This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Efficacy and Safety of Fingolimod (FTY720) in Patients With Relapsing-remitting Multiple Sclerosis (NCT00355134), a randomized phase 3 study. The primary outcome is Aggregate Annualized Relapse Rate (ARR) Estimate up to Month 24.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00355134 |
| Phase | Phase 3 |
| Condition | Multiple Sclerosis |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Enrollment | 1083 |
| Arms in the primary outcome | 3 |
| Lead sponsor | Novartis |
| Results posted | Yes |
2. Clinical Question
Population
Patients with relapsing-remitting multiple sclerosis were eligible for the study. The primary outcome population was the full analysis set: randomized patients who took at least one dose of study drug. The registry record gives an enrollment of 1083.
Intervention
During the core phase, participants received oral fingolimod once daily at either 0.5 mg or 1.25 mg, for up to 24 months. The extension phase continued fingolimod for participants in the active-treatment groups, subject to a protocol amendment affecting the 1.25 mg group.
Comparator
During the core phase, the comparator group received placebo capsules orally once daily. In the extension phase, participants originally assigned to placebo received either 1.25 mg or 0.5 mg fingolimod, so the extension did not preserve a placebo group as described in the core phase.
Primary question
Over 24 months, what was the estimated annualized relapse rate in each randomized treatment group? The primary outcome was the aggregate annualized relapse rate, expressed as relapses per year and estimated using a negative binomial regression model.
The trial was designed to assess safety, tolerability and efficacy of two oral fingolimod doses compared with placebo in people with relapsing-remitting multiple sclerosis. The principal outcome summarizes confirmed relapses over the core 24-month timeframe. It is a rate rather than a simple count: the analysis accounts for the total time participants spent on study, as well as the confirmed relapses observed in each group.
The registry record describes a randomized, triple-masked, parallel-group phase 3 study. Its official title specifies a 24-month double-blind core study with an optional extension phase. The primary outcome population included randomized participants who took at least one dose, rather than every randomized participant regardless of treatment exposure. The registry record does not report a separate primary comparison estimate or a primary-outcome p-value in the posted results.
3. Trial Design
This was a randomized, triple-masked, parallel-group, phase 3 study sponsored by Novartis. The registry record lists an enrollment of 1083 and identifies multiple sclerosis as the condition studied. The study began in 2006-06 and had a primary completion date of 2011-06; its status is completed. The official title describes a double-blind, placebo-controlled core phase lasting 24 months and an optional extension phase.
Three core-phase groups are listed: fingolimod 1.25 mg, fingolimod 0.5 mg, and placebo. Each treatment was taken orally once daily for up to 24 months during the core phase. In the extension, participants in the active-treatment groups continued fingolimod. Participants initially assigned to placebo received either fingolimod dose. The registry description also notes that a protocol amendment led to a switch for participants taking 1.25 mg, but the available arm description does not fully specify that switch.
Random allocation and masking are relevant to how group comparisons should be understood: they are design features intended to limit systematic differences in treatment assignment and in knowledge of assignment. They do not, by themselves, establish that all possible sources of bias or uncertainty were eliminated. Details of the allocation procedure, masking roles, concealment, and handling of deviations are not reported in the registry record.
The primary endpoint uses the 24-month core-phase timeframe. Extension outcomes have a longer timeframe, but the original placebo group received active treatment during the extension. Therefore, the extension experience does not represent a continued comparison with a group remaining on placebo as described for the core phase. The registry record does not report the planned sample-size justification or further detail about the extension’s analysis plan.
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 were randomized to receive fingolimod 0.5 mg, 1.25 mg or placebo for up to 24 months. Upon entry into the Extension phase, patients treated with fingolimod 0.5 mg or 1.25 mg during the Core phase continued treatment at the same dose, those previously treated with placebo were re-randomized in to receive one of the two doses of fingolimod.
