Study of Vitamin D in Untreated Metastatic Colorectal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Vitamin D in Untreated Metastatic Colorectal Cancer (NCT01516216), a randomized phase 2 study. The primary outcome is Median Progression-free Survival (PFS).

2. Clinical Question

The trial addresses a dose comparison, not vitamin D versus no vitamin D: both randomized groups received vitamin D and the same chemotherapy combination. Accordingly, the primary contrast concerns the higher-dose regimen relative to the standard-dose regimen in this study population. It does not directly estimate what would happen with chemotherapy alone or with vitamin D alone.

3. Trial Design

This was a randomized, double-blind, parallel-group phase II trial in participants with previously untreated metastatic colorectal cancer. The registry record identifies Dana-Farber Cancer Institute as the sponsor and lists the study status as completed. The study is described as evaluating the efficacy and safety of two vitamin D doses in combination with standard chemotherapy.

Both groups received FOLFOX-bevacizumab intravenously on Day 1, with timing described as +/- 7 days of every two-week cycle, according to institutional standards. The standard-dose group received 400 IU vitamin D3 orally once daily. The higher-dose group received 8000 IU daily for 2 weeks as a loading dose, followed by 4000 IU daily as maintenance. Participants were treated until disease progression, unacceptable toxicity, or withdrawal for other reasons, as described for the standard-dose arm; the higher-dose arm description also states treatment continued until disease progression or withdrawal, but its final text is truncated in the registry record.

Randomization supports comparison of the assigned regimens, while double masking is intended to reduce the influence of treatment knowledge on participants’ experiences and outcome assessment. The record does not report the randomization procedure, allocation concealment details, or which groups of people were masked. The primary analysis population is all randomized patients. The record describes 139 enrolled participants and provides arm-specific endpoint denominators, which can differ across outcomes.

The primary endpoint was median progression-free survival. Disease was evaluated every 4 cycles during treatment, and follow-up after treatment discontinuation is described as occurring every 8-16 weeks until progression or the start of non-protocol therapy. Participants were observed up to 28.5 months, with maximum follow-up of 56.7 months. The registry record does not report the planned sample-size justification or detailed 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: the analysis population is comprised of all randomized patients

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 Progression-free Survival (PFS), measured over Disease was evaluated every 4 cycles on treatment and off treatment every 8-16 weeks until PD or non-protocol therapy start if discontinued for reason other than PD. Participants were observed up to 28.5 months with maximum follow-up of 56.7 months..

Progression-free survival (PFS) is the duration from study entry until documented disease progression or death. The registry record says PFS was based on the Kaplan-Meier method. It defines progression using RECIST 1.1 criteria: for target lesions, progression includes at least a 20% increase in the sum of longest diameters relative to the smallest sum recorded since treatment began, or the appearance of one or more new lesions. For non-target lesions, progression includes new lesions or unequivocal progression of existing non-target lesions.

The median is the time by which the estimated survival curve reaches the point where half of the population is estimated to have experienced the event. It offers a clinically interpretable summary of a time-to-event distribution, while retaining information about the timing of events and allowing for censoring. Censoring occurs when a participant’s event time is not observed during the available follow-up. Patients who stopped treatment to pursue potentially curative resection were censored at surgery; those without documented progression or death were censored at their last known follow-up date.

PFS is useful in a trial of cancer treatment because it captures progression or death, rather than response alone. It does not by itself describe overall survival, quality of life, or the duration of benefit for every individual.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

How should the median and confidence interval be read?

The Kaplan-Meier method estimates the proportion of participants who remain free of the event over time, accounting for participants whose follow-up ends before progression or death is observed. The median PFS is the time corresponding to the point where the estimated proportion without an event reaches one half. A 95% confidence interval gives a range of values compatible with the data under the method’s assumptions and reflects uncertainty in the estimated median. It is not a range containing 95% of individual participants’ PFS times.

Separate confidence intervals are reported for the two arms. They help show the precision of each arm’s median, but they are not themselves a confidence interval for the difference between arms. A direct comparison requires a method that uses the outcome data from both groups together.

What does the log-rank test assess?

The log-rank test compares the timing of events across the groups over follow-up. It uses the information available at event times and accounts for censoring. The reported p-value of 0.07 is evidence about compatibility between the observed data and the test’s null hypothesis; it does not measure the size or clinical importance of a difference. The test’s assumptions and the detailed analysis plan are not reported in the registry record.

Time-to-event comparisons are often also presented with a hazard ratio, which describes the relative event rate over time under a proportional-hazards interpretation. No sponsor-reported hazard ratio for the primary comparison is listed in the registry record. A hazard ratio should not be inferred from the two medians or the log-rank p-value.

Why do denominators vary across outcomes?

