Efficacy and Safety of Everolimus (RAD001) in Patients of All Ages With Subependymal Giant Cell Astrocytoma Associated With Tuberous Sclerosis Complex (TSC)(EXIST-1): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Everolimus (RAD001) in Patients of All Ages With Subependymal Giant Cell Astrocytoma Associated With Tuberous Sclerosis Complex (TSC)(EXIST-1) (NCT00789828), a randomized phase 3 study. The primary outcome is Percentage of Participants With Best Overall Subependymal Giant Cell Astrocytomas (SEGA) Response.

2. Clinical Question

3. Trial Design

EXIST-1, registered as NCT00789828, was a Phase 3, randomized, quadruple-masked, parallel-group study of everolimus and placebo. Its official title describes a randomized, double-blind, placebo-controlled study in patients with SEGA associated with TSC. The registry lists Novartis Pharmaceuticals as sponsor and records the trial as completed. The registry enrollment is 117.

The intervention was oral everolimus, starting at 4.5mg/m^2 daily and titrated to attain a whole blood trough concentration of 5 to 15 ng/mL. Dose adjustments were permitted based on safety and whole blood trough concentrations. The comparator was matching oral placebo. Although the registry records quadruple masking, it does not identify which roles were masked or provide further operational details about masking.

The primary outcome is reported for the core period at Week 48 and for an extension period lasting up to 4 years. The core-period groups are labelled Everolimus and Placebo. The extension result is labelled Everolimus (Extension Period). The registry also includes a safety group labelled Placebo (Core) Then Everolimus Treated (Extension Period). These labels establish that the record reports results by period and treatment group, but do not by themselves describe the timing or rules for entering the extension or the analysis handling of participants across periods.

The registry records a start date of 2009-08 and a primary completion date of 2011-03. The protocol's detailed allocation procedures, masking procedures, extension-entry criteria, and any stratification factors are not reported in the registry record. The record also does not report the rationale for the enrollment target or a trial-specific sample-size calculation.

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 117 participants were enrolled and randomized into the core period. Only 111 participants completing the core period, continued in the open-label extension period of the study.

Core Period (48 Weeks)

Reasons for not completing this period:

Open-label Extension Period (4 Years)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The primary analysis was performed in Full Analysis Set (FAS) population, defined as all randomized participants involved in the study.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Best Overall Subependymal Giant Cell Astrocytomas (SEGA) Response, measured over End of core period (Week 48), and end of extension period (up to 4 years).

The primary endpoint is the percentage of participants whose best overall status was SEGA response, assessed by Independent Central Radiology Review. It is a binary outcome at the participant level: each participant is classified according to whether the stated response criteria were met. The endpoint is not simply a change in lesion volume. Response required a 50% reduction from baseline in the combined volume of target SEGA lesions, confirmed with a second scan performed approximately 12 weeks later, along with no unequivocal worsening of non-target lesions, no new SEGA lesions of at least 1 cm in longest diameter, and no new or worsening hydrocephalus. Multi-phase brain MRI was used to identify SEGA lesions.

These combined criteria make the endpoint a measure of radiographic response that also incorporates specified evidence of progression or worsening. The central review is the stated source for determining response, rather than a local investigator assessment. The registry reports the percentage with best overall response at the end of the core period, Week 48, and at the end of the extension period, up to 4 years.

The endpoint is useful for describing whether participants met a prespecified set of imaging-based conditions during the assessment period. It does not, by itself, describe symptoms, functional outcomes, quality of life, or the course of every lesion outside the criteria. The registry states that Kaplan-Meier estimation was used for determining time to SEGA response; that time-to-event analysis is distinct from the posted percentage with best overall response.

