The Exercise And Colorectal Cancer Treatment Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Exercise And Colorectal Cancer Treatment Trial (NCT03975491), a randomized na study. The primary outcome is The Percentage Change in High-sensitivity C-reactive Protein.

2. Clinical Question

The registry describes a randomized, parallel-group comparison. Its primary outcome is continuous: each group’s result is expressed as a mean percentage change, and the reported comparison is a net mean difference. The posted results therefore address the average change in this measure over the stated timeframe, rather than whether an individual participant experienced a particular change.

The clinical question should be kept close to that measurement. The registry record does not report a primary recurrence endpoint, and the posted primary result does not establish whether recurrence occurred or whether exercise altered recurrence risk. The record’s stated purpose supplies the broader biological context, while the primary endpoint supplies the specific quantity assessed.

3. Trial Design

The Exercise And Colorectal Cancer Treatment (EXACT) Trial is listed as a randomized, double-masked, parallel-group study. Its registry status is completed. The registry lists enrollment as 60 and identifies Pennington Biomedical Research Center as the sponsor. The recorded start date is 2019-08-01, and the primary completion date is 2023-05-01. The phase field is listed as NA.

The two intervention groups are moderate-intensity aerobic exercise and wait-list control. Both are classified as behavioral interventions. The registry’s arm descriptions identify the exercise arm as aerobic exercise and the control arm as wait-list control. Details of the exercise schedule, delivery, adherence expectations, and what participants on the wait-list were asked to do are not reported in the registry record.

The primary outcome is the percentage change in high-sensitivity C-reactive protein, with measurements specified at baseline and 12 weeks. It is reported as a mean with a 95% confidence interval. The outcome summaries list 31 participants in the exercise group and 28 in the control group. These are the group counts attached to the posted primary-outcome summaries, not a substitute for a detailed account of analysis populations or missing measurements.

The registry lists three secondary outcomes: percentage change in soluble tumor necrosis factor-alpha receptor two, change in circulating tumor cells per mL whole blood, and insulin resistance. It also reports eight outcome measures and five analyses. The record does not explain how these counts correspond to one another, or give a full schedule of outcome assessments.

Although masking is recorded as double, who was masked and how masking was implemented are not reported in the registry record. The allocation procedure, allocation concealment, eligibility criteria, and detailed study protocol are also not reported there. These omissions limit the detail available for assessing implementation, but do not change the basic registry description of a randomized parallel comparison.

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 The Percentage Change in High-sensitivity C-reactive Protein, measured over Baseline, 12 weeks.

The primary endpoint is the percentage change in high-sensitivity C-reactive protein, measured from baseline to 12 weeks. A percentage-change outcome expresses how a measure changes relative to its starting level; the registry reports the group summaries as means. This makes the endpoint a continuous measure rather than a count of events or a yes-or-no outcome.

The registry’s stated purpose is to examine biological processes through which exercise may prevent disease recurrence after colorectal cancer treatment. The primary endpoint provides a quantitative biological outcome for that purpose. It is not itself a measure of cancer recurrence, survival, symptoms, or quality of life. The registry record does not report why this particular measure was selected over other possible biological outcomes, nor does it state a threshold for interpreting a change as clinically important.

Values are reported for each group, followed by a mixed models analysis of the net mean difference. The arm-level means describe the average percentage change in each group. The net mean difference addresses the contrast between groups, which is the relevant randomized comparison. Its confidence interval describes the uncertainty around that estimated contrast. The timeframe is baseline and 12 weeks; longer-term results for this endpoint are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous, and the registry reports group means. A mean describes the average measured change among the observations included in the corresponding summary. It does not show how individual participants’ values were distributed, whether the distribution was symmetric, or whether a small number of observations strongly influenced the average. Those details are not reported in the registry record.

What does the confidence interval describe?

A confidence interval expresses uncertainty around an estimated quantity under the assumptions of the method used. For the arm-level summaries, the registry gives a 95% confidence interval for each mean. For the group comparison, it gives a 95% confidence interval for the net mean difference. The comparison interval is the more direct uncertainty summary for the randomized contrast. It runs from -17.1 to 76.2 around the reported net mean difference of 20.9. Because the interval includes negative and positive values, it does not settle the direction of the underlying difference. Its width conveys that the estimate remains imprecise.

A confidence interval is not a range containing a fixed proportion of individual participants’ measurements. It also does not assign a probability to the true value after the data have been observed. Its interpretation depends on the statistical model and assumptions used to construct it. The registry names a mixed models analysis for the comparison but does not provide model details, such as the covariance structure, adjustment variables, or handling of incomplete measurements.

What does the mixed models result represent?

The reported analysis is a mixed models analysis, with the comparison parameter described as a mean difference (net). In general, mixed models can account for relationships among repeated observations and can incorporate measured outcomes from participants with incomplete repeated measurements under specified assumptions. That general description does not establish which model features were used in this trial. The registry record does not report the exact model specification, the number of observations at each assessment, or the assumptions made about missing data.

