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 Stereotactic Radiotherapy in Locally Advanced Pancreatic Cancer With Different Biological Effective Doses (NCT04603586), a randomized na study. The primary outcome is Progression-Free-Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04603586 |
| Phase | NA |
| Condition | Treatment |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 119 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Changhai Hospital |
| Results posted | Yes |
2. Clinical Question
Population
People with locally advanced pancreatic cancer are the population described in the study title and summary. The registry record reports enrollment of 119, but the primary outcome population field is blank. Eligibility criteria and the characteristics of participants enrolled are not reported in the registry record.
Intervention
Arm A received stereotactic body radiotherapy (SBRT) with a biological effective dose (BED) <70Gy, combined with Gemcitabine and albumin-bound paclitaxel. The treatment was intended to compare a lower BED category with a higher one while the chemotherapy regimen remained the same.
Comparator
Arm B received SBRT with a BED >70Gy, combined with Gemcitabine and albumin-bound paclitaxel. The registry describes a randomized, parallel-group study, with 60 participants listed in Arm A and 59 in Arm B.
Primary question
The primary endpoint was median progression-free survival, measured in months, with a stated timeframe of 3 years. The practical question is how the two dose strategies compare on the reported median estimates and their uncertainty. The registry record does not report a formal statistical comparison between arms.
The study’s brief summary says its aim was to compare safety and efficacy across the BED categories, with the goal of identifying a dose range that could balance survival benefit and radiotherapy toxicity. That aim describes the study question, not a conclusion established by the posted outcome figures. The registry lists adverse events and overall survival as secondary outcomes, but secondary outcome results are not included in the registry entries.
The registry title calls the work a prospective phase II clinical study, while the phase field is recorded as NA. These descriptions are not reconciled in the registry record. The study is listed as completed and sponsored by Changhai Hospital.
3. Trial Design
The registry describes a randomized, parallel-group study with no masking. The two arms used the same named chemotherapy combination, Gemcitabine plus albumin-bound paclitaxel, and differed in the SBRT BED category: Arm A used BED <70Gy and Arm B used BED >70Gy. The posted primary outcome lists 60 participants for Arm A and 59 for Arm B.
Enrollment is recorded as 119. The study start date is 2020-11-10, the primary completion date is 2025-08-15, and the overall status is completed. The registry record does not report the date of final study completion, how participants were recruited, the randomization procedure, allocation concealment, or whether participants crossed between treatments.
No masking was used. This means the registry describes the study as open-label, but it does not report whether outcome assessment or data analysis involved any separate masking procedures. Details of the SBRT delivery, the chemotherapy schedule, dose modification rules, and treatment adherence are not reported in the registry record.
The registry includes a phase II description in the official title, but the phase field itself is NA. The record does not provide a sample-size justification or describe whether enrollment was based on a statistical power calculation. No interim analyses or stopping rules are reported. These omissions limit what can be inferred about how the study was planned and monitored; they do not establish that such procedures were or were not used.
Randomized allocation can, in general, support a fair comparison by reducing systematic differences between treatment groups at the time of assignment. That general rationale does not establish that the groups were balanced on every relevant characteristic in this study. Baseline characteristics and details of the randomization process 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 | Arm A | Arm B |
|---|---|---|
| STARTED | 60 | 59 |
| COMPLETED | 60 | 59 |
| NOT COMPLETED | 0 | 0 |
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 | Arm A | Arm B | Total |
|---|---|---|---|---|
| Participants analysed | 60 | 59 | 119 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 27 | 27 | 54 |
| Age, Categorical: >=65 years | Participants | 33 | 32 | 65 |
| Sex: Female, Male: Female | Participants | 27 | 28 | 55 |
| Sex: Female, Male: Male | Participants | 33 | 31 | 64 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Baseline characteristics | Participants | 60 | 59 | 119 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-Free-Survival, measured over 3 years.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | month |
| Endpoint classification | Time-to-Event Analysis |
The primary endpoint is progression-free survival (PFS), summarized as a median in months. In general, PFS measures the time from a defined starting point until disease progression or death, according to the study’s specified definitions. Such an endpoint can describe how long participants remain alive without documented progression. The registry record does not give the operational definition of progression, the event-assessment schedule, the starting point for measuring time, or details of how events were confirmed.
