Gemcitabine Hydrochloride With or Without Erlotinib Hydrochloride Followed by the Same Chemotherapy Regimen With or Without Radiation Therapy and Capecitabine or Fluorouracil in Treating Patients With Pancreatic Cancer That Has Been Removed by Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Gemcitabine Hydrochloride With or Without Erlotinib Hydrochloride Followed by the Same Chemotherapy Regimen With or Without Radiation Therapy and Capecitabine or Fluorouracil in Treating Patients With Pancreatic Cancer That Has Been Removed by Surgery (NCT01013649), a randomized phase 3 study. The primary outcome is Overall Survival (Percentage of Participants Alive) [Phase II].

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group Phase II-R/Phase III trial with no masking. The registry records enrollment of 546 and a completed status. Its official title describes a Phase II-R evaluation of erlotinib and a Phase III evaluation of chemoradiation as adjuvant treatment for resected head of pancreas adenocarcinoma.

The trial used two randomization steps. The primary endpoint belongs to step 1, which compared gemcitabine-based chemotherapy with gemcitabine plus erlotinib. Arm I is described as gemcitabine hydrochloride or allowable combination chemotherapy per standard of care; Arm II is gemcitabine hydrochloride with erlotinib hydrochloride. The registry describes treatment schedules, but the arm descriptions are incomplete in the record, so the full treatment protocols cannot be reconstructed from the posted information.

Step 2 addressed chemotherapy with or without subsequent chemoradiotherapy. The Phase III disease-free-survival comparison identifies Arm III as chemotherapy and Arm IV as chemotherapy followed by chemoradiotherapy. The description includes radiation therapy with capecitabine or fluorouracil as part of the broader trial question. These later-stage groups are not the groups used for the primary Phase II overall-survival result.

The primary analysis population is described as eligible participants randomized at step 1. The posted arm counts for the primary analysis are 163.0 and 159.0. The Phase III disease-free-survival analysis lists 174.0 and 180.0 in its two groups. These are endpoint-specific group counts; they should not be treated as a reconstruction of the full enrollment or combined across the two randomization steps.

The registry describes overall survival estimation by the Kaplan-Meier method, with survival measured from step 1 randomization until death from any cause or last known follow-up. The planned timing was after 200 deaths were reported. Maximum follow-up at the time of the Phase II analysis was 6.2 years. The registry record does not report the trial’s original sample-size justification or full statistical analysis plan.

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.

First Step

Reasons for not completing this period:

Second Step

Analysis population for the primary outcome, as defined in the registry: Eligible participants randomized at step 1.

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: Arm 1 and Arm 2: Phase II analysis population = eligible randomized Step 1 participants. Arm 3 and Arm 4: Phase III analysis population = randomized Step 2 participants. Phase II and phase III components are reported separately for clarity and so that totals can be seen for each component. Note that 354 participants were in both components.

6. Primary Endpoint

The registry lists the primary outcome as Overall Survival (Percentage of Participants Alive) [Phase II], measured over From step 1 randomization (gemcitabine vs. gemcitabine/erlotinib) to death or last follow-up. Maximum follow-up at the time of the phase II analysis was 6.2 years..

The primary endpoint is overall survival for the Phase II comparison. It measures the time from step 1 randomization until death from any cause. Participants without a recorded death by their last known follow-up are censored at that time, as described in the registry. The endpoint is summarized using Kaplan-Meier estimates, including median survival and a hazard ratio comparing the randomized groups.

Overall survival is a time-to-event endpoint: it uses both the timing of events and follow-up information, rather than treating every participant as though the outcome were known at a fixed date. The median is the time at which the estimated survival curve reaches the halfway point. It is a summary of the curve, not a measure of the survival time of every participant.

As a general statistical explanation, time-to-event methods are useful when participants have different lengths of follow-up or when some remain alive at the end of observation. Censoring allows their available follow-up to contribute without assuming that their later outcome is known. A hazard ratio compares the event rate over follow-up under a model; it is not the same quantity as a ratio of median survival times.

