A Long-term Safety Extension Study of Delta-9-tetrahydrocannabinol (THC) and Cannabidiol (CBD) in Multiple Sclerosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Long-term Safety Extension Study of Delta-9-tetrahydrocannabinol (THC) and Cannabidiol (CBD) in Multiple Sclerosis (NCT01610687), a na phase 3 study. The primary outcome is Incidence of Adverse Events as a Measure of Patient Safety.

2. Clinical Question

The official title refers to a double-blind, randomized, placebo-controlled study followed by an open-label assessment and extension. However, the registry fields for this record identify a single-group extension with no masking and no allocation value. These descriptions should not be treated as evidence of a randomized comparison within the extension. The posted record focuses on participants who continued into the extension and on adverse events reported during that period. It also lists secondary symptom and tolerability measures, but it does not provide their result values or analyses.

3. Trial Design

This is a phase 3 extension study record for GW-1000-02 in multiple sclerosis and spasticity. Its stated purpose is to assess long-term safety, tolerability and efficacy. The record lists 137 participants enrolled, one active-treatment group, a single-group model, allocation as NA and masking as none. The study status is completed. The registry record lists a start date of 2001-07 and a primary completion date of 2005-02.

The official title describes a larger sequence that includes a double-blind, randomized, parallel-group, placebo-controlled trial, then an open-label assessment and study extension. The entry analyzed is identified in its design fields as a single-group extension. Consequently, the randomized and placebo-controlled wording in the title does not supply a control arm for the extension results. The record gives no arm-level comparator data for the extension and explicitly indicates that the arms are not a like-for-like control comparison.

The intervention is GW-1000-02, classified as a drug and described as active treatment. The primary safety population is defined as subjects who took at least one dose after the end of the Part B open-label phase of the acute study and yielded on-treatment efficacy data. This population definition matters because it describes who contributes to the stated safety results; it is not necessarily interchangeable with every person enrolled in the broader study sequence.

The primary endpoint is the number of participants who experienced an adverse event during the extension, with a timeframe of up to 1206 days. Secondary outcomes include the number of sprays taken during the last 6 days of treatment and changes or assessments involving intoxication, global severity, pain, spasticity, tremor and bladder problems. Although these outcomes are named in the record, their numerical results are not reported in the registry record. A statistical analysis plan, details of follow-up visits, and the reasons for the stated timeframe are also not reported in the registry record.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects who took at least one dose of study medication after the end of the Part B open-label phase of the acute study, and yielded on-treatment efficacy data was classed as the efficacy and safety population.

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 Incidence of Adverse Events as a Measure of Patient Safety, measured over up to1206 days.

The primary endpoint counts participants who experienced an adverse event during the extension study. Its stated unit is participants, and the record describes the result as the number of patients who experienced an adverse event over a timeframe of up to 1206 days. The population is defined as subjects who took at least one dose of study medication after the end of the Part B open-label phase of the acute study and yielded on-treatment efficacy data.

An incidence count is useful for describing how many people in the treated group had an event during the period under observation. It gives a straightforward account of reported safety experience in this extension population. The registry record also distinguishes serious adverse events from other adverse events in its posted results, allowing those counts to be read separately.

A participant count does not describe every feature of the event experience. It does not indicate, by itself, how many events each participant had, when events occurred, how long they lasted, or how severe events were beyond the serious-event classification. The record does not give those details for the primary endpoint. Nor does a single-group count show what the event incidence would have been under a different treatment or without treatment. The endpoint is therefore descriptive for the participants observed in the extension, rather than a comparative estimate of treatment effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The results are best understood as descriptive safety statistics from a single treatment group. In a descriptive analysis, counts state what was observed in the recorded population. They do not automatically estimate a contrast between treatment strategies. Here, the registry lists participants at risk and participants affected for serious and other adverse events, but it does not report an inferential analysis or a comparator group for the extension.

What an incidence count represents

A participant-based adverse-event count records whether a person experienced an event in the period of interest. It differs from an event count, which tallies every occurrence. If one person experiences repeated events, a participant count may still count that person once for a given category. The registry endpoint is explicitly described as the number of patients who experienced an adverse event. The posted figures should therefore be read as affected participants, not as the total number of event episodes.

The at-risk count supplies the population context for the affected-participant count. In general, this helps readers understand the scale of the observed count. It does not by itself resolve differences in individual exposure duration or follow-up. The registry record gives a timeframe of up to 1206 days but does not report participant-specific time on treatment or a time-to-event analysis. A cumulative participant count can be difficult to compare across groups when follow-up differs, and no comparative group is listed for this extension.

Serious and other adverse events

Serious adverse events are presented separately from other adverse events in the registry results. This separation is useful because the categories communicate different classifications of safety events. The term-level percentages listed for serious events describe the reported terms, but the registry record does not provide the full coding rules, event narratives, or a complete list of all event terms. The categories should not be collapsed into a single measure unless a defined analysis supports doing so; no such combined analysis is reported in the registry record.

