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 Single-arm Trial to Evaluate the Role of the Immune Response to Talimogene Laherparepvec in Unresected Melanoma (NCT02366195), a na phase 2 study. The primary outcome is Odds Ratio of Baseline Intratumoral CD8+ Cell Density and Objective Response Rate.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02366195 |
| Phase | Phase 2 |
| Condition | Unresected Stage IIIb to IVM1c Melanoma |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 112 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Amgen |
| Results posted | Yes |
2. Clinical Question
Population
Adults with unresected stage IIIB to IVM1c melanoma were eligible for the study. The primary analysis population was narrower: participants who received at least one dose of talimogene laherparepvec and had non-missing baseline intratumoral CD8+ cell density data. The registry record states that data were available for 91 participants at primary completion.
Intervention
Participants received talimogene laherparepvec by intralesional injection into injectable cutaneous, subcutaneous, and nodal lesions. The record describes an initial dose of 10⁶ plaque-forming units per mL on day 1, followed by 10⁸ plaque-forming units per mL 21 days after the initial dose and every 14 days thereafter.
Comparator
There was no separate comparator group. This was an open-label, single-arm study, so participants receiving talimogene laherparepvec were not compared with a concurrent untreated group or another treatment group in this trial.
Primary question
The primary analysis asked whether baseline intratumoral CD8+ cell density was associated with objective response among treated participants with baseline CD8+ data. Specifically, a univariate logistic regression model evaluated baseline log2(CD8+ cell density) as a predictor of objective response, defined as complete or partial response under modified WHO criteria.
This is an association question within a treated group, not a randomized comparison of treatment effects. The clinical and statistical interpretation therefore depends on the estimated relationship between the biomarker and response, together with its uncertainty. The numerical odds ratio and its 95% confidence interval are not reported in the registry record, so the direction, size, and precision of that estimated association cannot be assessed from the record.
3. Trial Design
This was a phase 2, multicenter, open-label, single-arm trial of talimogene laherparepvec in adults with unresected stage IIIB to IVM1c melanoma. The registry record lists planned enrollment of 112 participants and records the study as completed. It gives a start date of 2015-04-07 and a primary completion date of 2017-06-26. Because there was no separate comparator arm, the trial design does not support a like-for-like comparison between talimogene laherparepvec and another treatment or control.
Talimogene laherparepvec was administered by intralesional injection into injectable cutaneous, subcutaneous, and nodal lesions. The record describes an initial dose of 10⁶ plaque-forming units per mL on day 1, followed by a dose of 10⁸ plaque-forming units per mL 21 days after the initial dose and every 14 days thereafter. The available arm description is incomplete after this dosing information, so further details are not reported in the registry record.
The primary objective was to evaluate the association between baseline intratumoral CD8+ cell density and objective response. The primary analysis included participants who received at least one dose and had non-missing baseline CD8+ data. The record states that this information was available for 91 participants at primary completion. Response was assessed every 12 weeks until disease progression beyond 6 months of treatment or the start of new anticancer therapy. Median follow-up at primary completion was 59 weeks, with a minimum of 3 weeks and a maximum of 116 weeks.
The study also planned to evaluate safety and tolerability, and the registry lists secondary outcomes concerning response, duration of response, tumor burden, and changes in CD8+ cell density. The detailed statistical plan, including any sample-size justification or interim-analysis plan, is 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
| Stage | Talimogene Laherparepvec |
|---|---|
| STARTED | 112 |
| Received Study Drug | 111 |
| COMPLETED | 58 |
| NOT COMPLETED | 54 |
Reasons for not completing this period:
| Reason | Talimogene Laherparepvec |
|---|---|
| Protocol-specified criteria | 3 |
| Withdrawal by Subject | 9 |
| Lost to Follow-up | 1 |
| Death | 41 |
Analysis population for the primary outcome, as defined in the registry: Participants who received at least 1 dose of talimogene laherparepvec with non-missing baseline CD8+ cell density data. Data is available for 91 participants who had data available for analysis at the time of primary completion (26 June 2017).
