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 Chemotherapy for Patients With Non-Small Cell Lung Cancer Who Are Non-Smokers (NCT00409006), a randomized phase 2 study. The primary outcome is Progression-Free Survival (PFS).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00409006 |
| Phase | Phase 2 |
| Condition | Non-small Cell Lung Cancer |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 70 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Eli Lilly and Company |
| Results posted | Yes |
2. Clinical Question
The central question is whether progression-free survival differed between two treatment sequences for Asian never-smoker patients with advanced non-small cell lung cancer receiving first-line treatment. One group received pemetrexed and cisplatin followed by gefitinib; the comparison group received pemetrexed and cisplatin followed by pemetrexed-based treatment. The registry describes the trial as a comparison of chemotherapy followed by gefitinib with chemotherapy alone.
Population
Asian patients classified as never smokers with advanced non-small cell lung cancer. The registry describes the primary analysis population as randomized and treated patients, including all randomized patients.
Intervention
Pemetrexed and cisplatin, followed by gefitinib. The registry describes the chemotherapy as intravenous treatment once every 3 weeks for 4 cycles, followed by daily oral gefitinib until progression or unacceptable toxicity.
Comparator
Pemetrexed and cisplatin, followed by pemetrexed, with optional cisplatin for up to 2 additional cycles. Further details of this arm's regimen are not reported in the registry record.
Primary question
How did median progression-free survival compare between the two groups, and how much uncertainty surrounds the estimates and the reported log-rank comparison?
The registry's stated rationale was that never-smoker patients might be more likely to respond favorably to gefitinib. That rationale frames the research question; it is not evidence that the treatment sequence produced a particular outcome. The trial also listed tumor response, duration of response, and overall survival as secondary outcomes. The posted results include a tumor-response comparison, while results for duration of response and overall survival are not reported in the registry record.
3. Trial Design
This was a randomized, parallel-group phase 2 trial with no masking. The registry lists enrollment of 70 participants and identifies Eli Lilly and Company as the sponsor. The study status is completed. These design features describe how the comparison was organized; they do not establish how precisely the trial could estimate treatment differences.
The two groups received pemetrexed and cisplatin initially. The experimental sequence then used gefitinib, while the comparator sequence continued with pemetrexed and could include cisplatin for up to 2 additional cycles. The registry gives doses and schedules for the initial chemotherapy and the gefitinib sequence. The comparator arm description ends before its full treatment instructions are stated, so additional regimen details are not reported in the registry record.
For the primary endpoint, the posted group sizes are 39 for pemetrexed/cisplatin/gefitinib and 31 for pemetrexed/cisplatin. The primary outcome is reported as median progression-free survival in months, with a 95% confidence interval for each group. The registry says the endpoint runs from randomization to the first observation of progression or death from any cause. Its stated timeframe is baseline to progression or death, 12 weeks up to 31 months.
The registry labels the primary analysis population randomized and treated, says it includes all randomized patients, and also states that patients are analyzed according to the treatment they actually received, describing this as an intent-to-treat analysis. Those statements do not fully clarify how treatment assignment and analysis population were operationalized. The record also reports censoring of 8 patients in the pemetrexed/cisplatin/gefitinib group and 11 in the pemetrexed/cisplatin group. More detail about randomization procedures, allocation concealment, and the handling of censoring is not reported in the registry record.
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: 86 participants entered study. 13 participants were screen failures. 73 participants (40 pemetrexed/cisplatin/gefitinib and 33 pemetrexed/cisplatin) were assigned to treatment. 3 participants did not receive study drug. These 16 participants (13 screen failures and 3 that did not receive drug) were not included in the analyses.
Overall Study
| Stage | Pemetrexed/Cisplatin/Gefitinib | Pemetrexed/Cisplatin |
|---|---|---|
| STARTED | 39 | 31 |
| COMPLETED | 19 | 19 |
| NOT COMPLETED | 20 | 12 |
Reasons for not completing this period:
| Reason | Pemetrexed/Cisplatin/Gefitinib | Pemetrexed/Cisplatin |
|---|---|---|
| Death | 15 | 6 |
| Lost to Follow-up | 1 | 2 |
| Withdrawal by Subject | 3 | 2 |
| Physician Decision | 1 | 2 |
Analysis population for the primary outcome, as defined in the registry: Randomized and treated (RT) population includes all randomized patients. Patients are analyzed according to the treatment they actually received (intent-to-treat \[ITT\] analysis). Patients were censored from this analysis (8 pemetrexed/cisplatin/gefitinib and 11 pemetrexed/cisplatin).
