Testing the Use of A Single Drug (Olaparib) or the Combination of Two Drugs (Cediranib and Olaparib) Compared to the Usual Chemotherapy for Women With Platinum Sensitive Ovarian, Fallopian Tube, or Primary Peritoneal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Testing the Use of A Single Drug (Olaparib) or the Combination of Two Drugs (Cediranib and Olaparib) Compared to the Usual Chemotherapy for Women With Platinum Sensitive Ovarian, Fallopian Tube, or Primary Peritoneal Cancer (NCT02446600), a randomized phase 3 study. The primary outcome is Progression Free Survival Determined Using Response Evaluation Criteria in Solid Tumors Version 1.1 Criteria.

2. Clinical Question

3. Trial Design

This is a phase III, randomized, open-label, parallel-group trial. The registry reports enrollment of 579. The primary outcome is described for all randomized subjects, and the arm-level progression-free survival results list 187 participants for platinum-based chemotherapy, 189 for olaparib, and 189 for olaparib plus cediranib. A separate Japanese cohort of 13 is included in toxicity assessments but excluded from the progression-free survival analysis. The registry record does not explain how these counts relate to the overall enrollment figure.

The three treatment groups are standard platinum-based chemotherapy, olaparib alone, and olaparib with cediranib maleate. The olaparib-alone arm is described as oral olaparib twice daily, in 28-day cycles. The combination arm uses oral olaparib twice daily and oral cediranib maleate once daily, also in 28-day cycles. Both descriptions specify continuing in the absence of disease progression or unacceptable toxicity. The detailed regimen for the chemotherapy arm is not reported in the registry record.

The trial began on 2016-03-28, and its primary completion date is listed as 2020-02-23. Its status is active, not recruiting. Randomization can help balance prognostic factors across groups, while the lack of masking means participants and study staff knew which treatment was assigned. The registry record does not report the randomization procedure, allocation ratio, or further design details. It also does not report the original sample-size justification or the assumptions used in planning the trial.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All Randomized subjects

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: Eligible and Treated patients

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival Determined Using Response Evaluation Criteria in Solid Tumors Version 1.1 Criteria, measured over The protocol required lesion assessments every 9 weeks from cycle 1, day 1 for the first year, then every 12 weeks thereafter until disease progression. An average of approximately 10 months..

Progression-free survival is the time from study enrollment until documented disease progression or death from any cause. Progression is defined using Response Evaluation Criteria in Solid Tumors, version 1.1. The registry reports the median in months for each treatment group and gives a 95% confidence interval for each median.

This endpoint combines two events that matter to the time-to-event analysis: documented progression and death. Participants who were alive and without documented progression at their last follow-up were censored at the date of their last CT scan, or at the CT scan date before two missed assessments. Censoring means that follow-up ends without a recorded event at that time; it does not mean the participant had an event on the censoring date.

Lesion assessments were required every 9 weeks from cycle 1, day 1 during the first year, then every 12 weeks thereafter until progression. The registry states an average of approximately 10 months. Because progression is detected at scheduled assessments, its recorded date may reflect the assessment schedule as well as the underlying disease course. The endpoint is used to compare the timing of progression or death between groups, but it is not the same as overall survival, which is listed as a secondary outcome.

7. Results Posted in the Registry

• Arm I (Platinum-based Chemotherapy) vs Arm III (Olaparib, Cediranib Maleate): hazard ratio 0.86 (95% CI 0.66–1.10), reported p = 0.077 (as reported in the registry).

8. Statistical Interpretation

9. Statistical Methods Explained

Progression-free survival is a time-to-event outcome: each participant contributes information about the time until progression or death, or about the duration of follow-up without a recorded event. The registry reports medians and confidence intervals for the three treatment groups, plus a hazard ratio for the comparison of chemotherapy and olaparib with cediranib. The principal statistical ideas behind those summaries are different, so they answer related but not identical questions.

What does the median describe?

The median is the point in the estimated progression-free survival distribution by which half of the group is estimated to have experienced progression or death. It gives a familiar time-based summary, in months, but it does not describe the whole distribution. Participants may have substantially shorter or longer times, and the median alone does not show how the groups compare at every point during follow-up.

