Trial of TRC105 and Sorafenib in Patients With HCC: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial of TRC105 and Sorafenib in Patients With HCC (NCT02560779), a na phase 1, phase 2 study. The primary outcome is Number of Patients Who Experience Dose Limiting Toxicities by Dose Level.

2. Clinical Question

3. Trial Design

This was an open-label phase 1b/2 clinical trial of carotuximab (TRC105) combined with standard-dose sorafenib in patients with hepatocellular carcinoma. The registry lists a single-group model, no masking, and allocation as not applicable. The record reports enrollment of 27 and a completed status. The study started in 2016-11, with primary completion in 2019-08.

The phase 1b portion was intended to evaluate safety and tolerability and determine a recommended phase 2 dose for TRC105 when added to sorafenib. The phase 2 portion was intended to estimate objective response rate according to RECIST 1.1. The registry description gives separate treatment caps for the two portions, but the actual division of enrollment between portions is not reported in the registry record.

The safety results identify two carotuximab schedules with sorafenib: 10 mg/kg weekly, and 10 mg/kg weekly followed by 15 mg/kg every other week. These groups are not a randomized intervention-versus-control structure. Differences between their reported event counts could reflect differences in group composition, exposure, follow-up, or other factors; the registry record does not report enough information to distinguish among these possibilities.

The primary outcome is the number of participants with dose-limiting toxicities by dose level, assessed during the first 4 months. The record’s description explains the dose-escalation rule, but dose-level outcome counts are not reported in the registry record. It also identifies pharmacokinetic profile and anti-product antibody assessment as secondary outcomes. The available record does not report their results or a detailed 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

Analysis population for the primary outcome, as defined in the registry: Patients must have received at least a portion of a dose of TRC105 or sorafenib to be evaluable for assessment of DLT.

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: Patients who received at least a portion of a dose of Carotuximab

6. Primary Endpoint

The registry lists the primary outcome as Number of Patients Who Experience Dose Limiting Toxicities by Dose Level, measured over 4 months.

The registered primary endpoint counts participants who experience a dose-limiting toxicity, or DLT, by dose level. The assessment window is the first 4 months of participation. A DLT was defined as one or more toxicities outlined in the protocol and considered at least possibly related to TRC105. The registry says that a participant had to receive at least a portion of a dose of TRC105 or sorafenib to be evaluable.

This kind of endpoint is used in early-phase dose evaluation to describe tolerability at the dose levels under study. The registry specifies that if 1 of 3 participants experienced a DLT, the dose level would be expanded to 6 participants. It also states that the maximum tolerated dose would have been exceeded if at least 33% of participants at a dose level experienced a DLT. These are decision rules described in the record, not reported findings.

The endpoint is a count of participants, rather than a continuous measurement or a time-to-event outcome. Its meaning depends on the specified toxicity definition, assessment window, and evaluable population. The actual number of DLTs by dose cohort is not reported in the registry record, so the primary endpoint result and the dose-level pattern cannot be interpreted from the posted information. Although the phase 2 objective was to estimate response rate by RECIST 1.1, that objective does not replace the registered primary endpoint listed in the outcome record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry presents an early-phase, open-label study with a single-group model. These design features shape how results can be interpreted. Open-label means masking was not used, as the record explicitly lists masking as none. A single-group designation means the study is not recorded as a randomized comparison between an intervention and a control. The safety record nevertheless separates participants into two carotuximab schedule groups. Their event counts are descriptive and cannot be treated as results from randomized arms.

What does a dose-limiting toxicity count describe?

A DLT count records how many evaluable participants meet the trial’s stated toxicity definition within its assessment window. Here the window is the first 4 months, and the registry ties the definition to toxicities outlined in the protocol that are considered at least possibly related to TRC105. The record also requires receipt of at least a portion of a dose of TRC105 or sorafenib for DLT evaluability. In general, a clear definition and consistent assessment window help make dose-level safety summaries interpretable. The actual dose-level counts are not reported in the registry record.

