Standardization of Surgery on the Pancreatic Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Standardization of Surgery on the Pancreatic Cancer (NCT00679913), a randomized phase 3 study. The primary outcome is Survival.

2. Clinical Question

3. Trial Design

This was a phase 3, randomized, parallel-group study comparing standard with extended pancreatoduodenectomy. The registry lists 244 participants as enrolled and marks the study completed. The listed start was 2005-11, and primary completion was 2012-12. Seoul National University Hospital is identified as the sponsor.

The registry describes the study as single-masked, but does not state who was masked or how masking was implemented. The allocation method is identified as randomized; further details about sequence generation, concealment, stratification, or the randomization process are not reported in the registry record. The planned allocation ratio is also not reported, even though the planning illustration for sample size assumes equal allocation.

For the primary outcome, the posted arm counts are 83 participants in the standard group and 86 in the extended group. The secondary morbidity outcome has the same group denominators. These outcome denominators are smaller than the registry’s overall enrollment count. The reasons for this difference, including whether it reflects eligibility, follow-up, missing outcome information, or another factor, are not reported in the registry record. It should not be assumed from these figures alone that every enrolled participant was randomized or included in the reported outcome counts.

The study summary states that the trial was designed to test whether more extensive nodal and soft-tissue clearance could improve survival without an increase in morbidity and mortality. The registry lists morbidity as a secondary outcome. Further protocol details, including the original sample-size justification and the planned statistical analysis, are not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: After enrollment, some patients were excluded due to (1) unresectable condition or metastasis found during surgery; (2) surgical rule violation; (3) inadequate case report form; or (4) pathologic diagnosis other than conventional ductal adenocarcinoma.

Overall Study

Reasons for not completing this period:

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 Survival, measured over 2 year after surgery.

The primary endpoint is the number of participants surviving two years after surgery. The registry describes it as a comparison of two-year overall survival between the standard and extended procedure groups. In the posted results, 37 of 83 participants in the standard group and 31 of 86 in the extended group are recorded as surviving at that time point.

A fixed-time survival outcome provides a straightforward comparison of the proportion of each group alive at a defined follow-up point. It can communicate how many participants reached that point, but a count at one time does not describe when deaths occurred or how survival changed throughout follow-up. The registry does not post a time-to-event analysis, information about censoring, or a survival curve, so the posted endpoint should be interpreted as the recorded two-year status rather than as a full account of survival over time.

Survival is relevant to the study’s stated hypothesis about whether extended surgery could improve survival. The study summary also names morbidity and mortality as considerations in evaluating the more extensive procedure. The registry does not report a separate primary mortality analysis beyond the two-year survival outcome, and it does not describe how the primary outcome was ascertained or how missing status was handled. Those details are not reported in the registry record.

7. Results Posted in the Registry

• Extended Pancreatoduodenectomy: 36.0% (31/86) versus 44.6% (37/83) on control; risk difference -8.5 percentage points (95% CI -23.3 to 6.2), risk ratio 0.81 (0.56–1.17); unadjusted two-proportion z-test p = 0.258.

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.

Morbidity

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Extended Pancreatoduodenectomy: 43.0% (37/86) versus 32.5% (27/83) on control; risk difference 10.5 percentage points (95% CI -4.0 to 25.0), risk ratio 1.32 (0.89–1.96); unadjusted two-proportion z-test p = 0.160.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Extended Pancreatoduodenectomy and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists one secondary outcome, morbidity within two years after surgery. Its posted counts show morbidity in both treatment groups, with a higher observed count in the extended group. The unadjusted comparison is uncertain, as its interval includes zero. This secondary result addresses a different part of the study’s stated question from the primary survival outcome: it describes morbidity rather than survival at two years.

Across the primary and secondary outcomes, the posted estimates point in different directions for their respective event definitions: the observed survival proportion is lower in the extended group, while the observed morbidity proportion is higher in that group. Both comparisons have intervals that include zero. These results can be considered together as a descriptive pattern, but they do not establish a single overall treatment effect, and the morbidity result does not independently confirm or overturn the primary comparison.

Secondary outcomes generally carry less evidential weight than a prespecified primary outcome because they may be one of several outcomes examined and may not be the main basis for the study’s design. When multiple outcomes or comparisons are tested, the chance of at least one apparently notable result can increase unless the analysis accounts for multiplicity. The registry does not report a multiplicity strategy, a hierarchy of secondary outcomes, or reported secondary analyses. The posted secondary comparison is therefore best treated as exploratory context rather than as a definitive standalone conclusion.