The posted medians are 10.3 months for chemotherapy, 8.2 months for olaparib, and 10.4 months for olaparib plus cediranib. Each is accompanied by a 95% confidence interval. A confidence interval conveys uncertainty in the estimated median; it is not a range that contains the outcome time for most participants. The intervals for the group medians should also not be treated as a formal test of the difference between groups. The registry's reported between-group analysis is the hazard ratio and log-rank result for chemotherapy versus the combination.

What does a hazard ratio describe?

A hazard is the instantaneous rate of an event among participants who have not yet had that event, conditional on remaining under observation and event-free up to that time. A hazard ratio compares hazards between groups over follow-up. A value of 1 represents the null value of equal hazards; values on either side indicate different estimated hazards according to the specified comparison direction. The registry reports a hazard ratio of 0.856, but does not specify the numerator and reference-group coding. The direction of the estimate should therefore not be translated into a claim that one named arm had a lower hazard.

A hazard ratio is not a ratio of median times, nor is it a percentage change in the median. It summarizes a relative event rate over time under the analysis used. Interpreting it as a constant relative effect requires assumptions about how the hazards relate over follow-up; the registry record does not report an assessment of those assumptions. The posted hazard ratio should be read alongside its confidence interval and p-value rather than as a complete description of the survival curves.

Why use the log-rank method?

The log-rank method is a common approach for comparing time-to-event outcomes. It uses the timing of events and the number of participants still at risk as follow-up proceeds, while allowing participants with incomplete event information to contribute until censoring. The registry identifies the log-rank method for the reported hazard-ratio analysis. It does not report further details about the model, covariate adjustment, or whether additional assumptions were checked.

The reported hazard ratio is 0.856, its 95% confidence interval is 0.663 to 1.105, and the p-value is 0.077. A p-value describes how incompatible the observed data are with a specified null hypothesis under the analysis assumptions; it is not the probability that the null hypothesis is true. The confidence interval includes 1, so the posted estimate remains uncertain about whether the event hazard was lower or higher in the numerator group relative to the reference group. The registry record does not report a prespecified significance threshold for the primary analysis.

How do censoring and assessment timing matter?

Participants who were alive and disease-free at their last follow-up were censored at the date of the last CT scan, or at the CT scan date before two missed assessments. Censored participants are not counted as having had progression or death at that time. Their observed follow-up still contributes information up to the censoring point. Time-to-event methods rely on assumptions about censoring, including that the reason follow-up ends does not systematically distort the event comparison after accounting for the analysis. The registry does not report additional analyses of censoring or missing follow-up.

Progression was assessed on a schedule: every 9 weeks from cycle 1, day 1 in the first year, then every 12 weeks thereafter. Because scans occur at intervals, the exact time progression begins may be unknown; the recorded time is tied to when progression is documented. This timing is relevant when interpreting differences in progression-free survival, especially when assessment schedules or adherence to assessments could differ. The registry record does not report comparative assessment completion or missing-scan counts.

What does randomization contribute?

Random assignment is intended to make treatment groups comparable on average with respect to prognostic factors, including factors that may not have been measured. This supports interpretation of group differences as treatment-assignment comparisons, subject to the conduct of the trial and the analysis. The trial is described as randomized and open-label. The registry record does not report the random sequence generation or allocation concealment methods, so those details cannot be assessed from the record.

The primary outcome population is described as all randomized subjects, with the Japanese cohort excluded from progression-free survival analysis. The registry record does not give more detailed analysis-population rules, such as the handling of treatment discontinuation or protocol deviations. It also does not report covariate-adjusted analyses or subgroup analyses for this endpoint. The posted primary comparison should therefore be understood in the context of the specified arm comparison and the analysis details that are actually recorded.

10. Confidence Intervals and Precision

The 95% confidence intervals for median progression-free survival are 8.7 to 11.2 months for platinum-based chemotherapy, 6.6 to 8.7 months for olaparib, and 8.5 to 12.5 months for olaparib plus cediranib. These intervals indicate uncertainty around each estimated median. The combination-arm interval spans a broader range of times than the chemotherapy-arm interval, and the olaparib-alone interval also expresses uncertainty around its median. The intervals are not individual participant ranges, and their overlap or non-overlap is not a replacement for a direct comparison analysis.

