Efficacy of 400 Mg Versus 800 Mg Imatinib in Chronic Myeloid Leukemia in Chronic Phase Patients - TOPS (Tyrosine Kinase Inhibitor Optimization and Selectivity): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of 400 Mg Versus 800 Mg Imatinib in Chronic Myeloid Leukemia in Chronic Phase Patients - TOPS (Tyrosine Kinase Inhibitor Optimization and Selectivity) (NCT00124748), a randomized phase 3 study. The primary outcome is Percentage of Participants With Major Molecular Response (MMR) Rates at 12 Months.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, open-label, parallel-group study of imatinib 400 mg versus 800 mg in people with newly diagnosed, previously untreated chronic myeloid leukemia in chronic phase. The registry reports enrollment of 476 participants. Randomization assigned participants to one of the two dose groups; masking was listed as none, so the study was open-label.

The 400 mg group received an oral dose once daily. The 800 mg group was to receive 400 mg orally twice daily, in the morning and evening. Both groups started their assigned dose on Day 0. Dose changes were possible under protocol guidelines: the 400 mg dose could be interrupted, reduced, or escalated, while the 800 mg dose could be interrupted or reduced. The registry does not report how often these changes occurred.

The primary outcome was the percentage of participants with a major molecular response at 12 months. The primary population was intent-to-treat, described as all randomized patients. The registry lists secondary outcomes covering response measures at later time points, molecular transcript status, time to first responses, and event-free survival. The posted record is marked terminated. The reason for termination and the trial’s original statistical design assumptions are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat (ITT) population consisted of all patients who are randomized into the 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 Percentage of Participants With Major Molecular Response (MMR) Rates at 12 Months, measured over 12 months.

The primary endpoint is the percentage of participants with a major molecular response at 12 months. The registry defines major molecular response as a Bcr-Abl transcript ratio of ≤0.1%, corresponding to a ≥ 3 log reduction of BCR-ABL transcripts from a standardized baseline. The measurement was made using reverse transcriptase polymerase chain reaction performed centrally.

This endpoint records whether each participant met the response definition by the specified time point, then expresses the result as a percentage of the analyzed population. Molecular response is used in this trial as a measure of change in the disease-associated transcript. It differs from outcomes that count symptoms, survival, or time until an event: the primary endpoint is a defined laboratory response at a particular follow-up time.

As a binary outcome, the endpoint does not describe how far beyond the response threshold a participant’s transcript level may have changed, nor does it alone describe the course of response over time. The registry’s definition and ITT population clarify what was counted, but the record does not provide further details about handling participants without a 12-month measurement.

7. Results Posted in the Registry

• Imatinib 800 mg: 45.1% (144/319) versus 38.9% (61/157) on control; risk difference 6.3 percentage points (95% CI -3.1 to 15.7), risk ratio 1.16 (0.92–1.46); unadjusted two-proportion z-test p = 0.193.

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.

Mean Actual Dose Intensity Per Day

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Imatinib 800 mg: mean 643.30 versus 399.30 on control; difference 244.00 (95% CI 222.14 to 265.86), standardized effect d = 1.76; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Imatinib 800 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Imatinib Pharmacokinetic Trough Plasma Concentration (Cmin) at Month 12

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Imatinib 800 mg: mean 739.58 versus 1,458 on control; difference -718.62 (95% CI -1,241 to -196.37), standardized effect d = -0.42; z-test p = 0.007.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Imatinib 800 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The listed secondary outcomes span several kinds of questions: molecular response percentages at later time points, cytogenetic and hematological response, undetectable Bcr-Abl transcripts, time to first response, and event-free survival. The record also gives detailed results for actual dose intensity and a pharmacokinetic concentration measure. These outcomes add context about response over time and treatment exposure, but they are not substitutes for the prespecified primary comparison at 12 months.

The two detailed secondary comparisons concern treatment exposure rather than the primary molecular response. Actual dose intensity differs between assigned groups, while the pharmacokinetic concentration result is a separate exposure measure. Such outcomes may help describe how the assigned regimens were delivered and measured, but they do not by themselves establish that a difference in exposure explains the primary endpoint result.

The registry lists 18 secondary outcomes, but detailed quantitative results are not reported in the registry record for most of them. As a result, the overall pattern across all secondary response and time-to-event outcomes cannot be assessed from the information recorded. The primary result remains the clearest quantitative efficacy comparison available in the record.

In general, multiple outcomes create more opportunities for apparently notable findings to arise by chance, particularly when analyses are unadjusted. The registry reports no analyses in the listed secondary-outcome records and does not describe a multiplicity strategy. Secondary results therefore carry less confirmatory weight than a clearly specified primary comparison, especially where methods, estimates, and uncertainty are not fully reported.