The registry’s brief summary identifies morbidity and mortality as concerns in evaluating more extensive surgery, but only morbidity is listed as a secondary outcome. The registry does not report further secondary endpoints or a separate mortality comparison. The counts should not be extended to outcomes that are not listed or to a general conclusion about safety beyond the recorded measures and their uncertainty.

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

• Morbidity

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is binary at the participant level: each participant is counted as surviving or not recorded as surviving two years after surgery. The registry reports the number surviving and the group denominator. This makes it possible to describe each arm’s observed outcome proportion and to compare those proportions. The numbers 37 of 83 and 31 of 86 are the posted primary counts. The registry does not give a detailed outcome-ascertainment procedure or state how an unknown two-year status would be classified.

What does a risk difference express?

In general, a risk difference subtracts the event proportion in one group from the event proportion in the other. For the primary outcome, the comparison is extended relative to standard surgery, and the reported risk difference is -0.085. Because the outcome is survival, a negative value indicates a lower observed survival proportion in the extended group. The risk difference is expressed on the proportion scale, which helps describe the absolute contrast between groups. It is an unadjusted comparison of the posted counts, not an estimate adjusted for other participant characteristics.

The secondary morbidity outcome uses the same general type of comparison, but the event is morbidity rather than survival. Its risk difference is 0.105 for extended relative to standard surgery. The direction of a risk difference depends on how the outcome is defined: a positive morbidity difference indicates a higher observed event proportion in the extended group, while a negative survival difference indicates a lower observed survival proportion in that group. The two outcomes should not be conflated, even though both are reported as participant counts.

What do the confidence intervals add?

In general, a confidence interval describes the range of effect estimates reasonably compatible with the observed data and the statistical procedure used. The primary 95% interval runs from -0.233 to 0.062. It includes zero, so the posted data do not tightly distinguish a difference from no difference. The secondary 95% interval runs from -0.040 to 0.250 and also includes zero. An interval is not a guarantee that the true value lies within its endpoints, and it does not describe the spread of individual participant outcomes. Its width reflects statistical uncertainty under the calculation’s assumptions.

How should the p-values be read?

The unadjusted p-value for the primary comparison is 0.258; for morbidity it is 0.16. In general, a p-value measures how often results at least as incompatible with a specified null model would arise if that model and the test assumptions held. It is not the probability that the null hypothesis is true, nor does it quantify the probability that a result will replicate. A p-value does not replace the effect estimate or its confidence interval: the estimate conveys direction and scale, while the interval conveys uncertainty about those quantities.

The registry lists no reported analyses. The arm comparisons are unadjusted recalculations from the posted counts, rather than a reported sponsor analysis. The statistical test, model, covariate adjustments, handling of protocol deviations, analysis population, and approach to missing data are not reported in the registry record. The record also does not state whether the primary analysis was based on intention to treat, a per-protocol population, or another defined group.

Risk ratios and odds ratios are alternative ways to express a binary-outcome comparison. In general, a risk ratio compares the outcome proportions between groups, while an odds ratio compares their odds. They may express the same data on different scales and should not be treated as interchangeable. The posted primary risk ratio is 0.809, with interval 0.559 to 1.17; the odds ratio is 0.701, with interval 0.378 to 1.299. For morbidity, the risk ratio is 1.323, with interval 0.892 to 1.961; the odds ratio is 1.566, with interval 0.837 to 2.932. These are unadjusted measures derived from the posted counts and do not resolve the uncertainty indicated by their intervals.

In general, randomization is intended to make treatment groups comparable on average, reducing systematic differences at assignment. It does not guarantee that the groups are identical in a particular trial, and interpretation still depends on how randomization was implemented and how participants were followed and analyzed. Those procedural details are not reported in the registry record. Likewise, the registry’s single-masking label does not identify the masked party, so the possible influence of knowledge of treatment assignment cannot be assessed from that label alone.

11. Confidence Intervals and Precision

The primary estimate is a risk difference of -0.085, with a 95% confidence interval from -0.233 to 0.062. The interval crosses zero and extends to values on either side of it. This means the observed counts are not precise enough to pin down whether the underlying two-year survival proportion is lower or higher in the extended group, or whether the difference is close to zero. The point estimate gives the direction of the observed contrast, but the interval shows substantial uncertainty around its magnitude.

In general, a confidence interval’s endpoints represent uncertainty in an estimated group contrast, not a range containing most individual outcomes. The interval is calculated under statistical assumptions, and its interpretation depends on the analysis method. The registry does not report the original statistical method or a sponsor analysis, so the recalculated interval should be read as an unadjusted summary based on the posted counts.

The morbidity estimate also has an interval that includes zero: 0.105, with a 95% confidence interval from -0.040 to 0.250. This leaves uncertainty about both the size and direction of the underlying difference. Neither interval establishes equivalence or proves that there is no effect. Evidence of equivalence would require a defined equivalence margin and an appropriate analysis; these are not reported in the registry record.

