Study of TAK-228 In Patients With Previously Treated Metastatic Renal Cell Carcinoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of TAK-228 In Patients With Previously Treated Metastatic Renal Cell Carcinoma (NCT03097328), a na phase 2 study. The primary outcome is Best Overall Response Rate.

2. Clinical Question

The question is descriptive: it concerns the proportion of participants in the treated group who met the response definition during assessment. It is not a direct comparison between TAK-228 and another treatment or a control condition, because none is listed. Consequently, the study design cannot isolate how much of any observed response might be attributable to TAK-228 rather than other influences.

The primary endpoint uses the best response recorded on treatment. It therefore summarizes whether a participant met the criteria for complete response or partial response at any relevant assessment, rather than describing the response at only one selected visit. The response criteria depend on changes in target and non-target lesions and the absence of new lesions. The endpoint was assessed through imaging every eight weeks (2 cycles). The registry reports a median of 2 cycles administered, with a range of <1-15, and says participants were assessed up to ~14 months.

The registry record also lists median progression-free survival, median overall survival, and the percentage of participants with any Grade 3 or higher treatment-related adverse events as secondary outcomes. Their numerical results are not reported in the registry record.

3. Trial Design

This is a Phase II, single-group study of TAK-228 in patients with previously treated metastatic renal cell carcinoma. The registry record lists enrollment of 39 and describes 38 eligible and treated patients in the primary response analysis. One participant was deemed ineligible and did not receive treatment, and was excluded from that analysis.

The allocation field is not applicable, the study model is single group, and masking is listed as none. No concurrent comparator arm is described. These design features mean that the study can report outcomes in the TAK-228 group, but it does not provide a randomized comparison against another treatment or control. The registry record does not report a protocol for allocation or a blinding procedure; given the single-group, unmasked design, no such comparison or masking is listed.

TAK-228 is described as an oral drug taken weekly for 4 weeks per cycle. Dosage was determined by the study team. The primary response endpoint was evaluated using imaging assessments every eight weeks (2 cycles). The median number of cycles administered was 2 (range <1-15), and the endpoint description says participants were assessed up to ~14 months. These are details of treatment and response assessment; the registry record does not report additional scheduling rules or dose-adjustment criteria.

The study is marked completed. The listed start date is 2017-08-01, and the primary completion date is 2021-07-20. The registry record lists 4 outcome measures, with best overall response rate as the primary outcome and median progression-free survival, median overall survival, and the percentage of participants with any Grade 3 or higher treatment-related adverse events as secondary outcomes.

The registry record does not report the planned sample-size justification, a statistical analysis plan, interim analyses, or methods for handling protocol deviations. Allocation and masking details should not be read as evidence about unreported procedures. The single-group design is the central interpretive feature: observed outcomes describe this treated cohort and do not, by themselves, establish a treatment effect relative to a counterfactual 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.

Before assignment: 1 patient was deemed as ineligible and did not receive protocol treatment. 38 eligible and treated patients were included in the analysis.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Best overall response was assessed in 38 eligible and treated patients. 1 patient who was deemed as ineligible and did not receive treatment was excluded from response analysis.

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.

Population: Baseline data was assessed in 38 eligible and treated patients. 1 patient who was deemed as ineligible and did not receive treatment was excluded from baseline analysis.

6. Primary Endpoint

The registry lists the primary outcome as Best Overall Response Rate, measured over Imaging assessments occurred every eight weeks (2 cycles) for response evaluation. The median number of cycles administered was 2 (range <1-15), thus participants were assessed up to ~14 months..

The primary endpoint is best overall response rate, defined as the percentage of participants whose best response recorded on treatment was complete response (CR) or partial response (PR), according to RECIST 1.1. It is a binary outcome at the participant level: each assessed participant either meets the response definition or does not.

For target lesions, CR requires disappearance of all target lesions, with pathological lymph nodes reduced to a short axis <10 mm. PR requires at least a 30% decrease in the sum of target-lesion diameters, using the baseline sum as the reference. For non-target lesions, CR requires disappearance of all non-target lesions, normalization of tumor marker level, and lymph nodes that are non-pathological in size (<10 mm short axis). The endpoint description also specifies a non-CR/non-progressive disease category for persistent non-target lesions and/or tumor marker levels above normal limits, with PR in the target lesion. Neither CR nor PR can include new lesions.

