Proof of Concept Study Concerning Efficacy of P03277 MR Imaging in HCC Diagnosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Proof of Concept Study Concerning Efficacy of P03277 MR Imaging in HCC Diagnosis (NCT02973516), a na phase 2 study. The primary outcome is Diagnostic Value Evaluation for HCC.

2. Clinical Question

The study asked whether P03277 triphasic liver imaging could help identify hepatocellular carcinoma (HCC) in people with chronic liver disease and suspected small liver nodules. Its stated diagnostic measures were sensitivity and specificity, assessed against a site-determined standard of reference. The registry describes an exploratory proof-of-concept study, rather than a randomized comparison of imaging strategies.

3. Trial Design

This was a phase IIa clinical study, described as a proof-of-concept and exploratory evaluation of P03277 MR imaging for HCC diagnosis. The registry identifies a single-group model, no masking, and allocation as not applicable. The study therefore was not a randomized, blinded comparison between imaging and a control treatment. The interventions listed are P03277 at 0.1 mmol/kg and P03277 at 0.05 mmol/kg; the arm description says P03277 was administered to acquire triphasic liver imaging.

The brief study description planned to include 30 subjects with suspected small nodules, with HCC confirmed or not by previous enhanced CT and/or MRI, before any biopsy for histology analysis. It also describes 10 additional subjects for exploration of diagnostic efficacy at half dose. The registry reports enrollment of 41. It does not explain the relationship between the planned numbers and the reported enrollment, nor does the primary outcome description fully specify how participants were allocated to the listed doses.

The reference for diagnosis was determined by each site according to its standard of care and adapted from EASL/EORTC diagnostic criteria. The possible information sources included prior contrast-enhanced imaging, biopsy analysis performed on-site, and the most recent available AFP results. The registry does not report whether the reference process was applied identically to every nodule or whether its assessors were masked to the P03277 imaging findings.

The primary outcome timeframe was a one-day procedure. The primary analysis was at the nodule level: FASN included 56 suspected nodules, with 43 from cohort 1 and 13 from cohort 2. Thus the nodule count is not interchangeable with the participant enrollment count. The registry does not report details of within-participant clustering or how any such dependence was handled in the analysis.

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: Among the 41 screened patients, one failed to meet randomization criteria and was screen failure. Therefore, 40 patients were included: 30 in Cohort 1 (0.1 mmol/kg) and 10 in Cohort 2 (0.05 mmol/kg),

Overall Study

Analysis population for the primary outcome, as defined in the registry: Analysis performed on the Full Analysis Set Nodules (FASN). A total of 56 suspected nodules were characterized: 43 from cohort 1 and 13 from cohort 2.

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: Analysis was performed for the Full Analysis Set (FAS). One patient from cohort 1 was not included in the FAS for not presenting any nodule at the time of inclusion.

6. Primary Endpoint

The registry lists the primary outcome as Diagnostic Value Evaluation for HCC, measured over One day procedure.

The primary outcome was the diagnostic value of imaging for HCC, expressed as sensitivity and specificity of diagnosis obtained by gadopiclenol imaging versus the standard of reference. Sensitivity describes how often a diagnostic test identifies disease among cases classified as having the disease by the reference. Specificity describes how often it classifies as negative those cases classified as not having the disease by the reference. These are distinct aspects of diagnostic performance, so one cannot stand in for the other.

Both measures are relevant because a diagnostic method can perform differently in recognizing reference-positive and reference-negative cases. Their interpretation depends on the definition and quality of the reference classification, as well as the analyzed cases. In this study, the reference was site-determined and adapted from diagnostic criteria, with possible use of prior imaging, biopsy analysis, and AFP results. The registry does not report a single uniform reference procedure or enough detail to assess how reference classifications were established for every nodule.

The outcome population was FASN, described as 56 suspected nodules, including 43 from cohort 1 and 13 from cohort 2. The registry identifies the endpoint and population but gives no numerical sensitivity or specificity estimates, confidence intervals, or counts of reference-positive and reference-negative nodules. Consequently, the posted outcome description identifies what was intended to be evaluated but does not allow the diagnostic estimates or their precision to be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Diagnostic studies compare a test classification with a reference classification. In general, sensitivity concerns identification of reference-positive cases, while specificity concerns identification of reference-negative cases. These measures address different kinds of classification and should be interpreted together when both are reported. In this study, the registry names both measures as primary diagnostic outcomes but does not report their estimates, confidence intervals, or the counts used to calculate them.

What does the reference standard represent?

A reference standard is the basis against which the index test is evaluated. It may be a definitive procedure, a set of criteria, or a combination of information sources. Here, the registry says the site determined the reference according to its standard of care, adapted from EASL/EORTC diagnostic criteria. Possible inputs included prior contrast-enhanced CT and/or MRI, on-site biopsy specimen analysis, and the most recent AFP results available. The record does not report the exact reference components used for each nodule or whether the same process was applied to all cases. If reference classifications vary in how they are established, the resulting diagnostic measures may also depend on those choices.