The registry identifies overall survival as the primary Phase II endpoint, while disease-free survival is reported as a secondary endpoint in both Phase II and Phase III. The primary endpoint therefore concerns death from any cause, whereas disease-free survival also counts a disease event. The registry record does not report additional detail about how disease events were defined or adjudicated.

7. Results Posted in the Registry

• Arm I (Gemcitabine Hydrochloride or Combination Chemotherapy) vs Arm II (Gemcitabine Hydrochloride, Erlotinib Hydrochloride): hazard ratio 1.04 (95% CI 0.79–1.38), reported p = 0.620 (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.

Disease-free Survival (Percentage of Participants Alive Without Disease) [Phase II]

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Arm I (Gemcitabine Hydrochloride or Combination Chemotherapy) vs Arm II (Gemcitabine Hydrochloride, Erlotinib Hydrochloride): hazard ratio 1.02 (95% CI 0.80–1.31) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Arm I (Gemcitabine Hydrochloride or Combination Chemotherapy) vs Arm II (Gemcitabine Hydrochloride, Erlotinib Hydrochloride) and the control. The interval gives the range of effects that remain compatible with the data.

Disease-free Survival (Percentage of Participants Alive Without Disease) [Phase III]

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Arm III (Chemotherapy) vs Arm IV (Chemotherapy, Chemoradiotherapy): hazard ratio 0.82 (95% CI 0.68–0.99) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for Arm III (Chemotherapy) vs Arm IV (Chemotherapy, Chemoradiotherapy) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary time-to-event results do not form a single uniform pattern. The Phase II disease-free-survival comparison has a hazard-ratio estimate close to 1, while the Phase III disease-free-survival estimate is below 1 and its reported 90% confidence interval does not include 1. These comparisons differ in randomization step and interval level, so their apparent contrast should not be read as a direct inconsistency or combined into one effect.

The secondary endpoints address questions beyond the primary Phase II overall-survival comparison. Disease-free survival includes a disease event as well as death, whereas overall survival concerns death from any cause. The registry also lists adverse-event outcomes and overall survival by baseline fatigue group. The secondary endpoint listing does not establish that every analysis was prespecified or confirmatory; those details are not reported in the registry record.

Secondary findings generally carry less weight when many outcomes or subgroup comparisons are examined without a reported adjustment for multiplicity. As a general statistical point, repeated testing increases the chance that at least one result will appear notable by chance. The registry record does not report a multiplicity adjustment, a hierarchy for testing the secondary endpoints, or the plan for interpreting them jointly. This uncertainty supports treating the secondary results as contextual rather than as replacements for the primary result.

The secondary results can help describe the broader pattern of outcomes, but they do not change what the primary analysis measures. In particular, a secondary estimate with an interval that excludes 1 does not establish that the primary overall-survival result also differs between groups. The endpoints, comparisons, and uncertainty levels need to remain distinct when interpreting the posted findings.

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

• Disease-free Survival (Percentage of Participants Alive Without Disease) [Phase II]
• Disease-free Survival (Percentage of Participants Alive Without Disease) [Phase III]
• Number of Participants by Highest Grade Adverse Event Reported (Phase II)
• Number of Participants by Highest Grade Adverse Event Reported (Phase III)
• Overall Survival (Percentage of Participants Alive) by Baseline Fatigue Group

9. Statistical Interpretation

10. Statistical Methods Explained

What does the Kaplan-Meier method summarize?

The registry says overall survival was estimated with the Kaplan-Meier method. In general, Kaplan-Meier estimation uses the observed event times and the available follow-up to estimate the probability of remaining event-free over time. Participants who have not had the event by their last known follow-up are censored. Their observed follow-up contributes information up to that point, while their later event status remains unknown. The registry defines the primary event as death from any cause and identifies last known follow-up as the censoring time.

