A Study to Evaluate Pazopanib in Comparison to Pemetrexed in Maintenance Setting in Non-progressing Subjects With Metastatic Stage IVA and IVB Non-squamous Non-small Cell Lung Cancer (NSCLC) Population: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate Pazopanib in Comparison to Pemetrexed in Maintenance Setting in Non-progressing Subjects With Metastatic Stage IVA and IVB Non-squamous Non-small Cell Lung Cancer (NSCLC) Population (NCT01313663), a randomized phase 2 study. The primary outcome is Progression Free Survival (PFS).

2. Clinical Question

3. Trial Design

This was a randomized, open-label, parallel-group, multi-center Phase II trial. Participants were assigned to maintenance pazopanib or maintenance pemetrexed after induction chemotherapy and a non-progressing response. Because treatment assignment was open-label, participants and investigators were not masked to the assigned treatment. Randomization can support a fair comparison by reducing systematic differences between groups on average, but the registry record does not report the randomization method or allocation-concealment procedures.

The brief summary describes an intended enrollment of approximately 200 participants randomized in a 1:1 ratio. The registry lists enrollment of 20 and status as terminated. The reason for termination is not reported in the registry record. These figures describe the planned and recorded enrollment, not the number contributing to every endpoint: the reported primary outcome gives arm sizes of 9 for pemetrexed and 11 for pazopanib. The registry identifies the primary analysis population as intent-to-treat, including everyone randomized regardless of whether treatment was administered.

The study started in 2011-02, with primary completion in 2012-07. The registry describes the study population in its title and summary as people with metastatic non-squamous NSCLC, while its condition listing says Lung Cancer, Small Cell. This is an inconsistency in the registry record; the displayed primary endpoint and trial description identify non-squamous NSCLC.

Safety and efficacy assessments were to be performed regularly, according to the brief summary. Their precise schedule, the randomization procedures, and the detailed statistical analysis plan are not reported in the registry record. The registry also does not explain why enrollment was lower than the approximately 200 described in the brief summary or why the trial was terminated.

4. Participant Flow and Accounting

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

Before assignment: Participants (par.) must have received at least 4-6 cycles of induction therapy with carboplatin + pemetrexed or cisplatin + pemetrexed and should have had stable disease, partial response, or complete response as best response at the time of screening/enrollment. Par. were stratified by disease stage and cycles of induction therapy at enrollment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Population: all participants who were randomized to study treatment regardless of whether or not treatment was administered

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival (PFS), measured over From randomization until the first documented sign of investigator-assessed disease progression or death, whichever occurred first (average of 10 study weeks).

Progression-free survival (PFS) measures the time from randomization until the first documented disease progression assessed by the investigator or death, whichever occurs first. It combines these events into a time-to-event outcome, capturing how long participants remain alive without documented progression during the study observation period. The registry specifies RECIST version 1.1 for assessing progression.

The registry's definition describes progression as at least a 20% increase in the sum of target-lesion diameters, using the smallest recorded sum since treatment started as the reference. For radiological progression, the date is the lesion-evaluation date. For symptomatic progression, the date is the symptomatic progression date, with radiological confirmation required. If a participant began subsequent anti-cancer therapy before documented progression or death, PFS was to be censored at the last adequate assessment before that therapy. Participants without a documented progression or death were also to be censored at their last adequate assessment.

PFS is useful in a maintenance comparison because it accounts for both whether an event occurred and when it occurred. The registry reports medians, which describe the time by which half of the observed group has experienced the endpoint under the analysis. It also reports a hazard ratio, a relative measure based on event timing over follow-up. These summaries answer related but different questions. The registry describes the timeframe as from randomization until progression or death, and labels it average of 10 study weeks.

7. Results Posted in the Registry

• Pemetrexed 500 mg/m^2 vs Pazopanib 800 mg: hazard ratio 0.92 (95% CI 0.32–2.65) (as reported in the registry).

8. Statistical Interpretation

9. Statistical Methods Explained

What does a median PFS represent?

