Stereotactic Body Radiation Therapy in Treating Patients With Stage I Non-Small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Stereotactic Body Radiation Therapy in Treating Patients With Stage I Non-Small Cell Lung Cancer (NCT00750269), a non-randomized phase 1, phase 2 study. The primary outcome is (Phase II) Primary Tumor Control Rate at the Maximum Tolerated Dose (MTD).

2. Clinical Question

The study focuses on a treatment setting where the tumor location and patients’ medical inoperability are central to the clinical question. Rather than comparing SBRT with a separate control treatment, the study evaluates a sequence of dose levels. This supports a dose-finding purpose in phase I and an assessment of tumor control at the selected maximum tolerated dose in phase II. The registry also notes that Dose Level 8 was analyzed, in addition to the maximum tolerated dose, because of the quantity of patients enrolled at that level and safety concerns.

The primary endpoint is not a general measure of survival or all cancer progression. It concerns control of the primary treated tumor, using a protocol-defined failure assessment. Other listed outcomes include progression-free survival, overall survival, local and nodal progression, distant metastases, and toxicity. Since no randomized comparator is listed, differences between dose levels cannot be interpreted as though patients had been assigned at random to comparable groups. The registry record does not report the primary endpoint result needed to assess the estimated rate or its precision.

3. Trial Design

This is a completed, non-randomized phase I/II study with a single-group model and no masking. The registry lists an enrollment of 120. That is the recorded enrollment figure; it should not be read as the number of participants included in each endpoint analysis. The primary endpoint population is described separately as all eligible patients who started study treatment.

The study evaluates stereotactic body radiation therapy in patients with stage I non-small cell lung cancer. Its official title specifies centrally located disease in medically inoperable patients. The intervention descriptions present escalating dose levels. The arm descriptions list nine levels, from Level 1 at 8.0 Gy/FX to Level 9 at 12.0 Gy/FX. The intervention list, however, names total doses only through SBRT 57.5 Gy, while the arm descriptions also list SBRT 60.0 Gy for Level 9. This difference in the registry descriptions should be kept in mind when interpreting the listed treatment levels.

The primary endpoint is the tumor-control rate at the maximum tolerated dose, measured from the start of SBRT to 2 years. The record says the protocol intended to analyze only the maximum tolerated dose, but Dose Level 8 was also analyzed because of the quantity of patients enrolled at that dose and safety concerns. It does not state the full dose-selection algorithm or the criteria used to determine the maximum tolerated dose; those details are not reported in the registry record.

Listed secondary outcomes cover survival, progression patterns, and toxicity. Safety figures are recorded for Dose Levels 5 through 9, rather than every dose level described. The registry lists the study as completed and gives start and primary-completion dates of 2009-02 and 2014-09. These design features describe the trial structure, but do not establish the magnitude or precision of tumor control. The numerical primary endpoint result is not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All eligible patients who started study treatment

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: All eligible patients receiving protocol treatment

6. Primary Endpoint

The registry lists the primary outcome as (Phase II) Primary Tumor Control Rate at the Maximum Tolerated Dose (MTD), measured over From start of SBRT to 2 years..

The primary endpoint is the primary tumor control rate at the maximum tolerated dose, assessed from the start of SBRT to 2 years. Control is defined as the absence of primary tumor failure. The protocol’s failure definition requires both a specified increase in tumor dimension for local enlargement and supporting PET imaging with uptake similar in intensity to the pretreatment staging PET, or biopsy confirmation of viable carcinoma. Marginal failures and involved lobe failures are also counted as primary tumor failures.

This endpoint is intended to capture whether the treated primary tumor remains controlled under the study’s specified assessment rules. The criteria combine measured tumor change with imaging or biopsy evidence, rather than treating any single observation as sufficient. Including marginal and involved lobe failures broadens the endpoint beyond failure confined to the center of the treated tumor.

The registry says the cumulative incidence method was used to estimate the primary tumor-control rate, and that a 90% confidence interval for local control was calculated using bootstrapping. A confidence interval describes uncertainty around an estimate under the analysis method; it does not show the range of outcomes for individual patients. The record does not give the estimate or interval limits, so the observed rate and its precision are not reported in the registry record.

