Study Assessing Activity of Molecularly Matched Targeted Therapies in Select Tumor Types Based on Genomic Alterations: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Assessing Activity of Molecularly Matched Targeted Therapies in Select Tumor Types Based on Genomic Alterations (NCT02795156), a non-randomized phase 2 study. The primary outcome is Overall Response Rate (ORR) in Each Arm Receiving Targeted Therapy Based on Relevant Genomic Alterations.

2. Clinical Question

The study asks whether genomic profiling can guide selection among commercially available targeted therapies for patients whose cancer has progressed beyond first-line treatment. The four tumor groups are considered separately in the arm descriptions. Within each group, the drug choice is described as depending on the participant’s specific genomic alterations. This means the study question concerns activity in molecularly matched treatment settings, rather than a direct randomized comparison of one drug against another.

The primary endpoint is overall response rate (ORR), defined by confirmed complete or partial response. The registry describes assessments every 8 weeks until tumor progression or treatment discontinuation, up to 45 months. Clinical benefit rate, time to treatment failure, and progression-free survival are listed as secondary outcomes for each arm. Their names indicate additional measures of clinical course, but the registry record does not provide their results or detailed definitions.

The record does not report the arm-specific ORR values or their 95% confidence intervals. Consequently, it does not support a numerical conclusion about the response proportion in any tumor group, nor a comparison of response across groups or drugs.

3. Trial Design

This is a phase II, non-randomized, open-label study with a parallel model. The registry lists an enrollment of 100 and status as completed. The study began on 2016-09-28, and primary completion was recorded as 2022-08-17. The sponsor is SCRI Development Innovations, LLC.

Four arms correspond to four tumor groups: non-small cell lung cancer, urothelial carcinoma, non-colon gastrointestinal cancer, and upper aerodigestive tract cancer. In each arm, participants who had failed first-line treatment could receive regorafenib, afatinib, or cabozantinib at the recommended dose level, depending on their genomic alterations. The registry does not describe random assignment or masking; masking is listed as none.

The parallel structure means that the arms represent distinct groups being followed within the same study. It does not, by itself, make their outcomes directly comparable. The groups differ by tumor type, and drug selection is linked to genomic alterations. The registry explicitly cautions that this is not a randomized comparison and that the arms are not a like-for-like control comparison. Differences in outcomes across arms, if reported, would therefore not isolate a treatment effect from differences in tumor group, genomic profile, or other factors.

The primary outcome is ORR in each arm. It is defined as the percentage of patients with confirmed complete response or confirmed partial response under RECIST v1.1. The secondary outcomes listed are clinical benefit rate, time to treatment failure, and progression-free survival in each arm. The response assessment schedule is every 8 weeks until tumor progression or treatment discontinuation, up to 45 months.

The registry record does not report the detailed protocol, eligibility criteria beyond the descriptions given, allocation procedures beyond non-randomization, or a statistical analysis plan. It also does not provide the number enrolled in each tumor arm as part of the primary outcome description. Safety group counts are reported separately and should not be assumed to supply the missing response denominators.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All enrolled patients are included.

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 Overall Response Rate (ORR) in Each Arm Receiving Targeted Therapy Based on Relevant Genomic Alterations, measured over every 8 weeks until tumor progression or treatment discontinuation, up to 45 months..

The primary endpoint is overall response rate (ORR), expressed as the percentage of participants with a confirmed complete response or confirmed partial response. A complete response means disappearance of all target and non-target lesions. A partial response means at least a 30% decrease in the sum of diameters of target lesions, using the baseline sum as the reference. The registry requires confirmation, defined as two complete and/or partial responses at least 4 weeks apart, according to RECIST v1.1.

Response is used as a measure of observable tumor change during treatment. In general, a defined response endpoint can provide a clinically interpretable summary of tumor shrinkage, while requiring confirmation can help distinguish a sustained response from a single assessment. The registry describes assessments every 8 weeks until progression or treatment discontinuation, up to 45 months.

The registry says all enrolled patients are included in the primary outcome population. However, it does not report the observed number or percentage of participants with confirmed responses in any arm, nor the 95% confidence intervals. It also does not describe how missing or unevaluable response assessments were handled. The endpoint definition is therefore available, but the posted record does not provide the result needed to quantify response in these groups.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a binary participant-level event: a participant either meets the registry’s definition of confirmed complete or partial response or does not. The outcome is summarized as a percentage. In general, a percentage expresses the share of a defined group with the event; its interpretation depends on a clear denominator and a consistent event definition. The record says all enrolled patients are included, but does not report the arm-specific response counts or percentages.

What does confirmation mean?

The registry defines a confirmed response as two complete and/or partial responses at least 4 weeks apart. This makes response dependent on repeated assessments, not just a single scan. Complete response requires disappearance of all target and non-target lesions. Partial response requires at least a 30% decrease in the sum of target-lesion diameters relative to baseline. These are the registry’s stated RECIST v1.1 criteria. The listed assessment schedule is every 8 weeks until progression or treatment discontinuation, up to 45 months.

