Study of Cobimetinib in Participants With Solid Tumors: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Cobimetinib in Participants With Solid Tumors (NCT00467779), a non-randomized phase 1 study. The primary outcome is Stage 1 and 1A: Number of Participants With Dose Limiting Toxicities (DLTs).

2. Clinical Question

3. Trial Design

This was a Phase 1, non-randomized, open-label study with a parallel model. Its official title describes a dose-escalation study of the safety and pharmacokinetics of oral GDC-0973/XL518 in participants with solid tumors. The record identifies cobimetinib as the study drug and lists midazolam and dextromethorphan as additional drugs used in a later stage.

Stage 1 assessed escalating cobimetinib doses on a 21/7 schedule: once daily on Days 1-21 of each 28-day cycle, followed by 7 days off treatment. Stage 1A used a 14/14 schedule: once daily on Days 1-14 of each 28-day cycle, followed by 14 days off treatment. The registry also describes expansion stages at doses established in the corresponding earlier dose-escalation stage. The descriptions state that treatment continued until progressive disease or unacceptable toxicity, with some duration details truncated in the registry record.

Stage 3 involved a single dose of midazolam and dextromethorphan in the absence of cobimetinib, followed by a washout and cobimetinib dosing. The record says this stage assessed pharmacokinetic effects; it is not a randomized comparison of the drugs. The primary endpoint concerned only Stage 1 and Stage 1A, not the expansion or Stage 3 cohorts.

The study is listed as completed, with a start in 2007-05 and primary completion in 2014-04. The registry record does not report a randomization procedure, masking, or a concurrent control arm. The overall enrollment entry does not establish how many participants were included in each endpoint analysis.

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: Study included following stages: Stage 1, Stage 1A, Stage 2, Stage 2A, and Stage 3. Different participants were recruited in each stage.

Only the first 6 of 15 groups are shown.

Stage 1

Reasons for not completing this period:

Stage 1A

Reasons for not completing this period:

Stage 2

Reasons for not completing this period:

Stage 2A

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety population; Stages 1 and 1A participants only.

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: Safety population consisted of all participants who received at least 1 dose of study drug.

Only the first 6 of 16 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Stage 1 and 1A: Number of Participants With Dose Limiting Toxicities (DLTs), measured over Stage 1 and 1A: Days 1 to 28 of Cycle 1.

The primary endpoint was the number of participants with dose-limiting toxicities (DLTs) in Stage 1 and Stage 1A. Its stated timeframe was Days 1 to 28 of Cycle 1, and its population was the safety population in those stages. The endpoint is a count of participants meeting the DLT definition, rather than a measurement of tumor response or a pharmacokinetic quantity.

Adverse events were graded using National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0. A DLT included a drug-related adverse event that the cohort review committee judged potentially clinically significant in a way that could expose participants in higher-dose cohorts to irreversible medical harm or require medical treatment to avoid such harm. The description also lists specified severe non-hematologic and hematologic toxicities, including Grade 3 or 4 events, Grade 4 thrombocytopenia, and specified Grade 4 neutropenia events.

In dose escalation, this endpoint is used to characterize early-cycle tolerability at each dose and to inform decisions about further escalation. The count is tied to the stated eligibility criteria, observation window, and committee judgement; it is not a complete summary of all safety events over the full study. The registry record does not report the primary endpoint count or its cohort-specific results, so the endpoint cannot be interpreted quantitatively from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a dose-escalation endpoint capture?

In general, a dose-escalation study examines safety as dose levels change, often using early-cycle toxicities to guide whether escalation should continue. A DLT endpoint is a rule-based outcome: the event has to meet the protocol’s definition within a specified observation window. The definition can combine clinical severity criteria with review or attribution by a study committee. Here, the stated window is Days 1 to 28 of Cycle 1, and the definition includes committee judgement about potential clinical significance as well as specified graded toxicities.