Precision is affected by the amount of outcome information and by the variability of the observed outcomes. The registry’s outcome denominators are 83 and 86 participants, and it does not report details about missing status, follow-up completeness, or analysis populations. Those omissions limit assessment of whether the posted counts represent all randomized participants and how those factors may affect precision.

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 illustrations are expressed as participants per arm for the binary outcomes in this trial. They do not represent total events. 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 set at the stated proportion of the observed effect. These are planning assumptions for the illustration, not the original trial design assumptions; the trial’s own sample-size justification is not reported in the registry record.

The 80% power column gives the estimated number of participants per arm needed, under those assumptions, for an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column gives the corresponding per-arm planning number for a 90% chance. In general, choosing a higher chance of detecting an effect requires more participants, all else being equal. The table’s rows consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller target effects need more participants because they are harder to distinguish from random variation.

For example, for planning for 100% of the observed primary effect, the illustration gives 518 participants per arm for 80% power and 693 participants per arm for 90% power. For planning for 50% of the observed primary effect, it gives 2,107 participants per arm for 80% power and 2,820 participants per arm for 90% power. These figures are planning illustrations based on the observed effect, not the registry’s original design calculation. They are not observed power: observed power is not an appropriate substitute for interpreting the estimates and their confidence intervals.

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

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.446; treatment proportion p₁ = 0.360; average p̄ = 0.403.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6937 = 1.3596.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6911 = 0.5816.
4. n = (1.3596 + 0.5816)² / (-0.0853)² = 517.7, rounded up to 518 participants per arm 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
p0 <- 0.44578313253012                          # control proportion
p1 <- 0.36046511627907                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(679913)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 518 (80% power), 693 (90% power)
 75% of observed effect: n per arm = 929 (80% power), 1,243 (90% power)
 50% of observed effect: n per arm = 2,107 (80% power), 2,820 (90% power)
power.prop.test: 517.7 
Simulated power at n = 518 per arm: 0.801 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a statistical analysis plan, the original sample-size justification, or any reported analyses. In particular, it does not state which statistical test or model was planned for the primary outcome, whether the analysis was adjusted for baseline characteristics, or how the treatment effect would be estimated in the original analysis. The comparison statistics discussed for the posted counts are unadjusted recalculations, not a substitute for an unreported sponsor analysis.

The record does not describe a multiplicity plan. It lists survival as the primary outcome and morbidity as a secondary outcome, but does not report whether hypotheses were ordered, whether a family of outcomes was defined, or whether any adjustment to account for multiple testing was planned. In general, examining several outcomes can increase the chance of obtaining a seemingly unusual result by chance. Without the planned approach, the extent to which the reported secondary comparison was intended to support a confirmatory claim cannot be assessed.

Missing-data procedures and analysis populations are also not reported in the registry record. The primary and morbidity outcomes each list 83 participants in the standard group and 86 in the extended group, while total enrollment is listed as 244. The record does not explain the relationship between enrollment and these outcome denominators, identify any participants without outcome data, or say how such participants were handled. It does not state whether the analyses followed an intention-to-treat principle or used another population definition.

Further information not reported in the registry record includes the allocation-concealment process, the identity of the masked party, the follow-up completeness, protocol deviations, and whether interim analyses or stopping rules were used. These omissions do not show that such procedures were absent; they mean that their presence and details cannot be established from the registry record. They limit assessment of the potential for bias and of how closely the posted comparisons reflect the trial’s planned analyses.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry’s serious adverse event figures list 0 affected participants in the standard group, with 83 at risk, and 0 in the extended group, with 86 at risk. It also lists 0 participants affected for the other adverse-event category in each group, using the same group denominators. No serious adverse event terms are listed.

These figures describe what is recorded in the registry’s adverse-event fields. They should not be read as evidence that no adverse events of any kind occurred, or that the procedures have no risks. The registry does not describe the event definitions, collection and reporting procedures, follow-up completeness for these fields, or how deaths were classified in this adverse-event listing. Death counts are not reported in these fields. The primary survival outcome is a separate result and should not be conflated with the serious adverse-event figures.

As with the efficacy outcomes, the adverse-event counts do not provide a complete account of safety without information about ascertainment and reporting. They are limited to the categories and denominators recorded in the registry.

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

The posted results describe a randomized comparison of standard and extended pancreatoduodenectomy, with two-year survival as the primary outcome and morbidity as a secondary outcome. They show the observed numbers of participants recorded as surviving and as experiencing morbidity in each group. The unadjusted primary comparison has a negative risk difference for extended relative to standard surgery, while the secondary morbidity comparison has a positive risk difference. Both confidence intervals include zero, so neither comparison precisely determines the underlying difference.

