A Phase I/Ib Study of MEK162, a MEK Inhibitor, in Combination With Carboplatin and Pemetrexed in Patients With Non-squamous Carcinoma of the Lung: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase I/Ib Study of MEK162, a MEK Inhibitor, in Combination With Carboplatin and Pemetrexed in Patients With Non-squamous Carcinoma of the Lung (NCT02185690), a na phase 1 study. The primary outcome is Recommended Dose in Milligrams Per Day for Binimetinib..

2. Clinical Question

The question is characteristic of early-phase dose-finding research: whether a combination can be taken forward at a dose selected with attention to tolerability. It differs from a randomized efficacy question, which would compare outcomes between assigned treatment groups. The registry lists a secondary objective response rate outcome under RECIST v1.1, but a numerical response result is not reported in the registry record. Nor does the record provide an estimate comparing the combination with another treatment.

The registry identifies the condition as lung cancer and gives a study summary focused on non-small cell lung cancer and KRAS mutations. It does not report enough detail to establish how closely the enrolled group matched that summary, or to characterize the participants beyond the stated disease setting. Accordingly, the result addresses the registered dose-selection question in the study population; it should not be read as an estimate of benefit for all people with lung cancer.

3. Trial Design

The registry describes a Phase I/Ib, open-label, dose-escalation, multicenter, non-randomized study of binimetinib combined with carboplatin and pemetrexed. Its phase classification is Phase 1. The study is recorded as single-group, with no masking and no random allocation. It enrolled 13 participants and is marked completed. The recorded study dates are 2017-03-07 for start and 2019-07-04 for primary completion.

The arm description identifies a binimetinib efficacy/safety group and says the study was designed to evaluate the safety and tolerability of oral binimetinib with the two chemotherapy drugs. It states that a standard 3+3 dose-escalation approach would be used in Phase I part A to determine the maximum administered dose. The description is truncated in the registry record, so the complete dose-escalation rules and the full distinction between study parts are not reported.

The adverse-event groups are labeled Binimetinib Dose Level 1 (30mg BID), Binimetinib Dose Level 2 (45mg BID), and Binimetinib Dose Level -1 (30mg BID). These labels describe dose-level groups, not randomized treatment arms. Because assignment was not randomized and there was no control group, differences between dose-level groups are not controlled treatment comparisons. Participants at different levels may also represent different points in the escalation process.

The primary outcome is the recommended binimetinib dose in milligrams per day, determined by toxicity over a timeframe of 22 weeks. The registry says the outcome was 30 mg and that both arms were combined to report one recommended single dose. The dose-level labels in the safety record use BID, while the primary outcome is described in milligrams per day. The registry does not explain how these reporting units relate, so they should not be treated as interchangeable or reconciled by assumption.

The sponsor is University Health Network, Toronto. The record does not report a sample-size justification, a detailed statistical analysis plan, or a separate comparator intervention.

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: Out of 16 patients screened, 13 were enrolled who met the eligibility criteria

Binimetinib Dose Level 1 (30mg BID)

Binimetinib Dose Level 2 (45mg BID)

Analysis population for the primary outcome, as defined in the registry: One patient was not evaluable for DLT; therefore, an additional patient was enrolled at Dose Level 2. The outcome measure for the recommended does is 30 mg, both arms were combined in the analysis here in order to report 1 recommended single dose (30mg).

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: Patients 18 years or older with histologically confirmed stage IV non squamous NSCLC were eligible. Patients were treated with binimetinib in combination with first-line carboplatin and pemetrexed chemotherapy.

6. Primary Endpoint

The registry lists the primary outcome as Recommended Dose in Milligrams Per Day for Binimetinib., measured over 22 weeks.

The primary endpoint is the recommended dose of binimetinib, expressed in milligrams per day, for use with standard therapy of pemetrexed and carboplatin. The recorded value is 30 mg. The endpoint description says the recommendation was determined by toxicity, and the timeframe is 22 weeks. The record also states that the dose outcome combines both arms to report one recommended single dose.

A dose recommendation is used in early-phase research to identify a regimen for further evaluation, based on the observed tolerability of the study combination. In general, toxicity-based dose selection focuses on whether adverse effects at a dose make it unsuitable for continued development. This is different from an efficacy endpoint such as objective response: the dose value itself does not quantify tumor shrinkage, time to progression, or survival.

The record says one patient was not evaluable for dose-limiting toxicity (DLT), and that an additional patient was enrolled at Dose Level 2. It does not define the DLT assessment criteria or give a participant-level account of the dose-selection decision. The endpoint therefore communicates the registry's recommended dose, but the information needed to reproduce the decision from individual toxicity observations is not reported in the registry record.

