This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
POD1UM-201 was a phase 2, non-randomized, open-label, parallel-design treatment study in participants with metastatic Merkel cell carcinoma. The registry reports 107 enrolled participants and two arms, with retifanlimab listed as the study intervention.
| Feature | POD1UM-201 |
|---|---|
| Phase | Phase 2 |
| Condition | Metastatic Merkel Cell Carcinoma |
| Intervention | Retifanlimab |
| Allocation | Non-randomized |
| Design model | Parallel |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 107 |
| Number of arms | 2 |
| Status | Completed |
| Lead sponsor | Incyte Corporation |
| ClinicalTrials.gov | NCT03599713 |
2. Clinical Question
The primary statistical question was whether participants with metastatic Merkel cell carcinoma would achieve a confirmed objective response following treatment with retifanlimab, with response assessed using RECIST version 1.1 and determined by Independent Central Radiographic Review.
Population
Participants with metastatic Merkel cell carcinoma enrolled in the phase 2 POD1UM-201 study.
Intervention
Retifanlimab.
Comparator
The registry describes the study as non-randomized and lists no randomized comparator intervention in the ClinicalTrials.gov record.
Primary question
What percentage of participants achieved a confirmed complete or partial response according to RECIST v1.1 during the specified assessment period?
3. Trial Design
The absence of randomization is an important statistical feature. A non-randomized treatment study can describe the observed response experience in its enrolled population, but it does not create the same treatment-versus-control causal comparison that random allocation provides.
Two-arm structure
The registry identifies two arms and describes the overall allocation as non-randomized. The intervention listed in the registry is retifanlimab. The available record does not provide additional intervention descriptions that would support assigning separate treatment regimens to the two arms.
4. Endpoints
| Endpoint | Registry definition and time frame |
|---|---|
| Objective Response Rate (ORR) | ORR was defined as the percentage of participants with a confirmed overall response of complete response (CR) or partial response (PR), per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (v1.1), as determined by Independent Central Radiographic Review (ICR), at any post-Baseline visit until the first progressive disease (PD) or new anti-cancer therapy. Time frame: up to 26.8 months. |
Complete response
A complete response was defined as disappearance of all target and non-target lesions and no appearance of any new lesions. Any pathological lymph nodes, whether target or non-target, had to have a reduction in the short axis to <10 millimeters.
Partial response
A partial response was defined as complete disappearance or at least a 30% decrease in the sum of the diameters of target lesions, using the baseline sum of diameters as the reference, with no new lesions and no progression of non-target lesions.
5. Results Reported in the Registry
The registry reports that results have been posted, but the available record does not contain a posted statistical analysis for the primary endpoint. The numerical registry information available for this analysis therefore consists of the study-level enrollment and safety information rather than an estimated ORR, confidence interval, or p-value.
Enrollment
Total enrolled population
Participants enrolled in POD1UM-201.
Because the registry record does not provide an ORR estimate, the corresponding confidence interval, or a hypothesis-test result, there is no numerical treatment-effect estimate to interpret for ORR from the available record.
Safety results
The registry reports serious adverse events by the listed affected and at-risk populations as follows:
| Registry safety group | Affected / at risk |
|---|---|
| Chemotherapy: Naïve | 26/101 |
| Chemotherapy: Refractory | 2/6 |
| Total | 28/107 |
The total registry count is 28 affected participants among 107 at risk. The available record does not provide a randomized comparator-based safety estimate, so these counts should be read as descriptive safety information rather than as a comparative treatment effect.
6. Planned Analysis
The primary endpoint is an objective response rate, a binary endpoint defined by whether each participant achieves a confirmed complete response or partial response under the specified RECIST v1.1 framework. For this type of endpoint, the primary descriptive analysis would normally estimate the proportion of evaluable participants meeting the response definition and report a confidence interval for that proportion.
The denominator and response count used in a formal analysis must follow the prespecified analysis population and endpoint rules. A confidence interval would quantify uncertainty around the estimated response proportion.
For a single-arm or non-randomized response-rate analysis, an exact binomial confidence interval or another prespecified binomial method is commonly used. If the study protocol specifies a null response rate, a one-sample binomial hypothesis test can also be used to compare the observed response proportion with that benchmark.
The registry's endpoint definition also makes the timing of assessment important. Response is evaluated at post-Baseline visits and continues until the first progressive disease or new anti-cancer therapy. Consequently, the analysis population, confirmation rules, censoring or treatment-discontinuation rules, and handling of unevaluable participants can materially affect the estimated ORR.
