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 NBI to Detect Post-RT Mucosal Residual NPC (NCT02584790), a na na study. The primary outcome is Positive Nasopharynx Biopsy Results Detected by NBI System in Those Post-radiotherapy 8th Week NPC Patient.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02584790 |
| Phase | NA |
| Condition | Nasopharyngeal Carcinoma |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 55 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Pamela Youde Nethersole Eastern Hospital |
| Results posted | Yes |
2. Clinical Question
Population
The study concerns patients with nasopharyngeal carcinoma assessed at the post-radiotherapy 8th week. The registry describes a post-radiotherapy 8 weeks NPC patient group. The recorded enrollment is 55, but the available description does not establish that every enrolled participant contributed to the primary endpoint or to the adverse-event summaries.
Intervention
The device is a laryngoscope with a narrow-band imaging (NBI) system. NBI vessel patterns were graded from A through D, from normal vessel size and regular patterns to thickened, elongated, distorted, earthworm-like vessels. The registry describes using NBI during the post-radiotherapy examination.
Comparator
The registered study model is single-group, with allocation recorded as NA and masking recorded as NONE. The arm description also refers to routine laryngoscope examination using a white-light (WL) system and nasopharynx biopsy at the same time. The record does not describe a randomized or like-for-like control comparison, nor does it report a separate comparator group for the primary endpoint.
Primary question
The primary endpoint is the number of participants with positive nasopharynx biopsy results detected by the NBI system at the post-radiotherapy 8th week. The registry describes biopsy as the means of determining whether residual NPC is present. The clinical question is therefore about identifying biopsy-positive cases in this setting, not about comparing treatment effects. The primary result itself is not reported in the registry record.
3. Trial Design
This is a completed, single-group device study of narrow-band imaging for detecting mucosal residual nasopharyngeal carcinoma after radiotherapy. The sponsor is Pamela Youde Nethersole Eastern Hospital. The registry lists the phase as NA, allocation as NA, and masking as NONE. It records enrollment of 55 and gives the study period as 2015-11 to primary completion in 2017-04.
The intervention is a laryngoscope with an NBI system. The arm description concerns patients at the post-radiotherapy 8th week. It says routine examination uses a white-light laryngoscope and that a nasopharynx biopsy is taken at the same time to determine whether residual NPC is present; the study uses the NBI system during the examination. The arm description ends before its full procedure is available, so details beyond that description are not reported in the registry record.
Because the registered model is single-group, the study is not a randomized comparison. The record does not describe an independent control arm or a between-group treatment contrast. NBI vessel patterns were classified into four grades, A through D, with the descriptions ranging from normal, regular patterns to thickened and distorted patterns. The registry lists secondary outcomes concerning residual NPC in NBI suspicious and non-suspicious pattern groups, and the sensitivity, specificity, positive predictive value, and negative predictive value of NBI pattern classification.
The study is marked COMPLETED. Completion status does not itself establish that all planned outcomes are reported. Although the record lists the primary endpoint and secondary outcomes, it does not report their numerical results or explain the statistical analysis used to produce them.
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 | Post Radiotherapy 8 Weeks NPC Patient |
|---|---|
| STARTED | 55 |
| COMPLETED | 40 |
| NOT COMPLETED | 15 |
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 | Post Radiotherapy 8 Weeks NPC Patient |
|---|---|---|
| Participants analysed | 55 | |
| Age, Customized: < 18 years old | Participants | 0 |
| Age, Customized: >= 18 years old | Participants | 55 |
| Sex: Female, Male: Female | Participants | 8 |
| Sex: Female, Male: Male | Participants | 47 |
| Race and Ethnicity Not Collected | Participants | — |
| Region of Enrollment: Hong Kong | participants | 55 |
| NPC patients | Participants | 40 |
6. Primary Endpoint
The registry lists the primary outcome as Positive Nasopharynx Biopsy Results Detected by NBI System in Those Post-radiotherapy 8th Week NPC Patient, measured over Post-radiotherapy 8th week.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is titled “Positive Nasopharynx Biopsy Results Detected by NBI System in Those Post-radiotherapy 8th Week NPC Patient.” Its unit is participants, its parameter is NUMBER, and its timeframe is the post-radiotherapy 8th week. In practical terms, it is intended to count participants whose nasopharynx biopsy was positive at that assessment.
The endpoint uses biopsy results as the reference for whether residual NPC is present, while NBI provides the vessel-pattern assessment being evaluated. This distinction matters: the endpoint is not simply a count of suspicious NBI appearances. It is specifically framed as positive biopsy results detected in the NBI examination context. The registry description says biopsy is taken to determine whether residual NPC is present.