The estimated curve can be summarized by its median, which is the time at which the curve reaches the halfway point. The reported median is therefore a property of the estimated survival curve, not an average and not a statement that every participant experienced that survival time. A median may be imprecise when the curve is estimated from limited information around that point. The confidence interval gives information about uncertainty in the estimate.

What does a hazard ratio represent?

In general, a hazard ratio compares the event hazards between groups over follow-up. A value of 1 represents equal hazards under the specified group coding. Values above or below 1 indicate a relative difference according to that coding, but the direction depends on which group is in the numerator and how the comparison is defined. The registry reports the comparison order and the hazard-ratio estimate, but does not describe further coding details.

A hazard ratio is not a ratio of median survival times and does not directly tell how many months an individual gains or loses. It also compresses information across follow-up into a single relative measure. Its interpretation can be less straightforward if the relative hazards change over time; the registry record does not report whether that assumption was assessed. This is a general limitation of interpreting a single hazard ratio, not evidence that the assumption failed in this trial.

How should the log-rank result be read?

The primary comparison used a log-rank method, according to the registry. In general, the log-rank test compares the observed timing of events between groups over the follow-up period. The reported p-value of 0.62 is evidence from that test, interpreted under its null hypothesis and assumptions. It is not a probability that the groups are identical, and it does not measure the clinical importance of a difference.

The p-value and confidence interval answer related but distinct questions. The test result assesses compatibility with a null hypothesis; the interval presents a range of plausible values for the effect estimate under the interval procedure. Reading both helps avoid treating a single cutoff as a complete interpretation. In this result the hazard-ratio interval includes 1, consistent with the non-small p-value, but neither summary proves there is no difference.

What does censoring mean in this analysis?

As a general explanation, censoring occurs when a participant’s exact event time is not observed during the available follow-up. For this endpoint, the record states that survival time ends at death or last known follow-up. A censored participant is not counted as though death occurred at the censoring date; instead, the analysis uses information through that date. Valid interpretation depends on assumptions about the relationship between censoring and the chance of the event. The registry record does not report diagnostic checks or further handling details.

The record says the primary analysis was to occur after 200 deaths were reported. It does not state the observed number of deaths at the analysis, whether the planned event count was reached, or whether interim analyses were conducted. Those details should not be inferred from the timing of the analysis or from the posted results.

Why do the different trial stages matter?

The primary endpoint is the Phase II overall-survival comparison from step 1. The Phase II disease-free-survival result also uses step 1 groups, but measures a different event definition. The Phase III disease-free-survival result comes from step 2 and compares chemotherapy with chemotherapy followed by chemoradiotherapy. Different stages and endpoints answer different questions; their hazard ratios should not be combined or interpreted as though they came from one randomized comparison.

The Phase II overall-survival intervals are reported at 95%. The Phase III disease-free-survival interval is reported at 90%. Confidence intervals at different levels have different coverage properties, so their widths and whether they include the null value are not directly comparable without accounting for the level. The registry record does not report a multiplicity procedure for interpreting the several outcomes and comparisons.

Finally, randomization supports comparison of the groups as assigned, while the absence of masking means participants and trial personnel were not masked according to the registry. Masking is generally more consequential for outcomes involving subjective assessment than for an objective event such as death, but the record does not provide details about outcome adjudication. These design features belong in the context of the result rather than serving as automatic evidence for or against its validity.

11. Confidence Intervals and Precision

The primary overall-survival estimate is accompanied by a 95% confidence interval for each median and for the hazard ratio. In general, a confidence interval expresses uncertainty arising from sampling under the method used. It is not a guarantee that the true value lies within the particular interval, nor does it describe the range of individual participant outcomes.

The hazard-ratio interval, 0.79 to 1.38, includes 1. This means the posted data are compatible with no relative hazard difference as well as with values on either side of that null value. The interval is also wide enough to leave uncertainty about the magnitude of a possible difference. The point estimate, 1.04, should not be read in isolation from that uncertainty. The log-rank p-value is 0.62, but it does not turn the result into evidence of equivalence.

