CAMPFIRE: A Study of Ramucirumab (LY3009806) in Children and Young Adults With Desmoplastic Small Round Cell Tumor: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for CAMPFIRE: A Study of Ramucirumab (LY3009806) in Children and Young Adults With Desmoplastic Small Round Cell Tumor (NCT04145349), a randomized phase 1, phase 2 study. The primary outcome is Progression Free Survival (PFS).

2. Clinical Question

3. Trial Design

CAMPFIRE was a randomized, open-label Phase 1/2 study of ramucirumab in pediatric patients and young adults with relapsed, recurrent, or refractory desmoplastic small round cell tumor. The registry lists an enrollment of 30 and a parallel-group design. Masking was none. The record status is completed. The study is described as part of the CAMPFIRE master protocol, a platform intended to accelerate development of treatments for pediatric and young adult participants with cancer.

The two groups received cyclophosphamide and vinorelbine, with ramucirumab added in one group. The registry lists 20 participants in the ramucirumab combination group and 10 in the cyclophosphamide and vinorelbine group. Treatment was described in 28-day cycles and continued until disease progression or a criterion for discontinuation was met. The listed regimen for the comparator specifies oral cyclophosphamide daily from days 1 to 28 and intravenous vinorelbine on days 1, 8, and 15. The combination-arm description gives the ramucirumab dose and schedule but is incomplete in the registry record, so its remaining administration details cannot be fully characterized from that record.

Random assignment is intended to make treatment groups comparable on average, while an open-label design means treatment assignment was not masked. The primary endpoint relied on investigator-determined progression, so knowledge of assignment could be relevant to how progression was assessed. Randomization does not eliminate uncertainty, particularly with the group sizes recorded for this study. The registry does not report the randomization procedure, allocation concealment, or further details of the Phase 1 and Phase 2 components. The original sample-size rationale and the trial’s prespecified statistical plan are 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least one dose of study drug (including the censored participants). Number of participants censored in Ramucirumab + Cyclophosphamide + Vinorelbine = 4, Cyclophosphamide + Vinorelbine = 2.

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: All randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival (PFS), measured over Randomization to Objective Progression or Death Due to Any Cause (Up To 23 Months).

The primary endpoint was progression-free survival, measured in months from randomization to investigator-determined objective progression or death from any cause in the absence of documented progression. Progression was defined using RECIST version 1.1. The posted definition includes a specified increase in the sum of target-lesion diameters, unequivocal progression of non-target lesions, or one or more new lesions.

Time-to-event endpoints such as progression-free survival retain information about both whether an event occurred and when it occurred. They can also account for participants who have not had a recorded event by the end of their available follow-up. In this study, participants known to be alive and without progression were censored at their last adequate tumor assessment or at randomization, whichever was later. Censoring means that the exact time to progression or death is not observed for that participant during the available record; it does not mean that an event occurred at the censoring date.

The registry describes a posterior median and 98% credible interval for the endpoint, but does not provide the arm-specific median estimates in the posted results. The reported comparison is a posterior mean hazard ratio from a Bayesian hierarchical model. These are related summaries of time-to-event experience but are not interchangeable: a median describes a point in the estimated event-time distribution, while a hazard ratio compares event rates over follow-up under the model.

7. Results Posted in the Registry

• Ramucirumab + Cyclophosphamide + Vinorelbine vs Cyclophosphamide + Vinorelbine: hazard ratio 0.69 (95% CI 0.25–1.69) (as reported in the registry).

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Overall Response Rate (ORR): Percentage of Participants Who Achieved a Complete Response (CR) or Partial Response (PR)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Cyclophosphamide + Vinorelbine: 10.0% (1/10) versus 10.0% (2/20) on control; risk difference 0.0 percentage points (95% CI -22.8 to 22.8), risk ratio 1.00 (0.10–9.75); unadjusted two-proportion z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cyclophosphamide + Vinorelbine and the control. The interval gives the range of effects that remain compatible with the data.

Complete Response (CR) : Percentage of Participants Who Achieved a CR

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Cyclophosphamide + Vinorelbine: 0.0% (0/10) versus 5.0% (1/20) on control; risk difference -5.0 percentage points (95% CI -14.6 to 4.6); unadjusted two-proportion z-test p = 0.472.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cyclophosphamide + Vinorelbine and the control. The interval gives the range of effects that remain compatible with the data.