A median is the point at which half of the observed time-to-event distribution lies on either side, subject to how censoring and follow-up are handled. It describes the center of the observed distribution, rather than the experience of every participant. The posted estimates are 8.6 months for Arm A, with a 95% confidence interval from 7.5 to 9.6, and 11.3 months for Arm B, with a 95% confidence interval from 10.1 to 12.5. The registry gives a timeframe of 3 years.
In general, a confidence interval expresses uncertainty around an estimated quantity under the assumptions of the analysis. It is not a range that contains the true value with certainty, nor does an interval for one arm by itself test whether the arms differ. The registry record does not report a formal comparison for the primary endpoint, a hazard ratio, or a p-value. It also does not specify the analysis population or explain how censoring was handled.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary endpoint estimates are median PFS of 8.6 months for Arm A and 11.3 months for Arm B. Arm A’s 95% confidence interval is 7.5 to 9.6 months; Arm B’s is 10.1 to 12.5 months. These figures describe the arm-specific median estimates and the uncertainty reported around each estimate. They do not, by themselves, quantify the uncertainty in a direct comparison between arms.
A comparison of the two median values is not a substitute for an inferential analysis. A formal analysis could account for the full time-to-event data, including event timing and censoring, and report a treatment contrast with its uncertainty. The registry record does not report a hazard ratio, a confidence interval for a between-arm effect, or a p-value for PFS. No sponsor analysis is posted among the outcome analyses. Consequently, the record does not establish whether the observed difference in the median estimates is statistically distinguishable from chance under a specified analysis.
The trial is described as randomized, which is relevant to the interpretation of comparative results. Randomization is generally intended to make treatment groups comparable on average at assignment. The record does not include baseline characteristics, the allocation method, or an account of adherence and treatment changes, so the degree to which the groups were comparable in practice cannot be assessed from these registry entries. The study is also listed as having no masking, which can matter for care and assessment decisions; the record does not describe any steps taken to reduce potential assessment bias.
The numerical outcome is limited to the median PFS estimates. It does not provide survival probabilities at particular times, a full survival curve, or information about the distribution of progression-free times beyond the reported summaries. The stated 3-year timeframe does not clarify how many participants were followed for that duration or whether follow-up was complete. Those details are not reported in the registry record.
The study summary identifies safety and efficacy as aims, and the registry lists adverse events and overall survival as secondary outcomes. However, the record does not include posted results for overall survival. Adverse-event counts appear in the safety record, but details of event definitions and attribution are limited. The primary endpoint figures therefore cannot answer the entire study question about both efficacy and toxicity.
Interpretation should remain tied to this registered comparison: SBRT at BED <70Gy versus BED >70Gy, with the same named chemotherapy combination. The outcome estimates do not isolate the effect of radiotherapy from chemotherapy, and they do not establish the optimal dose range. The record’s stated aim to identify a range that balances benefit and toxicity is a research objective, not itself evidence that such a range was identified.
9. Statistical Methods Explained
The primary result is a time-to-event endpoint summarized by a median. In general, time-to-event methods are designed for outcomes where the time until an event matters, rather than simply whether an event occurred. Participants may have different lengths of follow-up, and some may leave observation before an event is recorded. Such observations are commonly treated as censored, meaning the available follow-up contributes information up to a point without establishing a later event time. The registry record does not state how censoring was handled in this study.
What does a median PFS estimate describe?
The median is a summary of the event-time distribution. It is not the average follow-up time and does not tell how long each participant remained progression-free. The reported medians are 8.6 months for Arm A and 11.3 months for Arm B. Because the median compresses a distribution into one value, it cannot reveal all features of the underlying time-to-event experience. Survival curves, event counts, and additional follow-up summaries would provide other useful context, but those details are not reported in the registry record.
How should the confidence intervals be read?
Each median is accompanied by a 95% confidence interval: 7.5 to 9.6 months for Arm A and 10.1 to 12.5 months for Arm B. In general, a confidence interval reflects the precision of an estimate under the statistical method and assumptions used. A narrower interval indicates greater precision than a wider interval on the same scale, but precision is not the same as freedom from bias. The interval also does not mean that there is a 95% probability that the fixed true median lies within that particular calculated interval. The registry does not describe the method used to construct these intervals.
Confidence intervals for each arm are not the same as a confidence interval for the difference between arms. A direct comparison requires an analysis of the between-arm contrast. For time-to-event data, a model-based comparison might use a hazard ratio, while other analyses may compare survival distributions or specified time points. These are general examples, not a description of the method used in this study. The registry record does not report a hazard ratio or other formal PFS treatment contrast.
Why can’t the medians alone establish a treatment effect?
