This page separates information reported in the ClinicalTrials.gov record from statistical interpretation. The ClinicalTrials.gov record contains a registered primary endpoint and confirm that results were posted, but no formal statistical-analysis records or numeric primary-endpoint estimates are included in the ClinicalTrials.gov record.
1. Trial at a Glance
APPLE was a phase 3, single-group clinical trial evaluating an adenosine intervention after pulmonary vein isolation. The registry describes the intervention as an adenosine arm and defines the primary endpoint as reconnection of the pulmonary vein electrogram after adenosine infusion, assessed 5 minutes after intravenous adenosine.
| Feature | APPLE |
|---|---|
| Trial name | APPLE |
| Brief title | Adenosine Effect on Persistent Pulmonary Vein Isolation Line Evaluation (APPLE Trial) |
| Phase | Phase 3 |
| Status | COMPLETED |
| Therapeutic area | Cardiovascular |
| Conditions | Atrial Fibrillation; Persistent Common Pulmonary Vein |
| Allocation | NA |
| Design model | SINGLE_GROUP |
| Masking | NONE |
| Primary purpose | TREATMENT |
| Enrollment | 378.0 |
| Intervention | Adenosine arm (drug) |
| Lead sponsor | Yong Seog Oh |
| Sponsor type | OTHER |
| ClinicalTrials.gov | NCT01932112 |
2. Clinical Question
The registered primary endpoint asks whether pulmonary vein electrogram reconnection can be identified after intravenous adenosine following pulmonary vein isolation. Because the registry design is single-group, this is not a randomized comparison of adenosine against a concurrent control intervention.
Population
The registry identifies the conditions as atrial fibrillation and persistent common pulmonary vein. The ClinicalTrials.gov record does not provide a more detailed baseline eligibility description.
Intervention
The study intervention is an adenosine arm. The registered endpoint definition specifies 20mg intracardiac adenosine after pulmonary vein isolation.
Comparator
None is identified. The allocation field is NA and the design model is SINGLE_GROUP.
Primary question
Does administration of adenosine allow evaluation of pulmonary vein reconnection after pulmonary vein isolation?
3. Trial Design
Adenosine arm
- Adenosine is the registered intervention.
- The primary endpoint definition specifies 20mg intracardiac adenosine after pulmonary vein isolation.
- Pulmonary vein reconnection is evaluated following adenosine administration.
- The registered endpoint time frame is 5 minutes after IV adenosine.
Interpretive consequence
- The registry-reported design contains one study arm.
- There is no randomized between-arm treatment effect to estimate.
- A proportion of participants meeting the endpoint would be a descriptive quantity rather than a treatment-versus-control effect.
- Causal comparison with an untreated or alternative-treatment group cannot be derived from this design alone.
4. Trial Timeline
Study start
The ClinicalTrials.gov record identifies 2012-11 as the trial start date.
Primary completion
The ClinicalTrials.gov record identifies 2015-04 as the primary completion date.
Registry status
The trial status is recorded as COMPLETED, with results posted.
5. Primary Endpoint
| Endpoint | Registered definition | Time frame | Endpoint type |
|---|---|---|---|
| Reconnection of Pulmonary Vein Electrogram After Adenosine Infusion | After pulmonary vein isolation, 20mg Intracardiac adenosine will be given to treatment group, will evaluate pulmonary vein reconnection. | 5 minutes after IV adenosine | Other / unclear |
The endpoint is fundamentally a binary or categorical clinical-observation problem if each participant is classified as having or not having pulmonary vein reconnection at the specified assessment. However, the ClinicalTrials.gov record does not provide the number of participants with reconnection, the number without reconnection, or a formal statistical estimate for the endpoint.
For a binary endpoint, x would represent the number of participants with pulmonary vein reconnection and n the number evaluated. A confidence interval for the proportion could then quantify sampling uncertainty. Neither x nor a corresponding confidence interval is reported in the ClinicalTrials.gov record provided for APPLE.
6. Results
The ClinicalTrials.gov data posted on ClinicalTrials.gov for APPLE state Results Posted: Yes and identify 2 outcome measures posted. However, no formal statistical analyses were posted, and the registered primary endpoint ClinicalTrials.gov record contain no numeric endpoint estimate, event count, confidence interval, or p-value.
Primary endpoint: formal statistical estimate
No numeric reconnection count, proportion, confidence interval, or p-value is provided in the ClinicalTrials.gov record extract.
The absence of a statistical-analysis record means a treatment effect or formal between-group comparison cannot be reported from these data.
The important statistical distinction is between results being posted and a formal comparative statistical analysis being available in the ClinicalTrials.gov record. APPLE is recorded as a completed study with results posted, but the ClinicalTrials.gov record does not contain a statistical-analysis entry for the primary endpoint.