The posted response outcomes offer a limited view of tumor response alongside the primary time-to-event endpoint. Overall response was reported at 10.0% in each group. Complete response was reported at 5.0% in the ramucirumab combination group and 0.0% in the comparator group. The confidence intervals for these outcomes are broad, consistent with considerable uncertainty in estimates from the recorded group sizes. The registry also lists duration of response, pharmacokinetic measures, and treatment-emergent anti-drug antibodies, but detailed numerical results for those outcomes are not included in the posted results described by the record.

These secondary outcomes are related to, but not interchangeable with, progression-free survival. Response rate measures whether a participant achieved a specified response; progression-free survival incorporates time to progression or death. A response result therefore does not directly answer the primary time-to-event question. Nor does a similar response percentage determine the hazard ratio over follow-up. The primary and secondary outcomes can contribute different information, but their differences in definition and analysis should remain clear.

Unadjusted secondary results generally carry less evidential weight when many outcomes or comparisons are examined. In general, testing multiple hypotheses increases the chance of finding an apparently notable result by chance unless the analysis plan accounts for multiplicity. The registry does not report a multiplicity adjustment or a hierarchy of secondary hypotheses. The response comparisons are unadjusted recalculations and have no reported sponsor analyses in the record; they should be treated as exploratory descriptions rather than as independent confirmatory evidence.

The registry lists several additional secondary outcomes, but their numerical findings and analysis details are not reported in the registry record. Their absence prevents an overall assessment of the full secondary-outcome pattern. The available secondary data do not resolve the uncertainty in the primary estimate, and the primary estimate should not be replaced by selective emphasis on a secondary finding.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Overall Response Rate (ORR): Percentage of Participants Who Achieved a Complete Response (CR) or Partial Response (PR)
• Duration of Response (DoR)
• Complete Response (CR) : Percentage of Participants Who Achieved a CR
• Pharmacokinetics (PK): Maximum Serum Concentration of Ramucirumab (Cmax)
• PK: Minimum Serum Concentration of Ramucirumab (Cmin)
• Number of Participants With Treatment-Emergent Anti-Drug Antibodies (TE-ADA)

9. Statistical Interpretation

10. Statistical Methods Explained

Progression-free survival is a time-to-event outcome. For each participant, the relevant observation is the time from randomization until progression or death, if such an event is recorded during follow-up. A participant without a recorded event contributes follow-up until the last adequate assessment specified by the endpoint definition. This is called censoring. Censored observations are not treated as if progression occurred on the censoring date; instead, they inform the time-to-event analysis up to that point.

The registry’s progression definition follows RECIST version 1.1 and uses investigator-determined assessment. It includes a threshold increase in the sum of target-lesion diameters, unequivocal progression of non-target lesions, or the appearance of one or more new lesions. Death from any cause is also an event when it occurs in the absence of documented progression. Defining the event this way makes clear what counts as progression for the primary analysis, though the registry record does not provide details about assessment schedules or independent review.

The registry reports a Bayesian hierarchical model for the comparison and a posterior mean hazard ratio. A hazard ratio compares modeled event hazards between groups over follow-up. It is not a ratio of median times, and it is not a direct estimate of how many participants had progression. The reported value is 0.69, with a 98% credible interval from 0.25 to 1.69. The interval represents uncertainty about the model-based comparison given the data and the model. A Bayesian credible interval is interpreted in terms of the posterior distribution; it is not the same construction as a frequentist confidence interval.

The endpoint description also says that the posterior median and 98% credible interval were estimated using Bayesian analysis. However, arm-specific median progression-free survival values are not reported in the registry record. The comparative result available in the record is the posterior mean hazard ratio. The registry does not describe the model’s prior distributions, the hierarchy used, or the exact computational and diagnostic procedures. Those details would be needed for a fuller independent appraisal of the Bayesian analysis.

For a general explanation, a hierarchical model can share information across related groups or components through a model structure, rather than estimating every quantity in isolation. Whether and how that sharing affects an estimate depends on the model specification and the data. The registry names a Bayesian hierarchical model but does not report enough detail to determine the extent of borrowing or evaluate its assumptions. The result should therefore be read as the sponsor-reported model estimate, not as a simple calculation from the group summaries.

The comparison is reported as a superiority analysis. A superiority question asks whether the data support a difference in the direction favoring one group under the chosen analysis. The posterior interval includes hazard-ratio values below and above 1, and no p-value is reported. A single point estimate cannot convey that uncertainty. Nor does the label of a superiority analysis by itself establish that a group is superior; interpretation depends on the estimate, its uncertainty, and the analysis plan.

