This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study of MetMAb Administered to Patients With Advanced Non-Small Cell Lung Cancer, in Combination With Tarceva (Erlotinib) (NCT00854308), a randomized phase 2 study. The primary outcome is Progression-free Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00854308 |
| Phase | Phase 2 |
| Condition | Non-Small Cell Lung Cancer |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 137 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Genentech, Inc. |
| Results posted | Yes |
2. Clinical Question
The trial asked whether adding MetMAb to erlotinib was associated with different progression-free survival than adding placebo to erlotinib in people with advanced non-small cell lung cancer receiving second- or third-line treatment. It also recorded objective response outcomes, including for participants whose tumors were classified as MET diagnostic-positive.
Population
All randomized intent-to-treat patients with advanced non-small cell lung cancer. The registry describes the study as evaluating second- and third-line treatment. The primary outcome population is stated as all randomized intent-to-treat patients.
Intervention
MetMAb by intravenous infusion every 3 weeks, alongside erlotinib taken orally once daily. The arm description gives MetMAb as 15 mg/kg and erlotinib as 150 mg.
Comparator
Placebo by intravenous infusion every 3 weeks, alongside erlotinib taken orally daily at 150 mg.
Primary question
How did the posted progression-free survival results compare between the MetMAb-plus-erlotinib arm and the placebo-plus-erlotinib arm?
This comparison concerns the addition of MetMAb to a shared erlotinib regimen; it is not a comparison of erlotinib with no treatment. The primary outcome was time from randomization to progression or relapse, or death on study under the registry definition. The registry also records objective response outcomes, but those do not replace the primary time-to-event question.
3. Trial Design
The registry describes NCT00854308 as a randomized, Phase II, multicenter, double-blind, placebo-controlled, parallel-group study. Its stated purpose was to preliminarily evaluate the activity and safety of MetMAb plus erlotinib against placebo plus erlotinib in second- and third-line non-small cell lung cancer. The planned allocation was 1:1, and the record states that up to 180 patients would be randomized.
Enrollment is listed as 137. The posted primary outcome results give 69 participants in the MetMAb-plus-erlotinib arm and 68 in the placebo-plus-erlotinib arm. The study status is completed. The registry lists the start as 2009-04 and primary completion as 2010-11.
Both arms received erlotinib at 150 mg orally once daily. The experimental arm also received MetMAb at 15 mg/kg by intravenous infusion every 3 weeks; the comparator arm received placebo infusion every 3 weeks. The arm descriptions say treatment continued until disease progression or unacceptable toxicity. The registry lists erlotinib, MetMAb, and placebo as the interventions.
Double masking and placebo control are design features intended to limit differences in expectations and care between arms, although the registry record does not describe how masking was maintained or assessed. The specific randomization sequence-generation and allocation-concealment procedures are not reported in the registry record. A detailed protocol, sample-size justification, and a complete statistical analysis plan are also not reported in the registry record, so those procedures should not be inferred from the design label or the posted results.
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: Twenty-seven patients in the placebo + erlotinib arm with disease progression in the blinded treatment stage elected to receive MetMAb + erlotinib in the optional open-label phase of the study.
Overall Study
| Stage | MetMAb + Erlotinib | Placebo + Erlotinib |
|---|---|---|
| STARTED | 69 | 68 |
| Received Study Drug | 69 | 67 |
| COMPLETED | 7 | 1 |
| NOT COMPLETED | 62 | 67 |
Reasons for not completing this period:
| Reason | MetMAb + Erlotinib | Placebo + Erlotinib |
|---|---|---|
| Disease progression | 42 | 50 |
| Adverse Event | 8 | 3 |
| Death | 2 | 5 |
| Physician Decision | 6 | 6 |
| Withdrawal by Subject | 4 | 2 |
| Sponsor's decision to terminate study | 0 | 1 |
Analysis population for the primary outcome, as defined in the registry: All randomized intent-to-treat patients.
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.
| Characteristic | Unit | MetMAb + Erlotinib | Placebo + Erlotinib | Total |
|---|---|---|---|---|
| Participants analysed | 69 | 68 | 137 | |
| Age, Continuous | years | 63.6 (SD 9.4) | 62.7 (SD 10.6) | 63.1 (SD 10.0) |
| Sex: Female, Male: Female | Participants | 29 | 26 | 55 |
| Sex: Female, Male: Male | Participants | 40 | 42 | 82 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-free Survival, measured over Time from randomization to the first occurrence of progression/relapse or death on study. (Up to 20 months).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | months |
| Endpoint classification | not reduced to a single comparison |
Progression-free survival measures the time from randomization until the first recorded progression or relapse, or death on study. In this trial, progression or relapse was defined using RECIST 1.0 and assessed by the site radiologist or investigator. Death from any cause was included if it occurred within 30 days of the last treatment. The registry gives the outcome timeframe as up to 20 months.