For this single-group design, a natural analysis would describe the number and proportion of participants in whom pulmonary vein electrogram reconnection was observed after adenosine. A confidence interval would describe uncertainty around that estimated proportion. A p-value would require a prespecified null hypothesis, such as a comparison with a reference proportion, which is not reported here.
Accordingly, no hazard ratio, risk ratio, odds ratio, treatment difference, or comparative p-value should be constructed from the available information. Such quantities would imply a comparison that is not part of the registry-reported design.
What the posted-results status does mean
It means the registry identifies outcome measures as having been posted. It does not, by itself, establish that the registry-reported extract contains every numerical result or every statistical procedure used to analyze those outcomes. The ClinicalTrials.gov record specifically contain 2 outcome measures posted, but only 1 registered primary endpoint is described in the endpoint-specific information provided here.
7. Secondary Outcome Information
the ClinicalTrials.gov record states that 2 outcome measures were posted. The ClinicalTrials.gov record does not provide the names, definitions, time frames, numerical results, or statistical analyses for the additional outcome measure.
8. Statistical Methodology
Why a single-group analysis changes the statistical question
In a randomized parallel-group trial, the principal estimand often compares outcomes between treatment groups. APPLE is different: the registry design is SINGLE_GROUP with 1 arm and allocation recorded as NA. The central statistical object is therefore a within-study descriptive outcome rather than a randomized treatment contrast.
Binary endpoint analysis
If pulmonary vein reconnection is classified as present or absent, the basic descriptive quantity is the observed proportion. The numerator would be the number of participants with reconnection and the denominator would be the number of participants assessed at the specified time point.
A confidence interval for p̂ would quantify uncertainty caused by sampling variation. The appropriate interval method depends on the analysis plan and sample size; the registry-reported APPLE data do not specify the confidence-interval procedure.
Confidence intervals
A confidence interval would be useful because a single observed proportion is only an estimate of the underlying probability in the population represented by the study. A narrow interval would indicate greater statistical precision, while a wider interval would indicate greater uncertainty. The interval would not describe the probability that the true parameter lies inside the particular calculated interval.
Hypothesis testing
A p-value is meaningful only relative to a specified null hypothesis and analysis procedure. For example, a one-sample test could compare an observed proportion with a prespecified reference value. No such reference value is posted on ClinicalTrials.gov for APPLE, so a p-value cannot be responsibly reconstructed.
Denominator definition
For a binary endpoint, the denominator matters. If some enrolled participants were not evaluable at the endpoint assessment, the analysis could distinguish the full enrolled population from the endpoint-evaluable population. The ClinicalTrials.gov record gives enrollment of 378.0, but do not provide an endpoint-evaluable denominator.
9. Statistical Methods Explained
Why is a proportion a natural analysis for the primary endpoint?
The registered endpoint concerns whether pulmonary vein electrogram reconnection is observed after adenosine. If each participant receives a binary classification of reconnection versus no reconnection, the natural summary is the proportion with reconnection. This describes the frequency of the observed phenomenon without creating an artificial treatment comparison.
Why is there no hazard ratio?
A hazard ratio is ordinarily used for a time-to-event comparison between groups or modeled covariate levels. APPLE's registry-reported primary endpoint is assessed at a defined post-adenosine time frame of 5 minutes, and the trial has only one study arm. The ClinicalTrials.gov record therefore provide no basis for a treatment-versus-control hazard ratio.
Why would a confidence interval be useful?
A confidence interval would show how precisely the study estimates the proportion of participants with reconnection. Two studies can have the same observed proportion but different precision because their sample sizes or numbers of evaluable observations differ. APPLE's enrollment is 378.0, but the registry-reported endpoint-specific data do not establish the corresponding number evaluated.
Why is a p-value not automatically required?
Descriptive estimation does not require a hypothesis test. If the scientific goal is to estimate how frequently reconnection is observed after adenosine, an estimated proportion and confidence interval can be more directly aligned with the question than a p-value. A p-value becomes appropriate only when a prespecified null hypothesis is part of the study question.
What does the single-group design imply about causality?
Without a concurrent randomized comparator, the study can characterize observations following the intervention, but the registry-reported design does not isolate an effect by comparing outcomes under adenosine with outcomes under a different randomized condition. Any causal interpretation therefore requires assumptions or evidence beyond a simple between-arm comparison, because there is no between-arm comparison in this design.
Why does the endpoint time frame matter?
The registry specifies 5 minutes after IV adenosine. That time frame defines when the endpoint is assessed. A result measured at another time point would represent a different estimand unless the protocol explicitly treated the measurements as equivalent.
