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 RADVAX: A Stratified Phase I Trial of Pembrolizumab With Hypofractionated Radiotherapy in Patients With Advanced and Metastatic Cancers (NCT02303990), a na phase 1 study. The primary outcome is Number of Adverse Events.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02303990 |
| Phase | Phase 1 |
| Condition | Metastatic Cancers |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 59 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Abramson Cancer Center at Penn Medicine |
| Results posted | Yes |
2. Clinical Question
Population
The registry describes people with metastatic cancers. Its brief summary distinguishes patients with melanoma or non-small cell lung cancer who had failed anti-PD-1 therapy from patients with breast, pancreatic, and other metastatic cancers whose disease had progressed after at least one regimen of systemic therapy. The record does not report how many participants belonged to each disease or treatment-history group.
Intervention
The single arm received hypofractionated radiotherapy to an isolated index lesion together with pembrolizumab. The arm description abbreviates the combination as hypofractionated radiotherapy and pembrolizumab. Further details of the radiation schedule, dose, treatment duration, or pembrolizumab schedule are not reported in the registry record.
Comparator
There was no separate comparator arm. The registry lists a single-group model and no allocation. Consequently, the study does not provide a randomized, like-for-like comparison with another treatment or with no treatment.
Primary question
The listed primary outcome is the number of adverse events over 2 years. That is a safety-focused outcome, not a measure of tumor response or survival benefit. The registry record does not report an observed value for this primary count or a treatment-effect estimate. It separately records serious and other adverse-event counts, which can describe reported safety events but do not establish whether the combination caused them.
In practical terms, the record describes a phase I investigation of a treatment combination in a heterogeneous metastatic-cancer population. It can be read for what the registry says about the study design and recorded adverse events. It cannot answer a comparative question about whether the combination performs better than another approach, because there is no control group. Nor should the separate disease groups be treated as randomized strata or as independent comparisons: the registry does not give their sizes, outcome results by group, or a detailed analysis plan.
3. Trial Design
RADVAX is listed as a phase I trial with a single-group model. The registry reports enrollment of 59 and a completed status. The listed start date is 2015-01-08, and primary completion is recorded as 2018-12. The sponsor is Abramson Cancer Center at Penn Medicine. These registry fields describe the study record; they do not, by themselves, explain how participants were recruited or how long each participant was observed.
The intervention combined pembrolizumab, a drug, with radiotherapy directed to an isolated index lesion. The arm description calls the radiation hypofractionated. The brief summary describes different prior-treatment circumstances across cancer types: patients with melanoma and non-small cell lung cancer had failed anti-PD-1 therapy, while patients with breast, pancreatic, and other cancers had progressed after at least one systemic-therapy regimen. The record does not report enrollment totals within these categories or detailed eligibility criteria.
Allocation is listed as not applicable, and masking is listed as none. In context, those fields accord with the single-arm design: there is no randomized treatment assignment or blinded comparison reported. The study title calls it stratified, but the registry record does not report the stratification factors, how strata were used in analysis, or whether separate stratum-specific results were planned.
The primary outcome is the number of adverse events, with a 2 years timeframe. The record lists one outcome measure and no analyses. It does not report a protocol-level statistical plan, an interim-analysis plan, a sample-size justification, or detailed rules for outcome assessment. The planning illustration described in the registry is not the trial's design justification: it is a hypothetical calculation based on stated assumptions and is not an observed-power result.
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
| Stage | Single Arm |
|---|---|
| STARTED | 59 |
| COMPLETED | 59 |
| NOT COMPLETED | 0 |
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | Single Arm |
|---|---|---|
| Participants analysed | 59 | |
| Age, Categorical: <=18 years | Participants | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 32 |
| Age, Categorical: >=65 years | Participants | 27 |
| Age, Continuous | years | 60 (27 to 90) |
| Sex: Female, Male: Female | Participants | 35 |
| Sex: Female, Male: Male | Participants | 24 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 59 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 0 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 3 |
| Race (NIH/OMB): White | Participants | 56 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Number of Adverse Events, measured over 2 years.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | Adverse Events |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is titled Number of Adverse Events, with adverse events as its unit and a timeframe of 2 years. A count endpoint records how many events are observed over a specified period. It differs from an endpoint that records only whether a participant experienced at least one event: a participant may experience multiple events, so the total event count and the number of affected participants answer different questions.
