Pegasys® in Patients With Myeloproliferative Diseases: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pegasys® in Patients With Myeloproliferative Diseases (NCT00452023), a non-randomized phase 2 study. The primary outcome is Number of Patients With Complete Response (CR) or Partial Response (PR).

2. Clinical Question

3. Trial Design

This was a phase 2, nonrandomized, parallel study sponsored by M.D. Anderson Cancer Center. The record lists enrollment of 83 and an intervention of IFN-alpha2a. It describes five starting-dose groups: 90, 180, 270, 360, and 450 micrograms weekly, with each dose given by injection under the skin. Masking was none. The study status is completed.

The listed study start was 2005-04-07, and primary completion was 2023-05-26. The primary endpoint timeframe is up to 18 years. These dates and the endpoint timeframe describe the registry record; they do not establish how long each participant was followed or when each participant entered the study. Participant-level follow-up details are not reported in the registry record.

Because allocation was nonrandomized, the dose groups are not randomized treatment arms and should not be treated as equivalent groups differing only by dose. Differences in observed outcomes across groups could reflect differences in the participants or circumstances associated with group assignment, as well as dose. The registry does not report how dose assignment was made, whether participants changed dose, or how treatment exposure was handled in the analysis.

The record lists a single primary outcome measure and no outcome analyses. It does not report a control condition, an allocation procedure, masking, a statistical analysis plan, or a sample-size justification. The design therefore describes the treatment groups and the intended outcome, but does not provide the information needed to make a causal comparison between doses or against an untreated or alternative-treatment group.

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

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 Number of Patients With Complete Response (CR) or Partial Response (PR), measured over Up to 18 years.

The primary endpoint is the number of participants with complete response (CR) or partial response (PR). It is a count of participants meeting either response definition, not a measurement of the average change in a laboratory value. The definitions use platelet count, abbreviated PLT, and thromboembolic events, and specify that response is assessed without use of anagrelide or hydroxyurea.

For CR, the registry definition requires reduction of PLT to <440x109/l and disappearance of thromboembolic events. For PR, it allows reduction of PLT by 50% while still >440x109/l, or reduction of thromboembolic events by 50%. The registry record does not explain how these criteria are applied if a participant meets one component but not another, or how response is classified when information is incomplete.

General: a response endpoint can translate several clinical or laboratory criteria into a participant-level outcome that is easier to interpret than a collection of measurements. A count alone, however, does not show the proportion responding unless its denominator is also given, nor does it communicate how uncertain the estimate is. The registry lists the outcome definition and timeframe but does not report the response count, its denominator, or an uncertainty interval. Consequently, the endpoint’s intended meaning can be described, but the posted record does not establish the observed response result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General: clinical-trial methods determine how observations are collected, converted into outcomes, and compared. Interpretation depends not only on the endpoint definition but also on group assignment, follow-up, missing observations, and the statistical model. The registry record identifies some design features and the primary endpoint but leaves many analysis details unreported.

What does nonrandomized allocation imply?

In a randomized study, a chance procedure assigns participants to groups, helping balance measured and unmeasured prognostic characteristics on average. This study is identified as nonrandomized. The record does not describe how participants entered the different dose groups. As a result, observed differences between groups could be related to dose, to differences in who received each dose, or to other factors. A comparison across dose groups is therefore descriptive unless a suitable analysis and its assumptions are reported.

The record also states that masking was none. General: masking can reduce some forms of expectation or assessment bias, depending on who is masked and which outcomes are measured. The record does not report additional procedures for blinded outcome assessment. The response definition includes clinical and laboratory criteria, but the registry does not state who assessed response or whether assessment was independent.

How should a participant count be interpreted?

The primary endpoint is a count of participants with CR or PR. A count describes how many participants met the defined outcome, but its interpretation depends on the relevant denominator and observation window. A count without a denominator cannot be translated into an event frequency. The record does not report the primary response count or denominator, so neither a response proportion nor a group contrast can be described from the posted results.

General: a confidence interval expresses uncertainty around an estimate under a specified statistical method and its assumptions. Wider intervals indicate less precision, while narrower intervals indicate greater precision. A p-value addresses compatibility between data and a null hypothesis under a statistical model; it does not measure the size or importance of an effect. No confidence interval or p-value for the primary outcome is reported in the registry record.

What does the planning illustration represent?

General: sample-size planning asks how many participants per arm might be needed to detect a chosen effect with specified assumptions about significance level, allocation, and statistical method. The record’s planning assumptions specify a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the target effect is expressed as a stated fraction of the observed effect. These are planning assumptions, not evidence that the trial was designed or conducted using those assumptions.

A planning illustration is not observed power. Power is a design property under an assumed effect and model; calculating power after observing study results does not show that a result is reliable. In this record, the numerical primary effect is not reported, so the effect used for a numerical planning illustration cannot be checked against the posted primary outcome. The record also does not report a sample-size justification for the trial itself.

Why do event definitions and dispersion matter?

