This page separates reported trial results from statistical interpretation. Trial-specific numerical results and design facts are restricted to the ClinicalTrials.gov record for NCT03504397. The registry provides the official trial record.
1. Trial at a Glance
SPOTLIGHT was a randomized, double-blind, parallel phase 3 trial evaluating mFOLFOX6 plus zolbetuximab versus mFOLFOX6 plus placebo in adults with locally advanced unresectable or metastatic gastric or gastroesophageal junction adenocarcinoma.
| Feature | SPOTLIGHT |
|---|---|
| Trial name | SPOTLIGHT |
| NCT ID | NCT03504397 |
| Phase | Phase 3 |
| Status | Active, not recruiting |
| Design | Randomized, double-blind, parallel |
| Allocation | Randomized |
| Primary purpose | Treatment |
| Enrollment | 565 |
| Primary endpoint | Progression Free Survival (PFS) |
| Primary endpoint type | Time-to-event |
| Primary hypothesis | Superiority |
| Lead sponsor | Astellas Pharma Global Development, Inc. |
| Sponsor type | Industry |
| Start | 2018-06-21 |
| Primary completion | 2022-09-09 |
2. Clinical Question
The primary statistical question was whether mFOLFOX6 plus zolbetuximab produced a different progression-free survival experience from mFOLFOX6 plus placebo in adults with locally advanced unresectable or metastatic gastric or gastroesophageal junction adenocarcinoma.
Population
Adults with locally advanced unresectable gastroesophageal junction adenocarcinoma or cancer, locally advanced unresectable gastric adenocarcinoma or cancer, metastatic gastric adenocarcinoma or cancer, or metastatic gastroesophageal junction adenocarcinoma.
Intervention
Zolbetuximab administered with mFOLFOX6, consisting of oxaliplatin, folinic acid, and fluorouracil.
Comparator
Placebo administered with mFOLFOX6, consisting of oxaliplatin, folinic acid, and fluorouracil.
Primary question
Does mFOLFOX6 plus zolbetuximab improve progression-free survival compared with mFOLFOX6 plus placebo?
3. Trial Design
mFOLFOX6 + Zolbetuximab
- Zolbetuximab
- Oxaliplatin
- Folinic acid
- Fluorouracil
mFOLFOX6 + Placebo
- Placebo
- Oxaliplatin
- Folinic acid
- Fluorouracil
The design is randomized, parallel, and double-blind. Randomization is important because it establishes the treatment groups before outcomes are observed, while blinding is intended to reduce the influence of treatment knowledge on trial conduct and assessment.
4. Endpoints
| Endpoint | Registry definition / time frame | Type | Analysis |
|---|---|---|---|
| Progression Free Survival (PFS) | From date of randomization until the date of first documented radiological progression or date of death from any cause. PFS was defined as time from randomization until radiological progressive disease per RECIST 1.1 by independent review committee or death from any cause, whichever was earliest. | Time-to-event | Log-rank test; HR |
| Overall Survival (OS) | From the date of randomization until 62 months and 18 days. | Time-to-event | Log-rank test; HR |
| Time to Confirmed Deterioration — Physical Functioning | From the date of randomization until 62 months and 18 days, using physical functioning as measured by EORTC QLQ-C30. | Time-to-event | Log-rank test; HR |
| Time to Confirmed Deterioration — Abdominal Pain and Discomfort | From the date of randomization until 62 months and 18 days, using the Oesophago-gastric Questionnaire (OG25) on abdominal pain and discomfort as measured by EORTC QLQ-OG25. | Time-to-event | Log-rank test; HR |
| Time to Confirmed Deterioration — Global Health Status | From the date of randomization until 62 months and 18 days, using Global Health Status as measured by EORTC QLQ-C30. | Time-to-event | Log-rank test; HR |
| Duration Of Response (DOR) | From date of first response (CR/PR) until 62 months and 18 days. | Time-to-event | Log-rank test; HR |
| Objective Response Rate (ORR) | From the date of randomization until 62 months and 18 days. | Binary | Cochran-Mantel-Haenszel test |
The registry lists Progression Free Survival (PFS) as the single primary endpoint. The other posted analyses are secondary endpoints. This distinction matters because a primary endpoint is the outcome around which the principal confirmatory statistical question is organized, whereas secondary endpoints provide additional evidence about survival, response, and patient-reported deterioration.