7. Results Posted in the Registry

• Everolimus (Core Period): 34.6% (27/78) versus 0.0% (0/39) on control; risk difference 34.6 percentage points (95% CI 24.1 to 45.2); unadjusted two-proportion z-test p = <0.001.
• Everolimus (Extension Period): 57.7% (64/111) versus 0.0% (0/39) on control; risk difference 57.7 percentage points (95% CI 48.5 to 66.8); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in Frequency of Total Seizure Events Per 24 Hours at Week 24 in Both Core and Extension Period

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Everolimus (Core Period): mean -1.24 versus -0.24 on control; difference -1.00 (95% CI -3.25 to 1.25), standardized effect d = -0.17; z-test p = 0.383.
  • Everolimus (Extension Period): mean -6.07 versus -0.24 on control; difference -5.83 (95% CI -9.55 to -2.11), standardized effect d = -0.74; z-test p = 0.002.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Everolimus (Core Period) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Everolimus (Extension Period) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With Skin Lesions Assessed Using Physician's Global Assessement Overall Score

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Everolimus (Core Period): 41.7% (30/72) versus 10.5% (4/38) on control; risk difference 31.1 percentage points (95% CI 16.1 to 46.1), risk ratio 3.96 (1.51–10.40); unadjusted two-proportion z-test p = <0.001.
  • Everolimus (Extension Period): 58.1% (61/105) versus 10.5% (4/38) on control; risk difference 47.6 percentage points (95% CI 34.0 to 61.1), risk ratio 5.52 (2.15–14.14); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Everolimus (Core Period) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Everolimus (Extension Period) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes covering seizure frequency, SEGA progression and response timing, response duration and worsening, skin lesions, and everolimus blood concentration. The posted secondary result details include a change in seizure frequency and a skin-lesion assessment. The registry record does not report detailed numerical results for the other listed secondary outcomes in the information available for this analysis, so their pattern cannot be characterized from these data.

The secondary results address different questions from the primary imaging response. Seizure frequency is a repeated measure of events over time, while the skin-lesion outcome is a separate assessment. Those outcomes can add context about domains beyond SEGA imaging, but they do not validate the primary endpoint or establish that the same participants experienced the same pattern across outcomes. The secondary findings should be considered alongside the primary result rather than treated as interchangeable evidence.

The seizure-frequency comparison in the core period has an unadjusted recalculated p-value of 0.383, while the extension-period comparison has an unadjusted recalculated p-value of 0.002. These are not reported sponsor analyses. The posted skin-lesion comparisons have unadjusted recalculated p-values of p < 0.001. Their interpretation remains subject to the design, analysis, and multiplicity limitations described for secondary outcomes. Per-endpoint estimates and uncertainty are reported with their corresponding results rather than repeated in this overview.

Unadjusted secondary results carry less inferential weight because a registry record can list multiple outcomes and comparisons, while no multiplicity adjustment or prespecified testing hierarchy is reported. General: when several hypotheses are tested, an apparently noteworthy result can arise even if the underlying effects are absent. This does not make secondary outcomes uninformative; it means they are better treated as supportive or exploratory unless a prespecified analysis strategy establishes stronger confirmatory status. The record does not report that strategy or reported analyses for the secondary outcomes.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in Frequency of Total Seizure Events Per 24 Hours at Week 24 in Both Core and Extension Period
• Time to SEGA Progression
• Time to SEGA Response
• Duration of SEGA Response
• Time to SEGA Worsening
• Percentage of Participants With Skin Lesions Assessed Using Physician's Global Assessement Overall Score
• Duration of Skin Lesion Response in Everolimus Treated Participants
• Everolimus Blood Concentration (C2h) at 2 Hours Post Dose

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: participants either met the stated SEGA response criteria or did not. A percentage summarizes the share classified as responders in each group. This makes the result readily interpretable as an observed response proportion, but it depends on the response definition, assessment schedule, and analysis population. The registry defines the primary analysis population as the Full Analysis Set, described as all randomized participants involved in the study.

What does a risk difference describe?

A risk difference subtracts the response proportion in the comparator group from the response proportion in the treatment group. It expresses the contrast on the original proportion scale. The recalculated core-period risk difference is 0.346, with a confidence interval from 0.241 to 0.452. The extension comparison has a recalculated risk difference of 0.577, with a confidence interval from 0.485 to 0.668. These recalculations are unadjusted and use the posted summary statistics. They are not evidence that the registry reported an adjusted model; no reported analyses are listed.

General: a risk difference is often easier to relate to absolute event frequencies than a relative measure because it describes the separation between proportions. Its value depends on the event definition and the population being compared. Comparisons across periods require particular care when the groups may differ in their timing or composition. The registry does not describe enough extension-period details to determine whether those groups are comparable in the same way as the randomized core-period groups.

