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 Study of Capmatinib and Spartalizumab Combination Therapy vs Docetaxel in Non-small Cell Lung Cancer (NCT03647488), a na phase 2 study. The primary outcome is Randomized Part: Overall Survival (OS).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03647488 |
| Phase | Phase 2 |
| Condition | Carcinoma, Non-Small-Cell Lung |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 18 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Novartis Pharmaceuticals |
| Results posted | Yes |
2. Clinical Question
Population
Adults with locally advanced or metastatic non-small cell lung cancer that was EGFR wild-type for the specified mutations and ALK rearrangement negative, after failure of platinum doublet and checkpoint inhibitor treatment. The record describes locally advanced disease as stage IIIB when definitive chemo-radiation was not an option, or metastatic stage IV disease.
Intervention
The combination was capmatinib 400 mg twice daily and spartalizumab 400 mg intravenously once every 28 days. It was given in the run-in part and was also planned for the randomized part.
Comparator
Docetaxel 75mg/m2 intravenously once every 21 days, following local guidelines and product labels, was planned as the comparator in the randomized part. That part never started, so the record has no observed comparison with docetaxel.
Primary question
The planned primary endpoint was overall survival in the randomized part: time from treatment start to death from any cause, with survival censored at the last known date alive when death was not known. The registry record reports that no randomized participants were enrolled and no results were available for this endpoint.
The central clinical question was whether the capmatinib and spartalizumab combination could be evaluated against docetaxel in this pretreated population. The record supports describing the study plan and the run-in safety experience, but it cannot answer the comparative question. The combination was administered to participants in the run-in part; the docetaxel arm and randomized combination arm remained planned rather than observed. Consequently, there is no randomized estimate of the difference in overall survival, and no basis in this record for attributing an outcome difference to either treatment.
3. Trial Design
This was a phase II multicenter study described in its title as a randomized, two-arm comparison of capmatinib plus spartalizumab against docetaxel in pretreated adults with advanced or metastatic non-small cell lung cancer. The record lists no allocation method and no masking. Its design fields identify a single-group model, and its summary explains why the title's randomized comparison did not yield comparative results: the randomized part never started.
The study included a run-in part in which participants received capmatinib 400 mg twice daily and spartalizumab 400 mg intravenously once every 28 days. The record lists 18 participants enrolled overall, and the run-in safety group is reported with 18 at risk. In the planned randomized part, the combination regimen was to be compared with docetaxel 75mg/m2 intravenously once every 21 days. The record states that no participants were enrolled in that part.
The study is marked completed. Its listed start date is 2018-12-26, and its primary completion date is 2020-09-07. The planned primary endpoint concerned overall survival, assessed from treatment start until death from any cause or censoring at the last known date alive. Other listed outcomes included response and disease-control measures, progression-free survival, time to response, duration of response, and capmatinib pharmacokinetic measures.
Because only the run-in part generated participant safety data, this record is not a like-for-like treatment comparison. The design description should therefore be read as a distinction between the intended randomized study and the part that actually enrolled participants, rather than as evidence of a completed randomized trial.
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: Based on preliminary results of run-in part, the decision was not to open randomized part of the study. Therefore, no participants were enrolled in randomized part.
Overall Study
| Stage | Run-in Part: Capmatinib + Spartalizumab | Randomized Part: Capmatinib + Spartalizumab | Randomized Part: Docetaxel |
|---|---|---|---|
| STARTED | 18 | 0 | 0 |
| COMPLETED | 0 | 0 | 0 |
| NOT COMPLETED | 18 | 0 | 0 |
Reasons for not completing this period:
| Reason | Run-in Part: Capmatinib + Spartalizumab | Randomized Part: Capmatinib + Spartalizumab | Randomized Part: Docetaxel |
|---|---|---|---|
| Progressive disease | 10 | 0 | 0 |
| Adverse Event | 5 | 0 | 0 |
| Clinical progression | 2 | 0 | 0 |
| Participant refused to take investigational product | 1 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: All participants from the Safety Set in the run-in who experienced DLT during the first 8 weeks of dosing or met the minimum exposure criterion (subject received at least 1 infusion of spartalizumab and took at least 50% of the planned dose of capmatinib within the first 8 weeks of treatment) and had sufficient safety evaluations (subjects were observed for ≥56 days following the first dose, and are considered to have enough safety data to conclude that a DLT did not occur).