5. Statistical Methodology
Analysis populations
The primary PFS analysis was performed in the Full Analysis Set (FAS). The same FAS was used for the posted OS, time-to-confirmed-deterioration, and ORR analyses. The DOR analysis used the FAS — All Objective Responders.
| Analysis population | Endpoint use in the posted analyses |
|---|---|
| FAS | PFS, OS, physical-functioning TTCD, abdominal-pain/discomfort TTCD, global-health-status TTCD, and ORR |
| FAS — All Objective Responders | Duration of Response |
Time-to-event framework
Six of the seven posted statistical analyses use a time-to-event framework. These endpoints do not simply ask whether an event occurred; they consider when the event occurred. That distinction is important because two treatment groups can have the same eventual event proportion while having materially different event timing.
Log-rank testing
The registry reports the log-rank test for the primary PFS comparison and each of the posted secondary time-to-event comparisons. The log-rank test evaluates whether the event-time distributions differ between randomized groups over the observed follow-up.
For SPOTLIGHT, the reported hypothesis type is superiority. The effect measure accompanying the log-rank analysis is the hazard ratio.
Hazard ratio
The hazard ratio is the reported effect measure for the six posted time-to-event analyses. A hazard ratio below 1 indicates a lower estimated instantaneous event rate in the zolbetuximab group relative to the placebo group, whereas a value above 1 indicates a higher estimated instantaneous event rate.
This is a relative time-to-event measure. It is not an absolute risk difference, a probability of benefit for an individual patient, or a statement that every patient experiences the same proportional change in event risk.
Cochran-Mantel-Haenszel analysis
Objective response rate was analyzed using the Cochran-Mantel-Haenszel (CMH) test. The registry specifies a 1-sided CMH test and identifies Region, Number of Metastatic Sites, and Prior Gastrectomy as the stratification factors.
The CMH approach is useful when a binary outcome is compared across treatment groups while accounting for prespecified strata. Conceptually, it avoids treating the trial as though all participants belonged to one completely homogeneous population when the analysis has explicitly defined clinically relevant strata.
Stratified time-to-event analysis
The registry specifically documents stratification for the global-health-status TTCD and DOR analyses. The three stated stratification factors were Region (Asia vs Non-Asia), Number of organs with metastatic sites (0 to 2 vs ≥ 3), and Prior gastrectomy (Yes or No).
| Stratification factor | Levels reported |
|---|---|
| Region | Asia vs Non-Asia |
| Number of organs with metastatic sites | 0 to 2 vs ≥ 3 |
| Prior gastrectomy | Yes or No |
6. Results: Primary Endpoint — Progression-Free Survival
The registry reports a formal primary analysis of PFS in the FAS comparing mFOLFOX6 plus zolbetuximab with mFOLFOX6 plus placebo. The analysis used a two-sided log-rank test and reported the effect as a hazard ratio.
Hazard ratio for progression or death
95% CI: 0.591–0.910 · P = 0.0024
Analysis population: FAS · Hypothesis: Superiority
| Primary endpoint | Zolbetuximab + mFOLFOX6 | Placebo + mFOLFOX6 | Statistical result |
|---|---|---|---|
| Progression Free Survival (PFS) | Compared with placebo group | Reference treatment group | HR 0.734; 95% CI 0.591–0.910; P = 0.0024 |
An HR of 0.734 means that the estimated instantaneous rate of progression or death was approximately 73.4% of the corresponding rate in the mFOLFOX6-plus-placebo group under the time-to-event analysis. Equivalently, 1 − 0.734 = 0.266, so the estimate corresponds to a 26.6% lower estimated hazard of progression or death.
The HR does not mean that 26.6% of patients avoided progression, that every patient experienced a 26.6% reduction in risk, or that the probability of remaining progression-free at a particular time was exactly 26.6% higher. Hazard ratios describe relative event rates over follow-up rather than a single absolute probability.
The 95% confidence interval of 0.591–0.910 describes uncertainty around the estimated hazard ratio. It indicates that the observed estimate is not being presented as an exact treatment effect. The interval also remains below 1, which is consistent with the reported superiority hypothesis test.
The P = 0.0024 value addresses the statistical evidence against the null hypothesis under the reported testing framework. It does not measure the size of the treatment effect, does not say that there is a 0.24% probability that the null hypothesis is true, and does not describe the probability that an individual patient will benefit.
Because PFS is a time-to-event endpoint, interpretation also depends on censoring and the assumptions underlying the hazard-ratio representation. The registry reports the log-rank method and HR but does not provide sufficient information here to independently assess proportional hazards, the censoring distribution, or the detailed analysis plan.
