Modular Study to Evaluate CT7001 Alone in Cancer Patients With Advanced Malignancies: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Modular Study to Evaluate CT7001 Alone in Cancer Patients With Advanced Malignancies (NCT03363893), a randomized phase 1, phase 2 study. The primary outcome is Treatment-Emergent Adverse Events and Laboratory Abnormalities (Safety and Tolerability).

2. Clinical Question

The central question was whether CT7001 could be administered and studied in participants with advanced malignancies, with safety and tolerability assessed across several study modules. The registry describes CT7001 as monotherapy in advanced solid tumours and in groups with triple-negative breast cancer and castrate resistant prostate cancer. It also describes CT7001 with fulvestrant in hormone receptor-positive and HER2-negative breast cancer, and a food-versus-fasting crossover module in advanced solid tumours.

The study was modular rather than one uniform comparison. Dose-escalation cohorts, disease-specific expansion cohorts, combination treatment and the food-effect module address different questions and involve different groups. Accordingly, the cohort results should be read as descriptions within each group, not as if every participant had been randomized between the same two treatment options. The record reports no posted inferential analysis comparing the modules or establishing a common treatment effect across them.

3. Trial Design

NCT03363893 is listed as a completed Phase I/II study sponsored by Carrick Therapeutics Limited. The record gives a total enrollment of 124 and identifies advanced solid malignancies as the condition. The study began on 2017-11-14, and its primary completion date is recorded as 2022-12-15. It is described as multicentre, open-label and modular, with multiple parts and arms.

The first module included ascending-dose monotherapy cohorts and paired-biopsy breast cancer expansion cohorts. The descriptions state that CT7001 was used to explore dose levels and to identify a maximum tolerated dose, minimally biologically active dose and recommended dose for Phase II testing. The registry record does not report the final dose-selection conclusions in the information summarized for the primary outcome. Further Module 1 cohorts enrolled participants with triple-negative breast cancer or castrate resistant prostate cancer and used a dose schedule recommended from the earlier part.

Module 2 Part A combined CT7001 with fulvestrant for participants with locally advanced or metastatic hormone receptor-positive, HER2-negative breast cancer. Module 4 enrolled participants with advanced solid tumours in a randomized crossover involving fed and fasted conditions, followed by continuous once-daily dosing. These modules differ in population, treatment regimen and purpose, so cohort counts do not constitute a single set of interchangeable treatment arms.

The registry lists randomized allocation and a parallel model, while the Module 4 description specifically identifies a randomized crossover. Masking is listed as none, making the study open-label. The record does not describe the randomization sequence, allocation concealment, crossover washout details, or the exact statistical plan. It also does not report a single randomized control group used across the full study. Consequently, the overall design labels should be interpreted alongside the distinct module descriptions rather than as evidence that all reported cohorts were directly compared.

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: A total of 174 participants were recruited, of which 124 participants were enrolled and went on to receive study treatment with CT7001. Of the 174 participants screened, 50 participants did not receive study treatment, of which 46 participants did not meet the study enrolment criteria and 4 participants withdrew from the study.

Only the first 6 of 15 groups are shown.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Results are presented for Participants with one or more related TEAEs. Module 4 crossover cohorts are presented in terms of patients fed/fasted correctly in accordance with the protocol. One participant in 120mg OD Fed cohort deviated from the protocol and data for this participant is therefore reported under 120mg Fasted cohort.

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.

Only the first 6 of 16 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Treatment-Emergent Adverse Events and Laboratory Abnormalities (Safety and Tolerability), measured over Screening to end of study (28-35 calendar days after end of treatment). End of treatment at disease progression, unacceptable toxicity or withdrawal of consent. Average time on study 142.6 days (range 9 - 1135 days).

The primary outcome is treatment-emergent adverse events and laboratory abnormalities, recorded as counts of participants. Treatment-emergent adverse events are defined in the registry as adverse events occurring from Cycle 0 Day 1 or Cycle 1 Day 1 of a study module through 28 days after the last CT7001 dose in that module. The stated observation period extends from screening to the end of study, which is described as 28-35 calendar days after the end of treatment. Treatment ends at disease progression, unacceptable toxicity or withdrawal of consent. The record gives an average time on study of 142.6 days, with a range of 9 - 1135 days.