7. Statistical Methodology
Objective response rate as a binary endpoint
ORR converts the longitudinal tumor-assessment process into a binary participant-level outcome: response or no response. This makes the final summary straightforward, but the statistical validity depends on applying the RECIST criteria consistently and defining the analysis population in advance.
Here, x represents the number of participants classified as responders and n represents the relevant response-analysis population.
Confidence intervals for ORR
A confidence interval around an ORR estimate describes statistical uncertainty arising from the finite sample. For a binomial response endpoint, an exact binomial interval is often preferred when sample sizes or response proportions make normal approximations less reliable.
The interval should be interpreted as uncertainty around the estimated population response probability under the specified statistical model and sampling framework. It is not a range containing the response status of individual participants.
Independent central radiographic review
The endpoint was determined by Independent Central Radiographic Review. Central review can provide a consistent assessment framework across study sites and can reduce dependence on investigator-level interpretation of radiographic changes.
RECIST v1.1 classification
The endpoint uses explicit radiographic response categories. Complete response requires disappearance of target and non-target lesions and no new lesions, while partial response requires complete disappearance or at least a 30% decrease in the sum of target-lesion diameters from baseline, together with no new lesions and no progression of non-target lesions.
Time-limited response assessment
The endpoint is assessed at any post-Baseline visit until the first progressive disease or new anti-cancer therapy, with a registry time frame of up to 26.8 months. This makes the analysis dependent not only on tumor measurements but also on the event that terminates the specified assessment window.
8. Statistical Methods Explained
Why is ORR treated as a binary endpoint?
Although tumor measurements are collected repeatedly, the registered primary endpoint reduces those observations to whether a participant achieved a confirmed CR or PR. This allows the principal result to be expressed as a proportion rather than as a continuous change in tumor size.
What does a response rate actually measure?
An ORR estimates the proportion of participants meeting the predefined response criteria. It does not directly measure how long responses last, how long participants live, or whether treatment is superior to another treatment.
Why does the RECIST definition matter?
Without a standardized definition, different investigators could classify similar tumor changes differently. RECIST v1.1 provides explicit rules for target lesions, non-target lesions, lymph nodes, new lesions, and the degree of tumor shrinkage required for partial response.
Why use independent central review?
Independent central review provides a separate radiographic assessment of response. In an open-label study, this can be particularly relevant because treatment assignment is not masked to participants or investigators.
Why is a confidence interval useful for ORR?
A point estimate alone gives no direct indication of sampling precision. A confidence interval shows how much statistical uncertainty surrounds the estimated response proportion and is therefore more informative than the percentage alone.
Why does non-randomization matter?
Randomization balances measured and unmeasured prognostic factors in expectation. In a non-randomized study, observed response can be influenced by the characteristics of the enrolled population and other factors in addition to treatment, so the response proportion should not automatically be interpreted as a causal comparison with an untreated population.
9. Interpreting the Registry's Safety Data
26/101
The registry reports serious adverse events affecting 26 of 101 participants in the chemotherapy-naïve group.
2/6
The registry reports serious adverse events affecting 2 of 6 participants in the chemotherapy-refractory group.
28/107
The registry reports 28 affected participants among 107 participants overall.
Descriptive evidence
These counts describe serious adverse events and do not by themselves establish a comparative safety effect.
Safety and efficacy answer different statistical questions. ORR describes tumor response under a prespecified radiographic definition, whereas serious adverse-event counts describe an important category of harm. Neither measure can substitute for the other.
10. Clinical Biostats Interpretation
The primary endpoint is the proportion of participants who achieved a confirmed CR or PR according to RECIST v1.1 during the specified assessment period. A higher ORR would mean that a larger proportion of the analyzed population met those predefined response criteria.
It would not by itself mean that every responder experienced the same magnitude of tumor shrinkage, that responses were durable, or that treatment improved survival.
If an ORR estimate and confidence interval were reported, the interval would quantify statistical precision around the observed response proportion. A narrower interval would generally indicate greater precision than a wider interval, all else being equal.
If a hypothesis test were performed against a prespecified response-rate benchmark, its p-value would quantify evidence against that null hypothesis under the chosen statistical model. It would not tell us how large the response rate is, how clinically important the response is, or how long responses last.
Because allocation was non-randomized and masking was absent, the study is primarily informative about response and safety in the enrolled population. A response proportion cannot by itself establish what the response rate would have been under a different treatment in the same participants.
