection. These are planning concepts: power is a property of a proposed design under assumed conditions, not a probability that the trial's conclusion is correct.

For example, the primary-outcome planning illustration lists 10 participants per arm for 80% power and 13 participants per arm for 90% power when planning for 100% of the observed effect. For planning at 75% of the observed effect, it lists 18 and 23 participants per arm; for 50%, it lists 39 and 52. The lower assumed effects require larger sample sizes because smaller differences are harder to distinguish from random variation.

The plotted planning curve represents the same relationship between assumed effect size and required participants. It should be read as an illustration based on the observed result, not as a reconstruction of the original trial design. This is not observed power.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.340; pooled standard deviation σ = 0.266.
2. n = 2 × (1.96 + 0.842)² × 0.266² / 0.340² = 9.6, rounded up to 10 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.34                      # observed difference in means
sigma <- 0.266222248721413                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1280682)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 10 (80% power), 13 (90% power)
 75% of observed effect: n per arm = 18 (80% power), 23 (90% power)
 50% of observed effect: n per arm = 39 (80% power), 52 (90% power)
power.t.test: 10.7 
Simulated power at n = 10 per arm: 0.769 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when investigators assess several outcomes, time points, subgroups, or comparisons. Even if each test uses a conventional threshold, conducting many tests can increase the chance that at least one result appears statistically strong by chance. Common approaches include identifying a single primary endpoint, ordering tests hierarchically, or adjusting inference across a family of comparisons. These are general methods; which approach was planned or used in this trial is not reported in the registry record.

The registry identifies a primary endpoint and lists secondary outcomes, but it does not describe a multiplicity adjustment or testing hierarchy. It reports unadjusted p-values for the posted C-peptide comparisons. The record lists HbA1c as a secondary outcome, but its numerical result is not reported in the registry record. Without the protocol's outcome hierarchy and complete results, it is not possible to determine the intended family of tests or whether selective reporting is a concern.

Missing data can also affect interpretation. If outcome measurements are absent, the analyzed participants may differ from all randomized participants. The registry record does not report endpoint-specific missingness, reasons for missing measurements, imputation, or sensitivity analyses. The primary outcome group counts are not enough to establish participant flow because enrollment, allocation, follow-up, and analysis counts may differ.

The record does not define an intention-to-treat population, a per-protocol population, or another analysis population. It also does not report whether participants were analyzed according to randomized assignment, how treatment discontinuations were handled, or whether the posted means include all available measurements. These details matter because the effect of randomization is best preserved by a suitable analysis that accounts for all randomized participants and handles missing data transparently.

The record lists no reported analyses, despite posting outcome summaries and unadjusted comparisons. This leaves the statistical plan, model specification, and missing-data procedures unclear. Consequently, the primary and secondary findings can be described, but their robustness to alternative analyses cannot be assessed from the registry record.

15. 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 non-serious adverse event terms

The registry lists serious adverse events affecting 0 participants in the rituximab group and 0 in the parallel control group. It also lists deaths as 0 in each group. These are the posted counts for the reported risk groups; they do not establish that no event could have occurred outside the registry's reporting definitions or observation period.

For other adverse events, the record lists 44 affected participants in the rituximab group and 0 in the parallel control group. The registry does not provide event terms or severity details for these other events in the information, so their clinical nature cannot be characterized. The figures are counts, not rates, and should be read alongside the at-risk group counts listed in the registry rather than interpreted as measures of individual risk without further detail.

The record does not report definitions for serious events, ascertainment methods, follow-up duration for safety, or an event-by-event breakdown. A count of zero serious events in the posted data does not prove that a treatment is safe, just as a count of other events does not by itself establish causation. These summaries describe what was recorded, with limited detail for assessing comparative safety.

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

The posted primary comparison indicates a less negative mean change in three-hour C-peptide AUC in the rituximab group than in the insulin-only control group at the recorded assessment. The size of the listed mean difference is 0.34, and the reported unadjusted p-value is p < 0.001. The confidence interval lies above zero, supporting a nonzero contrast under the assumptions used for that calculation.

This is evidence about a measured biological outcome in a randomized trial, not a complete account of the treatment's effects. C-peptide is used as an indicator of endogenous insulin secretion, but the posted result does not directly quantify symptoms, complications, insulin requirements, quality of life, or long-term health outcomes. The registry record does not report follow-up results beyond the stated assessment that would show whether the contrast persisted.