The two arm-specific medians are descriptive estimates. Their numerical values cannot determine the probability that a between-arm difference arose by chance, nor do they account on their own for the precision of a direct contrast. The registry record lists no analysis results for the primary endpoint. A p-value, a treatment-effect estimate, and its confidence interval are not reported. It would therefore be inappropriate to label the difference statistically significant or not statistically significant on the basis of these medians alone.
In general, time-to-event analyses often use information from event timing and censoring, not merely the median from each arm. A hazard ratio, when suitable and reported, compares event rates over follow-up under model assumptions. Such a ratio is not equivalent to a ratio of medians. No hazard ratio is reported in the registry record, and no analysis method or model assumptions are described.
What does randomization contribute?
In general, randomization is intended to reduce systematic differences between groups at allocation, so that outcome differences can be interpreted more credibly as related to treatment assignment, subject to appropriate conduct and analysis. The registry records randomized allocation but does not provide the sequence-generation method, concealment procedures, baseline characteristics, or details of protocol deviations. It is therefore possible to describe the planned comparison as randomized, but not to assess these design features from the available record.
The study is listed as parallel-group and unmasked. A parallel design compares groups receiving separate strategies over the study period. Lack of masking can potentially influence treatment decisions, participant reporting, or outcome assessment, depending on the endpoint and procedures. These are general methodological considerations. The registry record does not describe outcome adjudication or measures to limit assessment bias.
What can sample-size planning tell readers?
In general, a planning calculation estimates the number of participants or events needed to detect a specified effect under chosen assumptions. It is not evidence that a study actually had a particular probability of detecting an effect after the results are observed. A planning illustration based on an observed effect is also not the same as the trial’s original design justification. The registry record does not report the study’s own sample-size rationale, target effect, or power calculation. No trial-specific planning conclusion can be drawn from the outcome figures alone.
The record specifies that any planning illustration is based on a fraction of the observed effect, uses a two-sided significance level of 0.05, assumes equal allocation, and uses a normal approximation. For time-to-event outcomes, the described illustration concerns a total number of events. These are features of a planning illustration, not evidence of the study’s original design assumptions or observed power. Numerical planning results are not reported in the registry record.
Overall, the available statistical reporting supports a description of arm-specific median PFS and confidence intervals. It does not support a reconstruction of the sponsor’s analysis, a formal test of the primary comparison, or a complete assessment of precision and robustness. Details of endpoint definitions, censoring, analysis population, and model choice are not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists PFS as the primary outcome and adverse events and overall survival as secondary outcomes. It also records three outcome measures, but the registry entries do not give a full hierarchy of hypotheses or explain how multiple outcomes were handled. Multiplicity arises when several endpoints or comparisons are assessed: without an appropriate plan, repeated testing can increase the chance of a false-positive finding. This is a general statistical concern, not a claim that a particular adjustment was or was not used in this study.
No multiplicity adjustment, testing sequence, or prespecified hierarchy is reported in the registry record. The record also does not describe interim analyses, subgroup analyses, or any planned strategy for handling multiple time points. The absence of those details prevents assessment of how the study intended to control interpretation across outcomes and analyses.
Missing-data procedures and the primary analysis population are not reported. The outcome population field is blank, and the record does not specify whether the analysis was intention-to-treat, per-protocol, or based on another population. It does not describe withdrawals, loss to follow-up, missing endpoint assessments, or sensitivity analyses. For time-to-event outcomes, incomplete follow-up is often addressed through censoring, but the method and assumptions used are not reported.
The registry includes adverse-event counts and primary endpoint summaries, but those entries do not establish that every randomized participant contributed complete data to every analysis. The study’s completion status likewise does not define the completeness of individual outcome data. Readers should treat unreported analysis-population and missing-data procedures as limits on assessing the strength and robustness of the posted results.
