S0016 Combination Chemotherapy With Monoclonal Antibody Therapy in Newly Diagnosed Non-Hodgkin's Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for S0016 Combination Chemotherapy With Monoclonal Antibody Therapy in Newly Diagnosed Non-Hodgkin's Lymphoma (NCT00006721), a randomized phase 3 study. The primary outcome is Progression-free Survival at 2 Years.

2. Clinical Question

The trial's brief description frames the study as a comparison of monoclonal antibodies given with combination chemotherapy. It does not establish that either regimen is more effective; that is the question the randomized comparison was intended to address. The registry lists 571 enrolled participants, but the primary outcome population is defined separately as eligible participants who started treatment. The posted primary results cover the two combination regimens, not the CHOP-only arm.

The question has a time-to-event component: progression or relapse, or death from any cause, marks the outcome event, while last contact is also part of the registry's endpoint description. This matters because progression-free survival is not the same as response at a particular assessment. A response outcome asks whether a response was recorded; progression-free survival concerns the time until a specified event or the last recorded contact. Those measures can inform different aspects of treatment comparison and should not be treated as interchangeable.

3. Trial Design

This is a randomized, phase III, parallel-group trial with no masking reported. The registry lists SWOG Cancer Research Network as sponsor, an enrollment of 571, a start date of 2001-03, primary completion in 2011-09, and a completed status. The trial title describes a comparison of CHOP plus rituximab with CHOP plus iodine-131-labeled monoclonal anti-B1 antibody, tositumomab. A CHOP-only arm is also described in the treatment information.

The primary comparison posted for progression-free survival at 2 years includes CHOP plus rituximab and CHOP plus tositumomab. The registry explicitly says that participants in the CHOP-only arm were not assessed for this endpoint due to early closure. Thus, the existence of a third treatment arm in the trial description should not be mistaken for a third group in the reported primary comparison.

The arm descriptions identify cyclophosphamide, doxorubicin, vincristine, and prednisone as the CHOP drugs. In the CHOP plus rituximab arm, rituximab is added to that chemotherapy; in the CHOP plus tositumomab arm, tositumomab is added. The dosing descriptions include schedules, but the registry text is incomplete in places, so the full treatment schedules are not reported in the registry record.

Randomization helps form treatment groups for comparison, while the parallel design means the groups receive their assigned regimens during the trial rather than crossing between them. No masking is reported, so the record describes an unmasked study. The registry record does not report the randomization method, allocation ratio, stratification factors, or the original design assumptions for the trial's sample size. Those details should not be inferred from the posted outcome counts or from planning illustrations based on observed results.

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: All eligible patients who started treatment were included in the analysis. The participants in CHOP alone arm were not assessed due to early closure.

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 Progression-free Survival at 2 Years, measured over 0-2 years.

The primary endpoint is progression-free survival at 2 years, reported as a percentage of participants, with a timeframe of 0-2 years. The registry defines the measurement from registration to the first observation of progression or relapse, death due to any cause, or last contact. In this setting, the endpoint captures whether participants remain free of the specified progression, relapse, or death event over the stated period, with contact information contributing to follow-up.

Progression-free survival is used in clinical trials because it combines disease progression or relapse with death into an outcome that reflects the time participants remain without either event. It differs from an objective response measure, which records whether a tumor response was observed. It also differs from overall survival, which concerns death rather than progression or relapse. General statistical context: time-to-event endpoints can use information about when events occur, while a percentage at a specified time gives a more focused snapshot of the proportion meeting the endpoint by that time.

The registry record does not provide a confidence interval for either group’s posted percentage. It does give a sponsor-reported Cox regression hazard ratio with a confidence interval and p-value. The exact procedure used to derive the displayed percentages, including handling of incomplete follow-up for that display, is not reported in the registry record. The event definition and time window are clear, but the reported materials do not fully specify every analytic detail.

7. Results Posted in the Registry

• CHOP + Tositumomab: 80.0% (212/265) versus 76.0% (203/267) on control; risk difference 4.0 percentage points (95% CI -3.1 to 11.0), risk ratio 1.05 (0.96–1.15); unadjusted two-proportion z-test p = 0.269.

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.

