Serial Ultrasound in Metastatic Renal Cell Carcinoma (mRCC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Serial Ultrasound in Metastatic Renal Cell Carcinoma (mRCC) (NCT04508725), a non-randomized na study. The primary outcome is Initial Objective Response- First Participation.

2. Clinical Question

3. Trial Design

This is a completed, nonrandomized, single-group study sponsored by Stanford University. The registry lists 22 participants enrolled. The study began on 2020-12-05, and its primary completion date was 2023-11-30. The listed phase is NA. Although the model is recorded as single group, the intervention descriptions identify treatment categories, including a TKI plus ICI and non-ICI therapy, as well as a category for participants enrolled consecutively in both arms.

The central study question concerns whether early changes in tumor perfusion measured by power Doppler ultrasound can predict initial objective response. The stated assessment times for ultrasound changes are after 3 or 6 weeks of treatment. Response was defined by RECIST v1.1 and assessed at the first on-treatment evaluation, described as 8-16 weeks after treatment initiation. The primary outcome timeframe is listed as 12 weeks.

Because treatment allocation was not randomized and masking was none, treatment groups should not be read as interchangeable comparison groups. Treatment choice and participant characteristics may differ across categories, and the registry does not describe a randomized assignment process. In particular, the category of participants enrolled consecutively in both arms is not a conventional independent treatment arm. The registry itself notes that the arms are not a like-for-like control comparison.

The record lists Doppler ultrasound as a diagnostic test, the ultrasound devices, and standard-of-care treatment options. It does not report a protocol-level account of how treatment selection was made, whether ultrasound measurements changed clinical decisions, or how the predictive relationship was analyzed. Those details 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 of 22 participants were enrolled in the study. Twelve participants started in Arm 1 and five participants started in Arm 2. Additional two participants consecutively participated in both arms; these participants are reported in Arm 3. Three participants were enrolled but did not start treatment and are not included in any arm.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Two of fourteen participants in Arm 1 were not evaluable because they were removed from the study during active participation without undergoing any response evaluation.

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.

6. Primary Endpoint

The registry lists the primary outcome as Initial Objective Response- First Participation, measured over 12 weeks.

The primary endpoint is initial objective response at the first participation, measured as a count of participants. Response means a complete response or partial response according to RECIST v1.1 at the first on-treatment response evaluation. The registry describes that evaluation as taking place 8-16 weeks after treatment began and lists the endpoint timeframe as 12 weeks.

This endpoint supplies a clinical response classification against which early ultrasound measurements could be assessed. The study’s stated aim is to evaluate whether changes in quantitative tumor perfusion after 3 or 6 weeks predict that later response classification. In general, a binary response endpoint reduces a complex clinical assessment to categories that can be compared with an earlier measurement, but the usefulness of that comparison depends on the response definition and on complete, clearly reported assessments.

The registry record says that two of fourteen participants in Arm 1 were not evaluable because they were removed during active participation without undergoing any response evaluation. It does not report the number with complete response, partial response, or no objective response, nor does it report a predictive association between ultrasound change and response. Consequently, the endpoint’s observed result cannot be described from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the endpoint count?

The primary endpoint is a count of participants meeting the stated objective-response definition. Complete response and partial response are grouped together as response; the registry does not report their separate counts or the count without objective response. The endpoint is tied to the first participation and first on-treatment response evaluation. Although the timeframe is listed as 12 weeks, the description gives an evaluation window of 8-16 weeks after treatment starts. Those statements describe the endpoint but do not give its observed result.

What would a predictive analysis address?

General statistical context: a predictive analysis asks whether a measurement available earlier can help distinguish later outcome categories. Here, the intended early measurements are changes in quantitative tumor perfusion on power Doppler ultrasound after 3 or 6 weeks, and the later outcome is objective response. An analysis might assess whether ultrasound measurements differ across response categories or whether they classify response with useful accuracy. The registry record does not report which predictive measures, thresholds, models, or validation procedures were used, so no such method should be attributed to the study.

Why does allocation matter?

General statistical context: random allocation helps balance measured and unmeasured participant characteristics across treatment groups on average. In a nonrandomized study, treatment categories may differ for reasons related to treatment choice or participant characteristics. A raw difference between categories can therefore combine treatment effects with those other differences. This record is explicitly nonrandomized and notes that the arms are not a like-for-like control comparison. It does not report an adjustment method or a causal comparison that would resolve this issue.

What does masking contribute?