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 | Pemetrexed/Cisplatin/Gefitinib | Pemetrexed/Cisplatin | Total |
|---|---|---|---|---|
| Participants analysed | 39 | 31 | 70 | |
| Age Continuous | years | 55.6 (SD 8.45) | 55.7 (SD 12.43) | 55.7 (SD 10.32) |
| Sex: Female, Male: Female | Participants | 30 | 25 | 55 |
| Sex: Female, Male: Male | Participants | 9 | 6 | 15 |
| Race/Ethnicity, Customized: East Asian | Participants | 39 | 31 | 70 |
| Region of Enrollment: Taiwan | participants | 10 | 11 | 21 |
| Region of Enrollment: China | participants | 16 | 14 | 30 |
| Region of Enrollment: Korea, Republic of | participants | 13 | 6 | 19 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 0 | Units on a scale | 15 | 9 | 24 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 1 | Units on a scale | 24 | 22 | 46 |
| History of Smoking: Yes | Participants | 0 | 2 | 2 |
| History of Smoking: No | Participants | 39 | 29 | 68 |
| Basis of Diagnosis: Histopathological | Participants | 30 | 22 | 52 |
| Basis of Diagnosis: Cytological | Participants | 9 | 9 | 18 |
| Pathological Diagnosis: Mixed Cell Carcinoma, Lung | Participants | 1 | 1 | 2 |
| Pathological Diagnosis: Adenocarcinoma | Participants | 30 | 24 | 54 |
| Pathological Diagnosis: Carcinoma, Squamous Cell | Participants | 7 | 4 | 11 |
| Pathological Diagnosis: Non-Small Cell Lung Carcinoma | Participants | 1 | 2 | 3 |
| Disease Stage: Stage IIIB | Participants | 6 | 4 | 10 |
| Disease Stage: Stage IV | Participants | 33 | 27 | 60 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-Free Survival (PFS), measured over Baseline to first observation of disease progression or death, 12 weeks up to 31 months.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | Months |
| Endpoint classification | Time-to-Event Analysis |
Progression-free survival measures the time from randomization until the first recorded disease progression or death from any cause. It combines these events into a time-to-event outcome: a participant's time is counted until one of the specified events occurs, or until the participant is censored if the event has not been observed during follow-up.
The endpoint is useful in a trial of cancer treatment because it reflects how long participants remain alive without documented progression. It differs from a response measure, which records whether a tumor meets a response definition at an assessment, and from overall survival, which concerns time to death. Progression-free survival can therefore describe a distinct aspect of the clinical course, although it does not by itself answer every question about symptoms, quality of life, or survival.
The registry reports median progression-free survival for each group and a 95% confidence interval around each median. A median is the point in the estimated event-time distribution by which half the participants have experienced the event. The confidence interval communicates uncertainty in that estimate; it is not a range containing the progression-free survival times of all participants. The registry's definition includes death due to any cause and specifies time from randomization, making the endpoint's event and starting point explicit.
Because some participants were censored, the median is interpreted alongside the time-to-event comparison and the follow-up information. The exact censoring rules and additional details of the endpoint assessment are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
Median progression-free survival was 9.95 months in the pemetrexed/cisplatin/gefitinib group, with a 95% confidence interval from 5.85 to 16.49 months. In the pemetrexed/cisplatin group, the median was 6.83 months, with a 95% confidence interval from 5.78 to 7.98 months. These are group-level estimates, and their confidence intervals indicate uncertainty around each median. The intervals overlap, so the medians alone do not establish a clear between-group difference.
The registry reports a log-rank p-value of 0.062 for the primary comparison. This is close to, but above, the 0.05 significance level used in the posted planning assumptions. It is not strong evidence against the null hypothesis under that conventional threshold. At the same time, a p-value above that threshold does not prove the groups are equivalent, nor does it show that any possible difference is absent. The trial's estimates and their uncertainty remain important to interpretation.
