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 Comparison of In-House Methods and Cobas BRAF V600 Mutation Assay in Melanoma Tumor Samples (NCT01744860), a registered phase not stated study. The primary outcome is BRAF Mutation Status According to Cobas 4800 BRAF V600 Mutation Test vs. INCa Laboratories Molecular Genetics Laboratories.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01744860 |
| Phase | Phase not stated |
| Condition | Malignant Melanoma |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 420 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Hoffmann-La Roche |
| Results posted | Yes |
2. Clinical Question
Population
The study concerns melanoma tumor samples meeting the inclusion criteria and not meeting the exclusion criteria. The registry lists enrollment of 420, while the brief summary says that no patients will be enrolled and describes data collection from tumor samples. The record does not clarify the relationship between the enrollment entry and the number of samples included in the analysis.
Intervention
The assessment method on one side of the comparison was the Cobas 4800 BRAF V600 mutation test. The record describes BRAF mutation status as being determined using this assay on tumor samples.
Comparator
The other method was the in-house testing used by INCa molecular genetics laboratories. The record also says that the type of V600 mutation was evaluated only by the INCa in-house method, so mutation subtype was not described as a corresponding outcome for both methods.
Primary question
The study asks how BRAF V600 mutation classifications from the two testing approaches compare in a real-life melanoma sample setting. The recorded categories are V600 mutation, no V600 mutation, and non-evaluable. The registry does not report the resulting category counts, a numerical agreement estimate, or a criterion for deciding what degree of agreement would be acceptable.
This is a comparison of laboratory methods, not a randomized comparison of treatments. Its central statistical issue is whether the methods assign matching mutation-status categories to the same eligible tumor samples. Understanding that question requires information about the sample-level results, including how often both methods give the same classification, how often they disagree, and how non-evaluable results are handled. Those results and handling details are not reported in the registry record.
3. Trial Design
This is described as a non-interventional study comparing the Cobas BRAF V600 mutation assay with in-house methods used in molecular laboratories. The material of interest is melanoma tumor samples; the brief summary explicitly says no patients will be enrolled. Accordingly, the named approaches are testing methods, not treatment arms assigned to patients. The registry labels the study arms by method, but does not describe random allocation, and the record warns that the arms are not a like-for-like control comparison.
The study is framed as taking place in a real-life setting and evaluates BRAF mutation status determined with each method. The registry gives a primary outcome timeframe of up to 6 months. It lists a start date of 2012-12, a primary completion date of 2013-04, and a completed status. Data collection is described as lasting approximately 6 months in the brief summary. The record does not explain how these date entries relate to the approximate collection period.
The registry lists enrollment of 420, but the brief summary says that no patients will be enrolled. The record does not clarify whether the enrollment entry refers to samples, another study unit, or a planned value. It therefore should not be treated as a reported number of patients or as a confirmed number of analyzed samples.
The registry describes 29 outcome measures and lists one analysis. It also records secondary outcomes concerning tumor sample characteristics and pre-analytical handling, including sample source, fixation or embedding, time to fixation, fixative, fixation duration, slice thickness, and dewaxing. The protocol details, study design features beyond those stated, and statistical analysis plan 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.
Before assignment: A total 420 melanoma tumor samples were analysed in 12 platform laboratories in France. The paraffin-embedded tumour samples were taken from biopsy or surgical specimen from the internal or external pathology laboratory.
Overall Study
| Stage | Melanoma Tumor Sample With BRAF V600 Mutation |
|---|---|
| STARTED | 420 |
| COMPLETED | 420 |
| NOT COMPLETED | 0 |
Analysis population for the primary outcome, as defined in the registry: All samples meeting the inclusion criteria and not meeting the exclusion criteria were considered for analysis. n = number of samples with BRAF V600 mutation by "in-house method".
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: Biological samples were analysed rather than samples
| Characteristic | Unit | Melanoma Tumor Sample With BRAF V600 Mutation |
|---|---|---|
| Participants analysed | 420 | |
| Age, Categorical: <=18 years | Participants | NA |
| Age, Categorical: Between 18 and 65 years | Participants | NA |
| Age, Categorical: >=65 years | Participants | NA |
| Sex: Female, Male: Female | Participants | NA |
| Sex: Female, Male: Male | Participants | NA |
6. Primary Endpoint
The registry lists the primary outcome as BRAF Mutation Status According to Cobas 4800 BRAF V600 Mutation Test vs. INCa Laboratories Molecular Genetics Laboratories, measured over Up to 6 months.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | number of samples |
| Endpoint classification | not reduced to a single comparison |
The primary outcome concerns BRAF mutation status according to the Cobas 4800 test compared with the status determined by INCa molecular genetics laboratories using in-house methods. It is expressed as a number of samples. For each method, the listed classifications are V600 mutation, no V600 mutation, and non-evaluable. The type of V600 mutation—such as E, K, R, D, E2, another V600 mutation, or not specified—is evaluated only by the in-house method.