The result does not mean that every participant had a better outcome, nor does it determine the clinical importance of the average difference. The record does not state a threshold for a clinically meaningful change. In addition, the endpoint description says the data were log-transformed for analysis, while the displayed outcome and comparison are labeled with a pmol/L unit. The precise interpretation of the estimate on the original scale is not explained.

Random allocation supports a between-group comparison, but the registry does not describe the randomization procedure, allocation concealment, analysis population, or missing-data handling. These omissions make it difficult to judge whether the posted contrast captures the effect of assignment for all randomized participants. The record also does not report a multiplicity strategy for secondary outcomes.

The secondary C-peptide findings share the primary result's general direction, but their unadjusted status and incomplete reporting limit their confirmatory strength. The trial's posted results therefore support a difference in the measured C-peptide summaries at the stated follow-up, while leaving questions about analysis details, clinical relevance, durability, and broader outcomes unanswered.

17. Why This Trial Matters Statistically

This trial is statistically informative because it compares a biological marker over time in randomized treatment groups and reports both arm-level changes and a between-group estimate. Showing the contrast alongside the group means helps readers distinguish the intervention group's change from the comparative effect. The result also illustrates why uncertainty measures matter: a mean difference and its confidence interval convey information that a p-value alone cannot.

The primary and secondary C-peptide results point in a broadly consistent direction, offering a chance to consider how related outcomes can reinforce a pattern while still requiring cautious interpretation. The secondary outcomes are unadjusted, and the registry record does not report a multiplicity plan. This makes the distinction between a primary test and exploratory or supporting analyses especially important.

The record also highlights the value of complete statistical reporting. It identifies randomization and masking, but does not describe the allocation procedure, primary analysis model, analysis population, missing-data methods, or multiplicity handling. It reports a log transformation without clarifying how the displayed estimates relate to that scale. These gaps limit reproducibility and the ability to assess robustness, even when the posted contrast appears precise.

Finally, the sample-size illustrations demonstrate how planning calculations depend on the effect size assumed: smaller target effects require more participants per arm. Because those illustrations are based on the observed effect rather than the original design, they cannot substitute for the trial's design rationale. Separating observed results from planning assumptions is essential to avoid treating post hoc calculations as evidence of achieved power.

18. 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.

Several limitations are specific to the registry record for this trial. The primary endpoint's timing is described both as six months from baseline and as six months after participants completed the injection. The record does not clarify the assessment anchor, which makes the timing less straightforward to interpret.

The endpoint description states that mean C-peptide AUC data were log-transformed for analysis, but the posted means and mean difference are labeled in pmol/L. The exact analysis scale, model, and any conversion to the displayed scale are not reported. The registry also does not identify the statistical test used for the posted unadjusted comparison.

Participant flow is not reported in enough detail to reconcile enrollment with the outcome-group counts. The record does not give reasons for missing measurements, the number assigned to each arm, or the analysis population. It is therefore not possible to assess attrition or determine whether all randomized participants contributed to the reported endpoint.

The registry does not report the original sample-size rationale, the prespecified analysis plan, multiplicity handling, or sensitivity analyses. It lists HbA1c as a secondary outcome without numerical HbA1c results in the posted outcomes. The adverse-event information also lacks event terms and detailed definitions. These limitations do not negate the posted summaries, but they restrict independent assessment of the analyses, clinical interpretation, and completeness of outcome reporting.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Immune Intervention With Rituximab to Preserve Beta Cell Function in Early Onset Type 1 Diabetes — NCT01280682.
• Registry record last updated 2023-12-11; status last verified 2023-03.
• Results first posted 2023-12-11.
• 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.

22. Record Summary

The registry record describes a completed, randomized, single-masked, parallel-group phase 4 trial comparing rituximab plus insulin with insulin alone in newly diagnosed type 1 diabetes. Enrollment is listed as 120, with start in 2010-07 and primary completion in 2016-12.

The primary outcome was change in three-hour mean C-peptide AUC. The posted unadjusted mean difference was 0.34, with p < 0.001; the reported confidence interval lies above zero. Fasting and peak C-peptide results show a similar overall direction, while numerical HbA1c results are not reported in the registry record.

The record does not report the original statistical plan, analysis population, missing-data methods, or multiplicity procedure. The posted results describe the measured outcomes and their uncertainty, but do not establish long-term effects or clinical importance.