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

• Percentage of Participants With Major Molecular Response (MMR) Rates at 24, 36, and 42 Months
• Percentage of Participants With Complete Cytogenetic Response (CCyR) Rate at 12, 24, 36, 42 Months
• Percentage of Participants With Complete Hematological Response (CHR) Rates at 12, 24, 36, and 42 Months
• Percentage of Patients With Undetectable Levels of Bcr-Abl (A Fusion Gene of the Breakpoint Cluster Region [Bcr] Gene and Abelson Proto-oncogene [Abl] Genes) Transcripts
• Time to First Major Molecular Response
• Time to First Complete Cytogenetic Response
• Time to First Complete Hematological Response (CHR)]
• Estimated Rate of Event Free Survival (EFS) in Two Treatment Arms

9. Statistical Interpretation

10. Statistical Methods Explained

The primary comparison is between two randomized groups and a binary endpoint: whether a participant met the major molecular response definition by 12 months. For this kind of outcome, a risk or response percentage describes the proportion meeting the definition in each group. The risk difference expresses the absolute contrast between those proportions. The posted comparison uses a risk difference of 0.063, with the 400 mg group as the reference and the 800 mg group as the compared group.

What does the confidence interval describe?

In general, a confidence interval gives a range of values compatible with the observed data and the statistical method used. It communicates uncertainty in the estimated group contrast, rather than a range containing a known true value with certainty. For the primary comparison, the interval runs from -0.031 to 0.157. Because it includes zero, the posted data do not rule out no difference under this interval-based approach. The interval also includes values on either side of zero, so the data are compatible with different directions of the group contrast.

What does the p-value add?

The registry comparison gives an unadjusted p-value of 0.193. In general, a p-value is calculated under a specified null model and describes how often results at least as extreme as the observed result would occur if that model were true and the test assumptions held. It does not give the probability that the null hypothesis is true, and it is not a measure of effect size. It also does not account for multiple testing unless a multiplicity procedure is explicitly included. The registry does not report an adjustment for multiple comparisons.

Why distinguish the risk difference from relative measures?

The risk difference compares response percentages on an absolute scale. Relative measures instead compare one group’s response with the other group’s response as a ratio. The posted record includes a risk ratio of 1.162 with an interval from 0.923 to 1.463, and an odds ratio of 1.295 with an interval from 0.877 to 1.911. These measures express the same group comparison on different scales; they are not interchangeable in interpretation. Both intervals include the value corresponding to no group contrast for their respective scale. The absolute difference is often easier to relate directly to the observed percentages, while relative measures can emphasize proportional change.

What does randomization contribute?

In general, random allocation is intended to make the assigned groups comparable on average at the start of a trial, including for factors that were not measured. It therefore supports interpretation of a between-group difference as an effect of assignment, provided the trial’s conduct and analysis preserve the randomized comparison. Here, the registry identifies the study as randomized and lists the primary population as all randomized participants. It does not give details of the randomization sequence or concealment process, or explain how missing primary outcomes were handled.

How should the posted recalculations be read?

The comparison statistics are unadjusted estimates based on the posted group summaries. They describe the data without adjustment for additional covariates. General statistical practice distinguishes such simple comparisons from a prespecified model that might account for baseline factors or trial design features. No adjusted primary analysis is reported in the registry record. The listed unadjusted comparison should therefore be interpreted as the comparison actually available in the record, not as evidence that no other analysis was conducted.

Why do outcome type and analysis population matter?

The primary outcome is binary and measured at a fixed time point, so it does not use the timing of response or later changes in response status. The registry lists time-to-response and event-free survival as secondary outcomes, but those analyses require methods that account for event timing and censoring. The registry record does not provide their detailed results or analysis methods. The stated ITT population includes all randomized patients, but the record does not explain whether the posted endpoint counts include every randomized participant or how unavailable measurements were classified.

These distinctions matter because a numerical estimate depends on the outcome definition, population, and analysis method. General statistical interpretation should keep those features visible: a response proportion at 12 months answers a different question from a time-to-response analysis, and an unadjusted comparison does not automatically answer questions about adjusted effects or other time points.

11. Confidence Intervals and Precision

The primary risk-difference estimate is 0.063, and its 95% confidence interval extends from -0.031 to 0.157. The interval is relatively broad around the point estimate and crosses zero. In practical statistical terms, the observed data do not pin down the size or direction of the group difference precisely: values consistent with the interval include a difference favoring either group as well as no difference.

The 400 mg group had 61 responses among 157 participants, and the 800 mg group had 144 among 319. The registry also reports the corresponding percentages as 38.9% and 45.1%. The larger number of participants in the 800 mg group does not by itself eliminate uncertainty, and the confidence interval is the direct summary of uncertainty for the stated contrast.