For the chemotherapy-versus-combination comparison, the reported hazard ratio is 0.856 and its 95% confidence interval is 0.663 to 1.105. This interval includes 1, the value corresponding to equal hazards. It is compatible with the null value and with hazard ratios on either side of it. The p-value is 0.077. Together, these figures indicate uncertainty in the estimated relative hazard; they do not establish that the groups are equivalent.

Precision depends on the amount and distribution of event information, not only on the number enrolled. Follow-up length, censoring, and the timing of events also affect how much information is available. The registry reports arm counts for the primary outcome but does not provide the number of progression-free survival events in each arm or the numbers censored. Those details are not reported in the registry record, limiting a fuller assessment of why the interval has its observed width.

The hazard-ratio interval is the uncertainty measure for the reported between-group analysis. The registry does not post a corresponding hazard-ratio interval for olaparib alone versus chemotherapy. The primary result therefore offers less precision information for that pairwise question than it does for the chemotherapy and combination comparison.

11. 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 planning figures are based on the observed effect and are expressed as total events, not participants per arm. For planning at 100% of the observed effect, the figures are 1,299 total events for 80% power and 1,739 total events for 90% power. For planning at 75% of the observed effect, they are 2,309 total events for 80% power and 3,091 total events for 90% power. For planning at 50% of the observed effect, they are 5,195 total events for 80% power and 6,955 total events for 90% power.

The assumptions for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. For a time-to-event outcome, the event totals are based on the Schoenfeld approach. The trial's own original sample-size assumptions and justification are not reported in the registry record.

In plain terms, the 80% power column gives the event total associated with an 80% chance of detecting the specified effect under the planning assumptions if that effect is present. The 90% power column gives the corresponding event total for a higher planned chance of detection. These are planning illustrations, not observed power for this trial. A smaller target effect requires more event information; reaching more events may require more participants, longer follow-up, or both, depending on event occurrence and censoring.

12. 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 = 0.856, so ln(HR) = -0.1555.
2. d = 4 × (1.96 + 0.842)² / [ln(0.856)]² = 1,298.6, rounded up to 1,299 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 <- 0.856                              # 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(2446600)
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 = 1,299 (80% power), 1,739 (90% power)
 75% of observed effect: events = 2,309 (80% power), 3,091 (90% power)
 50% of observed effect: events = 5,195 (80% power), 6,955 (90% power)
Simulated power with 1,299 events: 0.801 (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.

13. Group-Sequential Design Analysis

Many time-to-event trials, especially in oncology, stop or report early when interim data are overwhelming, so a single fixed-sample number understates the real design problem. The calculation below is an illustrative group-sequential design built from this trial's posted control median and hazard ratio. The registry record does not state the planned number of looks or the spending function, so the design is a planning illustration, not a reconstruction of the protocol. Assumed: one-sided α = 0.025, 90% power, three equally spaced looks on information, Lan-DeMets O'Brien-Fleming spending, 24 months of accrual, 60 months in total and a 1% monthly dropout hazard.

Under these assumptions the design needs about 2,117 participants in total, with the boundaries below.

Fixed design versus group-sequential alternatives

Group-sequential designs cost a small increase in maximum sample size and events compared with the fixed design, in exchange for the chance to stop early for overwhelming efficacy. O'Brien-Fleming spending keeps early boundaries very conservative and the final boundary close to the fixed-design value, whereas Pocock spending spends more alpha early.

Sensitivity to a smaller true effect

Planning for exactly the observed effect is optimistic, because observed effects in published trials tend to be somewhat larger than the true effect. The table repeats the main design (three looks, O'Brien-Fleming spending, 90% power) for smaller true hazard ratios.

For context, this trial enrolled 579 participants.

Cross-check, upper boundaries: gsDesign 3.71, 2.511, 1.993 | rpact 3.71, 2.511, 1.993. Agreement between the two packages confirms the boundary calculation.