This endpoint is relevant because the study is intended to compare the classifications produced by two laboratory approaches on melanoma tumor samples. In a method-comparison study, it is important to know not only how often each method assigns a category, but also whether both methods assign the same category to the same sample. A count of mutation-positive samples from one method alone would not, by itself, describe agreement between methods.
The registry defines the population for analysis as all samples meeting inclusion criteria and not meeting exclusion criteria. It describes the outcome population notation as the number of samples with BRAF V600 mutation by the in-house method. The registry does not report the actual counts, the denominator used for any comparison, a cross-classification of results, or a summary measure of agreement. It also does not state how the non-evaluable classifications enter any agreement calculation. These omissions prevent a numerical interpretation of the primary outcome from the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry identifies a direct laboratory-method comparison, but it does not report the primary outcome results. There are no posted counts for samples classified as having a V600 mutation, as having no V600 mutation, or as non-evaluable by either method. Nor does the record report how often the methods agreed or disagreed on individual samples. Without those results, the degree and pattern of concordance cannot be described quantitatively.
The study’s stated purpose is to evaluate concordance in melanoma tumor samples. Concordance is a comparison of classifications, so interpretation ordinarily depends on the paired outcomes for each sample: whether the two methods agree, and, when they do not, which classifications differ. A total count from only one method cannot establish that pattern. The record does not provide a sample-level cross-tabulation, a concordance percentage, a chance-adjusted agreement statistic, confidence intervals, or a hypothesis test for the primary outcome.
The non-evaluable category matters to interpretation. It is listed as an outcome category, but the record does not state whether such samples are excluded from a concordance denominator, retained as a separate mismatch category, or handled in another way. These choices can change the population represented by an agreement estimate. The registry also does not report the number of samples tested by both methods or the number with results available from each method.
The enrollment entry of 420 should be read alongside the brief summary, which states that no patients will be enrolled. The record does not clarify the unit represented by enrollment or report the analyzed sample count. Thus, the entry cannot resolve the missing denominator for the primary comparison.
Although one analysis is listed, the analytical procedure and its output are not reported in the registry record. No uncertainty interval or precision estimate is available to show how reliably an observed agreement estimate might reflect the samples under study. General statistical interpretation of method comparisons should distinguish the observed level of agreement from its precision, and should consider the distribution of categories and non-evaluable results. In this record, those quantities are not reported.
Planning illustrations are based on the observed effect and assume a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Those illustrations are not observed power and do not replace the trial’s own design rationale. The record does not report the underlying observed effect values or the trial’s sample-size justification. As the outcome estimates themselves are not reported, no numerical conclusion about the assay comparison follows from the registry entry.
9. Statistical Methods Explained
Method-comparison studies ask whether different measurement procedures produce sufficiently similar classifications or measurements on the same material. This is a general statistical explanation. For a categorical outcome such as BRAF V600 status, the key information is the joint classification: the result from the in-house method paired with the result from the Cobas test for each sample. Merely reporting separate totals for each method would not reveal how many samples matched across methods.
What does concordance describe?
Concordance refers to agreement between methods. A simple descriptive approach is to count matching classifications and differing classifications among samples with results from both methods. The specific classification categories in this study are V600 mutation, no V600 mutation, and non-evaluable. Reporting a cross-classification would show whether disagreements occur between mutation and no-mutation results, involve a non-evaluable result, or take another form permitted by the recorded categories. The registry does not provide that cross-classification or state which agreement statistic was used.
A percentage agreement can be useful as a direct description of the proportion of paired results that match. However, its meaning depends on which samples are included and how non-evaluable results are treated. A chance-adjusted agreement measure, such as kappa, is sometimes used in categorical method comparisons, but it is not reported in the registry record. No single agreement measure should be inferred from the study title or purpose.
