Long-term Safety and Efficacy Study of Fentanyl Sublingual Spray for the Treatment of Breakthrough Cancer Pain: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Long-term Safety and Efficacy Study of Fentanyl Sublingual Spray for the Treatment of Breakthrough Cancer Pain (NCT00538863), a non-randomized phase 3 study. The primary outcome is Percentage of Patients That Experienced 1 or More Adverse Events.

2. Clinical Question

The study’s stated purpose was to assess 90-day safety in patients using around-the-clock opioids for persistent cancer pain. That purpose and the primary endpoint are related but not identical descriptions: the purpose refers to a 90-day safety assessment, while the endpoint timeframe is baseline to study end, up to 116 days. The registry record does not explain how that endpoint timeframe relates to the stated 90-day purpose.

The registry describes one drug intervention and reports titration and maintenance stages. It does not report a randomized comparison with placebo, usual care, or another active treatment. Therefore, the central question is descriptive: what adverse-event experience was recorded among treated patients? The design does not estimate a randomized treatment contrast.

The record identifies cancer and pain as conditions. It does not report further population characteristics in the registry record for this analysis, such as age distribution, cancer types, or baseline pain measures. Those details should not be inferred from the study title or from the intervention description.

3. Trial Design

This completed Phase 3 study evaluated fentanyl sublingual spray in an open-label, non-randomized, single-group design. The registry reports enrollment of 319. Because there was no random allocation and no separate control treatment, the titration and maintenance labels should be read as treatment stages, not as randomized arms suitable for a direct efficacy or safety comparison.

During titration, patients could receive doses starting at 100, 200, or 400 µg, with upward titration to a maximum dose of 1600 µg. The description states that treatment could be used for up to a maximum of 4 breakthrough pain episodes per day, with a minimum separation of 4 hours between treatments. The stopping criterion for titration is cut off in the registry description and is not reported in the registry record.

During maintenance, patients received treatment for 90 days, up to a maximum of 4 times per day and with a minimum separation of 4 hours between treatments. The maintenance dose ranged from 100 to 1600 µg and was determined in an earlier study or during the open-label titration period. The record does not establish whether the stage-specific groups represent separate participants or overlapping sets of patients. Their counts should not be added or treated as independent groups.

The primary endpoint concerns the percentage of patients experiencing one or more adverse events from baseline to study end, up to 116 days. The safety population includes all patients who received study medication. The record identifies one primary outcome measure and no registered analysis entries. It does not report details of masking, which is consistent with the stated absence of masking, nor does it give a statistical analysis plan.

The study started in 2007-12, reached primary completion in 2010-10, and is marked completed. Dates and design details describe the conduct recorded in the registry; they do not establish how patients were selected, how missing observations were handled, or how the reported safety summaries 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: Five patients who completed the titration period did not start the maintenance period due to failure to determine a dose in the titration period (n=1), withdrawal from the study (n=3), and adverse event (n=1).

Open-label Titration

Open-label Maintenance

Analysis population for the primary outcome, as defined in the registry: Safety population: All patients who received study medication.

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: The total number of patients enrolled in the study is 319.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Patients That Experienced 1 or More Adverse Events, measured over Baseline to end of the study (up to 116 days).

The primary endpoint is the percentage of patients who experienced one or more adverse events. Its timeframe is baseline to the end of the study, up to 116 days, and its stated population is the safety population: all patients who received study medication. This endpoint reduces each patient’s experience to a yes-or-no status for whether at least one event occurred during the observation period.

As a general statistical concept, an any-event percentage is useful for describing how common it was for patients to experience at least one event. It is straightforward to interpret at the patient level and does not count repeated episodes in the same patient as separate people. However, it also discards information: it does not by itself describe the number of events per patient, when events occurred, how long they lasted, or whether they were serious or non-serious.

The registry identifies the endpoint and its timeframe, but a separate numerical result for the primary endpoint is not reported in the registry record. The adverse-event summary includes counts in categories labeled serious and other, but those entries are not a substitute for a separately reported primary-endpoint result. The record also does not explain whether the endpoint’s event definition includes all events in those categories or how an event was attributed to treatment.

