Berzosertib and Irinotecan in Treating Patients With Progressive, Metastatic, or Unresectable TP53 Mutant Gastric or Gastroesophageal Junction Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Berzosertib and Irinotecan in Treating Patients With Progressive, Metastatic, or Unresectable TP53 Mutant Gastric or Gastroesophageal Junction Cancer (NCT03641313), a na phase 2 study. The primary outcome is Objective Response Rate (ORR) According to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 Response Criteria.

2. Clinical Question

The central statistical question is therefore descriptive: how many participants in this treated group met the registry’s definition of objective response? A complete response means disappearance of all target lesions. A partial response is defined in the registry as at least a 30% decrease in the sum of the longest diameters of target lesions. The overall response definition combines complete and partial responses.

Because no comparator group is described, the study cannot directly estimate a treatment contrast against another assigned treatment. The registry summary gives a rationale for combining the drugs, but that rationale is not evidence of a comparative effect. The response endpoint can characterize observed tumor response in the study group; interpretation also depends on the number of participants assessed, the response evaluation process, and how missing or unevaluable assessments were handled. Those details, along with the response count itself, are not reported in the registry record.

3. Trial Design

This was a phase II, single-group study sponsored by the National Cancer Institute (NCI). The registry lists enrollment of 17 and no masking. Allocation is recorded as not applicable, consistent with the single-group design; there was no randomized assignment between treatment arms described in the registry record. The study status is completed. Its recorded start date is 2020-11-16, and its primary completion date is 2024-07-01.

Participants received irinotecan intravenously over 90 minutes and berzosertib intravenously over 60 minutes on days 1 and 15. Treatment cycles repeated every 28 days in the absence of disease progression or unacceptable toxicity. Endoscopic or CT-assisted biopsy and MRI were also listed as study procedures. The record does not give further details about how imaging assessments were scheduled or centrally reviewed.

The primary outcome was objective response rate, defined through the count of participants with a complete or partial response under RECIST 1.1. Its stated timeframe was up to 1 year. Secondary outcomes listed were duration of response, time to progression, progression-free survival, and overall survival. Although these outcomes are named, their numerical results and analysis details are not reported in the registry record.

Single-group studies can describe outcomes among participants who receive a regimen, but they do not provide a concurrent control group for separating treatment-associated outcomes from other influences. In this design, comparisons with an external benchmark would require a clearly specified and suitable reference, which is not reported in the registry record. The record also does not report a protocol-level sample-size justification, the planned statistical analysis, or details of any interim analyses. Those omissions limit how fully the design and intended inferential approach can be assessed.

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:

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 Objective Response Rate (ORR) According to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 Response Criteria, measured over Up to 1 year.

The primary endpoint counts participants who achieved either a complete response or a partial response under RECIST 1.1. The registry specifies that response is assessed by CT or MRI and gives a timeframe of up to 1 year. A complete response is disappearance of all target lesions. A partial response is at least a 30% decrease in the sum of the longest diameters of target lesions. Combining these two categories gives the stated overall response definition.

Response endpoints are used to describe whether measurable disease has decreased according to predefined criteria. General explanation: using standardized criteria helps make response classification more consistent than relying on an informal impression of change. A count of responders is also straightforward to interpret as a frequency within the treated group, provided the assessment population and handling of missing evaluations are clear.

The registry record does not report the number who responded, the number assessed, or an estimated response proportion with a confidence interval. It also does not report how confirmation of response, unevaluable scans, or missing assessments were handled. Thus the endpoint definition is available, but its observed result and precision cannot be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical information in the registry record is limited. It identifies a primary response endpoint and several secondary time-to-event outcomes, but it does not report endpoint results or a detailed analysis plan. The explanations below describe general statistical concepts and distinguish them from what is documented for this study.

What does a response count represent?

The primary endpoint is a count of participants who achieved either complete or partial response under RECIST 1.1. General explanation: a count is a frequency, not a treatment-effect comparison. To turn a count into a response proportion, the analysis must specify the denominator, such as all enrolled participants or only those with an evaluable assessment. Different denominator rules can produce different summaries. The registry record does not report the observed count, the denominator used for analysis, or a response proportion.

The endpoint definition sets a threshold for classifying partial response and a criterion for complete response. General explanation: prespecified criteria can reduce ambiguity in categorizing outcomes, but the classification also depends on the timing and availability of imaging and on assessment procedures. The registry record names CT or MRI as assessment methods but does not report the assessment schedule, review process, or rules for confirming response.

How should uncertainty be understood?

General explanation: a confidence interval is a range of values that reflects uncertainty in an estimate under a specified statistical procedure. For a response proportion, the interval depends on the number of participants and the number meeting the response definition, as well as the method used to calculate it. A wider interval indicates less precision than a narrower interval. No response estimate or confidence interval is reported in the registry record, so the precision of the primary result cannot be judged.

A p-value, if reported, would address compatibility of data with a specified statistical null model; it would not establish clinical importance or show that a treatment caused an outcome. No analysis entries or primary endpoint test result are reported in the registry record. It is therefore not possible to identify a null hypothesis, test procedure, or inferential conclusion from these entries.