Randomization is a design method intended to reduce systematic differences between groups at baseline. Open-label treatment means assignment was not masked. Because the primary event includes investigator-determined progression, open-label assessment is a possible source of measurement influence in general. The registry does not report whether progression was reviewed independently or whether additional safeguards were used. The posted information therefore supports describing the design, but not making a stronger claim about the effect of open-label assessment.

Sample-size planning figures are illustrations based on the observed effect, rather than a report of the trial’s original design rationale. For the time-to-event planning figures, the quantities are total events, calculated using a Schoenfeld approach. They are not participant counts per arm. The illustration specifies a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect set to a stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record. These planning calculations do not describe the actual analysis of the trial and must not be confused with observed power.

More generally, power is a design-stage probability of detecting a specified effect under assumptions about the outcome, analysis, and error rate. It is not a measure of how likely the observed result is to be true. A post hoc power calculation based on the observed effect does not add information beyond the estimate and its uncertainty. The planning figures are most useful as illustrations of how demanding a future study could be under the stated assumptions, not as evidence that the present study had a particular probability of success.

The registry includes other outcomes, including overall response, complete response, duration of response, pharmacokinetic measures, and treatment-emergent anti-drug antibodies. The posted detailed results include estimates for overall response and complete response, but the registry record does not provide detailed numerical results for every listed secondary outcome. Analysis methods and missing-data procedures for those secondary outcomes are not reported in the registry record. Their results should not be assumed to have been analyzed with the same model as progression-free survival.

11. Confidence Intervals and Precision

The primary comparison is accompanied by a 98% credible interval from 0.25 to 1.69 around the posterior mean hazard ratio of 0.69. The interval is wide and includes values below and above 1. This indicates substantial uncertainty about the comparative hazard under the reported model. It does not support a narrowly determined estimate of the size or direction of the difference. The absence of arm-specific median progression-free survival values further limits direct interpretation in months.

Credible intervals summarize uncertainty in a Bayesian posterior distribution, conditional on the data and model. They should be read together with the point estimate and the model assumptions. The interval does not mean that the true value has a 98% probability of being inside a fixed frequentist interval; it is a posterior probability statement under the Bayesian analysis. The record does not describe the prior distributions or enough model detail to examine how those choices may affect the interval.

The secondary response estimates also show uncertainty. Overall response was reported as 10.0% in each group, with 80% confidence intervals from 2.7% to 24.5% for the ramucirumab combination and from 1.1% to 33.7% for the comparator. Complete response was 5.0% and 0.0%, with respective 80% confidence intervals from 0.5% to 18.1% and from 0.0% to 20.6%. These are confidence intervals, not credible intervals, and they are reported at a different interval level from the primary endpoint. Their widths reflect limited precision in the observed response estimates.

Small group sizes make estimates sensitive to the outcomes of individual participants. That is a general statistical concern, not a claim that any particular result would change under a different analysis. The intervals are more informative about precision than a point estimate alone. The registry does not report a planned precision target or a margin for judging a clinically important difference.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures for progression-free survival are total event counts, not participants per arm. They use a Schoenfeld approach. For the other planning illustrations, the record’s stated convention is participants per arm. These figures are planning illustrations based on the observed effect, not the original trial design rationale.

The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the specified fraction of the observed effect. At 100% of the observed effect, the planning illustration gives 229 total events for 80% power and 306 total events for 90% power. At 75% of the observed effect, it gives 406 total events for 80% power and 543 total events for 90% power. At 50% of the observed effect, it gives 913 total events for 80% power and 1,222 total events for 90% power.

In plain language, the 80% power column describes the event count that the planning calculation associates with an 80% chance of detecting the specified effect under its assumptions. The 90% power column gives the corresponding calculation for a higher chance of detection, again assuming the stated effect and methods. These are not guarantees about what a study will find. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The pattern illustrates a general feature of trial planning: smaller effects are harder to distinguish from random variation, so more events are needed under otherwise unchanged assumptions. These figures are not observed power. They do not measure the probability that the posted result is correct or the probability that a future study will succeed.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.690, so ln(HR) = -0.3711.
2. d = 4 × (1.96 + 0.842)² / [ln(0.690)]² = 228.0, rounded up to 229 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 0.69                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(4145349)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 229 (80% power), 306 (90% power)
 75% of observed effect: events = 406 (80% power), 543 (90% power)
 50% of observed effect: events = 913 (80% power), 1,222 (90% power)
Simulated power with 229 events: 0.791 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

The registry’s sponsor-reported primary comparison is a Bayesian hierarchical-model posterior mean hazard ratio of 0.69, with a 98% credible interval from 0.25 to 1.69. The registry reports no p-value for this comparison. It also does not provide arm-specific median progression-free survival values, despite describing posterior median estimation for the endpoint.