General statistical context: for event data, uncertainty depends on the number of participants observed, how events are defined and ascertained, and the amount of observation time. A confidence interval can express uncertainty around an estimated event frequency, but no confidence interval is reported for these counts. There is also no reported adjustment for follow-up duration. The figures are therefore descriptive, rather than precision estimates from a stated statistical model.

Why a comparison matters

General statistical context: a concurrent control group can help separate differences associated with treatment assignment from differences due to the condition, background care, participant characteristics, or the way outcomes are measured. Random allocation can support a causal comparison when the trial design and analysis are appropriate. Those principles do not create a randomized comparison where one is not reported. Although the official title describes a randomized placebo-controlled trial followed by an extension, the design fields for this record identify a single group, allocation as NA and masking as none. The extension results therefore do not provide a randomized control contrast.

For the same reason, a p-value or a treatment-effect estimate cannot be inferred from the counts. A p-value would require a specified question, statistical model or test, and comparison. A confidence interval would require a defined estimand and method. Those analyses are not reported in the registry record. The presence of counts and percentages alone is not evidence that a formal comparison was conducted.

Planning illustrations are not trial results

General statistical context: sample-size planning asks how many participants a future study might need under stated assumptions about the effect to detect, variability, significance level and allocation. A planning illustration based on an observed effect is not an observed trial outcome and does not establish that the original study had a particular power. The assumptions described for planning illustrations include a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and effects expressed as fractions of an observed effect. Those are planning assumptions, not evidence that this extension had two arms or used those assumptions in its own design.

Power is a property of a planned design under specified assumptions; it is not a measure of whether a result is true. In particular, calculating a sample size from a posted result cannot repair the absence of a comparison group or an unreported analysis. The registry record does not report its sample-size justification, interim analyses, or statistical analysis plan. Any planning illustration should therefore be read as a hypothetical design aid, not as a description of the extension's protocol or observed performance.

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("GW-1000-02")
n      <- c(137)
events <- c(126)
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  GW-1000-02  137  126  92.0  86.1  95.9  86.2  95.5  86.2  95.4

11. Multiplicity, Missing Data and Analysis Populations

The record lists a primary safety endpoint and several secondary outcomes, including medication sprays, intoxication, investigator-assessed global severity, pain, spasticity, tremor and bladder problems. It also reports eight outcome measures overall and zero analyses. The registry record does not explain how multiple outcomes were handled statistically. It does not report whether any adjustment for multiple testing was planned, whether outcomes were tested in a hierarchy, or which secondary measures were included in formal analyses. Without those details or posted analysis results, no multiplicity-adjusted interpretation is possible.

Missing-data methods are not reported in the registry record. The primary population definition refers to participants who took at least one dose after the open-label phase and yielded on-treatment efficacy data. This identifies a stated population, but it does not explain how missing safety observations, withdrawal, discontinuation, or incomplete follow-up were handled. The record also does not report the amount or pattern of missing data for the endpoint or secondary outcomes.

The registry does not report an intention-to-treat analysis, a per-protocol analysis, imputation methods, sensitivity analyses, or a rule for assigning participants with incomplete outcome information. These methods should not be inferred from the population description. The results are therefore interpreted as the posted counts for the stated population, with uncertainty about how incomplete observation may have affected them.

The extension is a single-group record, so the usual distinction between intention-to-treat and per-protocol comparative analyses cannot be evaluated from the posted results. The record does not report a statistical analysis plan, interim analyses, or a sample-size justification. Consequently, the analytical handling of the multiple outcomes and missing observations remains unclear in the registry record.

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 3.6% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: urinary tract infection not otherwise specified (nos) (2.9%), multiple sclerosis relapse (2.2%), balance impaired nos (1.5%), muscle spasms (1.5%), muscle weakness nos (1.5%).

The serious adverse-event figures list 23 affected participants among 137 at risk in the GW-1000-02 group. The other-adverse-event figures list 127 affected participants among 137 at risk. These are participant counts; they should not be read as totals of individual event episodes. Serious-event terms include Urinary Tract Infection Not Otherwise Specified (NOS) at 2.9% and Multiple Sclerosis Relapse at 2.2%. Balance Impaired NOS, Muscle Spasms, Muscle Weakness NOS and Vomiting NOS are each listed at 1.5%.

The registry record does not report deaths. This absence of a reported value is not the same as a reported zero. It also does not give event-level timing, duration, outcomes, exposure-adjusted rates, or a complete account of the handling of follow-up. The serious-event terms and percentages should therefore be understood as the listed registry results, not as a complete characterization of all safety experience.

Because there is no concurrent control group in this extension record, these counts describe events during active treatment but do not show how many would have occurred under another treatment or without treatment. The figures are useful for describing what was recorded among the participants observed, but cannot establish comparative risk or causation.

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

The results establish a descriptive account of adverse events recorded among participants in the GW-1000-02 extension. In the listed group, 23 participants were recorded with serious adverse events and 127 with other adverse events, with 137 at risk for each category. The registry also names several serious-event terms and their percentages. These details provide a view of the safety experience documented in the record, within the stated population and observation timeframe.

