Testing the Combination of Olaparib and Durvalumab, Cediranib and Durvalumab, Olaparib and Capivasertib, and Cediranib Alone in Recurrent or Refractory Endometrial Cancer Following the Earlier Phase of the Study That Tested Olaparib and Cediranib in Comparison to Cediranib Alone, and Olaparib Alone: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Testing the Combination of Olaparib and Durvalumab, Cediranib and Durvalumab, Olaparib and Capivasertib, and Cediranib Alone in Recurrent or Refractory Endometrial Cancer Following the Earlier Phase of the Study That Tested Olaparib and Cediranib in Comparison to Cediranib Alone, and Olaparib Alone (NCT03660826), a randomized phase 2 study. The primary outcome is Progression-free Survival.

2. Clinical Question

3. Trial Design

This is a randomized, open-label, parallel-group phase II trial. The registry lists an enrollment of 288 and seven treatment arms. The study began on 2018-09-27, has a primary completion date of 2024-12-06 in the record, and is listed as active, not recruiting. The sponsor is the National Cancer Institute.

The arms include cediranib maleate alone in two reference groups, olaparib alone, cediranib with olaparib, olaparib with capivasertib, olaparib with durvalumab, and cediranib with durvalumab. The listed treatment schedules are not identical: for example, the record describes some drugs as daily oral treatment and others as treatment on specified days or as an intravenous infusion on day 1. Cycles are generally described as repeating every 28 days while there is no disease progression or unacceptable toxicity. The record does not provide a complete schedule for every arm in the available arm descriptions.

For progression-free survival, the listed analysis population is all randomized patients, or intent-to-treat. The registry gives median progression-free survival for each arm and reports five log-rank hazard-ratio comparisons. Two comparisons use cediranib reference group 1; three use cediranib reference group 2. The registry does not report a comparison for every possible pair of arms.

The record also lists overall survival, objective tumor response, and adverse events as secondary outcomes. Objective response counts are posted, but detailed overall-survival results are not reported in the registry record. The trial’s original sample-size justification, allocation ratio, randomization details, and complete statistical analysis plan are 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.

Only the first 6 of 7 groups are shown.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All patients randomized (Intent-to-Treat).

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.

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Progression-free Survival, measured over Scans were done every 8 weeks for the first year and then every 12 weeks thereafter until disease progression. Vital status was assessed every 3 months for 2 years and then every 6 months for 3 years (5-years total)..

Progression-free survival is the primary endpoint. The registry defines it as the time from study enrollment to investigator-determined disease progression or death from any cause, whichever occurs first. For patients who are alive and progression-free, the time at risk is censored at their last radiographic disease assessment. This definition combines disease worsening and death into one time-to-event outcome, so either event can determine the endpoint.

The registry reports the median progression-free survival in months for each arm, with a 95% confidence interval. In general, a median time-to-event estimate describes the time by which half of the relevant group is estimated to have experienced the event. It is useful when follow-up times differ among participants and when the outcome is not simply a yes-or-no status at one fixed visit. Participants who have not had progression or death by their last assessment can still contribute follow-up information through their censoring time.

Scans were scheduled every 8 weeks for the first year and every 12 weeks thereafter until disease progression. Vital status was assessed every 3 months for 2 years and then every 6 months for 3 years, for 5-years total. The timing and frequency of assessments matter because the progression date is determined from disease assessments rather than continuously observed. Further details about adjudication, censoring rules beyond the stated definition, or handling of missed assessments are not reported in the registry record.

7. Results Posted in the Registry

• Arm I (Cediranib maleate_Reference Group 1) vs Arm II (Olaparib): hazard ratio 1.45 (95% CI 0.91–2.30), reported p = 0.930 (as reported in the registry).
• Arm I (Cediranib maleate_Reference Group 1) vs Arm III (Cediranib Maleate, Olaparib): hazard ratio 0.70 (95% CI 0.43–1.14), reported p = 0.060 (as reported in the registry).
• Arm IV (Olaparib, Capivasertib) vs Arm VII (Cediranib maleate_Reference Group 2): hazard ratio 1.06 (95% CI 0.67–1.70), reported p = 0.620 (as reported in the registry).
• Arm V (Olaparib, Durvalumab) vs Arm VII (Cediranib maleate_Reference Group 2): hazard ratio 0.92 (95% CI 0.58–1.46), reported p = 0.360 (as reported in the registry).

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.