The confidence interval for the gefitinib-sequence median is comparatively broad, extending from 5.85 to 16.49 months. That breadth signals substantial uncertainty about the location of that median. The comparator median's interval is narrower, but precision of an individual median is not a substitute for an estimate and interval for the between-group effect. A hazard ratio or confidence interval for a direct time-to-event effect is not reported in the registry record.
The primary result should be read as an estimate from a small phase 2 randomized comparison, not as a definitive answer about all patients with this disease. The enrollment was 70, and the primary analysis includes censored participants. Differences in follow-up, event timing, and censoring can affect time-to-event estimates. The registry's statement that the analysis follows treatment actually received, while also describing the population as including all randomized patients and calling it intent-to-treat, leaves some ambiguity about the analysis population.
The secondary tumor-response comparison is less conclusive than a simple comparison of response counts might suggest: its recalculated risk-difference interval spans no difference, and the sponsor's reported logistic-regression p-value is 0.369. Results for duration of response and overall survival are not reported in the registry record. Taken together, the posted information supports a cautious reading: the two median estimates differ numerically, but the reported primary comparison does not meet the stated conventional threshold, and uncertainty remains substantial.
9. Statistical Methods Explained
Progression-free survival is a time-to-event endpoint because it records time until progression or death. Time-to-event methods are designed to use information from participants who have experienced an event as well as participants whose event time is not observed during the available follow-up. A participant without an observed event may be censored: the analysis uses the time through which the participant's status is known, without treating that time as if it were an event.
What does the median describe?
The median summarizes the estimated distribution of event times. It is the time by which half of the group is estimated to have experienced progression or death. The reported medians are 9.95 months for pemetrexed/cisplatin/gefitinib and 6.83 months for pemetrexed/cisplatin. These summaries are not averages of every participant's observed follow-up, and they do not reveal how event times are distributed across the entire follow-up period.
Each median has a 95% confidence interval: 5.85 to 16.49 months for the gefitinib sequence and 5.78 to 7.98 months for the comparison sequence. In general, a confidence interval describes the precision of an estimate under the statistical method used. A wider interval indicates greater uncertainty about the corresponding estimate. The intervals are for the separate group medians; they are not a confidence interval for a difference or ratio between groups.
What does the log-rank test assess?
The registry reports a log-rank comparison. In general, the log-rank test compares the event-time experience of groups across follow-up, using information about when events occur and accounting for censoring under its assumptions. It tests evidence against a null hypothesis of no difference in the event-time distributions. Its p-value does not measure the size of a treatment effect, the probability that the null hypothesis is true, or the chance that the result will replicate.
The reported log-rank p-value is 0.062. The posted planning assumptions use a two-sided significance level of 0.05, so this comparison is above that threshold. This classification should not be confused with proof of equivalence or absence of an effect. A p-value also does not replace an estimate of the between-group effect with its confidence interval. A hazard ratio and its confidence interval are not reported in the registry record, so the log-rank p-value and the separate median estimates are the available primary comparison information.
Why does censoring matter?
The registry reports that 8 participants in the pemetrexed/cisplatin/gefitinib group and 11 in the pemetrexed/cisplatin group were censored from the primary analysis. Censoring is not the same as an observed progression or death. It indicates that the event time was not observed in the analysis, while some follow-up information may still contribute. Time-to-event methods rely on assumptions about censoring; the detailed censoring rules and reasons are not reported in the registry record. If censoring is related to prognosis in ways the analysis does not account for, estimates can be difficult to interpret.
What does the secondary recalculation add?
For tumor response, the posted recalculation is an unadjusted risk difference based on the reported group summaries. It is not an adjusted model estimate. The recalculated comparison has a risk difference of -0.107, with a confidence interval from -0.337 to 0.123, and an unadjusted p-value of 0.368. The interval includes zero, which is the value corresponding to no risk difference. The sponsor's logistic-regression comparison reports p = 0.369. These analyses use different methods, so close but not identical p-values are not unexpected.
