Long Term Study of RBP 7000 in the Treatment of Subjects With Schizophrenia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Long Term Study of RBP 7000 in the Treatment of Subjects With Schizophrenia (NCT02203838), a na phase 3 study. The primary outcome is Participants With Treatment-Emergent Adverse Events (TEAE).

2. Clinical Question

3. Trial Design

This was a Phase 3, open-label, long-term safety and tolerability study of RBP-7000 in participants with schizophrenia. The registry lists a single-group model, no masking, and allocation as NA. The record describes enrollment of 500 and lists the study status as completed. The study start is recorded as 2014-06, and primary completion as 2016-09. Indivior Inc. is listed as the sponsor.

The described treatment was RBP-7000 120-mg subcutaneous injections every 28 days for 13 doses. Participants could enter as rollover participants from study RB-US-09-0010 or as de novo participants; the record notes that pre-study procedures varied for de novo participants depending on previous therapy. These different entry pathways matter when reading the safety groups: prior treatment history was not uniform, and the reported categories do not constitute randomized treatment arms.

The stated purpose was to assess long-term safety and tolerability and to continue collecting clinical outcome data. Listed clinical outcomes include changes in PANSS total and subscale scores and CGI-S scores at specified follow-up points. The record lists 6 outcome measures and 0 analyses. It does not report a statistical analysis plan or explain how outcome data were analyzed. The design therefore supports a descriptive reading of the posted safety counts, but it does not establish a controlled comparison of RBP-7000 against another treatment.

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: A total of 820 subjects were screened for study participation at 53 sites in the US. Overall, 500 participants were included in the Safety Population: 92 rollover participants and 408 de novo participants.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety population -- participants who received at least 1 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.

Population: Safety population

6. Primary Endpoint

The registry lists the primary outcome as Participants With Treatment-Emergent Adverse Events (TEAE), measured over Day 1 up to week 52.

The primary endpoint counts participants with treatment-emergent adverse events (TEAEs) during the period from Day 1 up to week 52. An adverse event is defined in the record as a study-related event representing a change in frequency or severity from a baseline event, if any, regardless of causal relationship or medical significance. A TEAE is an adverse event whose start date is on or after the first study dose date. Investigators determined whether events were related or not related to study drug.

This endpoint is used to describe safety experience during treatment. Counting participants rather than event episodes prevents a participant with repeated experiences in the same category from being counted repeatedly in that category. The record also states that a participant may appear in different categories. Serious adverse events are defined by specified outcomes, including death, a life-threatening event, hospitalization or prolonged hospitalization, persistent or significant disability or incapacity, or a congenital anomaly or birth defect.

These definitions do not make the endpoint a causal estimate. The count records events arising after treatment began, but timing alone does not demonstrate that treatment caused them. Nor does a participant count convey event severity beyond the separate serious-event classification. The registry’s safety population includes participants who received at least 1 dose; the reported results should be understood in that context.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a single-group, open-label study. That design determines what statistical interpretations are available. Descriptive counts can report how many participants experienced events in the observed study categories. Without randomized allocation to a comparison treatment, differences between the categories cannot be treated as unbiased estimates of the effect of RBP-7000. The rollover categories are defined by prior-study treatment history, while de novo participants entered through a different pathway. Those histories may accompany differences in participant characteristics or follow-up, but the registry record does not report an adjustment strategy.

What does a participant count represent?

The primary outcome is a count of participants with TEAEs, not a count of every event episode. The event definition is based on a start date on or after the first study dose. The record explains that a participant is counted only once within a category even if that participant had 2 or more adverse experiences, while the same participant may appear in different categories. Consequently, these counts answer a participant-level question: whether participants were recorded with an event in a category during the stated period. They do not describe the number of episodes, duration, or recurrence of events.

The denominator is the number at risk reported for each category, and the endpoint population is the safety population of participants who received at least 1 dose. Descriptive counts are informative about what was recorded in those groups. A proportion can sometimes aid interpretation when its numerator and denominator are clearly defined, but no new proportions are calculated in this explanation. Comparisons also require attention to differences in group composition and follow-up, not just the displayed counts.

Why does the lack of a comparator matter?

In a randomized comparison, allocation can help make treatment groups comparable on average, supporting an estimate of the treatment contrast. This study has allocation recorded as NA and a single-group model. Its rollover categories are not a substitute for randomized groups: prior treatment history defines them, and the registry explicitly cautions that they are not a like-for-like control comparison. Open-label conduct also means that treatment knowledge could influence reporting or assessment. These features do not invalidate the observed counts, but they limit causal interpretation.