The record describes four vessel-pattern grades. Grade A is normal in vessel size and length with a regular pattern; Grade B has normal vessel size with a short, regular spiderweb-like pattern; Grade C has irregular vessel size and length with a distorted, irregular pattern; Grade D has thickened vessel size and an elongated, distorted, earthworm pattern. It does not state in the primary endpoint description how those grades were grouped to define a positive NBI finding. The primary count and its denominator are not reported in the registry record, so the endpoint’s observed value cannot be interpreted from the record alone.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a clinically focused diagnostic question: whether NBI assessment at the post-radiotherapy 8th week can identify mucosal residual NPC, with biopsy results used to determine residual disease. The listed primary endpoint is a count of participants with positive nasopharynx biopsy results detected by NBI. However, the record does not report the count, denominator, or any estimate of diagnostic performance. It is therefore not possible to quantify the observed frequency of positive biopsies or the precision of that estimate from the registry record.
The secondary outcomes specify comparisons between NBI suspicious and non-suspicious pattern groups and measures of sensitivity, specificity, positive predictive value, and negative predictive value. These would address distinct aspects of diagnostic classification. Sensitivity describes how often a test identifies disease among those who have it; specificity describes how often it classifies as negative those who do not have it. Predictive values describe the probability of disease or no disease given a positive or negative test classification, and are affected by the prevalence of disease in the assessed population. These are general statistical definitions. None of the corresponding numerical estimates or uncertainty intervals is reported in the registry record.
The NBI grades are described, but the record does not specify the exact rule that maps grades to suspicious versus non-suspicious categories. Without that classification rule and the outcome counts, readers cannot reconstruct diagnostic accuracy measures or assess how borderline patterns were handled. The biopsy procedure is described as occurring at the same time as the routine WL examination; details on blinding of biopsy interpretation or the order and independence of assessments are not reported in the registry record.
There is no randomized comparator, and the study is registered as single-group. Consequently, the design as described does not estimate a causal treatment effect or establish whether NBI is better than another diagnostic strategy. Even if diagnostic-performance results were available, their interpretation would depend on the study population, the biopsy reference standard, the pattern classification, and the completeness of outcome ascertainment. The registry record does not report enough information to evaluate those features fully.
The record also does not report confidence intervals, a p-value, or a statistical test for the primary endpoint. A confidence interval, as a general concept, expresses the uncertainty around an estimated quantity under the assumptions of the analysis; it would help show how precisely a diagnostic-performance measure or event proportion was estimated. With neither an endpoint estimate nor an interval reported, no conclusion about precision can be drawn. The registered completion status and enrollment figure do not substitute for outcome results, and no clinical conclusion should be inferred from them.
9. Statistical Methods Explained
The registry identifies the study as single-group, with allocation recorded as NA and masking recorded as NONE. It lists one analysis, but does not report the statistical methods used for that analysis. In particular, the registry record does not report the analysis population, handling of missing observations, the statistical test, confidence-interval method, or any adjustment for covariates. The following concepts explain how diagnostic studies of this kind are commonly interpreted; they are general explanations, not claims about methods used in this study.
What does the primary endpoint count?
The primary endpoint is a participant count for positive nasopharynx biopsy results at the post-radiotherapy 8th week. A count alone does not communicate the proportion of assessed participants with a positive result unless its denominator is also known. The registry record does not report either the endpoint count or its analysis denominator. Enrollment is listed as 55, but enrollment should not automatically be treated as the number assessed for this endpoint: the record does not state how many participants completed the assessment, had evaluable biopsy results, or entered the primary analysis.
For a diagnostic endpoint, it is also important to distinguish the test result from the reference-standard result. Here the endpoint title refers to positive biopsy results detected by NBI, while the study description says biopsy determines whether residual NPC is present. The record does not give a participant-level classification or cross-tabulation of NBI findings against biopsy results. Without those paired classifications, diagnostic accuracy cannot be derived.
How are diagnostic measures interpreted?
In general, sensitivity is the fraction of people with the target condition who have a positive test result; specificity is the fraction without the condition who have a negative result. Positive predictive value is the fraction of positive test results that correspond to the condition, while negative predictive value is the fraction of negative results that correspond to its absence. These measures answer different questions and depend on the classification threshold. Predictive values also depend on the frequency of disease in the population under study. They should not be treated as interchangeable summaries.
The registry lists sensitivity, specificity, positive predictive value, and negative predictive value as secondary outcomes, but reports no values or intervals for them. It describes vessel grades A through D and separately refers to suspicious and non-suspicious pattern groups. The rule assigning grades to those groups is not reported in the registry record. That rule is essential to reproduce the diagnostic classification and understand how each accuracy measure was defined.
Why does uncertainty matter?