The record lists objective response rate (ORR) per RECIST v1.1 as a secondary outcome. A numerical ORR result is not reported in the registry record, so this secondary measure cannot supplement the dose recommendation with a numerical tumor-response estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several design features determine what can be inferred from this study. It is described as open-label, non-randomized, and single-group. Open-label means that masking was not used; the registry lists masking as none. Non-randomized means participants were not assigned by chance to a treatment or control condition. Single-group means there was no separate randomized control group for estimating a treatment contrast. These design features are consistent with a dose-escalation study, but they limit the scope of comparisons that can be made from its results.

What does dose escalation estimate?

In general, a dose-escalation study examines tolerability as dose levels are explored and uses the observed toxicity information to guide a dose recommendation. The registry says Phase I part A would use a standard 3+3 dose-escalation approach to determine the maximum administered dose. A 3+3 approach is a rule-based design: decisions about whether to escalate, remain at a dose, or stop depend on observed dose-limiting toxicities under the protocol's criteria. The registry description does not provide the full rules, criteria, or decision sequence for this study. As a result, the particular dose-selection path cannot be reconstructed from the registry record.

The primary endpoint is a recommended dose, rather than a mean or proportion measured on a conventional clinical outcome scale. The record gives 30 mg per day as the recommendation, based on toxicity, over 22 weeks. It also states that one patient was not evaluable for DLT and an additional patient was enrolled at Dose Level 2. That note signals that evaluability affected enrollment at that level, but the record does not explain why the participant was unevaluable, how missing DLT information was handled analytically, or whether the assessment window was completed for every participant.

Why is this not a controlled comparison?

A randomized comparison aims to make treatment groups comparable, on average, through allocation by chance. Here, allocation is recorded as not applicable, and the study is described as non-randomized. Dose groups are stages of dose exploration, not substitutes for randomized arms. Participants receiving different dose levels may have entered at different points in escalation and may have different exposure histories. A difference in observed adverse events between those groups cannot therefore be interpreted as an isolated dose effect without further design and analysis information. The registry record does not report an adjusted comparison between dose levels.

In general, masking can reduce the influence of expectations on outcome assessment or reporting. This study was open-label, so participants and study personnel were not masked to treatment. The registry does not report how this feature affected assessment, nor does it describe any independent or blinded review process. This matters especially for outcomes requiring judgment, although the record does not give enough detail to assess the procedures used for the listed outcomes.

What can a confidence interval tell a reader?

In general, a confidence interval describes the statistical uncertainty around an estimated quantity under the assumptions of the analysis. A narrower interval indicates greater precision, while a wider interval indicates less precision. An interval does not by itself address bias, confounding, or whether a result is clinically important. The registry record does not report a confidence interval for the recommended dose, ORR, or the serious-adverse-event entries. No precision estimate for those outcomes can therefore be interpreted from the registry result.

Likewise, a hypothesis test and its p-value address compatibility between observed data and a specified statistical model and null hypothesis. They do not establish causation, nor do they measure the size or importance of an effect. The registry lists no analyses in its analyses count, and it does not report a p-value or a statistical test for the primary dose recommendation. No formal inferential conclusion should be attributed to the reported dose value.

How should planning calculations be read?

In general, a sample-size calculation describes how many participants might be needed under chosen assumptions to target a stated precision or power for a specified analysis. The planning assumptions associated with this record describe an illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Such an illustration is not the trial's own design justification. It does not show that the trial was planned to detect that effect, and it is not observed power. The registry does not report a sample-size justification or enough detail to verify the assumptions against a protocol-defined analysis.

Over-dispersion is a general concern when outcome variability exceeds what a simple statistical model assumes. If it is ignored, uncertainty can be understated and intervals or tests can be too optimistic. This concept is relevant to count outcomes in general, but the registry does not report a count-outcome model, an over-dispersion assessment, or an adjustment for it. It would be inappropriate to infer that such a method was used.

Similarly, a model-based analysis requires details about the outcome distribution, covariates, missing observations, and model checks. Those details are not reported in the registry record. The outcome listings identify the primary recommended-dose measure and a secondary ORR measure, but no fitted model or adjusted estimate is given. Readers should distinguish the recorded result from statistical procedures that might be used in other studies but are not documented.