The posted primary results are described as participants with one or more related treatment-emergent adverse events. A participant count is a direct way to describe how many people in a cohort had at least one event meeting the stated definition; it does not show how many events each participant experienced, how severe every event was, or how long an event lasted. Laboratory abnormalities are included in the outcome title, but the record does not provide detailed laboratory-specific counts in the information summarized for these results.

Safety and tolerability are central outcomes in a Phase I/II study that includes dose escalation and several treatment regimens. They help characterize observed adverse events while doses and populations are being investigated. Counts alone, however, do not establish causation or provide a comparative estimate of risk. The registry does not report confidence intervals or a formal comparative analysis for this endpoint. The Module 4 population description also notes that results concern participants who followed fed or fasted instructions correctly, which complicates a uniform reading of its outcome counts.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The reported primary outcome is a binary participant-level measure: a participant either had one or more related treatment-emergent adverse events during the defined period or did not. The registry reports counts, not an event rate model or an adjusted estimate. This is a descriptive summary, and it answers a narrower question than a comparison of treatment strategies: it records how many participants in each cohort met the outcome definition.

What does a participant count tell the reader?

A count is straightforward to interpret when its denominator and outcome definition are clear. It does not distinguish a participant with one event from a participant with several events, because both satisfy the “one or more” criterion. It also does not communicate event duration, severity beyond any separately reported categories, or the timing of events. Different cohorts have different recorded participant counts, and the registry’s safety groups can have at-risk denominators that are not identical to the primary outcome cohort sizes. Denominators should therefore be taken from the relevant cohort and outcome rather than assumed to be common.

Time at risk can also matter in safety interpretation. The record states that treatment ended at progression, unacceptable toxicity or withdrawal of consent and gives an average time on study of 142.6 days, with a range of 9 - 1135 days. Participants observed for different lengths of time have different opportunities for events to be detected. A count does not adjust for this difference. No person-time analysis or adjustment for exposure duration is reported in the registry record.

Why are the cohorts not a common comparison?

The study combines dose-escalation groups, cancer-specific expansions, CT7001 with fulvestrant, and a fed-versus-fasted crossover module. They differ in dose, clinical group and treatment context. Although the overall record lists randomized allocation, a parallel model and no masking, the Module 4 description separately identifies a randomized crossover. Those design features do not make every cohort a randomized comparison against the same control. The registry record does not report an analysis that combines these groups or adjusts comparisons for their differences.

Randomization, in general, is used to make assigned groups comparable on average, while masking can reduce some risks of biased assessment. These are general statistical concepts, not claims about how well this study achieved those aims. In an open-label study, participants and study staff are not masked to treatment assignment. That may be relevant when events are reported or assessed, especially for outcomes that depend on judgment. The registry does not report blinded outcome assessment.

What would uncertainty measures add?

A confidence interval is a general statistical tool for showing the range of effect estimates reasonably compatible with the data under a specified method. It can make clear that a small cohort count may be imprecise, even when the count itself is easy to read. No confidence intervals are reported for the primary outcome counts. No p-values or model-based comparative estimates are posted either. Without those, the cohort descriptions do not quantify uncertainty through a formal inferential method.

A p-value, in general, measures how unusual data at least as incompatible with a specified null model would be if that model were true. It is not the probability that a treatment works or does not work, nor is it a measure of clinical importance. Because no comparative p-values are posted for this endpoint, the record does not provide a formal test of a treatment contrast. Lack of a posted test is not evidence that groups are equivalent.

How should serious events be distinguished?

The primary endpoint concerns related treatment-emergent adverse events, while the separate serious-event summaries list the number of participants at risk and affected by serious adverse events, other adverse events and deaths. Seriousness is a distinct classification from whether an event is considered related to treatment. The record gives counts for these categories, but the presented information does not establish that every serious event was treatment-related. The listed top serious adverse-event terms each have a frequency of 16.7%; the record does not provide further detail in the summarized data about their timing, recurrence or attribution.