The endpoint is a safety measure, not a measure of pain relief or comparative efficacy. Its interpretation is limited to the treated safety population and the specified follow-up period. Without a control group, the endpoint cannot distinguish events that occurred during treatment from events that would have occurred without the study treatment.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record describes a single-group, open-label study. In general, a single-group study observes outcomes in people receiving an intervention without a concurrent randomized control group. Open-label means masking was absent, as the registry states. These design features make event summaries useful for describing what was recorded during the study, but they do not provide the protection against confounding that random allocation can offer in a treatment comparison.

What does the primary endpoint summarize?

The primary endpoint is a percentage of patients who experienced one or more adverse events. Statistically, this is a binary patient-level outcome: each patient is classified as having experienced at least one event or not during the specified observation period. The endpoint title, unit, timeframe, and safety-population definition are reported. A separate numerical result for this endpoint is not reported in the registry record.

As a general concept, binary summaries can make an outcome easy to communicate, but they compress the underlying experience. They do not distinguish one event from repeated events in the same patient, do not represent event timing, and do not show how long patients were observed. For that reason, a binary endpoint is not interchangeable with an event count or a time-to-first-event analysis. The registry does not report a time-to-event endpoint.

How should stage-specific adverse-event counts be read?

The adverse-event summary lists patients at risk and patients affected in serious and other categories. The titration summary reports 33 serious-event patients among 229 at risk, and the maintenance summary reports 178 among 269 at risk. For the other-event category, it reports 194 among 229 at risk during titration and 269 among 269 at risk during maintenance. These are counts in the registry’s stage-specific summaries, not independent randomized groups.

As a general statistical principle, denominators matter because a count without its population at risk can be misleading. But having denominators does not make two groups comparable. In this study, allocation was non-randomized and the design was single-group. The registry does not explain whether the stage populations overlap, how patients moved between stages, or whether the duration of observation was identical across summaries. Therefore, stage-to-stage differences cannot be interpreted as treatment contrasts.

Over-dispersion is a general issue for event-count analyses: it occurs when observed counts vary more than a model’s assumed distribution allows, potentially making uncertainty estimates too narrow if the model is not suited to the data. It is relevant in principle when analyzing repeated events, since patients may differ in their tendency to experience events. However, no repeated-event model, dispersion estimate, or event-level analysis is reported in the registry record. The primary endpoint is patient-level and binary, and the record does not describe a model used to analyze it.

What do confidence intervals and significance tests add?

In general, a confidence interval gives a range of estimates compatible with a specified statistical procedure and the observed data. It helps communicate precision, which a point estimate alone cannot show. A significance test evaluates data against a defined null hypothesis under stated assumptions. Neither can be reconstructed responsibly without the analysis specification and necessary summary results. Confidence intervals, p-values, and an inferential analysis of the primary endpoint are not reported in the registry record.

The registry record also lists no analysis entries. It does not report whether any planned interim analyses, covariate adjustments, subgroup analyses, or sensitivity analyses were conducted. Their presence or absence in practice should not be guessed from the available record. The statistical analysis plan is not reported in the registry record.

How should planning illustrations be understood?

General sample-size planning asks how many participants may be needed to detect a chosen effect under specified assumptions. The planning assumptions described for this study use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such illustrations are hypothetical planning calculations, not observed power. They do not describe the actual single-group, non-randomized design, and they should not be interpreted as evidence that the trial had a particular probability of finding a result.

The registry does not report the trial’s own sample-size justification or a statistical plan connecting any planning illustration to the actual study design. The illustrations also cannot turn the titration and maintenance stages into a valid randomized comparison. In general, a planning calculation depends on the target contrast, outcome definition, allocation, assumptions, and analysis method. If those do not match the trial being conducted, the resulting illustration does not establish the adequacy of that trial’s design.