The stated dose-escalation rule says that a level would be expanded if 1 of 3 participants had a DLT, and that the maximum tolerated dose would have been exceeded if at least 33% of participants at a dose level had a DLT. These rules explain how the protocol intended to use observed toxicity information. They are not evidence that an expansion or stopping criterion was met. The record does not report the dose-level counts, dose decisions, or the selected recommended phase 2 dose.

Why are the schedule groups not a control comparison?

In a randomized trial, random assignment is intended to balance prognostic factors across treatment groups on average. This record does not describe randomized assignment: allocation is listed as not applicable and the design is single-group. The two safety group labels identify dosing schedules, but there is no basis in the record to interpret the groups as exchangeable control arms. Differences in their observed counts could reflect participant characteristics, follow-up, treatment exposure, or chance, among other factors. The registry does not report information sufficient to assess those explanations.

In general, a confidence interval expresses uncertainty around an estimated quantity under a specified statistical model. A narrow interval indicates greater precision than a wide interval, while neither width nor statistical significance alone establishes causation. No confidence intervals or inferential comparisons are reported for the DLT or serious adverse-event counts in the registry record. No effect estimate for the primary endpoint is available in the record.

How should counts and percentages be read?

Counts retain the observed number of participants recorded with an event and are especially relevant for small early-phase safety groups. A percentage can make a count easier to compare across group sizes, but its meaning depends on a clear denominator and event definition. The serious adverse-event information gives group counts and also lists selected event terms with percentages. The denominator and specific counting basis for those term-level percentages are not reported in the registry record. They should not be used to derive further estimates.

In general, event counts may reflect whether a participant had one or more events, rather than the total number of event episodes. A participant-level count and an event-level count answer different questions. The posted record describes the primary outcome as a number of participants with DLTs and presents serious adverse-event participant counts. Further details on recurrent events, duration, severity grading, and exposure-adjusted rates are not reported in the registry record.

What can the analysis record support?

The registry lists 4 outcome measures and 0 analyses. This does not establish that no analyses were conducted; it means formal analysis results are not recorded in the information available. The record does not report the statistical methods used for the primary endpoint, any adjustment for covariates, handling of missing outcomes, or inferential uncertainty. The phase 2 objective of estimating response rate by RECIST 1.1 is stated, but response counts and analysis details are not reported in the registry record.

In general, over-dispersion matters when count data vary more than a statistical model expects; using an unsuitable model can make standard errors and uncertainty estimates unreliable. That concern is relevant to some group-level count analyses, but the registry record does not report a model for the DLT or safety outcomes, and it gives no basis to claim that over-dispersion was assessed. Similarly, a formal sample-size justification, interim analysis plan, and stopping boundaries beyond the stated dose rules are not reported in the registry record. The information supports a design-based reading and descriptive reading of the posted safety counts, not an independent efficacy or comparative analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Carotuximab (10 mg/kg wk to 15 mg/kg Bi-wk) Plus S", "Carotuximab (10 mg/kg Weekly) Plus Sorafenib")
n      <- c(14, 13)
events <- c(1, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Carotuximab (10 mg/kg wk to 15 mg/kg Bi-wk) Plus S  14  1  7.1  0.2  33.9  1.3  31.5  1.7  31.9
ARM  Carotuximab (10 mg/kg Weekly) Plus Sorafenib  13  0  0.0  0.0  24.7  0.0  22.8  0.2  23.2

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies multiple objectives and outcomes: the primary dose-limiting toxicity count by dose level, a phase 2 objective of estimating response rate by RECIST 1.1, and secondary pharmacokinetic and immunogenicity outcomes. Considering several outcomes can raise multiplicity concerns because the more results examined, the greater the chance that at least one apparently notable finding could occur by chance. The registry record does not report a multiplicity adjustment, a hierarchy for interpreting outcomes, or a detailed statistical analysis plan. It is therefore not possible to determine from the record how multiple outcomes were handled analytically.

