Aspirin in Preventing Recurrence of Cancer in Patients With HER2 Negative Stage II-III Breast Cancer After Chemotherapy, Surgery, and/or Radiation Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Aspirin in Preventing Recurrence of Cancer in Patients With HER2 Negative Stage II-III Breast Cancer After Chemotherapy, Surgery, and/or Radiation Therapy (NCT02927249), a randomized phase 3 study. The primary outcome is Median Invasive Disease-free Survival (iDFS).

2. Clinical Question

3. Trial Design

The ABC Trial, registered as NCT02927249, is described as a randomized, phase III, double-blind, placebo-controlled study with a parallel design. The sponsor is the Alliance for Clinical Trials in Oncology. The registry reports enrollment of 3021 and lists the status as terminated. The study start date is 2017-01-06, and the primary completion date is 2021-12-13. The reason for termination is not reported in the registry record.

Participants were assigned to aspirin or placebo. The regimen was once-daily treatment by mouth for five years, unless disease progression or unacceptable toxicity occurred. The registry does not report further details of the randomization process or how allocation was concealed. It also does not report the original sample-size justification, planned interim analyses, or the full statistical plan.

The primary outcome was median invasive disease-free survival, assessed over a timeframe of five years. The registry defines the endpoint as time from randomization to the first specified invasive disease event or death from any cause. The posted comparison is a hazard ratio for aspirin versus placebo as reported by the sponsor, alongside a 95% confidence interval. No p-value is posted for this analysis, and the group median values themselves are not reported in the registry record.

The trial also listed overall survival, distant disease-free survival, cardiovascular disease incidence, and toxicities among its secondary outcomes. The posted results include hazard-ratio comparisons for the survival outcomes and counts for cardiovascular disease. The registry lists outcomes for toxicities, but the posted safety information summarized for serious and other adverse events does not provide a complete account of all toxicity analyses.

4. Participant Flow and Accounting

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

Overall Study

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 Median Invasive Disease-free Survival (iDFS), measured over 5 years.

Invasive disease-free survival (iDFS) measures the time from randomization until the first qualifying event. In this trial, qualifying events included distant recurrence, locoregional recurrence, cancer in the same or opposite breast, a second primary invasive cancer outside the breast, or death from any cause. Participants were censored at the date they were last known to be alive and free from invasive breast cancer and second invasive primaries, according to the registry definition.

This composite endpoint captures several clinically important ways that invasive disease can recur or a new invasive cancer can arise, while also counting death from any cause. A time-to-event measure uses information about both whether an event occurred and when it occurred during follow-up. In general, this can make it useful when participants have different lengths of observation or when not everyone experiences an event before follow-up ends.

The registry names median iDFS as the primary outcome, with a 95% confidence interval, in months, over a five-year timeframe. However, it does not report the median value for either group. The posted primary comparison is a hazard ratio of 1.19 with a 95% confidence interval from 0.91 to 1.56. The interpretation therefore rests on the reported relative event-rate comparison and its uncertainty, not on a directly reported difference between median times.

7. Results Posted in the Registry

• Arm I (Aspirin) vs Arm II (Placebo): hazard ratio 1.19 (95% CI 0.91–1.56) (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.

Median Overall Survival (OS)

• 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 (Aspirin) vs Arm II (Placebo): hazard ratio 1.12 (95% CI 0.74–1.69) (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 (Aspirin) vs Arm II (Placebo) and the control. The interval gives the range of effects that remain compatible with the data.

Median Distant Disease Free Survival (DDFS)

• 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 (Aspirin) vs Arm II (Placebo): hazard ratio 1.12 (95% CI 0.83–1.52) (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 (Aspirin) vs Arm II (Placebo) and the control. The interval gives the range of effects that remain compatible with the data.

Incidence of Cardiovascular Disease (Including Cerebrovascular Events, Myocardial Infarction, or Coronary Artery Disease Requiring Stent Placement, Angioplasty, or Bypass Surgery)

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm I (Aspirin): 0.1% (2/1393) versus 0.1% (2/1415) on control; risk difference 0.0 percentage points (95% CI -0.3 to 0.3), risk ratio 1.02 (0.14–7.20); unadjusted two-proportion z-test p = 0.987.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm I (Aspirin) and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover overall survival, distant disease-free survival, cardiovascular disease incidence, and toxicities. The survival outcomes use hazard-ratio comparisons, while cardiovascular disease is reported as a count of participants. The posted point estimates for overall survival and distant disease-free survival are both 1.12, and their 95% confidence intervals include the no-difference value. The cardiovascular outcome has very few recorded events in both groups, which limits the precision of a comparison based on those counts.