No discrepancy between a sponsor result and a recalculated result is flagged in the record. For this time-to-event outcome, the analysis uses the sponsor’s reported hazard ratio; it is not an independent re-estimation from the posted group summaries. A hazard ratio cannot be recovered from the group sizes and censored counts alone. The sponsor’s model-based analysis is therefore the relevant comparative estimate in the registry record, and its interval should be considered alongside its point estimate.

The posted response percentages have simple unadjusted comparisons in the analysis data, but those calculations are not the sponsor’s reported primary analysis. They should not be substituted for the Bayesian time-to-event result or treated as confirmation of it. The registry does not report sponsor analyses for the posted secondary response endpoints. These distinctions matter because analysis methods and endpoint definitions differ, and a summary statistic from one outcome does not independently validate a result for another outcome.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes in addition to progression-free survival. It does not report whether the protocol controlled the overall chance of false-positive findings across multiple outcomes, whether a hierarchy of testing was used, or whether secondary analyses were intended to be confirmatory. A multiplicity procedure is not reported in the registry record. Without that information, secondary comparisons should be read cautiously, particularly when their results are unadjusted.

The primary endpoint population included randomized participants who received at least one dose of study drug, including censored participants. The registry records 4 censored participants in the ramucirumab combination group and 2 in the comparator group. The censoring rule states that participants known to be alive and without progression were censored at their last adequate tumor assessment or at randomization, whichever was later. The registry does not give reasons for each censoring event, describe loss to follow-up, or provide additional sensitivity analyses for censoring.

Time-to-event analysis generally relies on assumptions about censoring, including that censoring does not systematically conceal a different event experience after accounting for the analysis. This is a general statistical consideration. The registry does not report analyses that test or vary this assumption. Further detail about missing endpoint assessments and the handling of missing data for the secondary outcomes is not reported in the registry record.

The stated primary population is not simply all randomized participants: it is randomized participants who received at least one dose. The registry does not report the number of randomized participants who did not receive a dose, or a separate intention-to-treat analysis. It also does not report detailed analysis populations for every secondary endpoint. The record therefore does not allow a complete comparison of results across alternative analysis populations.

Interim analyses, stopping rules, and any adjustment for repeated looks at accumulating data are not reported in the registry record. The original statistical analysis plan, model priors, and planned sensitivity analyses are also not reported. These omissions do not show that such procedures were absent; they limit what can be evaluated from the registry record.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: febrile neutropenia (10.0%), small intestinal obstruction (10.0%), pyrexia (10.0%), bacteraemia (10.0%), cellulitis (10.0%).

The registry records serious adverse events in 6 of 20 participants in the ramucirumab, cyclophosphamide, and vinorelbine group, and in 3 of 10 participants in the cyclophosphamide and vinorelbine group. It records other adverse events in 20 of 20 participants and 10 of 10 participants, respectively. Deaths are recorded for 12 participants in the combination group and 7 in the comparator group. These are group-level counts as posted; the record excerpt does not establish the cause or timing of the deaths.

The serious-event terms listed most often in the registry are febrile neutropenia, small intestinal obstruction, pyrexia, bacteraemia, cellulitis, and vascular access complication, each reported as 10.0%. The listing is descriptive and does not establish that a treatment caused an event. The record does not provide enough information in these summary figures to assess severity, timing, duration, or each participant’s relationship to treatment.

Adverse-event counts should be interpreted in light of the group sizes and the study’s specific population and follow-up. They are not a randomized comparison designed, on their own, to prove a difference in safety. The registry summary does not report detailed event-level narratives or a causal assessment for every event, so no causal conclusion follows from these figures alone.

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

The primary posted result is a model-based comparison of progression-free survival: a posterior mean hazard ratio of 0.69 with a 98% credible interval from 0.25 to 1.69. The point estimate lies below 1, but the interval spans values on both sides of 1. The result therefore has substantial statistical uncertainty about the direction and magnitude of the comparative hazard. No p-value is reported, and the registry does not provide arm-specific median progression-free survival estimates.

This result does not state how many additional months participants lived without progression, because the median values are not reported. It also does not give a direct difference in the percentage of participants who progressed. The hazard ratio is a model-based relative time-to-event statistic, and interpreting it requires attention to the model and endpoint definitions. The registry names a Bayesian hierarchical model but does not provide its prior specifications or enough detail to independently assess all assumptions.

