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 Study of Atezolizumab in Relapsed or Refractory Hodgkin Lymphoma (NCT03120676), a na phase 2 study. The primary outcome is Progression-free Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03120676 |
| Phase | Phase 2 |
| Condition | Hodgkin Lymphoma |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 3 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Memorial Sloan Kettering Cancer Center |
| Results posted | Yes |
2. Clinical Question
Population
Patients with relapsed or refractory Hodgkin lymphoma. The registry record lists an enrollment of 3.
Intervention
Atezolizumab, administered at 1200 mg intravenously every 3 weeks. Treatment was described as continuing until progression, intolerable toxicity or death, for a maximum of 36 cycles.
Comparator
There was no comparator group. The study is recorded as single-group, with allocation listed as NA.
Primary question
Progression-free survival at 12 months, expressed as a proportion with an exact 95% confidence interval. The registry record says the primary endpoint data were not collected.
The clinical question was whether progression-free survival could be described for people with relapsed or refractory Hodgkin lymphoma receiving atezolizumab in this study. Because the design is single-group, it does not set up a direct comparison with another treatment or a control condition. It can therefore describe the study group, but the record does not support a comparative estimate of treatment effect.
The registry summary also describes safety and efficacy as study aims and mentions biomarker research. Those aims do not change the primary endpoint recorded for the study. The recorded primary endpoint is progression-free survival at 12 months; the registry does not include collected data for it or an analysis result. The exact definition and assessment procedures for progression or survival are not reported in the registry record, so the endpoint’s operational details cannot be reconstructed from the record.
3. Trial Design
This was a Phase II, single-group study of atezolizumab in relapsed or refractory Hodgkin lymphoma. The registry lists no allocation method and no masking. A single-group design means the study intervention is not contrasted with a separate control group in the registered design. Accordingly, any interpretation is limited to the experience recorded for the study group, rather than a randomized or otherwise like-for-like comparison.
The registry describes atezolizumab at 1200 mg intravenously every 3 weeks. A cycle was defined as 3 weeks of therapy. Treatment was to continue until disease progression, intolerable toxicities or death, with a maximum treatment duration of 36 cycles. These are the registered treatment instructions; the record does not provide treatment exposure or adherence data.
The listed enrollment is 3. The study status is terminated. The record gives a start date of 2017-04-14 and a primary completion date of 2018-11-12. It does not report why the study was terminated, so no reason can be inferred from the status alone. Nor does the record state whether the enrollment reflected the planned target or why that enrollment was chosen.
The registry lists one outcome measure and no analyses. Its primary endpoint description specifies proportions with exact 95% confidence intervals at 12 months, but its population entry says data were not collected. Thus the design and endpoint can be described, but a primary endpoint estimate cannot be interpreted from the registry record. Details such as eligibility criteria beyond the disease description, follow-up procedures, and the statistical analysis 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
| Stage | Atezolizumab |
|---|---|
| STARTED | 3 |
| COMPLETED | 2 |
| NOT COMPLETED | 1 |
Reasons for not completing this period:
| Reason | Atezolizumab |
|---|---|
| Not treated | 1 |
Analysis population for the primary outcome, as defined in the registry: Data not collected
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 | Atezolizumab |
|---|---|---|
| Participants analysed | 3 | |
| Age, Continuous | years | 33 (28 to 37) |
| Sex: Female, Male: Female | Participants | 1 |
| Sex: Female, Male: Male | Participants | 2 |
| 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 | 0 |
| Race (NIH/OMB): White | Participants | 2 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 1 |
| Region of Enrollment: United States | Participants | 3 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-free Survival, measured over 12 months.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is progression-free survival at 12 months. In general, progression-free survival concerns the time during which a participant remains alive without disease progression. A fixed-time proportion describes the share of a defined group meeting the endpoint status at the specified time. The registry describes the endpoint as proportions with exact 95% confidence intervals.
In general, a confidence interval communicates uncertainty around an estimated quantity: it indicates a range of values compatible with the data under the method’s assumptions. An exact interval for a proportion uses a probability model for discrete outcomes rather than relying only on a large-sample approximation. With limited information, an interval may be imprecise; its actual width depends on the observed data and cannot be stated without an estimate.
The record does not report the precise operational definition used for progression-free survival, including the event criteria, assessment schedule, censoring rules or handling of participants without a recorded event. Those details matter because endpoint construction determines which observations count and how follow-up contributes to the estimate.
Most importantly, the registry states that primary endpoint data were not collected. No proportion or confidence interval is therefore reported, and the endpoint cannot be estimated from this record. The endpoint’s stated purpose and method should not be mistaken for an observed result.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record supports a description of the study design and its intended primary endpoint, but it does not provide a progression-free survival result. The primary endpoint is named as progression-free survival at 12 months, and the planned reporting format is a proportion with an exact 95% confidence interval. The population entry explicitly says data were not collected. The registry lists no analyses.