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 |
|---|---|---|---|
| Arm A | 26 | 60 | 43.3% |
| Arm B | 25 | 59 | 42.4% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Arm A | 60 of 60 |
| Arm B | 55 of 59 |
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 |
|---|---|
| Arm A | 58 of 60 |
| Arm B | 57 of 59 |
Most frequent serious adverse event terms
| Event term | Organ system | Arm A | Arm B |
|---|---|---|---|
| Decreased appetite | Gastrointestinal disorders | 14 of 60 | 10 of 59 |
| Leukopenia | Blood and lymphatic system disorders | 11 of 60 | 10 of 59 |
| Weight decreased | Metabolism and nutrition disorders | 11 of 60 | 8 of 59 |
| Thrombocytopenia | Blood and lymphatic system disorders | 9 of 60 | 9 of 59 |
| Nausea or vomiting | Gastrointestinal disorders | 8 of 60 | 9 of 59 |
| Fatigue | General disorders | 7 of 60 | 8 of 59 |
| Elevated aspartate aminotransferase | Hepatobiliary disorders | 8 of 60 | 5 of 59 |
| Peripheral neurotoxicity | Nervous system disorders | 6 of 60 | 7 of 59 |
| Elevated alanine aminotransferase | Hepatobiliary disorders | 4 of 60 | 8 of 59 |
| Anemia | Blood and lymphatic system disorders | 2 of 60 | 7 of 59 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Arm A | Arm B |
|---|---|---|---|
| Alopecia | General disorders | 51 of 60 | 53 of 59 |
| Leukopenia | Blood and lymphatic system disorders | 49 of 60 | 50 of 59 |
| Fatigue | General disorders | 51 of 60 | 48 of 59 |
| Decreased appetite | Gastrointestinal disorders | 48 of 60 | 51 of 59 |
| Weight decreased | Metabolism and nutrition disorders | 40 of 60 | 46 of 59 |
| Thrombocytopenia | Blood and lymphatic system disorders | 41 of 60 | 39 of 59 |
| Elevated alanine aminotransferase | Hepatobiliary disorders | 37 of 60 | 42 of 59 |
| Nausea or vomiting | Gastrointestinal disorders | 38 of 60 | 40 of 59 |
| Elevated aspartate aminotransferase | Hepatobiliary disorders | 35 of 60 | 41 of 59 |
| Hypoalbuminemia | Metabolism and nutrition disorders | 17 of 60 | 22 of 59 |
Serious events with the highest arm-level rate: decreased appetite (23.3%), leukopenia (18.3%), weight decreased (18.3%), thrombocytopenia (15.3%), nausea or vomiting (15.3%).
The adverse-event record lists serious adverse events affecting 26 of 60 participants in Arm A and 25 of 59 in Arm B. It also lists other adverse events affecting 58 of 60 in Arm A and 57 of 59 in Arm B. These are counts as recorded; the registry does not provide definitions of seriousness, severity, attribution, event duration, or whether the counts represent participants with at least one event under a specified analysis rule.
The registry lists deaths as 60 for Arm A and 55 for Arm B in its adverse-event data. The record does not clarify the death definition, timing, relationship to treatment, or whether these counts refer to a particular reporting category. They should not be interpreted as treatment-caused deaths or as a complete mortality comparison without that context.
Listed serious-event terms include decreased appetite at 23.3%, leukopenia at 18.3%, weight decreased at 18.3%, thrombocytopenia at 15.3%, nausea or vomiting at 15.3%, and fatigue at 13.6%. The registry record does not state the denominators or whether these percentages apply to both arms combined or to another reporting set. They therefore cannot be assigned to a particular arm or used to make a direct arm-to-arm safety comparison.
Safety counts describe what was entered in the registry, but the reporting details are insufficient to establish comparative risk or causality. The record does not include a formal safety analysis.
12. What the Result Does, and Does Not, Mean
The posted results offer a descriptive comparison of median PFS for two radiotherapy dose categories used with the same named chemotherapy combination. Arm A’s median is 8.6 months, with a 95% confidence interval of 7.5 to 9.6; Arm B’s median is 11.3 months, with a 95% confidence interval of 10.1 to 12.5. These estimates describe the registered primary outcome and its reported uncertainty at the arm level.
They do not, on their own, establish that one dose category produces a statistically reliable improvement over the other. The registry record does not report a formal between-arm estimate, confidence interval for a contrast, or p-value. The medians also do not provide a complete picture of event timing, censoring, or outcomes over the stated 3-year timeframe.
The study’s summary describes an aim of comparing efficacy and safety to identify a dose range balancing survival benefit and toxicity. The posted PFS figures alone do not establish that such a range was identified. The registry lists overall survival as a secondary outcome, but no overall survival results are reported in the available record. Safety reporting includes event counts and listed terms, but lacks information needed for a full comparative assessment.
More generally, a randomized design can support causal interpretation when the trial is conducted and analyzed appropriately. The record does not provide enough detail to assess baseline balance, randomization procedures, adherence, endpoint assessment, missing data, or analysis choices. Randomized allocation is an important design feature, but it cannot replace those details or a direct statistical comparison.