The secondary response estimates provide additional descriptive information but have broad confidence intervals and are not reported as sponsor analyses. They do not eliminate uncertainty in the primary result. The registry’s record of adverse events likewise describes events observed in the groups but does not establish treatment causation from these counts alone.

The trial was randomized but open-label, and progression was investigator-determined. Randomization is a strength for comparative inference, while lack of masking is relevant to assessment in general. The registry does not report independent progression review or sufficient detail to assess all safeguards. The group sizes and interval width also constrain precision. These features call for measured interpretation rather than a categorical claim of benefit or lack of benefit.

The findings apply to the study population and regimen described in the registry. They do not establish how the result would generalize to other tumor types, populations, treatment combinations, or clinical settings. They are statistical descriptions of a completed trial, not medical advice or a treatment recommendation.

18. Why This Trial Matters Statistically

This trial raises a familiar statistical challenge: evaluating a time-to-event outcome in a small randomized study while some participants have incomplete event histories. Progression-free survival incorporates event timing and censoring, which can be more informative than a simple count of participants with events, but it also depends on assumptions about how incomplete follow-up is handled.

The reported hazard ratio is a compact comparison, yet the 98% credible interval is broad and includes values on either side of 1. This makes precision central to interpretation. A point estimate alone could suggest a clear direction, while the interval shows that the data and model leave considerable uncertainty about the comparison. Reporting both the estimate and interval is therefore essential.

The Bayesian hierarchical model is another important feature. Such a model can be useful in settings where related information is modeled together, but interpretation depends on the model structure and prior choices. The registry does not report enough detail to evaluate those choices. Clear reporting of model assumptions, sensitivity analyses, and the relationship between primary and secondary analyses would help readers assess how robust the conclusions are.

Finally, the trial illustrates why planning calculations must be separated from observed results. The posted time-to-event planning figures are event-count illustrations under specified assumptions, not a retrospective measure of success. Smaller effects require more events under the same assumptions. Distinguishing design-stage planning from trial evidence prevents the planning numbers from being mistaken for a conclusion about what the observed study established.

19. 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 does not report arm-specific median progression-free survival values, although the endpoint description refers to posterior median estimation. As a result, the primary comparison cannot be translated into a comparison of medians in months from the posted results. The reported hazard ratio is accompanied by a broad 98% credible interval, limiting precision.

The registry identifies a Bayesian hierarchical model but does not report its prior distributions, full specification, diagnostics, or sensitivity analyses. It also does not provide detailed sponsor analyses for the response outcomes. Simple unadjusted comparisons of secondary outcomes should not be treated as equivalent to the primary model-based result, especially because a multiplicity strategy is not reported.

The trial was open-label, and progression was determined by investigators. The registry does not say whether progression was independently reviewed or describe additional safeguards against assessment bias. Censoring counts and rules are provided, but reasons for censoring, follow-up completeness, and sensitivity analyses addressing censoring are not reported in the registry record.

Detailed information about the original sample-size rationale, randomization procedure, allocation concealment, interim analyses, and stopping rules is not reported in the registry record. The ramucirumab combination-arm regimen description is truncated, so its full treatment schedule cannot be recovered from the available record. Numerical results for several listed secondary outcomes are also not reported in the registry record.

These reporting gaps limit independent assessment; they do not establish that a procedure was not performed. The trial’s recorded group sizes and the uncertainty in its estimates also mean that the posted results should be interpreted as specific to this study rather than as a definitive answer for broader populations or other regimens.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: CAMPFIRE: A Study of Ramucirumab (LY3009806) in Children and Young Adults With Desmoplastic Small Round Cell Tumor — NCT04145349.
• Registry record last updated 2026-08-07; status last verified 2026-07.
• Results first posted 2025-07-20.
• 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.

23. Record Summary

CAMPFIRE was a completed randomized, open-label Phase 1/2 study in children and young adults with relapsed, recurrent, or refractory desmoplastic small round cell tumor. It compared ramucirumab plus cyclophosphamide and vinorelbine with cyclophosphamide and vinorelbine.

The primary endpoint was progression-free survival. The sponsor-reported Bayesian hierarchical-model posterior mean hazard ratio was 0.69, with a 98% credible interval from 0.25 to 1.69. Arm-specific median progression-free survival values and a p-value are not reported. The broad interval reflects substantial uncertainty in the comparative estimate.

Posted response outcomes are descriptive and have broad 80% confidence intervals. The registry also records serious and other adverse events, but the summary counts do not establish treatment causation. Important details of the statistical plan, missing-data assessment, and several secondary results are not reported in the registry record.