Cilengitide, Temozolomide, and Radiation Therapy in Treating Patients With Newly Diagnosed Glioblastoma Multiforme: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cilengitide, Temozolomide, and Radiation Therapy in Treating Patients With Newly Diagnosed Glioblastoma Multiforme (NCT00085254), a randomized phase 1, phase 2 study. The primary outcome is Dose Limiting Toxicities of EMD + RT and TMZ.

2. Clinical Question

The trial’s stated rationale was to study the side effects and best dose of cilengitide when used with temozolomide and radiation therapy, and to compare how well the regimens work in newly diagnosed glioblastoma multiforme. The registry record therefore combines a dose-finding safety question with a later dose-level comparison. The safety endpoint is not an efficacy endpoint, and the Phase II survival outcome does not replace the primary toxicity outcome.

The registry record does not report the detailed eligibility criteria, the exact randomized allocation procedure, or the planned statistical analysis for the Phase II comparison. Those details are not reported in the registry record, so the extent to which the listed population reflects narrower eligibility criteria cannot be determined from this record alone.

3. Trial Design

NCT00085254 is recorded as a completed Phase I/II, randomized, parallel-group trial with no masking. Enrollment was 112. The study start is listed as 2005-04, and primary completion as 2012-11. The sponsor was the National Cancer Institute (NCI). These registry dates describe the record; they do not, by themselves, establish when every analysis was conducted or posted.

The design included a safety run-in and a Phase II dose comparison. During the safety run-in, the listed cohorts were 500mg, 1000mg, and 2000mg of cilengitide, with 3, 4, and 6 participants, respectively, in the primary endpoint results. The record describes dose escalation in a stepwise fashion and says at least 3 participants per cohort would be used to review dose-limiting toxicity. It also says 6 participants would be enrolled to ensure 3 were evaluable because of a high drop-out rate. The record does not explain how these statements were applied to the reported cohorts.

For Phase II, the posted overall survival results are for a 500mg arm with 46 participants and a 2000mg arm with 48 participants. Both arms received temozolomide and radiation therapy as well as cilengitide. The listed Phase II initiation regimen gave cilengitide intravenously on days 1 and 4, repeating weekly for 10 weeks; temozolomide and radiation therapy were also given during the initiation course. The maintenance regimen descriptions are truncated in the registry record, so their full schedules are not reported in the registry record.

The record lists laboratory biomarker analysis and a pharmacological study in addition to the drug and radiation interventions. It does not report the randomization ratio, allocation concealment, masking of outcome assessment, or a detailed participant flow. The record also does not establish from the information posted whether every participant enrolled contributed to each endpoint.

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

Analysis population for the primary outcome, as defined in the registry: at least 3 pts per cohort will be used to review DLT rate for dose escalation in stepwise fashion. we will enroll 6 pts to ensure that 3 pts are evaluable due to high drop out rate.

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: pt had undergone previous surgery and had histological diagnosis of glioblastoma. All pt 18 years or older and KPS greater or equal to 60%

6. Primary Endpoint

The registry lists the primary outcome as Dose Limiting Toxicities of EMD + RT and TMZ, measured over 10 weeks.

The primary endpoint was the number of participants with dose-limiting toxicities (DLTs) over 10 weeks. The description says participants were evaluated from the first dose through the end of the initiation cycle, described as 6 weeks of radiation therapy, temozolomide, and EMD, followed by 4 weeks of EMD alone. DLT was assessed using Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. The description excludes known hematological toxicities of temozolomide from the DLT definition and identifies nonhematological toxicities of Grades 3-4 severity, with an exception for nausea and vomiting without sufficient antiemetic prophylaxis.

This endpoint is used in dose-finding because treatment-related toxicity can inform whether a dose is suitable for further study. In general, counting participants who meet a prespecified DLT definition provides a direct summary of the frequency of qualifying toxicities during a defined observation period. It is not a measure of tumor response or survival. The cohort design described in the record used DLT review to support stepwise dose escalation.

The posted primary results list 0 participants with a DLT in the 500mg cohort (3 participants), the 1000mg cohort (4 participants), and the 2000mg cohort (6 participants). These are the recorded cohort counts; they do not establish that no other adverse events occurred. The result also needs to be read in light of the small cohort sizes and the defined DLT criteria. The registry record does not report additional details about evaluability, missing endpoint observations, or how participants who discontinued were handled.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary count measure?