Safety event counts are used in early-phase trials to describe the types and frequency of harms observed during treatment and follow-up. As a general statistical point, counts can help characterize the overall event burden, but their interpretation depends on the event definitions, observation period, and population included in the analysis. If participants contribute different amounts of follow-up, a raw total may not represent comparable event rates unless exposure time is accounted for. The registry record does not report whether the primary count is a total across participants, a per-participant measure, or an exposure-adjusted rate.
The endpoint is listed without an observed primary count, a confidence interval, or a detailed analysis method. Its 2 years timeframe should not be taken to imply that every participant completed the same amount of follow-up; that detail is not reported in the registry record. Separate safety fields report the number of participants affected by serious and other adverse events, but those fields are not interchangeable with the primary endpoint's event count.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry describes a completed phase I study with enrollment of 59 and one treatment arm combining hypofractionated radiotherapy with pembrolizumab. Its primary outcome is a count of adverse events over 2 years. However, the registry record does not report the observed value for that primary outcome, a confidence interval, or an inferential analysis of it. The available safety fields therefore should not be mistaken for a complete result for the primary endpoint.
For the single arm, 21 of 59 participants at risk are listed as affected by serious adverse events, and 55 of 59 are listed as affected by other adverse events. The record also lists 8 deaths. These are descriptive registry counts. The serious-event terms include Death at 13.6%, Not Otherwise Specified at 10.2%, and Anemia, Dehydration, Dizziness, and Fatigue at 1.7% each. The record does not provide enough detail to determine whether these terms refer to unique participants, event episodes, or a particular attribution category. It also does not report whether the deaths were judged treatment-related.
Because there is no comparator arm, none of these counts estimates the difference in risk between the combination and another treatment. The single-arm counts describe what was recorded among participants in this study; they do not isolate the contribution of radiotherapy, pembrolizumab, underlying cancer, or other care. A phase I label indicates the trial's phase, but the registry record does not provide further detail about dose escalation, dose-limiting toxicity rules, or the protocol's decision criteria.
In general, confidence intervals express uncertainty around an estimate; narrower intervals indicate greater precision under the model used. No interval for the primary adverse-event count is reported. Likewise, a p-value would address a defined statistical comparison or hypothesis, but the record lists no analyses and supplies no such test result. The absence of an analysis result in the registry is not evidence that an event did or did not occur; it limits what can be concluded from this record.
The record supports a cautious descriptive reading of the listed safety counts and study structure. It does not report a primary endpoint estimate or comparative efficacy result. The disease mix and prior-treatment histories also mean that one should not assume a single homogeneous population or generalize a subgroup result: no subgroup outcome estimates are reported.
9. Statistical Methods Explained
Methods determine how observations become statistical summaries and what conclusions those summaries can support. The registry record identifies a single-group phase I study, one primary count endpoint, a 2 years timeframe, and no analyses. It does not report the full statistical methods used in the trial. The concepts below are general explanations, not claims that a particular method was used in RADVAX.
What does a single-group design allow?
A single-group study describes outcomes among participants who received the same study regimen. It can document observed events and their timing if those details are collected and reported. Without a separate comparison group, it cannot directly estimate how outcomes would have differed under another treatment. Differences between the observed experience and an external expectation can be affected by patient selection, cancer type, prior treatment, follow-up, and other factors. This is why the counts in the registry should be read as descriptive rather than as a randomized treatment contrast.
The record lists no allocation and no masking. Those features mean there is no reported random assignment between arms and no reported blinded treatment comparison. The study title uses the word stratified, but the factors defining strata and the statistical use of strata are not reported in the registry record. It would therefore be inappropriate to infer specific stratified analyses from the title alone.
How is a count endpoint different from a binary endpoint?