These results do not establish that the procedures are equivalent. A confidence interval that includes zero does not prove that the true effect is zero; it indicates that the data are compatible with zero as well as with a range of other values. Nor do the results establish that either procedure is superior. The estimates are uncertain, and the registry does not report the planned statistical analysis, analysis population, or missing-data approach needed to fully assess the original inferential conclusions.

The results also do not describe survival throughout follow-up. The primary endpoint is a count of participants alive two years after surgery, and the registry does not post a time-to-event analysis or report when deaths occurred. The posted counts cannot be extended to long-term survival, recurrence, quality of life, or other outcomes not listed in the registry record.

The study summary states an aim of testing whether more extensive nodal and soft-tissue clearance could improve survival without increasing morbidity and mortality. The posted outcomes provide some information relevant to that aim, but the available record does not provide a complete account of the study’s protocol, surgical details, outcome ascertainment, analysis methods, or participant flow. The interpretation should therefore remain limited to the outcomes and uncertainty represented by the posted counts, rather than implying a broader clinical conclusion.

17. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison should be read through both its estimated effect and its uncertainty. A directionally negative primary risk difference does not by itself establish a meaningful or reliable difference. Its confidence interval includes zero and spans a range of possible values, making precision central to interpretation. The p-value alone cannot convey that range or explain the size of the observed contrast.

It also demonstrates the distinction between a fixed-time binary survival outcome and a time-to-event analysis. Counting participants alive at two years gives a clear endpoint at that time point, but does not show the timing of deaths or how the groups’ survival experiences evolved. The registry record does not provide a time-to-event analysis, so claims about survival patterns across the whole follow-up period would go beyond the posted outcome.

The secondary morbidity comparison illustrates why multiple outcomes require context. It can add descriptive information relevant to the study’s stated safety question, but its uncertainty and secondary status mean it should not be treated as a separate definitive answer. The registry does not report how multiplicity was handled, which further limits confirmatory interpretation.

Finally, the contrast between total enrollment and outcome denominators highlights the importance of participant flow and analysis-population reporting. Without an explanation of which participants contributed to each outcome and how missing information was handled, readers cannot fully assess the representativeness of the posted counts. The trial is therefore a useful example of how statistical interpretation depends not only on numerical estimates, but also on transparent reporting of design and analysis details.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific reporting gaps constrain interpretation. Although the study is marked randomized and single-masked, the registry does not describe sequence generation, allocation concealment, the masked party, or the implementation of masking. The surgical procedures are named, but further details about the extent of the operations and consistency of technique are not reported in the registry record.

The registry lists 244 participants enrolled, while the primary and secondary outcome denominators are 83 in the standard group and 86 in the extended group. The record does not explain this difference or provide a participant-flow account. It also does not identify an analysis population, report missing outcome data, or describe how incomplete follow-up was handled. This limits assessment of selection and attrition concerns.

The primary outcome is recorded as survival status two years after surgery, with no posted time-to-event analysis, survival curve, or censoring information. The registry does not report how survival status was ascertained. Likewise, the morbidity listing does not define the events or explain their ascertainment. The adverse-event figures should be interpreted in light of these reporting limits, rather than as a complete inventory of clinical events.

No sponsor-reported statistical analyses are listed. The registry does not report the original sample-size rationale, statistical test, adjustment strategy, multiplicity approach, interim analyses, stopping rules, or handling of protocol deviations. These details are not reported in the registry record and cannot be inferred from the posted counts or the planning illustrations. The available estimates therefore support descriptive and unadjusted interpretation, while the original confirmatory analysis and its assumptions remain unclear.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Standardization of Surgery on the Pancreatic Cancer — NCT00679913.
• Registry record last updated 2014-06-06; status last verified 2014-05.
• Results first posted 2014-06-06.
• 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.

22. Record Summary

The registry identifies NCT00679913 as a completed phase 3 randomized, parallel-group study sponsored by Seoul National University Hospital. It compared standard with extended pancreatoduodenectomy for pancreatic cancer and lists 244 participants enrolled. The primary outcome was the number of participants surviving two years after surgery; morbidity within two years was listed as a secondary outcome.

The primary counts were 37 of 83 in the standard group and 31 of 86 in the extended group. The secondary morbidity counts were 27 of 83 and 37 of 86, respectively. Unadjusted recalculated comparisons have confidence intervals that include zero. No reported analyses are listed in the registry record, and details about the original analysis plan, participant flow, and missing-data handling are not reported in the registry record.