They do not establish whether GW-1000-02 caused any particular event. Participants with multiple sclerosis may experience health events for many reasons, and a single-group extension does not separate treatment-associated events from events that might arise from the condition, other care or other factors. The record does not offer a concurrent control group, a comparative effect estimate, or an inferential analysis of adverse-event incidence.

The findings also do not provide a complete description of how often individual event episodes occurred or how event experience varied over time. Participant counts do not show recurrence, duration, severity beyond the serious classification, or the time at which events began. The record does not report individual treatment exposure, deaths, the handling of missing observations, or a full listing of adverse-event terms. These omissions limit how far the counts can be generalized beyond the specific recorded population and period.

The listed secondary outcomes suggest that symptom measures and treatment use were also of interest, but their numerical results are not reported in the registry record. Therefore, this record does not support conclusions about changes in pain, spasticity, tremor, bladder problems, intoxication, global severity, or medication use. The stated study purpose includes efficacy and tolerability, but the presence of that purpose does not substitute for posted outcome estimates.

Overall, the record supports a careful description of the reported safety counts and their limits. It does not establish a comparative treatment effect, causal attribution, or the precision of an underlying event rate. Those distinctions are important when interpreting extension-study results without a concurrent control group.

14. Why This Trial Matters Statistically

This trial record is statistically instructive because its official title and its extension-level design fields describe different parts of a study sequence. The title refers to a double-blind, randomized, placebo-controlled trial followed by an open-label assessment and extension. The record fields for the extension identify a single group, no masking and allocation as NA. Correct interpretation depends on distinguishing the broader sequence named in the title from the specific data recorded for the extension. A controlled design in an earlier phase does not automatically make extension results comparative.

The case also illustrates the difference between an adverse-event count and a treatment-effect estimate. Counts describe what was observed in a group. To estimate whether event incidence differs between treatment strategies, a suitable comparison and analysis are needed. With no concurrent comparator reported, the safety results are descriptive. This distinction prevents a reader from treating event counts as evidence that a treatment caused or prevented an event.

Another statistical issue is the relationship between participant counts and time under observation. The endpoint spans a timeframe of up to 1206 days, but individual exposure durations and exposure-adjusted rates are not reported. The same participant count may have different implications when observation times differ, so the available information does not support a time-standardized comparison. The registry record also does not report confidence intervals, hypothesis tests or a statistical analysis plan.

Finally, the record names multiple secondary outcomes but gives no numerical results or analyses for them. This highlights why readers need outcome estimates, missing-data methods and a stated approach to multiplicity to interpret a multi-outcome study fully. Naming an outcome is not the same as reporting a result. The extension therefore offers a useful example of how registry fields, endpoint definitions and analysis reporting jointly shape the strength of statistical conclusions.

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.

This extension has no listed concurrent control group, and the registry identifies it as a single-group study with no masking and allocation as NA. The adverse-event counts cannot therefore be interpreted as comparative effects. The official title describes a larger randomized and placebo-controlled study sequence, but the extension-level record does not provide a randomized control contrast.

The posted safety results are counts of participants, not event episodes. The record does not state how recurrent events were handled, provide individual event timing or duration, or report exposure-adjusted rates. Although the timeframe is described as up to 1206 days, participant-specific follow-up and treatment exposure are not reported. This limits interpretation of differences in opportunity to experience an event.

The registry record does not report deaths. It also does not provide confidence intervals, hypothesis tests, a statistical analysis plan, missing-data methods, or details of how participants with incomplete follow-up were handled. The stated safety population is defined by treatment exposure after the open-label phase and availability of on-treatment efficacy data, but the consequences of that selection are not described.

The record lists secondary outcomes but does not report their results or analyses. It does not report how multiplicity was handled, whether there were interim analyses, or the sample-size justification. Serious-event terms are listed, but a complete event-term listing and event-level narratives are not reported. These limitations constrain the conclusions to the descriptive safety counts posted for the extension population.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Long-term Safety Extension Study of Delta-9-tetrahydrocannabinol (THC) and Cannabidiol (CBD) in Multiple Sclerosis — NCT01610687.
• Registry record last updated 2023-01-10; status last verified 2022-12.
• Results first posted 2012-08-17.
• 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 extension study of GW-1000-02 in multiple sclerosis and spasticity. It lists 137 participants enrolled, one active-treatment group, a single-group model, allocation as NA and masking as none. The stated primary endpoint is the number of participants who experienced an adverse event during a timeframe of up to 1206 days.

Posted results list 23 participants with serious adverse events and 127 with other adverse events, with 137 at risk in each category. The record lists several serious-event terms and percentages but does not report deaths. It names secondary outcomes involving medication sprays, intoxication, global severity, pain, spasticity, tremor and bladder problems, but their results are not reported in the registry record.

The extension provides descriptive safety counts, not a randomized comparison. Comparative analyses, confidence intervals, missing-data methods, multiplicity handling, a statistical analysis plan and sample-size justification are not reported in the registry record.