A Proof-of-principle Study of HBO-SRS for Brain Metastases: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Proof-of-principle Study of HBO-SRS for Brain Metastases (NCT01850563), a na na study. The primary outcome is The Average Time From HBO Chamber Exit to SRS beam-on.

2. Clinical Question

The study addresses a practical sequencing question: whether hyperbaric oxygen could be followed by SRS within a clinically constrained visit. The registry describes transferring the patient and placing a Gill-Thomas-Cosman head frame after HBO; its brief summary says those steps need to be completed within a 15-minute time frame. The primary endpoint definition uses a different stated criterion: feasibility is achieved if more than 50% of 20 evaluable patients start SRS within 30 minutes of leaving the HBO tank. The registry does not explain the relationship between these two timing descriptions.

The secondary objectives concern outcomes and quality of life. Listed outcomes include overall survival in years, time until local recurrence, distant intracranial recurrence, time between SRS and whole-brain radiation therapy, and lesions exhibiting radionecrosis. Quality-of-life and mini mental status questionnaires were also described as part of follow-up. The posted record does not report the primary timing result or numerical results for these secondary outcomes, so it cannot establish from the registry information whether the feasibility criterion was met.

3. Trial Design

This was a completed, single-group study of hyperbaric oxygen before stereotactic radiosurgery for brain metastases. The registry lists phase as NA, allocation as NA, and masking as none. The recorded enrollment was 22. The sponsor was Dartmouth-Hitchcock Medical Center. The recorded start date was 2013-07-01, and the primary completion date was 2019-10-03.

The intervention was HBO at 2.4 ATA for 30 minutes immediately before SRS. The study description focuses on whether the sequence could be carried out within the time constraints of a treatment visit, as well as possible effects on outcomes and quality of life. The primary outcome is the average time from HBO chamber exit to SRS beam-on, measured in minutes. The record says three participants were screen failures, did not receive study treatment, and were excluded from the primary analysis. However, the feasibility definition refers to 20 evaluable patients. The relationship between the enrollment count, screen failures, and evaluable-patient count is not explained.

This design does not include a randomized concurrent control group. Although historical controls appear in the adverse-event information, the record explicitly cautions that the arms are not a like-for-like control comparison. As a result, the posted information does not support treating those groups as equivalent treatment and control arms or attributing differences between their counts to HBO.

The registry lists 10 outcome measures and 2 analyses. It does not report a detailed statistical analysis plan, a sample-size justification for the study, or whether interim analyses were planned. Those details are not reported in the registry record. The study status is completed, but completion status alone does not show whether the planned feasibility threshold was reached or whether secondary outcomes differed from expectations.

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.

Before assignment: Adverse events were not collected for historic controls.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Three (3) participants were screen failures, and did not receive study treatment, and not included in the 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.

6. Primary Endpoint

The registry lists the primary outcome as The Average Time From HBO Chamber Exit to SRS beam-on, measured over Approximately 30 minutes during SRS Treatment Visit.

The primary endpoint is the average time, in minutes, from leaving the HBO chamber to the start of the SRS beam. This interval captures whether the sequence of HBO, transfer, head-frame placement, and preparation for SRS can be completed within the stated timing constraints. It is a process measure: it describes treatment timing rather than tumor response, survival, or quality of life.

The registry specifies a mean with full range as the summary approach and gives a timeframe of approximately 30 minutes during the SRS treatment visit. Its feasibility definition is that more than 50% of 20 evaluable patients start SRS within 30 minutes of leaving the HBO tank. The full range would describe the spread from the shortest to the longest observed interval; the mean would summarize the average interval. These summaries answer related but distinct questions: the mean describes the group average, while the feasibility rule focuses on how many patients meet a time threshold.

The registry does not report the observed average, full range, or the number of patients starting SRS within 30 minutes. It therefore does not show whether the stated feasibility criterion was met. The description also refers to completing transfer and head-frame placement within 15 minutes, while the primary feasibility definition uses a 30-minute window to SRS beam-on. The record does not reconcile these timing criteria.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a mean and full range for the primary timing endpoint, but does not report the observed values or a detailed analysis plan. The following explains the statistical concepts relevant to interpreting that endpoint; it does not imply that any particular unreported method was used.

What does the mean represent?

In general, a mean is the arithmetic average of the observed times. It can be useful for describing the overall time burden of a sequence such as leaving the HBO chamber and beginning SRS. However, an average alone does not show how many participants met a specific time limit. A few unusually long or short times can also affect a mean, so it is useful to read it together with information about the distribution of individual times. The registry lists full range as the dispersion summary, but the actual range is not reported in the registry record.

What does the feasibility threshold ask?

The stated criterion is based on the proportion of evaluable patients who begin SRS within 30 minutes of leaving the HBO tank. It is not identical to asking whether the average time is within 30 minutes. A group could have an average that does not reveal how many individuals met the threshold, and a threshold count does not describe how far outside the limit other individual times fell. Both the number meeting the limit and the denominator assessed are needed to interpret the operational result. Neither the observed count nor the primary result is reported in the registry record.