General: a binary endpoint classifies each participant as meeting or not meeting a specified outcome. Its analysis often focuses on a proportion or a contrast between proportions, but those quantities require observed counts and denominators. For count outcomes that represent numbers of events, repeated events within a participant can create over-dispersion: variability greater than a simple model expects. An analysis may then need a model that accounts for this extra variability. The primary endpoint is described as a count of participants with a response, rather than a count of repeated response events; the registry lists no analysis model or dispersion estimate.

The response definition specifies platelet thresholds and reductions in thromboembolic events, but the record does not describe the data collection schedule, confirmation requirements, or rules for resolving conflicting criteria. These details can affect the classification of a participant as a responder. They are not reported in the registry record, so the endpoint definition cannot be expanded beyond the criteria stated.

What can be inferred from the listed safety figures?

General: safety counts are descriptive unless the study design and analysis support a comparison. Counts at risk and affected can clarify the reporting base, while event terms identify what was reported. They do not by themselves establish that a treatment caused an event, and they do not adjust for differences between nonrandomized groups. The registry provides counts by dose group but does not report adjusted comparisons, exposure-adjusted rates, or uncertainty estimates. The safety figures should therefore be read as recorded observations, not as estimates of comparative risk.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("IFN-alpha2a 450 Microgram (mcg)", "IFN-alpha2a 360 Microgram (mcg)", "IFN-alpha2a 270 Microgram (mcg)", "IFN-alpha2a 180 Microgram (mcg)", "IFN-alpha2a 90 Microgram (mcg)")
n      <- c(3, 3, 19, 26, 32)
events <- c(3, 3, 14, 23, 23)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  IFN-alpha2a 450 Microgram (mcg)  3  3  100.0  29.2  100.0  43.9  100.0  39.8  99.4
ARM  IFN-alpha2a 360 Microgram (mcg)  3  3  100.0  29.2  100.0  43.9  100.0  39.8  99.4
ARM  IFN-alpha2a 270 Microgram (mcg)  19  14  73.7  48.8  90.9  51.2  88.2  50.9  88.1
ARM  IFN-alpha2a 180 Microgram (mcg)  26  23  88.5  69.8  97.6  71.0  96.0  70.8  95.8
ARM  IFN-alpha2a 90 Microgram (mcg)  32  23  71.9  53.3  86.3  54.6  84.4  54.5  84.4

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary response endpoint and lists no outcome analyses. It does not report a statistical analysis plan, a hierarchy of endpoints, planned subgroup analyses, or a method for controlling multiplicity. Multiplicity arises when several outcomes, comparisons, subgroups, or repeated analyses are considered: the more opportunities to identify a nominally unusual result, the more important it becomes to specify how those analyses will be interpreted. No multiplicity procedure is reported in the registry record.

The absence of reported analyses does not establish that no analyses were conducted; it means that their methods and results are not reported in the registry record. No primary-endpoint estimate, confidence interval, or test result is listed, so there is no reported inferential result to evaluate for multiplicity.

Missing data handling is also not reported in the registry record. The record does not state how participants without an evaluable response assessment were handled, whether any outcome observations were imputed, or what analysis population was used. Enrollment is listed as 83, but the endpoint denominator is not reported; enrollment should not be assumed to equal the number with evaluable response information.

General: missingness can affect an estimate when the likelihood of having an observed outcome is related to the outcome itself or to participant characteristics. Methods such as complete-case analysis or imputation rely on assumptions, and those assumptions should be made explicit. Without a reported missing-data rule, readers cannot determine how incomplete response assessments might have influenced the endpoint. The registry also does not describe an intention-to-treat population, a per-protocol population, or any other analysis population.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: seizure (33.3%), vasculitis (33.3%), ankle fracture (5.3%), carcinoma-facial (5.3%), cardiac surgery (5.3%).

The safety record lists serious-event and other-event counts by starting-dose group, together with deaths. In the 450-microgram group, 1 participant was affected by a serious event and 2 by other events, with 3 at risk for each measure; deaths were 0. In the 360-microgram group, no participants were affected by serious events and 3 by other events, with 3 at risk for each measure; deaths were 0.

In the 270-microgram group, 3 participants were affected by serious events and 9 by other events, with 19 at risk for each measure; deaths were 0. In the 180-microgram group, 3 were affected by serious events and 8 by other events, with 26 at risk for each measure; deaths were 2. In the 90-microgram group, 7 were affected by serious events and 7 by other events, with 32 at risk for each measure; deaths were 1.

The most frequent listed serious-event terms are Seizure and Vasculitis, each reported as 33.3%. Ankle Fracture, carcinoma-facial, Cardiac Surgery, and Lung Infection are each reported as 5.3%. These terms and percentages are registry descriptions; the record does not provide event-level details or explain their denominators.

General: these figures are descriptive, not a randomized safety comparison. The record does not report adjusted estimates, exposure duration, or uncertainty intervals, so the counts cannot establish comparative risk or causation.