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: Based on preliminary results of run-in part, randomized part was not opened. Therefore, no participants were enrolled in the randomized part of the study.
| Characteristic | Unit | Run-in Part: Capmatinib + Spartalizumab | Randomized Part: Capmatinib + Spartalizumab | Randomized Part: Docetaxel | Total |
|---|---|---|---|---|---|
| Participants analysed | 18 | 0 | 0 | 18 | |
| Age, Continuous | Years | 61.2 (SD 10.52) | — | — | 61.2 (SD 10.52) |
| Sex: Female, Male: Female | Participants | 7 | 0 | 0 | 7 |
| Sex: Female, Male: Male | Participants | 11 | 0 | 0 | 11 |
| Race/Ethnicity, Customized: White | Participants | 17 | 0 | 0 | 17 |
| Race/Ethnicity, Customized: Missing | Participants | 1 | 0 | 0 | 1 |
6. Primary Endpoint
The registry lists the primary outcome as Randomized Part: Overall Survival (OS), measured over From start of treatment to death due to any cause, assessed until the end of the study (up to a planned duration of 18 months).
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
Overall survival is the time from the start of treatment to death from any cause. The registry record says that participants not known to have died would be censored at the last date they were known to be alive. In general, overall survival is used because it captures a clinically direct time-to-event outcome and does not require deciding whether a death was caused by the cancer or treatment.
For a randomized comparison, the endpoint could describe survival experience in each assigned group and support an estimate of the contrast between treatments. Its interpretation depends on the study having participants and outcome follow-up in those groups, as well as an analysis that accounts for censoring. Censoring is not itself a death or proof of continued survival indefinitely; it marks the end of observed survival information for that participant.
In this study, overall survival was designated as the primary endpoint for the randomized part, with assessment planned through the end of the study and a planned duration of 18 months. The record says that the randomized part never started and that no participants were enrolled in it. Results for the primary endpoint are therefore unavailable in the registry record. The endpoint definition and planned follow-up do not constitute an observed treatment effect.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The main interpretive point is that the record describes a planned randomized comparison but reports no randomized-part results. The primary endpoint was overall survival, yet the registry record states that no participants entered the randomized part. There is consequently no posted overall-survival estimate for the planned comparison and no treatment contrast to interpret.
The enrollment figure of 18 and the run-in safety group describe the part in which the combination was administered. They do not represent enrollment into the randomized comparison. The listed docetaxel regimen is a planned comparator regimen, not an observed group with outcomes in this record. A comparison of the run-in combination experience with docetaxel would not be a randomized or like-for-like comparison and should not be inferred from these data.
The record includes safety information for the run-in group: 10 participants were reported as affected by serious adverse events, 18 were reported as affected by other adverse events, and 5 deaths were listed. The registry record does not establish from those figures that deaths were caused by treatment, nor do aggregate counts indicate the timing, duration, or clinical course of events. Serious adverse-event terms and percentages provide additional descriptive information, but the listed terms may not represent mutually exclusive participants or events.
In general, an estimate describes the observed data, while its uncertainty reflects how much the data can distinguish among plausible values. A confidence interval, when reported, expresses uncertainty around an estimate under the analysis assumptions. No primary estimate or confidence interval for randomized overall survival is available in this record, so comparative precision cannot be assessed.
The study lists secondary outcomes including objective response rate and disease control rate by RECIST 1.1 and investigator assessment, progression-free survival, time to response, duration of response, and capmatinib exposure measures. The record identifies 22 outcome measures, but it reports no analyses. Naming an outcome is not the same as reporting a result, and the listed endpoint definitions cannot establish the direction or magnitude of any treatment effect.