These recalculated quantities are descriptive checks based on posted summary statistics, not replacements for the sponsor's analysis. They do not adjust for covariates or other trial features. A confidence interval describes uncertainty under the method used, and the unadjusted p-value assesses compatibility with a null comparison under that method. Neither establishes that a treatment sequence caused an observed difference.
How should planning illustrations be read?
The planning illustrations use the observed tumor-response effect as a reference, then consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The illustrations assume a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation. For planning for 100% of the observed effect, the estimates are 332 participants per arm for 80% power and 444 participants per arm for 90% power. For 75% of the observed effect, they are 597 and 799 participants per arm, respectively. For 50% of the observed effect, they are 1,355 and 1,813 participants per arm, respectively.
These are planning illustrations, not observed power and not the trial's original sample-size justification. The sample-size justification and its assumptions are not reported in the registry record. Planning estimates based on an observed effect can be unstable, particularly when that effect comes from a small study; smaller assumed effects require larger sample sizes under the stated assumptions. The illustrations should therefore be treated as conditional scenarios, not as evidence about the trial's probability of detecting an effect.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Log Rank | not reported | 0.062 |
For progression-free survival, the registry reports the sponsor's log-rank comparison with p = 0.062. The primary posted information consists of the group medians, their 95% confidence intervals, and this log-rank p-value. A recalculated hazard ratio or other direct time-to-event effect is not reported in the registry record. The time-to-event interpretation therefore rests on the reported sponsor comparison and the separate group estimates, rather than on a reconstructed hazard ratio.
For tumor response, the sponsor reports a logistic-regression p-value of 0.369. The unadjusted recalculation from the posted summary statistics gives a risk-difference estimate of -0.107, a confidence interval from -0.337 to 0.123, and p = 0.368. The small difference between the p-values is consistent with the use of different methods: logistic regression for the sponsor's analysis and an unadjusted comparison for the recalculation. Neither result is adjusted for other variables in the unadjusted recalculation.
The registry does not flag a discrepancy between the posted sponsor results and recalculations. The sponsor's analyses and the recalculated summaries should still be understood as answering related but not identical statistical questions. The sponsor's model-based result is the posted trial analysis; the recalculation is a simple check using the available group summaries and should not be treated as a substitute for the sponsor's analysis.
11. Multiplicity, Missing Data and Analysis Populations
Multiplicity arises when a study makes several statistical comparisons or examines several outcomes. In general, each additional test creates another opportunity for a result that appears unusual under a null hypothesis, even when no true effect exists. Methods that control multiplicity can limit that risk, but they may also affect how evidence is judged. The statistical plan for handling multiple outcomes or comparisons is not reported in the registry record.
The primary outcome is progression-free survival, and the registry also lists tumor response, duration of response, and overall survival as secondary outcomes. A sponsor analysis is posted for the primary comparison and for tumor response. Results for duration of response and overall survival are not reported in the registry record. It is not reported whether the primary endpoint was the only confirmatory test, how secondary analyses were ordered, or whether any adjustment was planned.
Missing information about multiplicity is important when interpreting secondary findings. An unadjusted p-value for one outcome does not account for other tests that may have been conducted. The recalculated tumor-response p-value is explicitly unadjusted. The sponsor's statistical approach to multiplicity is not reported in the registry record, so it is not possible to determine from the posted record whether the sponsor controlled the overall false-positive risk across outcomes.
There is also incomplete information about missing outcome data and analysis populations. The registry says that 8 participants in the gefitinib-sequence group and 11 in the comparison group were censored in the primary analysis. The reasons for censoring and detailed handling of participants with incomplete follow-up are not reported in the registry record. The description of the primary population says it includes all randomized patients, but also says patients are analyzed according to the treatment they actually received and calls the approach intent-to-treat. Those phrases do not fully resolve whether the analysis followed randomized assignment or treatment received.
Interim analyses, stopping rules, and a detailed statistical analysis plan are not reported in the registry record. Their absence from the record does not establish that they were or were not used. It limits how fully the analysis can be assessed and reinforces the need to interpret the posted results as an incomplete account of the statistical procedures.