At a broad level, the survival estimates do not provide a more decisive picture than the primary iDFS result: their uncertainty intervals also span the no-difference value. This is consistent with interpreting the primary result cautiously rather than treating the secondary outcomes as independent confirmation of a clear group difference. Secondary endpoints can offer context about other outcomes, but they do not replace the primary endpoint.

Multiple outcomes create multiple opportunities to observe an apparently notable result by chance. If many outcomes are examined without an appropriate adjustment or a clearly prespecified hierarchy, isolated findings carry less weight than a result from a single primary question. The registry does not report a multiplicity adjustment, a testing hierarchy, or details of how secondary analyses were prioritized. Thus, the secondary results should be treated as supportive or exploratory evidence, not as a basis for strong stand-alone claims.

The registry lists toxicity as a secondary outcome and posts some safety counts, but the information does not provide a complete set of toxicity comparisons or a detailed analysis plan. Differences in follow-up, event definitions, or analysis populations could affect interpretation; those details are not reported in the registry record. Taken together, the secondary outcomes add breadth to the record but do not remove the uncertainty in the primary estimate.

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

• Median Overall Survival (OS)
• Median Distant Disease Free Survival (DDFS)
• Incidence of Cardiovascular Disease (Including Cerebrovascular Events, Myocardial Infarction, or Coronary Artery Disease Requiring Stent Placement, Angioplasty, or Bypass Surgery)
• Incidence of Toxicities, Graded Using the National Cancer Institute's Common Terminology Criteria for Adverse Events Version 4.3

9. Statistical Interpretation

10. Statistical Methods Explained

Several statistical ideas help explain what the registry reports and what cannot be concluded from it. The outcome is time-to-event, the primary contrast is a hazard ratio, and uncertainty is expressed with a confidence interval. The registry does not name the statistical method used to estimate the sponsor’s hazard ratio, so the explanation below describes these concepts generally rather than asserting which specific model or procedures were used.

What does a time-to-event outcome represent?

A time-to-event outcome records the time from a defined starting point—in this trial, randomization—to the first event that meets the endpoint definition. Participants who have not experienced an event by the end of their observed follow-up may contribute information up to that point without being counted as having an event. This is called censoring. The iDFS definition states that censoring occurs at the last date the participant was known to be alive and free from the specified invasive cancers.

Time-to-event methods are designed to use the observed follow-up while accounting for participants whose event time is not observed during that period. Their validity depends on assumptions about censoring and the analysis method. The registry provides the endpoint’s censoring definition but does not report the analysis method or details about how censoring was handled in the sponsor’s analysis. Those details are not reported in the registry record.

What does a hazard ratio compare?

A hazard is an event rate at a particular point in follow-up among participants who have not yet experienced the event. A hazard ratio expresses the relative hazards for the two groups under the analysis. It is not a ratio of the numbers of participants who experienced an event, nor is it a direct comparison of median times. A hazard ratio can be read as a single relative summary, but that summary may require assumptions about how the group hazards relate over time. The registry does not report whether those assumptions were assessed or the exact method used.

The sponsor reports the primary hazard ratio as 1.19 for aspirin versus placebo. The 95% confidence interval runs from 0.91 to 1.56. A ratio at the no-difference point corresponds to equal hazards under this measure. Because the interval extends across that point, the estimate does not distinguish clearly between a lower, similar, or higher hazard in the aspirin group. The ratio does not state how many months participants lived without an event, and the registry does not report the group medians.

How should a confidence interval be read?

A confidence interval conveys uncertainty around an estimate. In general, a narrower interval indicates greater precision, while a wider interval indicates that a broader range of values is compatible with the data and analysis. A 95% confidence interval is a statistical procedure with a repeated-sampling interpretation; it is not a statement that there is a 95% probability that a particular fixed effect lies inside this specific interval.

For this primary comparison, the interval includes the no-difference value and spans effects in both directions. That means the reported estimate is not sufficiently precise to settle whether the groups differ on the hazard-ratio scale. It does not establish that there is no effect, and it does not prove that the groups are equivalent. The registry does not report a pre-specified equivalence margin or a non-inferiority margin.

Why are p-values and analysis details relevant?

