Tivantinib With or Without Erlotinib Hydrochloride in Treating Patients With Metastatic or Locally Advanced Kidney Cancer That Cannot Be Removed by Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tivantinib With or Without Erlotinib Hydrochloride in Treating Patients With Metastatic or Locally Advanced Kidney Cancer That Cannot Be Removed by Surgery (NCT01688973), a randomized phase 2 study. The primary outcome is Response Rate (Confirmed Complete Response or Partial Response), Determined According to Response Evaluation Criteria in Solid Tumors.

2. Clinical Question

3. Trial Design

This was a randomized, phase II, parallel-group trial. The registry lists enrollment of 55 and identifies the National Cancer Institute as the sponsor. Masking is listed as none, so the study was open-label. The trial status is completed. The recorded start date is 2012-08-20 and the primary completion date is 2017-04-30.

Arm I received tivantinib PO BID on days 1-28. Arm II received tivantinib PO BID and erlotinib hydrochloride PO QD on days 1-28. The comparison therefore keeps tivantinib in both regimens and adds erlotinib hydrochloride in one arm. The registry also lists laboratory biomarker analysis, without describing its procedures or how it informed the treatment comparison.

The registry lists recurrent renal cell carcinoma, stage III and stage IV renal cell cancer, and type 1 and type 2 papillary renal cell carcinoma among the conditions. Its summary describes metastatic or locally advanced kidney cancer that cannot be removed by surgery. Further eligibility criteria, baseline participant characteristics, and the number assigned to each arm are not reported in the registry record. The overall enrollment should not be taken as a group-specific count.

The primary outcome was response rate, defined around confirmed complete or partial response, with a timeframe of up to 3 years. Secondary outcomes included frequency and severity of toxicities graded using the National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.0, and progression-free survival. The record does not describe the randomization procedure, allocation concealment, or a statistical analysis plan.

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

Reasons for not completing this period:

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: Two patients in Arm I and 3 patients in Arm II were ineligible and excluded from the analyses.

6. Primary Endpoint

The registry lists the primary outcome as Response Rate (Confirmed Complete Response or Partial Response), Determined According to Response Evaluation Criteria in Solid Tumors, measured over Up to 3 years.

The primary endpoint counts participants whose best response met the registry definition of confirmed complete response or partial response, assessed according to Response Evaluation Criteria in Solid Tumors. The recorded timeframe is up to 3 years. The response classification depends on the sequence of objective disease statuses, rather than on a single observation alone.

For complete response, the definition requires at least two objective statuses of complete response at least four weeks apart, documented before progression or symptomatic deterioration. Partial response similarly requires at least two objective statuses of partial response or better at least four weeks apart, before progression or symptomatic deterioration, and not meeting the complete-response definition. The registry also describes stable/no response, increasing disease, and symptomatic deterioration categories. These categories make clear that the primary outcome is based on a defined best-response classification.

As a general statistical matter, an objective response endpoint can describe whether tumors meet prespecified criteria for shrinkage or disappearance. Confirmation requirements help distinguish a sustained response classification from an isolated assessment. The record does not state the rationale for selecting this endpoint, nor does it provide group-specific response counts or a comparative response estimate. The endpoint definition alone therefore describes what would be measured, not what the trial found.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

In general, randomization assigns participants to treatment groups using a chance-based process. Its purpose is to make the groups comparable on average at the time of assignment, including on factors that may not have been measured. It does not guarantee identical groups, especially in a phase II study, and it does not remove the need to describe participant characteristics or account for missing outcome information. This registry identifies the study as randomized but does not report the method used to create the allocation sequence or how assignment was concealed.

The parallel design means that participants were assigned to one regimen rather than crossing between regimens according to a crossover schedule. The recorded masking status is none, so the trial was open-label. As a general matter, lack of masking can affect behavior, clinical management, and outcome assessment, depending on the endpoint and safeguards. The registry does not describe blinding of response assessors or other steps to limit assessment bias.

How should a response endpoint be read?

The primary outcome is a count of participants meeting confirmed complete- or partial-response criteria. The registry defines these categories using repeated objective statuses and specifies that the qualifying statuses must be documented before progression or symptomatic deterioration. Requiring confirmation is intended, in general, to make a response classification depend on more than a transient or isolated assessment.

For a binary comparison, a statistical analysis would ordinarily summarize the response count in each group and estimate a contrast between groups. Uncertainty intervals would describe the range of effect sizes compatible with the data under the analysis assumptions. A test result alone would not convey the size or precision of the effect. In this case, the registry information does not present a single response comparison, so neither the effect estimate nor its uncertainty can be interpreted from the primary outcome record.