This endpoint is used in cancer trials because it captures time until disease worsening or death, rather than only whether a participant experiences a response at a particular assessment. A time-to-event analysis can account for participants whose event time is not observed during the available follow-up through censoring, although the registry record does not explain how censoring was handled in detail for this analysis.
The registry reports arm-level progression-free survival values of 2.2 months for MetMAb plus erlotinib and 2.6 months for placebo plus erlotinib, with respective 95% confidence intervals of 1.38 to 2.86 and 1.45 to 2.76. It does not label the precise summary statistic represented by those arm-level values, so they should not be described more specifically than the registry permits. The reported between-arm analysis is a log-rank analysis with a hazard ratio and confidence interval.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary analysis reports a hazard ratio of 1.086 for the comparison of MetMAb plus erlotinib with placebo plus erlotinib. Its 95% confidence interval is 0.727 to 1.622, and the reported p-value is 0.687. The interval spans the value representing equal hazards, and the estimate is compatible with different directions of association. The record therefore does not establish a clear between-arm difference in progression-free survival.
The arm-level values are 2.2 months for MetMAb plus erlotinib and 2.6 months for placebo plus erlotinib. Their confidence intervals overlap: 1.38 to 2.86 for the MetMAb arm and 1.45 to 2.76 for the placebo arm. Those arm-level summaries do not by themselves provide the same information as the log-rank comparison, which uses the event-time data. The record does not identify the summary statistic behind the arm-level values, so a more specific interpretation is not warranted.
A confidence interval describes a range of estimates compatible with the data and the statistical method, under its assumptions. In this case, the interval around the hazard ratio is wide enough to include both lower and higher hazards for the MetMAb arm relative to placebo. The p-value likewise does not indicate strong evidence against the null hypothesis of no difference under the reported analysis. Neither quantity proves that the treatments are equivalent, and neither establishes that a clinically meaningful difference is absent.
The primary analysis is the most direct registered comparison for the trial's principal question. The secondary objective-response outcomes show a similar broad caution: their point estimates are numerically higher in the MetMAb arm, but those estimates are uncertain and do not provide a basis for overriding the primary time-to-event result. The record does not report a multiplicity adjustment, which further limits how much weight should be placed on separate secondary findings.
These results describe the posted trial data and their uncertainty. They do not determine the effect in every patient group, establish the reason for any observed pattern, or support a treatment recommendation. Details such as the handling of missing outcome data and the full analysis plan are not reported in the registry record, so the results cannot be fully audited from the registry entries alone.
9. Statistical Methods Explained
The primary comparison uses a log-rank analysis and reports a hazard ratio (HR). In general, a hazard ratio compares the instantaneous event rates between groups over follow-up, subject to the assumptions of the model and analysis. An HR of 1 corresponds to equal hazards; a value on either side represents a different estimated hazard in the numerator group relative to the reference group. The registry reports an HR of 1.086, with a 95% confidence interval of 0.727 to 1.622.
A confidence interval expresses uncertainty about an estimated quantity under the statistical method. A narrower interval indicates more precision, while a wider interval leaves more possible values compatible with the data. Here the interval includes 1, the no-difference value for a hazard ratio, and allows estimates on both sides of that value. The p-value of 0.687 is the reported test result for the log-rank comparison; it is not the probability that either treatment is effective, nor does it measure the probability that the null hypothesis is true.
The log-rank test is generally used to compare time-to-event experience between groups across follow-up. Unlike a simple comparison of arm-level summaries, it draws on the timing of events and the participants still under observation at each event time. Its interpretation depends on the quality and completeness of event-time information and on the analysis procedures. The registry reports the method but does not report detailed censoring rules, diagnostic checks, or additional model assumptions.
For objective response, the registry reports percentages with 95% confidence intervals and a Cochran-Mantel-Haenszel analysis. In general, this method can compare categorical outcomes between groups while accounting for strata, but the registry entry does not specify the stratification details. A separate simple recalculation from the posted summary counts gives an unadjusted risk difference and an unadjusted p-value. These are not a substitute for the sponsor's reported analysis: the recalculation does not reproduce any stratification or other adjustment that may be part of the sponsor's method.