For a comparative safety claim, statistical methods would need to address the comparison being made, relevant differences between groups, and uncertainty around the estimate. The registry record does not report such a comparative analysis. It also does not report confidence intervals or hypothesis tests for the primary endpoint. A confidence interval, in general, expresses the range of estimates compatible with the data and a specified procedure; it helps communicate precision, but does not by itself remove design limitations or establish causation.

What is reported about analysis?

The record lists 6 outcome measures and 0 analyses. It describes the primary TEAE measure and lists secondary PANSS and CGI-S changes, but does not give the statistical methods used to analyze those outcomes. The analysis plan, including any model choices, covariate adjustments, handling of repeated observations, or inferential procedures, is not reported in the registry record. Those methods should not be inferred from the outcome descriptions.

For count data, a descriptive summary can be appropriate when the aim is to state how many participants experienced an event. If the goal is to compare event occurrence, the analysis would need to match the endpoint and account for the study design and relevant observation time. General statistical methods may also account for over-dispersion, which occurs when variability exceeds what a simple count model assumes. That concept is general; the record does not say that an over-dispersion model or any other model was used.

The registry includes a planning-illustration description based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect and distinguishes participant-per-arm illustrations from total-event illustrations for time-to-event outcomes. These are planning assumptions, not evidence of the trial’s original design or observed power. The record does not report the trial’s own sample-size justification, and no numerical planning estimates are available in the information.

How should uncertainty and event timing be considered?

Safety event counts are shaped by the amount of observation as well as by event occurrence. Participants may have different exposure or follow-up durations; the registry record does not report person-time summaries or detail how incomplete follow-up was handled. A count alone therefore cannot be interpreted as an exposure-adjusted rate. General statistical practice distinguishes the number of affected participants from a rate that incorporates time at risk, but no such rate is reported for this endpoint.

Likewise, an event beginning after treatment starts is treatment-emergent under the registry definition, but that timing rule does not settle causality. The investigator’s relatedness classification is conceptually separate from whether the event meets the TEAE timing definition. Reading the outcome as a safety description, while keeping its design and reporting limits in view, is the most defensible statistical interpretation.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists several clinical outcome measures: changes in PANSS total score to Days 29, 169 and End of Study; changes in PANSS subscale scores to End of Study; and changes in CGI-S scores to Days 29, 169 and End of Study. Multiple outcomes and assessment times create the general possibility of multiplicity: when many statistical tests are conducted, the chance of at least one apparently notable result can increase. Whether and how that issue was addressed is not reported in the registry record.

The record lists 6 outcome measures and 0 analyses. It does not report an analysis hierarchy, a multiplicity adjustment, or an inferential procedure for these outcomes. The absence of reported analyses means there is no basis in the record to judge whether statistical testing was performed or how any testing was interpreted.

The primary endpoint population is specified as the safety population, defined as participants who received at least 1 dose. The record does not report the analysis population for each secondary outcome, whether all enrolled participants contributed to each assessment, or how missing outcome observations were handled. Missing-data rules, reasons for missingness, and sensitivity analyses are not reported in the registry record. These details matter because incomplete follow-up can affect which participants contribute to a particular time point and can influence interpretation of change-from-baseline summaries.

The registry’s reported endpoint timeframe is Day 1 up to week 52. The record does not state whether participants with shorter observation were handled differently in the primary safety summary. No assumptions about missingness, imputation, censoring, or interim analyses can be made from the information reported.

11. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: accidental overdose (3.6%), schizophrenia (3.6%), pulmonany embolism (3.6%), agitation (3.2%), suicidal ideation (3.2%).

The serious adverse-event counts are reported separately for the de novo and rollover categories. The record gives both the number at risk and the number affected: de novo participants, 25 among 408; rollover placebo, 3 among 28; rollover RBP-7000 90 mg, 3 among 31; and rollover RBP-7000 120 mg, 3 among 33. These counts describe participants recorded with serious adverse events in each category. They are not a randomized comparison, and the group sizes and treatment histories differ.

Seriousness follows the registry definition, which includes events resulting in death, a life-threatening event, hospitalization or prolonged hospitalization, persistent or significant disability or incapacity, or a congenital anomaly or birth defect. The record also reports deaths by group: 3 for de novo participants, 1 for rollover placebo, and 0 for each of the rollover RBP-7000 groups. Those figures are not identified as treatment-related deaths.

Listed serious-event terms include percentages, but the reported information does not clarify their denominator. They should be read as registry-reported term percentages, not as directly comparable group-specific risks. Counts, seriousness classifications, and investigator assessments of relatedness answer different questions; none alone establishes that treatment caused an event.

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

The results establish what the registry reports about observed adverse events in the safety population during the stated follow-up period. They offer descriptive information by de novo and rollover category, including serious-event counts, counts in the other adverse-event category, and deaths. The event definition and category-counting rule help clarify what those values represent: participants are counted within a category, not every episode, and a participant may be represented in different categories.