A binary endpoint indicates whether a participant had an event at least once. A count endpoint can preserve repeated events, so one participant can contribute more than one event. Counts may be analyzed with methods designed for nonnegative event totals. General examples include count regression models, but the registry record does not report any model used for the primary endpoint.
Some count models assume that the event count varies according to a particular distribution and that its variance is linked to its mean. In general, over-dispersion means the observed variability is greater than the model allows. If it is ignored, standard errors may be too small and confidence intervals or tests may be misleading. Analysts may use alternative models or robust uncertainty estimates when appropriate. This is a general concern for count data, not a finding about this trial: the record provides neither the primary event count nor an analysis model with which to assess dispersion.
Why does follow-up matter for event counts?
A total event count can be influenced by how long participants are observed. If follow-up differs, a participant observed longer has more opportunity to experience an event. A rate analysis can account for exposure time, but it requires suitable information about observation time and a defined approach to events. The registry gives a 2 years timeframe for the endpoint, but does not report participant-level follow-up, censoring rules, or whether the endpoint was analyzed as a total count or rate. No exposure-adjusted result is reported.
General count-data methods may also distinguish the probability of any event from the number of events among those with events. Those components can be useful when many participants have no event or when repeated events are common. There is no basis in this record to say that RADVAX used such a method. The primary endpoint result and detailed analysis plan are not reported in the registry record.
What do the registry safety fields summarize?
The record gives a number at risk and a number affected for serious and other adverse events in the single arm. These are participant-level summaries as labeled in the safety fields; they do not provide a full event-by-event distribution. The serious-event term percentages add descriptive detail, but the record does not provide event counts by term, attribution, timing, or overlap among participants. General statistical interpretation should distinguish the proportion of participants with at least one event from the total number of event episodes. The registry does not provide enough information to reconstruct that distinction for every listed term.
What are the planning illustrations?
A sample-size calculation asks how many participants would be needed under specified assumptions to detect a particular effect with a chosen error rate and power. General planning calculations depend on the effect being targeted, outcome variability, design, and analysis method. The registry's planning illustration is explicitly a hypothetical calculation, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is stated as a fraction of the observed effect. Those assumptions do not describe the actual single-arm design, and the illustration is not observed power. The record does not report that these assumptions were used to plan enrollment of 59.
For count outcomes, a planning method also needs assumptions about event frequency and variability. The record does not report a trial-specific sample-size justification or the inputs used to plan enrollment. Since no randomized comparison or primary analysis is reported, hypothetical two-arm planning assumptions should not be interpreted as evidence that the trial had a control arm.
Finally, statistical summaries depend on analysis populations and data handling. For example, an analysis could include all enrolled participants or a defined subset with evaluable follow-up. The registry does not report the population used for the primary endpoint, missing-data rules, or handling of participants who did not complete follow-up. These omissions limit reproducibility and constrain interpretation of the available counts.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists one primary outcome measure and no analyses. It does not report a multiplicity plan, such as whether tests were adjusted for multiple outcomes, subgroups, or repeated looks at the data. The number of analyses listed as zero should not be interpreted as proof that no analyses were conducted; it means that analysis results are not included in the registry record. Any formal testing plan is not reported in the registry record.
As a general statistical concept, multiplicity arises when several hypotheses are tested or several outcomes are examined. Without an appropriate plan, the chance of at least one false-positive finding can increase across a family of tests. Methods for controlling that risk depend on which outcomes and comparisons are confirmatory. Here, the registry provides no reported inferential results, so there is no basis for identifying a particular adjustment method or judging its effect.
Missing-data rules and analysis populations are also not reported in the registry record. The record does not state how missing event reports, incomplete follow-up, withdrawals, or participants with no evaluable endpoint data were handled. Nor does it identify whether the primary outcome was analyzed among all enrolled participants or a defined subset. Those details matter because excluding participants or using different assumptions about missing outcomes can alter a summary.