Finally, safety interpretation depends on more than statistical calculations. Definitions of adverse events, ascertainment procedures, follow-up completeness, and rules for attributing events to treatment all affect what a count means. Those details are not reported in the registry record. The posted summaries are therefore best treated as descriptive registry entries, with the design and reporting limits kept explicit.

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("Fentanyl Sublingual Spray Titration", "Fentanyl Sublingual Spray Maintenance")
n      <- c(229, 269)
events <- c(135, 217)
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  Fentanyl Sublingual Spray Titration  229  135  59.0  52.3  65.4  52.5  65.1  52.5  65.1
ARM  Fentanyl Sublingual Spray Maintenance  269  217  80.7  75.4  85.2  75.5  84.9  75.5  84.9

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary outcome measure and lists no analysis entries. It does not report a statistical analysis plan, prespecified testing hierarchy, or multiplicity adjustment. Multiplicity concerns arise in general when many outcomes, subgroups, or time points are tested, because repeated testing can increase the chance of a false-positive finding. Whether such testing occurred in this study is not reported in the registry record, so no conclusion about its handling is warranted.

The registry does not describe the amount or pattern of missing outcome data, the number of patients with incomplete follow-up, or how missing observations were handled. It also does not report a missing-data sensitivity analysis. These omissions matter because the primary endpoint concerns whether a patient experienced an event over an observation period. In general, incomplete observation can affect whether an event is captured; the direction and size of any resulting bias depend on why data are missing and on the analysis approach.

The stated safety population includes all patients who received study medication. The record does not provide a detailed analysis-population flow or explain how this population relates to the titration and maintenance summaries. The stage-specific denominators should not be combined: the trial is single-group, and whether the stage populations overlap is not reported in the registry record.

Definitions for adverse events and serious adverse events, event-attribution procedures, and rules for counting recurrent events are also not reported in the registry record. As a result, the summaries cannot be used to infer unreported methods or to reconstruct a complete analysis. No assumptions about interim analyses, exclusions, imputation, or statistical testing should be made from the registry entries alone.

12. 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: malignant neoplasm progression (22.3%), disease progression (1.9%), anaemia (1.5%), pneumonia (1.5%), cancer pain (1.5%).

The registry’s adverse-event summary gives stage-specific numbers at risk and numbers affected. During titration, the serious-event entry is 33 among 229 at risk; during maintenance, it is 178 among 269 at risk. In the other-event category, the corresponding entries are 194 among 229 and 269 among 269. These are counts of patients represented in the registry’s categories, not counts of event episodes.

The distinction between serious and other adverse events is a classification, not a complete ranking of how severe every event was. The registry record does not give the event definitions, counting rules, or the method used to determine whether an event was related to treatment. It also does not explain whether patients could appear in both stage summaries. The design is single-group and non-randomized, so the stage entries are not a control comparison.

The listed serious-event terms include malignant neoplasm progression at 22.3%, disease progression at 1.9%, anaemia at 1.5%, pneumonia at 1.5%, cancer pain at 1.5%, and nausea at 1.1%. The denominator and method behind those term percentages are not reported in the registry record. Event terms and percentages should therefore be read as descriptive entries, not as estimates of treatment-caused risk.

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

The posted results describe adverse events recorded during a completed, open-label study of fentanyl sublingual spray. They provide stage-specific counts for serious and other adverse-event categories and list several serious-event terms with percentages. This information can describe what the registry records among treated patients, but it does not establish that the intervention caused any particular event.

The primary endpoint is the percentage of patients with at least one adverse event during baseline to study end, up to 116 days. The registry identifies the endpoint but does not report a separate numerical result for it in the record available for this analysis. The adverse-event category counts should not be substituted for that missing endpoint result without a stated definition linking them.

The results do not show how fentanyl sublingual spray compares with no treatment, placebo, or another treatment. There was no randomized control group. Nor do the titration and maintenance summaries constitute a like-for-like comparison: the study is listed as single-group and non-randomized, and overlap between stage populations is not described.