The primary outcome is a binary event measure: each participant either met the DLT definition during the stated observation period or did not. The registry reports counts, rather than a modeled risk estimate or a comparison between randomized groups. In general, a binary endpoint is often summarized by the number of participants experiencing the event and the number evaluated. This allows the reader to see the event count in relation to the cohort size, while recognizing that small cohorts leave substantial uncertainty about the underlying event frequency.

The DLT definition matters because a count is only interpretable in relation to the events that qualify. The registry refers to CTCAE version 4.0, excludes known temozolomide hematological toxicities, and describes qualifying nonhematological toxicity severity. The record’s wording includes an exception concerning nausea and vomiting without sufficient antiemetic prophylaxis. It does not report further adjudication procedures, whether the DLT definition was amended, or how incomplete observation was treated. Those details are not reported in the registry record.

How should the survival estimate be read?

Overall survival is a time-to-event outcome. A median survival is the time by which half of the observed population has experienced the event, as estimated from the survival data. It describes the center of the estimated time-to-event distribution; it does not describe every participant’s time or the total length of follow-up. A confidence interval communicates uncertainty around the estimate under the analysis method. Wider intervals indicate less precision than narrower intervals. The registry gives medians and 95% confidence intervals for the Phase II dose arms, with follow-up stated as up to 3 years.

The Cox regression hazard ratio compares the event hazard between groups over follow-up, under the Cox model. A hazard ratio of 1 represents no relative difference in the modeled hazards; values on either side indicate different estimated directions of the comparison, depending on which arm is in the numerator. The reported comparison is Arm 1 versus Arm 2, and the registry gives the hazard ratio as 0.83 with a 95% confidence interval of 0.5 to 1.3 and p=0.4. The interval’s width reflects uncertainty in the estimated relative hazard. The record does not report the model covariates, checks of model assumptions, censoring details, or an explanation of how missing survival information was handled.

A p-value is a measure of how incompatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. The reported p=0.4 is not evidence that the two dose arms are equivalent. Equivalence or noninferiority would require an appropriate design and prespecified margin, neither of which is reported in the registry record.

What do the planning illustrations mean?

The registry’s planning illustration is based on the observed effect, rather than being the trial’s own design justification. It assumes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It varies the effect to be detected across 100%, 75%, and 50% of the observed effect. For time-to-event outcomes, the figures are total events required under the Schoenfeld approach, not participants per arm. The posted illustrations list 905 and 1,211 total events for 80% and 90% power when planning for 100% of the observed effect; 1,608 and 2,153 when planning for 75%; and 3,618 and 4,843 when planning for 50%.

These figures are planning illustrations only. They are not the number of events observed in the trial, do not describe the trial’s achieved power, and do not establish the trial’s original sample-size target. A calculation that uses an observed effect as its basis is especially dependent on that estimate and the listed assumptions. The trial’s own sample-size justification is not reported in the registry record.

Why can a comparison be unavailable?

A comparative estimate needs an outcome structure and analysis that support a contrast between groups. The primary DLT results are from separate safety run-in cohorts at three dose levels, rather than the two Phase II survival arms. The registry record indicates that the primary binary endpoint does not have an estimable two-arm comparison. No recalculated two-arm estimate is therefore appropriate to describe for that endpoint. In general, recalculated estimates from summary statistics are unadjusted and may differ from a sponsor’s analysis when methods or analysis populations differ. The registry does not report a competing sponsor analysis for this primary endpoint.

For the secondary survival outcome, the registry reports a Cox regression comparison. That sponsor-reported estimate and p-value take precedence over any simple unadjusted reconstruction. The record does not report enough detail to reproduce the fitted model or independently assess its assumptions. No additional adjusted effect estimate should be inferred from the posted medians alone.