For the overall objective-response comparison, the simple unadjusted recalculation gives a risk difference of 0.014, with a 95% confidence interval from -0.060 to 0.087 and an unadjusted p-value of 0.713. For the MET diagnostic-positive subgroup, the recalculated risk difference is 0.053, with a 95% confidence interval from -0.058 to 0.165 and an unadjusted p-value of 0.364. These intervals include no difference. The calculated values describe unadjusted comparisons of the posted summary counts; they should not be conflated with the Cochran-Mantel-Haenszel p-values reported in the registry.
The sample-size illustrations are planning calculations based on the observed effect, not the trial's own design or evidence of achieved power. The stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation. The illustrations consider detecting stated fractions of the observed effect. In general, planning for a smaller effect requires more participants than planning for a larger effect, all else equal. These calculations do not show that the observed study had adequate power, and they do not establish the effect size that a future study should expect.
Over-dispersion is a general concern when count outcomes vary more than a model assumes; ignoring it can make standard errors too small and uncertainty look understated. The primary outcome is time to event, and the posted secondary outcomes include binary response percentages. The registry record does not report an over-dispersion assessment or a count-outcome model, so that concept should not be used to infer an unreported analysis choice.
What does the uncertainty say?
Both the primary hazard-ratio interval and the recalculated response risk-difference intervals leave room for effects in more than one direction. This means the estimates are not precise enough to distinguish confidently between possible benefit, little difference, or possible disadvantage. The intervals are not proof of equivalence: equivalence requires a prespecified margin and an analysis designed for that question, neither of which is reported in the registry record.
What can be checked from the registry?
The registry identifies the primary outcome definition, the intent-to-treat population, the reported log-rank method, and a reported hazard ratio with its confidence interval and p-value. It also supplies arm-level outcome summaries and secondary outcome analyses. The statistical analysis plan, detailed missing-data procedures, and multiplicity approach are not reported in the registry record. Those omissions prevent a complete assessment of whether all analysis choices followed a prespecified plan.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Log Rank | Hazard Ratio (HR) 1.086 | 0.727 to 1.622 | 0.687 |
The registry reports the sponsor's primary log-rank analysis as an HR of 1.086, with a 95% confidence interval of 0.727 to 1.622 and p = 0.687. The simple recalculation described for binary outcomes does not apply to this time-to-event primary endpoint, so it does not provide an alternative primary HR or p-value. There is no recalculated primary result to compare with the sponsor's analysis.
For objective response, the sponsor-reported Cochran-Mantel-Haenszel p-values are 0.710 for the overall population and 0.367 for the MET diagnostic-positive subgroup. The simple unadjusted recalculations give p-values of 0.713 and 0.364, respectively. These small differences are consistent with the fact that the calculations use different methods: the registry identifies the sponsor method as Cochran-Mantel-Haenszel, while the recalculations are unadjusted comparisons from posted summary counts.
The recalculated results are useful as a check on the posted summary counts, but they are not a replacement for the sponsor's analysis or for any prespecified stratified analysis. The registry record does not provide enough detail to reproduce the sponsor's analysis fully. The record flags no discrepancy. Where a recalculation differs because it uses a simpler method, the sponsor's reported analysis remains the relevant registered inferential result.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists 5 outcome measures and 4 analyses, with 3 named secondary outcomes. Considering multiple outcomes or subgroup analyses can increase the chance of a false-positive finding if each is interpreted using an unadjusted threshold. A prespecified hierarchy, family-wise error control, or another multiplicity strategy can help address that issue. Such a strategy is not reported in the registry record, so the secondary p-values should not be read as if multiplicity had been controlled.
The primary outcome population is specified as all randomized intent-to-treat patients. This indicates that the primary analysis is intended to include participants according to randomized assignment rather than according to treatment received. The registry does not give a detailed account of exclusions, protocol deviations, or the handling of participants who did not contribute evaluable outcome data. Those details are not reported in the registry record.
Progression-free survival requires a rule for participants whose progression, relapse, or death is not observed before follow-up ends or who stop contributing information. Such observations are commonly censored in time-to-event analysis. The outcome definition and timeframe are recorded, but detailed censoring procedures and the extent of missing time-to-event data are not reported in the registry record. The registry also does not report how missing response assessments were handled.