For a time-to-event outcome such as PFS, each participant contributes information about the time from randomization until progression or death, or until the last assessment at which the endpoint status is known. The median is the time at which the estimated event-time curve reaches the halfway point. It is a compact description of the distribution, but it does not show every participant's event time or the full pattern over follow-up. The registry reports median PFS in weeks for each arm and gives a 90% confidence interval only as a lower limit in each arm; an upper limit is not reported in the registry record.

The median does not account for the entire shape of a time-to-event curve in the way a complete curve can. Two groups can have similar medians while differing at other points in follow-up, or have different medians without a uniform separation over time. No curve-specific numerical summaries beyond the posted values should be inferred from the medians alone.

What does censoring mean in this analysis?

Censoring is used when a participant's exact time to progression or death is not observed during the relevant period. The endpoint definition says that participants who started subsequent anti-cancer therapy before documented progression or death were censored at their last adequate assessment before that therapy. Those without a documented progression or death were censored at the date of their last adequate assessment. Censoring retains partial follow-up information rather than treating the last assessment as an event.

Time-to-event analyses generally rely on assumptions about censoring. In general, the analysis is most straightforward when censoring is not related to the participant's unobserved future event time, after accounting for relevant information. The registry states the censoring rules but does not report how many observations were censored, why participants had no further assessments, or whether the assumptions were assessed.

How should the hazard ratio be read?

A hazard ratio compares the event hazard between groups over time. In general, the hazard is the instantaneous event rate among participants who have not yet had the event. A value of 1 represents equal hazards under the comparison's coding; values on either side represent different relative hazards. The posted estimate is 0.92 and its 90% confidence interval is 0.32 to 2.65. Because the registry lists the comparison as pemetrexed versus pazopanib but the objective describes pazopanib relative to pemetrexed, the precise numerator and denominator convention is not clear in the record. The estimate should therefore be quoted as reported rather than assigned a directional interpretation beyond the registry's description.

In general, a hazard ratio summarizes relative event rates across follow-up and is often interpreted under a proportional-hazards model, which assumes the relative hazard is sufficiently stable over time. If that assumption does not hold, a single hazard ratio may be an incomplete summary. The analysis method and any assessment of this assumption are not reported in the registry record.

What role does randomization play?

Random assignment is intended to make treatment groups comparable at the point of allocation, including with respect to measured and unmeasured prognostic factors on average. It does not guarantee balance in a small trial. The registry describes randomization and a parallel design, but does not report the sequence-generation procedure, concealment method, stratification factors, or baseline characteristics in the registry record used for this interpretation. Open-label assignment means treatment was not masked, which can affect aspects of care or assessment. The primary endpoint was investigator-assessed, and the registry does not report whether outcome assessment was independently reviewed or blinded.

Why report both medians and a hazard ratio?

Medians give each group's event-time summary in the original time units, weeks. A hazard ratio provides a relative comparison between groups over time. Their combination can be informative because each expresses a different aspect of the outcome. Neither removes uncertainty: the medians have incomplete confidence limits in the registry record, and the hazard-ratio interval is broad. The p-value and method used for the sponsor's hazard-ratio analysis are not reported in the registry record, so the posted result cannot be fully evaluated as a formal hypothesis test from these details alone.

10. Confidence Intervals and Precision

The main comparative estimate is the hazard ratio of 0.92, with a 90% confidence interval from 0.32 to 2.65. This interval is wide relative to the point estimate. It indicates that the trial's estimate is not precise enough to narrowly characterize the relative event hazard. The interval includes 1, the value corresponding to equal hazards under the comparison's coding, and also includes values substantially below and above 1. The record does not report a p-value, so no p-value-based conclusion should be added to this interpretation.

Confidence intervals express uncertainty from sampling and the analysis method. In general, a narrower interval indicates greater precision, while a wider interval indicates that a broader range of effects remains compatible with the data and method. An interval is not a guarantee that the true value lies within its bounds, nor does it assign a probability to the true value after observing the data. The registry does not identify the hazard-ratio method or its assumptions, which limits assessment of the interval's construction.

The reported median PFS is 10.3 weeks for pemetrexed and 12.7 weeks for pazopanib. The respective reported lower confidence limits are 5.3 weeks and 6.0 weeks; upper limits are not reported in the registry record. These arm-specific summaries do not replace the comparative hazard ratio, and they should not be used to infer a precise between-group difference.