Core Phase
| Stage | Fingolimod 1.25 mg | Fingolimod 0.5 mg | Placebo (Core) | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg |
|---|---|---|---|---|---|
| STARTED | 370 | 358 | 355 | 0 | 0 |
| COMPLETED | 251 | 272 | 255 | 0 | 0 |
| NOT COMPLETED | 119 | 86 | 100 | 0 | 0 |
Reasons for not completing this period:
| Reason | Fingolimod 1.25 mg | Fingolimod 0.5 mg | Placebo (Core) | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg |
|---|---|---|---|---|---|
| Withdrawal by Subject | 35 | 24 | 35 | 0 | 0 |
| Adverse Event | 28 | 22 | 16 | 0 | 0 |
| Lost to Follow-up | 17 | 13 | 21 | 0 | 0 |
| Abnormal laboratory value(s) | 19 | 14 | 2 | 0 | 0 |
| Unsatisfactory therapeutic effect | 10 | 6 | 17 | 0 | 0 |
| Administrative problems | 5 | 3 | 5 | 0 | 0 |
| Protocol Violation | 3 | 2 | 2 | 0 | 0 |
| Abnormal test procedure result(s) | 1 | 2 | 1 | 0 | 0 |
| Condition no longer requires study drug | 1 | 0 | 1 | 0 | 0 |
Extension Phase
| Stage | Fingolimod 1.25 mg | Fingolimod 0.5 mg | Placebo (Core) | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg |
|---|---|---|---|---|---|
| STARTED | 203 | 217 | 0 | 105 | 107 |
| COMPLETED | 172 | 180 | 0 | 89 | 88 |
| NOT COMPLETED | 31 | 37 | 0 | 16 | 19 |
Reasons for not completing this period:
| Reason | Fingolimod 1.25 mg | Fingolimod 0.5 mg | Placebo (Core) | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg |
|---|---|---|---|---|---|
| Withdrawal by Subject | 7 | 11 | 0 | 7 | 5 |
| Adverse Event | 13 | 9 | 0 | 3 | 5 |
| Lost to Follow-up | 2 | 6 | 0 | 2 | 4 |
| Abnormal laboratory value(s) | 4 | 2 | 0 | 2 | 2 |
| Administrative problems | 2 | 4 | 0 | 1 | 1 |
| Unsatisfactory therapeutic effect | 3 | 4 | 0 | 0 | 1 |
| Abnormal test procedure result(s) | 0 | 1 | 0 | 0 | 1 |
| Protocol Violation | 0 | 0 | 0 | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: Full analysis set, including all patients who were randomized and took at least one dose of study drug.
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.
| Characteristic | Unit | Fingolimod 1.25 mg | Fingolimod 0.5 mg | Placebo (Core) | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg | Total |
|---|---|---|---|---|---|---|---|
| Participants analysed | 370 | 358 | 355 | 105 | 107 | 1295 | |
| Age Continuous | years | 40.9 (SD 8.90) | 40.6 (SD 8.39) | 40.1 (SD 8.42) | NA (SD NA) | NA (SD NA) | 40.5 (SD 8.58) |
| Age Continuous | years | 40.6 (SD 8.71) | 40.8 (SD 7.96) | NA (SD NA) | 39.8 (SD 8.32) | 41.1 (SD 8.11) | 40.6 (SD 8.28) |
| Gender: Female | participants | 281 | 275 | 288 | 0 | 0 | 844 |
| Gender: Male | participants | 89 | 83 | 67 | 0 | 0 | 239 |
| Gender: Female | participants | 148 | 160 | 0 | 88 | 85 | 481 |
| Gender: Male | participants | 55 | 57 | 0 | 17 | 22 | 151 |
6. Primary Endpoint
The registry lists the primary outcome as Aggregate Annualized Relapse Rate (ARR) Estimate up to Month 24, measured over 24 months.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | relapses per year |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the aggregate annualized relapse rate (ARR) estimate up to Month 24, measured in relapses per year. The registry defines a relapse as a new neurological abnormality, or worsening of a previously stable or improving abnormality, separated by at least 30 days from the preceding clinical demyelinating event. The abnormality must last at least 24 hours, occur without fever or known infection, and be confirmed by an independent evaluating physician.