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("Arm 1 500mg (Safety Run In)", "ARM 2 1000mg (Safety run-in)", "Arm 3 2000mg (Safety Run-In)")
n      <- c(3, 4, 6)
events <- c(0, 0, 0)
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  Arm 1 500mg (Safety Run In)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  ARM 2 1000mg (Safety run-in)  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Arm 3 2000mg (Safety Run-In)  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0

11. Multiplicity, Missing Data and Analysis Populations

Interpretation depends not only on an effect estimate but also on which participants were analyzed, how missing observations were handled, and how multiple outcomes were treated. The registry record reports 112 enrolled participants, the three safety run-in cohort sizes for the primary outcome, and the Phase II arm sizes for overall survival. It does not provide a full participant flow or explain how enrollment relates to the endpoint-specific analysis populations. Whether all enrolled participants were included in each analysis is not reported in the registry record.

The record does not report the amount or pattern of missing primary endpoint data, the number of participants evaluable for DLT, or how participants who discontinued before the end of the 10-week observation period were classified. Although the primary endpoint description mentions a high drop-out rate when explaining the intended cohort size, it does not give the observed drop-out count or a missing-data procedure. The effect of incomplete follow-up on the posted DLT counts therefore cannot be assessed from the record.

For overall survival, the registry reports a Cox regression result, but does not describe the censoring rules, the number of deaths, the handling of lost follow-up, or whether the analysis followed an intention-to-treat population. These are not reported in the registry record. The absence of these details limits independent assessment of the survival analysis, even though the point estimate, confidence interval, and p-value are posted.

Multiple outcomes and dose comparisons can create multiplicity concerns: the more questions examined, the greater the possibility of chance findings. The registry record does not report a prespecified multiplicity strategy, a hierarchy of secondary outcomes, or whether any p-values were adjusted. It also does not report interim analyses or stopping rules, so their use cannot be inferred.

The study’s own sample-size justification and detailed statistical analysis plan are not reported in the registry record. The posted planning illustrations are not a substitute for either item. Without the analysis plan, it is not possible to establish whether the posted analyses followed prespecified methods or to distinguish planned analyses from later exploratory ones.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: extraocular muscle paresis (1.9%), nausea (1.9%), vomiting (1.9%), edema face (1.9%), death nos (1.9%).

The serious adverse-event figures are organized by cilengitide dose group. The record reports 0 serious adverse events among 52 participants at Dose 500, 0 among 6 at Dose 1000, and 1 among 54 at Dose 2000. The corresponding other adverse-event counts are 11 among 52, 0 among 6, and 14 among 54. These are counts of participants affected as reported in the registry; they are not DLT counts and should not be substituted for the primary endpoint.

The listed serious-event terms each have a reported frequency of 1.9%: Extraocular muscle paresis, Nausea, Vomiting, Edema Face, Death NOS, and Seizure. The record does not provide a dose assignment for each listed term, and the listed terms should not be attributed to a particular dose group. Death NOS is listed as a serious-event term; the separate deaths field is not reported for the dose groups.

These safety summaries are descriptive. The registry record does not explain whether the dose-group populations overlap across phases, how event coding and attribution were determined, or whether every event was considered treatment-related. A serious adverse event is not automatically a DLT, because the DLT endpoint uses its own definition and observation window. The figures also do not establish a causal relationship between a regimen and any event.

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

The primary result says that no DLTs were recorded in the listed safety run-in cohorts during the specified assessment period: 0 of 3 at 500mg, 0 of 4 at 1000mg, and 0 of 6 at 2000mg. This is a description of the observed counts under the registry’s DLT definition. It does not show that the risk of DLT is zero, establish that all dose levels have the same safety profile, or mean that participants experienced no adverse events. The separate adverse-event summaries include events beyond the primary DLT count.

The Phase II survival results describe median overall survival by dose level and a Cox regression comparison. The reported hazard ratio is 0.83 (95% confidence interval 0.5 to 1.3; p=0.4). The estimate is uncertain, and the reported result does not establish that one dose level produces longer survival than the other. It also does not demonstrate that the dose levels are equivalent. The trial record does not report a comparison with no cilengitide or with a different treatment regimen, so these results cannot establish the effect of adding cilengitide relative to such a comparator.

The two components answer different questions. The safety run-in supports description of DLT counts in small dose cohorts; the Phase II results provide a dose-level comparison of survival. Neither result alone establishes a broad clinical conclusion, and they should not be merged into one overall claim about benefit or harm. The registry record’s outcome definitions, analysis details, and participant flow are incomplete, which further constrains interpretation.