In general, an estimate of a proportion is more informative when accompanied by uncertainty, such as a confidence interval. A confidence interval describes a range of values compatible with the observed data under the method used; it does not make the estimate certain, and its width depends in part on the amount of information available. The record does not report a confidence interval or specify an inferential method for the feasibility proportion. It is therefore not possible to assess the statistical precision of that proportion from the posted results.

Why does the study design matter?

The study is listed as single-group, with allocation NA and masking none. In general, a single-group study can describe a procedure’s observed timing and participant outcomes, but it does not provide a randomized comparison that separates the intervention from other influences. A historical group is not automatically comparable with participants treated in the study. The registry specifically cautions that the historical controls shown in the adverse-event information are not a like-for-like comparison. Differences in counts between such groups are descriptive and do not establish that HBO caused a difference.

The intervention’s practical sequence also matters for interpretation. HBO was given at 2.4 ATA for 30 minutes immediately before SRS. The primary outcome begins at chamber exit and ends at SRS beam-on. The brief summary separately mentions a 15-minute limit for transfer and head-frame placement, while the feasibility rule uses a 30-minute interval to beam-on. These are not interchangeable endpoints unless the protocol explains how the steps and timing windows correspond. That explanation is not reported in the registry record.

How should the other outcomes be read?

The registry lists outcomes involving survival, local recurrence, distant intracranial recurrence, whole-brain radiation therapy, and radionecrosis, along with quality-of-life and mini mental status questionnaires in the study description. These outcomes differ in their definitions and time scales. In general, time-to-event outcomes depend on both when an event occurs and how long participants are observed without that event. A participant who has not experienced an event by the end of follow-up may contribute information differently from a participant with an observed event. The record does not give enough detail about analysis methods, censoring, or follow-up for these outcomes to assess how they were handled.

The sample-size material is described as a planning illustration based on the observed effect, not as the trial’s own design justification. It assumes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; time-to-event illustrations use total events, while other illustrations use participants per arm. These are planning assumptions, not observed power, and they do not turn this single-group feasibility study into a randomized comparison. The record does not report the study’s own sample-size justification or establish that the planning illustration was used to design the trial.

In general, over-dispersion occurs when observed variability is greater than a model assumes. If present, uncertainty estimates based on an overly restrictive model can be too narrow. This is a general statistical consideration rather than a reported finding for this study: the registry does not provide enough outcome data or model details to assess over-dispersion. More broadly, a statistical method can only be evaluated when its assumptions, analysis population, and outcome data are described. Those details are incomplete for the posted outcomes.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures and 2 analyses, including the primary timing endpoint and several secondary outcomes related to survival, recurrence, treatment, and radionecrosis. It also describes quality-of-life and mini mental status questionnaires. The registry record does not report how the outcomes were prioritized beyond identifying the primary endpoint, whether multiple testing was considered, or whether any adjustment for multiplicity was planned.

In general, evaluating many outcomes creates more opportunities for apparently notable findings to arise by chance. Multiplicity methods can manage that issue, but the appropriate approach depends on the questions and analysis plan. Without a reported plan, the secondary outcomes should be read as separate descriptive or exploratory questions unless the record gives a basis for stronger interpretation. No numerical results for these outcomes are posted in the information.

Missing-data handling is also not reported in the registry record. The primary endpoint description says three participants were screen failures, did not receive study treatment, and were excluded from the analysis; it also defines feasibility using 20 evaluable patients. The relationship among these counts is not explained. The record does not state how missing timing observations, incomplete questionnaires, or participants with incomplete follow-up were handled.

Beyond the screen-failure statement, the analysis populations are not described in sufficient detail to identify exactly who contributed to each outcome. In general, the choice of population and the treatment of missing outcomes can affect an estimate and its interpretation. Without those details and the relevant outcome counts, the posted record does not permit an assessment of potential bias from exclusions or incomplete follow-up.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event information lists an HBO Feasibility group and a Historic Controls group. Each group has 19 participants at risk for the serious adverse event count, and each has 0 participants affected. The other adverse-event count is 5 in the HBO Feasibility group and 6 in Historic Controls. The deaths count is 8 in HBO Feasibility and 0 in Historic Controls.

These figures should be read as counts in the registry record, not as evidence that the groups were comparable. The study is single-group, and the registry explicitly states that the arms are not a like-for-like control comparison. The historical-control counts therefore do not provide a reliable causal comparison of safety outcomes between HBO and an alternative treatment.

The record does not report details needed to interpret the adverse-event counts, including event descriptions for the other adverse events, timing, attribution, or follow-up context for the deaths. The serious adverse event listing has no top serious terms. A count of zero affected participants in the posted category does not establish that no safety concerns occurred outside the reported category or observation period. These counts alone do not show whether HBO caused an event or death.