Why are paired results important?
When two methods assess the same sample, the results are paired rather than independent. The comparison should preserve that pairing. Separate summaries of mutation-positive totals do not show whether the methods classified the same samples as positive. For example, identical totals could arise even if different individual samples were classified as positive. Conversely, unequal totals do not by themselves explain the direction or composition of disagreements. These are general properties of paired categorical data, not reported findings from this study.
For a method comparison, interpretation may also depend on the intended reference standard. The record names the INCa in-house method and the Cobas assay but does not identify either as a definitive reference standard. It therefore does not support treating one method’s classification as truth or describing the other method’s results as diagnostic errors. The record also does not report sensitivity, specificity, predictive values, or a clinical adjudication process.
How should uncertainty be assessed?
An estimate such as agreement calculated from observed samples is subject to sampling uncertainty. A confidence interval is a general way to describe the range of values compatible with the observed data under a specified statistical procedure. It helps distinguish a precise estimate from one that remains uncertain. The registry does not report an agreement estimate, confidence interval, standard error, or the procedure used to quantify uncertainty.
Any uncertainty assessment would need to correspond to the paired design and to the analysis population. If some samples were not evaluable by one or both methods, the analysis should make clear whether those observations are excluded, separately described, or included under a defined rule. The record lists non-evaluable as a category but does not report its frequency or analytical handling. Consequently, neither the effective denominator nor the precision of the comparison can be assessed from the record.
What do the planning assumptions mean?
The planning illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect set as a stated fraction of the observed effect. In general, such calculations describe planning scenarios under their assumptions; they do not measure the probability that the completed study detected an effect. The record says the illustrations are based on the observed effect, but it does not report the effect values or the calculated participant requirements in the registry record for interpretation. It also does not report that these assumptions were the study’s prespecified design or that the study used a randomized allocation. The arms represent testing methods, and the registry states that they are not a like-for-like control comparison.
In general, over-dispersion is relevant when data vary more than a statistical model’s assumed distribution allows, because ignoring it can make uncertainty estimates too optimistic. The primary outcome is described as counts of samples, but the registry does not report a model, a dispersion assumption, or a rationale for using a count model. It would therefore be inappropriate to claim that over-dispersion was present or that a particular adjustment was applied.
The registry lists 29 outcome measures and one analysis. Those entries alone do not identify the statistical method, the contrast evaluated, the handling of repeated or related outcomes, or the uncertainty procedure. The analysis plan, including any prespecified endpoint hierarchy, is not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 29 outcome measures and one analysis. These counts do not establish how many statistical hypotheses were tested or whether all outcomes were analyzed together. The primary outcome concerns BRAF V600 status by the two methods; listed secondary outcomes concern tumor sample characteristics and pre-analytical processing. The record does not describe an outcome hierarchy, a primary analysis model, or whether any secondary outcomes were exploratory.
When multiple outcomes or comparisons are assessed, the chance of at least one apparently unusual result can increase. General approaches to multiplicity include prespecifying a primary comparison, adjusting significance thresholds, or interpreting secondary analyses as exploratory. The registry does not report whether any multiplicity approach was planned or used. No adjustment should be assumed from the count of analyses alone.
The analysis population is described as all samples meeting the inclusion criteria and not meeting the exclusion criteria. Inclusion and exclusion criteria themselves, and the number of samples meeting them, are not reported in the registry record. The brief summary says no patients will be enrolled, while enrollment is listed as 420; the record does not resolve the unit or analyzed denominator.
Non-evaluable is an explicit outcome category, but the record does not state how non-evaluable results or missing test results were handled. It does not report missing-data frequencies, reasons for missing results, exclusions from analysis, imputation, or sensitivity analyses. These details matter because an agreement estimate based only on samples evaluable by both methods could represent a different subset from the full eligible sample population. No conclusion about missingness or its effect on the comparison can be drawn from the registry record.
11. What the Result Does, and Does Not, Mean
The study addresses a relevant statistical question: whether two approaches classify BRAF V600 mutation status consistently in melanoma tumor samples. The registry identifies the methods and the broad outcome categories, but does not report the result counts or any numerical agreement measure. The posted record therefore does not establish how closely the classifications matched, how often they differed, or how precise any observed comparison was.