These results should be understood as registry-reported estimates for this study, not as individualized guidance or a definitive statement about treatment. The registry record does not report the clinical characteristics of the enrolled participants or enough detail to establish how widely the estimates apply to other settings. A conclusion about comparative efficacy or safety would require appropriate analyses and complete reporting beyond the arm-specific summaries available.
13. Why This Trial Matters Statistically
This study addresses a statistical challenge common in clinical research: comparing treatment strategies that differ in one component while keeping other named components similar. Both arms combine SBRT with Gemcitabine and albumin-bound paclitaxel, while the BED category differs. The randomized, parallel-group description provides a framework for a comparative question, and the median PFS estimates give readers an initial view of the outcomes reported for each arm.
The record also illustrates why descriptive estimates and inferential comparisons answer different questions. An arm-specific median describes the center of the observed time-to-event distribution. A formal treatment comparison would quantify the contrast and its uncertainty while addressing event timing and follow-up. Because no such analysis is reported, the numerical medians are informative descriptions but do not settle the comparative question.
Reporting uncertainty matters as well. The confidence intervals communicate precision around the arm-specific medians, but they do not supply the missing confidence interval for a between-arm effect. This distinction is important whenever readers are tempted to infer a treatment difference from separate summaries alone.
Safety is another part of the study’s stated purpose, yet the record’s adverse-event information has limited context. Counts without clear definitions, denominators for the listed terms, or a comparative analysis cannot establish relative safety. The registry also lists overall survival as a secondary outcome without posted results, leaving an important aspect of the stated research question unresolved in this record.
Overall, the study is statistically noteworthy as a randomized dose-category comparison, while the available reporting supports only a partial reading of its findings. It demonstrates the value of complete methods and outcome reporting: endpoint definitions, analysis populations, missing-data handling, direct effect estimates, and uncertainty are necessary to move from descriptive results to a well-supported comparative conclusion.
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.
Several trial-specific details needed to evaluate the reported PFS results are not reported in the registry record. These include the operational definition of progression, the time origin for PFS, how progression was assessed, how events were confirmed, and how censoring was handled. Without these details, the endpoint’s implementation and analysis cannot be independently evaluated.
The record gives arm-specific medians and confidence intervals but no formal between-arm analysis. It does not report a hazard ratio, a confidence interval for the treatment contrast, or a p-value. The analysis population is not specified, and the record does not describe missing outcome data, participant withdrawals, treatment adherence, or protocol deviations.
Although allocation is recorded as randomized, the randomization method and allocation concealment are not described. The study is listed as having no masking, but the record does not say how outcome assessment was conducted or whether any independent adjudication was used. Baseline participant characteristics are also not reported in the registry entries, limiting assessment of group comparability.
The registry identifies adverse events and overall survival as secondary outcomes, but secondary results are not available in the record. Safety event definitions and attribution are not described, and the denominators for listed serious-event percentages are unspecified. These limitations prevent a full evaluation of the study’s combined efficacy and safety aim.
The registry title describes the study as phase II, while the phase field is NA. It is listed as completed, but the record does not give a final completion date or details of follow-up completeness. The stated timeframe of 3 years does not by itself show how many participants contributed data through that period. These reporting gaps constrain interpretation without proving that the underlying trial lacked the corresponding procedures or data.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Stereotactic Radiotherapy in Locally Advanced Pancreatic Cancer With Different Biological Effective Doses — NCT04603586.
- Registry record last updated 2025-10-20; status last verified 2025-09.
- Results first posted 2025-10-20.
- 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
The registry record describes a randomized, parallel-group, unmasked study comparing SBRT at BED <70Gy with SBRT at BED >70Gy, with Gemcitabine and albumin-bound paclitaxel in both arms. It reports enrollment of 119, lists Changhai Hospital as sponsor, and records the study as completed. The start date is 2020-11-10, and the primary completion date is 2025-08-15.
The primary outcome is median PFS in months over a stated timeframe of 3 years. The reported estimate is 8.6 months for Arm A, with a 95% confidence interval of 7.5 to 9.6, and 11.3 months for Arm B, with a 95% confidence interval of 10.1 to 12.5. No formal between-arm primary analysis is reported in the registry record.
Adverse events and overall survival are listed as secondary outcomes. The record includes adverse-event counts and selected serious-event terms, but does not provide overall survival results or enough detail for a complete comparative safety assessment. The title describes phase II, while the phase field is NA.