The relative-effect intervals give another view of precision. The risk ratio is 1.162, with an interval from 0.923 to 1.463; the odds ratio is 1.295, with an interval from 0.877 to 1.911. Each interval includes its no-association value, consistent with the uncertainty evident in the risk-difference interval. The measures use different scales and should not be read as interchangeable.

Confidence intervals describe statistical uncertainty under the method used. They do not account for every possible source of uncertainty, such as incomplete reporting or unreported analysis choices. The registry does not report missing-data handling or a detailed primary analysis plan, so the interval should be interpreted in the context of those information gaps.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations show participants per arm for the binary primary outcome, not total events. They are planning calculations based on the observed effect, with the assumed effect reduced to 100%, 75%, or 50% of that observed effect. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

For each assumed effect, the 80% power column gives the estimated participants per arm for a design with an 80% chance of detecting that specified effect under the planning assumptions, if the effect is real and the assumptions hold. The 90% power column describes the corresponding planning target for a 90% chance. These are prospective design illustrations, not a measure of the probability that the trial’s observed result is correct. The sample sizes are not observed power.

The illustrations require more participants per arm when the target effect is smaller: detecting 75% or 50% of the observed effect requires larger planned groups than detecting the full observed effect. This follows from a general statistical principle: smaller differences are harder to distinguish from random variation, so a study usually needs more information to detect them with the same power.

These calculations should not be treated as the actual enrollment target or as evidence about why the trial ended. They use the observed comparison as a planning reference and assume equal allocation, whereas the posted primary groups contain different numbers of participants. The registry does not report whether the original trial planning used these assumptions or whether any other planning procedure was used.

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 = 0.389; treatment proportion p₁ = 0.451; average p̄ = 0.420.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6980 = 1.3680.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6966 = 0.5863.
4. n = (1.3680 + 0.5863)² / (0.0629)² = 966.1, rounded up to 967 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 <- 0.388535031847134                          # control proportion
p1 <- 0.45141065830721                          # 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(124748)
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 = 967 (80% power), 1,293 (90% power)
 75% of observed effect: n per arm = 1,710 (80% power), 2,288 (90% power)
 50% of observed effect: n per arm = 3,824 (80% power), 5,119 (90% power)
power.prop.test: 966.1 
Simulated power at n = 967 per arm: 0.798 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary endpoint and lists numerous secondary outcomes, but a detailed statistical analysis plan is not reported in the registry record. In particular, the record does not describe whether the primary analysis was adjusted for covariates, whether there were planned subgroup analyses, or whether an analysis hierarchy governed the secondary outcomes.

Multiplicity refers, in general, to the increased chance of a false-positive finding when many hypotheses are tested. A prespecified procedure can control that risk, for example by adjusting the significance threshold or defining an ordered testing strategy. The registry does not report whether any such procedure was used. The unadjusted p-value for the primary comparison should therefore be understood as unadjusted, and results across multiple outcomes should not be treated as a set of independent confirmatory tests.

The primary population is defined as all randomized participants, which is consistent with an intent-to-treat approach. However, the record does not describe how many randomized participants had an available 12-month measurement, how missing measurements were handled, or whether any participants were excluded from the posted endpoint calculation. These omissions matter because missing outcome data can affect the interpretation of a response proportion.

The registry also does not report details of interim analyses, stopping rules, or whether the termination status was related to efficacy, safety, recruitment, or another reason. No such explanation should be inferred from the status alone. Without these details, readers cannot determine whether the displayed analyses reflect a prespecified plan or what role any unreported analyses may have had. These gaps limit assessment of multiplicity, missing data, and the completeness of the analysis record.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pyrexia (4.1%), neutropenia (3.8%), thrombocytopenia (2.8%), anaemia (1.9%), febrile neutropenia (1.6%).

The serious adverse event summaries use different numbers at risk in the two groups: 157 for imatinib 400 mg and 316 for imatinib 800 mg. The record reports 42 participants affected in the 400 mg group and 121 in the 800 mg group. These are counts of participants with serious adverse events, not counts of distinct event types or a comparison adjusted for time under observation.

The most frequently listed serious adverse event terms include pyrexia at 4.1%, neutropenia at 3.8%, thrombocytopenia at 2.8%, anaemia at 1.9%, febrile neutropenia at 1.6%, and appendicitis at 1.6%. These term-specific percentages should be read as descriptive registry figures. The record does not establish whether participants could contribute to more than one term, so the percentages should not be added together.

The registry does not report death counts in these summaries. The serious adverse event counts alone do not establish a causal difference between assigned groups or fully describe safety, and the record’s group sizes and follow-up details should be considered when interpreting them.