Even if a numerical agreement estimate were available, agreement would need to be interpreted in context. A general measure of agreement does not by itself establish which method is correct, and it does not necessarily describe clinical performance. That interpretation would depend on a defined reference standard, the distribution of mutation categories, the handling of non-evaluable samples, and the uncertainty around the estimate. The registry does not report those details or the relevant estimates.
The study is non-interventional and compares laboratory methods rather than assigning treatments. It cannot be read as a treatment-effect comparison. The registry’s enrollment entry does not resolve the analyzed sample count, particularly because the brief summary says no patients will be enrolled. The record also does not provide the denominator needed to judge the extent of data available for the comparison.
The registry’s completed status and date entries describe study administration, not the strength or direction of the findings. The primary completion date is 2013-04, and the primary outcome timeframe is up to 6 months, but these registry entries do not supply the missing outcome results. The record supports describing the study question and its planned categories; it does not support a conclusion about whether the methods agreed well or poorly, or whether either method should be preferred.
12. Why This Trial Matters Statistically
Comparisons between laboratory methods raise statistical issues that differ from comparisons of treatments. The central unit is the sample tested by both approaches, and the relevant evidence is the paired pattern of classifications. Separate method-specific totals can conceal mismatches, so a useful report should make clear how often categories align and where they differ.
This study also highlights the importance of defining the analysis population. The primary outcome includes a non-evaluable category, while the registry does not report how those results enter the comparison. Without a clear denominator and rules for incomplete or non-evaluable observations, readers cannot judge which samples an agreement statistic represents. The discrepancy between the enrollment entry and the statement that no patients will be enrolled adds uncertainty about the unit counted.
Pre-analytical factors are also included among the secondary outcomes: sample source and characteristics, fixation or embedding procedures, time to fixation, fixative, fixation duration, slice thickness, and dewaxing. These are relevant descriptors of how samples were handled. The record does not report their observed values or explain whether they were examined as predictors of disagreement, so no relationship between processing and assay classification can be inferred.
For statistical interpretation, a clear account of the paired results, chosen agreement measure, uncertainty, non-evaluable handling, and analysis population would be needed. The registry identifies the comparison but does not report those elements. This makes the record useful for understanding the study question and its scope, while leaving the quantitative evidence needed to assess concordance unavailable.
13. 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 limitation of the registry record is that it does not report the primary outcome counts or an agreement estimate. It also does not provide a paired classification summary, confidence interval, hypothesis test, or details of the statistical method. As a result, the magnitude and precision of concordance cannot be assessed.
The analysis denominator is unclear. Although enrollment is listed as 420, the brief summary says that no patients will be enrolled. The record does not say whether the enrollment entry refers to samples or another unit, and it does not state how many eligible samples were analyzed. The description of the analysis population gives eligibility criteria in general terms but does not report the criteria themselves or the number of samples meeting them.
Non-evaluable is among the primary outcome categories, yet its frequency and handling are not reported. The record does not explain how missing results, samples tested by only one method, or exclusions were handled. It also does not identify a reference standard, define an acceptable level of agreement, or report the method used to quantify concordance.
The record lists 29 outcome measures, but does not describe an outcome hierarchy or multiplicity procedure. It lists one analysis without specifying its statistical model or uncertainty method. The secondary outcomes cover sample and pre-analytical characteristics, but no results or analyses connecting these characteristics to test classifications are reported. These limitations concern the information available in the registry record; they do not establish that the underlying study failed to collect or analyze these data.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Comparison of In-House Methods and Cobas BRAF V600 Mutation Assay in Melanoma Tumor Samples — NCT01744860.
- Registry record last updated 2016-03-28; status last verified 2015-11.
- Results first posted 2016-03-28.
- 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.
17. Record Summary
NCT01744860 is a completed, non-interventional study comparing BRAF V600 mutation classifications from INCa molecular genetics laboratories’ in-house methods with the Cobas 4800 mutation test in melanoma tumor samples. The stated primary outcome is the number of samples classified as having a V600 mutation, having no V600 mutation, or being non-evaluable; mutation subtype is evaluated only by the in-house method.
The registry lists enrollment of 420 and says no patients will be enrolled. It describes a primary outcome timeframe of up to 6 months, with a start date of 2012-12 and primary completion date of 2013-04. The record lists 29 outcome measures and one analysis, but does not report the primary outcome counts, a numerical agreement measure, or the analysis details needed to interpret concordance. The unit represented by enrollment and the handling of non-evaluable or missing results are not reported in the registry record.