10. Safety Results
The registry profile reports serious adverse events by arm as Adenosine Arm: 2/378.
| Safety measure | Adenosine arm |
|---|---|
| Serious adverse events | 2/378 |
The reported safety quantity is an affected-to-at-risk count: 2/378. It identifies 2 participants with serious adverse events among 378 participants at risk as reported in the ClinicalTrials.gov record.
This count should not be converted into a comparative risk estimate because APPLE has no concurrent control arm in the registry-reported design. It also should not be treated as evidence that all other participants were completely free of any adverse event; the registry-reported measure is specifically serious adverse events.
Because no comparator is available, the safety result is descriptive. A comparative risk difference or risk ratio would require another defined population or treatment group.
11. What the Primary Endpoint Can and Cannot Establish
What it can describe
The frequency with which pulmonary vein electrogram reconnection is observed after adenosine at the registered assessment time.
What it cannot provide by itself
A randomized treatment effect comparing adenosine with another intervention, because no concurrent comparator arm is identified.
What a confidence interval would add
It would quantify uncertainty around the estimated endpoint proportion rather than simply reporting one observed percentage.
What the p-value would require
A prespecified null hypothesis and corresponding statistical test. Neither is posted on ClinicalTrials.gov for the APPLE primary endpoint.
This distinction is particularly important in clinical-trial reporting. A registry endpoint can be clinically meaningful even when its appropriate statistical summary is descriptive rather than comparative. The correct analysis follows the design and estimand rather than forcing a familiar randomized-trial statistic onto a single-group study.
12. Analysis Population and Missing Data
the ClinicalTrials.gov record reports 378.0 enrolled participants. It does not specify an intention-to-treat population, per-protocol population, safety population definition, endpoint-evaluable population, or missing-data strategy.
| Population / issue | Information reported | Statistical implication |
|---|---|---|
| Enrollment | 378.0 | Total enrollment is known. |
| Study arms | 1 | No randomized treatment comparison is available. |
| Endpoint-evaluable count | Not reported | The denominator for a primary-endpoint proportion cannot be confirmed. |
| Missing-data method | Not reported | No imputation procedure should be assumed. |
| Analysis population definitions | Not reported | Population-specific estimates cannot be reconstructed. |
Why missing data matter here
Suppose a binary endpoint is observed for fewer participants than were enrolled. The resulting proportion depends on how participants with missing assessments are handled. A complete-case analysis, a conservative assumption, or a prespecified imputation method can produce different results.
13. Design Features Not Supported by the Supplied Data
Several statistical design topics commonly relevant to clinical trials are not documented in the registry-reported APPLE data. They should therefore not be presented as features of this trial.
| Design topic | What can be concluded from the ClinicalTrials.gov record |
|---|---|
| Non-inferiority margin | Not reported; no non-inferiority comparison is identified. |
| Crossover | Not reported. |
| Factorial design | Not reported; the design model is SINGLE_GROUP. |
| Multiplicity adjustment | Not reported. |
| Interim analysis | Not reported. |
| Missing-data / imputation method | Not reported. |
| Stratification | Not reported. |
| Bayesian methods | Not reported. |
14. Statistical Interpretation of a Single-Group Trial
The most direct estimand for a binary reconnection endpoint would be the probability of observed pulmonary vein reconnection after the specified adenosine administration and assessment procedure in the study population.
Because there is one study arm, the analysis does not estimate a difference between randomized treatment assignments. The observed endpoint frequency therefore should not be described as a relative reduction, increase, or treatment effect versus a control group.
An estimate without a confidence interval gives less information about statistical precision. For a binary endpoint, an exact or otherwise appropriate binomial confidence interval can be considered when the numerator and denominator are available. Those values are not posted on ClinicalTrials.gov for the primary endpoint in the present data.
The clinical meaning of an observed reconnection frequency depends on how the endpoint is defined, how consistently it is measured, the population in which it is measured, and what decision the measurement is intended to inform. Statistical estimation should preserve those distinctions rather than converting the observation into an unsupported comparative claim.
15. Limitations
- No concurrent comparator: the registry-reported design is SINGLE_GROUP with 1 arm and allocation recorded as NA, so a randomized treatment comparison cannot be estimated.
- Primary endpoint result not numerically available in the ClinicalTrials.gov record: the registry data state that results were posted, but no statistical-analysis record or primary-endpoint estimate is reported here.
- Endpoint denominator not established: enrollment is 378.0, but the ClinicalTrials.gov record does not identify the number of participants evaluable for the primary endpoint.
- No confidence interval or p-value: neither is included for the registered primary endpoint in the ClinicalTrials.gov record.
- Secondary outcomes incompletely characterized: 2 outcome measures are reported as posted, but only the primary endpoint's definition and time frame are reported.