7. Secondary Endpoint Results
Overall Survival
Hazard ratio for overall survival
95% CI: 0.644–0.954 · P = 0.0075
Analysis population: FAS · Time frame: from randomization until 62 months and 18 days
The OS analysis used a two-sided log-rank test with a hazard ratio as the effect measure. An HR of 0.784 corresponds to an estimated instantaneous rate of death approximately 78.4% of that in the placebo comparison group, or approximately a 21.6% lower estimated hazard based on 1 − 0.784.
The 95% CI of 0.644–0.954 quantifies uncertainty around the estimated HR. It does not provide a range of individual survival times and should not be interpreted as saying that individual patients have between 64.4% and 95.4% of the risk of death.
The reported P = 0.0075 is evidence from the stated two-sided log-rank comparison. It is not a measure of the magnitude of the survival difference. The magnitude is described by the HR, while absolute survival probabilities or median survival times would provide different clinical perspectives; those quantities are not reported in the ClinicalTrials.gov record.
Time to Confirmed Deterioration — Physical Functioning
Hazard ratio for confirmed deterioration in physical functioning
95% CI: 0.994–1.687 · P = 0.0280
Analysis population: FAS · EORTC QLQ-C30 physical functioning
For this endpoint, the HR is above 1. Under the reported hazard-ratio framework, that corresponds to a higher estimated instantaneous rate of confirmed deterioration in physical functioning in the zolbetuximab group relative to the placebo group. An HR of 1.295 corresponds to approximately a 29.5% higher estimated hazard of the event.
The 95% CI of 0.994–1.687 is relatively wide and lies very close to 1 at its lower boundary. The reported P-value is 0.0280. These quantities should be read together rather than treating the P-value as a measure of effect size.
Time to Confirmed Deterioration — Abdominal Pain and Discomfort
Hazard ratio for confirmed deterioration in abdominal pain and discomfort
95% CI: 0.475–1.071 · P = 0.0508
Analysis population: FAS · EORTC QLQ-OG25
The HR of 0.713 corresponds to an estimated instantaneous event rate approximately 71.3% of that in the placebo group, or approximately a 28.7% lower estimated hazard of confirmed deterioration under this endpoint definition.
However, the 95% CI of 0.475–1.071 crosses 1. The reported P = 0.0508 also illustrates why a P-value should not be read in isolation. The effect estimate, its confidence interval, and the endpoint's place among multiple secondary outcomes all contribute to the appropriate interpretation.
Time to Confirmed Deterioration — Global Health Status
Hazard ratio for confirmed deterioration in global health status
95% CI: 0.874–1.492 · P = 0.1685
Analysis population: FAS · Stratified log-rank analysis
The HR of 1.142 corresponds to an estimated instantaneous event rate approximately 14.2% higher in the zolbetuximab group under the reported analysis. The confidence interval of 0.874–1.492 includes 1, and the reported P-value is 0.1685.
This analysis was explicitly stratified by Region, Number of organs with metastatic sites, and Prior gastrectomy. That stratification means the comparison was not treated as a completely unadjusted pooled comparison across all participants.
Duration Of Response
Hazard ratio for duration of response
95% CI: 0.573–1.087 · P = 0.0721
Analysis population: FAS — All Objective Responders
The DOR analysis begins at first response rather than at randomization. An HR of 0.789 corresponds to an estimated instantaneous event rate approximately 78.9% of that in the comparator group among objective responders, or approximately a 21.1% lower estimated hazard of the DOR event.
The 95% CI of 0.573–1.087 crosses 1, and the reported P-value is 0.0721. An important statistical distinction is that DOR is analyzed only among participants who achieved an objective response. It therefore addresses durability conditional on response and does not replace the randomized primary PFS analysis.
Objective Response Rate
Cochran-Mantel-Haenszel comparison
1-sided CMH test · Analysis population: FAS
Stratification factors: Region, Number of Metastatic Sites, and Prior Gastrectomy
Unlike the other posted results, ORR is a binary endpoint rather than a time-to-event endpoint. The registry reports a 1-sided CMH test but does not provide an ORR estimate for either randomized group in the ClinicalTrials.gov record. Accordingly, no treatment-group response percentages are inferred here.
The absence of an effect estimate is important: P = 0.4536 is the reported result of the specified CMH test, but a P-value alone does not communicate the magnitude or direction of a difference in response rates.