The registry's primary dose result combines both arms to report one recommended dose. A combined value answers the stated reporting purpose, but it does not show separate primary dose estimates by group. The record does not provide the analysis steps that generated the combined result. Consequently, the value can be reported as the registry's conclusion, while its underlying calculation cannot be independently checked from the information recorded.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies the recommended binimetinib dose as the primary outcome and lists ORR per RECIST v1.1 as a secondary outcome. It records 4 outcome measures, but the names and results for all of them are not included in the available outcome details. A numerical ORR result and results for the other unlisted outcome measures are not reported in the registry record.

Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are examined. In general, making many inferential comparisons can increase the chance of a chance finding if each is interpreted separately. A statistical plan may specify which outcomes are primary, how secondary outcomes are handled, and whether adjustments are used. The registry record does not report a multiplicity adjustment or a complete hierarchy for the outcomes, so no claim about controlled overall error rates can be made.

The record reports 0 analyses and does not provide a detailed statistical analysis plan. It therefore does not describe planned hypothesis tests, prespecified covariate adjustment, subgroup analyses, or sensitivity analyses. The absence of a listed analysis is not proof that no analysis occurred; it means the methods and numerical results are not reported in the registry record.

Missingness is specifically noted for DLT evaluation: one patient was not evaluable, and an additional patient was enrolled at Dose Level 2. The reason for unevaluability, the handling of that participant's information, and the rules used to define an evaluable participant are not reported. The record does not state whether other outcome data were missing or how any such missingness was handled.

The primary result combines both arms to give one dose recommendation. The record does not define an analysis population for this endpoint or explain whether the combined result used every enrolled participant, only evaluable participants, or another set. The safety groups have separate at-risk counts, but the registry does not explain their relationship to the analysis populations for other outcomes. These omissions limit reproducibility and make it important not to substitute enrollment for an endpoint-specific denominator.

11. 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: confusion (16.7%), diverticulitis (grade 3) (14.3%), febrile neutropenia (grade 3) (14.3%), pulmonary infection (grade 2) (14.3%), anemia (grade 3) (14.3%).

The registry reports serious adverse events separately by binimetinib dose-level group. For Binimetinib Dose Level 1 (30mg BID), 6 participants were listed as at risk and 1 as affected by a serious adverse event; deaths are listed as 0. For Binimetinib Dose Level 2 (45mg BID), 7 were listed as at risk and 3 as affected; deaths are listed as 1. The Dose Level -1 group has 0 at risk, 0 affected, and 0 deaths recorded. These are descriptive entries, not evidence that one dose caused more or fewer events than another.

The record also lists serious-event terms with percentages: Confusion, 16.7%; Diverticulitis (grade 3), 14.3%; Febrile neutropenia (grade 3), 14.3%; Pulmonary infection (grade 2), 14.3%; Anemia (grade 3), 14.3%; and Cerebral metastases, 14.3%. The registry does not link each term to a specific dose group in this list or give corresponding event counts and denominators there. The percentages should not be used to infer group-specific rates or to reconstruct counts.

In general, an at-risk count is the denominator associated with the record's safety grouping, while an affected count indicates the recorded number of participants affected under that entry. The registry does not provide event timing, duration, recurrence, attribution, or detailed definitions for each safety measure. The entries therefore give a limited account of reported serious events, not a complete safety profile or a randomized comparison.

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

The result does mean that the registry records 30 mg per day as the recommended binimetinib dose for combination with carboplatin and pemetrexed, with the recommendation determined by toxicity over 22 weeks. The record says both arms were combined to report one dose. This is the central numerical result for the primary endpoint, and it should be understood as a dose recommendation from this early-phase study.

The result does not mean that 30 mg per day was shown to improve tumor response, survival, or another clinical outcome. The registry lists ORR as a secondary outcome but does not report a numerical result. It also provides no control-group estimate, treatment-effect estimate, confidence interval, or hypothesis test for efficacy. The dose result alone cannot answer whether the combination performs better than another regimen.

The result does not establish that the recommended dose is appropriate for every patient or every clinical setting. The registry record describes a specific study population and combination, but provides limited eligibility detail. The study summary says it targeted patients with KRAS mutations; mutation-level characteristics of participants are not reported. The record also does not report a protocol, complete dose-escalation rules, or enough participant-level toxicity information to reproduce the recommendation.

Safety counts likewise need to be interpreted in context. The groups were not randomized controls, and the dose-level groups are not necessarily comparable. The listed affected counts and deaths describe the record's group entries; they do not by themselves establish that dose level caused an event or that one dose level has a different risk. The serious-event term percentages lack linked group denominators in the registry listing.

In general, an early-phase dose recommendation can provide a basis for further evaluation, while a comparative efficacy conclusion requires appropriate outcome data and a design suited to comparison. For this record, a registered dose recommendation is available, but the registry does not report the results needed to make a numerical claim about efficacy or comparative benefit. The distinction is between what the record states—the recommended dose and selected safety entries—and what it does not establish: a treatment effect against a comparator.