A count of participants with DLTs is different from a count of all adverse-event episodes. A participant can experience more than one event, but the endpoint asks whether the participant meets the DLT criteria. The registry’s primary endpoint is a participant count, while the adverse-event entries separately report participants affected by serious events and other events for each cohort. The registry record does not report the primary count, so no DLT rate or confidence interval can be described.

Why does the design matter?

In general, randomization helps make treatment groups comparable, on average, with respect to factors that could affect outcomes. This study was non-randomized and open-label, and its cohorts were organized by dose, schedule, or study stage. That makes a simple comparison between cohort outcomes vulnerable to differences beyond the dose itself. For example, cohorts in an escalation stage and an expansion stage have different roles in the study. A difference in observed safety counts across such cohorts cannot by itself show that one dose caused more or fewer events.

In general, masking can reduce the influence of knowledge about assigned treatment on outcome assessment or reporting. There was no masking in this study. The primary endpoint did include a committee judgement component for certain events, but the registry record does not provide details of procedures intended to limit assessment bias. Those procedures are not reported in the registry record.

How should counts and percentages be read?

A participant count is a numerator: it says how many participants met the specified event definition. It is not interpretable as a risk or rate without knowing the relevant denominator and population. Cohort safety entries list participants at risk alongside participants affected, but those cohort counts do not replace the missing primary endpoint results. Term-level serious adverse-event percentages are also presented, but their denominators are not reported in the registry record. They should therefore be read as the percentages recorded for those terms, not reconstructed into broader study-wide rates.

In general, a confidence interval expresses the range of values compatible with the data under a specified statistical procedure and assumptions. It communicates precision as well as the observed estimate. No confidence intervals for the primary endpoint are reported in the registry record. For sparse event counts, intervals can be wide; a small or absent observed count would not, by itself, establish that the underlying event risk is negligible. This is a general statistical caution, not a numerical finding about this study.

What can pharmacokinetic outcomes add?

Pharmacokinetic outcomes describe how drug concentrations change over time, using quantities such as maximum concentration, time to maximum concentration, area under the concentration-time curve, clearance, and half-life. The registry lists secondary pharmacokinetic outcomes for cobimetinib, including several steady-state measures and a time-to-maximum-concentration measure for Stage 1A. These outcomes address exposure and drug handling rather than the DLT count. Although the record lists 44 outcome measures, the pharmacokinetic outcome values are not included in the registry record for this explanation; their numerical results are not reported in the registry record.

General statistical methods can also address unequal variability, repeated measurements, or time-to-event outcomes, but the registry record does not specify such methods for the listed outcomes. It does not report a statistical analysis plan, model, missing-data procedure, or interim-analysis approach. Those details should not be inferred from the study’s phase or design. No recalculated analysis is appropriate from the cohort summaries alone, and no comparative estimate can be derived without making unsupported assumptions.

What do planning calculations mean?

In general, a sample-size calculation is a planning exercise based on assumptions about the effect to detect, significance level, allocation, and statistical method. It is not a description of what occurred in the trial, and a planned power value is not observed power. The registry record contains no randomized comparison, and it does not report a primary endpoint effect estimate on which a meaningful comparative planning illustration could be based. A planning calculation cannot repair the absence of primary endpoint results or create a controlled treatment contrast.