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 |
|---|---|---|---|
| Pemetrexed/Cisplatin/Gefitinib | 6 | 39 | 15.4% |
| Pemetrexed/Cisplatin | 3 | 31 | 9.7% |
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 |
|---|---|
| Pemetrexed/Cisplatin/Gefitinib | 38 of 39 |
| Pemetrexed/Cisplatin | 31 of 31 |
Most frequent serious adverse event terms
| Event term | Organ system | Pemetrexed/Cisplatin/Gefitinib | Pemetrexed/Cisplatin |
|---|---|---|---|
| Nausea | Gastrointestinal disorders | 1 of 39 | 1 of 31 |
| Vomiting | Gastrointestinal disorders | 1 of 39 | 1 of 31 |
| Headache | Nervous system disorders | 1 of 39 | 1 of 31 |
| Death | General disorders | 1 of 39 | 0 of 31 |
| Pyrexia | General disorders | 1 of 39 | 0 of 31 |
| Acute tonsillitis | Infections and infestations | 1 of 39 | 0 of 31 |
| Infection | Infections and infestations | 0 of 39 | 1 of 31 |
| Anorexia | Metabolism and nutrition disorders | 1 of 39 | 0 of 31 |
| Myalgia | Musculoskeletal and connective tissue disorders | 1 of 39 | 0 of 31 |
| Coma | Nervous system disorders | 1 of 39 | 0 of 31 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Pemetrexed/Cisplatin/Gefitinib | Pemetrexed/Cisplatin |
|---|---|---|---|
| Nausea | Gastrointestinal disorders | 27 of 39 | 20 of 31 |
| Vomiting | Gastrointestinal disorders | 19 of 39 | 15 of 31 |
| Anorexia | Metabolism and nutrition disorders | 19 of 39 | 15 of 31 |
| Neutrophil count decreased | Investigations | 8 of 39 | 13 of 31 |
| Haemoglobin decreased | Investigations | 8 of 39 | 11 of 31 |
| Pruritus | Skin and subcutaneous tissue disorders | 14 of 39 | 5 of 31 |
| Rash | Skin and subcutaneous tissue disorders | 13 of 39 | 6 of 31 |
| Fatigue | General disorders | 10 of 39 | 5 of 31 |
| White blood cell count decreased | Investigations | 6 of 39 | 8 of 31 |
| Dyspnoea | Respiratory, thoracic and mediastinal disorders | 6 of 39 | 8 of 31 |
Serious events with the highest arm-level rate: nausea (3.2%), vomiting (3.2%), infection (3.2%), headache (3.2%), asthma (3.2%).
The posted serious adverse event figures report 6 affected participants among 39 at risk in the pemetrexed/cisplatin/gefitinib group and 3 among 31 at risk in the pemetrexed/cisplatin group. The figures for other adverse events are 38 among 39 at risk and 31 among 31 at risk, respectively. These are counts of participants, not a comparison adjusted for exposure time or other differences between groups. The registry does not report deaths in these safety summaries.
The listed serious adverse event terms are nausea, vomiting, infection, headache, asthma, and pneumonitis, each reported as 3.2%. A listed term is not necessarily a complete account of every event or its clinical course. The registry summary does not specify event duration, recurrence, severity beyond the serious classification, or treatment attribution for each term. The figures describe what was posted and do not by themselves establish that a treatment caused an event or that one sequence has a different overall safety profile.
Safety counts should be read alongside the size and duration of exposure in each group and the rules used to collect and classify adverse events. Additional detail about those aspects is not reported in the registry record. The posted numbers are descriptive and do not support medical advice.
13. What the Result Does, and Does Not, Mean
The result describes a randomized phase 2 comparison in Asian never-smoker patients with advanced non-small cell lung cancer. The group medians for progression-free survival were 9.95 months with pemetrexed/cisplatin followed by gefitinib and 6.83 months with pemetrexed/cisplatin. The sponsor's log-rank comparison has a p-value of 0.062, above the 0.05 significance level used in the posted planning assumptions. These findings do not establish a statistically significant primary comparison under that threshold.
That conclusion is not the same as proving that the treatments are equivalent, or that no difference exists. A non-significant result can occur when the true difference is small, when estimates are imprecise, or when a study has limited information. The confidence intervals around the medians—particularly the interval from 5.85 to 16.49 months for the gefitinib sequence—show that the estimated group medians are uncertain. The registry does not report a direct hazard-ratio estimate and interval that could quantify a relative time-to-event effect.