These findings apply to the outcomes and populations as recorded for this trial. They do not, by themselves, establish how the regimens compare in other populations, under other schedules, or against other treatment approaches. The trial is recorded as completed, but completion status does not resolve the uncertainty in the estimates or supply methodological details that are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial illustrates why a clinical study can have distinct statistical questions within a single record. The primary endpoint concerns dose-limiting toxicity during a dose-finding period, while the Phase II secondary outcome concerns overall survival by dose level. The event counts, median survival estimates, and Cox hazard ratio each answer different questions. Keeping these endpoints separate prevents a safety result from being mistaken for an efficacy result, or a survival comparison from being treated as a direct answer to the dose-finding question.

The primary endpoint also shows the limits of interpreting zero observed events. Small cohorts can be useful for an early safety review, but a count of zero still leaves uncertainty about the underlying event frequency. The number of participants and the operational definition of DLT matter as much as the count itself. Here, the registry describes a specified evaluation period and a definition that excludes certain known temozolomide hematological toxicities. The reported counts should be understood within those boundaries.

The Phase II analysis illustrates the value of reading an effect estimate together with its confidence interval and p-value. The hazard ratio alone gives an incomplete impression; its interval communicates substantial uncertainty, while the p-value does not establish equivalence. Median survival is a useful distributional summary but is not interchangeable with a hazard ratio. Each statistic has a distinct interpretation.

Finally, registry records can leave important analysis questions unanswered. Details about missing data, analysis populations, multiplicity, interim analyses, and sample-size justification are not reported. Recognizing those omissions is part of statistical interpretation: conclusions should be limited to what the posted outcomes and methods can support, without filling gaps by assumption.

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.

Several trial-specific limitations constrain interpretation of the posted findings. The primary DLT result comes from small safety run-in cohorts, with 3, 4, and 6 participants at the listed dose levels. Those counts are informative as observed events but provide limited precision about uncommon toxicities. The registry describes a planned minimum number of evaluable participants and a concern about drop-out, but does not report how many participants were ultimately evaluable or how incomplete observation affected the results.

The record lists 112 participants enrolled, while the survival analyses include 46 participants in the 500mg arm and 48 in the 2000mg arm. The reason for the difference between total enrollment and the posted Phase II analysis groups is not reported in the registry record. Participant flow, exclusions, and analysis-population definitions are not reported, limiting assessment of selection or attrition.

The survival estimate is imprecise, with a wide confidence interval for the hazard ratio and a wide interval around the 2000mg arm’s median. The registry does not report model covariates, assumption checks, event counts, censoring details, or missing-survival-data handling. The full maintenance treatment schedules are truncated in the arm descriptions, and the record does not report whether treatment exposure differed between groups.

The registry record also does not report a detailed statistical analysis plan, sample-size justification, multiplicity approach, interim analyses, or complete eligibility criteria. These omissions limit assessment of prespecification and generalizability. The listed serious adverse-event terms are not assigned to dose groups, and the record does not provide treatment-relatedness determinations. These limits mean the posted results should be interpreted as registry summaries rather than a complete account of trial conduct and analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Cilengitide, Temozolomide, and Radiation Therapy in Treating Patients With Newly Diagnosed Glioblastoma Multiforme — NCT00085254.
• Registry record last updated 2016-02-25; status last verified 2015-10.
• Results first posted 2016-02-25.
• 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

NCT00085254 is a completed randomized, parallel Phase I/II trial with no masking, sponsored by the National Cancer Institute (NCI). The record lists enrollment of 112 participants and evaluates cilengitide with temozolomide and radiation therapy for newly diagnosed glioblastoma multiforme. The safety run-in examined DLT counts across 500mg, 1000mg, and 2000mg cohorts; no DLTs were recorded in the listed cohort results.

The Phase II secondary survival outcome compares the 500mg and 2000mg arms. Median overall survival was 17.4 months and 20.8 months, respectively. The reported Cox hazard ratio was 0.83 (95% confidence interval 0.5 to 1.3; p=0.4). The uncertainty around this comparison does not establish superiority or equivalence. Important details about analysis populations, missing data, multiplicity, and the statistical plan are not reported in the registry record.