8. Safety Results
The ClinicalTrials.gov record reports serious adverse events by arm as affected participants divided by participants at risk. These counts are presented exactly as reported.
| Intervention | Serious adverse events | Interpretation of reported count |
|---|---|---|
| Zolbetuximab | 133 / 279 | 133 affected among 279 at risk |
| Placebo | 129 / 278 | 129 affected among 278 at risk |
| 5-fluorouracil | 262 / 557 | 262 affected among 557 at risk |
| Folinic Acid / Leucovorin | 262 / 557 | 262 affected among 557 at risk |
| Oxaliplatin | 262 / 557 | 262 affected among 557 at risk |
The chemotherapy-component entries have the same affected and at-risk counts in the ClinicalTrials.gov record. They should not be interpreted as independent randomized treatment-group comparisons because oxaliplatin, folinic acid/leucovorin, and 5-fluorouracil are components of the chemotherapy regimen.
9. Statistical Methods Explained
Why was a log-rank test used for PFS?
PFS is a time-to-event endpoint, and some participants may not have experienced progression or death by the time of analysis. A log-rank test is designed to compare the event-time distributions between groups while incorporating the timing of events and censoring. It therefore uses more information than a simple comparison of the proportion of participants who eventually experienced progression.
What does an HR of 0.734 mean?
The primary PFS HR of 0.734 indicates an estimated instantaneous progression-or-death rate about 73.4% of that in the comparator group. The complementary quantity, 1 − 0.734, is 0.266, so the estimate can be described as a 26.6% lower estimated hazard. This does not mean that 26.6% more patients remained progression-free at every time point.
Why does the confidence interval matter?
The point estimate is only one summary of the observed treatment comparison. The 95% CI of 0.591–0.910 communicates uncertainty around the primary HR estimate. A narrow interval would indicate greater precision than a wide interval, all else being equal. The interval does not describe the range of treatment effects that individual patients experience.
Why does the P-value not measure effect size?
The P-value reflects how compatible the observed data are with a specified null hypothesis under the statistical test. It is affected by both the size of an observed difference and the amount of information available. The HR and its confidence interval are therefore needed to understand the estimated magnitude and precision of the PFS comparison.
Why does an HR above 1 matter for a deterioration endpoint?
For the physical-functioning TTCD endpoint, the HR was 1.295. Because the event being analyzed is confirmed deterioration, an HR above 1 corresponds to a higher estimated event rate in the zolbetuximab group. The meaning of the direction depends on what the event represents; an HR below 1 is not automatically beneficial for every possible endpoint.
Why was a CMH test used for ORR?
ORR is binary: a participant either meets the response definition or does not. The CMH test provides a way to compare treatment groups while accounting for specified strata. In SPOTLIGHT, the registry identifies Region, Number of Metastatic Sites, and Prior Gastrectomy as the stratification factors for the 1-sided CMH analysis.
Why is DOR analyzed in responders?
DOR begins at the first documented response, so participants who never respond do not enter the responder-based time origin. This makes DOR a measure of the persistence of response rather than a replacement for an intention-to-treat endpoint such as PFS. Conditioning on response also means DOR should be interpreted differently from a randomized comparison beginning at treatment assignment.
10. Stratified Analysis
Several posted analyses explicitly use stratification. The registry specifies the same three stratification factors for the global-health-status TTCD and DOR analyses, and the ORR analysis uses those factors in the CMH framework.
| Factor | Strata | Role reported in the registry |
|---|---|---|
| Region | Asia vs Non-Asia | Stratification factor |
| Number of organs with metastatic sites | 0 to 2 vs ≥ 3 | Stratification factor |
| Prior gastrectomy | Yes or No | Stratification factor |
Stratification can improve the alignment between the analysis and the design when important baseline factors are expected to influence the endpoint. It does not mean that the treatment effect is separately estimated with equal precision inside every stratum. The overall stratified result remains the principal summary for the specified analysis.