What do the adverse-event counts represent?

The adverse-event information is recorded as counts of participants with an event. The registry gives the number at risk and the number affected for serious and other events in each arm, along with death counts. These are descriptive summaries; without a stated analysis, they should not be read as adjusted comparisons or as evidence of causation. In general, adverse-event comparisons also depend on how events were collected, the period of observation, exposure time, and rules for attributing events. Those details are not reported in the registry record presented for this trial.

The registry names a toxicity grading system, the National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.0. Grading provides a structured way to describe severity. It does not, by itself, establish whether an event was caused by treatment. The record does not provide event timing, duration, exposure-adjusted rates, or arm-specific counts for each serious-event term in the list.

What does over-dispersion mean in general?

For count outcomes, a simple model may assume that variation is explained by the average count. Over-dispersion means the observed variation is greater than that assumption allows. If it is ignored, standard errors may be too small and uncertainty may appear narrower than it is. This is a general statistical concern for some count analyses; the primary outcome is described as a participant count and the registry does not report a model or an over-dispersion assessment.

How should planning calculations be interpreted?

The planning illustration is explicitly described as being computed from an observed effect, not as the trial's own design justification. It assumes a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such calculations illustrate how a hypothetical future design might relate to an assumed effect and target power. They are not observed power, do not show that the trial was adequately powered, and do not substitute for the sponsor's original sample-size rationale. That rationale is not reported in the registry record. The primary response comparison is not reduced to a single estimate in the posted results, so those planning assumptions should not be treated as evidence about the trial's observed treatment effect.

More broadly, analysis choices matter. The registry does not report a statistical analysis plan, prespecified model, confidence-interval method, handling of missing assessments, or analysis population. Without those details and without a primary comparative estimate, readers cannot assess whether the posted outcome summaries follow a prespecified strategy or how sensitive conclusions might be to alternative assumptions.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures, including the primary response outcome and secondary outcomes for toxicities and progression-free survival. It records zero analyses. The primary response outcome is described through several categories or time points and is not reduced to one comparison in the posted results. A statistical analysis plan, including a plan for handling multiple outcomes or comparisons, is not reported in the registry record.

Multiplicity is a general concern when several outcomes, time points, or subgroup comparisons are examined. The more opportunities there are to identify a seemingly notable result, the more important it is to distinguish prespecified analyses from exploratory ones and to explain any adjustment strategy. No multiplicity procedure or hierarchy among outcomes is reported. It is therefore not possible to determine how the investigators planned to interpret multiple statistical tests.

The registry does not describe missing outcome assessments, reasons for missingness, imputation methods, or sensitivity analyses. It also does not identify the analysis population used for response or safety summaries. In general, missing outcomes can affect estimates if the probability of missingness is related to prognosis or treatment experience; the consequences depend on the pattern and analysis assumptions. Those assumptions cannot be evaluated from this record.

The registry's planning illustration is described as separate from the trial's own design, and the trial's sample-size justification is not reported. The posted information therefore does not establish a prespecified statistical strategy for multiplicity, missing data, or analysis populations.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dyspnea (16.0%), anemia (8.0%), abdominal pain (8.0%), small intestinal obstruction (8.0%), fatigue (8.0%).

The registry reports serious adverse-event counts and other adverse-event counts for each arm, with 25 participants at risk in each. In the tivantinib arm, 9 participants were recorded as affected by serious adverse events and 22 by other adverse events; the death count is 20. In the combination arm, 16 participants were recorded as affected by serious adverse events and 23 by other adverse events; the death count is 21. These are counts, not adjusted comparisons.

The listed serious-event terms include dyspnea at 16%, and anemia, abdominal pain, small intestinal obstruction, fatigue, and increased alanine aminotransferase at 8% each. The record does not assign these listed term percentages to a particular arm. A percentage attached to a term is not, on its own, evidence that the event was caused by either regimen. The registry does not give the timing, duration, or treatment attribution of these events, nor does it explain the causes or timing of deaths.

As a general principle, safety summaries should be interpreted alongside definitions, observation periods, exposure, and event attribution. The posted counts do not establish that one regimen is safer than the other.

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

The trial record describes a randomized phase II comparison of tivantinib alone with tivantinib plus erlotinib hydrochloride in patients with advanced papillary renal cell carcinoma. It specifies a confirmed response endpoint and gives a detailed classification rule for complete and partial response. These details clarify what the primary outcome was intended to measure.