The findings also do not establish how the sequences compare for every outcome. Tumor response is a separate measure, while duration of response and overall survival results are not reported in the registry record. The listed secondary analyses do not replace the primary endpoint, and no multiplicity strategy is reported. Safety summaries are descriptive counts and do not determine causation.
Generalizability is limited to the population and treatment context studied. The registry identifies Asian never-smoker patients with advanced disease; results should not be assumed to apply to other populations or settings. The record also leaves some uncertainty about the analysis population and censoring details. The evidence is best understood as the posted estimates and their uncertainty from this particular trial, not as a universal conclusion about treatment.
14. Why This Trial Matters Statistically
This trial illustrates several core features of clinical biostatistics. First, a trial can report a numerical difference between group medians while the formal time-to-event comparison remains above a conventional significance threshold. Statistical interpretation needs both the estimated outcomes and uncertainty, rather than a simple focus on which median is larger.
Second, progression-free survival is a censored time-to-event outcome. Participants do not all necessarily experience progression or death during observed follow-up, so analysis must distinguish observed events from incomplete event times. The registry reports censoring in both groups, but not the reasons or detailed rules. This makes the reporting of censoring and analysis populations central to evaluating the result.
Third, the precision of an effect estimate matters. The intervals around the reported medians show that the estimates are uncertain, and a direct between-group hazard ratio with a confidence interval is not reported. A p-value alone cannot convey the magnitude or precision of a treatment effect. Conversely, separate confidence intervals around medians are not the same as an interval for a contrast between groups.
Finally, the study highlights why registry results must be read with attention to analysis plans and completeness. The secondary outcome record is partial, multiplicity handling is not reported, and the primary population description combines language about all randomized participants with analysis according to treatment received. Such details affect what can be concluded. Statistical evidence is strongest when outcome definitions, analysis populations, missing-data procedures, and planned comparisons are clearly documented alongside 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.
Several trial-specific limitations affect interpretation. Enrollment was 70 in a phase 2 trial, and the posted primary group sizes were 39 and 31. This scale can leave estimates imprecise, as reflected in the confidence intervals around the progression-free survival medians. The registry's sample-size justification is not reported in the registry record, so it is not possible to evaluate the original design assumptions from this information.
The primary comparison is reported through separate medians and a log-rank p-value. A hazard ratio and confidence interval for the time-to-event contrast are not reported in the registry record. As a result, the posted primary results do not give a direct relative-effect estimate with its precision. The record reports censoring of 8 and 11 participants but does not state the reasons or detailed censoring rules.
The analysis-population description is difficult to reconcile: it says the randomized and treated population includes all randomized patients, while also stating that participants are analyzed according to treatment actually received and describing the analysis as intent-to-treat. The registry does not clarify this wording. Masking was absent, and further details about outcome assessment procedures are not reported in the registry record.
Reporting is incomplete across outcomes. Tumor response has a posted comparison, but results for duration of response and overall survival are not reported in the registry record. The multiplicity strategy, interim analyses, stopping rules, and detailed statistical plan are also not reported. The comparator regimen description ends before all its instructions are stated. These gaps do not determine the direction of any treatment effect, but they restrict assessment of the trial methods and the breadth of conclusions that can be drawn.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Chemotherapy for Patients With Non-Small Cell Lung Cancer Who Are Non-Smokers — NCT00409006.
- Registry record last updated 2010-09-09; status last verified 2010-08.
- Results first posted 2010-09-09.
- 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
The registry record describes a completed, randomized, open-label, parallel-group phase 2 trial with enrollment of 70. It compared pemetrexed and cisplatin followed by gefitinib with pemetrexed and cisplatin followed by pemetrexed-based treatment in Asian never-smoker patients with advanced non-small cell lung cancer.
The primary endpoint was median progression-free survival, defined from randomization to first progression or death from any cause. The reported medians were 9.95 months and 6.83 months, respectively; the sponsor's log-rank p-value was 0.062. The record also includes a tumor-response comparison and safety counts. Results for duration of response and overall survival, the sample-size justification, multiplicity strategy, and several analysis details are not reported in the registry record.