The registry lists enrollment of 20, and the primary endpoint arm counts are 9 for pemetrexed and 11 for pazopanib. The small recorded enrollment and broad interval are relevant context for precision, but the registry does not report the reason for termination or a detailed account of missing or censored observations. Precision should therefore be judged from the posted uncertainty estimates rather than from a binary claim about whether an effect was established.

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 illustrations use the observed effect as a starting point and show planning under 100%, 75% and 50% of that effect. Because the outcome is time-to-event, the entries are total event counts, not participants per arm. The displayed 80% power column gives the total events used in a planning illustration designed to have an 80% chance of detecting the stated effect under the assumptions; the 90% power column gives the event count for a 90% chance. These are planning targets, not observed trial outcomes.

The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The calculation uses the stated fraction of the observed effect as the effect to detect and applies the Schoenfeld approach for total events. The trial's own original design assumptions are not reported in the registry record. The brief summary describes approximately 200 intended participants, but it does not provide an original event target or a detailed sample-size justification.

The displayed planning values rise as the effect to detect becomes smaller: at 100% of the observed effect, the listed total event requirements are 4,516 for 80% power and 6,046 for 90% power; at 75%, they are 8,028 and 10,748; at 50%, they are 18,063 and 24,182. Smaller effects are harder to distinguish from random variation, so they require more events under these assumptions. These figures are not observed power. They are illustrative design calculations based on the observed effect and the specified assumptions, not an assessment of the probability that the completed trial detected an effect.

Event counts and participant counts are not the same. How many participants would be needed to accumulate a given number of events depends on event occurrence and follow-up, details that are not established by these illustrations. The planning figures should therefore be read as event targets, not as the actual enrollment or as a statement about the enrollment required for a future study.

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 = 0.920, so ln(HR) = -0.0834.
2. d = 4 × (1.96 + 0.842)² / [ln(0.920)]² = 4,515.7, rounded up to 4,516 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 <- 0.92                              # 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(1313663)
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 = 4,516 (80% power), 6,046 (90% power)
 75% of observed effect: events = 8,028 (80% power), 10,748 (90% power)
 50% of observed effect: events = 18,063 (80% power), 24,182 (90% power)
Simulated power with 4,516 events: 0.798 (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 6,336 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 20 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 <- 2.37692307692308          # control-arm median, months (converted from the posted unit)
hr          <- 0.92          # 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: 2990         p (1-sided)   0.0001
 Events: 2039        ~HR at bound   0.8485
    Month: 11    P(Cross) if HR=1   0.0001
              P(Cross) if HR=0.92   0.0338
    IA 2: 67%                   Z   2.5114
      N: 5132         p (1-sided)   0.0060
 Events: 4078        ~HR at bound   0.9244
    Month: 19    P(Cross) if HR=1   0.0060
              P(Cross) if HR=0.92   0.5603
        Final                   Z   1.9930
      N: 6336         p (1-sided)   0.0231
 Events: 6117        ~HR at bound   0.9503
    Month: 60    P(Cross) if HR=1   0.0250
              P(Cross) if HR=0.92   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 the sponsor's hazard-ratio analysis as 0.92, with a 90% confidence interval from 0.32 to 2.65. No p-value or analysis method is reported. The comparison is labelled Pemetrexed 500 mg/m^2 vs Pazopanib 800 mg, while the stated objective describes pazopanib relative to pemetrexed; the registry record does not clarify the direction convention. The estimate is therefore best quoted in the form reported.

Simple recalculations from posted summary statistics can serve as checks, but they are unadjusted and cannot recreate an analysis that uses participant-level event times or other model details. No discrepancy is flagged between the reported result and a recalculated result. The sponsor's own analysis takes precedence if a discrepancy is present; no such discrepancy is indicated for this result. The posted medians should not be used to derive a replacement hazard ratio.

The reported confidence interval is broad, and the registry does not provide the details needed to independently verify the analysis method, censoring implementation or model assumptions. The result should therefore be understood as the sponsor-reported estimate with its stated uncertainty, rather than as a fully reproducible analysis based on the registry record alone.