How should a confidence interval be read?

The registry specifies a 95% confidence interval for the primary outcome. In general, a confidence interval communicates uncertainty around an estimated response percentage. Its width reflects precision: estimates based on less information often have greater uncertainty, while more information can yield greater precision. The confidence interval does not establish that a result is clinically important, and it does not eliminate possible bias arising from study design or outcome assessment. In this case, the interval limits are not reported in the registry record, so neither its width nor the precision of the response estimate can be assessed.

Why does the design affect comparisons?

Allocation is non-randomized, masking is listed as none, and the study has a parallel model. Treatment choice depended on genomic alterations, while the four arms represent different tumor groups. In general, random assignment can help balance measured and unmeasured participant characteristics between treatment groups. Without randomization, groups may differ for reasons connected to treatment selection or prognosis. Since the registry also cautions that the arms are not like-for-like controls, an arm-level contrast should not be read as a randomized treatment comparison.

Open-label design means masking is absent in the registry description. In general, knowledge of treatment can influence clinical management, assessment, or reporting, although the extent of any such influence cannot be determined from this record. The outcome is based on RECIST criteria, but the record does not describe independent review, assessor masking, or procedures for resolving differing assessments.

What is known about statistical planning?

The registry record does not report the study’s sample-size justification or detailed statistical plan. Sample-size illustrations associated with the analysis are planning illustrations based on the observed effect, not observed power. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the illustration gives a total number of events using the Schoenfeld approach; other illustrations give participants per arm. These planning assumptions do not establish the trial’s original design assumptions, and the record does not report that the study used them prospectively.

In general, over-dispersion occurs when observed variability exceeds what a chosen statistical model expects. It can make standard errors and confidence intervals too small if a model does not account for the extra variability. The registry record does not report a model for the primary endpoint or indicate whether over-dispersion was assessed. This concept is relevant as a general statistical caution, not as a claim about the analysis used in this study.

The registry lists secondary time-to-event outcomes, but does not report their analysis methods, censoring rules, or estimates. In general, time-to-event methods account for both the timing of an event and incomplete follow-up, but their interpretation depends on the event definition and handling of censoring. Neither is detailed for these secondary outcomes in the 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("Arm 1", "Arm 2", "Arm 3", "Arm 4")
n      <- c(39, 10, 42, 9)
events <- c(4, 0, 3, 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  Arm 1  39  4  10.3  2.9  24.2  4.1  23.6  4.2  23.7
ARM  Arm 2  10  0  0.0  0.0  30.8  0.0  27.8  0.2  28.5
ARM  Arm 3  42  3  7.1  1.5  19.5  2.5  19.0  2.6  19.1
ARM  Arm 4  9  0  0.0  0.0  33.6  -0.0  29.9  0.3  30.8

11. Multiplicity, Missing Data and Analysis Populations

The record lists ORR in each arm as the primary outcome and clinical benefit rate, time to treatment failure, and progression-free survival in each arm as secondary outcomes. Considering multiple outcomes or multiple arm-level comparisons can increase the chance of observing a seemingly notable result by chance. In general, multiplicity procedures may adjust the interpretation or testing across several comparisons. The registry record does not report whether any such procedure was planned or used, and it lists no analyses.

The primary outcome description states that all enrolled patients are included. It does not explain how participants without an evaluable response assessment, participants who discontinue treatment, or participants with missing measurements were handled. Nor does it report missing-data counts or sensitivity analyses. The observation schedule continues every 8 weeks until tumor progression or treatment discontinuation, up to 45 months, but the record does not give further rules for follow-up after those events.

The study population for the primary outcome is described broadly as all enrolled patients. The record does not provide a separate analysis-population definition, such as rules for inclusion in an evaluable subset, or clarify how the named population was implemented in the analysis. It does not report whether any participants were excluded from an outcome analysis.

These omissions matter because denominators, missing outcomes, and analysis populations affect how a percentage should be interpreted. The registry’s stated population definition is useful, but without arm-specific enrollment, response counts, and missing-data handling, the response proportion and its precision cannot be independently understood from the posted record. No inference should be made about the direction or size of a missing-data effect.

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: upper gastrointestinal haemorrhage (11.1%), haemoptysis (11.1%), pneumonia klebsiella (11.1%), nervous system disorder (11.1%), syncope (11.1%).

The safety figures are presented by arm. For Arm 1, the record lists 39 participants at risk for serious adverse events and 16 affected; for other adverse events it lists 39 at risk and 39 affected, with 14 deaths. For Arm 2, the corresponding counts are 10 and 6 for serious events, 10 and 10 for other events, and 1 death. For Arm 3, they are 42 and 14 for serious events, 42 and 41 for other events, and 13 deaths. For Arm 4, they are 9 and 5 for serious events, 9 and 8 for other events, and 3 deaths.

The record also lists serious-event terms and percentages: Upper gastrointestinal haemorrhage, 11.1%; Haemoptysis, 11.1%; Pneumonia klebsiella, 11.1%; Nervous system disorder, 11.1%; Syncope, 11.1%; and Ileus, 10.0%. The record does not give further detail about the denominators or treatment attribution for these terms.