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: Participants who received at least 1 dose of talimogene laherparepvec.
| Characteristic | Unit | Talimogene Laherparepvec |
|---|---|---|
| Participants analysed | 111 | |
| Age, Continuous | years | 65.7 (SD 15.1) |
| Sex: Female, Male: Female | Participants | 62 |
| Sex: Female, Male: Male | Participants | 49 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 1 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 110 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Race/Ethnicity, Customized: American Indian or Alaska Native | Participants | 0 |
| Race/Ethnicity, Customized: Asian | Participants | 0 |
| Race/Ethnicity, Customized: Black or African American | Participants | 0 |
| Race/Ethnicity, Customized: Multiple | Participants | 0 |
| Race/Ethnicity, Customized: Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race/Ethnicity, Customized: White | Participants | 111 |
| Race/Ethnicity, Customized: Other | Participants | 0 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 0 (Fully active, no restrictions) | Participants | 87 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 1 (Restricted but ambulatory) | Participants | 24 |
| Log2(Intratumoral Cluster of Differentiation 8-positive (CD8+) Cell Density): Primary completion | Log2 CD8+ cells/mm² | 8.041 (SD 1.903) |
| Log2(Intratumoral Cluster of Differentiation 8-positive (CD8+) Cell Density): Final analysis | Log2 CD8+ cells/mm² | 8.109 (SD 1.940) |
6. Primary Endpoint
The registry lists the primary outcome as Odds Ratio of Baseline Intratumoral CD8+ Cell Density and Objective Response Rate, measured over Intratumoral CD8+ cell density: Baseline. Response: every 12 weeks until disease progression beyond 6 months of treatment or the start of new anticancer therapy; median time on follow-up at primary completion = 59 weeks (min 3 to max 116 weeks).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | ratio |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the odds ratio relating baseline intratumoral CD8+ cell density to objective response. Objective response was defined as complete response or partial response according to a modified version of the WHO response criteria. A univariate logistic regression model evaluated baseline log2(CD8+ cell density) as a predictor of that response outcome. The reported odds ratio is described as unadjusted.
In general, an odds ratio expresses how the odds of an outcome relate to a predictor. In this analysis, the predictor is baseline log2(CD8+ cell density), and the outcome is whether a participant had an objective response. The confidence interval communicates uncertainty around the estimated association; a wider interval would indicate less precision than a narrower one. Neither the numerical odds ratio nor its 95% confidence interval is reported in the registry record, so the association’s magnitude and precision cannot be interpreted from the posted results.
A biomarker-response association can help characterize which baseline features coincide with response among people receiving a treatment. It does not, by itself, establish that the biomarker causes a response or that treatment changes outcomes relative to an alternative. The single-arm design is important to that distinction: the primary endpoint concerns association within the treated group, rather than a between-group treatment effect.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record describes an analysis of whether baseline intratumoral CD8+ cell density predicts objective response among participants treated with talimogene laherparepvec. It specifies a univariate logistic regression using baseline log2(CD8+ cell density) and an unadjusted odds ratio with a 95% confidence interval. However, the numerical odds ratio, confidence interval, and associated test result are not reported in the registry record. Consequently, the posted information does not establish whether the estimated association was positive or negative, how large it was, or how precisely it was estimated.
The primary analysis population consisted of treated participants with non-missing baseline CD8+ data, with data available for 91 participants at primary completion. This population is not identical by definition to everyone enrolled or everyone included in safety reporting. The registry record lists enrollment of 112, while the safety group lists 111 participants at risk for serious and other adverse events. These figures refer to different recorded populations and should not be treated as interchangeable. The record does not explain the reasons for each population difference.
Objective response was defined as complete or partial response under modified WHO criteria. Response assessment was planned every 12 weeks until disease progression beyond 6 months of treatment or the start of new anticancer therapy. The stated median follow-up at primary completion was 59 weeks, ranging from 3 weeks to 116 weeks. These details describe the assessment context, but they do not fill in the missing primary estimate or show how many participants experienced each response category.
Because this was a single-arm, open-label trial, there is no concurrent control group for estimating a comparative treatment effect. Even if the biomarker association estimate were available, it would describe variation in response within the treated population; it would not show that treatment caused responses or that the treatment outperformed another approach. The design also cannot separate treatment effects from other factors that may be related to both baseline CD8+ density and response.
Planning illustrations based on an observed effect are not evidence of observed power, and they are not the trial’s original sample-size rationale. The registry record does not report a sample-size justification or a full statistical analysis plan. In particular, it does not report how missing biomarker or response data were handled, whether covariates were considered in other models, or whether adjustments were made for multiple analyses. Those omissions limit assessment of the analysis beyond the details explicitly described for the primary endpoint.