Objective Tumor Response

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm II (Olaparib): 12.5% (4/32) versus 24.1% (7/29) on control; risk difference -11.6 percentage points (95% CI -31.0 to 7.7), risk ratio 0.52 (0.17–1.59); unadjusted two-proportion z-test p = 0.238.
  • Arm III (Cediranib Maleate, Olaparib): 31.4% (11/35) versus 24.1% (7/29) on control; risk difference 7.3 percentage points (95% CI -14.6 to 29.2), risk ratio 1.30 (0.58–2.93); unadjusted two-proportion z-test p = 0.518.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm II (Olaparib) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm III (Cediranib Maleate, Olaparib) and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists overall survival, objective tumor response, and adverse events among the secondary outcomes. Detailed results are not equally available for each. Objective tumor response is reported as participant counts by arm, and the available recalculated comparisons cover olaparib and cediranib plus olaparib against cediranib reference group 1. Detailed overall-survival results are not reported in the registry record. Safety information is posted separately as counts of participants with serious and other adverse events, along with selected serious-event terms.

Across the secondary measures, the evidence is incomplete for drawing a unified conclusion about how the treatment groups compare. The response outcome captures tumor response rather than the timing of progression or death, so it is related to but distinct from the primary endpoint. A response count does not convey how long a response lasted, and it does not include the full time-to-event information used for progression-free survival. The registry’s response denominators also differ from the primary endpoint group counts; the reasons are not reported.

In general, secondary outcomes can help describe other aspects of a trial, but their results often carry less confirmatory weight than a prespecified primary endpoint. This is particularly important when multiple outcomes and treatment comparisons are examined. If many unadjusted comparisons are considered, some apparent differences can arise by chance. The registry record does not report a multiplicity adjustment or identify which secondary analyses, if any, were designated confirmatory. The response recalculations are unadjusted and have no sponsor-reported analysis accompanying them in the record.

Accordingly, the secondary findings provide context rather than a substitute for the primary progression-free survival results. They should not be combined into a single claim about the treatments. Detailed secondary endpoint definitions, missing-data handling, and analysis populations beyond the information stated in the record are not reported in the registry record.

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

• Overall Survival
• Objective Tumor Response
• Incidence of Adverse Events

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports progression-free survival as a time-to-event endpoint, with median estimates and 95% confidence intervals by arm. It reports the pairwise comparisons as hazard ratios analyzed using the log-rank method. These statistical summaries answer related but distinct questions: the medians describe event timing within arms, while a hazard ratio compares event rates between two groups over follow-up under a time-to-event framework.

What does a hazard ratio describe?

In general, a hazard is the instantaneous event rate among participants who have not yet had the event at a given time. A hazard ratio compares those rates between two groups over time. A value of 1 corresponds to equal hazards under the comparison’s orientation; values on either side indicate different relative hazards. Which arm is in the numerator and how the registry encoded each comparison determine the direction. The registry labels the contrasts by arm order and reports the hazard ratios, but it does not provide a fuller explanation of the coding direction. It is therefore safest to retain the reported pair order and estimate rather than infer a treatment advantage from the number alone.

What does the log-rank method contribute?

In general, the log-rank test compares the observed timing of events between groups across follow-up, taking account of the time at which participants are observed and censored. It is commonly used in time-to-event analyses. The registry names log-rank as the method for the five reported comparisons and gives a hazard ratio, confidence interval, and p-value for each. The record does not describe model specifications, proportional-hazards checks, stratification factors, or whether the reported hazard ratios came from a particular fitted model. Those details should not be inferred from the method label alone.

How should the confidence intervals and p-values be read?

In general, a confidence interval represents uncertainty around an estimate under the assumptions of the analysis. A wide interval indicates that a broad range of values is compatible with the data and method; a narrow interval indicates greater precision. For a hazard ratio, an interval including 1 is compatible with equal hazards as well as with differences in either direction. A p-value measures how unusual the data, or more extreme data, would be under a specified null hypothesis and model. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect.

The five reported hazard-ratio intervals each include 1. The estimates therefore leave meaningful uncertainty about the direction and magnitude of the pairwise differences. Interpretation also depends on the number of comparisons: testing several arms can increase the chance of a chance finding if results are interpreted without an appropriate multiplicity plan. The registry record does not report whether or how multiplicity was addressed.

What do the median estimates add?