How should the p-values be read?

The posted p-values for the primary comparisons are unadjusted recalculations: p < 0.001 for the core-period comparison and p < 0.001 for the extension comparison. A p-value summarizes how incompatible the observed data are with a specified null model, under that model's assumptions. It is not the probability that the null hypothesis is true, and it does not quantify the probability that a result is due to chance. The registry record does not report an analysis plan specifying a primary hypothesis test, adjustment variables, or a multiplicity procedure. The recalculated p-values should not be treated as substitutes for a trial-reported inferential analysis.

General: unadjusted testing considers a comparison without accounting for other tests. When many outcomes or comparisons are examined, the chance of at least one apparently notable result can increase. Whether and how to account for that depends on the prespecified testing strategy, which is not reported in the registry record. The primary endpoint has a designated primary status, but the record does not state how its inference was protected against testing of other outcomes.

Why are there confidence intervals?

The arm-specific primary results are reported with 95% confidence intervals. The recalculated risk differences also have confidence intervals. General: a confidence interval conveys sampling uncertainty around an estimate under the method used to calculate it. It is not a guarantee that the true value lies in that particular interval, nor does it incorporate every source of uncertainty, such as outcome measurement, missing data, or limitations in study conduct. A wider interval indicates less precision than a narrower interval on the same scale.

The reported intervals for the extension comparison describe the recalculated contrast using the posted group summaries; they do not address whether the extension group and core-period placebo group are exchangeable. Statistical precision and design comparability are separate issues. A narrow interval cannot compensate for uncertainty about whether the groups are suitable for a causal comparison.

What role does the Kaplan-Meier method have?

The endpoint description says Kaplan-Meier estimation was used to determine time to SEGA response. General: Kaplan-Meier methods estimate the timing of an event while accounting for participants whose event time is not observed within the follow-up period, under assumptions about censoring. The registry does not post a time-to-response estimate or its analysis details in the primary result. The arm-level percentage with best overall response and a time-to-event estimate answer related but distinct questions: whether response occurred by an assessment period, and how long it took to occur.

The registry also lists a sample-size planning illustration for selected secondary comparisons. Its stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation, with planning based on fractions of the observed effect. These are planning illustrations based on observed effects, not the trial's own design calculation and not observed power. The registry record does not report a primary-endpoint planning illustration or the trial's original sample-size justification.

11. Confidence Intervals and Precision

The core-period everolimus response estimate is 34.6%, with a 95% confidence interval from 24.2% to 46.2%. The placebo estimate is 0.0%, with a 95% confidence interval from 0.0% to 9.0%. The interval around the placebo estimate is important: an observed percentage of 0.0% does not establish that the underlying response probability is exactly zero. The interval expresses uncertainty in the group estimate.

The recalculated core-period risk difference is 0.346, with a confidence interval from 0.241 to 0.452. This gives a range of values compatible with the unadjusted comparison under the calculation method. It is an interval for the contrast, not an interval for an individual's chance of response. The registry does not report the method used for the arm-specific confidence intervals or for the recalculated risk-difference intervals.

For the extension period, the everolimus response estimate is 57.7%, with a 95% confidence interval from 47.9% to 67.0%. The recalculated risk difference against the placebo core-period group is 0.577, with a confidence interval from 0.485 to 0.668. These intervals describe statistical precision around the estimates, but they do not establish that the extension comparison has the same design validity as the randomized core-period comparison.

General: confidence intervals reflect sampling uncertainty as modeled by the method used. They do not automatically account for missing outcome data, systematic measurement differences, or the consequences of selecting groups by follow-up period. Precision should therefore be considered alongside the population definition, endpoint rules, analysis methods, and study design. The registry record does not report enough information to evaluate several of those factors in detail.

12. Multiplicity, Missing Data and Analysis Populations

The primary analysis population is described as the Full Analysis Set, defined in the registry as all randomized participants involved in the study. The record does not give further detail about how that definition was implemented, how participants with unavailable assessments were handled, or whether all randomized participants contributed to each outcome. Missing-data methods, imputation procedures, and sensitivity analyses are not reported in the registry record.