9. Statistical Methods Explained
The primary analysis used univariate logistic regression. In general, logistic regression is used when the outcome has two categories, such as whether a participant did or did not meet a response definition. The model relates the odds of the outcome to one or more predictors. Here, the registry record identifies baseline log2(CD8+ cell density) as the predictor and objective response as the outcome. “Univariate” indicates that the stated primary model evaluates this predictor without listing additional covariates in that model.
What the odds ratio represents
The odds ratio describes the modelled association between the predictor and the odds of objective response. Because the predictor is specified on the log2 scale, the interpretation is tied to that transformed measure rather than to an untransformed cell-density value. The registry record says the reported odds ratio is unadjusted. In general, an unadjusted estimate does not account for other variables through adjustment in the model. It should therefore be read as the model’s stated association, not as a causal effect or an estimate adjusted for potential confounding.
The record does not include the numerical odds ratio. It also does not give the confidence limits or a p-value for the primary estimate. As a result, it is not possible to describe the strength, direction, statistical uncertainty, or compatibility of the estimate with no association. A confidence interval, in general, represents uncertainty around an estimate under the assumptions of the analysis; its actual limits are needed to assess precision in this trial.
How response was defined
Objective response was the endpoint outcome and was defined as complete response or partial response using a modified version of the WHO response criteria. This combines the two response categories into a binary outcome for the logistic regression. The record states that response assessments were scheduled every 12 weeks until disease progression beyond 6 months of treatment or the start of new anticancer therapy. It does not report the number of participants in each response category or further details on how assessments were adjudicated.
In general, a binary response endpoint makes the analysis straightforward to express as odds of response, but it does not retain all the detail that may be present in the course of tumor change over time. Other listed outcomes address duration of response, tumor burden, and changes in CD8+ cell density. Their presence does not substitute for the missing numerical result of the primary analysis, and their detailed estimates are not reported in the registry record.
Analysis population and follow-up
The primary analysis population was restricted to participants who received at least one dose and had non-missing baseline CD8+ cell density data. The record states that 91 participants had data available for analysis at primary completion. This is a defined analysis population, not a statement that every enrolled participant contributed to the primary model. The reasons for missing baseline data and the handling of any missing response outcomes are not reported in the registry record.
The median time on follow-up at primary completion was 59 weeks, with a minimum of 3 weeks and a maximum of 116 weeks. In general, differences in follow-up duration can affect the opportunity to observe outcomes, especially outcomes that depend on time. For the primary binary response analysis, the record gives the assessment schedule and follow-up description but does not explain whether or how differences in assessment availability were handled in the model.
Single-arm design and interpretation
The analysis estimates an association within a group receiving the study treatment. There is no randomized comparator, and allocation is listed as not applicable. In general, randomization can help balance measured and unmeasured baseline characteristics between groups; this design does not make that comparison. A relationship between CD8+ density and response therefore cannot establish that CD8+ density causes response, nor can it isolate the effect of treatment from other influences.
The study was open-label, with no masking. The registry record does not provide additional information about blinded assessment or other procedures intended to limit assessment bias. The detailed statistical analysis plan, including model diagnostics, assumptions, and any sensitivity analyses, is not reported in the registry record. These details matter because the validity and interpretation of a model depend not only on its name but also on its implementation and the adequacy of its assumptions.
Planning illustrations are not trial results
Any sample-size planning illustrations based on an observed effect are hypothetical planning exercises rather than observed power. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such illustrations do not change the actual single-arm design, and they should not be read as the trial’s original design rationale. The registry record does not report the trial’s own sample-size justification.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists a primary endpoint and multiple secondary outcomes, including associations involving baseline or change in CD8+ cell density, objective or durable response, duration of response, and tumor burden. It also reports 15 outcome measures and 15 analyses. The presence of multiple outcomes raises a general multiplicity question: when many hypotheses are examined, the chance of at least one apparently notable result can increase. The registry record does not state whether multiplicity adjustments, a testing hierarchy, or other procedures were used. It is therefore not possible to determine how the reported analyses, if any, were interpreted jointly.
The primary analysis population required receipt of at least one dose and non-missing baseline CD8+ cell density data. The record states that 91 participants had data available for analysis at primary completion. It does not describe why baseline data were missing for others, how missing response assessments were handled, whether participants were excluded for other reasons, or whether any imputation or sensitivity analyses were performed. These details are not reported in the registry record.