In general, the median time-to-event estimate gives a readily interpretable summary of when the event has occurred for half of the group, accounting for censoring through the time-to-event analysis. It does not describe the full shape of the event-time distribution, nor does it quantify the uncertainty in a comparison between groups by itself. The arm-specific confidence intervals describe uncertainty around those medians. Comparing medians informally is not a substitute for the reported pairwise hazard-ratio analyses, because the latter use follow-up across the event times rather than only a single point on each arm’s distribution.

Progression-free survival is based on radiographic assessments and vital-status follow-up. In general, censoring allows a participant who has not had a recorded event by the last available assessment to contribute observed follow-up without treating that participant as event-free indefinitely. The registry gives the censoring endpoint for participants alive and progression-free, but does not report additional rules for missed scans, withdrawals, or other missing observations.

The registry identifies the primary population as all randomized patients, or intent-to-treat. In general, intent-to-treat analysis retains participants in their randomized groups for analysis, which preserves the comparison created by randomization. The record does not give a detailed account of exclusions, protocol deviations, or how the intent-to-treat designation was implemented for each posted measure. The reported arm counts for the primary outcome and the denominators for the secondary response outcome differ; reasons for those differences are not reported in the registry record.

11. Confidence Intervals and Precision

Precision is reflected in the width of an estimate’s confidence interval. For the arm-specific median progression-free survival estimates, the intervals vary in width, showing that the amount of uncertainty is not the same in every arm. For example, the estimate of 3.8 months for cediranib reference group 1 has a 95% confidence interval from 3.0 to 5.4 months, while the estimate of 5.6 months for olaparib plus capivasertib has an interval from 2.6 to 6.0 months. The latter interval covers a broader range of median values, indicating less precision for that estimate.

The pairwise hazard-ratio confidence intervals also span substantial ranges. The intervals for the five comparisons all include 1, so the posted results are compatible with no relative hazard difference as well as with differences in either direction. For instance, the interval for cediranib reference group 1 versus cediranib plus olaparib extends from 0.43 to 1.14, and the interval for olaparib plus capivasertib versus cediranib reference group 2 extends from 0.67 to 1.70. Such intervals do not identify a single precise relative effect.

Confidence intervals are not guarantees that the true value lies within the displayed limits. In general, they describe the long-run coverage behavior of a method under its assumptions. Their interpretation depends on study design, analysis choices, and the validity of the statistical model. The registry record reports 95% intervals but does not provide enough detail to assess every underlying assumption or how the intervals were calculated. The randomized group sizes for the primary endpoint range from 40.0 to 43.0, while the registry’s reported analytic counts vary across outcomes. Modest group sizes and limited event information can contribute to uncertainty, but the record does not report a formal explanation for the observed interval widths.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on observed effects, not the trial’s original sample-size justification. The planning assumptions listed are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row considers detecting the full observed effect or a stated fraction of it. For time-to-event comparisons, the figures are total events required, using the Schoenfeld approach. For other comparisons, the figures are participants per arm.

The 80% power column gives the number of total events, or participants per arm where applicable, that the planning calculation estimates would be needed for an 80% chance of detecting the specified effect under those assumptions. The 90% power column gives the corresponding requirement for a 90% chance. These are hypothetical planning targets, not estimates of the probability that the completed trial detected an effect. This is not observed power.

The calculation starts from an effect observed in the posted comparison and considers smaller fractions of that effect. A smaller effect is harder to distinguish from random variation, so the planning calculation requires more events or participants. The figures are not a retrospective claim that the trial should have enrolled a particular number, because the trial’s own design assumptions and target effect are not reported in the registry record.

For progression-free survival, event requirements are totals across the two compared groups, not per-arm counts. For objective response, the planning quantities are participants per arm. The illustrations rely on equal allocation and the other stated assumptions; they do not account for the complete multi-arm design, possible multiplicity adjustments, or all features of the actual trial unless those assumptions happen to match. The figures should therefore be read as general planning references for the specific pair and effect fraction, not as a verdict on trial adequacy.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 1.450, so ln(HR) = 0.3716.
2. d = 4 × (1.96 + 0.842)² / [ln(1.450)]² = 227.4, rounded up to 228 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 1.45                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(3660826)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 228 (80% power), 305 (90% power)
 75% of observed effect: events = 405 (80% power), 542 (90% power)
 50% of observed effect: events = 910 (80% power), 1,218 (90% power)
Simulated power with 228 events: 0.809 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

The five primary comparisons are reported as log-rank analyses in the registry record, with hazard ratios, 95% confidence intervals, and p-values. The independently recalculated material does not replace or contradict those sponsor-reported estimates; no discrepancy is flagged. Where a recalculation and sponsor analysis differ, the sponsor’s own analysis takes precedence, but that issue does not arise for the posted primary comparisons.