General safety analysis may also consider event severity, exposure duration, recurrent events, laboratory changes and treatment discontinuation. Those approaches can address questions that a participant count cannot. The registry record does not report detailed models or analysis methods for these aspects in the information summarized. No inferential analyses are listed for the outcomes, and the record does not describe interim analyses or a statistical analysis plan. These methods cannot be inferred from the study phase or from the fact that the design includes dose escalation.

What do the sample-size illustrations mean?

The planning illustrations are based on the observed effect and are not the study’s design justification or observed power. In general, a planning calculation estimates the sample size needed under specified assumptions to detect a chosen effect with a stated error threshold and power. The assumptions described for the illustrations include a two-sided significance level of 0.05, equal allocation, a normal approximation and a target effect defined as a fraction of the observed effect. They do not describe the actual allocation or a formal analysis of this multi-module study. The registry record does not report its sample-size justification. As no valid two-arm recalculation is available for this binary endpoint, no new comparison-based sample-size conclusion should be drawn from the posted counts.

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("Module 1 PartA-Cohort 1 120mg OD", "Module 1 Part A-Cohort 2 240mg OD", "Module 1 Part A-Cohort 3 480mg OD", "Module 1 Part A-Cohort 4 360mg OD", "Module 1 Part A-Cohort 5 180mg BID", "Module 1 Part A Paired Biopsy Breast Cancer Expans")
n      <- c(6, 7, 6, 6, 8, 5)
events <- c(6, 7, 6, 6, 8, 5)
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  Module 1 PartA-Cohort 1 120mg OD  6  6  100.0  54.1  100.0  61.0  100.0  59.0  99.6
ARM  Module 1 Part A-Cohort 2 240mg OD  7  7  100.0  59.0  100.0  64.6  100.0  63.1  99.7
ARM  Module 1 Part A-Cohort 3 480mg OD  6  6  100.0  54.1  100.0  61.0  100.0  59.0  99.6
ARM  Module 1 Part A-Cohort 4 360mg OD  6  6  100.0  54.1  100.0  61.0  100.0  59.0  99.6
ARM  Module 1 Part A-Cohort 5 180mg BID  8  8  100.0  63.1  100.0  67.6  100.0  66.4  99.7
ARM  Module 1 Part A Paired Biopsy Breast Cancer Expans  5  5  100.0  47.8  100.0  56.6  100.0  54.1  99.6

11. Multiplicity, Missing Data and Analysis Populations

The study contains numerous cohorts, dose levels, cancer groups, treatment contexts and outcomes. The registry lists pharmacokinetic outcomes such as maximum plasma concentration, area under the curve and trough concentrations, as well as anti-tumour outcomes according to RECIST v1.1, including best objective response, progression-free survival and clinical benefit rate. It reports 7 secondary outcomes posted. A collection of outcomes and cohort-specific summaries can raise multiplicity concerns when many hypotheses are tested: the more tests conducted, the greater the chance of at least one apparently notable result by chance. The registry record does not report a multiplicity adjustment, hierarchy of hypotheses or prespecified testing sequence. No inferential analyses are posted for the outcome measures, so the record does not establish how multiplicity would have been handled in any formal testing.

Missing-data handling is also not reported in the registry record. The primary outcome counts do not describe how participants with incomplete follow-up, missing assessments or early treatment discontinuation were handled, beyond the stated observation window and stopping conditions. It is not reported whether missing outcomes were treated as absent, excluded, imputed or handled by another approach. The Module 4 note specifically identifies a participant in the 120mg OD Fed cohort who deviated from protocol and says that participant’s data were handled accordingly, but the summarized record does not provide more detail about the analysis consequence. The Module 4 results are also described as restricted to participants fed or fasted correctly.

