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
GALAHAD was a phase 2, single-group study evaluating niraparib in men with metastatic castration-resistant prostate cancer and DNA-repair anomalies. The registry reports an enrollment of 289 participants, one intervention, and a binary primary endpoint measuring objective response rate in participants with measurable metastatic castration-resistant prostate cancer and a BRCA mutation.
| Feature | GALAHAD |
|---|---|
| Trial name | GALAHAD |
| Brief title | An Efficacy and Safety Study of Niraparib in Men With Metastatic Castration-Resistant Prostate Cancer and DNA-Repair Anomalies |
| Phase | Phase 2 |
| Status | Completed |
| Study start | 2016-08-31 |
| Primary completion | 2021-01-26 |
| Condition | Prostatic Neoplasms |
| Design model | Single-group |
| Allocation | NA |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 289 |
| Intervention | Niraparib |
| Lead sponsor | Janssen Research & Development, LLC |
| Sponsor type | Industry |
| Results posted | Yes |
| Outcome measures posted | 11 |
2. Clinical Question
The registered study evaluates the efficacy and safety of niraparib in men with metastatic castration-resistant prostate cancer and DNA-repair anomalies. The primary efficacy question is expressed through objective response rate among the specifically defined population with measurable metastatic castration-resistant prostate cancer and a BRCA mutation.
Population
Men with metastatic castration-resistant prostate cancer and DNA-repair anomalies. The registered primary endpoint further specifies participants with measurable mCRPC and a BRCA mutation.
Intervention
Niraparib.
Comparator
There is no separate comparator arm. The registered design model is single-group.
Primary question
What objective response rate is observed under the registered response definition in participants with measurable mCRPC and a BRCA mutation?
3. Trial Design
GALAHAD uses a single-group phase 2 design with no masking and one intervention, niraparib. This is fundamentally different from a parallel-group randomized trial: there is no randomized control group against which a between-arm treatment effect can be estimated.
Why the single-group structure matters
A single-group study can estimate the proportion of participants who meet a prespecified response definition. It cannot, by itself, provide a randomized estimate of the difference between niraparib and another treatment because no concurrent randomized comparator is part of the registered design.
4. Endpoints
Primary Endpoint
| Endpoint | Time frame | Type | Registered definition |
|---|---|---|---|
| Objective Response Rate (ORR) for Participants With Measurable Metastatic Castration-resistant Prostate Cancer (mCRPC) and Breast Cancer Gene (BRCA) Mutation | Up to 52 months | Binary | ORR defined as percentage of participants with BRCA DNA-repair anomalies and measurable disease whose best response is either complete response (CR) or partial response (PR) per Response Evaluation Criteria in Solid Tumors (RECIST 1.1) and with no evidence of bone progression per Prostate Cancer Working Group 3 (PCWG3) criteria. |
The registry therefore defines the primary outcome as a percentage, with each participant ultimately classified according to whether the specified response criteria are met. The definition combines tumor-response assessment under RECIST 1.1 with the requirement that there be no evidence of bone progression according to PCWG3 criteria.
Understanding the endpoint definition
Complete response
A participant qualifies through the registry's stated CR category under the RECIST 1.1 response framework, subject to the additional bone-progression requirement.
Partial response
A participant qualifies through the registry's stated PR category under the RECIST 1.1 response framework, subject to the additional bone-progression requirement.
Bone progression
The registered endpoint specifically requires no evidence of bone progression according to PCWG3 criteria.
Analysis unit
The primary result is naturally expressed as a proportion or percentage of participants meeting the complete-response or partial-response definition.
5. Results Status and Planned Analysis
The primary endpoint is a binary objective response outcome assessed over up to 52 months. For a single-group binary endpoint, the fundamental descriptive analysis is the proportion of participants meeting the prespecified response definition. A confidence interval for that proportion would quantify statistical uncertainty around the observed response rate.
The numerator is the number of participants whose best response satisfies the registered CR-or-PR definition and whose disease has no evidence of bone progression under PCWG3. The denominator must follow the prespecified analysis population for the endpoint.
Depending on the prespecified statistical analysis plan, a single-group binary endpoint may also be evaluated against a reference response rate using an exact or other appropriate binomial hypothesis test. However, the registry-reported GALAHAD trial data do not specify a reference response rate, hypothesis-test procedure, confidence-interval procedure, or formal statistical analysis. Those elements therefore cannot be presented as registered GALAHAD methods.
6. Statistical Methodology
Binary outcome analysis
The primary endpoint is explicitly classified as binary. Each participant is classified according to whether the prespecified response criterion is met. The resulting endpoint is summarized as a percentage.
The observed response proportion estimates the probability of meeting the endpoint within the defined study population and analysis framework. It is a within-study estimate, not automatically a comparative treatment effect.
Confidence intervals for a proportion
A confidence interval around a binary response rate describes statistical uncertainty associated with estimating an underlying response probability from a finite number of participants. Exact binomial methods are commonly useful when the endpoint is a proportion, particularly when response counts are not large or when proportions approach the boundaries of 0 or 1.