R code and output

suppressWarnings(suppressMessages({ library(gsDesign); library(rpact) }))
med_control <- 10.3          # control-arm median, months (converted from the posted unit)
hr          <- 0.856          # hazard ratio used for planning
lambdaC     <- log(2) / med_control
eta         <- 0.01               # dropout hazard per month (assumption)
acc_months  <- 24                 # accrual duration (assumption)
study_months<- 60                 # total study duration (assumption)

design <- function(k, sfu, beta, h = hr) gsSurv(k = k, test.type = 1, alpha = 0.025, beta = beta, sfu = sfu,
  lambdaC = lambdaC, hr = h, hr0 = 1, eta = eta, gamma = 1, R = acc_months, T = study_months, minfup = study_months - acc_months)

# Main design: 3 looks, Lan-DeMets O'Brien-Fleming spending, one-sided alpha 0.025, 90% power
x <- design(3, sfLDOF, 0.10)
cat(sprintf("N_TOTAL|%s\n", format(ceiling(sum(x$gamma * x$R)), big.mark = ",")))
for (i in 1:3) cat(sprintf("GS|%d|%.1f|%.0f|%.3f|%.5f|%.5f\n", i, x$T[i], x$n.I[i], x$upper$bound[i], pnorm(-x$upper$bound[i]), cumsum(x$upper$spend)[i]))
cat("\n"); print(gsBoundSummary(x))

# Comparison of designs
cmp <- list(list("Fixed design, one analysis", 1, sfLDOF), list("Two looks, O'Brien-Fleming", 2, sfLDOF),
            list("Three looks, O'Brien-Fleming", 3, sfLDOF), list("Three looks, Pocock", 3, sfLDPocock))
cat("\n")
for (b in c(0.20, 0.10)) for (d in cmp) {
  y <- design(d[[2]], d[[3]], b)
  cat(sprintf("DES|%s|%d|%.0f|%s|%.1f\n", d[[1]], round(100 * (1 - b)), tail(y$n.I, 1), format(ceiling(sum(y$gamma * y$R)), big.mark = ","), tail(y$T, 1)))
}

# Sensitivity: what if the true effect is smaller than the observed one? (3 looks, O'Brien-Fleming, 90% power)
cat("\n")
for (fr in c(1, 0.75, 0.5)) {
  y <- design(3, sfLDOF, 0.10, exp(log(hr) * fr))
  cat(sprintf("SENS|%.2f|%.3f|%.0f|%s\n", fr, exp(log(hr) * fr), tail(y$n.I, 1), format(ceiling(sum(y$gamma * y$R)), big.mark = ",")))
}

# Independent cross-check of the boundaries with rpact
r <- getDesignGroupSequential(kMax = 3, alpha = 0.025, beta = 0.10, sided = 1, typeOfDesign = "asOF", informationRates = x$timing)
cat(sprintf("\nCross-check, upper boundaries: gsDesign %s | rpact %s\n", paste(round(x$upper$bound, 3), collapse = ", "), paste(round(r$criticalValues, 3), collapse = ", ")))

Method: LachinFoulkes 
     Analysis               Value Efficacy
    IA 1: 33%                   Z   3.7103
      N: 1560         p (1-sided)   0.0001
  Events: 586        ~HR at bound   0.7360
    Month: 18    P(Cross) if HR=1   0.0001
              P(Cross) if HR=0.86   0.0338
    IA 2: 67%                   Z   2.5114
      N: 2118         p (1-sided)   0.0060
 Events: 1172        ~HR at bound   0.8635
    Month: 28    P(Cross) if HR=1   0.0060
              P(Cross) if HR=0.86   0.5603
        Final                   Z   1.9930
      N: 2118         p (1-sided)   0.0231
 Events: 1758        ~HR at bound   0.9093
    Month: 60    P(Cross) if HR=1   0.0250
              P(Cross) if HR=0.86   0.9000



Cross-check, upper boundaries: gsDesign 3.71, 2.511, 1.993 | rpact 3.71, 2.511, 1.993

14. Analyses Reported by the Sponsor

The registry reports a hazard ratio of 0.856, with a 95% confidence interval of 0.663 to 1.105 and p = 0.077, for the comparison of platinum-based chemotherapy and olaparib plus cediranib. The analysis method is log-rank. These are the sponsor's reported results in the registry record.

No discrepancy is flagged between the posted analysis and the sponsor's reported result. The statistical interpretation therefore rests on the reported hazard ratio, interval, and p-value. There is no separate recalculated hazard-ratio estimate posted in the registry record. In particular, the reported result should not be replaced by an estimate inferred from the medians: a hazard ratio cannot be recovered simply by comparing the reported median times.