11. P-Values, Confidence Intervals, and the Superiority Framework
The posted analyses identify superiority as the hypothesis type. The primary PFS result therefore asks whether the randomized treatment comparison provides evidence of a difference in the direction represented by the superiority hypothesis.
| Endpoint | Effect measure | 95% CI | P-value | Hypothesis |
|---|---|---|---|---|
| PFS | HR 0.734 | 0.591–0.910 | 0.0024 | Superiority |
| OS | HR 0.784 | 0.644–0.954 | 0.0075 | Superiority |
| Physical functioning TTCD | HR 1.295 | 0.994–1.687 | 0.0280 | Superiority |
| Abdominal pain/discomfort TTCD | HR 0.713 | 0.475–1.071 | 0.0508 | Superiority |
| Global health status TTCD | HR 1.142 | 0.874–1.492 | 0.1685 | Superiority |
| DOR | HR 0.789 | 0.573–1.087 | 0.0721 | Superiority |
| ORR | CMH test | Not reported | 0.4536 | Superiority |
These seven results should not be read as though they were all interchangeable primary tests. Only PFS is identified as the primary endpoint in the ClinicalTrials.gov record. The other analyses are secondary. The ClinicalTrials.gov record does not provide a multiplicity-adjustment procedure or endpoint hierarchy beyond identifying the primary endpoint, so no additional multiplicity conclusion is imposed here.
12. Reading the Primary Hazard Ratio Carefully
The primary PFS HR of 0.734 indicates a lower estimated instantaneous rate of progression or death in the zolbetuximab-plus-mFOLFOX6 group under the reported analysis. The corresponding 26.6% figure is a relative hazard interpretation, not an absolute percentage-point difference in PFS.
The 95% CI of 0.591–0.910 shows the uncertainty associated with the HR estimate. The confidence interval is particularly useful because it gives more information than the point estimate alone about how precisely the treatment effect was estimated.
The reported two-sided P = 0.0024 indicates evidence against the null hypothesis under the specified log-rank testing framework. It should not be translated into a probability that the treatment works or into the magnitude of benefit.
The ClinicalTrials.gov record does not provide median PFS, median OS, fixed-time survival probabilities, numbers of progression events, Kaplan-Meier coordinates, or a treatment-specific absolute PFS difference. Those quantities are therefore not calculated from the HR and confidence interval.
13. Interpreting the Secondary Endpoints as a Set
The secondary analyses cover several distinct clinical questions. OS addresses death from any cause. TTCD endpoints examine time until confirmed deterioration in specified patient-reported measures. DOR examines the duration of response among objective responders. ORR asks whether a response occurred. These are related but not interchangeable outcomes.
| Question | Endpoint | Statistical perspective |
|---|---|---|
| Does treatment affect time until death? | OS | Time-to-event; HR 0.784 |
| Does treatment affect time until physical-functioning deterioration? | Physical-functioning TTCD | Time-to-event; HR 1.295 |
| Does treatment affect time until abdominal pain/discomfort deterioration? | Abdominal pain/discomfort TTCD | Time-to-event; HR 0.713 |
| Does treatment affect time until global-health-status deterioration? | Global-health-status TTCD | Time-to-event; HR 1.142 |
| Among responders, how long does response persist? | DOR | Time-to-event; HR 0.789 |
| Does treatment affect the proportion achieving objective response? | ORR | Binary; 1-sided CMH P = 0.4536 |
This endpoint diversity is statistically useful because it prevents the entire evidence base from being reduced to a single number. At the same time, it means each result needs to be interpreted according to its own event definition and analysis population.
14. Limitations
- Registry-level detail: the ClinicalTrials.gov record provides the principal statistical results but not the full statistical analysis plan or all protocol-level analysis specifications.
- Limited absolute outcome reporting: the statistical analyses posted on ClinicalTrials.gov report HRs and confidence intervals for the time-to-event endpoints but do not provide median survival, fixed-time survival probabilities, or event counts.
- ORR effect estimate unavailable: the registry reports the CMH P-value but does not provide the treatment-group ORR percentages or a risk ratio/risk difference in the ClinicalTrials.gov record.
- Secondary-endpoint multiplicity: the ClinicalTrials.gov record identifies one primary endpoint and several secondary analyses but do not provide a multiplicity-adjustment strategy for the secondary endpoint family.
- Hazard-ratio assumptions: a single HR summarizes relative event rates over follow-up. Without the underlying event-time data or diagnostic information, proportional-hazards behavior cannot be independently evaluated from the ClinicalTrials.gov record.
- Censoring: time-to-event analyses depend on how censoring is defined and handled. The ClinicalTrials.gov record does not provide the complete censoring rules or individual follow-up records.
- Analysis populations: most analyses use the FAS, while DOR uses the FAS among all objective responders. These populations answer different statistical questions.
- Safety denominators: serious adverse-event data are reported by intervention component, and the chemotherapy components share the same affected/at-risk counts in the ClinicalTrials.gov record. They should not be treated as separate randomized comparisons.