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

The posted primary result describes a higher observed percentage with major molecular response in the 800 mg group than in the 400 mg group at 12 months. The estimated risk difference is 0.063, but its confidence interval includes zero and extends in both directions. The result therefore does not precisely establish the size or direction of the underlying group difference.

This finding does not show that the regimens are equivalent. An interval that includes zero can reflect uncertainty and does not prove the absence of a meaningful difference. Nor does the point estimate alone show that the higher dose caused a higher response percentage. The trial was randomized, which supports a comparison of assignment groups, but interpretation still depends on conduct, outcome availability, and analysis details that are not fully reported in the registry record.

The endpoint is a laboratory-defined response at one time point. It does not, by itself, establish the duration of response, survival, quality of life, or the balance of benefits and harms. The registry lists related secondary outcomes, but most of their detailed results are not reported in the registry record. The serious adverse event summaries provide descriptive counts and selected terms, not a complete account of all possible safety considerations.

The findings apply to the population and treatment regimens described in the record: newly diagnosed, previously untreated chronic-phase chronic myeloid leukemia and the assigned imatinib doses. They should not be generalized automatically to other populations, regimens, or outcomes. The record’s terminated status does not explain why the study ended, and no reason should be inferred.

In general, a confidence interval and endpoint definition help clarify what the data can support. Here, the main statistical message is an uncertain estimate of the 12-month molecular response contrast, alongside limited registry detail about analysis choices, missing data, and most secondary outcomes.

17. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison should be read through its effect estimate and uncertainty, not only through the direction of the observed percentages. The two response percentages differ descriptively, while the risk-difference interval spans values on both sides of no difference. That combination makes the precision of the estimate central to interpretation.

It also shows how endpoint choice shapes the question. A binary response measure at 12 months addresses whether a defined molecular threshold was met by that time. It does not answer the same question as a time-to-response endpoint, a later response percentage, or event-free survival. The registry lists these related outcomes, but their detailed results are not generally available in the record.

The contrast between randomized assignment and actual dose intensity is another useful statistical distinction. Assignment is the basis of the randomized comparison; actual delivered dose is a post-assignment measure that can be affected by dose interruption, reduction, or escalation. Descriptive dose-intensity results therefore provide context, but they do not replace the randomized comparison of assigned groups.

Finally, the record demonstrates why completeness of statistical reporting matters. Missing detail about multiplicity, missing outcomes, the original planning assumptions, and the reason for termination limits evaluation of how robustly the observed results answer the trial’s question. These are not reasons to dismiss the posted estimates, but they define the boundaries of what can be concluded from the registry record.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record is marked terminated, but the reason for termination is not reported in the registry record. The record does not explain whether the status affected recruitment, follow-up, or the availability of outcome data. The registered enrollment is 476, while the primary comparison lists 157 participants in one group and 319 in the other; the record does not provide further detail about this distribution or participant flow.

The primary endpoint population is described as all randomized patients, but the record does not explain how missing 12-month measurements were handled or whether every randomized participant contributed an observed endpoint value. It also does not report the original sample-size justification, interim analyses, stopping rules, or a detailed statistical analysis plan. These gaps prevent a full assessment of whether the posted analyses followed prespecified procedures.

The trial was open-label. The primary endpoint used a centrally performed molecular assay, but the registry does not report other safeguards against potential bias or provide detailed information about assessment procedures. Dose adjustments were permitted under protocol guidelines, yet the frequency and reasons for changes are not reported. Such changes may matter when distinguishing assigned dose from delivered dose.

Most listed secondary outcomes lack detailed quantitative results in the registry record. The posted safety summaries report selected serious adverse event counts and terms, but do not provide death counts or a complete accounting in the information available. These limitations constrain interpretation of the broader efficacy and safety profile; they do not change the numerical primary comparison itself.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Efficacy of 400 Mg Versus 800 Mg Imatinib in Chronic Myeloid Leukemia in Chronic Phase Patients - TOPS (Tyrosine Kinase Inhibitor Optimization and Selectivity) — NCT00124748.
• Registry record last updated 2012-02-03; status last verified 2012-01.
• Results first posted 2012-02-03.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry describes a Phase 3 randomized, open-label, parallel-group comparison of imatinib 400 mg once daily and 800 mg daily as 400 mg twice daily in newly diagnosed, previously untreated chronic-phase chronic myeloid leukemia. Enrollment was 476, and the record is marked terminated.

The primary endpoint was the percentage with major molecular response at 12 months in the intent-to-treat population. The posted response percentages were 38.9% and 45.1%, respectively. The risk difference was 0.063, with a 95% confidence interval from -0.031 to 0.157 and an unadjusted p-value of 0.193.

The registry lists 18 secondary outcomes, but detailed results are available for only a limited set in the record. The reason for termination, original sample-size assumptions, and several analysis details are not reported in the registry record.