The estimate's direction is not mapped to a numerator and reference arm in the registry record. The comparison is labeled as chemotherapy versus the combination, but the record does not clarify which group is the reference in the hazard-ratio coding. The confidence interval includes 1, and the p-value is 0.077; those are the appropriate posted measures for describing the uncertainty in this comparison. The record does not report a sponsor hazard ratio for olaparib alone versus chemotherapy.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists progression-free survival as the primary outcome and identifies overall survival, frequency and severity of adverse effects, and patient-reported disease-related symptom scores among the secondary outcomes. It reports six outcome measures and one analysis. Those counts do not establish how many hypotheses were formally tested or how testing was organized. The registry record does not report a multiplicity-control procedure, a testing hierarchy, or whether any outcomes were designated as exploratory.

When several outcomes or pairwise comparisons are examined, the chance of at least one apparently unusual result can increase if each is interpreted without accounting for the others. Methods such as a prespecified hierarchy or adjusted significance thresholds can address this issue, but the record does not say whether such methods were used. The posted p-value for the chemotherapy and combination comparison should therefore be read as the reported result for that analysis, without assuming that it was adjusted for other outcomes or comparisons.

The primary outcome population is described as all randomized subjects, with the Japanese cohort excluded from the progression-free survival analysis. The Japanese cohort is included in toxicity assessments. The registry record does not report further analysis-population definitions, including detailed rules for treatment exposure, protocol deviations, or withdrawal. It does not report an intention-to-treat label or a per-protocol analysis for this endpoint.

The record describes censoring for participants who remained alive and disease-free at last follow-up: they were censored at their last CT scan date, or at the scan date before two missed assessments. It does not describe other missing-data methods, imputation, sensitivity analyses for missing outcomes, or the extent of missing scan data. The reported assessment schedule is every 9 weeks in the first year and every 12 weeks thereafter, but assessment completion by arm is not reported in the registry record. Consequently, the effect of missing or delayed assessments on the comparison cannot be evaluated from the posted information.

The registry does not report an interim-analysis plan, stopping boundaries, or details of any monitoring related to efficacy. Those design features should not be inferred from the reported trial phase or the final p-value.

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: pericardial effusion (7.7%), small intestinal obstruction (7.7%), anorexia (7.7%), acute kidney injury (7.7%), hypertension (6.3%).

The adverse-event summary lists serious adverse events for 18 of 187 participants in the platinum-based chemotherapy arm, 22 of 189 in the olaparib arm, and 29 of 189 in the olaparib-plus-cediranib arm. In the Japanese cohort, 1 of 13 participants is listed as affected by a serious adverse event. These counts describe participants affected, not the total number of serious-event episodes.

The summary also reports other adverse events affecting 169 of 187 participants in the chemotherapy arm, 184 of 189 in the olaparib arm, 184 of 189 in the combination arm, and 13 of 13 in the Japanese cohort. The registry lists deaths of 72, 95, 84, and 4 in these respective groups. It does not specify in this summary whether these deaths were caused by treatment, occurred during a particular follow-up period, or were otherwise related to an adverse event.

Leading serious adverse-event terms include pericardial effusion, small intestinal obstruction, anorexia, and acute kidney injury, each listed as 7.7%, along with hypertension at 6.3% and anemia at 4.8%. The record does not make the population or arm scope of these term percentages clear in the summary. These figures should not be treated as causal comparisons between treatments. The Japanese cohort contributes to toxicity reporting but not to the primary progression-free survival analysis.

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

The posted results describe median progression-free survival in three randomized groups and a hazard-ratio analysis for chemotherapy compared with olaparib plus cediranib. The chemotherapy median was 10.3 months, and the combination-arm median was 10.4 months. The reported hazard ratio was 0.856, with a 95% confidence interval of 0.663 to 1.105 and p = 0.077. The interval includes 1, so the estimate does not provide a clear distinction in event hazard between the groups.

This does not show that the treatments are equivalent. Equivalence requires a prespecified margin and an analysis designed to assess whether the effect lies within that margin; such a margin is not reported in the registry record. Nor does the result prove that there is no difference. The confidence interval represents uncertainty and remains compatible with multiple possible relative hazard values.

The olaparib-alone group had a median progression-free survival of 8.2 months, with a 95% confidence interval of 6.6 to 8.7 months. The registry does not post a hazard ratio or p-value comparing this arm with chemotherapy. It is therefore not possible to draw a formal inferential conclusion about that pairwise comparison from the posted primary results.