Analysis populations are not fully specified in the registry record. The primary outcome description refers to participants with one or more related treatment-emergent adverse events, while the cohort rows provide participant counts and outcome values. The record does not clearly explain whether every enrolled participant, treated participant or another defined population contributed to each endpoint. Since enrollment, endpoint denominators and safety-group at-risk counts may differ by module, those values should not be treated as interchangeable. The record does not report a prespecified approach for protocol deviations, exclusions or analysis sets.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: diarrhoea (16.7%), atypical pneumonia (16.7%), bronchitis (16.7%), malignant neoplasm progression (16.7%), posterior reversible encephalopathy syndrome (16.7%).

The serious-adverse-event figures are cohort-specific participant counts. For each group, the registry separately lists participants at risk and participants affected by serious adverse events, other adverse events and deaths. These categories should not be conflated: a serious adverse event is not necessarily a treatment-related event, and a death recorded in a cohort is not, by itself, evidence that treatment caused the death. The summarized record does not provide attribution for each death.

In the Module 1 and Module 2 groups, the listed numbers of participants with serious adverse events include 1 in the 120mg OD cohort, 4 in the 240mg OD cohort, 1 in the 480mg OD cohort, 3 in the 360mg OD cohort, and 3 in the 180mg BID cohort. The disease-specific expansion counts are 9 for triple-negative breast cancer and 2 for castrate resistant prostate cancer. The Module 2 Part A counts are 2 in the 240mg cohort and 8 in the 360mg cohort. The Module 4 counts are 0 in the fed and fasted groups at 120mg and 360mg, 0 in the 240mg continuous-dosing group, and 4 in the 360mg continuous-dosing group. Their corresponding at-risk denominators differ by cohort, so these counts should be interpreted with the denominators recorded for each group.

The registry lists Diarrhoea, Atypical pneumonia, Bronchitis, Malignant neoplasm progression, Posterior reversible encephalopathy syndrome and Dyspnoea as top serious adverse-event terms, each at 16.7%. These descriptive entries do not show whether events recurred, when they occurred or whether they were attributed to treatment. They should not be read as comparative risks across regimens.

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

The registry record describes safety outcomes observed across a complex early-phase study of CT7001. It records participants with one or more related treatment-emergent adverse events in multiple dose and disease-specific cohorts, and separately lists serious adverse events, other adverse events and deaths. It also documents that the study included monotherapy, CT7001 with fulvestrant and a food-effect crossover module. These observations are useful as a record of what was reported in the study groups.

The counts do not establish that CT7001 caused an event, even when an event is categorized as related in the primary outcome. Causality is a broader judgment that depends on clinical context, timing, alternative explanations and the assessment process. Nor do the data establish that CT7001 does or does not work against cancer. The primary outcome is safety and tolerability, while anti-tumour response and time-to-event outcomes are listed as secondary outcomes. The registry record does not post analyses for these secondary outcomes in the information.

The results also do not support a single comparison between CT7001 and a common control. Cohorts vary by dose, disease group, regimen and module, and the posted record does not provide inferential comparisons or confidence intervals. Counts from separate groups cannot alone establish that one dose or regimen has a greater or smaller safety risk than another. Likewise, the listed serious adverse-event counts and deaths should not be treated as treatment-attributable effects without the relevant event-level assessments and analysis.

The Module 4 qualification is especially important for reading its primary outcome entries: the registry says these results are presented in terms of fed or fasted protocol adherence and notes a protocol deviation. That means those values cannot automatically be interpreted as comparable related-adverse-event counts. More generally, no reported sample-size justification, missing-data strategy, analysis-population definition or multiplicity approach is available in the registry record. These omissions limit how far statistical conclusions can be taken. The defensible interpretation is descriptive: the record reports cohort-specific counts, with substantial differences in context and no formal comparative estimate.

14. Why This Trial Matters Statistically

This study illustrates why statistical interpretation in modular early-phase research must begin with the structure of the cohorts. The phrase “CT7001 study” covers dose escalation, paired-biopsy expansion, disease-specific monotherapy groups, combination treatment with fulvestrant and a fed-versus-fasted crossover. Those modules do not all answer the same question. A reader who treats them as one homogeneous trial may mistake descriptive counts for direct treatment comparisons.