Masking can reduce the risk that knowledge of one result influences the interpretation of another. The registry lists masking as none. It does not say whether readers of P03277 images knew the reference classification, whether reference assessors knew the P03277 findings, or how diagnostic readings were adjudicated. Those procedural details are not reported in the registry record, so the possible influence of interpretation is difficult to assess.

Why does the analysis unit matter?

The primary analysis population is defined as FASN and contains 56 suspected nodules: 43 from cohort 1 and 13 from cohort 2. The registry separately reports enrollment of 41 participants. A nodule-level analysis can include more than one nodule from a participant. In general, measurements from the same participant may be correlated rather than independent. If that occurs, methods that treat every nodule as independent may understate uncertainty. The registry record does not report how many participants contributed multiple nodules or whether the analysis accounted for clustering.

The name Full Analysis Set Nodules is given, but the registry does not report its detailed inclusion rules, exclusions, or the number of nodules with complete information for each diagnostic measure. It also does not describe how missing imaging, missing reference classifications, or indeterminate results were handled. These details are important because the analyzed cases determine which nodules contribute to sensitivity and specificity.

How should confidence intervals be understood?

In general, a confidence interval expresses statistical uncertainty around an estimate under the assumptions of the analysis method. A narrower interval indicates greater precision under those assumptions; a wider interval indicates less precision. For diagnostic measures, interval estimates help distinguish a point estimate from the range of values compatible with the observed data and method. No confidence intervals are reported for the primary diagnostic outcome in the registry record, so precision cannot be evaluated from the posted results.

What do the sample-size illustrations mean?

General sample-size illustrations can show how many participants per arm might be needed under specified planning assumptions. The registry’s described illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. For time-to-event calculations, the illustration concerns a total number of events; for other calculations, it concerns participants per arm. These are planning scenarios, not observed power and not evidence that the study was designed around those assumptions.

The trial itself is single-group and not a randomized two-arm comparison, so an equal-allocation comparison is not a description of its design. The record does not report the study’s own sample-size justification or protocol assumptions. It also does not report a numerical observed diagnostic effect from which to interpret the planning illustrations. Accordingly, those illustrations cannot be used to infer what effect the trial was adequately powered to detect.

How should safety counts be interpreted?

Safety counts report the number of participants with an event among those at risk in a cohort. Serious adverse events are distinct from other adverse events, and event counts are not necessarily the same as counts of affected participants. The record gives affected-participant counts for serious and other events, plus deaths, and separately names two serious-event terms with percentages. Without additional detail, the named terms cannot be assigned to a particular cohort or interpreted as proof of causation. Descriptive counts in small cohorts also do not establish comparative safety.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies sensitivity and specificity as diagnostic outcomes, but it does not report a statistical analysis plan describing how they were tested or interpreted together. It does not report multiplicity adjustments, a hierarchy of outcomes, or how conclusions would be drawn if the measures differed. The record also gives no numerical outcome estimates or analysis results. Any procedure for controlling the chance of false-positive findings across multiple analyses is therefore not reported in the registry record.

Missing-data methods are also not described. The primary population is FASN, comprising 56 suspected nodules, but the record does not give the number with complete imaging and reference information for each endpoint. It does not state how indeterminate imaging interpretations, unavailable reference information, or missing values were treated. Nor does it report a participant or nodule disposition accounting that would show whether cases were excluded from the diagnostic analysis.

The registry states that the reference standard could draw on site-standard care adapted from diagnostic criteria, including prior imaging, biopsy analysis, and recent AFP results. It does not report whether every nodule had each type of information or how differences in available information were managed. This matters because missingness in the index test or reference classification can affect which cases contribute to diagnostic estimates.

The record names the analysis population but does not provide the detailed analysis rules, any sensitivity analyses, or the handling of multiple nodules per participant. It also does not report whether diagnostic outcomes were analyzed separately by dose or cohort. These omissions limit assessment of the robustness of the primary results, beyond the more fundamental limitation that numerical sensitivity and specificity results are not reported.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: gastric perforation (10.0%), procedural complication (10.0%).

The registry reports serious adverse events for two cohorts. In cohort 1, 0 of 30 participants at risk were affected by a serious adverse event; in cohort 2, 1 of 10 were affected. Deaths were reported as 0 in each cohort. These are descriptive counts, not estimates from a randomized safety comparison. The study was single-group and unmasked, and the record does not establish that the cohorts are comparable.

Other adverse events are reported separately: 8 of 30 participants at risk in cohort 1 and 0 of 10 in cohort 2 were affected. These figures should not be combined with the serious-event counts or treated as interchangeable categories. The registry lists gastric perforation and procedural complication among the top serious-event terms, each with 10.0%. It does not provide enough detail in this record to assign those terms to a cohort or infer a causal relationship with P03277 or the imaging procedure.

Small cohort counts make safety patterns difficult to characterize precisely. The record does not report event timing, duration, severity beyond the serious classification, or attribution. The figures therefore describe what was recorded, but do not establish comparative risk or causation.

