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 Testing the Combination of Anti-cancer Drugs Atezolizumab and Tiragolumab in People With Advanced Stage Rare Cancers, RARE3 Trial (NCT05715281), a na phase 2 study. The primary outcome is Percent Change of Active Cluster of Differentiation 8 (CD8+) T Cells Infiltrating the Tumor at Baseline and After Treatment With Atezolizumab and Tiragolumab.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT05715281 |
| Phase | Phase 2 |
| Condition | Advanced Rare Malignant Solid Neoplasm, Rare Malignant Solid Neoplasm, Refractory Rare Malignant Solid Neoplasm |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 12 |
| Arms in the primary outcome | n/a |
| Lead sponsor | National Cancer Institute (NCI) |
| Results posted | Yes |
2. Clinical Question
Population
People with advanced rare malignant solid neoplasms, including rare solid tumors described as advanced or refractory. The registry record lists an enrollment of 12. The primary endpoint analysis population was restricted to participants with an evaluable baseline and after-treatment tumor biopsy pair; that population is reported as 4/12 participants.
Intervention
Atezolizumab and tiragolumab are given together. The treatment description specifies intravenous administration on day 1 of each cycle, with cycles repeating every 21 days while there is no disease progression or unacceptable toxicity. The study also collects biopsy and blood samples to examine immune response and treatment-related biological changes.
Comparator
There is no separate comparator group in this single-group study. Allocation is listed as NA and masking as none. Consequently, the design does not provide a randomized, like-for-like comparison against another treatment or a control condition.
Primary question
The primary endpoint asks how the density of activated CD8+ T cells inside tumor tissue changes from baseline to after treatment with the combination. Activated cells are identified using specified markers in multiplex immunofluorescence assays. The registry record gives the measurement schedule and analysis population, but the endpoint estimate and its full range are not reported in the registry record.
This question is biological rather than a direct measure of tumor shrinkage or time until progression. The record also lists objective tumor response rate and progression-free survival among secondary outcomes. Those outcomes address different aspects of disease course; they do not substitute for a reported primary biomarker result. The trial therefore asks whether an immune-cell feature in available tumor samples changes following treatment, while the registry information does not establish the size or direction of that change.
3. Trial Design
RARE3 is a phase II trial sponsored by the National Cancer Institute (NCI). Its registry title describes testing the combination of atezolizumab and tiragolumab in people with advanced-stage rare cancers. The design is single group, with allocation listed as NA and masking listed as none. There is no randomized control arm, so outcomes cannot be interpreted as a direct comparison between the combination and another assigned treatment.
The registry record lists enrollment as 12 and status as active, not recruiting. The recorded start date is 2023-09-26, and the primary completion date is 2025-10-14. These dates describe the registry record; they do not themselves establish that all endpoint data have been collected or reported. The trial is classified as phase II.
Participants in the treatment group receive atezolizumab intravenously over 60 minutes and tiragolumab intravenously over 30-90 minutes on day 1 of each cycle. Cycles repeat every 21 days in the absence of disease progression or unacceptable toxicity. The intervention list also includes biopsy, biospecimen collection, computed tomography, and echocardiography. The treatment description states that an echocardiogram is performed at baseline. It also describes biopsies, but the available description is incomplete and does not give a full schedule for all procedures.
The primary endpoint uses tumor tissue collected at baseline within 28 days before treatment and at the beginning of cycle 3. The stated average interval is 45 days. Only participants whose baseline and after-treatment biopsy pair was evaluable using multiplex immunofluorescent microscopy were included in that analysis population, reported as 4/12 participants. This selection makes the primary analysis depend on having usable paired tissue, not simply on enrollment in the treatment group.
The registry lists five outcome measures and three named secondary outcomes: objective tumor response rate, progression-free survival, and a clinically promising change in CD8+ T-cell infiltration. It does not report a randomized comparison or an analysis entry for the outcomes. Statistical procedures beyond the endpoint description and planning assumptions 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 | Treatment (Atezolizumab, Tiragolumab) |
|---|---|
| STARTED | 12 |
| Completed Follow-up Period | 5 |
| COMPLETED | 9 |
| NOT COMPLETED | 3 |
Reasons for not completing this period:
| Reason | Treatment (Atezolizumab, Tiragolumab) |
|---|---|
| Hospitalized and unable to start the study | 1 |
| Initiation of another treatment | 1 |
| Withdrawal by Subject | 1 |
Analysis population for the primary outcome, as defined in the registry: 4/12 participants were analyzed because the analysis population includes all participants with a baseline and after treatment tumor biopsy pair that was evaluable using multiplex immunofluorescent microscopy analysis.