- Analysis populations not specified: the ClinicalTrials.gov record does not define ITT, per-protocol, or endpoint-evaluable populations.
- Missing-data methods not specified: no imputation or missing-assessment strategy is provided.
- Safety is descriptive: serious adverse events are reported as 2/378 in the adenosine arm, with no comparator arm posted on ClinicalTrials.gov for relative safety measures.
- Generalizability: the ClinicalTrials.gov record identifies atrial fibrillation and persistent common pulmonary vein but does not provide the detailed eligibility and baseline characteristics needed to define the broader population represented by the trial.
16. Why This Trial Matters Statistically
APPLE is useful as a teaching case because it illustrates an important principle in clinical-trial statistics: The statistical method should follow the design and endpoint. A phase 3 label does not automatically imply that the appropriate analysis is a randomized hazard ratio, risk ratio, or treatment difference.
| Concept | How it appears in APPLE |
|---|---|
| Single-group design | One registered study arm and SINGLE_GROUP design model. |
| Descriptive estimation | The primary endpoint naturally lends itself to reporting the frequency of pulmonary vein reconnection. |
| Binary endpoint | Reconnection can naturally be represented as present versus absent if classified that way for analysis. |
| Confidence interval | Would quantify uncertainty around an estimated endpoint proportion. |
| Null hypothesis | A formal p-value would require a prespecified reference hypothesis, which is not reported. |
| Denominator selection | Enrollment is 378.0, but the endpoint-evaluable denominator is not reported. |
| Safety analysis | Serious adverse events are reported descriptively as 2/378 in the adenosine arm. |
| Endpoint timing | The registered primary endpoint is assessed 5 minutes after IV adenosine. |
| Registry interpretation | Results are posted, but the registry-reported statistical-analysis array contains no formal analyses. |
17. Planned Statistical Interpretation of the Primary Endpoint
Because the ClinicalTrials.gov record confirms that results were posted but does not provide a formal statistical-analysis entry, the appropriate educational analysis is to describe what would normally be done for the registered endpoint without asserting an unreported outcome.
Step 1 · Define the endpoint
Classify whether pulmonary vein electrogram reconnection occurred following the specified adenosine procedure.
Step 2 · Define the analysis set
Specify which enrolled participants have a valid primary-endpoint assessment at the registered time frame.
Step 3 · Estimate the proportion
Calculate the observed proportion with reconnection using the prespecified numerator and denominator.
Step 4 · Quantify uncertainty
Report an appropriate confidence interval for the proportion if required by the statistical analysis plan.
This approach is preferable to introducing a comparison that the trial design does not contain. If the statistical analysis plan specified a particular reference proportion, exact test, or other inferential procedure, that method would need to be documented before a p-value could be interpreted.
18. Statistical Methods Explained: Reading the Safety Count
The serious-adverse-event result provides a useful example of the difference between a count, a proportion, and a comparative effect.
The registry supplies the affected count and at-risk count directly. A descriptive safety proportion could be calculated from these values, but the ClinicalTrials.gov record does not require or support a comparison with another randomized arm.
Even when a denominator is available, the statistical interpretation depends on the endpoint definition. "Serious adverse events" is not interchangeable with all adverse events, treatment discontinuations, or any other safety category. The safest interpretation is therefore to preserve the registry's stated safety category.
19. Sources
- ClinicalTrials.gov: APPLE — NCT01932112. Registry source for the ClinicalTrials.gov record, design, endpoint definition, status, enrollment, and posted-outcome information used on this page.
- Linked publication: PubMed record for PMID 26960977.
Continue with the underlying statistical concepts
Explore the statistical methods that help explain single-group clinical trials, binary endpoints, confidence intervals, and safety-event summaries.
20. Related Tutorials
Learn more about the methods used to understand this trial:
21. Related Calculators
22. Record Summary
APPLE is a phase 3, completed, single-group trial with 378.0 enrolled participants and one registered intervention arm involving adenosine. Its registered primary endpoint is Reconnection of Pulmonary Vein Electrogram After Adenosine Infusion, assessed 5 minutes after IV adenosine. The ClinicalTrials.gov record confirms that results were posted and that 2 outcome measures were posted, but the ClinicalTrials.gov record contains no formal analysis records or numeric primary-endpoint estimate.
The statistical lesson is therefore methodological rather than a fabricated efficacy conclusion. With a single-group design, the natural analysis of a binary reconnection endpoint would generally focus on an observed proportion and its uncertainty. A comparative effect measure would require a comparator, while a p-value would require a prespecified null hypothesis. The safety information is likewise descriptive: the ClinicalTrials.gov record reports 2/378 serious adverse events in the adenosine arm.