The outcome timeframe names baseline and 12 weeks. The posted summary does not make clear whether the reported group means are calculated from individual percentage changes or from another analysis representation. The registry’s own result labels the reported contrast as a net mean difference from a mixed models analysis. The analysis should be interpreted using that stated parameter rather than treating it as a simple subtraction of the posted arm means.

How should the p-value be read?

The reported p-value is 0.32. In general, a p-value describes how unusual data at least as inconsistent with a specified null model would be if that model and its assumptions held. It is not the chance that the result is due to chance, and it does not measure clinical importance. The registry record does not state the null hypothesis in detail, the sidedness of the test, or a significance threshold for this analysis. The p-value is best considered together with the estimated difference and its confidence interval.

What do planning illustrations mean?

Sample-size illustrations are planning calculations, not results from the trial and not observed power. The registry’s description of the illustrations states that they use a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a fraction of the observed effect. For continuous outcomes, such illustrations are expressed as participants per arm. The registry record does not report that these assumptions were the trial’s original design assumptions, nor does it give a sample-size justification for the trial. A planning illustration cannot establish the probability that this completed study detected an effect.

Any such illustration depends on its assumptions, including the chosen effect to detect and the statistical model used. Smaller target effects generally require more information to distinguish from random variation, while larger target effects are easier to detect. This is a general statistical relationship, not a statement about the trial’s original planning. The registry’s planning description expressly characterizes the calculation as an illustration based on the observed effect, rather than the trial’s own design.

What is not specified?

The registry does not report the full statistical analysis plan, model diagnostics, treatment of outliers, transformations, covariate adjustment, or handling of missing outcome values. It also does not explain the construction of the percentage-change measure or describe whether the analysis followed intention-to-treat principles. These details matter because model choices and analysis populations can affect an estimate and its uncertainty. Without them, the reported result can be described, but the analysis cannot be fully reproduced from the registry record alone.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s comparison as a mixed models analysis: a net mean difference of 20.9, with a 95% confidence interval from -17.1 to 76.2 and p = 0.32. This is the available inferential comparison for the primary endpoint. The arm-level means and their confidence intervals are descriptive summaries; they are not a replacement for the reported model-based contrast.

The registry does not provide a separate recalculated estimate for this continuous two-arm comparison. As a result, there is no independent recalculated contrast to compare numerically with the sponsor’s result. The sponsor’s analysis is not marked as discrepant in the registry record. Its estimate and uncertainty should therefore be read as the posted analysis, while recognizing that the model specification and supporting analysis details are not reported in the registry record.

The reported interval spans negative and positive values. This means the result does not precisely locate the group difference on one side of zero. The p-value should be read as part of the sponsor’s analysis, not as a measure of effect size or clinical importance. The record does not provide enough information to reconstruct the mixed models analysis or assess its assumptions independently.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures and five analyses, and it identifies three secondary outcomes: percentage change in soluble tumor necrosis factor-alpha receptor two, change in circulating tumor cells per mL whole blood, and insulin resistance. The record does not explain how the outcome-measure and analysis counts relate to the listed outcomes. It also does not report a multiplicity adjustment, a hierarchy of outcomes, or a prespecified strategy for interpreting several tests. Without that information, the risk of chance findings across multiple analyses cannot be assessed from the registry record.

For the primary endpoint, the record gives an outcome summary and a mixed models comparison. It does not report the number of observations available at each assessment, reasons for missing measurements, or the method used to handle missing data. Missingness can affect an estimate when the observed participants differ from those whose measurements are absent. Statistical methods can address some patterns of incomplete data only under assumptions; the relevant assumptions and checks are not reported in the registry record.

The registry record also does not describe the primary analysis population, such as whether analyses included all randomized participants or only participants with usable measurements. The posted arm counts for the primary outcome are 31 for exercise and 28 for control. Those counts do not, on their own, identify the analysis population or explain exclusions and missing observations.

Although randomization is listed, the record does not report whether the analysis preserved the randomized group assignments in all analyses, or whether any participants switched groups or stopped intervention. These omissions do not prove that the analysis was inappropriate; they mean that the registry record does not contain enough detail to evaluate these issues. The listed results should be interpreted as posted, with uncertainty about multiplicity handling, missing-data procedures, and analysis populations.

12. 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.

For the safety summaries, the registry lists 31 participants at risk in the exercise group and 29 in the control group. It reports zero serious adverse events in each group and zero other adverse events in each group. Deaths are listed as zero in the exercise group and one in the control group. These safety denominators differ from the primary-outcome summary counts; each set of figures should be read in the context of its own outcome.

Zero reported events in a group means no events of that category were recorded in the posted summary for that group. It does not establish that the risk is absent, particularly when the information available is limited. The death count is a descriptive registry figure and does not establish that the death was caused by the intervention or unrelated to it. Cause, timing, and relationship to study intervention are not reported in the registry record.

The registry does not list serious adverse event terms. The figures should therefore not be extended beyond the categories and counts posted. They do not give a detailed account of event severity, duration, clinical circumstances, or how events were monitored. These limitations are especially relevant when using event summaries to compare groups or draw conclusions about safety.