Different outcomes can have different numbers of participants with evaluable data. The posted results use outcome-specific denominators, and the analysis population for the primary endpoint is described as all randomized patients. A denominator smaller than the randomized arm size does not, by itself, explain why data are absent. Reasons for missing outcome data and any imputation procedures are not reported in the registry record.

For PFS, the record describes censoring at surgery for participants pursuing potentially curative resection, and at the last known follow-up for those without progression or death. Censoring allows such participants to contribute follow-up up to the point their outcome is no longer observed. Its validity depends on assumptions about the relationship between censoring and future event risk; additional diagnostics or sensitivity analyses are not reported in the registry record.

What do recalculated secondary comparisons represent?

For selected binary secondary outcomes, the posted recalculations use the arm-level counts to estimate simple, unadjusted comparisons. A risk difference expresses the difference between the observed proportions in the two groups. Its confidence interval describes uncertainty in that contrast. When the interval includes zero, the data are compatible with no difference as well as with differences in either direction at the interval’s confidence level.

These recalculations are not adjusted analyses and do not replace the sponsor’s own analysis. The registry reports sponsor p-values for some outcomes, using methods identified as chi-squared tests. For the primary median time-to-event endpoint, the recalculated comparison is not available because the posted summaries do not permit that two-arm recalculation. The sponsor’s log-rank analysis is the reported comparison for PFS.

How should planning illustrations be understood?

Planning illustrations based on observed secondary effects are hypothetical design calculations, not observed trial findings and not estimates of the trial’s power after completion. The record describes these illustrations as using a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect set at a stated fraction of the observed effect. Such calculations depend strongly on the assumed effect size. Smaller target effects generally require larger planned groups to achieve a given power under the calculation’s assumptions.

The sample-size justification used to design this trial is not reported in the registry record. Planning illustrations derived from observed data should not be used to claim that the trial was adequately powered or that a null result demonstrates equivalence.

10. Analyses Reported by the Sponsor

For the primary PFS comparison, the registry reports a log-rank p-value of 0.07. It does not report a primary hazard ratio or a recalculated comparison based on the arm-level median summaries. The registered time-to-event result is therefore the sponsor’s own reported comparison; a hazard ratio or an alternative primary p-value should not be reconstructed from the posted medians and confidence intervals.

For some binary secondary outcomes, simple unadjusted recalculations are available from the posted counts. These calculations are not necessarily identical to the sponsor’s analysis: the registry identifies chi-squared tests for objective response and baseline vitamin D deficiency, while the recalculations use a risk-difference approach. Their p-values need not match because the methods and estimands differ. The record does not report a sponsor p-value for the treatment-related neutropenia outcome.

The discrepancy flag is false. The recalculations should be treated as descriptive checks based on posted summary data, not as corrections to the registered analysis. Where the registry reports its own analysis, that remains the primary account of the sponsor’s analysis. Neither an unadjusted recalculation nor a sponsor p-value alone establishes clinical importance.

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary endpoint and 8 secondary outcomes. When multiple outcomes are tested, the chance of at least one apparently notable result arising by chance can increase if each test is interpreted without accounting for the others. Statistical approaches can control this issue, for example by adjusting thresholds or specifying a hierarchy of tests. Whether the trial used such an approach, and which secondary outcomes were considered confirmatory, is not reported in the registry record. The posted analyses should not be treated as though they were all independent confirmatory tests.

The primary analysis population is stated to comprise all randomized patients. However, the arm-level denominators vary across endpoints, and the registry does not explain all differences in the available data. The primary endpoint description specifies censoring at the time of surgery for patients who discontinued treatment to pursue potentially curative resection, and at the last known follow-up for participants without progression or death. The record does not report imputation methods, detailed missing-data assumptions, or sensitivity analyses addressing missing outcomes.

The registry record does not report the trial’s sample-size justification, a full prespecified statistical analysis plan, or whether interim analyses were conducted. It also does not report the intended alpha threshold for the primary comparison in the information. The planning illustrations associated with secondary outcomes use stated assumptions, but they are not evidence of the trial’s original design rationale or of observed power.

These omissions limit assessment of how closely the posted results follow the prespecified plan and how robust the results are to alternative missing-data or multiplicity approaches. They do not invalidate the reported values, but they mean that the p-values and confidence intervals should be read alongside the limits of the available analysis details.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: neutrophil count decreased (32.4%), hypertension (14.9%), peripheral sensory neuropathy (10.4%), diarrhea (9.0%), fatigue (7.5%).

The registry reports serious adverse events in 39 of 67 participants in the standard-dose arm and 47 of 68 in the higher-dose arm. It also reports other adverse events in 65 of 67 and 67 of 68 participants, respectively, and deaths of 54 and 45. These are counts within the safety groupings recorded in the registry; they are not, on their own, evidence that vitamin D dose caused an event or death. The record does not specify in this summary the timing, attribution, or relationship of the deaths to treatment.