The sample-size justification and assumptions underlying the original trial design are not reported in the registry record. The posted planning illustrations are explicitly based on the observed effect and are not a substitute for that missing design information. Interim analyses, stopping rules, and a complete statistical analysis plan are not reported in the registry record either. Their absence from the registry entry does not establish that they were absent from trial conduct; it means they cannot be evaluated from the registered information alone.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| MetMAb + Erlotinib | 29 | 69 | 42.0% |
| Placebo + Erlotinib | 22 | 67 | 32.8% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| MetMAb + Erlotinib | 68 of 69 |
| Placebo + Erlotinib | 67 of 67 |
Most frequent serious adverse event terms
| Event term | Organ system | MetMAb + Erlotinib | Placebo + Erlotinib |
|---|---|---|---|
| PNEUMONIA | Infections and infestations | 4 of 69 | 1 of 67 |
| PULMONARY EMBOLISM | Respiratory, thoracic and mediastinal disorders | 4 of 69 | 1 of 67 |
| ANAEMIA | Blood and lymphatic system disorders | 1 of 69 | 2 of 67 |
| DYSPNOEA | Respiratory, thoracic and mediastinal disorders | 3 of 69 | 0 of 67 |
| RASH | Skin and subcutaneous tissue disorders | 2 of 69 | 1 of 67 |
| LUNG INFECTION | Infections and infestations | 1 of 69 | 1 of 67 |
| DEHYDRATION | Metabolism and nutrition disorders | 1 of 69 | 1 of 67 |
| CHRONIC OBSTRUCTIVE PULMONARY DISEASE | Respiratory, thoracic and mediastinal disorders | 1 of 69 | 1 of 67 |
| HAEMOPTYSIS | Respiratory, thoracic and mediastinal disorders | 2 of 69 | 0 of 67 |
| INTERSTITIAL LUNG DISEASE | Respiratory, thoracic and mediastinal disorders | 0 of 69 | 2 of 67 |
Most frequent non-serious adverse event terms
| Event term | Organ system | MetMAb + Erlotinib | Placebo + Erlotinib |
|---|---|---|---|
| RASH | Skin and subcutaneous tissue disorders | 42 of 69 | 41 of 67 |
| DIARRHOEA | Gastrointestinal disorders | 28 of 69 | 35 of 67 |
| FATIGUE | General disorders | 22 of 69 | 24 of 67 |
| NAUSEA | Gastrointestinal disorders | 22 of 69 | 21 of 67 |
| DECREASED APPETITE | Metabolism and nutrition disorders | 14 of 69 | 16 of 67 |
| DYSPNOEA | Respiratory, thoracic and mediastinal disorders | 13 of 69 | 16 of 67 |
| COUGH | Respiratory, thoracic and mediastinal disorders | 13 of 69 | 13 of 67 |
| OEDEMA PERIPHERAL | General disorders | 16 of 69 | 5 of 67 |
| ANAEMIA | Blood and lymphatic system disorders | 10 of 69 | 10 of 67 |
| DERMATITIS ACNEIFORM | Skin and subcutaneous tissue disorders | 10 of 69 | 10 of 67 |
Serious events with the highest arm-level rate: pneumonia (5.8%), pulmonary embolism (5.8%), dyspnoea (4.3%), anaemia (3.0%), interstitial lung disease (3.0%).
The registry records serious adverse events for 29 of 69 participants at risk in the MetMAb-plus-erlotinib group and 22 of 67 participants at risk in the placebo-plus-erlotinib group. These figures describe recorded serious adverse events in the respective safety groups. They are not the same as the primary outcome analysis, and the safety denominators differ from the posted primary outcome arm counts for the placebo group.
The registry lists pneumonia and pulmonary embolism at 5.8% each, dyspnoea at 4.3%, and anaemia, interstitial lung disease, and respiratory distress at 3.0% each among the top serious-event terms. The record does not provide enough detail in these entries to assign those listed percentages to a particular arm. Serious-event definitions, event-specific timing, and a full assessment of causality are not reported in the registry record.
These figures should be read as descriptive safety data, not as proof that a particular event was caused by a study drug or as a complete comparison of overall safety. Death counts are not reported in the registry record. Differences in the number of participants at risk and the absence of detailed event-level information further limit direct interpretation.
13. What the Result Does, and Does Not, Mean
The registry's primary result is an uncertain comparison of progression-free survival between MetMAb plus erlotinib and placebo plus erlotinib. The reported HR is 1.086, and its 95% confidence interval, 0.727 to 1.622, includes the no-difference value. The log-rank p-value is 0.687. The data therefore do not show a clear difference under the reported primary analysis, but they also do not establish that the regimens have equivalent effects.