The median estimates also have uncertainty. Their 95% confidence intervals overlap, but overlap or non-overlap of separate intervals is not a substitute for the reported between-group test. The hazard ratio and log-rank result address the comparison across follow-up, while each median summarizes one group’s estimated survival curve.

For the Phase III secondary disease-free-survival comparison, the registry reports a 90% confidence interval, rather than the 95% intervals used for the primary endpoint and Phase II disease-free-survival result. A 90% interval is constructed at a different confidence level, so its apparent width and its relation to the null value should be interpreted at that level. The registry does not report a p-value for that secondary comparison.

Precision depends on the amount and timing of outcome information, follow-up, and the analysis method. Group counts alone do not determine precision in a time-to-event analysis. The registry record does not report enough detail to assess how those elements affected the posted estimates beyond the reported intervals and follow-up summaries.

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 the observed effect, not the trial’s original design calculations. For these time-to-event comparisons, the figures are total events, not participants per arm. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is the stated fraction of the observed effect.

The 80% power column describes the event total estimated to give an 80% chance of detecting the specified effect under those planning assumptions, if that effect and assumptions were true. The 90% power column makes the corresponding illustration for a 90% chance. These are prospective planning interpretations, not descriptions of what occurred in the trial. This is not observed power.

Rows for 100%, 75%, and 50% of the observed effect represent increasingly smaller target effects. Smaller effects are harder to detect reliably and therefore require more events under the same assumptions. The very large event totals in some rows illustrate that relationship; they should not be read as a claim about the number of participants needed for this specific trial, since these are event counts and the planning exercise is based on the observed result.

The record does not report the assumptions used for the trial’s own original sample-size design. The registry’s planned primary-analysis timing after 200 deaths is a separate statement and does not make these planning illustrations part of the original protocol. The event-based figures should therefore be read only as hypothetical planning quantities associated with the observed effect and stated assumptions.

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 = 1.040, so ln(HR) = 0.0392.
2. d = 4 × (1.96 + 0.842)² / [ln(1.040)]² = 20,409.7, rounded up to 20,410 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 <- 1.04                              # 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(1013649)
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 = 20,410 (80% power), 27,323 (90% power)
 75% of observed effect: events = 36,284 (80% power), 48,574 (90% power)
 50% of observed effect: events = 81,639 (80% power), 109,292 (90% power)
Simulated power with 20,410 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.

14. Analyses Reported by the Sponsor

The primary overall-survival comparison is reported using the sponsor’s log-rank analysis: hazard ratio 1.04, 95% confidence interval 0.79 to 1.38, and p-value 0.62. The registry also reports the Kaplan-Meier median estimates for the two groups. These are the sponsor-reported results to use when describing the trial’s primary analysis.

Any recalculation from posted summary statistics is unadjusted and cannot reproduce every feature of the sponsor’s analysis. The record does not flag a discrepancy between the sponsor result and a recalculation. Where a discrepancy is flagged, the sponsor’s own analysis takes precedence; in interpreting this record, the reported sponsor analysis remains the central inferential result.

The registry identifies the log-rank method but does not provide additional details about model specification, covariate adjustment, or diagnostics. A summary-statistic recalculation should not be treated as a replacement for the analysis performed by the sponsor or as a second independent trial result. The distinction is especially important for time-to-event data, where event timing, censoring, and analysis choices contribute to the estimate.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcome measures and analyses, including overall survival, disease-free survival, adverse-event outcomes, and overall survival by baseline fatigue group. Multiple endpoints and comparisons can raise multiplicity concerns: as a general statistical explanation, the more hypotheses examined, the greater the chance of at least one apparently notable result arising by chance if no adjustment or testing hierarchy is used.