ARR expresses relapse frequency in relation to time under observation. The registry’s definition takes the sum of confirmed relapses in a group, divides it by the sum of study days for that group, and multiplies by 365.25. This gives a rate intended to account for differences in time on study, rather than treating every participant as if observed for an identical period.
In general, rates are useful when the outcome is an event that can recur and when observation time may vary. A rate does not describe the probability that an individual participant will experience a relapse, nor does it indicate how relapses were distributed across participants. The endpoint is also not the same as the percentage who remained relapse-free; that is a separate secondary outcome. The primary analysis population included randomized patients who took at least one dose of study drug.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary estimates were 0.203 relapses per year for fingolimod 1.25 mg, with a 95% confidence interval from 0.165 to 0.249; 0.208 for fingolimod 0.5 mg, with a 95% confidence interval from 0.17 to 0.254; and 0.403 for placebo, with a 95% confidence interval from 0.342 to 0.475. These are model-based group estimates for the 24-month primary timeframe. Both fingolimod-group estimates are lower than the placebo estimate in the posted results.
The intervals describe uncertainty around each group’s estimated rate. In general, a confidence interval is a range of values compatible with the data and the statistical procedure under its assumptions; it is not a range containing a known fixed proportion of individual outcomes. These intervals help convey the precision of each estimate, but the registry record does not post a primary pairwise treatment-effect estimate, its confidence interval, or a primary p-value. The group estimates alone should therefore not be recast as a reported formal comparison.
The estimates were calculated using negative binomial regression, adjusted for treatment, pooled center, number of relapses in the previous 2 years before enrollment, and baseline expanded disability status scale (EDSS). The outcome description says the model estimated ARR from confirmed relapses and time on study. Consequently, the figures are adjusted rate estimates, not simple raw event counts divided by a single common follow-up duration. The registry record does not report further model diagnostics, the fitted dispersion parameter, or details of how missing outcome observations were handled.
Two secondary relapse-free outcomes are also posted: one through Month 24 and one through the end of study. Their estimates show a higher relapse-free percentage in each fingolimod group than in the placebo group for the core 24-month outcome, and higher percentages in the active-treatment groups than the placebo-labelled group for the end-of-study outcome. The latter spans an extension in which participants originally assigned to placebo received fingolimod. It is therefore not a comparison against a group that remained on placebo throughout follow-up.
These findings describe the registry’s posted estimates and their precision. They do not, by themselves, establish the clinical importance of a difference, resolve uncertainties about unreported analysis choices, or establish how results apply beyond the trial population. The registry record does not post reported analyses for the secondary outcomes. Unadjusted secondary comparisons should be read as exploratory rather than as a substitute for a prespecified, multiplicity-controlled analysis.
9. Statistical Methods Explained
What does an annualized relapse rate represent?
The endpoint counts confirmed relapses relative to time under study. The registry describes its calculation as the group’s summed confirmed relapses divided by the group’s summed study days, multiplied by 365.25. This converts the event frequency to a yearly rate scale. It is a group-level rate estimate, not a prediction that each participant will have the same number of relapses, and it does not reveal whether events were concentrated among a smaller number of participants.
Why use negative binomial regression?
The primary outcome is a count of recurrent events, with study time contributing to the rate calculation. In general, count outcomes can vary more than a simple Poisson model assumes; this is called over-dispersion. A negative binomial model allows more variability in the count outcome than the basic Poisson model, which can make it appropriate when such extra variability is present. That is a general explanation of the method, not a claim about which diagnostic checks were performed in this trial. The registry record names the model but does not report a dispersion estimate, goodness-of-fit assessment, or model diagnostics.