The listed serious adverse event terms include neutrophil count decreased at 32.4%, hypertension at 14.9%, peripheral sensory neuropathy at 10.4%, diarrhea at 9.0%, fatigue at 7.5%, and hyponatremia at 7.4%. The percentages are reported as terms in the registry and are not presented by treatment arm in the available information. They should not be interpreted as comparative risks between the two groups.

Safety counts describe events observed during the study, with adverse events evaluated on day 1 of each cycle and up to 30 days after the last dose. They do not by themselves establish comparative causality or capture the full clinical context of each event.

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

The primary result is a comparison of median PFS between two randomized vitamin D dosing regimens, each given with FOLFOX-bevacizumab. The posted medians are estimates, and their confidence intervals show uncertainty. The log-rank p-value describes the statistical comparison recorded for the trial; it does not tell readers the probability that a treatment works, nor does it establish that the two regimens are equivalent.

The result does not show the effect of vitamin D compared with no vitamin D, because both groups received supplementation. It also does not isolate the effect of chemotherapy, since both groups received the same chemotherapy combination. The comparison is specific to the assigned dose regimens and the study population described in the registry record.

A group median is not a prediction of an individual participant’s outcome. Participants can have substantially different progression-free times, and censored observations mean that not every participant’s event time was observed. The posted data do not establish how the result applies to other settings or populations.

The secondary outcomes add context but do not settle the primary comparison. Some measures concern response or toxicity; others concern vitamin D levels or survival. Differences in outcome definitions, denominators, uncertainty, and the number of outcomes examined affect how much weight each result can bear. No medical decision should be based on these statistics alone.

Overall, the registry results support a careful description of the observed arm-specific PFS estimates and the sponsor-reported test. They do not warrant a claim that the higher-dose regimen is definitively better, worse, or equivalent, and they do not establish a causal benefit from vitamin D relative to no supplementation.

14. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison should be read through both its estimated outcomes and the uncertainty around them. Median PFS is readily interpretable, but the medians alone do not quantify the full between-group effect. The log-rank test provides a comparison of event timing, while the absence of a reported primary hazard ratio limits the available description of relative event rates.

It also shows why endpoint-specific denominators matter. The primary analysis population is described as all randomized patients, but not every posted outcome uses the same denominator. Without a full account of missing observations and analysis decisions, readers should distinguish the randomized population from the participants contributing data to each secondary measure.

The design tests dose intensity within a shared chemotherapy regimen. That makes the comparison focused, but it also limits what can be inferred: the trial does not include a chemotherapy-only or no-vitamin-D arm. Statistical interpretation must follow the actual contrast randomized, rather than a broader question about supplementation.

Finally, secondary measures and planning calculations can be informative without being decisive. Multiple outcomes can generate a varied pattern, and calculations based on observed effects are not observed power or proof that the original design was adequate. A careful reading keeps the primary comparison, secondary evidence, and limits of the registry record distinct.

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 does not provide the primary hazard ratio or its confidence interval, so the relative event-rate contrast cannot be assessed from the reported medians alone. It also does not report a detailed statistical analysis plan, the sample-size justification, the primary testing threshold, or whether interim analyses were conducted. These gaps restrict evaluation of the prespecification and operating characteristics of the analysis.

Outcome denominators vary, and reasons for missing endpoint data are not reported in the registry record. Although censoring rules for PFS are described, no sensitivity analyses or diagnostics for censoring assumptions are reported. The record also does not describe imputation procedures for missing data.

Only some of the listed secondary outcomes have numerical results in the record. Multiplicity handling is not reported, and multiple unadjusted comparisons can increase the chance of findings that appear notable by chance. Secondary analyses therefore have limited confirmatory weight.

The trial compares two vitamin D doses alongside the same chemotherapy; it cannot estimate a vitamin D-versus-no-vitamin-D effect. The serious adverse event terms and percentages are not shown by arm in the available information, and the relationship of reported deaths to treatment is not specified. Finally, the higher-dose arm description is truncated in the registry record, limiting the detail available about its treatment discontinuation conditions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Vitamin D in Untreated Metastatic Colorectal Cancer — NCT01516216.
• Registry record last updated 2022-04-26; status last verified 2022-04.
• Results first posted 2022-03-23.
• 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

NCT01516216 is a completed randomized, double-blind, parallel-group phase II trial of vitamin D supplementation with FOLFOX-bevacizumab in previously untreated metastatic colorectal cancer. It compares a standard-dose vitamin D regimen with a higher-dose loading and maintenance regimen. The primary endpoint is median PFS, defined as time from study entry to documented progression or death and analyzed using the Kaplan-Meier method.

The registry reports median PFS of 11.0 months in the standard-dose arm and 13.0 months in the higher-dose arm, with a log-rank p-value of 0.07. The record lists secondary outcomes and safety results, but does not report all listed secondary results or a primary hazard ratio. Interpretation is limited by unreported details on multiplicity, missing data, and the original sample-size justification.