The population is described across several cancer types and prior-treatment circumstances, but subgroup analysis methods and subgroup results are not reported. The single-arm design also means there is no randomized comparison within those groups. A reader should not infer that missing multiplicity, missing-data, or analysis-population information was handled in any particular way. These gaps prevent a full assessment of the statistical plan and the robustness of any unreported primary endpoint 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 |
|---|---|---|---|
| Single Arm | 21 | 59 | 35.6% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Single Arm | 8 of 59 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Single Arm | 55 of 59 |
Most frequent serious adverse event terms
| Event term | Organ system | Single Arm |
|---|---|---|
| Death | General disorders | 8 of 59 |
| Not Otherwise Specified | General disorders | 6 of 59 |
| Anemia | Blood and lymphatic system disorders | 1 of 59 |
| Dehydration | Gastrointestinal disorders | 1 of 59 |
| Dizziness | General disorders | 1 of 59 |
| Fatigue | General disorders | 1 of 59 |
| Fever | General disorders | 1 of 59 |
| Hypotension | Cardiac disorders | 1 of 59 |
| Nausea | General disorders | 1 of 59 |
| Respiratory Failure | Respiratory, thoracic and mediastinal disorders | 1 of 59 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Single Arm |
|---|---|---|
| Fatigue | General disorders | 28 of 59 |
| Not otherwise specified | General disorders | 24 of 59 |
| Nausea | General disorders | 22 of 59 |
| Anorexia | General disorders | 21 of 59 |
| Cough | General disorders | 15 of 59 |
| Constipation | General disorders | 13 of 59 |
| Abdominal Pain | General disorders | 11 of 59 |
| Dyspnea | General disorders | 9 of 59 |
| Depression | General disorders | 8 of 59 |
| Dyspepsia | General disorders | 8 of 59 |
Serious events with the highest arm-level rate: death (13.6%), not otherwise specified (10.2%), anemia (1.7%), dehydration (1.7%), dizziness (1.7%).
The single-arm safety record lists 59 participants at risk for both serious and other adverse events. It reports 21 participants affected by serious adverse events and 55 affected by other adverse events. These are separate registry summaries; the record does not explain whether the groups overlap, so their counts should not be added together. Nor should the affected-participant counts be treated as the total number of event episodes.
Among the serious adverse-event terms listed, Death is recorded at 13.6%, Not Otherwise Specified at 10.2%, and Anemia, Dehydration, Dizziness, and Fatigue at 1.7% each. The record also lists 8 deaths. It does not specify in this summary how the death count relates to the serious-event term percentage, whether the percentages are participant-based, or whether any event was considered related to study treatment. These details are not reported in the registry record.
In general, serious describes a regulatory category of event and does not necessarily mean that every event was judged caused by the treatment. Event attribution, severity grading, timing, and the possibility of multiple terms for one participant all affect interpretation. Because those details are not reported, the figures should be treated as registry safety descriptions rather than causal estimates or a complete account of individual event histories.
12. What the Result Does, and Does Not, Mean
The registry record establishes that RADVAX was a completed, single-arm phase I trial enrolling 59 participants and evaluating pembrolizumab with hypofractionated radiotherapy. It reports one primary endpoint, the number of adverse events over 2 years, and participant counts for serious and other adverse events. These details describe the structure of the study and some recorded safety experience.
The record does not report the observed primary endpoint count, an estimate with a confidence interval, or a statistical analysis of that outcome. It also does not provide a comparative efficacy result. Accordingly, the information cannot establish whether the combination improves an outcome, whether it is better than another treatment, or whether it caused the reported events. The safety counts are not substitutes for a controlled comparison.
General interpretation of safety data requires attention to who was enrolled, what events were counted, how long participants were observed, and whether events were attributed to treatment. The study included metastatic cancers across multiple cancer types and treatment histories. The record does not give group sizes or results by cancer type, and it does not provide event timing or treatment attribution. Those omissions limit how specifically the counts can be interpreted.
The figures also should not be taken as a complete estimate of risk for every patient group or treatment setting. A single-group record cannot separate the contribution of the study regimen from the underlying disease, prior therapies, or other care. Nor can the absence of a reported primary outcome result be read as a favorable or unfavorable result. It means that this registry record does not provide that result.