13. Why This Trial Matters Statistically

This study is statistically informative because it illustrates a different purpose for a clinical trial than estimating comparative efficacy. Its central outcome is a recommended dose selected with regard to toxicity. The key statistical question is therefore how the observed tolerability information supports a dose decision, not whether a treatment group's outcome differs from a control group's outcome.

The design features define the limits of interpretation. The study was open-label, non-randomized, and single-group, with dose-level groups used during escalation. Such a structure is suited to early dose exploration, but it does not create a randomized control comparison. Descriptive safety entries can be reported by dose level, yet comparisons between them are vulnerable to the fact that dose assignment and escalation stage were not randomized.

The record also shows why endpoint-specific denominators matter. Enrollment is 13, but the safety entries give at-risk counts by dose group, and the dose-selection account notes one participant unevaluable for DLT. These pieces of information serve different purposes. Treating overall enrollment as the denominator for every outcome would ignore the record's endpoint-specific details.

Reporting completeness is another statistical issue. The primary dose recommendation is available, while a numerical ORR result, detailed dose-escalation rules, confidence intervals, and an analysis plan are not reported. Without those items, readers cannot reproduce the recommendation or assess the precision of a response estimate. Clear reporting of estimands, analysis populations, missing-data handling, and uncertainty would make the evidential meaning easier to evaluate.

Finally, the record separates a clinically important dose-selection aim from claims that require further evidence. A dose recommendation is useful information for understanding what the study selected, but it is not a measure of comparative benefit. Keeping those questions distinct prevents a dose-finding result from being mistaken for an efficacy result that the registry does not report.

14. 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 contains a brief, truncated arm description. It names a standard 3+3 approach for Phase I part A, but does not give the complete escalation rules, DLT definitions, dose decision criteria, or the relation between the Phase I and Ib parts. This prevents independent reconstruction of how the recommended dose was reached.

The primary endpoint is described as 30 mg per day, while safety groups are labeled with doses of 30mg BID and 45mg BID. The record does not explain the difference in units or how the labels correspond to the primary recommendation. Those entries should be quoted as recorded rather than reconciled through an assumption.

The record notes one patient was not evaluable for DLT and an additional patient was enrolled at Dose Level 2, but does not state why the patient was unevaluable or how the missing assessment was handled. It does not define analysis populations for the dose recommendation, ORR, or safety outcomes. Although enrollment is 13, that figure alone does not establish the denominator for every endpoint.

The registry does not report a numerical ORR result, confidence intervals, hypothesis tests, or a detailed analysis plan. It records 4 outcome measures, but the available outcome details identify only the primary dose measure and the secondary ORR outcome. The other outcome names and their results are not reported in the registry record.

Safety detail is limited. The registry lists serious-event affected counts and deaths by dose-level group, and a set of serious-event terms with percentages. It does not report the term-specific denominators or assign each listed term to a dose group. Event timing, duration, recurrence, and attribution are also not reported. Since the study was not randomized, the dose-level safety entries should not be interpreted as a causal comparison.

Finally, the record does not report a sample-size justification or enough information to assess whether the enrollment was selected for a particular precision or inferential objective. These reporting limitations do not alter the recorded dose recommendation, but they constrain how fully its basis and uncertainty can be evaluated.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Phase I/Ib Study of MEK162, a MEK Inhibitor, in Combination With Carboplatin and Pemetrexed in Patients With Non-squamous Carcinoma of the Lung — NCT02185690.
• Registry record last updated 2025-08-21; status last verified 2025-08.
• Results first posted 2025-08-21.
• 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.

18. Record Summary

This completed, open-label Phase I/Ib dose-escalation study evaluated binimetinib with carboplatin and pemetrexed in patients with non-squamous carcinoma of the lung. It is recorded as Phase 1, non-randomized, and single-group, with enrollment of 13. The study started on 2017-03-07 and reached primary completion on 2019-07-04.

The primary outcome was the recommended binimetinib dose in milligrams per day, determined by toxicity over 22 weeks. The registry reports 30 mg and says both arms were combined to report one recommended dose. It notes one patient was not evaluable for DLT and an additional patient was enrolled at Dose Level 2.

The registry lists ORR per RECIST v1.1 as a secondary outcome but does not report a numerical result. It also gives serious-adverse-event entries by dose-level group and selected serious-event terms, without the complete detail needed for a comparative safety analysis. A statistical analysis plan, confidence intervals, and sample-size justification are not reported in the registry record.