The results do not establish how the observed safety experience compares causally with another treatment. The study was open-label and single-group, and the registry cautions that the reported categories are not like-for-like controls. Different entry pathways and prior treatment histories further limit direct comparison. The record does not provide an adjusted comparative estimate or an inferential analysis of the primary endpoint.

Nor do the counts establish that an adverse event was caused by study drug. A TEAE is defined by its timing relative to the first dose, while investigators separately determined relatedness. The deaths recorded by category are not identified as caused by treatment. Serious-event classification describes regulatory seriousness criteria; it does not itself identify causation.

The study’s purpose also included continued collection of PANSS and CGI-S clinical outcome data. The registry lists those outcome measures but reports 0 analyses. Without reported results and analysis methods, the listed measures cannot support conclusions about symptom change or comparative clinical effects. The appropriate interpretation is therefore limited: the record documents safety-event observations and study aims, while leaving comparative effectiveness, causal attribution, and statistical uncertainty unresolved.

13. Why This Trial Matters Statistically

This trial illustrates an important distinction in clinical biostatistics: recording safety events and estimating a treatment effect are different tasks. A single-group open-label study can describe the adverse events observed among participants receiving treatment. That information may be useful for characterizing the study’s safety experience, particularly when the event definitions and analysis population are stated clearly.

At the same time, descriptive counts do not answer what would have happened to comparable participants under another treatment or without the study treatment. Randomized allocation is not part of this design, and the registry explicitly warns that the categories are not like-for-like controls. Differences between de novo and rollover groups can reflect their entry pathways and histories, in addition to any differences in observed events. Treating such group contrasts as treatment effects would overstate what the design can show.

The record also demonstrates why outcome definitions matter. TEAEs are identified by timing after the first study dose, while relatedness is assessed separately. The participant-level counting rule avoids counting repeated experiences multiple times within a category, but it does not describe recurrence or exposure-adjusted event rates. Seriousness, occurrence, and causation are distinct concepts and should not be collapsed into a single interpretation.

Finally, the record lists clinical outcome measures but does not report analyses for them. Statistical interpretation depends not only on which outcomes were planned, but also on how observations were analyzed, how missing data were handled, and how multiple outcomes were considered. Where those details are absent, restraint is part of sound analysis. The trial’s statistical value lies in clarifying both the descriptive safety information available and the comparative questions that this record cannot resolve.

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 design and reporting limitations constrain interpretation. The study is open-label, has a single-group model, and records allocation as NA. It therefore does not provide randomized control groups for estimating a treatment contrast. The registry specifically states that the adverse-event categories are not a like-for-like control comparison. De novo and rollover participants entered through different pathways, and rollover groups reflect different prior-study treatment histories.

The safety endpoint counts participants with events, not event episodes or exposure-adjusted rates. The information reported does not provide person-time, detailed follow-up duration by group, or a method for handling participants with incomplete observation. The primary timeframe is Day 1 up to week 52, but the registry does not explain whether all participants contributed the same observation period.

Although the record lists PANSS and CGI-S outcomes, it reports 0 analyses. Statistical methods, confidence intervals, hypothesis tests, missing-data procedures, multiplicity handling, and any adjustment for group differences are not reported in the registry record. The information also does not clarify the denominator for the listed serious-event term percentages. These gaps limit assessment of precision and prevent stronger comparative conclusions.

The record does not report the trial’s sample-size justification or explain any interim analyses or statistical plan. Enrollment is listed as 500, but that figure alone does not establish the precision of a particular endpoint estimate or the adequacy of a comparison. The planning assumptions described in the record are illustrative and should not be treated as the study’s original design rationale or as observed power.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Long Term Study of RBP 7000 in the Treatment of Subjects With Schizophrenia — NCT02203838.
• Registry record last updated 2018-09-28; status last verified 2018-08.
• Results first posted 2018-09-28.
• 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.

18. Record Summary

NCT02203838 is a completed Phase 3, open-label, single-group study sponsored by Indivior Inc. It studied RBP-7000 in participants with schizophrenia, with enrollment recorded as 500. The listed regimen was RBP-7000 120-mg subcutaneous injections every 28 days for 13 doses. Participants entered as de novo or rollover participants.

The primary safety outcome was participants with treatment-emergent adverse events from Day 1 up to week 52, in a safety population defined by receipt of at least 1 dose. The registry reports serious and other adverse-event counts by entry category, along with deaths and selected serious-event terms. The design does not support a randomized comparison across those categories.

PANSS and CGI-S clinical outcomes are listed, but the record reports 0 analyses. Comparative estimates, inferential methods, missing-data handling, and the statistical analysis plan are not reported in the registry record.