The reported hazard ratios should be interpreted in the order of the named arms and alongside their confidence intervals. The record does not give additional information about the hazard-ratio model, coding orientation, or analysis-plan details, so the recalculations cannot supply those missing design details. The planning illustrations use the observed results as a starting point and make stated simplifying assumptions; they are not the sponsor’s analysis and do not revise the trial’s reported p-values or confidence intervals.

For objective tumor response, the registry posts participant counts by arm, and the derived pairwise estimates are unadjusted recalculations rather than sponsor-reported analyses. The registry’s own response-analysis results are not reported in the registry record. Those recalculations should be treated as descriptive, simple comparisons based on the posted counts, not as replacements for a prespecified or adjusted sponsor analysis.

15. Multiplicity, Missing Data and Analysis Populations

The study has seven treatment arms and the registry reports five pairwise primary endpoint comparisons. It also lists secondary outcomes, including overall survival, objective tumor response, and adverse events. In general, making multiple comparisons increases the chance that one or more results may appear unusual by chance, even when no true differences exist. Multiplicity methods can address that concern by adjusting thresholds or controlling a chosen error rate. Whether any such approach was used is not reported in the registry record.

The registry labels the reported pairwise tests as superiority analyses and names the log-rank method. It does not report a hierarchy of hypotheses, a family-wise error strategy, or whether the five comparisons were prespecified. The p-values should therefore be read as reported for individual comparisons, without assuming that they account for the full set of arms, outcomes, or analyses.

For the primary endpoint, the record identifies the analysis population as all randomized patients, or intent-to-treat. It defines censoring for patients alive and progression-free as the time to the last radiographic disease assessment. The scan schedule and vital-status schedule are stated. Further details on missed assessments, withdrawal, loss to follow-up, or imputation are not reported in the registry record. In general, time-to-event analyses rely on assumptions about censoring, and the record does not give enough information to evaluate those assumptions fully.

The original statistical analysis plan, original sample-size justification, interim analyses, and rules for handling missing endpoint data are not reported in the registry record. The planning illustrations are separate from those original design choices and should not be used to infer them. The registry also does not explain why two cediranib reference groups are used or how the analysis plan handled the multi-arm structure.

16. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (10.8%), hypertension (10.8%), diarrhea (5.4%), weight loss (5.1%), dehydration (5.1%).

The safety results report counts of participants at risk and counts affected by serious adverse events and other adverse events in each arm. For serious adverse events, the posted affected counts range from 11 to 19 participants across groups; the at-risk counts range from 37 to 42. For other adverse events, the affected counts range from 37 to 42 among groups with 37 to 42 participants at risk. These are counts, not estimates of causal effects or adjusted comparisons between arms.

The registry lists abdominal pain and hypertension at 10.8% each among the leading serious adverse-event terms. Diarrhea is listed at 5.4%, and weight loss, dehydration, and acute kidney injury are each listed at 5.1%. These selected terms do not represent a complete account of every event or its severity. The record also gives death counts by arm but does not report in this summary whether deaths were treatment-related or how they were classified.

In general, adverse-event counts should be interpreted in light of the number of participants observed, exposure duration, event definitions, and how events were collected. Further details on these points, attribution to study treatment, and statistical comparisons of safety outcomes are not reported in the registry record. The counts alone do not establish whether an event was caused by a treatment.

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

The posted results describe progression-free survival in a randomized phase II study with seven treatment arms. They show arm-specific median estimates and five pairwise hazard-ratio analyses. The median estimates range from 2.0 months to 6.0 months, and the reported hazard-ratio confidence intervals all include 1. The comparisons therefore leave uncertainty about the direction and size of any pairwise differences.

These results do not support a simple ranking of all seven regimens. Only five selected comparisons are reported, and two distinct cediranib reference groups serve as comparators. The registry does not explain the selection of those comparisons, report an overall test across all arms, or provide a multiplicity strategy. Arm-specific medians should not be treated as if every arm had been formally compared with every other arm.