The registry records masking as none. General statistical context: masking can reduce the influence of knowledge about treatment or outcome on measurement and assessment. Its relevance depends on which people are masked and how outcomes are determined. The record does not give further details about masking procedures or who performed ultrasound measurements and response assessments. The absence of masking information beyond “none” should not be replaced by assumptions about the conduct of assessments.

How should counts and uncertainty be read?

General statistical context: a count describes how many participants meet a definition, while a confidence interval communicates uncertainty around an estimated quantity. Counts alone do not show the precision of a response rate or predictive association, and a count cannot establish that an ultrasound measure predicts response. The registry record reports no primary response count, effect estimate, or confidence interval. It also lists zero analyses. Accordingly, inferential conclusions and precision cannot be reconstructed from the reported endpoint description.

What is the role of the sample-size illustration?

The planning illustration is described as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect. Such planning quantities are illustrative rather than observed evidence and are not observed power. The record does not report the trial’s own sample-size justification. No numerical planning sample sizes are stated in the registry record, so they cannot be discussed as calculated values.

What would over-dispersion mean in general?

General statistical context: over-dispersion occurs when the variability in data is greater than a model assumes. If it is ignored in a setting where it is present, uncertainty can be understated and inferential results can become too optimistic. The primary endpoint is a participant count, but the registry does not identify an over-dispersion model or report a dispersion estimate. There is no basis in the record to say that over-dispersion was assessed or that a particular count model was used.

What can be inferred from the analysis listing?

The registry lists eight outcome measures and zero analyses. This indicates that the posted record does not include analyses for those outcomes; it does not establish whether analyses were performed elsewhere. The listed secondary outcomes include relative change in tumor burden, per-lesion response, and 12-month progression-free survival, each described for first and second participation. The registry does not report results for those outcomes in the information available. Without estimates and their uncertainty, the statistical strength and clinical relevance of any relationship remain unassessable.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures: the primary initial objective-response outcome and secondary outcomes covering relative change in tumor burden, per-lesion response, and 12-month progression-free survival, with first- and second-participation versions among the secondary outcomes. It lists zero analyses. The record does not report whether a multiplicity strategy was planned or used, such as prioritizing outcomes, adjusting significance thresholds, or treating secondary outcomes as exploratory. Without that information and without analysis results, the chance of false-positive findings across multiple tests cannot be evaluated.

Missing response information is explicitly noted for Arm 1: two of fourteen participants were not evaluable because they left during active participation without any response evaluation. The record does not report how these participants were treated in the primary analysis, whether any outcome values were imputed, or whether a sensitivity analysis was performed. Their response status should not be assumed to be either response or nonresponse.

The analysis population is not reported in the registry record. It is therefore unclear from this record which enrolled participants were included in each endpoint analysis, whether inclusion required an evaluable ultrasound examination, or how repeat participation was handled. The availability and handling of missing ultrasound measurements and missing secondary outcomes are also not reported. These gaps limit assessment of potential bias from incomplete outcome data and make it difficult to connect the enrollment count to any analysis denominator.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse event figures for three categories. In Arm 1, 12 participants were at risk for serious adverse events and zero were affected; in Arm 2, 5 were at risk and zero were affected; and in Arm 3, 2 were at risk and zero were affected. No serious adverse event terms are listed. These are counts as recorded, not estimates of comparative safety.

For other adverse events, the record lists 2 affected participants among 12 at risk in Arm 1, zero among 5 in Arm 2, and zero among 2 in Arm 3. Death counts are 1, zero, and zero, respectively. The registry figures do not establish whether a death was caused by treatment, related to an adverse event, or otherwise connected to study participation. Such attribution is not reported in the registry record.

The categories are nonrandomized, and Arm 3 is described as participants enrolled consecutively in both arms. Their counts should not be interpreted as a randomized safety comparison. The registry does not provide event-level detail, timing, severity beyond the serious-event grouping, or further exposure information in these figures. Zero reported serious events in a category means none are listed as affected in these counts; it does not by itself establish absence of risk.

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

The study addresses a clinically relevant measurement question: whether changes in tumor perfusion observed early during treatment can anticipate the first objective response assessment. The registry identifies the ultrasound approach, the early measurement times of 3 or 6 weeks, and the response definition based on complete or partial response per RECIST v1.1. This makes the intended relationship understandable.

However, the registry record does not report the primary response count or any analysis of ultrasound measurements against response. It also reports zero analyses. There is therefore no posted estimate of predictive performance, no uncertainty interval, and no basis to characterize the strength or direction of a relationship. The stated study aim is not itself evidence that prediction was demonstrated.