- Incomplete treatment details: the ClinicalTrials.gov record identifies the intervention drugs but do not provide dosing, schedule, treatment duration, or treatment-modification rules.
- No unsupported reconstruction: median PFS, median OS, subgroup estimates, Kaplan-Meier curves, and other quantities not present in the ClinicalTrials.gov record is intentionally not reconstructed from the reported HRs.
15. Why This Trial Matters Statistically
SPOTLIGHT is a useful teaching case because it illustrates how a randomized phase 3 trial can combine a primary time-to-event endpoint with several secondary time-to-event and binary endpoints. The statistical story is not contained in the P-value alone; it requires attention to endpoint definition, analysis population, effect measure, confidence interval, stratification, and the distinction between primary and secondary analyses.
| Concept | How it appears in SPOTLIGHT |
|---|---|
| Randomization | Randomized parallel-group phase 3 design with 565 enrolled participants |
| Blinding | Double-blind trial design |
| Time-to-event analysis | PFS, OS, three TTCD endpoints, and DOR |
| Hazard ratio | Effect measure for the posted time-to-event analyses |
| Log-rank test | Reported method for PFS and the secondary time-to-event analyses |
| Confidence interval | 95% two-sided intervals accompany the reported HR estimates |
| Stratified analysis | Region, metastatic-site count, and prior gastrectomy are identified as stratification factors in specified analyses |
| CMH test | 1-sided Cochran-Mantel-Haenszel analysis for ORR |
| Binary endpoint | ORR contrasts with the time-to-event endpoints |
| Analysis population | FAS for most analyses; FAS among objective responders for DOR |
| Superiority testing | All seven posted statistical analyses identify superiority as the hypothesis type |
16. A Practical Workflow for Analyzing SPOTLIGHT-Type Trials
This workflow highlights why statistical analysis begins before the P-value is read. First identify what constitutes the event and when follow-up starts. Then identify the analysis population and statistical test. Next interpret the effect estimate and its uncertainty. Finally, place the result within the endpoint hierarchy and trial design.
17. Time-to-Event Endpoints: What Is Being Compared?
For PFS, the registry defines the endpoint from randomization until the earliest of radiological progression or death from any cause. This creates a composite event definition: either component can end a participant's progression-free follow-up.
The registry definition specifies radiological progressive disease according to RECIST 1.1 by independent review committee or death from any cause, whichever occurs first.
This matters because a participant who dies before documented radiological progression still experiences a PFS event. PFS therefore cannot be interpreted as simply "time until a scan shows progression."
18. What the Registry Results Do and Do Not Establish
What is directly reported
The primary PFS HR is 0.734 with a 95% CI of 0.591–0.910 and P = 0.0024. The registry also reports six secondary statistical analyses, their methods, populations, and available effect estimates or P-values.
What requires caution
The ClinicalTrials.gov record does not provide every quantity needed to reconstruct the full survival experience, including median event times, event counts, survival probabilities at selected time points, or individual participant follow-up.
What should not be inferred
An HR should not be converted into an absolute probability of benefit. A P-value should not be interpreted as a probability that the treatment hypothesis is true.
Why endpoint context matters
PFS, OS, TTCD, DOR, and ORR have different time origins, event definitions, and analysis populations. Their effect estimates therefore cannot be treated as interchangeable measures of the same outcome.
19. Related Tutorials
Learn more about the methods used in this trial:
20. Related Calculators
21. Sources
- ClinicalTrials.gov: NCT03504397 — SPOTLIGHT.
- PubMed: PMID 42095986.
- PubMed: PMID 40680855.
- PubMed: PMID 38861294.
- PubMed: PMID 37068504.
Continue with the statistical methods behind SPOTLIGHT
Explore survival-analysis tutorials, statistical calculators, and additional Clinical Biostats clinical-trial analyses.
22. Record Summary
SPOTLIGHT provides a detailed example of randomized clinical-trial analysis centered on a time-to-event primary endpoint. The primary PFS analysis used a two-sided log-rank test in the FAS and reported an HR of 0.734 with a 95% CI of 0.591–0.910 and P = 0.0024. Secondary analyses extended the statistical framework to OS, confirmed deterioration in several patient-reported measures, DOR, and ORR using log-rank or CMH methods as specified in the registry.
The most important statistical lesson is that these results should be interpreted as a collection of endpoint-specific analyses rather than as one universal treatment-effect number. The primary PFS result has a defined time origin, event definition, analysis population, effect measure, confidence interval, and hypothesis test. The secondary results answer different questions and use different statistical populations or methods.