The primary safety outcome is participant-based, while serious adverse events and deaths appear in separate summaries. These categories answer different questions and use cohort-specific at-risk counts. The distinction between an event count, participants with at least one event, seriousness and relatedness is fundamental: collapsing them into a single notion of “safety” would lose information. Module 4 adds a further interpretive issue because its outcome description emphasizes adherence to fed or fasted conditions.

Statistically, the record also shows the limits of what a registry result can establish. There are no posted inferential analyses for the outcomes, no confidence intervals for the primary counts, and no reported sample-size justification or missing-data strategy. Without a formal comparison, numerical variation across cohorts is descriptive and may reflect different populations, treatment schedules or follow-up as well as chance. Small, unequal groups further limit precision, but the registry does not quantify that uncertainty with interval estimates.

These considerations matter beyond this specific study. Clinical researchers need to match the estimand, denominator, outcome definition and analysis population to the question they intend to answer. Registry summaries can document planned modules and posted observations, but they may not include the statistical detail needed to reconstruct every analysis decision. The appropriate interpretation is therefore neither to overstate nor dismiss the results: state what was counted, preserve the distinctions among cohorts and endpoints, and identify the comparative and causal questions the record does not resolve.

15. Limitations

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

The registry record presents many cohorts but does not provide a common comparison group across the full study. Some cohorts are dose-escalation groups, others are disease-specific expansions, one combines CT7001 with fulvestrant, and Module 4 uses fed and fasted conditions followed by continuous dosing. The overall labels of randomized and parallel do not explain the allocation details for each module; Module 4 is separately described as a randomized crossover. This limits any attempt to interpret all cohort counts as a single comparative experiment.

Outcome interpretation is further constrained by the Module 4 population note. It says results are presented in terms of participants fed or fasted correctly according to protocol and identifies a protocol deviation in the 120mg OD Fed cohort. The summarized record does not fully explain the resulting analysis population or how that deviation affected the primary endpoint count. The reported counts and at-risk values in the separate serious-event summaries also belong to different reporting contexts and should not be substituted for one another.

The registry record does not report confidence intervals, p-values or models for the primary outcome; it also does not provide a sample-size justification, interim-analysis details, a missing-data plan, a detailed analysis-population definition or a multiplicity strategy. Although pharmacokinetic and anti-tumour outcomes are listed, analyses for those outcomes are not posted in the summarized record. The listed average study time and range indicate variation in time on study, but there is no reported person-time adjustment for the safety counts.

Finally, the outcome definition focuses on participants with one or more related treatment-emergent adverse events. That count does not convey event frequency per participant, duration, severity distribution or attribution details for every event. The registry’s serious adverse-event terms and death counts likewise do not provide enough information to establish causation. These limitations mean the posted results should be treated as descriptive cohort-level observations rather than precise comparative estimates.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Modular Study to Evaluate CT7001 Alone in Cancer Patients With Advanced Malignancies — NCT03363893.
• Registry record last updated 2026-06-23; status last verified 2026-05.
• Results first posted 2026-06-23.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT03363893 is a completed, multicentre Phase I/II study of CT7001 in advanced malignancies. The registry lists randomized allocation, an open-label design and 124 participants enrolled. Its modules cover CT7001 monotherapy, disease-specific expansions, CT7001 with fulvestrant, and a fed-versus-fasted crossover followed by continuous dosing.

The primary outcome concerns treatment-emergent adverse events and laboratory abnormalities; the posted results focus on participants with one or more related treatment-emergent adverse events. The registry also lists pharmacokinetic and anti-tumour secondary outcomes, but no analyses are reported for them in the summarized record. Cohort-specific safety counts are available, including separate serious adverse-event and death counts, but no formal comparative analyses or confidence intervals are posted.

Interpretation is limited by differences among modules and populations, the Module 4 protocol-adherence qualification, and the absence of reported details about the statistical plan, missing-data handling and sample-size justification. The record supports descriptive reading of the reported groups, not a unified comparative or causal conclusion.