15. Multiplicity, Missing Data and Analysis Populations

The record lists 15 outcome measures and 14 posted secondary outcomes. The primary outcome is PFS; secondary outcomes include overall survival, best overall response, adverse events, treatment exposure and dosing measures. A single posted analysis is recorded for the primary comparison. The registry does not report whether the trial's statistical plan controlled for multiple comparisons across outcomes, whether any outcome hierarchy was prespecified, or how secondary analyses were intended to affect interpretation.

Multiplicity arises when several outcomes or comparisons are examined: even if each test uses a conventional significance threshold, the chance of at least one chance finding across a set of tests can increase. In general, multiplicity procedures can adjust thresholds or define a prespecified hierarchy. No such procedure is reported in the registry record, and the p-value for the primary comparison is not reported. It is not possible to infer an adjustment strategy from the list of outcomes alone.

The primary endpoint population is described as intent-to-treat, including all randomized participants whether or not study treatment was administered. The reported arm counts for the primary outcome are 9 and 11. The registry does not report whether all randomized participants contributed an evaluable PFS time, how missing assessments were handled beyond the stated censoring rules, or the number and reasons for censoring. A detailed account of missing outcome data is not reported in the registry record.

The record describes censoring at the last adequate assessment in specified circumstances, including initiation of subsequent anti-cancer therapy before progression or death. This rule explains the intended endpoint handling, but it does not describe the amount of censoring or whether censoring differed between arms. The randomization method, analysis software, detailed model specification, interim analyses and full statistical analysis plan are not reported in the registry record. Without those details, the posted result can be described, but several aspects of its analysis cannot be independently assessed.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: respiratory failure (11.1%), chest pain (11.1%), pneumonia (11.1%), dyspnoea (10.0%), pneumothorax traumatic (10.0%).

The registry reports serious adverse events among 2 participants in the pemetrexed group, with 9 at risk, and among 2 participants in the pazopanib group, with 10 at risk. These are counts of participants affected, not counts of all serious-event episodes. The record does not report deaths for these safety groups.

Listed serious-event terms include respiratory failure, chest pain and pneumonia, each reported as 11.1%; dyspnoea is reported as 10.0%, as is traumatic pneumothorax. These are the registry's listed percentages. The record does not provide the term-specific denominators or event counts in the registry record used for this description, so the percentages should not be converted into counts or treated as a complete account of safety.

In general, adverse-event comparisons in small groups are descriptive and can be imprecise. The registry also lists non-serious on-therapy adverse events and other safety outcomes, but those detailed results are not included in the posted serious-event figures. The registry record does not report enough information to determine comparative safety from these counts alone.

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

The posted result is an estimate of PFS in a randomized comparison of maintenance pazopanib and pemetrexed among non-progressing participants with metastatic non-squamous NSCLC after induction treatment. The registry reports median PFS of 12.7 weeks for pazopanib and 10.3 weeks for pemetrexed. The comparative hazard ratio is 0.92, with a 90% confidence interval from 0.32 to 2.65. The wide interval leaves substantial uncertainty about the relative hazard, and the registry does not report a p-value.

These data do not establish a precise comparative effect. The medians describe group-level event-time summaries and do not, on their own, show that the difference between treatments is attributable to treatment rather than sampling variation. The hazard ratio is the comparative estimate, but its interval spans values on both sides of 1. In addition, the registry does not make the comparison's numerator and denominator convention fully clear, and it does not report the analysis method.

The registry lists enrollment of 20 and a terminated status, although the brief summary described approximately 200 planned participants. The reason for termination is not reported in the registry record. The recorded enrollment and broad interval are important context, but they do not by themselves reveal why the trial ended or how the result would have differed under another design.

The findings apply to the recorded trial population and endpoint definition, not automatically to other patient groups, treatment settings or outcomes. PFS is not the same outcome as overall survival, response, tolerability or quality of life. The registry lists these as secondary outcomes but the figures described in the primary result do not answer those questions. The result should be read as a limited, uncertain estimate from the posted trial record, not as medical advice or a definitive statement about treatment effects.