Objective Response (Confirmed and Unconfirmed Complete and Partial Responses)

• 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:
  • CHOP + Tositumomab: 84.2% (223/265) versus 83.9% (224/267) on control; risk difference 0.3 percentage points (95% CI -6.0 to 6.5), risk ratio 1.00 (0.93–1.08); unadjusted two-proportion z-test p = 0.936.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CHOP + Tositumomab and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists objective response and adverse events among the secondary outcomes. The objective-response results are posted for the same two combination regimens used in the primary comparison, while the adverse-event record includes the CHOP-only group as well. Taken together, these outcomes address different dimensions of the trial: response describes tumor response, and safety outcomes describe recorded harms. They do not substitute for the progression-free survival endpoint.

The objective-response comparison does not show a clear separation between the two combination regimens in the posted results. This broad pattern is consistent with the primary comparison's lack of a statistically clear difference, but the endpoint definitions differ, so similarity across these outcomes should not be treated as independent confirmation of the primary result. Response and progression-free survival capture distinct features and time structures.

Secondary outcomes are generally less definitive when many outcomes or comparisons are examined, particularly if no adjustment for multiplicity is reported. The registry does not report a multiplicity procedure or identify which secondary hypotheses were prespecified. Without that information, unadjusted secondary p-values can have an elevated chance of apparently notable results arising from the number of comparisons. They are best read as supportive or exploratory evidence rather than as stand-alone confirmation.

The registry lists six outcome measures and two analyses, but the full analysis plan and the status of all planned outcomes are not reported. The number of posted secondary outcomes is limited, and the record does not establish whether omitted or unreported outcomes were absent from the trial plan. The overall secondary-outcome pattern should therefore be interpreted cautiously and alongside the primary endpoint, rather than as a separate basis for a strong treatment claim.

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

• Objective Response (Confirmed and Unconfirmed Complete and Partial Responses)
• Number of Patients With Grade 3 Through Grade 5 Adverse Events That Are Related to Study Drug

9. Statistical Interpretation

10. Statistical Methods Explained

Randomization is intended to make treatment groups comparable, on average, with respect to factors that could influence outcomes. It does not guarantee exact balance in every characteristic, and it cannot prevent bias from issues arising after assignment, such as differences in follow-up or outcome assessment. The registry identifies the study as randomized and parallel, but does not report the sequence-generation method, concealment procedure, or stratification details. Those parts of the randomization process are therefore not reported in the registry record.

What does the primary percentage represent?

The endpoint is progression-free survival at 2 years, expressed as a percentage of participants. The registry's event definition includes progression or relapse and death from any cause, with last contact also identified in the measurement description. A percentage at a fixed time provides an interpretable summary of the outcome at that time. It should not automatically be read as a raw count divided by the listed group size: the registry does not explain the exact calculation used for the posted percentages or the treatment of participants whose outcome status was not observed at that time.

General statistical explanation: time-to-event analyses are useful when participants have different lengths of observation or when some are last known to be event-free before the end of follow-up. Such participants may be censored rather than treated as having experienced the event. The registry reports a Cox regression analysis, a common model for time-to-event data, but does not describe the censoring rules, model covariates, or diagnostic checks. Accordingly, the model result can be described, but its full implementation cannot be reconstructed from the record.

How should the risk difference be understood?

The recalculated risk difference is 0.040 for CHOP plus tositumomab compared with CHOP plus rituximab. A risk difference expresses the absolute contrast between the two posted group proportions. Its 95% confidence interval is -0.031 to 0.11. General statistical explanation: an interval estimates a range of values compatible with the observed data under the calculation's assumptions. An interval spanning zero includes no difference, while its width conveys uncertainty. The recalculation is unadjusted and based on the posted summary statistics; it is not a replacement for the trial's own model-based analysis.

The accompanying unadjusted p-value is 0.269. A p-value measures how unusual the data, or more extreme data, would be under a specified null hypothesis and statistical procedure. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. Its interpretation depends on the analysis being tested. Here, it pertains to the unadjusted recalculation and should not be conflated with the p-value from the Cox regression.

Why are the reported hazard ratio and risk difference different?

The hazard ratio of 0.79 comes from a Cox regression analysis, with a 95% confidence interval from 0.6 to 1.05 and p = 0.11. A hazard ratio compares instantaneous event rates over the follow-up period, subject to the model and its assumptions. The risk difference compares the posted percentages at a specified time. These measures summarize different features of the outcome, so their estimates and p-values need not match. The registry does not identify the reference coding for the hazard ratio or report covariate adjustment, so those details should not be inferred.

General statistical explanation: a Cox model commonly relies on a proportional-hazards assumption, meaning the relative hazard is treated as stable over time. Whether that assumption was assessed, and whether it was appropriate in this trial, is not reported in the registry record. A confidence interval for a hazard ratio is compared with 1 as the no-relative-difference value; the reported interval includes 1. This is a statement about statistical uncertainty, not proof of equal outcomes.