The record identifies a primary endpoint and lists a range of secondary outcomes, but a multiplicity strategy is not reported. In particular, the registry record does not specify a hierarchy for testing secondary endpoints, a family of hypotheses, or an adjustment for testing multiple outcomes. The posted p-values discussed for the comparisons are unadjusted recalculations, not reported sponsor analyses. They should not be read as controlling the overall false-positive risk across the outcome set.

The registry does not report the original statistical analysis plan, the primary hypothesis test, covariate adjustment, handling of protocol deviations, or interim analyses. It also does not report a trial-specific sample-size justification. These omissions limit assessment of whether the posted summaries correspond to prespecified analyses and how the study was designed to address uncertainty.

General: missing data can affect an estimate when the availability of an outcome is related to participant characteristics or outcomes. The direction and size of that effect cannot be inferred without information about the missingness pattern and analytic approach. Likewise, multiplicity is a design and interpretation issue: multiple tests may change the meaning of nominal p-values, but an appropriate correction depends on what was prespecified and which comparisons were intended to support formal conclusions. Those details are not reported in the registry record.

The extension-period response result deserves additional caution because the registry does not describe extension entry, participant-level follow-up, or the composition of the comparison groups in enough detail to evaluate selection or time-related differences. This is separate from missing-data handling and multiplicity. Each is an independent limitation on the strength of inference.

13. 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: convulsion (33.3%), pyrexia (16.7%), pneumonia (16.7%), raynaud's phenomenon (16.7%), urinary tract infection (6.1%).

The registry reports serious adverse events for groups defined by treatment period. In the Everolimus Treated (Core and Extension Period) group, 33 participants were reported with a serious adverse event among 78 at risk. In the Placebo (Core) Then Everolimus Treated (Extension Period) group, 17 were reported with a serious adverse event among 33 at risk. In the Placebo Treated (Core Period) group, 3 were reported with a serious adverse event among 6 at risk. These counts describe the registry's reported groups and should not be converted into a direct treatment comparison without accounting for the differing group definitions and periods.

The listed serious adverse event terms include convulsion, pyrexia, pneumonia, Raynaud's phenomenon, urinary tract infection, and gastroenteritis. The registry lists corresponding percentages, but the denominator for those term-specific percentages is not reported in the registry record. Death counts are not reported in the posted safety data. The registry also reports other adverse events in each group, but their counts do not distinguish how many events each participant experienced.

General: adverse event summaries describe events observed during the recorded exposure and follow-up; they do not by themselves establish that treatment caused an event. The groups span different periods and treatment histories, so their counts are not a randomized, like-for-like safety contrast. The registry record does not provide sufficient detail to assess event timing, exposure duration, severity beyond the serious classification, or attribution.

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

The primary result describes the proportion of participants who met a particular radiographic response definition during the reported assessment period. In the randomized core-period comparison, the observed response proportion is higher in the everolimus group than in the placebo group, and the unadjusted recalculated risk difference has a confidence interval that does not include zero. This is a summary of the posted data under the specified endpoint definition, not a complete account of clinical benefit or risk.

The response definition combines change in target SEGA volume with conditions concerning non-target lesions, new lesions, and hydrocephalus. It therefore represents a specific imaging-based outcome. It does not establish the effect on every manifestation of TSC, seizure burden, daily function, quality of life, or outcomes not measured by the endpoint. The secondary outcome list includes some of these other domains, but the posted results do not provide a complete account of every listed measure.

The core-period randomized comparison and the extension-period result should not be treated as having equal inferential status. The core period directly compares the randomized everolimus and placebo groups. The extension comparison uses the placebo core-period group as its comparator, but the registry record does not report enough detail about extension enrollment and group composition to establish that this is a comparable randomized contrast. The larger extension-period response estimate is therefore a period-specific observation, not by itself proof of a lasting or increasing treatment effect.

Statistical significance does not establish clinical importance, certainty about individual outcomes, or absence of bias. Confidence intervals describe uncertainty under their calculation methods, while missing-data handling, multiplicity, and analysis details are not reported. The primary result is informative evidence about the stated endpoint in the analyzed population, but should be interpreted with these reporting limitations in view.