The sample-size planning assumptions described in the record are illustrative and based on an observed effect. They are not evidence that the trial attained a particular power, and they do not supply missing randomized results. Without an observed randomized effect, the appropriate conclusion is limited: the study generated run-in safety information, while its intended randomized efficacy comparison is unanswered by this registry record.
9. Statistical Methods Explained
The registry record describes a phase II study with a run-in part and a planned randomized part. Those parts answer different statistical questions. The run-in can describe experience among participants who received the combination; a randomized comparison would require observed outcomes in the assigned treatment groups. Since the record says the randomized part never started, the planned comparison cannot be analyzed from the posted results.
What does randomization contribute?
In general, random assignment is intended to make treatment groups comparable with respect to both measured and unmeasured factors, on average. This supports interpretation of between-group differences as treatment comparisons, subject to adherence, follow-up, missing data, and the analysis plan. Here, randomization is part of the intended design, but no randomized participants were enrolled. The record therefore contains no randomized groups whose outcomes could support that interpretation.
The record lists allocation as not applicable and identifies a single-group model. It also lists no masking. Masking can reduce some risks of biased assessment or care when participants or investigators do not know assigned treatment, but its relevance depends on the design and outcome. The record does not describe a blinded randomized comparison in operation.
How should overall survival be understood?
Overall survival is a time-to-event outcome. Each participant contributes information from treatment start until death or the last date known alive. When a participant is alive at the end of observed follow-up, or their later status is unavailable, the observation can be censored at the last known date alive, as specified in the registry definition. In general, survival methods account for such incomplete event-time information rather than treating every censored participant as if an event occurred at the censoring time.
Time-to-event estimates depend on the observed event and follow-up information and on assumptions about censoring. The record does not report randomized-participant survival observations, an estimate, a confidence interval, or an analysis. Thus no survival curve, hazard contrast, or measure of comparative uncertainty can be interpreted from this endpoint in this trial.
What can the planning assumptions tell readers?
The record describes a sample-size illustration using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the planning tables use a total event count based on the Schoenfeld approach; for other outcomes, they express participants per arm. These are general design-planning concepts, not observed trial findings.
The registry record does not report a trial-specific sample-size justification or the full statistical plan. The planning illustration is explicitly not the trial's own design and does not establish that a planned sample size was enrolled, that an effect was detected, or that the study had observed power. A sample-size calculation is conditional on its assumptions; it cannot substitute for an estimate from actual outcome data.
Why does over-dispersion matter in general?
For outcomes such as event counts, a statistical model may assume a particular amount of variability. Over-dispersion means the observed variability exceeds that assumption. If it is ignored, standard errors can be too small and confidence intervals too narrow, making results appear more precise than the data warrant. This is a general issue in count-data analysis; the registry record does not report whether such a model or adjustment was used in this study.
Similarly, analysis choices such as covariate adjustment, the handling of protocol deviations, and the definition of analysis populations can affect estimates and their interpretation. Those details are not reported in the registry record. No recalculated treatment effect is available to compare with a sponsor analysis, and no sponsor analysis of randomized overall survival is reported.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists multiple secondary outcomes, including response, disease control, progression-free survival, time to response, duration of response, and capmatinib pharmacokinetic measures. It also identifies 22 outcome measures. When many outcomes or repeated statistical tests are examined, the chance of at least one apparently notable result can increase. In general, multiplicity procedures may control that risk or prioritize confirmatory endpoints, but the registry record does not report a multiplicity strategy or a statistical analysis plan.
There are no analyses reported in the registry record. The primary overall-survival results are unavailable because no participants entered the randomized part. The record does not say whether any secondary outcome data were collected for the run-in part or provide analysis populations for those outcomes. Outcome listings should not be taken as evidence that every endpoint was measured or analyzed.