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

This study addresses a diagnostic question: how P03277 triphasic liver imaging classified suspected nodules relative to a site-determined standard of reference. The registry identifies sensitivity and specificity as the primary measures and describes an analysis of 56 nodules. However, it does not report numerical estimates for either measure or their uncertainty. Readers cannot determine the observed diagnostic performance or how precisely it was estimated from the posted outcome information.

The design also matters to interpretation. The registry describes a phase IIa, exploratory, single-group study with no masking and no randomized allocation. It lists two P03277 doses, but does not report a randomized control comparison or numerical outcomes by dose. The presence of two dose entries should not be read as evidence that a controlled dose comparison was conducted.

The reference classification was determined at the site using its standard of care, adapted from diagnostic criteria, and could incorporate different types of information. This makes the reference clinically contextual, but the record does not specify the exact evidence used for each nodule. The absence of details on assessor masking, missing data, and handling of multiple nodules also limits assessment of bias and precision.

The safety counts describe adverse events recorded in the two cohorts, including serious events, other events, and deaths. They do not show whether events were caused by the study procedure, and the cohort sizes do not support a strong comparative safety interpretation. The named serious-event terms are not linked to a particular cohort in the registry information.

Accordingly, the record supports understanding the study’s diagnostic aim, broad design, analysis population, reference approach, and selected safety counts. It does not establish that P03277 imaging is accurate or inaccurate, and it does not support a treatment recommendation. The numerical diagnostic findings needed for that assessment are not reported in the registry record.

13. Why This Trial Matters Statistically

Diagnostic evaluations depend on more than a test name and a point estimate. Sensitivity and specificity answer different questions, and both require clear definitions of the cases used in their denominators. This registry describes both measures as primary outcomes, specifies a nodule-level analysis population, and explains that the reference diagnosis could draw on several sources. Those details identify important features a reader would need when judging a diagnostic study.

The record also illustrates why the unit of analysis matters. It reports 41 enrolled participants and describes 56 suspected nodules in the primary analysis population. These counts refer to different units. If participants contributed multiple nodules, those observations may be related; the record does not report whether this occurred or how it was addressed statistically. Reporting such details is important for understanding the precision of diagnostic estimates.

Reference standards are another central issue. A site-determined standard adapted from criteria may reflect usual diagnostic practice, but the record does not specify the evidence used for each nodule. The mix of prior imaging, biopsy information, and AFP results may affect how classifications are established. Clear reporting of reference procedures and assessor masking helps readers evaluate potential sources of bias.

Finally, this record shows the distinction between a stated endpoint and a reportable result. It names sensitivity and specificity but does not provide their values or uncertainty. Without those results, the statistical question is identifiable, but its empirical answer cannot be assessed. The study’s exploratory, single-group design further means that the listed dose interventions should not be mistaken for a randomized control comparison.

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 registry record does not report numerical sensitivity or specificity estimates, confidence intervals, or the classification counts underlying the primary diagnostic outcome. This prevents direct evaluation of diagnostic performance and its precision. Although it identifies 56 suspected nodules in FASN, it does not describe the number classified as HCC or not HCC by the reference standard.

The reference was determined at the site according to standard care and adapted from diagnostic criteria. Prior contrast-enhanced imaging, biopsy analysis, and recent AFP results could contribute, but the record does not specify which sources applied to each nodule or whether the process was consistent across cases. It also does not report masking of image readers or reference assessors.

The record reports enrollment of 41 participants and an analysis population of 56 nodules. It does not explain how nodules were distributed across participants, whether any participant contributed more than one nodule, or how within-participant dependence was handled. Detailed inclusion rules, exclusions, missing-data methods, and case disposition are not reported. This limits assessment of the analysis population and the validity of uncertainty estimates.

The study was single-group, unmasked, and not randomized. Although the registry lists two doses and describes additional subjects for half-dose exploration, it does not report dose-specific diagnostic results or a randomized dose comparison. The registry also does not report the study’s sample-size justification, analysis plan, or multiplicity procedures. These design and reporting limitations should be considered alongside the absence of numerical primary results.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Proof of Concept Study Concerning Efficacy of P03277 MR Imaging in HCC Diagnosis — NCT02973516.
• Registry record last updated 2021-10-26; status last verified 2021-10.
• Results first posted 2021-10-06.
• 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.

18. Record Summary

NCT02973516 is a completed phase IIa, exploratory study sponsored by Guerbet, with a reported enrollment of 41. The record lists a start date of 2016-12 and primary completion on 2019-04-04. It evaluates P03277 triphasic liver imaging for HCC diagnosis in people with chronic liver disease and suspected nodules.

The study is described as single-group, with no masking and allocation not applicable. The listed interventions are P03277 at 0.1 mmol/kg and P03277 at 0.05 mmol/kg. The primary diagnostic outcome is sensitivity and specificity versus a site-determined standard of reference. Its analysis population contains 56 suspected nodules: 43 from cohort 1 and 13 from cohort 2.

Numerical sensitivity and specificity results are not reported in the registry record. Safety information includes serious and other adverse-event counts and zero deaths in each cohort, but does not establish causation or comparative risk.