The registry says the ARR estimates were adjusted for treatment, pooled center, relapse count in the 2 years before enrollment, and baseline EDSS. Adjustment means that the model incorporates these listed factors when estimating the treatment-group rates. In general, adjustment can account for measured variables included in a model, but it cannot remove uncertainty from unmeasured factors or correct every possible model misspecification. The registry does not report the detailed coding of the adjustment variables or the exact model specification beyond the listed factors.
How should the confidence intervals be read?
Each primary rate is accompanied by a 95% confidence interval. In general, a confidence interval conveys the precision of an estimate under a statistical procedure and its assumptions. A narrower interval indicates greater precision than a wider interval on the same scale, while neither interval describes the spread of individual participants’ outcomes. The posted intervals apply to the group ARR estimates. The registry does not provide a confidence interval for a primary between-group contrast, so the individual group intervals should not be treated as though they were a direct interval for that contrast.
What is different about the secondary comparisons?
The posted relapse-free secondary outcomes are percentages rather than event rates. The registry’s secondary comparison calculations are unadjusted re-calculations from posted summary statistics. Such a calculation compares the posted proportions without incorporating the primary model’s listed adjustment factors. It can be useful for understanding the direction and precision of the posted summary contrast, but it is not the same as a sponsor-reported adjusted analysis. The registry record lists no reported analyses for these secondary outcomes.
The secondary comparison results include unadjusted p-values and confidence intervals for risk differences. A p-value is a measure, under a specified null model and assumptions, of how unusual data at least as discrepant as those observed would be. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. A confidence interval for a risk difference communicates uncertainty about that contrast on the difference scale. The posted secondary comparisons have p-values of <0.001, but their unadjusted status and the number of outcomes under consideration matter when interpreting them.
What do the sample-size illustrations mean?
The sample-size illustrations are planning calculations based on observed effects, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The rows describe planning for 100%, 75%, or 50% of an observed effect, and give participants per arm for 80% and 90% power. In general, such calculations estimate the group sizes a hypothetical future study might need under stated assumptions. They are not observed power, do not establish the probability that this trial detected a true effect, and are not evidence of the trial’s original design rationale. The registry record does not report the trial’s own sample-size justification.
These illustrations also depend on assumptions about the effect size and analysis. If a future study aims to detect a smaller effect, it generally requires more participants under otherwise comparable assumptions. That is a general planning principle, not a new calculation for this trial. The illustrations should not be treated as definitive requirements: actual planning also depends on the intended endpoint, follow-up, analysis, missingness, and other design features, which are not fully reported in the registry record.
What information is absent?
The registry record does not report a primary pairwise contrast, its p-value, or a detailed protocol-level statistical analysis plan. It also does not report whether interim analyses were planned or conducted. These omissions limit assessment of how the posted estimates relate to the complete prespecified inferential strategy. The primary estimates and secondary re-calculations remain interpretable as posted summaries, but they do not supply the missing protocol context.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary relapse-rate outcome and multiple secondary outcomes, including clinical relapse, disability, brain-volume, and lesion measures. It does not report whether the statistical plan controlled the overall type I error rate across multiple outcomes, doses, timeframes, or pairwise comparisons. The record also does not report a hierarchy of hypotheses or a specific multiplicity adjustment. Without those details, the inferential status of secondary comparisons cannot be established from the registry entry.
The secondary comparison p-values are unadjusted re-calculations. An unadjusted p-value evaluates an individual comparison under its stated assumptions, but does not account for the broader set of comparisons that may have been considered. In general, multiplicity matters because repeated testing can make at least one small p-value more likely even when no underlying effects are present. This does not show that a particular secondary finding is false; it means that isolated unadjusted results have less evidential weight than results interpreted under a prespecified multiplicity strategy.