These are descriptive registry counts. They do not show that an event was caused by treatment, and the death counts do not establish cause of death. The arms differ in tumor group and treatment selection, so their counts are not a like-for-like safety comparison. The record does not report additional details such as event timing, severity grades, or drug-specific attribution.

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

The registry record identifies a study of molecularly matched targeted therapies in four tumor groups and specifies a confirmed response endpoint. It also describes the assessment schedule and lists secondary outcomes. These details establish what the study set out to measure, but they do not by themselves show the observed activity of the therapies.

The primary ORR estimates and 95% confidence intervals are not reported in the registry record. It is therefore not possible to state what proportion responded in any arm, how uncertain those estimates were, or whether response differed across tumor groups. The absence of reported estimates is not a result of zero response and should not be interpreted as evidence for or against activity.

The non-randomized design and genomic matching also limit what a cross-arm comparison could establish. Each arm represents a different tumor group, and drug selection depended on genomic alterations. Even if arm results were available, differences would not automatically show that one drug was more effective than another. The registry cautions that these arms are not a like-for-like control comparison.

The safety counts provide a descriptive account of affected participants and deaths by arm. They do not establish treatment causation or support a direct comparative safety conclusion. The listed serious-event terms and percentages likewise do not provide, on their own, enough context to attribute events to a particular drug or assess differences across treatment groups.

In general, statistical evidence is best interpreted by considering the endpoint definition, estimate, uncertainty, population, and design together. Here, the endpoint definition and design are described, while the primary estimates, confidence interval limits, and detailed analysis methods are absent. Conclusions should therefore remain limited to the study’s stated aims and the descriptive information recorded, rather than extending to treatment effects or comparative efficacy.

14. Why This Trial Matters Statistically

This study illustrates a statistical challenge in precision oncology: treatment assignment can be tied to both tumor type and genomic profile. The clinical question is whether therapies selected to match genomic alterations show activity in several tumor settings, but the non-randomized structure means that arm-level results would not be direct randomized comparisons. Tumor group and treatment selection are part of the comparison context, not incidental details.

The endpoint is also a useful example of a binary outcome built from longitudinal assessments. A participant qualifies only with confirmed complete or partial response under the stated RECIST v1.1 criteria. This definition makes the event interpretable, while the repeated assessments and confirmation requirement mean that assessment timing and missing observations are important. The record gives the schedule but not the handling of missing assessments.

Statistical interpretation depends not only on whether an outcome is defined, but also on the estimated event proportion, its uncertainty, the denominator, and the analysis population. The registry specifies a 95% confidence interval but does not report its limits or the response estimates. This makes it impossible to judge precision or compare the observed response proportions from the posted record.

The study also highlights why registry descriptions and results should be read separately. The record contains substantial information about the study purpose, arms, endpoint, and some safety counts, while the primary response results and detailed analysis plan are not reported. Clear separation of design information, descriptive counts, and inferential results helps prevent a stated research question from being mistaken for evidence of an 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 statistical interpretation is that the registry record does not report the primary ORR results. It gives neither arm-specific response counts nor percentages, despite specifying the endpoint and a 95% confidence interval. The precision of the result and the extent of response in each tumor group therefore cannot be assessed.

The record describes four tumor arms and says drug choice depended on genomic alterations. It is non-randomized and explicitly cautions that the arms are not a like-for-like control comparison. Any cross-arm contrast would be difficult to separate from differences in tumor group, genomic profile, and treatment selection. No randomized comparator is described.

Important analysis details are also absent: the statistical analysis plan, sample-size justification, multiplicity approach, missing-data procedures, and detailed definitions or results for the secondary outcomes are not reported in the registry record. The primary outcome description states that all enrolled patients are included, but arm-level response denominators and handling of unevaluable outcomes are not provided.

The safety figures are descriptive and do not establish causality. The registry record does not report further details about treatment attribution, event timing, or severity grades in the safety summary. Accordingly, neither comparative treatment effects nor comparative safety conclusions can be drawn from these data alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study Assessing Activity of Molecularly Matched Targeted Therapies in Select Tumor Types Based on Genomic Alterations — NCT02795156.
• Registry record last updated 2023-12-05; status last verified 2023-08.
• Results first posted 2023-09-15.
• 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

NCT02795156 is a completed phase II, non-randomized, open-label, parallel study sponsored by SCRI Development Innovations, LLC. The record lists enrollment of 100, with a start date of 2016-09-28 and primary completion on 2022-08-17. It examines afatinib, regorafenib, and cabozantinib selected according to genomic alterations across four tumor groups after failure of first-line treatment.

The primary endpoint is confirmed ORR in each arm under RECIST v1.1, assessed every 8 weeks until progression or treatment discontinuation, up to 45 months. Clinical benefit rate, time to treatment failure, and progression-free survival are listed as secondary outcomes. The registry record does not report the primary response estimates or confidence interval limits. It provides safety counts by arm and cautions that the non-randomized arms are not like-for-like controls.