As a general statistical concept, a response proportion provides a compact summary of how many people in an assessed group met a specified clinical and imaging definition. It is useful for describing tumor response over the assessment period, but it does not describe how long a response lasts or establish a comparison with another treatment. Those questions require other outcomes or a suitable comparison design.

The registry record specifies a 90% confidence interval for the endpoint, but does not report the response-rate estimate or its interval. Best overall response was assessed in 38 eligible and treated patients; one ineligible patient who did not receive treatment was excluded from response analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a participant-level binary outcome expressed as a percentage: best overall response is achieved when the best recorded on-treatment response is complete response or partial response under RECIST 1.1. This kind of endpoint reduces each participant's assessment history to whether the prespecified response category was reached. It is therefore relatively straightforward to summarize as a proportion, but it does not retain details such as the timing or duration of response.

What does the confidence interval describe?

The registry record specifies a 90% confidence interval for best overall response rate. As a general statistical explanation, a confidence interval describes uncertainty in an estimate under the assumptions of the statistical method. It is not a guarantee that a particular interval contains the true value, nor does it represent the range of individual participant outcomes. The width of an interval depends on the amount of information and the method used. The response estimate, interval bounds, and interval method are not reported in the registry record, so the interval's width and implications cannot be assessed.

What does the single-group design allow?

The study is listed as single group, with allocation not applicable and masking none. A single-group proportion can describe the observed response frequency in the assessed participants. It cannot, by itself, distinguish a treatment-related effect from other explanations, because no concurrent comparator group is listed. In general, randomized comparisons help estimate differences between assigned groups by balancing measured and unmeasured influences in expectation; that comparison is not part of the design recorded for this study.

The absence of masking is also relevant to interpretation of an imaging-based outcome. As a general point, knowledge of treatment assignment can influence some decisions about assessment or reporting. The registry record does not describe a masking process or provide enough detail to characterize how response assessments were reviewed. It would be inappropriate to infer additional assessment safeguards or their absence beyond the recorded masking status.

How do the response criteria work?

The response definition follows lesion-based rules. A complete response requires disappearance of target and non-target lesions, with specified lymph-node size and tumor-marker conditions. A partial response for target lesions requires at least a 30% decrease in the sum of diameters relative to baseline. The endpoint description requires that there be no new lesions for CR or PR. These criteria make the outcome more explicit than an unstructured judgment, but their application depends on assessment information being available and interpreted according to the criteria.

Imaging assessments occurred every eight weeks (2 cycles) for response evaluation. The median number of cycles administered was 2 (range <1-15), and the record says participants were assessed up to ~14 months. As a general concept, assessment frequency and treatment exposure can affect which responses are observed: a participant must remain assessable long enough for a response to be detected, while best response can summarize multiple assessments. The registry record does not describe how early treatment discontinuation, missed imaging, or unevaluable scans were handled.

What can planning calculations tell readers?

As a general statistical concept, sample-size calculations are planning tools that describe the amount of information needed under specified assumptions. They are not a measure of the probability that a completed study found an effect, and they are not observed power. The planning assumptions stated for this record use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. However, this trial is single group, and the record does not include the numerical planning outputs or the observed primary response estimate. Those planning assumptions should not be mistaken for the design of this trial or for a result from it.

More generally, recalculations from posted summary statistics can differ from sponsor analyses when they use unadjusted methods or different assumptions. No recalculated primary response result is reported in the registry record, and no discrepancy is specified. No independently calculated estimate should be substituted for a sponsor analysis if the two differ; the sponsor's own analysis takes precedence when a discrepancy is identified.

Why do analysis populations matter?

The record states that 38 eligible and treated patients were assessed for best overall response and that one ineligible, untreated participant was excluded. This is a clear description of the primary analysis population at that level. As a general principle, results can differ depending on whether analyses include everyone enrolled, only those treated, or only participants with evaluable assessments. The record does not report a broader intention-to-treat analysis, an evaluable-patient rule beyond the stated exclusion, or methods for missing outcomes. Therefore, the precise relationship between the enrolled population and any other outcome-specific populations is not established.