Missing-data handling is not reported in the registry record. The primary population is described as randomized participants who took at least one dose, but the registry does not explain how missing relapse assessments, incomplete follow-up, treatment discontinuation, or missing covariates were handled in the model. It also does not describe sensitivity analyses for missing outcomes. These details matter because the amount and pattern of missing follow-up can affect estimates, particularly when outcome status is related to treatment or relapse experience.
The registry does not report whether interim analyses were planned or conducted, nor does it report the trial’s sample-size justification or a full statistical analysis plan. The posted primary outcome description does identify the model and adjustment factors, which gives useful information about the analysis, but it does not answer these additional design and analysis questions. Accordingly, the estimates can be described as posted, while claims about multiplicity control, missing-data robustness, or the full prespecified analysis strategy remain unverified.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Core: Fingolimod 1.25 mg | 53 | 370 | 14.3% |
| Core: Fingolimod 0.5 mg | 53 | 358 | 14.8% |
| Core: Placebo | 45 | 355 | 12.7% |
| Extension: Fingolimod 1.25 mg | 27 | 308 | 8.8% |
| Extension: Fingolimod 0.5 mg | 21 | 324 | 6.5% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Core: Fingolimod 1.25 mg | 335 of 370 |
| Core: Fingolimod 0.5 mg | 320 of 358 |
| Core: Placebo | 305 of 355 |
| Extension: Fingolimod 1.25 mg | 216 of 308 |
| Extension: Fingolimod 0.5 mg | 223 of 324 |
Most frequent serious adverse event terms
| Event term | Organ system | Core: Fingolimod 1.25 mg | Core: Fingolimod 0.5 mg | Core: Placebo | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg |
|---|---|---|---|---|---|---|
| Basal cell carcinoma | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 6 of 370 | 9 of 358 | 2 of 355 | 0 of 308 | 1 of 324 |
| Multiple sclerosis relapse | Nervous system disorders | 2 of 370 | 1 of 358 | 3 of 355 | 2 of 308 | 5 of 324 |
| Bradycardia | Cardiac disorders | 6 of 370 | 0 of 358 | 1 of 355 | 2 of 308 | 0 of 324 |
| Squamous cell carcinoma | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 3 of 370 | 1 of 358 | 2 of 355 | 1 of 308 | 1 of 324 |
| Syncope | Nervous system disorders | 2 of 370 | 2 of 358 | 1 of 355 | 1 of 308 | 1 of 324 |
| Abdominal pain | Gastrointestinal disorders | 1 of 370 | 2 of 358 | 2 of 355 | 0 of 308 | 0 of 324 |
| Migraine | Nervous system disorders | 1 of 370 | 1 of 358 | 2 of 355 | 1 of 308 | 0 of 324 |
| Angina pectoris | Cardiac disorders | 1 of 370 | 0 of 358 | 2 of 355 | 1 of 308 | 0 of 324 |
| Macular oedema | Eye disorders | 2 of 370 | 1 of 358 | 1 of 355 | 0 of 308 | 0 of 324 |
| Cholelithiasis | Hepatobiliary disorders | 0 of 370 | 1 of 358 | 2 of 355 | 0 of 308 | 1 of 324 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Core: Fingolimod 1.25 mg | Core: Fingolimod 0.5 mg | Core: Placebo | Extension: Fingolimod 1.25 mg | Extension: Fingolimod 0.5 mg |
|---|---|---|---|---|---|---|
| Nasopharyngitis | Infections and infestations | 88 of 370 | 84 of 358 | 85 of 355 | 50 of 308 | 52 of 324 |
| Upper respiratory tract infection | Infections and infestations | 92 of 370 | 87 of 358 | 86 of 355 | 49 of 308 | 40 of 324 |
| Headache | Nervous system disorders | 81 of 370 | 83 of 358 | 76 of 355 | 25 of 308 | 29 of 324 |
| Sinusitis | Infections and infestations | 45 of 370 | 55 of 358 | 45 of 355 | 31 of 308 | 27 of 324 |
| Cough | Respiratory, thoracic and mediastinal disorders | 51 of 370 | 52 of 358 | 53 of 355 | 23 of 308 | 24 of 324 |
| Nausea | Gastrointestinal disorders | 57 of 370 | 63 of 358 | 54 of 355 | 13 of 308 | 14 of 324 |
| Urinary tract infection | Infections and infestations | 45 of 370 | 47 of 358 | 54 of 355 | 23 of 308 | 24 of 324 |
| Diarrhoea | Gastrointestinal disorders | 52 of 370 | 48 of 358 | 43 of 355 | 13 of 308 | 16 of 324 |
| Dizziness | Nervous system disorders | 53 of 370 | 38 of 358 | 43 of 355 | 10 of 308 | 11 of 324 |
| Melanocytic naevus | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 33 of 370 | 38 of 358 | 44 of 355 | 18 of 308 | 21 of 324 |
Serious events with the highest arm-level rate: basal cell carcinoma (2.5%), bradycardia (1.6%), multiple sclerosis relapse (1.5%), major depression (0.8%), convulsion (0.8%).