The registry does not report a p-value for the primary analysis. A p-value, when reported, describes how compatible the observed data are with a specified null hypothesis under the analysis assumptions. It is not a measure of the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. Here, the interval gives a direct view of the uncertainty, while the absent p-value means that an exact reported significance test cannot be described.

The analysis method is listed as not reported in the registry record. The record also does not report whether the primary comparison was adjusted for covariates, the exact analysis population, how missing outcome information was handled, or whether the proportional-hazards assumption was evaluated. These omissions prevent a detailed assessment of the analysis beyond the posted estimate and interval.

What does over-dispersion mean in general?

For count or binary outcomes, a statistical model may assume a particular amount of variability. Over-dispersion means that the observed variability is greater than the model expects. If it is not accounted for when present, uncertainty can be understated and confidence intervals or tests can be misleading. This is a general consideration for some outcome analyses; the registry does not identify over-dispersion as an issue in this trial, and no claim about it can be made from the posted results.

The primary iDFS outcome is a time-to-event measure, so the relevant posted summary is a hazard ratio rather than a count-based risk difference. The secondary cardiovascular outcome is reported as counts of participants, but a full model description for that outcome is not reported. General statistical concepts should not be mistaken for evidence that a particular modeling choice or diagnostic procedure was used in this trial.

Why distinguish an estimate from a conclusion?

An estimate is the numerical summary calculated from the observed data. A conclusion also depends on uncertainty, the study design, analysis choices, and the question being asked. For this trial, the point estimate and its interval are both essential: the point estimate is 1.19, while the interval ranges from 0.91 to 1.56. Reading only the point estimate would ignore the substantial uncertainty expressed by the interval. Reading only whether the interval crosses the no-difference point would also omit the range of effects that remains compatible with the result.

The posted information supports a cautious description of the primary result as an imprecise comparison with an interval spanning the no-difference value. It does not support a claim of benefit, harm, or equivalence. Further methodological details that could refine interpretation are not reported in the registry record.

11. Confidence Intervals and Precision

The primary hazard-ratio estimate is 1.19, and its 95% confidence interval extends from 0.91 to 1.56. This interval communicates uncertainty around the estimated relative hazard. It includes the no-difference value and also includes values on either side, so the posted estimate is compatible with more than one interpretation of the direction of the group contrast.

The interval is not a direct range for the group median iDFS times. The registry reports median iDFS as an outcome but does not report the median values or their confidence intervals for either arm. It is therefore not possible to describe the difference in median time from the posted results.

Confidence intervals are useful because they show both the estimate and the precision with which it has been obtained. A point estimate can appear to favor one direction while the interval remains broad enough to include no difference and an effect in the opposite direction. In this case, the interval does not support a definitive conclusion that the groups differ, but it also does not establish that their outcomes are identical.

The registry does not report the primary p-value, analysis method, or details needed to assess the assumptions behind the interval. The confidence interval should therefore be interpreted as the sponsor’s posted uncertainty summary, not as a fully reproducible analysis. The reported interval is the central evidence about precision available for the primary comparison.

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 planning illustrations are calculations based on the observed effects, not the trial’s original design assumptions. For time-to-event outcomes, the figures are total events required; for the binary cardiovascular outcome, they are participants per arm. The planning assumptions listed are a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. The assumed effect to detect is a stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column describes a planning target under which a study would have an 80% chance of detecting the specified effect if that effect were real and the planning assumptions held. The 90% power column uses a higher probability of detection under the same kind of assumptions. These are design-planning quantities, not measures of the certainty of the posted result. This is not observed power.

For the primary time-to-event comparison, planning for 100% of the observed effect gives 1,038 total events for 80% power and 1,389 total events for 90% power. Planning for 75% of the observed effect gives 1,845 and 2,470 total events, respectively. Planning for 50% of the observed effect gives 4,151 and 5,556 total events, respectively. The pattern illustrates that a smaller effect is harder to detect and requires more information.

For the cardiovascular endpoint, the corresponding illustrations are participants per arm. For 100% of the observed effect, the figures are 44,814,426 participants per arm for 80% power and 59,993,801 for 90% power. These very large planning figures reflect the small observed difference used in the illustration; they should not be read as the trial’s actual enrollment target. More generally, smaller effects require more participants or events to detect with the same planned power.