The registry record lists no statistical analyses. It does not report the calculation method for the response proportion, interval construction, hypothesis tests, covariate adjustment, or a multiplicity strategy. These omissions limit reproducibility and prevent assessment of whether planned and reported methods aligned. They do not show that a particular method was or was not used outside the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("TAK-228")
n      <- c(38)
events <- c(2)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  TAK-228  38  2  5.3  0.6  17.7  1.5  17.3  1.6  17.3

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 4 outcome measures: best overall response rate as the primary outcome, plus median progression-free survival, median overall survival, and the percentage of participants with any Grade 3 or higher treatment-related adverse events as secondary outcomes. It lists no statistical analyses. The results for these outcomes are not reported in the registry record.

Multiplicity arises when several outcomes or statistical tests are considered together. As a general statistical concept, examining many endpoints can increase the chance that at least one result appears notable by chance, depending on the analysis plan and decision rules. The registry record does not report whether outcomes were tested formally, how any multiple testing was handled, or whether a hierarchy among secondary outcomes was specified. No multiplicity adjustment should be assumed.

The primary response population is described as 38 eligible and treated patients; one participant deemed ineligible and not treated was excluded. This does not fully document missing-data handling. The registry record does not report how unevaluable scans, missed visits, early discontinuation, or incomplete outcome data were classified or analyzed. It also does not describe imputation, sensitivity analyses, or rules for censoring time-to-event outcomes.

The record therefore supplies a specific primary analysis population and the exclusion of one untreated, ineligible participant, but not a complete account of analysis populations across all outcomes. It is not reported in the registry record whether the secondary outcomes used the same population. The lack of an analysis description limits assessment of selective reporting and of how missing observations may affect interpretation.

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: nausea (5.3%), fatigue (5.3%), dehydration (5.3%), hyperglycemia (5.3%), myocardial infarction (2.6%).

The safety record lists one TAK-228 group. For serious adverse events, 38 participants were at risk and 12 were affected. A separate other-event category lists 38 at risk and 38 affected. The record also lists 15 deaths. These are distinct entries in the safety summary and should not be added together: the categories may not represent mutually exclusive participants, and the registry record does not explain their overlap.

The listed serious-event terms and percentages are Nausea, 5.3%; Fatigue, 5.3%; Dehydration, 5.3%; Hyperglycemia, 5.3%; Myocardial infarction, 2.6%; and Mucositis oral, 2.6%. These are the terms and percentages recorded; the registry record does not further explain the denominator or whether these are the only serious-event terms.

As a general statistical point, an affected-participant count describes how many people were recorded with an event in the stated group; it is not the same as a count of event episodes. A percentage for an individual term should be interpreted according to its reported definition and denominator. Serious adverse events should also not be equated with Grade 3 or higher treatment-related adverse events: those are separately listed concepts, and the result for the Grade 3 or higher secondary outcome is not reported in the registry record.

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

The registry record describes a Phase II, single-group study of TAK-228 and defines best overall response using RECIST 1.1. It also identifies the response-assessment schedule and the population assessed: 38 eligible and treated patients, excluding one ineligible participant who did not receive treatment. These registry entries clarify what the primary endpoint was intended to measure and who was included in its stated assessment.

The record does not report the observed response rate, the number achieving complete or partial response, or the 90% confidence interval. It is therefore not possible to state how often the endpoint was met or how precisely the rate was estimated from the registry record. This absence is not a result of zero responses; it is an absence of the numerical result in the registry record.

Even a reported response proportion in this study would describe the treated group rather than establish a difference attributable to TAK-228. The study has no listed comparator, and its allocation is not applicable. Without a concurrent comparison, the record cannot show how the observed outcomes compare with outcomes under another treatment or control condition.

The endpoint also has a specific scope. Best overall response captures whether a participant met CR or PR criteria during on-treatment assessment. It does not by itself summarize response duration, progression-free survival, overall survival, or the balance of benefit and harm. Those are separate outcomes. Although the record lists median progression-free survival, median overall survival, and a Grade 3 or higher treatment-related adverse-event outcome, their numerical results are not reported in the registry record.

Safety entries provide information on recorded serious and other adverse events and deaths, but should be interpreted separately from efficacy. The safety record lists 12 participants affected by serious adverse events among 38 at risk and 15 deaths. The record does not provide the additional detail needed to infer the cause of deaths, event timing, or overlap among adverse-event categories.