Consequently, there is no posted estimate from which to assess the observed proportion, its uncertainty, or the number of participants contributing to the endpoint analysis. No conclusion about the size or direction of an observed effect can be drawn. This is different from an estimate that is close to a threshold or an estimate with a wide confidence interval: in this record, the endpoint result itself is absent.
The enrollment listed is 3, but that entry does not supply an endpoint numerator, an analysis denominator, follow-up information or an estimate. It cannot be used by itself to reconstruct progression-free survival. The record also does not explain whether the missing endpoint data reflect early termination, data availability, or another circumstance. The study status is terminated, but the reason for termination is not reported in the registry record.
The adverse-event record also requires careful reading. For the atezolizumab group, it lists zero participants at risk for serious events and zero affected, zero at risk for other events and zero affected, and zero deaths. Because the record lists zero at risk, these zero counts do not establish that no adverse events occurred among treated participants. They provide no usable event rate or estimate of risk in this study.
In general, a single-group study can describe outcomes observed among participants under its treatment conditions, but without a comparator it cannot separate treatment effects from other explanations for those outcomes. In this case, the absence of primary endpoint data further limits interpretation: even a descriptive estimate is unavailable. The registry record therefore documents a study question and treatment plan, not a result that answers the efficacy question.
No randomized comparison, adjusted effect estimate, or uncertainty interval is reported. Such quantities should not be inferred from the study’s stated aims, status, or enrollment. The record also does not report an analysis population, follow-up distribution, or reasons for missing endpoint data. These omissions prevent assessment of how representative any potential observed result would have been, even if one were available.
9. Statistical Methods Explained
The registry describes a proportion-based primary endpoint and specifies an exact 95% confidence interval. These statements describe the intended form of the analysis, not a reported analysis: the endpoint population is marked as data not collected, and the registry lists no analyses. The method therefore has no posted estimate to summarize.
What does a proportion estimate?
In general, a proportion is a count of participants meeting a defined condition divided by the relevant analysis population. For a fixed-time endpoint, the condition must specify the participant’s status at that time and how incomplete follow-up is handled. The registry identifies progression-free survival at 12 months but does not report the operational rules or the observations needed to calculate the proportion. It is not possible to infer a numerator or denominator from the enrollment entry alone.
What does an exact confidence interval add?
In general, a confidence interval describes statistical uncertainty around an estimate. An exact interval for a proportion uses the discrete probability structure of the data and avoids depending on a normal approximation for that interval. It does not remove uncertainty, correct missing data, or guarantee that the endpoint was measured consistently. Its width depends on the observed data. Because no endpoint estimate is posted, the interval’s limits and precision are not available.
What can a single-group design establish?
In general, a single-group design can report outcomes observed in the group receiving the intervention. Without a control group, the design does not directly estimate how outcomes would differ under another treatment or under no study intervention. Differences between observed outcomes and an external expectation would also require a defined comparison and assumptions not reported in this record. The registry lists no randomized comparison, and its allocation entry is NA.
How should the planning assumptions be read?
The registry’s planning assumptions describe an illustration, not the trial’s own design justification. They specify 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 stated fraction of the observed effect. For time-to-event tables, the assumptions refer to a total number of events using the Schoenfeld approach; other tables use participants per arm. These assumptions do not match the registered single-group design as a direct comparison, and the record does not report that they guided this study’s enrollment.
Any sample-size values generated under such assumptions are planning illustrations, not observed power and not evidence that the study had a particular chance of detecting an effect. The registry record does not report a study-specific sample-size justification, a target based on a stated clinical effect, or a statistical analysis plan. It also does not provide a concrete sample-size estimate in the recorded details.
Why do missing outcome data matter?
In general, an endpoint estimate depends on knowing which participants were eligible for the analysis, what endpoint information was observed, and how incomplete follow-up was treated. Missingness can affect both the estimate and its uncertainty, particularly when the reason for missing information is related to participants’ outcomes. No method can be assessed without the relevant data and analysis description. Here, the registry explicitly marks primary endpoint data as not collected, so there is no basis for evaluating a missing-data adjustment or a sensitivity analysis.
These are general statistical principles. They clarify what the intended method would ordinarily convey, but they do not fill gaps in the registry record or produce an outcome estimate.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists one outcome measure and no analyses. It identifies progression-free survival at 12 months as the primary endpoint, but does not report endpoint data or an analysis result. The number of listed outcomes alone does not establish how many statistical tests were planned, whether secondary outcomes were assessed, or how the study handled multiple comparisons. A multiplicity strategy is not reported in the registry record.
The primary endpoint population is described as data not collected. The record does not give an analysis population, a participant flow, follow-up completeness, or reasons for missing endpoint information. It also does not describe imputation, censoring, sensitivity analyses, or any other missing-data procedure. These methods cannot be assumed from the endpoint label or the planned use of an exact confidence interval.