An estimate based on a limited number of assessed participants may be imprecise, particularly when an outcome is uncommon or when few participants fall into a relevant diagnostic category. A confidence interval is a general way to represent uncertainty around an estimate. Its width depends on the amount and distribution of information and on the method used to calculate it. The registry record does not report confidence intervals for the primary or secondary outcomes, nor enough outcome data to assess precision independently.
For diagnostic measures, uncertainty can differ across sensitivity, specificity, and predictive values because each measure is based on different subsets of the diagnostic classifications. A complete analysis would ordinarily make clear which participants contribute to each calculation and how indeterminate or missing results are handled. Those details are not reported in the registry record. No precision statement should therefore be inferred from the listed enrollment alone.
What does the planning illustration mean?
The stated planning assumptions describe an illustration based on a fraction of an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The record characterizes these as planning assumptions rather than the trial’s own design. The trial itself is registered as single-group, so the equal-allocation, two-arm illustration is not a description of its enrolled design. A sample-size calculation is a planning exercise about what a future study might require under specified assumptions; it is not observed power and does not establish the precision or reliability of this study’s result.
The registry record does not report the effect estimate underlying that illustration, its numerical planning outputs, or a sample-size justification for the completed study. It also does not report a planned statistical analysis in enough detail to assess whether the analysis matched the design. The planning assumptions should therefore be kept separate from evidence about observed diagnostic performance.
Why does over-dispersion matter?
Over-dispersion is a general concern when counts vary more than a chosen statistical model expects. If a count model assumes less variability than is present, uncertainty can be understated and tests may appear more decisive than warranted. Whether this issue applies depends on the outcome structure and model. The registry record does not identify a count model or describe an assessment of over-dispersion, so it is not possible to say whether it was relevant or addressed in this analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary endpoint and secondary outcomes concerning the association of residual NPC with NBI suspicious and non-suspicious pattern groups, as well as sensitivity, specificity, positive predictive value, and negative predictive value. It also records three outcome measures and one analysis. The statistical plan for those outcomes, including whether any were designated exploratory, is not reported in the registry record.
When several outcomes or statistical tests are examined, the chance of at least one apparently unusual result can increase. Multiplicity procedures are general methods for controlling that risk, or analyses may instead clearly distinguish a primary confirmatory question from exploratory questions. The registry record does not report whether multiplicity was addressed, whether confidence intervals were adjusted, or which analyses were considered primary beyond the named primary endpoint. No inference about adjustment should be made.
The record does not describe missing-data handling, exclusions, or an analysis population. It does not say how participants with missing NBI assessments, unavailable biopsy results, or indeterminate classifications were handled. It also does not report whether all enrolled participants were assessed at the intended post-radiotherapy 8th week. These omissions matter because diagnostic measures can change if participants with incomplete or uninterpretable results are excluded.
The enrollment figure and the adverse-event at-risk counts refer to different registry entries and should not be assumed to describe identical analysis populations. The record does not explain the relationship between them. A participant flow description, numbers assessed for each endpoint, and explicit missing-data rules are not reported in the registry record.
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 |
|---|---|---|---|
| Post Radiotherapy 8 Weeks NPC Patient | 0 | 40 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Post Radiotherapy 8 Weeks NPC Patient | 0 of 40 |
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 |
|---|---|
| Post Radiotherapy 8 Weeks NPC Patient | 0 of 40 |
The adverse-event listing gives a group titled Post Radiotherapy 8 Weeks NPC Patient. For serious adverse events, the at-risk count is 40 and the affected count is 0. For other adverse events, the at-risk count is 40 and the affected count is 0. The recorded death count is 0. These are the safety figures available in the registry record.
These entries should be read as registry counts, not as proof that no adverse events occurred in every setting or outside the recorded observation. The record does not report the safety observation period, event definitions, ascertainment procedures, or whether all enrolled participants were included in safety follow-up. The adverse-event at-risk count should not be substituted for the enrollment figure when interpreting the primary endpoint.
The intervention is a device-based examination, but the registry safety summary does not provide event descriptions or a causal assessment. Zero affected events in the listed fields does not establish the absence of unrecorded or unreported events. Nor does the safety listing address diagnostic performance. It is a separate description of reported serious events, other events, and deaths in the named group.
12. What the Result Does, and Does Not, Mean
The registry record establishes that a completed, single-group study evaluated an NBI laryngoscope examination in patients with NPC at the post-radiotherapy 8th week. Its primary endpoint concerns positive nasopharynx biopsy results, and its secondary outcomes concern NBI pattern groups and diagnostic-performance measures. The record supplies the clinical framework and outcome definitions, but not the numerical primary or secondary results.
Accordingly, the record does not establish how many participants had positive biopsy results, how frequently each NBI grade occurred, or how accurately suspicious and non-suspicious patterns classified biopsy status. It does not provide a sensitivity, specificity, predictive value, uncertainty interval, or statistical test result. Without these, the magnitude and precision of the observed diagnostic findings cannot be evaluated from the registry record.