The arm-level progression-free survival values are 2.2 months and 2.6 months, but the registry does not identify the precise summary statistic behind those values. They should not be over-interpreted as a specific type of estimate. The primary time-to-event comparison is the more direct registered test of the trial question.
Objective-response estimates are numerically higher in the MetMAb arm, but their uncertainty and the absence of a reported multiplicity strategy limit the strength of that pattern. Secondary outcomes can inform questions for further study, but they do not independently confirm the primary outcome. Nor do these registry results establish what would happen in a different population, treatment setting, or protocol.
Statistical uncertainty is not a statement that there is no effect. It means the available data do not distinguish the possible effects with enough precision to support a firm conclusion. Conversely, a numerical difference in point estimates is not proof of a real or clinically important effect. The trial's registry results should be interpreted as evidence from this study, with the reported methods and the limitations of the available information, rather than as a treatment recommendation.
14. Why This Trial Matters Statistically
This trial is statistically informative because its primary question is framed as a randomized comparison with a shared background treatment. Erlotinib was used in both groups, while the randomized contrast was MetMAb versus placebo. That design makes the between-arm comparison central to understanding the added-treatment question, provided randomization and follow-up were implemented as intended.
The primary analysis also illustrates why time-to-event outcomes and arm-level summaries should not be treated as interchangeable. The registry gives arm-level progression-free survival values as well as a log-rank hazard-ratio analysis. The log-rank comparison uses event timing across follow-up, while a single arm-level summary cannot convey the full pattern of events and censoring. The record does not identify the exact summary statistic for the arm-level values, which is an important reason to avoid over-interpreting them.
The objective-response outcomes illustrate a second statistical issue: numerical differences can coexist with considerable uncertainty. The response point estimates lean in the same direction in the posted populations, yet the recalculated intervals include no difference, and no multiplicity strategy is reported. The pattern is suitable for cautious description, not a definitive claim based on secondary testing.
Finally, planning illustrations based on observed effects do not establish the adequacy of the original design. They answer a conditional planning question under stated assumptions, whereas the original sample-size rationale is not reported in the registry record. Separating observed results, uncertainty, and hypothetical planning requirements is essential to avoid mistaking a post hoc illustration for evidence about achieved power or a prespecified design target.
15. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The registry record does not provide a complete statistical analysis plan, a sample-size justification, or detailed information on interim analyses and stopping rules. It also does not report the randomization sequence-generation or allocation-concealment procedures. These omissions limit assessment of prespecification and trial conduct.
The exact summary statistic used for the arm-level progression-free survival values is not identified in the record. Although the primary endpoint definition and the log-rank hazard-ratio result are available, detailed censoring procedures, event counts, and missing-data handling are not reported in the registry record. These details are needed to assess how the time-to-event analysis was implemented and how robust it may be to incomplete follow-up.
The objective-response analyses have limited precision, particularly in the MET diagnostic-positive subgroup, and the registry does not report a multiplicity approach. The simple recalculations use posted summary counts and are unadjusted; they cannot reproduce the sponsor's Cochran-Mantel-Haenszel analysis or any unreported stratification. Duration of overall response is listed as a secondary outcome, but its results are not included in the registry record for this analysis.
The safety entries provide serious-event counts and a list of top terms with percentages, but the details do not support a complete arm-specific assessment of each listed event. Death counts are not reported in the registry record. More broadly, the registry results alone do not establish causality for adverse events or explain how every participant's outcome was handled.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: A Study of MetMAb Administered to Patients With Advanced Non-Small Cell Lung Cancer, in Combination With Tarceva (Erlotinib) — NCT00854308.
- Registry record last updated 2017-03-31; status last verified 2012-11.
- Results first posted 2011-10-18.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT00854308 is a completed randomized, double-blind, multicenter Phase II trial of MetMAb plus erlotinib versus placebo plus erlotinib for advanced non-small cell lung cancer. Enrollment is listed as 137, and the primary outcome is progression-free survival in all randomized intent-to-treat patients. The registry reports an HR of 1.086, with a 95% confidence interval of 0.727 to 1.622 and p = 0.687.
The primary comparison does not establish a clear difference between groups. Objective-response point estimates are numerically higher in the MetMAb arm, but those secondary results are uncertain and no multiplicity strategy is reported. Important details on the analysis plan, missing data, and safety-event attribution are not reported in the registry record.