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("Stage 1 Cohort 01 - Cobimetinib 0.05 mg/kg (21/7)", "Stage 1 Cohort 02 - Cobimetinib 0.10 mg/kg (21/7)", "Stage 1 Cohort 03 - Cobimetinib 0.20 mg/kg (21/7)", "Stage 1 Cohort 04 - Cobimetinib 10 mg (21/7)", "Stage 1 Cohort 05 - Cobimetinib 20 mg (21/7)", "Stage 1 Cohort 06 - Cobimetinib 40 mg (21/7)")
n      <- c(4, 3, 3, 3, 3, 6)
events <- c(0, 0, 0, 0, 0, 1)
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  Stage 1 Cohort 01 - Cobimetinib 0.05 mg/kg (21/7)  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Stage 1 Cohort 02 - Cobimetinib 0.10 mg/kg (21/7)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Stage 1 Cohort 03 - Cobimetinib 0.20 mg/kg (21/7)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Stage 1 Cohort 04 - Cobimetinib 10 mg (21/7)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Stage 1 Cohort 05 - Cobimetinib 20 mg (21/7)  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Stage 1 Cohort 06 - Cobimetinib 40 mg (21/7)  6  1  16.7  0.4  64.1  3.0  56.4  3.7  57.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 44 outcome measures, including a primary safety endpoint and secondary pharmacokinetic outcomes. When a study examines many outcomes or analyses, the chance of at least one apparently notable result can increase if each is assessed separately. In general, multiplicity can be addressed by prespecifying a primary outcome, using an adjustment procedure, or interpreting secondary findings as exploratory. The registry record does not report a multiplicity strategy or statistical analysis plan, so it is not possible to determine how the listed outcomes were prioritized or whether any adjustment was planned.

The record lists no analyses. Numerical results for the primary DLT endpoint are not reported in the registry record, and no inferential statistics or uncertainty intervals are available for interpretation. The record does not explain whether results were omitted, not entered, or unavailable for another reason. The reason is not reported in the registry record.

The primary endpoint population is identified as the safety population in Stages 1 and 1A, but the record does not state the number analyzed or provide participant-level accounting for that endpoint. It also does not report how missing observations, withdrawals, or participants with incomplete first-cycle follow-up were handled. No assumptions about these procedures should be made.

In general, missing outcome data can affect estimates if the probability of missingness is related to treatment, prognosis, or the outcome itself. The extent and consequences depend on why data are missing and how the analysis handles them. Since the registry record does not provide missing-data counts or methods, the direction or size of any potential bias cannot be assessed. The reported enrollment entry should not be substituted for an endpoint-specific analysis population.

12. Safety Results

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

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: anaemia (33.3%), supraventricular tachycardia (33.3%), diarrhoea (33.3%), gastrointestinal haemorrhage (33.3%), faecal incontinence (33.3%).

The adverse-event summaries are organized by study cohort and distinguish serious adverse events from other adverse events. Each cohort entry gives a count of participants at risk and a count of participants affected for each category. These are descriptive counts, not randomized comparisons. Cohorts differ in dose, schedule, and stage, so their entries should be considered separately rather than combined into a treatment effect.

The listed serious adverse-event terms are Anaemia, Supraventricular tachycardia, Diarrhoea, Gastrointestinal haemorrhage, and Faecal incontinence, each at 33.3%, and Abdominal pain at 25.0%. The registry record does not report the denominators associated with these term percentages. A term percentage is not necessarily a percentage of all enrolled participants, and it should not be used to infer the overall frequency of serious events.

The primary DLT endpoint has a specific definition and observation window. Serious adverse events are a separate safety category and should not be equated with DLTs. The registry record does not provide enough information to determine which serious events met the DLT definition or how the event summaries relate to the primary endpoint. No overall safety rate should be calculated from the cohort entries.

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

The record documents a Phase 1 study that examined cobimetinib dose escalation and expansion in participants with solid tumors, with both 21/7 and 14/14 schedules. Its primary endpoint was designed to capture DLTs during the first cycle in the Stage 1 and Stage 1A safety populations. The cohort-level adverse-event entries and serious adverse-event terms offer descriptive safety information, but they do not supply the missing primary endpoint result.

Because the numerical DLT result is not reported in the registry record, it is not possible to state how many participants met the primary endpoint, estimate its frequency, or assess its precision. Nor can a maximum tolerated dose be inferred from the listed cohort entries alone. The record says that expansion dosing used the maximum tolerated dose established in an earlier stage, but does not report the endpoint result or the basis for that determination.