The design also limits what type of conclusion could be drawn. A single-group diagnostic study is not a randomized comparison of interventions, and the registry does not describe a like-for-like control group. It therefore does not answer a causal question about whether one treatment produces better outcomes than another. The record’s description of routine WL examination and biopsy does not by itself establish a comparative analysis between WL and NBI.
Even if the endpoint results were reported, interpretation would require clarity about the NBI classification threshold, how biopsy findings were paired with NBI patterns, and how missing or indeterminate observations were handled. Those methods are not reported in the registry record. The listed adverse-event counts are separate safety information and do not fill the gap in diagnostic outcome reporting.
The careful conclusion is limited: the registry documents a study designed to assess NBI vessel patterns in this post-radiotherapy setting, but the available record does not report the outcome estimates needed to judge observed diagnostic performance. Neither a positive nor a negative conclusion about NBI’s performance can be drawn from the endpoint title, completion status, enrollment figure, or absence of reported adverse events alone.
13. Why This Trial Matters Statistically
This study raises a statistical question that is different from the familiar treatment-versus-control comparison: how well does an imaging classification identify a condition when biopsy is used to determine disease status? That requires paired information. Each participant’s NBI classification must be considered alongside the biopsy result, with clear definitions of positive and negative classifications. The registry identifies the intended diagnostic measures but does not report the paired results.
The four-grade description also illustrates why classification rules matter. Vessel patterns are described from normal and regular appearances through irregular, distorted, thickened patterns, while the secondary outcomes refer to suspicious and non-suspicious groups. A reproducible rule connecting these categories is necessary for interpreting sensitivity and specificity and for allowing another analyst to repeat the calculations. That rule is not reported in the registry record.
Precision is another central issue. Diagnostic estimates can be unstable when relatively few observations inform one part of the classification, and a point estimate alone would not show that uncertainty. Confidence intervals and denominators would help readers judge how much information supports each measure. The record reports neither the estimates nor their uncertainty, so the statistical strength of the findings cannot be assessed.
Finally, a single-group diagnostic evaluation should not be mistaken for evidence from a randomized comparison. The record’s registered design defines the scope of possible conclusions. Its completed status, enrollment, and safety entries are informative descriptors, but they cannot replace the missing diagnostic results. This distinction is important for researchers who use registry records to evaluate both study design and reporting completeness.
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 main trial-specific limitation of the available registry record is the absence of numerical primary and secondary outcome results. The primary endpoint is defined, but its count and denominator are not reported. The record also does not report the NBI grade distribution, biopsy classifications, or the listed diagnostic-performance measures. It is therefore not possible to evaluate the observed findings or their precision.
The description of the study procedure is incomplete: the arm description ends mid-sentence. It mentions routine WL laryngoscopy and concurrent nasopharynx biopsy, along with use of NBI, but does not fully document the examination process. The exact rule for combining the four NBI grades into suspicious and non-suspicious groups is not reported. The registry record also does not report how biopsy interpretation was conducted or whether readers were blinded to NBI findings.
Important analysis details are absent, including the statistical test or estimation method, confidence intervals, missing-data rules, and analysis population. The record does not explain whether participants with incomplete assessments or indeterminate results were included. It does not report a participant flow connecting enrollment of 55 with endpoint assessment or with the adverse-event group, for which the at-risk counts are 40.
The single-group design and lack of a like-for-like control comparison limit the scope of interpretation. The registry does not report a comparative diagnostic analysis against WL, despite mentioning WL as part of routine examination. Finally, it does not report multiplicity handling or a complete statistical plan. These gaps restrict assessment of reproducibility and certainty without implying that any unreported procedure was or was not performed.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: NBI to Detect Post-RT Mucosal Residual NPC — NCT02584790.
- Registry record last updated 2017-12-08; status last verified 2017-10.
- Results first posted 2017-10-20.
- 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 for NCT02584790, titled “NBI to Detect Post-RT Mucosal Residual NPC,” describes a completed, single-group device study sponsored by Pamela Youde Nethersole Eastern Hospital. It records enrollment of 55, phase NA, allocation NA, and masking NONE. The study period is listed as 2015-11 to primary completion in 2017-04.
The intervention is a laryngoscope with an NBI system used in patients at the post-radiotherapy 8th week. The primary endpoint is positive nasopharynx biopsy results detected by NBI, measured in participants. Secondary outcomes concern residual NPC by NBI pattern grouping and diagnostic-performance measures. Numerical outcome results and key analysis details are not reported in the registry record.
The adverse-event listing records 0 serious events, 0 other events, and 0 deaths, with 40 at risk for each event category. The record does not report enough information to connect those counts to the primary analysis population or to assess the observation period.