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 | Treatment (Atezolizumab, Tiragolumab) |
|---|---|---|
| Participants analysed | 12 | |
| Age, Categorical: <=18 years | Participants | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 10 |
| Age, Categorical: >=65 years | Participants | 2 |
| Age, Continuous | years | 55.08 (SD 13.85) |
| Sex: Female, Male: Female | Participants | 9 |
| Sex: Female, Male: Male | Participants | 3 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 2 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 10 |
| 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 | 2 |
| Race (NIH/OMB): White | Participants | 10 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Region of Enrollment: United States | participants | 12 |
6. Primary Endpoint
The registry lists the primary outcome as Percent Change of Active Cluster of Differentiation 8 (CD8+) T Cells Infiltrating the Tumor at Baseline and After Treatment With Atezolizumab and Tiragolumab, measured over At baseline within 28 days before start of treatment and beginning cycle 3 (each cycle is 21 days), an average of 45 days.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | percent change |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the percent change in active CD8+ T cells infiltrating the tumor at baseline and after treatment with atezolizumab and tiragolumab. The assay identifies activated CD8+ T cells through expression of T-cell receptor activation, measured by Zeta chain phosphorylation or phosphorylated Zap70 at tyrosine position 493. CD8+ cells inside the tumor section that test positive for these markers are quantified, and the change in active-cell density after treatment compared with baseline is to be reported.
The measurement is taken at baseline within 28 days before treatment and at the beginning of cycle 3, with an average interval of 45 days. The planned summary is the median percent change with the full range. The median describes the middle of the observed distribution and is less affected by extreme observations than the arithmetic mean. The full range conveys the spread from the lowest to highest observed values, although it does not show how values are distributed between those ends.
A tissue biomarker can help characterize a biological response within the tumor. It is not itself a direct measure of tumor shrinkage, symptom improvement, or survival. The registry record specifies the assay and summary approach but does not report the median change or full range. It therefore does not show whether the measured density increased, decreased, or had little change in the evaluable biopsies.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary outcome is defined in terms of change in activated CD8+ T-cell density within tumor tissue. The registry record gives an analysis population of 4/12 participants, meaning that the stated primary analysis includes only participants whose baseline and after-treatment biopsy pair was evaluable by multiplex immunofluorescent microscopy. The record does not report the median percent change or its full range. Without those results, the direction, magnitude, and variability of the observed biomarker change cannot be described.
The analysis population is an important part of the interpretation. Participants who lack an evaluable paired biopsy are not represented in this endpoint analysis, even though they are included in the trial's enrollment figure. Reasons for the absence of usable paired tissue, and whether those reasons are associated with participant characteristics or outcomes, are not reported in the registry record. The analyzed group may therefore differ from the enrolled group in ways that cannot be assessed from these registry entries.
The design is single group and unmasked, with no randomized comparator. Any before-and-after change in the biomarker would describe change during the treatment period among the analyzed participants. On its own, it would not separate a treatment-related change from temporal variation, measurement variability, or other influences. The design also does not establish how the biomarker change compares with what would have occurred under a different treatment or no study treatment.
The registry lists other outcomes, including objective tumor response rate and progression-free survival, but their results are not reported in the registry record. These outcomes would address clinical disease measures that differ from the tissue biomarker. The relationship between the biomarker and those outcomes is also not reported. The stated aim of studying immune response and future response patterns is a research rationale, not evidence that the biomarker predicts clinical benefit in this trial.
Safety entries describe participants with serious and other adverse events, as well as selected serious-event terms. They are descriptive counts and percentages, not a comparison of risks across treatment groups. The trial has no control group, and the registry record does not provide a causal comparison. The listed events should therefore be read as events recorded during the study, not as estimates of excess risk attributable to the combination.
Planning illustrations are described as based on the observed effect and use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Such illustrations concern hypothetical future planning; they are not observed power and are not the trial's own design justification. The actual numerical planning outputs and the design rationale are not reported in the registry record.