What can a single-group design estimate?

General explanation: a single-group design can describe outcomes observed among participants receiving the study regimen. Without a concurrent comparison group, it does not directly estimate the difference between that regimen and another treatment. Differences in participant characteristics, disease course, assessment practices, or other factors cannot be separated through random assignment to a comparator within this study. The registry record lists no control arm and no randomized comparison.

The study’s summary states a rationale for combining berzosertib and irinotecan. That rationale should be distinguished from a statistical comparison. A comparison with irinotecan alone would require outcome data from a suitable comparator group or another valid design, neither of which is described in this record.

What do the time-to-event outcomes require?

Duration of response, time to progression, progression-free survival, and overall survival are listed as secondary outcomes. General explanation: time-to-event methods account for the time until an event and may also account for participants whose event has not been observed by the end of follow-up. Such participants are commonly described as censored, meaning their event time is not fully observed. Interpretation depends on event definitions, follow-up, censoring rules, and the analysis method. Those details and numerical results are not reported in the registry record.

Over-dispersion is a general issue for some count-data models: observed variability can exceed the variability assumed by a simpler model, making standard errors and intervals too optimistic if that feature is ignored. It is not possible to determine from the registry record whether a count model was used or whether over-dispersion was relevant to any analysis.

How should planning illustrations be read?

General explanation: a sample-size calculation is a planning exercise based on assumptions about an effect, variability, significance level, power, and design. The registry record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. This illustration is not the trial’s own design justification, and the record does not report a protocol-level sample-size rationale.

Any sample-size values from such an illustration would be hypothetical planning quantities, not observed participants and not evidence of observed power. The actual study was single-group, whereas the described illustration assumes equal allocation between two arms. Those assumptions do not describe the trial’s allocation. No numerical illustration values are reported in the registry record, so no particular planning requirement can be assessed from it.

In general, multiplicity arises when several outcomes or statistical tests are considered, because the chance of at least one chance finding can increase with the number of tests. The registry lists multiple outcomes, but it does not report a multiplicity procedure or a hierarchy of testing. It is not possible to determine whether any adjustment was planned or applied.

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("Treatment (Irinotecan and M6620)")
n      <- c(17)
events <- c(0)
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  Treatment (Irinotecan and M6620)  17  0  0.0  0.0  19.5  0.0  18.4  0.1  18.5

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome and four secondary outcomes: duration of response, time to progression, progression-free survival, and overall survival. It does not report a statistical testing hierarchy, a multiplicity adjustment, or whether the secondary outcomes were intended to support formal confirmatory claims. General explanation: when several outcomes or tests are examined, multiplicity procedures can help control the overall chance of false-positive findings. Without a documented plan, the handling of multiple analyses cannot be assessed.

Missing-data methods are also not reported in the registry record. For the response endpoint, the record does not state how participants without an evaluable scan, with a missed assessment, or with an assessment after the response timeframe were handled. General explanation: the choice to treat missing outcomes as nonresponse, exclude them, or use another method affects the interpretation and denominator of a response summary. No such rule can be inferred from the endpoint definition.

The registry does not report the analysis population for the primary or secondary outcomes. It is not stated whether analyses included all enrolled participants, only treated participants, or a subset with evaluable assessments. The registry record also does not report details of protocol deviations or withdrawals that could inform population definitions.

The record contains no analysis entries for the listed outcomes. It therefore does not provide a statistical analysis plan, missing-data strategy, or population definitions that would allow these aspects to be evaluated. These omissions do not establish that no such methods existed outside the registry record; their details are not reported in the registry record.

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: febrile neutropenia (17.6%), abdominal pain (5.9%), death nos (5.9%), colonic obstruction (5.9%), aspiration (5.9%).

The registry lists 17 participants at risk for serious adverse events, of whom 10 were affected. It also lists 17 participants at risk for other adverse events, with 17 affected. The record lists 17 deaths. These are registry counts; the record does not explain the timing, causes, or attribution of the deaths, or whether the categories overlap.

The listed serious adverse-event terms include febrile neutropenia at 17.6%, and abdominal pain, death NOS, colonic obstruction, aspiration, and infectious enterocolitis at 5.9% each. These percentages describe the registry’s reported terms. The record does not provide further detail on event severity beyond the serious-event categorization, duration, recurrence, or whether events were considered related to study treatment. The terms may not represent mutually exclusive groups of participants, and the record does not clarify overlap.

General explanation: adverse-event counts describe events or participants reported in a study; by themselves, they do not establish that the treatment caused an event. Interpretation depends on definitions, follow-up, attribution procedures, and the reporting denominator. Those details are not reported in the registry record. The safety entries should therefore be read as recorded counts and terms, not as a causal comparison or a complete account of risk.