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

The result describes percentage change in high-sensitivity C-reactive protein between baseline and 12 weeks in a randomized comparison of moderate-intensity aerobic exercise with a wait-list control. The posted means are 12.8 for exercise and -6.7 for control. The sponsor-reported net mean difference is 20.9, with a 95% confidence interval from -17.1 to 76.2 and p = 0.32.

The comparison interval covers negative and positive values, so the direction and size of the underlying difference are uncertain. The point estimate is one estimate from the data, not a precise statement of what would occur in every participant or in another study. The p-value does not indicate the probability that exercise has or lacks an effect, and it does not quantify clinical value. The record does not report a clinically meaningful threshold for this endpoint.

The posted result concerns a biological measure over the specified timeframe. The registry’s stated purpose refers to biological processes through which exercise may prevent recurrence, but the primary result does not measure recurrence. It therefore cannot be read as evidence about recurrence, survival, or other clinical outcomes. The listed secondary outcomes include other biological measures and insulin resistance, but the primary result alone does not establish findings for those outcomes.

The registry record does not supply enough information to assess the full statistical model, missing-data handling, multiplicity strategy, or analysis population. Those limitations constrain how completely the estimate can be evaluated. The result is best understood as the reported mean comparison and its uncertainty, rather than as a definitive statement about the intervention or its broader clinical effects.

14. Why This Trial Matters Statistically

This trial illustrates why a randomized study’s primary question should be separated from the broader motivation for conducting it. The EXACT Trial states an interest in biological processes related to possible recurrence prevention after colorectal cancer treatment. Its posted primary outcome, however, is percentage change in high-sensitivity C-reactive protein from baseline to 12 weeks. That distinction keeps interpretation anchored to what was actually measured.

It also illustrates the importance of reading an estimated contrast with its confidence interval. The reported net mean difference is 20.9, while its 95% confidence interval extends from -17.1 to 76.2. The estimate alone might seem to point in one direction, but the interval shows substantial uncertainty about both direction and magnitude. A p-value of 0.32 adds a test-based summary; it does not replace the estimate, its uncertainty, or a judgment about clinical relevance.

The record also shows why methods transparency matters. The trial is identified as randomized, double-masked, and parallel-group, and a mixed models analysis is named. Yet the registry record does not describe the model specification, missing-data procedures, analysis populations, or multiplicity strategy. Those details are important for understanding how the result was obtained and how robustly it can be interpreted.

Finally, registry outcome and safety summaries answer narrower questions than a full clinical report might. The primary biomarker result is not a recurrence result, and the adverse-event counts do not provide detailed event context. Recognizing these boundaries helps readers distinguish the posted evidence from the study’s stated aims.

15. Limitations

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

The registry record provides a limited description of the intervention and comparator. It names moderate-intensity aerobic exercise and a wait-list control but does not report exercise frequency, duration, supervision, adherence, or the activities permitted in the control group. These details would help readers understand what exposure the comparison represents.

Masking is recorded as double, but the record does not identify who was masked or explain how masking was maintained. The allocation procedure and allocation concealment are also not reported. Eligibility criteria and detailed participant characteristics are not available in the registry record, limiting assessment of how closely the enrolled population matches the broader population of people who completed colorectal cancer treatment.

The primary endpoint is described as percentage change in high-sensitivity C-reactive protein at baseline and 12 weeks. The registry does not explain the calculation of percentage change, provide the full mixed models specification, or describe model diagnostics and covariate adjustment. It does not report the analysis population, missing-data approach, or reasons for unavailable measurements. These omissions make independent reproduction and evaluation of the primary analysis difficult.

The record lists multiple outcome measures and analyses but does not report how multiplicity was handled or how secondary outcomes were prioritized. It also does not provide a detailed interpretation threshold for the primary outcome. The posted group summaries and confidence intervals therefore cannot establish clinical importance on their own.

Finally, the primary endpoint is a 12-week biological measure, while the study’s stated purpose concerns biological processes related to recurrence prevention. The record does not post recurrence results as the primary outcome. The safety summary gives event counts but not event descriptions, timing, or attribution. These are limitations of what can be concluded from the registry record, not proof that the study lacked additional information elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The Exercise And Colorectal Cancer Treatment Trial — NCT03975491.
• Registry record last updated 2026-10-07; status last verified 2026-09.
• Results first posted 2025-09-22.
• 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.

19. Record Summary

The EXACT Trial is listed as a completed, randomized, double-masked, parallel-group study sponsored by Pennington Biomedical Research Center. The registry lists enrollment as 60 and compares moderate-intensity aerobic exercise with wait-list control in people who completed colorectal cancer treatment.

The primary outcome is percentage change in high-sensitivity C-reactive protein from baseline to 12 weeks. The reported mixed models net mean difference is 20.9, with a 95% confidence interval from -17.1 to 76.2 and p = 0.32. The interval spans negative and positive values, leaving the direction and magnitude of the difference uncertain.

The registry also lists three secondary outcomes. Important details of the statistical model, missing-data handling, multiplicity strategy, and analysis populations are not reported in the registry record. The primary endpoint is a biological measure, not a posted recurrence outcome.