The endpoint is distinct from overall survival, progression-free survival, nodal progression, distant metastases, and toxicity, which are listed separately. A tumor-control rate alone would not answer those other questions. The registry’s endpoint description therefore defines a focused measure of local primary-tumor outcome rather than a comprehensive summary of benefit or harm.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes several statistical ideas relevant to interpreting this study: a dose-finding design, a time-bounded tumor-control endpoint, cumulative incidence estimation, and a bootstrap confidence interval. It does not report enough detail to reconstruct the full analysis plan or reproduce the endpoint analysis. General explanations below describe what these methods mean, not unreported choices made by the investigators.

What does the single-group, non-randomized design imply?

A single-group study follows participants receiving the study intervention without a concurrent control group. Here, the registry describes escalating SBRT dose levels in a phase I/II study. Dose escalation is intended to support dose evaluation, but the listed levels are not randomized groups. Randomization helps balance measured and unmeasured participant characteristics on average; without it, differences across dose levels can reflect factors other than dose. A non-randomized contrast among levels is therefore not equivalent to a randomized treatment comparison.

The record says that only the maximum tolerated dose was intended for primary endpoint analysis under the protocol, with Dose Level 8 also analyzed because of enrollment quantity at that level and safety concerns. It does not report the dose-escalation rules, decision criteria, or how the maximum tolerated dose was selected. Those features are important to interpreting dose selection and are not reported in the registry record.

What does the primary tumor-control endpoint measure?

The endpoint is the rate of primary tumor control from the start of SBRT to 2 years. Its event definition is the absence of primary tumor failure. Failure depends on a combination of measured tumor enlargement and corroborating PET evidence or biopsy confirmation of viable carcinoma. Marginal failures and involved lobe failures are also treated as primary tumor failures. These rules make the endpoint more specific than a simple record of any change in tumor size.

For a time-bounded outcome, the analysis must account for when events occur and for participants who may not have complete observation through the stated follow-up horizon. The registry identifies the cumulative incidence method, but does not describe the competing events, censoring rules, or handling of incomplete follow-up. The endpoint description alone does not establish how those details were implemented.

Why use cumulative incidence?

General explanation: cumulative incidence estimates the probability of experiencing a particular event by a specified time while accounting for competing events that can prevent that event from occurring. It differs from methods that treat competing events as though they were ordinary loss to follow-up. The appropriateness and interpretation of a cumulative-incidence estimate depend on which competing events are defined and how they are handled. Those details are not reported in the registry record.

In this study, the registry states that the cumulative incidence method was used to estimate primary tumor control. Because the registry does not report the estimate, its interval limits, or the competing-event definitions, the numerical outcome and its sensitivity to those analysis choices cannot be assessed from the record.

What does a bootstrap confidence interval communicate?

General explanation: a confidence interval expresses uncertainty around an estimated quantity under the assumptions and procedures of the analysis. A bootstrap interval is built by repeatedly resampling observed data and recalculating the estimate, rather than relying only on a formula based on a particular theoretical distribution. Bootstrap methods can be useful when the sampling distribution of an estimate is difficult to represent with a simple formula. They do not remove uncertainty or correct for bias in study design, measurement, or follow-up.

The registry reports a 90% confidence interval for local control calculated using bootstrapping. It does not state the resampling procedure, the number of resamples, or other implementation details. More importantly for interpretation, the numerical interval itself is not reported in the registry record. Without its limits, readers cannot judge the estimate’s precision.

How should the safety counts be read?

The adverse-event entries present counts affected and counts at risk for serious and other adverse events, grouped by dose level. General explanation: a count of affected participants is meaningful only alongside its denominator and the event definition, follow-up period, and ascertainment process. Counts across groups should not be added or compared casually when groups differ in size or were not assigned at random. Participants may also contribute to more than one event category, so event counts do not necessarily represent unique people.

The registry’s adverse-event information includes Level 5 through Level 9 group entries. It lists serious adverse-event terms with percentages separately. The terms and percentages do not establish whether an event was caused by treatment, and serious classification is not interchangeable with clinical severity grading. The record’s listed serious-term percentages lack corresponding denominator and count detail in the registry record for this account.

What can planning illustrations establish?

General explanation: a sample-size calculation can illustrate how many participants might be needed under specified assumptions about the effect to detect, significance level, allocation, and statistical approximation. Such a calculation is a planning exercise, not a measurement of the trial’s achieved power. If planning illustrations are based on an observed effect, they depend on that effect estimate and should not be mistaken for evidence that the study was designed around the same assumptions.