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

The registry record documents a phase II study of berzosertib with irinotecan in a single group of participants with progressive, metastatic, or unresectable TP53 mutant gastric or gastroesophageal junction cancer. It defines objective response using RECIST 1.1 and lists several secondary outcomes. This establishes what the study intended to measure, but not what the participants’ response outcomes were.

The primary response count and any response estimate are not reported in the registry record. Therefore, the record does not show how many participants had a complete or partial response, how precise an estimate might be, or whether any statistical criterion was met. Enrollment of 17 is not a substitute for the missing outcome data. It is also not evidence of treatment benefit or lack of benefit.

The absence of a comparator is important. Even if an observed response result were available, a single-group study would describe the treated group rather than directly estimate an effect relative to another treatment. A treatment contrast, and the uncertainty around it, would require an appropriate comparison. The registry summary’s statement about the rationale for combining the agents does not supply such a comparison.

General explanation: statistical evidence and clinical importance are related but distinct. A precise estimate may still require context to judge its importance, while an imprecise estimate may leave a wide range of plausible values. Here, neither the response estimate nor its precision is reported, so neither dimension can be evaluated from the registry record.

The safety entries provide some recorded counts and event terms, but not timing, causes, attribution, or a comparative safety analysis. Those entries do not establish that every listed event or death was caused by study treatment. Taken together, the registry record supports a description of design and outcome definitions, but it does not support a conclusion about efficacy, comparative effectiveness, or causal safety effects.

14. Why This Trial Matters Statistically

This trial is statistically notable for the distinction between an endpoint definition and an endpoint result. The registry gives a clear primary outcome concept: participants with complete or partial response under RECIST 1.1. Yet it does not report the response count, estimate, or uncertainty. A defined outcome tells readers what was intended to be measured; it cannot replace the numerical findings needed to interpret the study.

The single-group design also makes the target of inference important. Such a study can characterize outcomes among participants receiving the combination, but it does not create a randomized estimate of how outcomes compare with an alternative. The registry summary discusses a rationale involving the combination and irinotecan alone, while the trial itself is recorded as having one treatment group. Keeping those concepts distinct prevents a proposed rationale from being mistaken for comparative evidence.

General explanation: readers often focus on a response proportion or a p-value, but interpretation also depends on the denominator, precision, missing-data handling, outcome assessment, and design. For this record, several of those elements are not reported. That limits what can be learned statistically, even though the study’s question and endpoint definition are identifiable.

The record also illustrates why registry entries should be read as a source with specific boundaries. It lists secondary outcomes and safety counts, but not a full analysis plan or outcome results. Clear reporting of those details is necessary to assess uncertainty, multiplicity, analysis populations, and the scope of conclusions. The missing result should not be replaced with a guess based on enrollment or the treatment rationale.

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 central limitation for interpreting this trial from the registry record is that the primary objective response result is not reported. There is no response count, estimate, confidence interval, or test result. The record therefore cannot establish the observed response frequency or its precision.

The single-group design has no concurrent comparator, so it cannot directly estimate a treatment difference against another regimen. The registry does not describe randomization or a control arm. This limits causal and comparative interpretation independently of whether response data might be available elsewhere.

Other details needed to assess the primary endpoint are also not reported in the registry record. These include the analysis denominator, the population included in the analysis, missing or unevaluable scan handling, timing and confirmation rules for response, and the assessment review process. The registry names CT or MRI, but does not describe those implementation details.

Although secondary outcomes are listed, their results, definitions beyond their names, analysis methods, follow-up details, and censoring rules are not reported. The statistical analysis plan, multiplicity handling, interim analyses, and protocol-level sample-size justification are also not reported. A planning illustration based on the observed effect is not a substitute for the trial’s own design rationale and does not establish observed power.

The safety record includes counts and selected serious-event terms, but lacks information on timing, causality, overlap, and the context of the deaths. These limitations mean the entries support only a restricted description of the study and its recorded safety data, not a complete assessment of benefits and harms.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Berzosertib and Irinotecan in Treating Patients With Progressive, Metastatic, or Unresectable TP53 Mutant Gastric or Gastroesophageal Junction Cancer — NCT03641313.
• Registry record last updated 2026-10-20; status last verified 2026-09.
• Results first posted 2025-07-08.
• 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

NCT03641313 is a completed phase II, single-group study sponsored by the National Cancer Institute (NCI), with enrollment of 17 and no masking. It studied berzosertib with irinotecan in participants with progressive, metastatic, or unresectable TP53 mutant gastric or gastroesophageal junction cancer. Treatment was described as intravenous administration on days 1 and 15 in repeating 28-day cycles, subject to disease progression or unacceptable toxicity.

The primary endpoint was the number of participants with complete or partial response under RECIST 1.1, assessed by CT or MRI over up to 1 year. The response result is not reported in the registry record. Secondary outcomes included duration of response, time to progression, progression-free survival, and overall survival, but their results are not reported.

The record lists serious and other adverse-event counts and selected serious-event terms. It does not report a detailed analysis plan, missing-data rules, analysis populations, or a comparator group. Interpretation is therefore limited to the recorded design, endpoint definition, and safety entries.