The summaries also do not by themselves establish the frequency of repeated events, the time to an event, the duration of events, or the uncertainty around a treatment effect. Confidence intervals and an inferential comparison are not reported in the registry record. The serious-event terms do not provide a causal attribution assessment, and their percentage denominators are not stated.

In general, safety findings from a treated cohort can help characterize events observed during study follow-up. Their interpretation depends on how events were defined and collected, how complete follow-up was, and whether the study included a suitable comparison group. Those details are not reported in the registry record. The appropriate conclusion is therefore limited to describing the posted registry counts and terms, rather than making a comparative claim about treatment risk or benefit.

14. Why This Trial Matters Statistically

This study illustrates why design labels are essential when reading clinical trial results. Although the registry uses titration and maintenance group labels, it identifies the overall study as single-group and non-randomized. Those labels do not create a randomized comparison. Treating the stage summaries as separate control and treatment groups would risk attributing differences to treatment when other explanations, including stage membership and follow-up, are not addressed in the record.

It also shows why an endpoint definition and its result need to be considered separately. The registry specifies a patient-level endpoint—experiencing one or more adverse events—but a separate numerical result for that endpoint is not reported in the registry record. The adverse-event counts offer related descriptive information, yet they do not automatically answer the endpoint question unless the definitions and analysis connecting them are stated.

The record highlights the importance of denominators and event categories. Counts among patients at risk are more interpretable than counts alone, while serious and other event categories address different aspects of safety reporting. Still, neither denominators nor labels resolve questions about overlap between stages, event definitions, causality, or missing follow-up.

Finally, this record demonstrates the limits of planning illustrations. General calculations can explore hypothetical sample sizes under explicit assumptions, but they do not represent observed power, validate a design that differs from those assumptions, or replace a reported statistical plan. A careful statistical reading keeps the calculations, study design, and observed registry summaries distinct.

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 design is open-label, non-randomized, and single-group, with no separate control treatment reported. Consequently, the registry summaries cannot estimate a controlled treatment contrast, and events observed during follow-up cannot automatically be attributed to the intervention. The record does not provide a method for assessing treatment causality.

The stage-specific summaries are not fully explained. The record does not state whether patients in titration also appear in maintenance, how patients entered or left each stage, or whether their observation periods were comparable. It would therefore be inappropriate to treat the two stage populations as independent or to interpret their counts as a direct comparison.

The primary endpoint is described, but a separate numerical result for it is not reported in the registry record. The record also does not report confidence intervals, p-values, analysis methods, or a statistical analysis plan. It does not state how missing data were handled, how adverse events were defined and collected, or how recurrent events were counted.

The serious-event term percentages are listed without a reported denominator or calculation method. The event terms alone do not show whether the events were considered related to treatment. In addition, the titration stopping criterion is cut off in the registry description, and the available record does not supply the missing detail.

These reporting gaps limit precision and interpretation. They do not justify assuming that unreported procedures were or were not performed. The findings should be confined to the descriptive registry information, without extrapolating to populations, comparisons, or treatment effects not represented in the record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Long-term Safety and Efficacy Study of Fentanyl Sublingual Spray for the Treatment of Breakthrough Cancer Pain — NCT00538863.
• Registry record last updated 2013-10-11; status last verified 2013-08.
• Results first posted 2013-10-11.
• 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 record describes a completed Phase 3, open-label, non-randomized, single-group study of fentanyl sublingual spray for breakthrough cancer pain. Enrollment was 319. The stated purpose was to assess 90-day safety among patients receiving around-the-clock opioids for persistent cancer pain.

The primary endpoint was the percentage of patients experiencing one or more adverse events from baseline to study end, up to 116 days, in the safety population of patients who received study medication. The record does not report a separate numerical result for that endpoint. Stage-specific serious and other adverse-event counts and a list of serious-event terms are reported, but the stages are not randomized comparison groups.

The registry lists no analysis entries. A statistical analysis plan, missing-data methods, endpoint confidence intervals, and the denominator for serious-event term percentages are not reported in the registry record. The available information supports descriptive interpretation only.