What can the event counts add?

The recalculation uses the posted summaries for the two assessed arms: 203 of 267 in the CHOP plus rituximab group and 212 of 265 in the CHOP plus tositumomab group. These counts accompany the percentage comparison, but the record does not give enough detail to establish whether they are simple observed successes or derived from a time-to-event estimator. They should therefore be read in the context of the registry's stated percentage endpoint, rather than used to reconstruct an unreported analytic procedure.

The posted analyses do not report adjustment for multiple comparisons, a complete list of prespecified statistical hypotheses, or an analysis plan. The registry lists six outcome measures and two analyses, but does not identify the full relationship between all planned outcomes and analyses. No assumption about interim analyses, stopping boundaries, or a prespecified covariate model is warranted from the information recorded.

11. Confidence Intervals and Precision

The primary group estimates are 76.0% for CHOP plus rituximab and 80.0% for CHOP plus tositumomab. The registry does not report confidence intervals around these separate percentages. The unadjusted risk-difference recalculation gives an estimate of 0.040 and a 95% confidence interval from -0.031 to 0.11. Because the interval includes zero and stretches across a range of possible differences, the comparison is not estimated with enough precision to rule out no difference or to define a narrow range of plausible effects.

The sponsor-reported Cox regression gives a hazard ratio of 0.79 with a 95% confidence interval from 0.6 to 1.05. The interval includes 1, the value representing no relative difference in hazard. Its width also indicates uncertainty in the estimated relative event rate. The confidence interval applies to the hazard ratio, not to the fixed-time percentages or the recalculated risk difference.

General statistical explanation: a confidence interval is a measure of uncertainty around an estimate under the assumptions of the analysis. A narrower interval indicates greater precision, while a wider interval leaves more possible values compatible with the data. An interval that includes a null value does not prove that the null is correct; an interval that excludes it does not by itself establish clinical importance. Interpretation also depends on the outcome, design, and the range of effects that would matter in context.

No equivalence or noninferiority margin is reported in the registry record. The results therefore should not be described as proving the regimens equivalent. The available intervals indicate that uncertainty remains around both the magnitude and, depending on the measure, the direction of the 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 figures are illustrations based on the observed effect, not the original trial design. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row considers detecting the stated fraction of the observed effect. The registry's own original sample-size assumptions are not reported in the registry record.

For the binary primary endpoint, the figures are participants per arm, not a total across arms. The 80% power column describes the planning number per arm that would give an 80% chance of detecting the specified effect under the listed assumptions, if that effect were the true effect and the assumptions held. The 90% power column gives the corresponding planning number per arm for a higher chance of detecting it. These are design calculations; they are not statements about the probability that the trial's result is correct.

The planning figures use the observed effect as their starting point and also consider planning for 75% and 50% of the observed effect. Smaller effects require more participants per arm to detect under the same assumptions. This relationship is why the figures increase as the target effect is reduced. A very small observed effect can lead to especially large planning illustrations, but those values do not establish that such a future trial is necessary or feasible.

These values are not observed power. They do not measure what the completed trial achieved after the outcome was known, and they should not be used to reinterpret a nonsignificant result as evidence of absence. They are prospective illustrations of sample size under stated assumptions. Because the original design assumptions and the complete analysis plan are not reported, the illustrations cannot be treated as a reconstruction of the trial's planning process.

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 = 0.760; treatment proportion p₁ = 0.800; average p̄ = 0.780.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5857 = 1.1479.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5850 = 0.4924.
4. n = (1.1479 + 0.4924)² / (0.0397)² = 1,707.1, rounded up to 1,708 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 <- 0.760299625468165                          # control proportion
p1 <- 0.8                          # 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(6721)
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 = 1,708 (80% power), 2,285 (90% power)
 75% of observed effect: n per arm = 3,085 (80% power), 4,129 (90% power)
 50% of observed effect: n per arm = 7,050 (80% power), 9,437 (90% power)
power.prop.test: 1707.1 
Simulated power at n = 1,708 per arm: 0.803 (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. Analyses Reported by the Sponsor

The sponsor's posted primary analysis is a Cox regression with a hazard ratio of 0.79, a 95% confidence interval from 0.6 to 1.05, and p = 0.11. It is the trial's reported model-based analysis of the time-to-event outcome. The record does not specify the reference-group coding, covariates, or other model details, which limits a more complete interpretation of the estimate.