The safety group lists 111 participants at risk, while planned enrollment was 112. The primary population is defined by treatment exposure and available baseline biomarker data, so the population sizes should be interpreted according to their stated purposes rather than assumed to be identical. The record does not provide a participant-flow account explaining the differences. It also does not report a full analysis plan or a sample-size justification. Without these details, readers cannot assess all planned versus conducted analyses, the impact of missing information, or the role of multiplicity in interpreting the findings.
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 |
|---|---|---|---|
| Talimogene Laherparepvec | 33 | 111 | 29.7% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Talimogene Laherparepvec | 41 of 112 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Talimogene Laherparepvec | 97 of 111 |
Most frequent serious adverse event terms
| Event term | Organ system | Talimogene Laherparepvec |
|---|---|---|
| Metastatic malignant melanoma | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 3 of 111 |
| General physical health deterioration | General disorders | 2 of 111 |
| Pyrexia | General disorders | 2 of 111 |
| Back pain | Musculoskeletal and connective tissue disorders | 2 of 111 |
| Metastases to central nervous system | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 2 of 111 |
| Cardiac failure | Cardiac disorders | 1 of 111 |
| Cardio-respiratory arrest | Cardiac disorders | 1 of 111 |
| Vertigo positional | Ear and labyrinth disorders | 1 of 111 |
| Faecaloma | Gastrointestinal disorders | 1 of 111 |
| Cholecystitis | Hepatobiliary disorders | 1 of 111 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Talimogene Laherparepvec |
|---|---|---|
| Pyrexia | General disorders | 52 of 111 |
| Chills | General disorders | 30 of 111 |
| Influenza like illness | General disorders | 29 of 111 |
| Nausea | Gastrointestinal disorders | 26 of 111 |
| Fatigue | General disorders | 24 of 111 |
| Headache | Nervous system disorders | 21 of 111 |
| Arthralgia | Musculoskeletal and connective tissue disorders | 20 of 111 |
| Injection site pain | General disorders | 19 of 111 |
| Asthenia | General disorders | 18 of 111 |
| Pain in extremity | Musculoskeletal and connective tissue disorders | 16 of 111 |
Serious events with the highest arm-level rate: metastatic malignant melanoma (2.7%), general physical health deterioration (1.8%), pyrexia (1.8%), back pain (1.8%), metastases to central nervous system (1.8%).
The safety record lists 111 participants at risk for serious adverse events and 33 participants affected by at least one serious adverse event. It lists the same 111 as at risk for other adverse events, with 97 affected. These are participant counts, not rates of events per person-time. A participant may experience more than one event, so affected-participant counts should not be read as a count of all event occurrences.
The most frequently listed serious-event terms include metastatic malignant melanoma at 2.7%; general physical health deterioration, pyrexia, back pain, and metastases to central nervous system each at 1.8%; and cardiac failure at 0.9%. These percentages are registry-reported values. The record also lists 41 deaths. It does not state in this summary whether those deaths were attributed to treatment, nor does it establish that a serious adverse event caused a death.
These figures describe events recorded in a single-arm study. Without a comparator group, they cannot show whether event occurrence was more or less common than it would have been under another treatment or no study treatment. The registry record does not provide additional detail on event timing, severity grading beyond the serious-event designation, duration, or causal attribution.
12. What the Result Does, and Does Not, Mean
The primary analysis was designed to examine whether baseline intratumoral CD8+ cell density was associated with objective response among participants treated with talimogene laherparepvec. The registry record specifies a univariate logistic regression model and identifies the analysis population as treated participants with non-missing baseline CD8+ data. It states that data were available for 91 participants at primary completion. However, the numerical odds ratio and its 95% confidence interval are not reported in the registry record. The result’s direction, magnitude, and precision therefore cannot be described from the posted information.
Even a fully reported association would answer a limited question: whether the baseline biomarker and response varied together within the treated group under the analysis used. It would not establish that baseline CD8+ cell density causes a response, or that it predicts a treatment-specific benefit. To establish a comparative treatment effect, a comparison with an appropriate control group would be needed; this trial had no separate comparator.
The single-arm design also means that changes or responses cannot be separated from other possible influences using a concurrent control comparison. The open-label nature of the trial is another feature to consider when interpreting assessments, although further information on assessment procedures is not reported in the registry record. The specified endpoint uses modified WHO criteria, and assessments were scheduled over follow-up, but the record does not provide the primary response counts or model estimate.