9. Statistical Methods Explained
The primary outcome is a paired tissue biomarker: active CD8+ T-cell density is assessed before treatment and after treatment, and the change is expressed as a percent change. The analysis is restricted to participants with an evaluable biopsy at both time points. The registry describes multiplex immunofluorescence as the assay method and specifies marker expression used to identify activated cells. The result is intended to summarize a biological feature within the tumor rather than a clinical endpoint such as tumor response or progression-free survival.
What does the median describe?
As a general statistical concept, the median is the middle value after observations are ordered. It can be a useful summary when measurements are skewed or when some values are unusually large or small, because extreme observations do not pull it in the same way they can pull an arithmetic mean. The registry specifies the median as the summary for percent change. The actual median is not reported in the registry record, so its value and direction cannot be interpreted.
What does the full range add?
As a general statistical concept, the full range gives the lowest and highest observed values. It communicates how far apart the extremes are, but does not show whether most values are near the middle or clustered near one end. The registry specifies a full range with the median. No endpoints of that range are reported in the registry record. With a small evaluable analysis population, individual observations can have substantial influence on the apparent range and the median; the registry’s stated analysis population is 4/12 participants.
Why does paired measurement matter?
As a general statistical concept, paired measurements compare each participant's later measurement with that same participant's earlier measurement. This can account for stable differences between participants when describing within-person change. However, pairing requires usable observations at both time points. In this trial, the primary endpoint population is defined by an evaluable baseline and after-treatment biopsy pair. The record does not describe how missing or unevaluable samples were handled beyond this population definition, nor does it report sensitivity analyses addressing missing tissue.
What does an assay definition establish?
The assay definition states how the biomarker is identified and counted. Here, activated CD8+ T cells are defined using T-cell receptor activation markers: Zeta chain phosphorylation or phosphorylated Zap70 at tyrosine position 493. CD8+ cells within the tumor section that are positive for these markers are quantified. As a general statistical concept, a clear operational definition supports consistency in measurement, but it does not by itself establish assay precision, reproducibility, or freedom from measurement error. Those characteristics are not reported in the registry record.
How should a confidence interval be understood?
As a general statistical concept, a confidence interval expresses uncertainty around an estimate under a specified statistical procedure. A narrow interval generally indicates greater precision than a wide interval, while neither interval width nor the point estimate alone establishes clinical importance. The registry entries do not report a confidence interval for the primary endpoint, nor do they specify an interval method. The endpoint description instead identifies the median and full range as the planned summary. Statistical uncertainty for the primary estimate therefore cannot be evaluated from the reported information.
Why does the single-group design matter?
As a general statistical concept, a comparison between measurements before and after treatment does not have the same interpretation as a randomized comparison between treatment groups. A within-person change may be described, but changes over time and other influences cannot be ruled out by a single-group design alone. This trial has one treatment group, allocation listed as NA, and masking listed as none. The registry record does not report an adjustment strategy or a causal analysis that would address the absence of a comparator.
What are the planning assumptions?
The registry describes sample-size planning illustrations as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. As a general statistical concept, a planning calculation estimates the sample size needed under chosen assumptions; it is not a measurement of the evidence in the current trial. The illustrations are explicitly not the trial's own design, and they should not be called observed power. Numerical planning outputs and the trial's sample-size justification are not reported in the registry record.
For time-to-event outcomes, the registry describes planning quantities as total events using the Schoenfeld approach; other planning quantities are expressed as participants per arm. These are conventions for interpreting the planning illustrations, not observed trial outcomes. The registry record does not provide outcome analysis procedures, interim-analysis details, or a full statistical analysis plan.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists five outcome measures, with the primary endpoint focused on change in active CD8+ T-cell density. Named secondary outcomes include objective tumor response rate, progression-free survival, and a clinically promising change in CD8+ T-cell infiltration. When multiple outcomes are assessed, statistical interpretation can be affected by multiplicity: the more questions examined, the greater the chance that an apparently notable finding arises by chance if no adjustment or hierarchy is specified. The registry entries do not state whether multiplicity adjustments, endpoint hierarchies, or confirmatory testing rules were planned. Those details are not reported in the registry record.
The primary analysis population consists of participants with an evaluable baseline and after-treatment biopsy pair, reported as 4/12 participants. This makes the availability and evaluability of tissue central to the endpoint. The record does not describe how missing biopsies, unusable specimens, or incomplete measurements were handled beyond defining the analysis population. It does not report imputation, sensitivity analyses, or reasons for missing or unevaluable tissue. These methods are not reported in the registry record.