The recalculated comparison is unadjusted and uses the posted endpoint summaries. It expresses the contrast as a risk difference of 0.040, with a 95% confidence interval from -0.031 to 0.11 and an unadjusted p-value of 0.269. This is not the same estimand or method as the Cox regression: one compares fixed-time percentages, while the other models event rates over time. Their numerical estimates and p-values are therefore not expected to be identical.

The posted data do not flag a discrepancy between the recalculation and the sponsor analysis. That does not make the two analyses interchangeable. The sponsor's Cox result should be read as the reported time-to-event model, while the recalculation is a transparent, simple comparison of the summary percentages. Both indicate uncertainty around the treatment contrast, but each answers its own statistical question.

15. Multiplicity, Missing Data and Analysis Populations

The registry record does not report whether the trial adjusted for multiple comparisons. This matters because the study describes more than one treatment arm and includes primary and secondary outcomes. When several hypotheses are tested, the chance of at least one apparently notable result can increase if each is interpreted without accounting for the collection of tests. General statistical explanation: multiplicity procedures control or describe this issue by adapting the analysis or interpretation to the number and structure of comparisons. No such procedure can be assumed.

The record also does not provide a full statistical analysis plan, identify all prespecified hypotheses, or describe any interim analyses. The posted primary result is described as an analysis of eligible participants who started treatment. The CHOP-only arm was not assessed for the primary endpoint because of early closure. How that early closure related to any monitoring or decision process is not reported in the registry record.

Missing-data handling is not described in enough detail to determine how unavailable outcome information was managed. The endpoint definition includes last contact, but the record does not state the censoring rules, the amount of incomplete follow-up, or whether sensitivity analyses were performed. The group percentages should not be interpreted as having complete follow-up unless that is independently established; such a statement is not reported in the registry record.

The reported primary population is not simply all enrolled participants: it is described as all eligible patients who started treatment, with the CHOP-only arm excluded from assessment. The registry does not provide a full accounting of screening, eligibility, treatment initiation, exclusions, or participant flow for the primary analysis. Nor does it state whether analyses followed randomized assignment regardless of treatment received. These omissions limit assessment of the analysis population and the potential effects of exclusions.

The absence of these details does not establish that the trial used inadequate methods. It means the registry record does not allow readers to evaluate those methods fully. Conclusions should be restricted to the reported populations, endpoints, and analyses.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: second primary (2.3%), hematologic-other (1.1%), febrile neutropenia (0.8%), infection w/o 3-4 neutropenia (0.8%), platelet transfusion (0.4%).

The serious adverse-event figures report 3 affected participants among 263 at risk for CHOP plus rituximab and 16 among 263 at risk for CHOP plus tositumomab. For the CHOP-only group, the record gives 0 affected among 13 at risk. The CHOP-only group is much smaller in the posted safety data, so its count should not be treated as directly comparable in precision with the counts for the combination groups.

The registry also reports other adverse events affecting 262 of 263 at risk in the CHOP plus rituximab group, 261 of 263 in the CHOP plus tositumomab group, and 13 of 13 in the CHOP-only group. These are counts of participants with reported events, not measures of event severity or duration by themselves. The listed serious-event terms include second primary at 2.3%, hematologic-other at 1.1%, febrile neutropenia at 0.8%, infection without grade 3-4 neutropenia at 0.8%, platelet transfusion at 0.4%, and LVEF decrease/CHF at 0.4%.

These registry counts do not alone establish comparative safety. The record does not provide exposure-adjusted rates or a full account of follow-up by group. Deaths are not reported in the registry record. Interpretation should account for the small CHOP-only safety group and the absence of additional detail about event timing and exposure.

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

The posted primary endpoint estimates describe progression-free survival at 2 years for the two assessed combination regimens. The sponsor-reported Cox result and the unadjusted risk-difference recalculation both leave substantial uncertainty around the comparison. The available information does not establish a clear difference between CHOP plus rituximab and CHOP plus tositumomab on this endpoint.

This is not evidence that the regimens are identical. A result that does not establish a difference may reflect uncertainty about the effect as well as the possibility that the treatments have similar outcomes. The confidence intervals allow a range of effects, and the registry does not report an equivalence or noninferiority margin. It would therefore be incorrect to interpret the result as proving equivalence or as ruling out effects that may matter.

The conclusion is also limited to the population and analysis actually reported. The primary comparison includes the two combination arms, while the CHOP-only arm was not assessed for the endpoint because of early closure. The registry's primary population includes eligible patients who started treatment. It does not give enough detail to extend the result to every enrolled participant or to all patients with lymphoma.