Missing-data methods are also not reported in the registry record. The DLT description defines an evaluable population in terms of receiving at least a portion of a dose of TRC105 or sorafenib, but the record does not state how participants with incomplete follow-up, unevaluable toxicity information, or withdrawals were handled in the analysis. It does not report endpoint-specific denominators for all outcomes or explain whether missing outcome information was imputed, excluded, or handled in another way.

The phase 1b/2 description gives separate objectives for safety and dose selection and for response-rate estimation, but the analysis populations for those objectives are not reported in the registry record. The number enrolled is recorded as 27, while the distribution by study portion and the number analyzed for each endpoint are not provided in the information available. Consequently, the posted results do not permit a complete assessment of missingness, population definitions, or the sensitivity of findings to alternative analysis choices.

12. 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: pleural effusion (15.4%), acute kidney injury (7.7%), alanine aminotransferase increased (7.7%), anemia (7.7%), aspartate aminotransferase increase (7.7%).

The serious adverse-event counts are reported separately for the two carotuximab-plus-sorafenib schedules. In the group titled carotuximab 10 mg/kg weekly followed by 15 mg/kg every other week plus sorafenib, 5 of 14 participants were reported with a serious adverse event. In the group titled carotuximab 10 mg/kg weekly plus sorafenib, 9 of 13 participants were reported with a serious adverse event. The registry reports 0 deaths in each group.

These counts describe the groups as recorded; they do not establish that the schedules differ in safety. This was not a randomized comparison, and the registry does not provide the contextual information needed to interpret a direct contrast as causal. Serious adverse events are also distinct from dose-limiting toxicities: a serious adverse event is a safety reporting category, while the DLT definition in this study depends on specified toxicities, their possible relationship to TRC105, and the first 4 months of participation.

The record lists pleural effusion at 15.4%, and acute kidney injury, alanine aminotransferase increased, anemia, aspartate aminotransferase increase, and blood bilirubin increased at 7.7% each among the leading serious-event terms. The denominator and counting basis for these term-level percentages are not reported in the registry record. The absence of reported deaths in these groups should not be generalized beyond the registry’s stated groups and results.

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

The registry record establishes that this completed, open-label phase 1b/2 study evaluated carotuximab (TRC105) with standard-dose sorafenib in patients with hepatocellular carcinoma. It describes a phase 1b purpose of evaluating safety and tolerability and determining a recommended phase 2 dose, and a phase 2 purpose of estimating objective response rate by RECIST 1.1. It also records serious adverse-event counts for two carotuximab schedules.

The posted information does not report the primary endpoint’s DLT counts by dose level. It therefore cannot show the observed toxicity pattern across levels, whether the stated dose-escalation criteria were met, or what recommended phase 2 dose was selected. Response-rate results and results for the listed pharmacokinetic and immunogenicity outcomes are also not reported in the registry record. The record’s stated objectives should not be mistaken for reported findings.

The serious adverse-event counts show that events were recorded in both safety groups, with 5 of 14 participants in one group and 9 of 13 in the other reported with a serious adverse event. The record reports 0 deaths in each group. These observations are descriptive. They do not establish that either schedule caused more or fewer serious events, and they cannot separate treatment-related events from events due to other causes.

In general, a trial’s observed results and the strength of conclusions are not the same thing. Interpretation depends on design, outcome definitions, completeness of follow-up, analysis methods, and precision. Here the lack of randomization precludes reading the regimen groups as treatment-versus-control comparisons, while the absence of reported DLT and response results limits conclusions about the trial’s primary and phase 2 objectives. Formal estimates of uncertainty and adjusted analyses are not reported in the registry record.