12. What the Result Does, and Does Not, Mean

This study asks whether a tightly sequenced treatment visit is feasible: HBO at 2.4 ATA for 30 minutes immediately before SRS, followed by a short interval from chamber exit to SRS beam-on. The registry defines a specific threshold, more than 50% of 20 evaluable patients starting SRS within 30 minutes. That definition makes the operational question clear, but the posted record does not provide the observed mean time, full range, number meeting the threshold, or a conclusion about whether feasibility was achieved.

The record also describes a 15-minute time frame for transfer and head-frame placement. Since the primary endpoint’s threshold is 30 minutes to SRS beam-on, these descriptions should not be treated as the same measure without further clarification. The registry does not explain their relationship.

Even if the timing result were available, a single-group feasibility study would principally describe the experience of participants assessed in this study. It would not establish that HBO improves tumor control, survival, or quality of life compared with SRS without HBO. No randomized comparison is recorded, and the historical controls listed for adverse events are explicitly not a like-for-like comparison. The registry’s secondary-outcome list identifies questions the study intended to examine, but numerical results are not reported in the record.

Accordingly, the posted information documents the study design, intervention, feasibility criterion, and some participant counts. It does not establish that the operational threshold was met, quantify uncertainty around the primary result, or support causal claims about clinical benefit or harm. The distinction between an intended endpoint and a reported result is central to interpreting this record.

13. Why This Trial Matters Statistically

The statistical interest of this study lies in separating a procedural feasibility question from a clinical effectiveness question. The primary endpoint measures elapsed time from HBO chamber exit to SRS beam-on. This makes treatment sequence and timing the focus, rather than directly measuring tumor response or survival. A clearly specified threshold can make a feasibility question operational, but it still needs a reported numerator, denominator, and uncertainty to show how consistently the procedure met that threshold.

The record also illustrates why analysis populations need to be unambiguous. It reports enrollment of 22, three screen failures excluded from the primary analysis, and a feasibility definition based on 20 evaluable patients. The relationship among these counts is not explained. Without clarity about who was assessed, a reader cannot fully interpret a rate or judge how exclusions might affect it.

The design further emphasizes the limits of historical comparisons. The adverse-event information includes historical-control counts, yet the registry warns that the arms are not like-for-like. This is an important distinction: numerical differences between groups do not establish an intervention effect when the comparison is not designed to support that inference.

Finally, a completed study record is not the same as a complete statistical report. The registry identifies the primary endpoint and secondary questions, but does not post the primary timing result, its variability, or the numerical secondary results. The case therefore demonstrates why registry descriptions, outcome definitions, and reported estimates must be read separately when assessing what evidence is available.

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 primary limitation for interpretation is that the registry record does not report the primary endpoint result. The average time, full range, number meeting the 30-minute threshold, and feasibility conclusion are not reported. This prevents assessment of the study’s central operational question from the posted information.

There is an unresolved difference between the feasibility definition’s 20 evaluable patients and the participant accounting that lists 22 enrolled and three screen failures excluded from analysis. The record does not explain how these numbers relate. The brief summary’s 15-minute transfer and head-frame timing description also differs from the primary endpoint’s 30-minute interval to SRS beam-on, and the record does not reconcile the two.

The single-group design has no randomized concurrent comparator, and the historical controls are explicitly not a like-for-like comparison. Consequently, the posted design cannot support causal comparisons between HBO and an alternative sequence. The lack of reported numerical results for the listed secondary outcomes also limits assessment of survival, recurrence, quality of life, and other outcomes.

Other details needed for statistical appraisal are not reported in the registry record. These include a detailed analysis plan, a study sample-size justification, interim-analysis plans, missing-data methods, the analysis population for each outcome, and multiplicity handling. The adverse-event counts also lack contextual details such as timing and attribution. These limitations concern what can be concluded from the registry record; they do not establish that additional information was not collected or analyzed.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Proof-of-principle Study of HBO-SRS for Brain Metastases — NCT01850563.
• Registry record last updated 2022-02-08; status last verified 2022-01.
• Results first posted 2021-07-29.
• 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.

18. Record Summary

NCT01850563 is a completed, single-group study of hyperbaric oxygen immediately before stereotactic radiosurgery for brain metastases. The registry lists 22 enrolled participants, no masking, and no randomized allocation. HBO was given at 2.4 ATA for 30 minutes. The primary endpoint was the average time from chamber exit to SRS beam-on, with feasibility defined as more than 50% of 20 evaluable patients starting SRS within 30 minutes.

The registry states that three participants were screen failures and excluded from the primary analysis, but does not explain the relationship between that count and the 20 evaluable patients in the feasibility definition. The primary result and numerical results for the listed secondary outcomes are not reported in the registry record. Historical-control adverse-event counts are listed, with an explicit warning that the groups are not a like-for-like comparison. The record describes the study question and design, but does not provide enough outcome information to determine whether the feasibility criterion was achieved or to support comparative clinical conclusions.