The record also does not support a treatment-effect comparison. Participants were not randomized, masking was none, and the registry cautions that the treatment arms are not a like-for-like control comparison. Differences across the treatment categories cannot be attributed to treatment alone on the basis of these descriptions. The category of participants enrolled consecutively in both arms further complicates simple group comparisons.

The two Arm 1 participants who were not evaluable for response are a documented source of incomplete primary-outcome information. The record does not state their response status or how they were handled analytically. Other analysis-population and missing-data details are not reported.

In short, the registry describes a completed study designed to examine an early imaging marker, but the posted information is insufficient to judge whether that marker predicts response, how accurately it does so, or how precise any estimate might be. These limits concern the evidence reported in the record and do not establish what analyses may exist elsewhere.

13. Why This Trial Matters Statistically

Prediction studies require more than an early measurement and a later outcome: they need a clearly defined outcome, an explicit relationship between measurements and outcomes, and enough reporting to judge performance and uncertainty. This record specifies objective response as complete response or partial response per RECIST v1.1 and identifies power Doppler ultrasound perfusion changes after 3 or 6 weeks as the proposed early marker. That framing makes the intended statistical question clear even though its answer is not reported.

The design also illustrates why treatment comparisons and prediction questions need to be kept distinct. The ultrasound question concerns whether an early measure is associated with a later response classification. A comparison between treatment categories asks a different question and is especially difficult to interpret because allocation was nonrandomized and the registry says the arms are not like-for-like controls. A marker can be studied without establishing a causal treatment contrast, and a treatment-group difference would not by itself establish predictive value.

Reporting the response count, analysis population, missing-data handling, effect estimates, and uncertainty would be important for evaluating the evidence. The registry identifies two Arm 1 participants as not evaluable and lists zero analyses, but does not report the primary result or the analysis methods. These omissions prevent readers from assessing precision, potential bias, or the robustness of any predictive finding.

The statistical importance of the study is therefore in its question and in the reporting needs that accompany it. Early monitoring may be evaluated as a predictor, but conclusions depend on transparent outcome and analysis reporting, particularly when the design is nonrandomized and outcome data are incomplete for some participants.

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 the primary endpoint result, including the number of participants with complete response or partial response. It also gives no estimate of the relationship between ultrasound perfusion changes and objective response, no confidence interval, and no predictive-performance measures. This prevents evaluation of both the direction and precision of the intended finding.

Two of fourteen participants in Arm 1 were not evaluable for response because they were removed during active participation without a response evaluation. The record does not report their subsequent response status or how they were handled in analysis. Analysis populations, ultrasound-measurement completeness, and missing-data methods are not reported.

The study is nonrandomized and unmasked. The registry explicitly warns that the treatment arms are not a like-for-like control comparison. The treatment categories also include participants enrolled consecutively in both arms, which does not constitute a conventional independent comparator. No adjustment strategy or approach to confounding is reported.

Although the study is marked completed, the registry lists zero analyses. It does not report the multiplicity approach, statistical model for the primary question, validation strategy for prediction, sample-size justification, or interim analyses. Those details are not reported in the registry record and should not be inferred from the stated objective. The planning illustration is not the trial’s own design justification and is not observed power.

The record’s safety figures give participant counts but not detailed event descriptions or attribution for the death count. As a result, these figures cannot establish causality or support a comparative safety conclusion.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Serial Ultrasound in Metastatic Renal Cell Carcinoma (mRCC) — NCT04508725.
• Registry record last updated 2025-07-25; status last verified 2025-07.
• Results first posted 2025-07-25.
• 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

NCT04508725 is a completed, nonrandomized, single-group study sponsored by Stanford University, with 22 participants enrolled. The record lists a start date of 2020-12-05 and a primary completion date of 2023-11-30. Its purpose was to assess whether changes in quantitative tumor perfusion measured by power Doppler ultrasound after 3 or 6 weeks of treatment could predict initial objective response.

The primary outcome is initial objective response, defined as complete response or partial response per RECIST v1.1 at the first on-treatment evaluation. The record describes that evaluation as 8-16 weeks after treatment initiation and lists a 12-week timeframe. Two of fourteen participants in Arm 1 were not evaluable for response.

The registry lists eight outcome measures and zero analyses. It does not report the primary response count or an analysis of ultrasound prediction. The treatment categories were not randomized and are not described as like-for-like controls, limiting comparative interpretation.