The results also do not establish why progression occurred or whether any differences in the posted outcomes were caused by a study treatment. The endpoint includes progression or death from any cause, and the record does not provide detailed causal analyses. The secondary response counts and adverse-event figures offer additional descriptive information, but they are not a substitute for the primary time-to-event comparison. Detailed overall-survival results are not reported in the registry record.

In general, a randomized design supports comparison of groups, but interpretation still depends on the analysis plan, completeness of follow-up, endpoint measurement, and precision. Important details about missing data, interim analyses, multiplicity, and original sample-size assumptions are not reported. The findings should therefore be understood as the estimates and uncertainty reported in this registry record, rather than as definitive evidence for or against a treatment.

18. Why This Trial Matters Statistically

This trial is statistically notable for its multi-arm structure and the use of two cediranib reference groups. The primary endpoint comparisons do not cover every possible pair of arms, making it important to keep each estimate tied to its named contrast. A numerical ranking of the arm medians would not answer the same question as a prespecified pairwise analysis and could obscure the uncertainty in the estimates.

The study also illustrates the difference between a time-to-event summary and a simple outcome count. Progression-free survival incorporates the timing of progression or death and allows follow-up to be censored at the last disease assessment for participants without an event. Objective response, by contrast, is posted as participant counts. Each measure addresses a different aspect of the observed clinical course, and differences in one should not be assumed to imply a similar difference in another.

Confidence intervals are central to interpreting the comparisons. The reported intervals span 1, while the arm-specific median intervals vary in width. This emphasizes that an estimate alone does not capture how much uncertainty surrounds it. The planned-power illustrations likewise help explain how detectable effects depend on event or participant counts, but they are not evidence of achieved power or a reconstruction of the original trial design.

Finally, the registry record leaves key methodological questions unanswered, including multiplicity handling, original sample-size rationale, and detailed missing-data procedures. This makes transparent reporting of the statistical plan especially important when interpreting results from multiple arms and outcomes.

19. 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 affect interpretation. The record reports only five primary endpoint pairwise comparisons among seven arms and does not explain why those comparisons were selected. It uses cediranib reference group 1 for two contrasts and reference group 2 for three others, but does not describe the relation between the reference groups or the reason for using both. No comparison across all arms is reported.

The arm labels and descriptions are not fully consistent. Arm VII is titled as cediranib reference group 2, while its description calls it reference group 1. Arm VI’s description ends after the words “Echo or MU,” leaving the listed assessment description incomplete. These record details do not change the reported numerical results, but they make the study documentation harder to interpret without clarification.

Primary endpoint arm counts range from 40.0 to 43.0, while objective response denominators range from 29.0 to 38.0. The registry does not explain why the outcome populations differ or detail the handling of participants without evaluable response data. The primary endpoint definition states a censoring rule, but additional rules for missed scans and other missing observations are not reported.

The reported hazard-ratio estimates have confidence intervals that include 1, and the record does not report multiplicity adjustment. This combination limits the strength of conclusions from individual pairwise p-values. Further, detailed overall-survival results, original sample-size justification, interim analyses, and a complete statistical analysis plan are not reported in the registry record. The available evidence thus supports cautious description of the reported estimates, not a complete assessment of all trial hypotheses or all sources of bias.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Testing the Combination of Olaparib and Durvalumab, Cediranib and Durvalumab, Olaparib and Capivasertib, and Cediranib Alone in Recurrent or Refractory Endometrial Cancer Following the Earlier Phase of the Study That Tested Olaparib and Cediranib in Comparison to Cediranib Alone, and Olaparib Alone — NCT03660826.
• Registry record last updated 2026-10-22; status last verified 2026-07.
• Results first posted 2025-09-23.
• 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.

23. Record Summary

NCT03660826 is a randomized, open-label, parallel-group phase II trial sponsored by the National Cancer Institute. It enrolled 288 participants and is listed as active, not recruiting. The study evaluates single-agent and combination regimens involving olaparib, cediranib, capivasertib, and durvalumab in recurrent, persistent, or metastatic endometrial cancer.

The primary endpoint is progression-free survival, defined from enrollment to investigator-determined progression or death. Median estimates and 95% confidence intervals are reported for seven arms, with five pairwise log-rank hazard-ratio comparisons. All five hazard-ratio confidence intervals include 1. Secondary outcomes include overall survival, objective tumor response, and adverse events; detailed overall-survival results are not reported in the registry record.

The registry does not report the original sample-size rationale, multiplicity handling, interim analyses, or full missing-data procedures. The planning illustrations are separate from the trial’s own design assumptions.