The registry record does not report the trial’s own sample-size justification. It also does not report the numerical primary endpoint estimate needed to inspect planning illustrations based on the observed effect. Any sample-size values should therefore be read only as conditional planning illustrations, not as observed power or as a substitute for the endpoint result.

Overall, the record identifies some analysis methods but leaves important details unspecified. The statistical analysis plan, missing-data methods, precise analysis populations for secondary outcomes, competing-event definitions, and implementation details are not reported in the registry record. Those absences constrain independent assessment of the analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Level 8: 11.5 Gy/FX", "Level 9: 12.0 Gy/FX")
n      <- c(38, 33)
events <- c(34, 29)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Level 8: 11.5 Gy/FX  38  34  89.5  75.2  97.1  75.9  95.8  75.8  95.7
ARM  Level 9: 12.0 Gy/FX  33  29  87.9  71.8  96.6  72.7  95.2  72.5  95.0

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary tumor-control endpoint and several secondary outcomes: progression-free survival, overall survival, local progression, nodal progression, distant metastases, and toxicity outcomes. It does not report a multiplicity strategy for interpreting results across these endpoints, dose levels, or analyses. The multiplicity plan is therefore not reported in the registry record.

General explanation: when many outcomes or comparisons are examined, the chance of at least one apparently notable result can increase. A prespecified strategy may distinguish a primary confirmatory question from exploratory analyses, or adjust interpretation for multiple testing. Without a reported plan, readers cannot determine whether any later nominal statistical findings were intended as confirmatory or exploratory. The registry’s list of outcomes alone does not establish such a strategy.

The record specifies the primary endpoint population as all eligible patients who started study treatment. It does not give analysis populations for each secondary outcome or explain how participants with missing endpoint assessments, incomplete follow-up, or protocol deviations were handled. Missing-data methods and outcome-specific analysis populations are not reported in the registry record.

The primary endpoint uses cumulative incidence and a bootstrap confidence interval, but the record does not specify competing-event definitions or detailed resampling procedures. It also does not provide a complete statistical analysis plan or interim-analysis information. Those items are not reported in the registry record, so they should not be inferred from the phase I/II design or from the fact that Dose Level 8 was analyzed in addition to the maximum tolerated dose.

12. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dysphagia (12.5%), esophagitis (12.5%), dyspnea (7.9%), sinus bradycardia (7.1%), respiratory failure (7.1%).

The serious adverse-event entries give, for each listed dose level, the number at risk and the number affected. At Level 5, the record lists 8 at risk and 1 affected; at Level 6, 7 at risk and 0 affected; at Level 7, 14 at risk and 2 affected; at Level 8, 38 at risk and 9 affected; and at Level 9, 33 at risk and 7 affected. These are group-level registry entries. They do not establish that events were caused by SBRT, nor do they support randomized comparisons among dose levels.

The other-adverse-event entries list 7 affected among 8 at risk at Level 5, 7 among 7 at Level 6, 13 among 14 at Level 7, 34 among 38 at Level 8, and 27 among 33 at Level 9. Serious and other adverse-event categories should be read as the registry presents them; they are not interchangeable measures. Counts should not be pooled across levels, and the groups should not be treated as like-for-like controls.

The record also names dysphagia and esophagitis at 12.5% each among the top serious adverse-event terms, with other listed terms at their recorded percentages. The denominator and event counts for those term percentages are not reported in the registry record. An adverse event classified as serious is not necessarily the same as an event graded as severe, and these entries do not by themselves establish event attribution.

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

The study addresses dose evaluation and primary tumor control after SBRT for stage I non-small cell lung cancer in medically inoperable patients, with the official title specifying centrally located disease. The registry’s endpoint definition is detailed: tumor control means absence of primary tumor failure under specified tumor measurement and confirmation criteria, with marginal and involved lobe failures counted as failures. This provides a defined outcome to evaluate, rather than relying on an unspecified impression of response.

However, the numerical primary tumor-control result and its 90% confidence interval are not reported in the registry record. The record therefore does not show the observed control rate or how precisely it was estimated. The listed enrollment of 120 cannot fill this gap and should not be interpreted as the analyzed population for the endpoint.

The results also cannot establish how SBRT compares with another treatment, because the study is non-randomized, single-group, and has no concurrent comparator. Dose levels are not randomized arms, and the registry cautions that they are not a like-for-like control comparison. Any differences across dose levels could reflect factors other than dose.