Missing-data methods are not reported in the registry record. The primary endpoint definition does specify censoring at the last known date alive for participants not known to have died; that is part of the endpoint definition, not a description of all missing-data procedures. The record does not describe follow-up completeness, reasons for missing observations, or sensitivity analyses. It also does not report intention-to-treat, per-protocol, or safety analysis population definitions. These omissions limit assessment of how outcomes would have been selected and handled in a randomized analysis.
11. 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 |
|---|---|---|---|
| Run-in Part: Capmatinib + Spartalizumab | 10 | 18 | 55.6% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Run-in Part: Capmatinib + Spartalizumab | 5 of 18 |
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 |
|---|---|
| Run-in Part: Capmatinib + Spartalizumab | 18 of 18 |
Most frequent serious adverse event terms
| Event term | Organ system | Run-in Part: Capmatinib + Spartalizumab |
|---|---|---|
| Fatigue | General disorders | 3 of 18 |
| Cardiac failure congestive | Cardiac disorders | 1 of 18 |
| Ventricular arrhythmia | Cardiac disorders | 1 of 18 |
| General physical health deterioration | General disorders | 1 of 18 |
| Pyrexia | General disorders | 1 of 18 |
| Anaphylactic reaction | Immune system disorders | 1 of 18 |
| Drug hypersensitivity | Immune system disorders | 1 of 18 |
| Abdominal infection | Infections and infestations | 1 of 18 |
| Pneumonia | Infections and infestations | 1 of 18 |
| Respiratory tract infection | Infections and infestations | 1 of 18 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Run-in Part: Capmatinib + Spartalizumab |
|---|---|---|
| Nausea | Gastrointestinal disorders | 7 of 18 |
| Diarrhoea | Gastrointestinal disorders | 5 of 18 |
| Vomiting | Gastrointestinal disorders | 5 of 18 |
| Blood creatinine increased | Investigations | 5 of 18 |
| Dyspnoea | Respiratory, thoracic and mediastinal disorders | 5 of 18 |
| Asthenia | General disorders | 4 of 18 |
| Fatigue | General disorders | 4 of 18 |
| Oedema peripheral | General disorders | 4 of 18 |
| Pyrexia | General disorders | 4 of 18 |
| Weight decreased | Investigations | 3 of 18 |
Serious events with the highest arm-level rate: fatigue (16.7%), cardiac failure congestive (5.6%), ventricular arrhythmia (5.6%), general physical health deterioration (5.6%), pyrexia (5.6%).
Safety results apply to the run-in combination group, not to a randomized comparison with docetaxel. The registry record lists 18 participants at risk for serious and other adverse events. It reports 10 participants affected by serious adverse events and 18 affected by other adverse events, with 5 deaths listed. These are descriptive counts; they do not establish that an event or death was caused by either study drug.
The most frequently listed serious adverse-event term was fatigue at 16.7%. Cardiac failure congestive, ventricular arrhythmia, general physical health deterioration, pyrexia, and anaphylactic reaction were each listed at 5.6%. The registry record does not provide event-level timing, duration, severity beyond the serious classification, or causal attribution for these terms in the information. Terms can occur in the same participant, so the percentages should not be added together or read as separate groups of participants.
General guidance: adverse-event counts describe what was recorded in the observed group, while comparison of risk between treatments requires suitable comparison groups and consistent follow-up. Those conditions are not present for a randomized treatment comparison in this record.
12. What the Result Does, and Does Not, Mean
The record documents a phase II study intended to assess capmatinib plus spartalizumab in pretreated adults with advanced or metastatic non-small cell lung cancer. It records a run-in group that received the combination and provides descriptive safety figures for that group. It also documents an intended randomized comparison against docetaxel, with overall survival as the primary endpoint.