18. Why This Trial Matters Statistically

This trial is statistically informative as an example of why a treatment comparison needs both an effect estimate and an account of its precision. The reported hazard ratio of 0.92 is accompanied by a 90% confidence interval from 0.32 to 2.65. The interval, rather than the point estimate alone, makes clear that the comparative estimate is uncertain. A point estimate near 1 should not be mistaken for proof that the treatments have identical effects.

The record also illustrates the distinction between an endpoint summary and a comparative analysis. Median PFS is reported separately for each arm, whereas the hazard ratio summarizes a relative time-to-event comparison. The medians do not substitute for the hazard ratio, and neither measure is fully interpretable without attention to censoring, follow-up and analysis assumptions. The registry states its censoring rules but provides limited detail about the statistical method and missing observations.

Trial status and enrollment matter to the context in which an estimate is read. The record reports enrollment of 20 and termination, while the brief summary describes approximately 200 participants as the intended enrollment. Since the reason for termination is not reported, it would be inappropriate to attribute a cause or infer what the result would have been with greater enrollment. The broad interval itself is direct evidence that the estimate is not precise.

Finally, registry fields need to be read carefully. The trial title and summary describe non-squamous NSCLC, but the condition listing says Lung Cancer, Small Cell. Noting such inconsistencies, alongside the incomplete reporting of analysis details, is part of responsible statistical interpretation. The record allows a clear description of the posted outcome but leaves important questions about the original design and detailed analysis unanswered.

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 has an internal population-label inconsistency: the title and brief summary describe metastatic non-squamous NSCLC, while the condition listing says Lung Cancer, Small Cell. The trial description supports the former wording, but the mismatch should be kept in mind when interpreting registry metadata.

The record lists enrollment of 20 and terminated status, whereas the brief summary describes approximately 200 planned participants. The reason for termination, whether the planned enrollment changed, and any consequences for the analysis are not reported in the registry record. It would be speculative to attribute the termination to recruitment, safety, efficacy or another cause.

Several details needed to assess the PFS analysis are not reported: the hazard-ratio method, the comparison's direction convention, the p-value, the randomization procedures, baseline prognostic information, the amount and pattern of censoring, and any assessment of model assumptions. The registry provides censoring rules but not the number of censored observations or a detailed description of missing assessments. The upper confidence limits for the arm-specific medians are also not reported.

The trial was open-label and the primary endpoint was investigator-assessed. The registry does not say whether progression assessments were independently reviewed or masked. Although the endpoint definition refers to RECIST version 1.1, additional details about assessment schedules and review procedures are not reported. These omissions limit independent evaluation of potential assessment and analysis issues.

The number of outcomes and secondary analyses raises questions about multiplicity, but the registry does not report a prespecified adjustment strategy or hierarchy. The original sample-size justification and design assumptions are also not reported in the registry record. These limitations call for cautious interpretation of the posted estimate and do not support conclusions beyond the outcomes and population described.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate Pazopanib in Comparison to Pemetrexed in Maintenance Setting in Non-progressing Subjects With Metastatic Stage IVA and IVB Non-squamous Non-small Cell Lung Cancer (NSCLC) Population — NCT01313663.
• Registry record last updated 2014-10-27; status last verified 2013-10.
• Results first posted 2013-04-29.
• 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

NCT01313663 was a randomized, open-label, parallel-group Phase II trial comparing maintenance pazopanib with pemetrexed after induction therapy in non-progressing participants described as having metastatic Stage IVA or IVB non-squamous NSCLC. The primary outcome was PFS from randomization to documented investigator-assessed progression or death.

The registry reports median PFS of 10.3 weeks for pemetrexed and 12.7 weeks for pazopanib. The reported hazard ratio is 0.92, with a 90% confidence interval from 0.32 to 2.65; a p-value is not reported. The record lists enrollment of 20 and status as terminated, while the brief summary describes approximately 200 intended participants. The reason for termination is not reported in the registry record.

The broad confidence interval indicates substantial uncertainty in the comparative estimate. The registry does not report several details needed for a full assessment of the analysis, including its method and the comparison's direction convention. The condition listing also conflicts with the non-squamous NSCLC description in the title and summary.