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 = 1.190, so ln(HR) = 0.1740.
2. d = 4 × (1.96 + 0.842)² / [ln(1.190)]² = 1,037.5, rounded up to 1,038 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 <- 1.19                              # 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(2927249)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 1,038 (80% power), 1,389 (90% power)
 75% of observed effect: events = 1,845 (80% power), 2,470 (90% power)
 50% of observed effect: events = 4,151 (80% power), 5,556 (90% power)
Simulated power with 1,038 events: 0.802 (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 sponsor’s reported primary analysis is a hazard ratio of 1.19 with a 95% confidence interval from 0.91 to 1.56. No p-value or named analysis method is reported in the registry record. The statistical interpretation should therefore start from this reported estimate and interval rather than attempt to infer a different sponsor result from incomplete information.

Any simple recalculation from posted summary statistics is unadjusted and may not reproduce the sponsor’s analysis. Such a recalculation cannot account for analysis details that are not reported, including any adjustments or modeling choices. In general, the sponsor’s own analysis takes precedence when comparing it with a recalculation. In this record, the discrepancy flag is false, so no discrepancy between the posted sponsor estimate and the recalculated result is indicated.

The hazard-ratio interval remains the key evidence about the precision of the sponsor’s comparison. It spans the no-difference value, and the registry does not post a p-value that could add a reported test result. The record does not contain enough methodological detail to independently reproduce the sponsor’s analysis or to assess all of its assumptions.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes or statistical comparisons are examined. Even if each test is conducted using a conventional threshold, carrying out many tests can increase the chance that at least one result appears notable by chance. Researchers may address this through pre-specified testing sequences, adjusted thresholds, or other procedures. The registry record does not report whether any multiplicity adjustment was planned or used, or whether a hierarchy governed testing across the primary and secondary outcomes.

This matters especially when interpreting secondary outcomes. Their results can be informative, but without a reported multiplicity strategy it is difficult to know whether a nominally notable secondary finding would retain the same evidential weight after accounting for the number of questions examined. No p-values are reported for the posted survival comparisons, and the registry does not provide a complete set of tests across all outcomes.

The record also does not report the analysis populations for the primary or secondary outcomes. It does not state whether the primary analysis followed an intention-to-treat principle, how participants with missing outcome information were handled, or how deviations from assigned treatment were addressed. The iDFS definition describes when censoring occurs, but the registry does not report additional details about missing data or censoring assumptions.

Other analysis details are also absent, including the named model used for the hazard ratios, whether covariate adjustment was applied, and whether model assumptions were evaluated. The original sample-size justification, any interim analysis plan, and the reason for termination are not reported in the registry record. These gaps do not alter the posted numerical estimate, but they limit independent evaluation of how it was obtained and how robustly the planned analysis was followed.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: infections and infestations - oth spec (0.5%), bruising (0.4%), dyspepsia (0.4%), thromboembolic event (0.4%), surgical and medical proced - oth spec (0.4%).

The registry reports serious adverse events in 99 of 1393 participants at risk in the aspirin arm and 91 of 1415 participants at risk in the placebo arm. It also reports other adverse events in 621 of 1393 and 726 of 1415, respectively. The record lists 48 deaths in the aspirin group and 43 in the placebo group. These are descriptive counts; no statistical comparison or confidence interval for these safety counts is reported in the information summarized.

The most frequently listed serious adverse event terms include infections and infestations, other specified, at 0.5%; bruising at 0.4%; dyspepsia at 0.4%; thromboembolic event at 0.4%; surgical and medical procedures, other specified, at 0.4%; and skin infection at 0.4%. The registry’s listed terms and percentages describe recorded events, but do not by themselves establish that an event was caused by the assigned treatment.

Safety counts should be read with their denominators and event definitions in mind. The registry does not report further details about exposure-adjusted rates, timing, severity breakdowns beyond the serious-event classification, or the statistical method for comparing groups. The counts alone do not establish a causal difference between the arms.

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

The primary result describes the comparison of time to the first event in a composite invasive disease-free survival endpoint. The reported hazard ratio is 1.19, with a 95% confidence interval from 0.91 to 1.56. This interval includes the no-difference value, so the result does not establish a clear difference between aspirin and placebo on the reported hazard-ratio scale.

That conclusion is narrower than saying the treatments are equivalent. An interval that includes no difference can arise when the data are compatible with a range of effects, and this interval includes values both below and above the no-difference point. The registry does not report an equivalence or non-inferiority design, a corresponding margin, or group median iDFS values. It therefore does not support a claim that the treatments produce the same outcome.