What the record does not establish is the relative effect of the combination and docetaxel. The randomized part never started, no participants were enrolled in it, and no randomized primary-endpoint results are available. The trial therefore does not provide a comparative estimate of overall survival or the precision of such an estimate. It also does not support treating the run-in experience as a control comparison.
The safety counts and serious-event terms describe recorded events in the run-in group. They do not by themselves show causation, comparative safety, or the balance of benefits and harms. The listed deaths are not identified in this record as treatment-related. A count of events is not equivalent to a causal assessment.
In general, the absence of a randomized result is not proof that the treatments have equal outcomes, nor does it show that one is better. It means the planned comparison is unanswered by these registry results. Likewise, listed secondary outcomes and planning assumptions do not supply results that are not reported.
The appropriate statistical reading is therefore narrow: the record offers study-design information and descriptive run-in safety data, while the planned randomized efficacy question remains unresolved. No clinical recommendation follows from this record alone.
13. Why This Trial Matters Statistically
This trial is statistically notable because its title and intended design describe a randomized comparison, while the registry record reports that the randomized part never began. That distinction matters: a planned comparison is not equivalent to an observed comparison, and the existence of a comparator regimen in the protocol does not create a comparator dataset.
The record illustrates why readers should identify which study part generated each result. The run-in group received capmatinib plus spartalizumab and contributes the reported safety information. Docetaxel was planned for the randomized part, but there are no enrolled randomized participants in the registry record. Combining those descriptions as though both were observed arms would misstate the evidence.
It also illustrates the difference between endpoint specification and endpoint evidence. Overall survival is defined and designated as primary, but no randomized results are available. A carefully specified outcome cannot answer a treatment question without observed outcome data and an appropriate analysis.
Finally, the record separates planning assumptions from trial results. Sample-size illustrations can explain how a future or hypothetical study might be designed under stated assumptions, but they do not recover a treatment effect or demonstrate observed power. For statistical interpretation, the most important finding is the limit imposed by the study's actual enrollment: descriptive run-in information is present, but the intended randomized estimate is not.
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 principal trial-specific limitation is that the randomized part never started. Although the study was described as a two-arm randomized comparison, the registry record reports no participants enrolled in that part. There is therefore no randomized comparison of capmatinib plus spartalizumab with docetaxel and no randomized result for the primary overall-survival endpoint.
The reported safety experience comes from the run-in combination group. Without an observed comparator group in the randomized part, these descriptive safety figures cannot establish relative safety. The registry record also does not provide causal attribution for the listed deaths or serious adverse-event terms in the information reported.
The record does not report the statistical analysis plan, a trial-specific sample-size justification, multiplicity procedures, analysis-population definitions, or missing-data methods beyond the stated censoring rule for overall survival. It reports no analyses. These gaps prevent assessment of planned versus actual analysis choices and leave uncertainty about the handling of outcomes beyond the endpoint definition.
The listed secondary outcomes and outcome-measure count do not establish which outcomes were collected or analyzed in the run-in group. No comparative interpretation should be drawn from the planned docetaxel regimen and the observed run-in experience. These limitations apply specifically to what this registry record can establish; they do not determine results that may or may not exist outside the record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Study of Capmatinib and Spartalizumab Combination Therapy vs Docetaxel in Non-small Cell Lung Cancer — NCT03647488.
- Registry record last updated 2022-01-24; status last verified 2022-01.
- Results first posted 2021-10-25.
- 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.
18. Record Summary
NCT03647488 is a completed phase II study sponsored by Novartis Pharmaceuticals. It was designed to evaluate capmatinib plus spartalizumab in pretreated adults with advanced or metastatic non-small cell lung cancer and planned a randomized comparison with docetaxel. The record lists 18 participants enrolled overall and describes a run-in combination group.
The randomized part never started, and no participants were enrolled in it. Overall survival was the planned primary endpoint for that part, but results are unavailable. The registry record reports run-in safety counts and serious adverse-event terms, but no analyses. The posted evidence therefore describes the study plan and run-in safety experience, not a randomized comparison of the treatments.