For GALAHAD, however, the ClinicalTrials.gov record does not identify the confidence-interval method used for the posted results. A specific interval calculation should therefore not be attributed to the trial.
Hypothesis testing in a single-group study
A one-sample binary endpoint can be compared with a prespecified reference response probability. Conceptually, the null hypothesis can be written as:
where p is the response probability under the registered endpoint definition and p0 is a prespecified reference value.
Such a comparison is meaningful only when the reference value and testing framework are specified in advance. The registry-reported GALAHAD data do not provide those values or procedures, so no hypothesis-test conclusion is reported here.
Response assessment
The registered endpoint uses RECIST 1.1 for complete and partial response classification and requires no evidence of bone progression under PCWG3 criteria. Statistically, this means the binary endpoint is not simply an unconstrained tumor-shrinkage measure: its classification incorporates both the radiologic response framework and the specified bone-progression condition.
Analysis time frame
The primary endpoint has a registered time frame of up to 52 months. A time frame is important because response classification depends on the observation period defined by the endpoint. The ClinicalTrials.gov record does not provide a formal landmark analysis, time-to-response analysis, or duration-of-response statistical analysis for the primary endpoint.
7. Statistical Methods Explained
What does a binary primary endpoint mean?
A binary endpoint places each participant into one of two outcome categories for the specified analysis: meeting the response definition or not meeting it. The primary summary is therefore a proportion or percentage rather than a mean, median, or hazard ratio.
Why is ORR expressed as a percentage?
The registered endpoint is defined as the percentage of participants whose best response is complete response or partial response under RECIST 1.1 and who have no evidence of bone progression under PCWG3. A percentage makes the frequency of qualifying responses directly interpretable within the defined study population.
Why does the RECIST 1.1 definition matter statistically?
A statistical analysis is only as reproducible as its endpoint definition. RECIST 1.1 supplies the response-classification framework referenced by the registry, while the additional PCWG3 requirement further determines which participants qualify for the primary endpoint.
Why is the PCWG3 condition part of the endpoint?
The registered definition does not count a RECIST-based complete or partial response alone. It also requires no evidence of bone progression under PCWG3 criteria. This creates a composite classification rule: both components of the definition matter when determining whether a participant is a responder.
Why is a confidence interval useful for ORR?
An observed response percentage is estimated from a finite sample. A confidence interval communicates uncertainty around the underlying response probability. It does not describe the probability that an individual participant will respond, nor does it establish a causal treatment effect.
Why is a single-group response rate not automatically a treatment effect?
Without a randomized comparator, the study does not directly estimate the difference between niraparib and another treatment under otherwise comparable conditions. A response proportion can describe what occurred in the study group, while a comparative causal interpretation requires an appropriate reference and assumptions that go beyond the observed proportion itself.
Why can a p-value not be reported here?
A p-value requires a specified null hypothesis and statistical testing procedure. The registry-reported GALAHAD data contain no formal statistical analyses and do not identify a reference response rate or formal hypothesis-testing method. Reporting a p-value would therefore require information not contained in the ClinicalTrials.gov record.
8. Analysis Population and Interpretation
The study enrolled 289 participants overall. The primary endpoint, however, is specifically defined for participants with BRCA DNA-repair anomalies and measurable disease. This distinction is statistically important: the overall enrollment number should not automatically be treated as the denominator for the primary endpoint without confirmation that all enrolled participants belong to the endpoint's defined analysis population.
| Feature | What the ClinicalTrials.gov record establishes | Statistical implication |
|---|---|---|
| Overall enrollment | 289 participants | Describes total study enrollment. |
| Primary endpoint population | Participants with BRCA DNA-repair anomalies and measurable disease | The endpoint definition is narrower than the broad enrollment description. |
| Endpoint type | Binary | Primary summary is a response proportion. |
| Time frame | Up to 52 months | Response classification is evaluated within the registered time frame. |
| Design | Single-group | No randomized between-arm treatment comparison is available. |
| Formal statistical analyses | None in the posted results | No trial-specific estimate, CI, or p-value is reproduced here. |
The distinction between enrollment and the primary endpoint population is central. A study can enroll 289 participants while a particular endpoint applies only to a defined subset. The denominator used for an ORR estimate must follow the endpoint's prespecified analysis population rather than being assumed from total enrollment.
9. Safety
The ClinicalTrials.gov record reports serious adverse events by arm as 134/289 for niraparib. Because GALAHAD is a single-group study, there is no separate comparator-arm serious-adverse-event count in the ClinicalTrials.gov record.
| Safety measure | Niraparib |
|---|---|
| Serious adverse events, affected / at risk | 134 / 289 |
The affected/at-risk representation is useful because it identifies both the number of participants affected and the population against which that count is expressed. The ClinicalTrials.gov record does not provide additional safety estimates, confidence intervals, event-specific breakdowns, severity categories, treatment-relatedness assessments, or comparative safety statistics.