Accordingly, the record is informative about the study’s intended questions, design, and selected safety counts, but it does not support a complete evaluation of dose selection or response. It does not establish a comparative treatment effect, and it should not be interpreted as proof for or against treatment benefit.

14. Why This Trial Matters Statistically

This trial illustrates an important statistical distinction between an early-phase dose-evaluation objective and a comparative efficacy question. Its phase 1b purpose focused on safety, tolerability, and identifying a recommended phase 2 dose for TRC105 with sorafenib. The primary endpoint was a participant count of DLTs by dose level during the first 4 months. That endpoint is designed to inform dose evaluation, not by itself to establish comparative treatment benefit.

The study also highlights how design labels constrain interpretation. The registry describes a single-group, open-label trial with allocation not applicable, while safety information is listed for two dose schedules. Those groups might invite direct comparison, but they do not constitute randomized treatment and control arms. Statistical contrasts cannot remove the limitations created by nonrandom allocation or unreported differences in participant characteristics and follow-up.

Another lesson is that a stated objective is not equivalent to a reported result. The record identifies response-rate estimation as the phase 2 aim, yet response results are not reported in the registry record. It also describes dose-escalation rules, but the dose-level DLT counts and dose-selection outcome are not reported. Readers need those observed outcomes, along with analysis methods and uncertainty estimates, to assess how well the stated questions were answered.

Finally, safety counts require careful interpretation. The serious adverse-event group counts and listed terms provide descriptive information, but do not establish causality or a difference between schedules. The absence of reported deaths in the listed groups does not resolve other safety questions. Together, these features make the record useful for understanding the planned early-phase framework while underscoring the limits of drawing conclusions from incomplete outcome reporting and nonrandomized group data.

15. 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 principal limitation for interpreting the primary endpoint is that dose-limiting toxicity counts by dose cohort are not reported in the registry record. The record describes the DLT definition and dose-escalation rules, but not the observed results or the recommended phase 2 dose. Without those counts, the primary endpoint cannot be assessed against the protocol’s stated decision rules.

The record gives the phase 2 objective as estimating response rate by RECIST 1.1 but does not report the response results. It also lists pharmacokinetic and anti-product antibody outcomes without reporting their findings. The available information therefore does not permit evaluation of those objectives or outcomes.

The design is recorded as single-group, open-label, with allocation not applicable and no masking. The two carotuximab schedule groups are not randomized comparators. Their serious adverse-event counts cannot support a causal comparison, and the registry does not report characteristics or exposure details that would help explain differences between groups.

Further limitations include the absence of a reported statistical analysis plan, analysis populations, missing-data methods, confidence intervals, and formal inferential results. The record lists 4 outcome measures and 0 analyses, but that does not show whether analyses were or were not conducted outside the registry. The denominator and counting basis for the listed serious-event percentages are also not reported in the registry record. These gaps limit assessment of precision, completeness, and the robustness of conclusions from the posted information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Trial of TRC105 and Sorafenib in Patients With HCC — NCT02560779.
• Registry record last updated 2020-07-17; status last verified 2020-07.
• Results first posted 2020-07-17.
• 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.

19. Record Summary

NCT02560779 is a completed, open-label phase 1b/2 trial of carotuximab (TRC105) with standard-dose sorafenib in patients with hepatocellular carcinoma. The registry records a single-group model, allocation not applicable, enrollment of 27, a start in 2016-11, and primary completion in 2019-08.

The primary outcome was the number of participants with DLTs by dose level during the first 4 months. The phase 1b objective was to assess safety and tolerability and determine a recommended phase 2 dose; the phase 2 objective was to estimate response rate by RECIST 1.1. Neither the dose-level DLT counts nor response results are reported in the registry record.

Serious adverse events are reported for two carotuximab schedules with sorafenib, with 5 of 14 and 9 of 13 participants affected, respectively. The record reports 0 deaths in each group. The groups are not randomized comparators, and the counts are descriptive rather than evidence of a causal difference.