The endpoint is progression-free survival, not overall survival, quality of life, or a comprehensive measure of treatment burden. The registry lists objective response and adverse events as secondary outcomes, but those outcomes answer different questions and should not be collapsed into a single judgment about benefit or harm. The posted safety counts also require caution, particularly because the CHOP-only group is small in the reported safety data and details on exposure and event timing are not available.

In short, the registry record supports a description of the observed outcomes and their uncertainty, not a clinical recommendation. It does not establish that one regimen works better or that either regimen does not work. The result is best understood as an imprecise comparative estimate with important limits in the publicly reported analysis details.

18. Why This Trial Matters Statistically

This trial illustrates why the choice of statistical measure matters. The primary outcome is progression-free survival at a fixed time, but the sponsor reports a Cox regression hazard ratio that uses time-to-event information. The unadjusted recalculation expresses an absolute difference in the posted percentages. Each measure addresses a related but distinct aspect of the comparison, and neither should be substituted for the other.

It also demonstrates the role of uncertainty. A point estimate alone can appear decisive, but confidence intervals reveal how much uncertainty remains. Here, the intervals for the risk difference and hazard ratio include their respective no-difference values. That does not prove the groups are equivalent; it shows that the data do not precisely resolve the contrast under those analyses. Distinguishing lack of clear evidence from proof of no effect is central to sound interpretation.

The early closure of the CHOP-only arm is another important design feature. Although that arm is part of the trial description and has safety information posted, it is not included in the primary efficacy comparison. Readers must keep the trial's overall structure separate from the population represented in a particular outcome analysis.

Finally, the record illustrates the limits of registry reporting. The outcome and broad analysis method are available, but key details such as missing-data handling, model specification, multiplicity control, and original sample-size assumptions are not. Statistical interpretation depends not only on estimates and p-values, but also on knowing the population, methods, and uncertainty behind them. When those details are absent, conclusions should remain correspondingly narrow.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The CHOP-only arm was not assessed for the primary endpoint because of early closure. This means the primary efficacy result does not compare all trial regimens, despite the trial description including that arm. The registry does not report the reason for early closure or how it affected the timing or conduct of other analyses.

The primary population is described as eligible patients who started treatment, but the record does not provide a complete participant-flow account or explain how exclusions and missing outcome information were handled. The percentages are posted without group-specific confidence intervals, and the registry does not specify the procedure used to produce those fixed-time estimates. The Cox result is reported without reference-group coding, covariates, or information about model assumption checks.

The record does not report a full statistical analysis plan, multiplicity adjustment, interim analysis details, or the trial's original sample-size assumptions. These omissions prevent a complete assessment of whether the reported analyses followed prespecified methods and how the study was planned to address its hypotheses.

The primary outcome is limited to progression-free survival. Other relevant outcomes, including overall survival and quality of life, are not presented in the posted results. The safety record provides counts, but does not report exposure-adjusted rates, timing, duration, or deaths. The CHOP-only safety group has only 13 participants at risk in the posted data, limiting what its counts can say about that arm.

These are limitations of what the registry record reports, not proof that the trial did not collect or analyze additional information. They constrain the conclusions that can be drawn from the posted record alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: S0016 Combination Chemotherapy With Monoclonal Antibody Therapy in Newly Diagnosed Non-Hodgkin's Lymphoma — NCT00006721.
• Registry record last updated 2026-03-16; status last verified 2026-02.
• Results first posted 2013-02-26.
• 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

The registry describes a completed randomized phase III parallel-group trial with no masking reported, sponsored by SWOG Cancer Research Network. It lists 571 enrolled participants and compares CHOP with antibody combinations in newly diagnosed lymphoma; the official title specifies newly diagnosed follicular non-Hodgkin's lymphoma.

The primary endpoint is progression-free survival at 2 years. The posted comparison reports 76.0% for CHOP plus rituximab and 80.0% for CHOP plus tositumomab. The sponsor's Cox regression gives a hazard ratio of 0.79, with a 95% confidence interval from 0.6 to 1.05 and p = 0.11. An unadjusted recalculation of the fixed-time comparison gives a risk difference of 0.040, with a 95% confidence interval from -0.031 to 0.11 and p = 0.269.

The CHOP-only arm was not assessed for the primary endpoint due to early closure. The record does not report several analysis details, including missing-data handling, multiplicity procedures, and original sample-size assumptions. The results therefore describe an uncertain comparison between the two assessed combination regimens, rather than establishing a clear difference or equivalence.