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

The registry record establishes that a completed phase 2, nonrandomized, parallel study evaluated weekly IFN-alpha2a in patients with chronic myeloproliferative diseases. It specifies five starting-dose groups and defines a primary endpoint based on complete or partial response. It also records safety counts and selected serious-event terms. These details describe the study’s scope and the information the registry reports.

The record does not give the numerical result for the primary endpoint. It therefore does not show how many participants met the response definition, what fraction of those assessed responded, or how response differed across dose groups. No estimate of precision or inferential result is reported. The absence of these results means the posted record cannot be used to quantify the primary outcome.

The trial’s nonrandomized allocation is important when considering any apparent differences between dose groups. General: groups formed without random assignment may differ in ways that affect both treatment received and outcome. An observed contrast cannot automatically be attributed to dose, even if outcome counts are available. The record does not report a method to adjust for such differences.

The safety counts provide a descriptive account of events recorded in each group, but they do not identify causation or establish comparative safety. The at-risk counts are specific to the safety reporting and cannot stand in for unreported response denominators. The record also does not describe exposure duration or how incomplete safety follow-up was handled.

Accordingly, the study record supports a description of the trial design, intended response criteria, and selected safety observations. It does not support a numerical conclusion about response frequency, an adjusted comparison between doses, or a causal claim about treatment effects. General: distinguishing the question a study set out to ask from the result it actually reports is essential; a defined endpoint is not itself evidence of an observed effect.

14. Why This Trial Matters Statistically

This trial illustrates how a registry record can provide a clear endpoint definition while leaving the endpoint result unavailable. The response criteria specify the role of platelet count and thromboembolic events, making the intended clinical classification understandable. Yet without the observed response count and denominator, readers cannot assess the frequency or precision of response. Endpoint definitions and endpoint results answer different questions.

The nonrandomized design also matters for statistical interpretation. Multiple starting-dose groups may invite comparisons, but the record does not describe random allocation or a control group. General: differences among nonrandomized groups can reflect baseline characteristics and treatment-selection processes as well as the intervention. A numerical contrast alone would not remove that concern; methods and assumptions would need to be reported.

The safety information demonstrates another distinction: counts of affected participants are useful descriptions, but they are not automatically comparative risk estimates. Without a clear analysis, denominators tied to exposure, and uncertainty measures, the counts cannot show whether groups differ beyond what might arise from their composition or follow-up.

Finally, the record lists no outcome analyses and does not report a statistical analysis plan, missing-data strategy, or analysis population. These omissions limit reproducibility and make it difficult to assess how the intended endpoint was translated into a result. General: transparent reporting of estimates, uncertainty, analysis choices, and participant flow helps readers judge both the strength and limits of clinical evidence. This record is therefore statistically informative about design and reporting gaps, even though it does not provide a numerical primary outcome.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary limitation for interpreting efficacy is that the registry does not report the numerical primary response result. It also does not provide a response denominator, an estimate comparing dose groups, or uncertainty around an estimate. The endpoint definition and the stated timeframe do not substitute for outcome data.

The design is nonrandomized, and the registry does not describe how dose groups were assigned. No like-for-like control comparison is identified. Any comparison across groups would therefore be vulnerable to differences in participant characteristics or treatment selection, and the record does not report an adjusted analysis to address these concerns.

The registry does not report the timing or schedule of response assessments, confirmation rules, individual follow-up, or how changes in treatment were handled. Although the primary timeframe is listed as up to 18 years, that does not establish the follow-up received by each participant. The handling of missing or unevaluable response data is also not reported.

The record lists no outcome analyses and does not report the statistical analysis plan, analysis populations, multiplicity approach, or sample-size justification. It is consequently not possible to assess whether analyses followed prespecified methods or how incomplete observations were treated. No numerical primary effect is available to assess the assumptions behind a planning illustration.

Safety interpretation has separate limitations. The registry lists counts at risk and affected, deaths, and selected serious-event terms, but does not report adjusted comparisons, exposure-adjusted rates, or uncertainty intervals. The event-term percentages are recorded without a stated denominator in the listed information. These figures should be understood as descriptive registry data, not as a causal or comparative assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pegasys® in Patients With Myeloproliferative Diseases — NCT00452023.
• Registry record last updated 2024-12-11; status last verified 2024-12.
• Results first posted 2024-12-11.
• 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.

19. Record Summary

The registry describes NCT00452023, a completed phase 2, nonrandomized, parallel study sponsored by M.D. Anderson Cancer Center. It lists enrollment of 83, no masking, and weekly IFN-alpha2a starting-dose groups of 90, 180, 270, 360, and 450 micrograms. The population is described as patients with chronic myeloproliferative diseases, with Philadelphia chromosome-positive chronic myeloid leukemia excluded in the official title.

The primary endpoint counts participants with complete or partial response over up to 18 years. The registry provides the response criteria but does not report the numerical primary outcome or any outcome analyses. Safety counts are listed by dose group, along with deaths and selected serious-event terms. Allocation procedures, missing-data handling, analysis populations, inferential methods, and the trial’s sample-size justification are not reported in the registry record.