Safety results likewise describe events in the treated group, not comparative risks. The listed counts and percentages identify events recorded in the registry, but do not show whether treatment caused them or how they compare with events under another strategy. Overall, the record documents the study question, design, analysis approach, and some safety counts, but does not report the main numerical efficacy association needed to judge its size or uncertainty.
13. Why This Trial Matters Statistically
This trial is statistically notable because its primary question concerns a biomarker-response association rather than a direct comparison of treatment groups. Baseline intratumoral CD8+ cell density was evaluated as a predictor of objective response in participants receiving talimogene laherparepvec. That question can be relevant to understanding variation among treated participants, but it is distinct from asking whether the treatment changes response relative to an alternative.
The design makes that distinction especially important. The study was single-arm and open-label, with no separate comparator. A model relating a baseline marker to response can quantify an association within the observed group, but it cannot by itself attribute that association to a treatment effect or establish that the marker identifies treatment-specific benefit. Interpretation must also account for the defined analysis population: treated participants with non-missing baseline CD8+ data.
The registry record gives several elements needed to understand the intended analysis: the endpoint definition, the use of univariate logistic regression, the follow-up description, and the eligible analysis population. Yet the numerical primary odds ratio and its confidence interval are not reported. Without them, readers cannot assess the estimated association or its precision. This is a useful reminder that a clearly stated statistical question and method are not enough to interpret a result when the estimate itself is absent.
The record also lists multiple secondary outcomes and analyses, while not reporting whether multiplicity was addressed. It does not report a complete statistical analysis plan, a sample-size justification, or detailed handling of missing data. These gaps matter because they affect how readers assess robustness, uncertainty, and the scope of conclusions. Safety counts offer descriptive context but, without a control group or causal attribution information, do not provide comparative risk estimates.
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 central limitation for interpreting the primary endpoint is that the numerical odds ratio and its 95% confidence interval are not reported in the registry record. The model is described, but without its estimate there is no basis for judging the direction, size, or precision of the biomarker-response association. The record also does not report a primary p-value or response counts.
The single-arm, open-label design has no concurrent comparator. It can describe outcomes and associations among participants receiving talimogene laherparepvec, but does not support a like-for-like estimate of treatment effect against another group. The primary analysis is additionally restricted to treated participants with non-missing baseline CD8+ data. The record states that data were available for 91 participants, but does not explain the missingness or report sensitivity analyses for its possible effects.
The registry lists multiple secondary outcomes and analyses, but does not state whether multiplicity was addressed or identify a testing hierarchy. It also does not report the full statistical analysis plan, model diagnostics, or the trial’s sample-size justification. These omissions limit assessment of the analysis choices and the robustness of conclusions.
For safety, the registry reports affected-participant counts, selected serious-event terms, and deaths, but does not provide comparative rates or causal attribution in the information summarized. The arm description is incomplete after the dosing schedule, and further details of response assessment and event follow-up are not reported in the registry record. These limitations should be kept separate from the described trial details: they identify what cannot be concluded from the registry record, rather than demonstrating that an unreported procedure or analysis did or did not occur.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Single-arm Trial to Evaluate the Role of the Immune Response to Talimogene Laherparepvec in Unresected Melanoma — NCT02366195.
- Registry record last updated 2021-11-30; status last verified 2021-11.
- Results first posted 2018-07-18.
- 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
The registry record describes NCT02366195, a completed phase 2, multicenter, open-label, single-arm trial of talimogene laherparepvec in unresected stage IIIB to IVM1c melanoma. Planned enrollment was 112. The primary question concerned the association between baseline intratumoral CD8+ cell density and objective response. A univariate logistic regression model evaluated baseline log2(CD8+ cell density) as a predictor of response, defined as complete or partial response under modified WHO criteria.
The primary analysis population comprised treated participants with non-missing baseline CD8+ data; the record states that data were available for 91 participants at primary completion. Median follow-up at primary completion was 59 weeks, with a minimum of 3 weeks and a maximum of 116 weeks. The record lists 111 participants at risk for safety events, 33 affected by serious adverse events, 97 affected by other adverse events, and 41 deaths. The primary odds ratio and its 95% confidence interval are not reported in the registry record.