The result also does not identify the effect on each component of the composite endpoint separately. The endpoint combines recurrences, new invasive cancers, and death from any cause. Without component-specific results, it is not possible to determine which events contributed most to the overall comparison.

Secondary survival outcomes also have confidence intervals that include the no-difference value, while the cardiovascular event counts are sparse. Those findings do not resolve the uncertainty in the primary comparison. The safety information describes events recorded in each group, but the posted counts do not by themselves establish causation or a reliable between-group difference.

Finally, the registry’s terminated status is a study-status fact, not an explanation of the result. The reason for termination and the relevant analysis details are not reported in the registry record. The most defensible interpretation is therefore limited to the posted estimates, intervals, and counts: the primary comparison is uncertain, and the record does not establish benefit, harm, or equivalence.

18. Why This Trial Matters Statistically

This trial illustrates why an estimate should be read together with its uncertainty. The primary hazard ratio of 1.19 might seem to suggest a particular direction if considered alone, but its 95% confidence interval from 0.91 to 1.56 extends across the no-difference value. The interval changes the interpretation from a simple reading of the point estimate to a more cautious account of what the data can distinguish.

It also illustrates the difference between a time-to-event endpoint and a median time. The registry names median iDFS as the primary outcome but reports the comparative result as a hazard ratio and does not provide the group medians. A relative hazard summary and a median survival time answer related but distinct questions; one cannot be substituted for the other.

The trial’s secondary outcomes highlight further statistical issues. Different outcomes use different scales, and sparse events can leave comparisons imprecise. Assessing several outcomes also raises multiplicity concerns, particularly when the registry does not report a testing hierarchy or adjustment strategy. These features make it important to distinguish a primary result from secondary observations and to avoid over-weighting isolated comparisons.

Finally, the planning illustrations show how required information depends on the effect one aims to detect. Smaller assumed effects require more events or participants for a given planning target. Those illustrations are not observed power and are not evidence about the original trial design, whose assumptions are not reported. The case demonstrates why planning calculations, sponsor analyses, confidence intervals, and secondary outcomes must be understood as different kinds of statistical information.

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 does not report the reason the trial was terminated. Without that information, termination cannot be linked to the result or interpreted as evidence about the intervention. The record reports enrollment of 3021, but does not provide the original sample-size justification or the trial’s own design assumptions.

The primary outcome is described as median iDFS, yet group median values are not reported. Instead, the posted comparison is a hazard ratio with a confidence interval. The named analysis method, p-value, analysis population, covariate adjustment, and evaluation of model assumptions are not reported in the registry record. These omissions limit independent assessment and reproduction of the sponsor’s analysis.

The outcome definition is composite, and the posted primary result does not include a breakdown by event component. That makes it difficult to determine whether the overall estimate reflects a consistent pattern across recurrences, new invasive cancers, and death, or is driven more by particular components. The registry also does not provide details about missing outcomes or how censoring assumptions were evaluated.

Several secondary outcomes are listed, but multiplicity procedures and a testing hierarchy are not reported. The cardiovascular outcome has few recorded events, limiting the information available for a precise comparison. Safety summaries are not accompanied by a complete account of exposure-adjusted analyses or detailed comparisons. These limitations do not invalidate the posted results, but they restrict how far conclusions can be taken beyond the estimates and counts reported.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Aspirin in Preventing Recurrence of Cancer in Patients With HER2 Negative Stage II-III Breast Cancer After Chemotherapy, Surgery, and/or Radiation Therapy — NCT02927249.
• Registry record last updated 2026-09-04; status last verified 2026-08.
• Results first posted 2023-03-27.
• 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

NCT02927249, the ABC Trial, was a randomized, phase III, double-blind, placebo-controlled parallel-group study sponsored by the Alliance for Clinical Trials in Oncology. It enrolled 3021 participants and is listed as terminated. Aspirin once daily was compared with placebo once daily, with treatment planned for five years absent progression or unacceptable toxicity.

The primary outcome was median invasive disease-free survival over five years. The posted hazard ratio for aspirin versus placebo is 1.19, with a 95% confidence interval from 0.91 to 1.56; group median values and a p-value are not reported in the registry record. The interval includes the no-difference value, leaving the primary comparison uncertain. Secondary results and adverse-event counts add context but do not resolve that uncertainty. The registry does not report the reason for termination, the original design assumptions, or sufficient analysis details for independent reproduction.