10. Study Timeline
| Milestone | Date |
|---|---|
| Study start | 2016-08-31 |
| Primary completion | 2021-01-26 |
| Study status | Completed |
The study therefore spans from its registered start date of 2016-08-31 through its registered primary completion date of 2021-01-26. The primary endpoint itself has a separate registered assessment time frame of up to 52 months.
11. What a Complete ORR Analysis Would Contain
For an educational statistical workflow, a complete analysis of this primary endpoint would begin by establishing the prespecified analysis population and classifying each participant according to the registered response definition.
The important point is that the statistical calculation comes after the clinical endpoint has been operationalized. A response rate calculated from an incorrect denominator or without the registered bone-progression condition would not represent the same endpoint.
The ClinicalTrials.gov record does not provide the responder count or endpoint-specific denominator, so the numerical ORR cannot be calculated from the ClinicalTrials.gov record.
12. Limitations
- Single-group design: there is no randomized comparator arm in the registered design, so the primary endpoint does not directly estimate a randomized treatment difference.
- Endpoint-specific population: the primary endpoint is restricted to participants with BRCA DNA-repair anomalies and measurable disease, whereas total enrollment is 289. The endpoint denominator should not be inferred from enrollment alone.
- No formal statistical analyses reported: no formal statistical analyses were posted. Consequently, the trial-specific response estimate, confidence interval, and p-value are not reproduced here.
- Reference value not reported: the data do not identify a null or historical response rate against which the single-group ORR was formally tested.
- Response definition: the primary endpoint requires both the RECIST 1.1 response classification and no evidence of bone progression according to PCWG3 criteria. A generic ORR calculation would not necessarily reproduce this endpoint.
- Time frame: the registered primary endpoint is assessed up to 52 months. A single response percentage compresses the response information within that defined period.
- Comparative safety: the ClinicalTrials.gov record contains only the niraparib affected/at-risk count of 134/289. There is no comparator-arm rate with which to calculate a between-group safety difference.
- External validity: interpretation should remain tied to the registered study population and endpoint definition rather than assuming that the observed response proportion applies to patients outside those criteria.
13. Why This Trial Matters Statistically
GALAHAD is a useful teaching example because its primary endpoint and study architecture illustrate an important distinction in clinical-trial statistics: estimating a response proportion is not the same statistical problem as estimating a randomized treatment effect.
| Concept | How it appears in GALAHAD |
|---|---|
| Single-group design | The registry identifies the design model as SINGLE_GROUP with one arm. |
| Binary endpoint | The registered primary endpoint is classified as binary. |
| Objective response rate | The primary endpoint is a percentage of participants meeting the CR-or-PR response definition. |
| RECIST 1.1 | Provides the response framework specified in the registered endpoint definition. |
| PCWG3 | Adds the registered requirement for no evidence of bone progression. |
| Confidence intervals | Would quantify uncertainty around a single-group response proportion if a confidence-interval method were specified. |
| One-sample hypothesis testing | Could compare ORR with a prespecified reference response rate, but no such reference value is reported in the ClinicalTrials.gov record. |
| Analysis population | The primary endpoint specifies BRCA DNA-repair anomalies and measurable disease, illustrating why total enrollment is not necessarily the endpoint denominator. |
| Safety analysis | Serious adverse events are reported as affected/at-risk for the single niraparib group. |
| Results transparency | Results are posted, but the ClinicalTrials.gov record contains no formal statistical analyses, limiting reproducible statistical interpretation. |
14. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The registered primary outcome is a binary response proportion evaluated within a defined population over up to 52 months. Because the study is single-group, the primary endpoint is descriptive of the study population unless an independently specified reference value and formal comparison are used.
Clinical interpretation
The endpoint is designed to identify participants whose best response meets the registered CR-or-PR definition under RECIST 1.1 while also having no evidence of bone progression according to PCWG3. That clinical definition determines who counts as a responder.
A response percentage can be clinically informative without being a randomized estimate of treatment benefit. In GALAHAD, the single-group design means that the observed response rate should be interpreted in the context of the study population, endpoint definition, and any prespecified reference framework rather than as a direct comparison with an unobserved control group.
15. Statistical Concepts in This Trial
Learn more about the methods used in this trial:
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: GALAHAD (NCT02854436).
- Linked publication: PubMed record for PMID 37404070.
- Linked publication: PubMed record for PMID 35131040.
Continue through the Clinical Biostats statistical pathway
Use the related tutorials and calculators to explore binary endpoints, single-group clinical-trial designs, confidence intervals for proportions, and response-rate analysis.
18. Record Summary
GALAHAD is a phase 2, completed, single-group study of niraparib in men with metastatic castration-resistant prostate cancer and DNA-repair anomalies. Its registered primary endpoint is a binary objective response rate assessed for participants with BRCA DNA-repair anomalies and measurable disease over up to 52 months, with response defined through RECIST 1.1 and the additional requirement of no evidence of bone progression under PCWG3 criteria.
The statistical structure is therefore centered on estimation of a single-group response proportion, rather than a randomized treatment-versus-control comparison. The ClinicalTrials.gov record reports an overall enrollment of 289 and serious adverse events affecting 134 of 289 participants in the niraparib group.