The registry also does not report the endpoint estimate, its full range, or an uncertainty interval. Without these results, readers cannot assess the observed variability or the precision of the primary estimate. The trial is single group, so multiplicity considerations do not create a missing comparator; they concern how multiple outcomes are interpreted. A statistical analysis plan, including any interim analyses or multiplicity strategy, is not reported in the registry record.
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 |
|---|---|---|---|
| Treatment (Atezolizumab, Tiragolumab) | 5 | 12 | 41.7% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Treatment (Atezolizumab, Tiragolumab) | 0 of 12 |
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 |
|---|---|
| Treatment (Atezolizumab, Tiragolumab) | 11 of 12 |
Most frequent serious adverse event terms
| Event term | Organ system | Treatment (Atezolizumab, Tiragolumab) |
|---|---|---|
| CPK increased | Investigations | 1 of 12 |
| Diarrhea | Gastrointestinal disorders | 1 of 12 |
| Fever | General disorders | 1 of 12 |
| Guillain-Barre syndrome | Nervous system disorders | 1 of 12 |
| Headache | Nervous system disorders | 1 of 12 |
| Hypertension | Vascular disorders | 1 of 12 |
| Myasthenia gravis | Nervous system disorders | 1 of 12 |
| Myositis | Musculoskeletal and connective tissue disorders | 1 of 12 |
| Pain | General disorders | 1 of 12 |
| Paresthesia | Nervous system disorders | 1 of 12 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Treatment (Atezolizumab, Tiragolumab) |
|---|---|---|
| Hyperglycemia | Metabolism and nutrition disorders | 8 of 12 |
| Aspartate aminotransferase increased | Investigations | 7 of 12 |
| Fatigue | General disorders | 7 of 12 |
| Alkaline phosphatase increased | Investigations | 5 of 12 |
| Anemia | Blood and lymphatic system disorders | 5 of 12 |
| Hyponatremia | Metabolism and nutrition disorders | 5 of 12 |
| Nausea | Gastrointestinal disorders | 5 of 12 |
| Anorexia | Metabolism and nutrition disorders | 4 of 12 |
| Diarrhea | Gastrointestinal disorders | 4 of 12 |
| Myalgia | Musculoskeletal and connective tissue disorders | 4 of 12 |
Serious events with the highest arm-level rate: cpk increased (8.3%), diarrhea (8.3%), fever (8.3%), guillain-barre syndrome (8.3%), headache (8.3%).
For the treatment group, the registry reports 5 participants with serious adverse events among 12 at risk, and 11 participants with other adverse events among 12 at risk. Deaths are recorded as 0. These are descriptive safety figures for the study group. They are not comparisons with an untreated group or another treatment group, since the trial has no randomized comparator.
The registry lists CPK increased, diarrhea, fever, Guillain-Barre syndrome, headache, and hypertension among the serious adverse-event terms, each with a reported percentage of 8.3%. These percentages are reproduced as recorded. The listed terms should not be treated as mutually exclusive categories or added together: a participant may have more than one event, and the registry information does not give a basis for summing terms. It also does not establish from these figures alone whether an event was caused by treatment.
As a general statistical point, safety counts describe what was recorded in the observed study group, while causal attribution and comparative risk require additional information and an appropriate comparison. The registry record does not provide detailed event timing, duration, severity breakdowns, or a causal analysis in the registry record. The counts and term percentages therefore describe reported safety observations without establishing the cause or comparative frequency of events.
12. What the Result Does, and Does Not, Mean
The trial's primary question concerns a tissue-based measure of immune activity: the percent change in active CD8+ T-cell density inside the tumor after treatment with atezolizumab and tiragolumab. The registry defines how these cells are identified, the biopsy time points, and the planned summary as a median with full range. It reports that 4/12 participants had an evaluable paired biopsy for the primary analysis.
The endpoint result itself is not reported in the registry record. There is therefore no posted median or range in the registry record from which to describe the observed change. It is not possible to say whether the biomarker increased or decreased, how large any change was, or how variable the measurements were. The analysis population alone does not establish a treatment effect.
Even if a before-and-after biomarker change were available, a single-group design would show change during treatment in the analyzed participants, not a randomized treatment contrast. It would not by itself show what would have happened under another treatment or without study treatment. It also would not establish that a biomarker change translates into tumor response, longer progression-free survival, or another clinical benefit.