The registry reports serious adverse events separately for the core and extension phases. In the core phase, serious adverse events affected 53 participants in the fingolimod 1.25 mg group, 53 in the fingolimod 0.5 mg group, and 45 in the placebo group. The corresponding groups at risk were 370, 358, and 355. In the extension, serious adverse events affected 27 participants in the 1.25 mg group and 21 in the 0.5 mg group; each group at risk was 308 and 324, respectively.
These figures are counts of participants affected, not event rates adjusted for time at risk. The registry does not provide serious-event time-at-risk estimates or a formal comparative analysis. The extension groups also reflect a different treatment context from the core phase, including placebo participants receiving fingolimod during the extension. Counts across phases should not be combined or compared as if they represented identical follow-up and treatment conditions. Death counts are not reported in the registry record.
The listed serious-event terms include basal cell carcinoma (2.5%), bradycardia (1.6%), multiple sclerosis relapse (1.5%), major depression (0.8%), convulsion (0.8%), and nephrolithiasis (0.8%). These term percentages should be read as registry-reported descriptions, not as a complete safety profile or proof of causation. The registry summary does not establish that a listed event was caused by treatment, and the absence of a term from the list does not show that it did not occur.
12. What the Result Does, and Does Not, Mean
The primary estimates describe the annualized rate of confirmed relapses during the 24-month core phase for each treatment group in the full analysis set. Both fingolimod dose groups had lower posted ARR estimates than the placebo group. The confidence intervals communicate uncertainty around those group estimates. These are useful descriptive results, but the registry record does not post a primary pairwise treatment-effect estimate or a primary p-value, so the group rates should not be presented as though they were a complete reported test of the primary comparison.
The results do not mean that every participant experienced the group-average rate, or that an individual participant’s future relapse experience can be predicted from the estimate. ARR is a group-level rate based on recurrent confirmed events and time on study. It does not describe the severity of relapses, their distribution among participants, or other aspects of disease course. The separate relapse-free percentages answer a different question and are subject to their own uncertainties and analysis limitations.
The posted secondary comparisons show unadjusted estimates and p-values for relapse-free outcomes, but the registry does not report analyses for those outcomes or a multiplicity procedure. They therefore should not be treated as equivalent to a fully documented confirmatory analysis. The end-of-study outcome is especially different in context because participants originally assigned to placebo received fingolimod during the extension.
The results also do not establish that the findings apply to people or settings beyond the study population. The registry record does not provide enough detail to assess every aspect of missing data, deviations from assigned treatment, or model fit. Safety counts are descriptive and do not establish causal relationships. Overall, the posted information supports a careful account of the reported group estimates and their limitations, not a treatment recommendation or a claim about an individual’s likely outcome.