11. Important Interpretation Issues
- Non-randomized allocation: treatment comparisons based on non-randomized groups can be affected by differences in participant characteristics and other sources of confounding.
- No masking: the registry describes the study as unmasked. Objective central radiographic review can provide an independent assessment of response, but the overall study remains open-label.
- Response is not survival: ORR measures achievement of CR or PR and does not directly quantify overall survival or progression-free survival.
- Confirmation matters: the registered endpoint requires a confirmed overall response rather than a single unconfirmed tumor measurement.
- Assessment window matters: the endpoint is defined at post-Baseline visits until the first PD or new anti-cancer therapy, with a time frame of up to 26.8 months.
- Analysis population matters: the denominator used for a formal ORR analysis can materially affect the estimate. The available registry information does not provide a numerical response-analysis population.
- Safety denominators differ: the registry reports serious adverse events separately for chemotherapy-naïve and chemotherapy-refractory groups, with 101 and 6 participants at risk, respectively.
- No comparative efficacy estimate: the available registry record does not provide an ORR estimate, confidence interval, or p-value in a posted statistical analysis.
12. Trial Timeline
Study start
POD1UM-201 began on 2019-02-25.
Study conduct
The phase 2 study evaluated retifanlimab in participants with metastatic Merkel cell carcinoma under a non-randomized, parallel, unmasked design.
Primary completion
The registry lists 2022-01-21 as the primary completion date.
Registry status
The ClinicalTrials.gov record identifies the study status as completed.
13. Why This Trial Matters Statistically
POD1UM-201 is useful as a teaching case because its design illustrates an important distinction between estimating an observed response rate and making a randomized comparative treatment claim.
| Concept | How it appears in POD1UM-201 |
|---|---|
| Binary efficacy endpoint | ORR classifies participants according to confirmed CR or PR. |
| RECIST v1.1 | Response depends on predefined rules for tumor disappearance and reduction in target-lesion diameters. |
| Independent central review | Primary response assessment is determined by ICR. |
| Non-randomized design | Participants were not randomly allocated to treatment groups. |
| Open-label design | Masking was listed as none. |
| Confidence intervals | A formal ORR analysis would normally quantify uncertainty around the response proportion. |
| Analysis population | The denominator used for ORR is critical because the endpoint is a proportion. |
| Safety analysis | Serious adverse events are reported descriptively by the listed chemotherapy-history groups. |
14. Limitations
Non-randomized allocation
The study does not provide the balance created by randomization, limiting causal interpretation of an observed response rate.
Unmasked treatment
The registry identifies masking as none. Independent central review is relevant to the objectivity of radiographic assessment.
Limited endpoint set in the record
The available registered primary endpoint is ORR. Other efficacy outcomes should not be inferred from the registry record.
No posted formal analysis
The registry record contains no posted statistical analysis with an ORR estimate, confidence interval, or p-value.
The most important limitation for quantitative interpretation is the absence of a posted numerical ORR analysis in the available registry record. Without the response count, response-analysis denominator, confidence interval, and hypothesis-test result, the primary endpoint cannot be summarized as a numerical treatment effect from the registry information available here.
15. Statistical Concepts in This Trial
The study connects several fundamental clinical-trial concepts: binary response endpoints, standardized radiographic criteria, central review, confidence intervals, non-randomized treatment evaluation, and descriptive safety analysis.
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: NCT03599713 — POD1UM-201.
Continue through the Clinical Biostats knowledge graph
Explore statistical methods, clinical-trial endpoints, and analysis tools that provide deeper context for interpreting clinical research.
18. Record Summary
POD1UM-201 is a completed phase 2, non-randomized, parallel, unmasked treatment study of retifanlimab in metastatic Merkel cell carcinoma. The registry identifies objective response rate as the primary endpoint, defined by confirmed complete or partial response according to RECIST v1.1 and determined by Independent Central Radiographic Review, with assessments continuing until the first progressive disease or new anti-cancer therapy and a time frame of up to 26.8 months.
The statistical interpretation of this study therefore centers on a single-arm-style response-rate framework rather than a randomized hazard-ratio comparison. A formal ORR analysis would ordinarily report the number of responders, the response-analysis denominator, a confidence interval, and, where prespecified, a hypothesis test against a benchmark response rate. The available registry record does not provide those numerical statistical results.
The registry does provide descriptive safety information: serious adverse events affected 26 of 101 participants in the chemotherapy-naïve group, 2 of 6 in the chemotherapy-refractory group, and 28 of 107 participants overall. Those figures provide a safety description but do not establish a comparative treatment effect.