The safety entries describe affected counts in listed dose groups and selected serious adverse-event terms. They provide context about reported events, but do not independently establish causation or comparative safety. The term percentages do not include denominator and count details in the registry information available for this account.

Thus, the record supports an understanding of the study’s question, design, endpoint definition, and selected safety reporting. It does not support a numerical conclusion about the primary tumor-control outcome, its precision, or comparative effectiveness. Missing analysis details also limit independent evaluation of follow-up, missing data, multiplicity, and endpoint implementation.

14. Why This Trial Matters Statistically

This trial is statistically notable for the way its clinical question and design differ from a conventional randomized comparison. It uses a non-randomized, single-group phase I/II structure to examine escalating SBRT dose levels, with dose evaluation followed by assessment of primary tumor control at the maximum tolerated dose. Understanding that structure is essential: dose levels in a dose-finding study are not automatically comparable treatment groups.

The endpoint definition illustrates how a clinical outcome can depend on explicit measurement rules. Primary tumor failure requires specified tumor enlargement along with PET or biopsy evidence, while marginal and involved lobe failures are also counted. Such definitions affect which observed events contribute to the outcome and therefore matter for reproducibility and interpretation.

The record also identifies cumulative incidence estimation and bootstrap confidence intervals. These methods highlight that time-to-event outcomes may require attention to competing events and that uncertainty can be estimated through resampling. Yet methods labels do not replace numerical results or full analytic detail. Here, the numerical control estimate and confidence limits are not reported in the registry record, preventing evaluation of the result’s magnitude and precision.

Finally, the registry’s distinction between the maximum tolerated dose analysis and the additional Dose Level 8 analysis underscores that analyses may be influenced by safety considerations and enrollment patterns. The record explains the reason for including Level 8, but does not provide the full statistical analysis plan. This makes it important to separate the stated design rationale from conclusions that require outcome estimates and complete methods.

15. 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 registry record does not report the numerical primary tumor-control estimate or the limits of its 90% confidence interval. This prevents assessment of the endpoint’s observed magnitude and precision. Although the record states that the primary endpoint population is all eligible patients who started treatment, it does not report the number meeting that definition or the number contributing to the endpoint analysis.

The study is non-randomized and single-group, with no concurrent control. Dose levels therefore cannot be treated as randomized comparisons, and observed differences across levels would be vulnerable to confounding. The record does not describe the dose-escalation rules or selection criteria for the maximum tolerated dose. It also reports an additional analysis of Dose Level 8 because of enrollment quantity and safety concerns, but does not provide full analysis details.

There is a mismatch within the treatment descriptions: the intervention list ends at SBRT 57.5 Gy, while the arm descriptions include Level 9 at SBRT 60.0 Gy. The reason for this difference is not reported in the registry record. Safety entries are listed only for Level 5 through Level 9 in the registry record, so they do not describe safety information for every listed level.

The registry does not report the full statistical analysis plan, multiplicity approach, missing-data methods, competing-event definitions, or detailed bootstrap implementation. It also does not give denominator and count detail for the named serious adverse-event percentages. These omissions constrain independent appraisal of endpoint analyses and safety reporting. No external control, causal treatment comparison, or complete assessment of all outcomes can be reconstructed from the registry description alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Stereotactic Body Radiation Therapy in Treating Patients With Stage I Non-Small Cell Lung Cancer — NCT00750269.
• Registry record last updated 2022-06-09; status last verified 2022-05.
• Results first posted 2017-05-10.
• 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.

19. Record Summary

The registry describes a completed, non-randomized phase I/II single-group study of SBRT for stage I non-small cell lung cancer, with the official title specifying centrally located disease in medically inoperable patients. Recorded enrollment is 120. The primary endpoint is tumor control at the maximum tolerated dose from the start of SBRT to 2 years, with a protocol-defined primary tumor failure assessment. Cumulative incidence estimation and a bootstrap 90% confidence interval are reported as the analysis methods.

The registry says Dose Level 8 was analyzed in addition to the maximum tolerated dose because of enrollment quantity and safety concerns. It lists secondary survival, progression, metastasis, and toxicity outcomes, as well as serious and other adverse-event counts for Dose Levels 5 through 9. The numerical primary tumor-control estimate and confidence limits are not reported in the registry record. The design does not provide a randomized control comparison, and further details on multiplicity, missing data, and the full analysis plan are not reported in the registry record.