In short, the registry record supports interpretation of the design and endpoint definition, but not a numerical judgment about response or comparative treatment effects. A missing registry result should not be taken to mean a positive result, a negative result, or no effect.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how a clearly defined endpoint and a single-group design answer different questions. Best overall response is defined through RECIST 1.1 criteria and expressed as a percentage of participants. The record specifies a 90% confidence interval, signaling that uncertainty around the response proportion was intended to be reported. Yet the registry record contains no response estimate or interval, so readers cannot assess the observed frequency or its precision.

The study's single-group structure is central to what can be inferred. A response proportion can describe outcomes among treated participants, but a control group is needed for a direct estimate of the difference between treatment conditions. Because no comparator is listed, there is no randomized treatment contrast in this record. This distinction remains important even when an endpoint is objectively defined.

The description of the assessment population is also consequential. The registry identifies 38 eligible and treated patients for response assessment and states that one ineligible, untreated participant was excluded. Reporting who enters an analysis helps readers understand the denominator, but full interpretation also depends on knowing how missing or unevaluable assessments were addressed. Those methods are not reported in the registry record.

The record lists several secondary outcomes alongside response. Considering multiple outcomes can broaden the clinical picture, but requires transparent reporting of analytic priorities and multiplicity handling. Neither a statistical analysis plan nor a multiplicity strategy is reported. In addition, the listed safety counts and terms should not be treated as substitutes for the unreported Grade 3 or higher treatment-related adverse-event percentage.

The key statistical lesson is that endpoint definitions, analysis populations, uncertainty measures, comparison structure, and missing-data rules all shape interpretation. Here, some design and endpoint details are available, while the principal numerical efficacy result and several analysis details are not. Keeping those boundaries clear avoids treating a design description as evidence of a particular outcome.

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 main limitation for interpreting efficacy is that the registry record does not report the primary best overall response rate, its participant count, or the specified 90% confidence interval. Without those values, the response frequency and its precision cannot be evaluated. Results for the listed secondary outcomes are also not reported in the registry record.

The single-group design provides no concurrent comparator. Consequently, any outcome observed in the TAK-228 group could not be directly contrasted with an outcome under another treatment within this study. The registry lists no allocation to multiple groups and no masking.

The primary response analysis includes 38 eligible and treated patients, with one ineligible, untreated participant excluded. The registry record does not report how missing imaging, unevaluable scans, early treatment discontinuation, or other incomplete assessments were handled. It also does not report whether the secondary outcomes used the same analysis population.

The assessment schedule is described as imaging every eight weeks (2 cycles), with a median of 2 cycles administered (range <1-15) and assessment up to ~14 months. The record does not provide further information about response-assessment review, adjudication, or rules for interpreting unavailable assessments. Those details could matter for consistent classification of best response.

The registry record does not report a statistical analysis plan, sample-size justification, interim analyses, or methods for controlling multiplicity. It lists no statistical analyses. These are limits on what can be independently evaluated from the record, not proof that analyses or procedures were absent outside it.

For safety, the record lists affected participants and event terms but does not explain overlap across categories, event timing, or attribution beyond the label of the listed secondary outcome. The result for any Grade 3 or higher treatment-related adverse events is not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of TAK-228 In Patients With Previously Treated Metastatic Renal Cell Carcinoma — NCT03097328.
• Registry record last updated 2024-04-29; status last verified 2024-04.
• Results first posted 2022-02-24.
• 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 registry record describes a completed Phase II, single-group study of TAK-228 in previously treated metastatic renal cell carcinoma. Enrollment is listed as 39. The intervention was oral TAK-228, taken weekly for 4 weeks per cycle, with dosage determined by the study team. Allocation is not applicable and masking is none.

The primary outcome was best overall response rate under RECIST 1.1, assessed in 38 eligible and treated patients. One ineligible participant who did not receive treatment was excluded from response analysis. Imaging assessments occurred every eight weeks (2 cycles); the median number of cycles administered was 2 (range <1-15), with assessment up to ~14 months.

Secondary outcomes listed are median progression-free survival, median overall survival, and the percentage of participants with any Grade 3 or higher treatment-related adverse events. The numerical results for the primary and secondary outcomes are not reported in the registry record. The record lists 12 participants affected by serious adverse events among 38 at risk, 38 affected in a separate other-event category among 38 at risk, and 15 deaths.