The appropriate conclusion is narrow: the registry documents the design and listed safety counts, but the reported information does not support a comparative treatment-effect conclusion. Further interpretation would require outcome details, event definitions, follow-up information, analysis methods, and attribution data that are not reported in the registry record.
13. Why This Trial Matters Statistically
This trial is statistically notable because it illustrates the difference between an early-phase, single-group safety description and a comparative treatment evaluation. A phase I label and a single treatment arm set important limits on the questions the observed data can answer. Counts from a treated group can describe events recorded in that group, but they cannot show how those counts compare with what would have happened under an alternative regimen.
The primary endpoint is itself a count, which raises questions that are common in clinical research: whether repeated events are included, how follow-up is handled, and whether the analysis accounts for different time at risk. The registry names the endpoint and timeframe but does not report the observed primary count or its analysis. This makes the distinction between a clearly specified outcome label and a fully interpretable statistical result especially important.
The safety record also demonstrates why event counts and affected-participant counts must not be conflated. The record lists participants affected by serious and other events, as well as serious-event terms and a death count. These summaries do not reveal event timing, overlap, attribution, or the total number of episodes. Without those details, even a numerical safety description has limited scope.
Finally, a planning illustration based on hypothetical two-arm assumptions is different from the design actually recorded as single group. Keeping planning assumptions separate from trial conduct prevents a sample-size calculation from being mistaken for evidence of comparative power. The trial therefore offers a useful statistical case study in matching conclusions to design, endpoint, and reporting detail.
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 registry record is limited for statistical interpretation because it does not report an observed result for the primary endpoint, despite naming the endpoint and its timeframe. No confidence interval or analysis result is listed. The record also lists no analyses, leaving the statistical methods for the primary count unreported.
The single-group design has no like-for-like control arm. Any observed events cannot be contrasted with an alternative treatment group within this trial, and their cause cannot be separated from cancer, prior treatment, or other care using the information reported. The record does not provide treatment attribution for the listed events.
Population detail is incomplete for interpreting a study spanning metastatic cancers and different prior-treatment histories. The number enrolled in each cancer type or history group, the stratification factors, and subgroup outcomes are not reported in the registry record. It is therefore not possible to assess whether the safety summaries differ across these groups.
The safety fields do not supply event-level detail, including timing, duration, severity grading, recurrence, or overlap among terms. The relationship between the 8 deaths and the serious-event percentages is not explained. The registry also does not report participant-level follow-up, censoring, event definitions, or whether the 2 years timeframe was achieved consistently.
Finally, the analysis population, missing-data handling, multiplicity approach, and trial-specific sample-size justification are not reported. These limitations do not show that the study lacked such procedures; they mean the registry record does not document them. Together, they restrict reproducibility and make the posted counts unsuitable for a broader comparative or causal conclusion.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: RADVAX: A Stratified Phase I Trial of Pembrolizumab With Hypofractionated Radiotherapy in Patients With Advanced and Metastatic Cancers — NCT02303990.
- Registry record last updated 2021-05-12; status last verified 2021-04.
- Results first posted 2021-04-22.
- 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
RADVAX is a completed phase I, single-group trial of pembrolizumab with hypofractionated radiotherapy to an isolated index lesion in patients with metastatic cancers. The registry reports enrollment of 59, a start date of 2015-01-08, and primary completion in 2018-12. Allocation is listed as not applicable and masking as none.
The primary outcome is the number of adverse events over 2 years. The registry does not report the observed primary count or an analysis result. It lists 21 of 59 participants affected by serious adverse events, 55 of 59 affected by other adverse events, and 8 deaths; serious-event terms and percentages are also listed. These are descriptive safety records, not a controlled comparison or a causal estimate.
Detailed statistical methods, missing-data rules, analysis populations, subgroup results, and event attribution are not reported in the registry record. The available information supports a limited description of trial design and recorded safety counts, not a conclusion about comparative efficacy or treatment causality.