The registry record does not report whether the investigators adjusted for multiplicity, designated a hierarchy among secondary endpoints, or treated any secondary comparison as confirmatory. It also does not report the full statistical analysis plan. Consequently, the secondary estimates should be interpreted with caution and should not be presented as though a family-wise testing strategy or other adjustment had been used.

The record describes the primary analysis population as eligible participants randomized at step 1. It does not provide a full accounting of exclusions after randomization, deviations from assigned treatment, or the exact population used for every secondary analysis. Those analysis-population details are not reported in the registry record.

For overall survival, the registry defines the endpoint as time to death from any cause or last known follow-up, with the latter treated as censored. This gives a basic description of censoring but does not provide the amount of missing follow-up, reasons for loss to follow-up, or sensitivity analyses for missing outcome information. The registry record does not report those details. They should not be inferred from the fact that Kaplan-Meier methods were used.

The primary analysis was planned after 200 deaths were reported, but the posted information does not establish whether interim analyses occurred or provide any stopping boundaries. The original sample-size justification, missing-data strategy, multiplicity strategy, and full prespecified analysis plan are not reported in the registry record. Their absence limits how fully the design and confirmatory status of the reported findings can be assessed.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: anemia (5.7%), fever (4.4%), abdominal pain (4.0%), neutrophil count decreased (3.8%), fatigue (3.2%).

The adverse-event information gives serious-event and other-event counts with the corresponding at-risk counts, by group. For Arm I, 38 of 160 were listed with serious adverse events; for Arm II, 45 of 158. Arm III lists 33 of 174, and Arm IV lists 31 of 180. These are descriptive counts as posted, not adjusted comparisons.

The registry also lists the most frequent serious adverse-event terms and their percentages: anemia, 5.7%; fever, 4.4%; abdominal pain, 4.0%; neutrophil count decreased, 3.8%; fatigue, 3.2%; and sepsis, 3.2%. The record gives no inferential comparison for these terms in the registry record reported.

Other adverse events were reported for 158 of 160 in Arm I, 157 of 158 in Arm II, 173 of 174 in Arm III, and 179 of 180 in Arm IV. Death counts are also listed by group, but the registry does not state in this summary whether each death was attributed to treatment or provide a safety comparison adjusted for follow-up. Differences in group counts alone do not establish differences in risk; interpretation would require consistent definitions, follow-up context, and an appropriate comparative analysis.

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

The primary result describes overall survival in the two step 1 randomized groups. The reported medians and hazard ratio summarize the observed data, while the confidence intervals show that estimates remain uncertain. The p-value does not provide evidence that the groups are identical, and the result does not establish equivalence. The appropriate reading is that the posted primary analysis did not demonstrate a clear difference under the reported test, while leaving a range of possible effects compatible with the data.

This is not a claim that either regimen works or does not work. A non-small p-value does not prove absence of an effect, and a point estimate near 1 does not rule out effects that the interval still allows. Conversely, a secondary estimate with a confidence interval excluding 1 does not establish a definitive clinical benefit or change the primary endpoint result.

The trial addresses adjuvant treatment after surgery in the population described in the registry, and its Phase II and Phase III comparisons concern different treatment strategies and endpoints. The findings should not be generalized beyond those populations and comparisons without additional evidence. The registry record does not report enough information to assess every aspect of eligibility, treatment delivery, adherence, or outcome ascertainment.

Statistical uncertainty is not the only consideration in interpreting trial results. Clinical importance depends on the magnitude and relevance of an effect, the outcome being measured, and the balance of benefits and harms. The posted registry information gives some outcome and adverse-event summaries, but it does not provide a complete comparative assessment of those considerations. The statistical results should therefore be understood as estimates with limitations, not as medical advice.

18. Why This Trial Matters Statistically

Statistically, this trial is notable for separating two randomized questions within a Phase II-R/Phase III program. The primary endpoint is the Phase II overall-survival comparison of gemcitabine-based chemotherapy with and without erlotinib. The Phase III disease-free-survival comparison evaluates chemotherapy with or without subsequent chemoradiotherapy. Keeping the comparisons distinct is essential: they differ in randomization step, treatment strategy, and endpoint.