Progression-free survival is not overall survival, and the registry lists overall survival separately as a secondary outcome. The posted information also does not establish the comparative effect on patient-reported symptoms or adverse effects beyond the event counts described. These outcomes have distinct meanings and should not be inferred from the progression-free survival estimate.

The findings apply to the analysis populations and outcome definitions recorded for this trial. The Japanese cohort was excluded from progression-free survival analysis and included in toxicity assessments. The registry status is active, not recruiting, and its record includes the posted outcome figures. The results should be understood as statistical summaries of this trial rather than as individual predictions or treatment guidance.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares three treatment strategies within a randomized phase III design and reports a time-to-event primary outcome. The registry provides median progression-free survival estimates with confidence intervals for all three groups, while the reported hazard-ratio analysis focuses on chemotherapy and the olaparib-plus-cediranib combination. Seeing both summaries helps distinguish a median time from a relative event-rate comparison.

The result also illustrates why uncertainty matters. The reported hazard ratio is 0.856, but its 95% confidence interval of 0.663 to 1.105 includes the null value of 1, and the p-value is 0.077. A point estimate alone can suggest a direction without resolving how much uncertainty surrounds it. The interval and p-value keep the interpretation anchored to the precision of the evidence rather than to a single estimate.

The comparison among arms is not equally documented in the posted results. Median progression-free survival is reported for chemotherapy, olaparib, and the combination, but the hazard-ratio result is reported only for chemotherapy versus the combination. This highlights the difference between having descriptive summaries for several groups and having a formal between-group analysis for each comparison.

The trial also demonstrates how analysis populations can differ across outcomes. The Japanese cohort is excluded from progression-free survival analysis but included in toxicity assessments. Statistical interpretation requires attention to such outcome-specific populations, as well as to censoring and scheduled lesion assessments. The registry record leaves several design and analysis details unreported, which makes careful limits on interpretation an essential part of reading the result.

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.

The registry record does not report the hazard-ratio coding direction, so the estimate of 0.856 cannot safely be attributed as a lower hazard in a named arm. It reports no hazard ratio or p-value for olaparib alone versus chemotherapy, limiting formal comparison of that arm with the control. The medians are descriptive summaries and cannot fill that gap.

The trial is open-label. Knowledge of assigned treatment may affect aspects of care or reporting, although progression was assessed using RECIST 1.1 and scheduled imaging. The registry does not report blinded central review, consistency of assessment across arms, or comparative adherence to the CT or MRI schedule. These details are not reported in the registry record.

The Japanese cohort is excluded from the progression-free survival analysis but included in toxicity assessments. The registry reports enrollment of 579, arm counts of 187, 189, and 189 for the primary outcome groups, and a Japanese cohort of 13 for toxicity. It does not explain how these figures reconcile with one another. The primary outcome's censoring rules are described, but the record does not report numbers censored, event counts by arm, or sensitivity analyses addressing missing assessments.

The trial's original sample-size justification, primary analysis significance threshold, multiplicity strategy, covariate adjustment, interim analyses, and detailed analysis-population rules are not reported in the registry record. The planning illustrations based on the observed effect are not the original design justification. The reported serious adverse-event summary also does not establish causation or clearly specify the scope of the leading event-term percentages.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Testing the Use of A Single Drug (Olaparib) or the Combination of Two Drugs (Cediranib and Olaparib) Compared to the Usual Chemotherapy for Women With Platinum Sensitive Ovarian, Fallopian Tube, or Primary Peritoneal Cancer — NCT02446600.
• Registry record last updated 2026-10-01; status last verified 2026-09.
• Results first posted 2021-03-12.
• 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

NCT02446600 is a randomized, open-label, phase III, parallel-group study of platinum-based chemotherapy, olaparib, and olaparib plus cediranib for women with recurrent platinum-sensitive ovarian, fallopian tube, or primary peritoneal cancer. The registry reports enrollment of 579 and status active, not recruiting.

The primary outcome is progression-free survival by RECIST 1.1, measured from enrollment to progression or death. Posted median estimates are 10.3 months for chemotherapy, 8.2 months for olaparib, and 10.4 months for the combination. For chemotherapy versus the combination, the reported hazard ratio is 0.856, with a 95% confidence interval of 0.663 to 1.105 and p = 0.077. The Japanese cohort is excluded from the progression-free survival analysis and included in toxicity assessments.