13. Why This Trial Matters Statistically
This trial illustrates why endpoint scale and model choice matter. The primary measure is a recurrent-event rate rather than a single binary outcome or a total count without regard to follow-up time. Its definition incorporates days on study, while its negative binomial regression accounts for the count-based structure and adjusts for listed baseline and center factors. Readers therefore need to distinguish the adjusted rate estimates from raw event totals and from the separate relapse-free percentages.
It also shows the value of reporting uncertainty alongside point estimates. Confidence intervals make the precision of each group estimate visible, while a formal treatment contrast would directly summarize the difference between groups. The registry provides the former for the primary endpoint but does not post the latter. That gap demonstrates why an endpoint description and group estimates do not fully replace a reported comparative analysis.
The study’s secondary outcomes raise another general statistical issue: multiple outcomes and comparisons need an interpretation plan. A set of unadjusted results may be informative, but without reported multiplicity control and analysis details it is harder to distinguish confirmatory findings from exploratory signals. Likewise, a longer extension can add follow-up experience, but treatment changes mean it does not preserve the original placebo-controlled comparison.
Finally, planning calculations based on observed effects should not be confused with the trial’s design justification or with observed power. They are illustrations under specified assumptions and can help readers understand how assumed effect size relates to planning. The registry does not report the original sample-size rationale. Together, these features make the record useful for teaching how model-based estimates, confidence intervals, secondary outcomes, extension design, and incomplete reporting shape statistical interpretation.
14. 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 specific to this registry record affect interpretation. First, the primary outcome results are group ARR estimates with confidence intervals, but no primary between-group contrast, confidence interval for that contrast, or primary p-value is posted. This restricts what can be said about the reported formal comparison and prevents reconstructing the sponsor’s primary inference from the available primary results alone.
Second, the analysis description gives the negative binomial model and its adjustment factors but not additional specification, diagnostics, or a detailed account of missing-data handling. The full analysis plan, interim-analysis information, and sample-size justification are not reported in the registry record. These omissions leave uncertainty about prespecified decisions and the robustness checks used.
Third, secondary numerical comparisons are unadjusted re-calculations, and reported analyses are not listed for those outcomes. The registry record does not describe multiplicity control. Consequently, the secondary findings deserve more cautious interpretation than a fully reported, prespecified analysis with a stated approach to multiple testing.
Fourth, the extension differs from the core comparison. The placebo group received fingolimod during the extension, and the record notes a protocol-amendment switch affecting participants taking 1.25 mg without fully describing that change in the arm text. End-of-study results therefore cannot be interpreted as a continuation of the original placebo comparison. Finally, serious adverse-event counts are descriptive, without a formal comparative analysis in the record; death counts are not reported. These limitations concern what the registry documents and do not establish that unreported analyses or information were absent from other study materials.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Efficacy and Safety of Fingolimod (FTY720) in Patients With Relapsing-remitting Multiple Sclerosis — NCT00355134.
- Registry record last updated 2012-08-07; status last verified 2012-08.
- Results first posted 2012-06-26.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT00355134 is a completed, randomized, triple-masked, parallel-group phase 3 study of oral fingolimod versus placebo in patients with relapsing-remitting multiple sclerosis. The registry lists an enrollment of 1083, a study start in 2006-06, and primary completion in 2011-06. The core phase compared fingolimod 1.25 mg, fingolimod 0.5 mg, and placebo for up to 24 months, with an optional extension.
The primary endpoint was the model-estimated annualized relapse rate through Month 24 in randomized participants who took at least one dose. The posted estimates were lower in both fingolimod groups than in the placebo group, with 95% confidence intervals reported for each group. A primary pairwise contrast and p-value are not reported in the registry record. Secondary relapse-free outcomes and descriptive serious adverse-event counts are also posted, with important differences between the core and extension treatment contexts. The registry does not report the full statistical analysis plan, missing-data approach, multiplicity strategy, interim analyses, or original sample-size justification.