In general, multiplicity refers to the interpretive problem that can arise when several hypotheses or outcomes are tested. Missing-data methods address how incomplete measurements are handled; analysis-population definitions specify which participants contribute to an estimate. They are separate issues, and neither can be evaluated without a recorded plan and the relevant observations.
The adverse-event entries list zero at risk and zero affected for both serious and other events, as well as zero deaths. With no participants listed as at risk, those entries do not provide a basis for assessing event frequency. The registry does not report further details about safety-analysis populations or missing safety information.
11. What the Result Does, and Does Not, Mean
The record establishes that a Phase II, single-group study of atezolizumab was registered for relapsed or refractory Hodgkin lymphoma, with progression-free survival at 12 months named as the primary endpoint. It also records a treatment schedule, an enrollment of 3, and a terminated status. These registry entries describe the study; they do not constitute an outcome result.
The primary endpoint data were not collected, according to the registry record. There is no reported proportion, exact 95% confidence interval, or analysis from which to assess progression-free survival. The absence of an estimate must not be read as evidence of benefit, lack of benefit, or a particular level of uncertainty. The record does not report why the endpoint data were not collected.
The single-group design also limits what a result could establish. In general, outcomes in one group do not provide a direct comparison with an alternative treatment. Here, even a descriptive outcome estimate is unavailable, so there is no basis for judging the endpoint against a comparator or a benchmark.
The adverse-event entries cannot be interpreted as evidence of no events because the record lists zero participants at risk. Nor does the terminated status explain why the study ended. The reason for termination is not reported in the registry record.
In general, a study’s planned question, its observed data, and the conclusions supported by those data are distinct. This record gives the question and some design details, but not the primary outcome observations needed to answer it. It should therefore be read as a limited account of a trial record rather than as evidence for a clinical conclusion.
12. Why This Trial Matters Statistically
This study illustrates why a registered endpoint and an observed result are not interchangeable. The registry specifies progression-free survival at 12 months and an intended reporting approach using a proportion with an exact 95% confidence interval. Yet it states that primary endpoint data were not collected and lists no analyses. A statistical method can describe how data would be analyzed, but it cannot yield an estimate without endpoint observations.
The single-group design is also important. In general, a single-group study can provide a descriptive account of participants receiving an intervention, while a comparative claim requires a suitable contrast. The registry contains no comparator group, so it does not set up a like-for-like treatment comparison. This design feature remains relevant regardless of the endpoint chosen.
The adverse-event entries provide another lesson in denominator awareness. Counts of zero affected participants do not establish absence of events when the record also lists zero at risk. Interpreting event counts requires knowing which participants were observed and eligible to contribute to the safety summary.
Finally, the record’s terminated status and small listed enrollment do not explain the missing endpoint data. The reason for termination and the study-specific enrollment rationale are not reported. Statistical interpretation should distinguish documented details from plausible but unverified explanations. In this case, the most defensible conclusion is that the registry records an intended clinical question but no collected primary endpoint result with which to answer it.
13. 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 is that the registry marks the primary endpoint population as data not collected. No progression-free survival proportion, confidence interval, or analysis is reported. This prevents assessment of the endpoint itself and its precision.
The listed enrollment is 3, and the design is single-group. With no comparator, the registered design does not support a direct comparison with another treatment. The record does not report whether enrollment met a target or provide a study-specific sample-size justification, so the enrollment cannot be assessed against a documented planning rationale.
The study is marked terminated, but no reason for termination is reported. The record also does not describe participant flow, treatment exposure, follow-up duration, endpoint assessment rules, censoring, analysis populations, or missing-data methods. These details are relevant to interpreting a progression-free survival endpoint, and their absence means that the intended analysis cannot be independently evaluated.
The safety summary lists zero participants at risk and zero affected for both serious and other events, with zero deaths. Those entries do not establish an absence of events among treated participants because no participants are listed as at risk in the summary. The record provides no further safety-analysis context.
These limitations are specific to the information recorded for this study. They do not establish why data are absent or why the study was terminated; those explanations are not reported in the registry record.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Study of Atezolizumab in Relapsed or Refractory Hodgkin Lymphoma — NCT03120676.
- Registry record last updated 2019-11-14; status last verified 2018-11.
- Results first posted 2019-11-14.
- 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.
17. Record Summary
The registry record describes a Phase II, single-group study of atezolizumab for relapsed or refractory Hodgkin lymphoma. Atezolizumab was to be administered at 1200 mg intravenously every 3 weeks, for up to 36 cycles. The primary endpoint was progression-free survival at 12 months, intended to be expressed as a proportion with an exact 95% confidence interval.
The listed enrollment is 3, and the study status is terminated. The record lists one outcome measure and no analyses, and states that primary endpoint data were not collected. It gives a start date of 2017-04-14 and a primary completion date of 2018-11-12. No comparator group is recorded. The reason for termination, study-specific sample-size rationale, and primary endpoint result are not reported in the registry record.