The non-randomized, open-label design also limits causal interpretation. The cohorts are not like-for-like randomized groups, and the study does not provide a concurrent randomized comparator. Differences in observed adverse-event counts cannot be attributed to dose or schedule alone. The Stage 3 pharmacokinetic component included midazolam and dextromethorphan, but no numerical pharmacokinetic findings are reported in the registry record.

The posted information therefore supports a description of the study’s design, endpoint definitions, and selected cohort safety entries. It does not establish a comparative benefit, a treatment effect, or a complete safety profile. These boundaries reflect the information recorded, not a conclusion that the study did or did not observe particular outcomes.

14. Why This Trial Matters Statistically

Early-phase dose-escalation studies illustrate the distinction between a design question and an estimable result. A protocol can define a clinically important endpoint, an observation window, and criteria for escalation, yet a registry entry without the endpoint count cannot answer how frequently the defined event occurred. The distinction matters because the design alone is not evidence of the result.

This study also highlights why dose cohorts should not automatically be treated as comparative arms. Escalation and expansion groups serve different purposes, while the schedules and doses vary. Without randomization and a concurrent control, descriptive differences across cohorts do not isolate the effect of dose or schedule. The posted adverse-event counts can still convey what was recorded in individual cohorts, but they do not create a controlled comparison.

The separation between DLTs and general safety summaries is another important statistical point. DLT is a specifically defined outcome in a particular timeframe, with some criteria involving committee judgement. Serious adverse events are reported as a distinct category. A reader should not use one category as a proxy for the other unless the record provides the necessary mapping, which it does not.

Finally, a registry record is useful for clarifying design and outcome definitions, but completeness of results matters for interpretation. Here, the primary endpoint result and analysis details are not reported in the registry record. That absence prevents quantitative assessment of the study’s central safety question and underscores the need to distinguish reported counts from conclusions that would require analysis, denominators, and uncertainty estimates.

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 is that the registry record does not report the numerical result for the primary DLT endpoint. The definition and timeframe are available, but without the participant count and endpoint-specific analysis population, the primary safety question cannot be answered quantitatively. No confidence interval or other measure of precision is reported.

The design was non-randomized and open-label, with cohorts differing by dose, schedule, and stage. These features limit interpretation of differences between cohorts as effects of dose or schedule. There is no randomized control group for a like-for-like comparison. The record does not report details of outcome-assessment safeguards beyond the stated cohort review committee role.

The overall enrollment entry does not give the number analyzed for each outcome. The registry does not report missing-data counts, reasons for missingness, or the handling of incomplete follow-up. It also does not report the statistical analysis plan, multiplicity approach, interim-analysis procedures, or sample-size justification. These items are not reported in the registry record.

The serious adverse-event term percentages lack stated denominators in the record, and cohort summaries should not be added together to create an overall rate. The listed pharmacokinetic outcomes likewise have no numerical results in the record available for this explanation. These limitations restrict the conclusions to the study design and the descriptive safety entries actually recorded.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Cobimetinib in Participants With Solid Tumors — NCT00467779.
• Registry record last updated 2016-08-26; status last verified 2016-07.
• Results first posted 2016-07-22.
• 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

NCT00467779 is a completed Phase 1, non-randomized, open-label study of cobimetinib in participants with solid tumors. The study included dose-escalation and expansion stages using 21/7 and 14/14 schedules, plus a Stage 3 pharmacokinetic component involving midazolam and dextromethorphan. Enrollment is listed as 119.

The primary endpoint was the number of participants with DLTs during Days 1 to 28 of Cycle 1 in the Stage 1 and 1A safety populations. The numerical primary endpoint result is not reported in the registry record. Cohort-level serious and other adverse-event counts and selected serious-event term percentages are listed, but they do not provide a randomized comparison or replace the missing primary result. Statistical analysis details and missing-data methods are not reported in the registry record.