However, the posted primary outcome information does not give a single comparative response estimate. Without group-specific response counts or an estimate of the between-group contrast, it is not possible to describe the magnitude or precision of any response difference. The record also does not report a confidence interval or a comparative test result for this endpoint. The design and endpoint definition should not be mistaken for evidence about the result.

The adverse-event information supplies descriptive counts in each arm and a list of serious-event terms with percentages. Such summaries can help characterize what the registry recorded, but they do not demonstrate that a treatment caused an event or establish a difference in safety between regimens. Death counts are also listed, without causes or timing. The record does not give a comparative progression-free survival result.

In general, a randomized trial can support a treatment comparison when outcomes are analyzed and reported with enough detail to assess the estimate and its uncertainty. Here, the available registry information is incomplete for that purpose. It does not establish that either regimen improved response, nor that the regimens were equivalent. It also does not support a conclusion that either regimen was safer. Those conclusions would require results and analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for the contrast between a randomized design and the limited comparative outcome information recorded. Randomization can support a fair comparison, but interpretation also depends on the outcome data, the analysis, and its uncertainty. The registry identifies the main question and how confirmed response was defined, yet it does not report a single response comparison.

The response definition illustrates why endpoint construction matters. Complete and partial responses require repeated objective statuses at least four weeks apart, and the qualifying statuses must precede progression or symptomatic deterioration. Such rules aim to make the outcome classification more consistent, but the usefulness of the definition for treatment comparison depends on complete and clearly reported assessments in each arm. The registry does not provide those arm-specific response results.

The study also highlights the difference between descriptive safety counts and comparative evidence. Counts of affected participants and listed serious-event terms are informative descriptions, but alone they do not resolve attribution or quantify a comparative effect with uncertainty. The registry's absence of a recorded analysis, statistical plan, and missing-data approach limits assessment of how the outcomes were evaluated.

Finally, planning illustrations based on an observed effect should not be confused with the original design rationale or with observed power. The record explicitly distinguishes the illustration from the trial's own design. This distinction is important when readers use a trial to think about future study planning: assumptions and estimates used for planning are not substitutes for an adequately reported trial result.

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 outcome as a single comparison. It therefore lacks the arm-specific response counts and comparative estimate needed to describe the observed response difference and its precision. The record also does not give a comparative progression-free survival result, despite listing that outcome as secondary.

Further limits concern the methods. The randomization procedure, allocation concealment, statistical analysis plan, analysis population, and handling of missing outcomes are not reported. Masking is listed as none, and the record does not state whether response assessors were blinded. The posted information does not explain how several response categories or time points were incorporated into an analysis, or how multiple outcomes were handled.

The adverse-event summaries do not include timing, duration, treatment attribution, or exposure-adjusted rates. Serious-event terms are not reported by arm in the listed term summary, and death counts are not accompanied by causes or timing. These details limit interpretation of the safety counts and prevent a reliable causal or comparative safety conclusion from these summaries alone.

The registry lists conditions and describes the population broadly as having metastatic or locally advanced kidney cancer that cannot be removed by surgery, but does not report detailed eligibility criteria or baseline characteristics. The distribution of participants between arms is also not stated in the record. The sample-size justification and rationale are not reported; the planning illustration is explicitly not the trial's own design. These gaps limit assessment of generalizability, precision, and the intended statistical strategy.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Tivantinib With or Without Erlotinib Hydrochloride in Treating Patients With Metastatic or Locally Advanced Kidney Cancer That Cannot Be Removed by Surgery — NCT01688973.
• Registry record last updated 2019-01-03; status last verified 2018-12.
• Results first posted 2019-01-03.
• 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

NCT01688973 is a completed, randomized, open-label, parallel phase II trial sponsored by the National Cancer Institute. The registry records enrollment of 55, with a start date of 2012-08-20 and primary completion date of 2017-04-30.

The trial compared tivantinib alone with tivantinib plus erlotinib hydrochloride in metastatic or locally advanced papillary renal cell carcinoma. The primary outcome was confirmed complete or partial response according to Response Evaluation Criteria in Solid Tumors, assessed up to 3 years. Secondary outcomes included toxicity frequency and severity and progression-free survival.

The registry gives adverse-event counts by arm but does not report a single comparative primary response estimate. It lists five outcome measures and zero analyses. The statistical plan, handling of missing outcomes, and sample-size justification are not reported in the registry record.