No result should be interpreted as a guarantee of response or nonresponse for an individual. The registry record does not support medical advice, nor does this statistical summary determine an appropriate treatment choice. It describes what was recorded for this trial and the limits of what can be concluded from those results.

15. Why This Trial Matters Statistically

EXIST-1 illustrates why a trial result should be read through its endpoint definition as well as its treatment labels. The primary outcome is not an unrestricted judgment of improvement: it is a centrally reviewed imaging response requiring a volume reduction and the absence of specified forms of worsening. Knowing those rules clarifies what the reported percentage means and what it leaves outside its scope.

The trial also illustrates the distinction between a randomized comparison and a later-period estimate. The core-period groups are the direct randomized comparison. The extension result adds follow-up information, but its comparison with the placebo core-period group cannot be assumed to preserve the same comparability without details on extension participation and analysis. A statistically precise estimate does not answer that design question.

Another statistical lesson is the value of reading effect estimates with confidence intervals rather than relying on p-values alone. The arm proportions and risk difference describe the size and direction of the observed contrast, while the intervals convey sampling uncertainty. The unadjusted p-values add a test-based summary but do not account for testing across the wider outcome set or for unreported analytic choices.

The record also demonstrates the importance of reporting methods that affect interpretation: missing-data handling, multiplicity, analysis plans, and sample-size justification. These details are not reported in the registry record. Their absence does not erase the posted results, but it limits the ability to assess whether the results align with prespecified analyses and how robust they are to alternative assumptions.

16. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to the information reported for this trial. The registry defines the primary analysis population as the Full Analysis Set, but does not describe how participants with missing imaging assessments were handled or whether all randomized participants contributed to each endpoint. This matters because the response classification depends on imaging at specified times, including confirmation with a second scan approximately 12 weeks later.

The primary endpoint is based on an imaging definition and central radiology review. The registry does not report further details about reviewer procedures, adjudication, or handling of indeterminate scans. The posted summaries also do not include the original analysis plan, adjustment methods, or a multiplicity strategy. Thus, the recalculated comparisons are useful summaries of posted counts and proportions, but not a replacement for a reported prespecified analysis.

The extension-period outcome is especially limited as a comparative estimate. The registry labels the group as Everolimus (Extension Period) and uses the placebo core-period group in the recalculated comparison. The record does not report enough information about who entered the extension, their treatment histories for the response analysis, or how follow-up differed by group. Consequently, the extension contrast may reflect differences in period, selection, or observation as well as treatment; the available information does not allow those influences to be separated.

Safety groups are also defined across different treatment periods, and the term-specific serious adverse event percentages have no reported denominator in the registry record. The registry does not provide enough detail on exposure duration, event timing, or attribution to support a like-for-like safety comparison. Finally, the registry does not report the trial's sample-size justification, interim analyses, or detailed statistical plan, limiting evaluation of design assumptions and analytic decisions.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Efficacy and Safety of Everolimus (RAD001) in Patients of All Ages With Subependymal Giant Cell Astrocytoma Associated With Tuberous Sclerosis Complex (TSC)(EXIST-1) — NCT00789828.
• Registry record last updated 2016-02-03; status last verified 2016-01.
• Results first posted 2012-05-01.
• 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.

20. Record Summary

NCT00789828, EXIST-1, is a completed Phase 3 randomized, quadruple-masked, parallel-group study sponsored by Novartis Pharmaceuticals. It enrolled 117 participants and compared oral everolimus with matching oral placebo in patients with SEGA associated with TSC. The registry records a start date of 2009-08 and a primary completion date of 2011-03.

The primary endpoint was the percentage of participants with best overall SEGA response, determined by Independent Central Radiology Review using imaging and specified lesion and hydrocephalus criteria. At the end of the core period, the reported response was 34.6% for everolimus and 0.0% for placebo. An extension-period everolimus result of 57.7% is also reported. Recalculated unadjusted comparisons are statistically distinguishable from zero, but the extension comparison is not established as equivalent to the randomized core-period comparison.

The registry does not report the detailed statistical plan, missing-data methods, multiplicity strategy, or sample-size justification. These omissions, together with limited extension-period and safety-reporting detail, constrain interpretation of the posted results.