The primary analysis illustrates why a point estimate and a significance test should not be used alone. The hazard-ratio estimate is near the null value, but its confidence interval spans values on both sides of 1. The p-value does not establish equivalence, and the interval communicates uncertainty about the possible effect size. This is a useful example of interpreting a time-to-event result through both estimation and testing.

The trial also highlights how follow-up and censoring shape survival analyses. Kaplan-Meier estimation and a log-rank comparison use event timing and follow-up rather than simply comparing the proportion of participants who experienced an event at one fixed time. The record specifies the primary endpoint’s time origin and censoring rule, while leaving other analysis details unreported.

Finally, the secondary findings demonstrate why outcome multiplicity and differences in confidence-interval levels matter. A secondary estimate cannot be read as a substitute for the primary result, and a finding from one trial stage should not be pooled informally with another stage. The registry record does not report a multiplicity strategy, so secondary results warrant cautious, context-specific interpretation.

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 posted primary result is specific to the Phase II step 1 comparison, while the broader trial also includes a separate Phase III comparison. The Phase III disease-free-survival result therefore does not provide a direct replication of the primary overall-survival analysis. The record’s arm descriptions are incomplete, limiting reconstruction of the full treatment regimens and their delivery.

Although the registry describes a planned analysis after 200 deaths and gives maximum follow-up for the Phase II analysis, it does not report whether that event target was reached, the observed event count for the primary analysis, or details of any interim monitoring. Original sample-size assumptions and the complete analysis plan are not reported. This limits assessment of how the posted analysis relates to the initial design.

The record does not report the missing-data strategy beyond describing censoring at last known follow-up for overall survival. It does not provide detailed information on loss to follow-up, outcome adjudication, treatment adherence, protocol deviations, or sensitivity analyses. It also does not describe a multiplicity adjustment for the several endpoints and comparisons.

The study was unmasked. This is part of the registry’s design description, but the record does not explain how outcome assessment was conducted or whether adjudicators were independent of treatment assignment. The primary endpoint is death from any cause, but the record does not provide enough detail to assess all possible sources of bias or the impact of design features on measurement.

Finally, adverse-event counts are presented descriptively, without a posted comparative analysis in the registry record reported. The available figures do not by themselves account for differences in follow-up or exposure. These limitations do not invalidate the reported results, but they constrain how much can be concluded from the registry summary alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Gemcitabine Hydrochloride With or Without Erlotinib Hydrochloride Followed by the Same Chemotherapy Regimen With or Without Radiation Therapy and Capecitabine or Fluorouracil in Treating Patients With Pancreatic Cancer That Has Been Removed by Surgery — NCT01013649.
• Registry record last updated 2026-02-23; status last verified 2026-02.
• Results first posted 2024-10-24.
• 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

NCT01013649 is a completed randomized, parallel-group, unmasked Phase II-R/Phase III trial sponsored by the National Cancer Institute. It enrolled 546 participants and examined adjuvant treatment strategies for resected pancreatic cancer, including gemcitabine with or without erlotinib and chemotherapy with or without subsequent chemoradiotherapy.

The primary posted endpoint is Phase II overall survival from step 1 randomization to death or last known follow-up. The sponsor-reported comparison used a log-rank method and gave a hazard ratio of 1.04, a 95% confidence interval of 0.79 to 1.38, and p = 0.62. Median overall survival was 29.9 months in Arm I and 28.1 months in Arm II.

The record also posts disease-free-survival results from Phase II and Phase III, along with adverse-event summaries. These secondary findings concern different endpoints or trial stages and should be interpreted separately from the primary result. Details of multiplicity handling, the full analysis plan, and several missing-data and analysis-population procedures are not reported in the registry record.