The registry identifies objective tumor response rate and progression-free survival as secondary outcomes, but their numerical results are not reported in the registry record. The relationship between those outcomes and the primary biomarker is also not reported. The study's stated interest in learning about immune response and which future patients may respond is a scientific aim; it is not evidence that the biomarker predicts response.
Safety information is likewise descriptive. The reported adverse-event counts and terms should not be interpreted as comparative risks or as proof that a particular event was caused by treatment. Overall, the record defines a biological endpoint and a limited analysis population, but does not report the primary result needed to judge the observed biomarker change.
13. Why This Trial Matters Statistically
This trial illustrates a common distinction in clinical research: a biological endpoint can be measured carefully without being interchangeable with a clinical outcome. The primary measure is activated CD8+ T-cell density in tumor tissue, quantified through marker expression in multiplex immunofluorescence assays. That can address whether an immune feature changes over the treatment period, but it answers a different question from whether tumors respond or progression-free survival changes.
The analysis population also matters. The primary endpoint is based on participants with an evaluable paired biopsy, reported as 4/12 participants, rather than on everyone enrolled. In tissue-based research, availability and evaluability can shape which observations contribute to the result. Interpretation therefore depends not only on the assay and summary measure, but also on how the analyzed participants relate to those enrolled. Reasons for the restricted analysis population and methods for handling missing samples are not reported in the registry record.
The design highlights a further statistical issue: measurement over time within a single group is not a randomized comparison. It can describe change in participants with paired measurements, but cannot isolate treatment effects from other explanations in the way a suitable controlled comparison can. This distinction is important when reading biomarker findings as evidence about a treatment.
Finally, the endpoint is summarized using a median and full range, which foregrounds both a central value and the extremes rather than an average alone. The registry record specifies that approach but does not report the actual summary. Consequently, the statistical design is identifiable, while the primary outcome's direction, magnitude, and precision remain unavailable in the registry information described.
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 main limitation for interpreting the primary endpoint is that its result is not reported in the registry record. Although the measure, assay, timing, and summary approach are specified, the median percent change and full range are absent. The available information therefore cannot establish the observed direction or size of biomarker change.
The primary analysis is limited to participants with an evaluable baseline and after-treatment biopsy pair, reported as 4/12 participants. The registry record does not explain why the remaining enrolled participants were not in this analysis population, or whether biopsy availability is related to participant or disease characteristics. It does not report a missing-data strategy or sensitivity analysis. This limits assessment of how representative the analyzed tissue measurements may be.
The single-group design has no randomized control arm, and masking is listed as none. Any reported within-participant change would not, by itself, distinguish treatment-associated change from other changes over time or measurement variation. The registry entries do not describe adjustment methods, assay reproducibility, or other procedures that might address these issues.
Several outcome and analysis details are also unavailable. The registry lists secondary outcomes, but numerical results for them are not reported in the registry record. A statistical analysis plan, multiplicity approach, interim-analysis details, and sample-size justification are not reported in the registry record. The planning assumptions described for illustrations should not be mistaken for evidence that the study was designed around those calculations. Safety information is descriptive and does not provide a randomized comparison or establish event causation.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Testing the Combination of Anti-cancer Drugs Atezolizumab and Tiragolumab in People With Advanced Stage Rare Cancers, RARE3 Trial — NCT05715281.
- Registry record last updated 2026-11-04; status last verified 2026-10.
- Results first posted 2026-10-14.
- 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.
18. Record Summary
RARE3 is a phase II, single-group study sponsored by the National Cancer Institute (NCI), evaluating atezolizumab plus tiragolumab in advanced rare solid tumors. The registry lists enrollment of 12, status as active, not recruiting, a start date of 2023-09-26, and primary completion date of 2025-10-14. Allocation is listed as NA and masking as none.
The primary endpoint is percent change in active CD8+ T-cell density from baseline to the beginning of cycle 3, summarized by median and full range. The stated average interval is 45 days. The analysis population is 4/12 participants with evaluable paired biopsies. The endpoint result is not reported in the registry record. Named secondary outcomes include objective tumor response rate and progression-free survival, whose results are also not reported in the registry record.
Safety entries include 5 participants with serious adverse events, 11 with other adverse events, and 0 deaths. The listed serious-event terms each have a reported percentage of 8.3%. These observations are descriptive and do not provide a comparison with another treatment group.