Pembrolizumab Plus CA-4948 for the Treatment of Patients With Progressive Metastatic Urothelial Cancer Despite Prior Immunotherapy

Dose limiting toxicities (DLTs)

Adverse events (AEs) will be graded according to National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) version (v) 5.0. Occurrence of DLTs along with count and percentage of subjects experiencing DLTs will be summarized. Safety of the combination regimen will be summarized by the number of adverse events as well as by the number and percentage of subjects experiencing the adverse events in both dose escalation and expansion cohorts. The safety summary will include count and percentage for overall and by AE type, seriousness, severity, attribution, anticipated or not. Furthermore, toxicity index will be calculated as a summary index for each patient to summarize multiple AEs.

Recommended phase 2 dose (RP2D)

The RP2D may be determined to be the highest dose level, the maximum tolerated dose, or it may be a lower dose based on the consensus of the investigators, Cancer Therapy Evaluation Program and pharmaceutical company collaborators.

Incidence of AEs

AEs will be graded using NCI CTCAE v 5.0. Safety will be summarized by the number of AEs as well as by the number and percentage of subjects experiencing the AEs in both the dose escalation and dose expansion phases of the study. The safety summary will include count and percentage for overall and by AE type, seriousness, severity, attribution, anticipated or not. Toxicity will be calculated as a summary index for each patient to summarize multiple AEs.

Objective response rate (ORR)

ORR will be defined as the proportion of patients achieving a complete response (CR) or partial response (PR) per Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 criteria at 9 weeks. The number of confirmed responses will be reported along with a corresponding two-sided 95% Wilson confidence interval (CI) and p-value will be reported from Binomial exact test.

Progression-free survival (PFS)

PFS will be defined as the time from initial treatment to any progression or death from any cause, whichever occurs first. Kaplan-Meier estimates of median PFS time will be presented together with two-sided 95% CIs calculated according to Brookmeyer and Crowley. The frequency (number and percentage) of participants with an event (progression or death) and censoring reasons will be presented (alive, withdrawal of consent, lost to follow-up).

Overall survival (OS)

OS will be defined as the duration from the first date of study treatment to the date of death. Kaplan-Meyer estimates of median OS will be presented together with two-sided 95% CIs calculated according to Brookmeyer and Crowley. the frequency of participants with an event and censoring reasons will be presented.

Duration of response (DOR)

DOR will be defined as the time from date of the first documentation of response to the first documentation of progression of disease or death due to any cause in the absence of documented progression of disease. Kaplan-Meier estimates of median DOR time will be presented together with two-sided 95% CIs calculated according to Brookmeyer and Crowley. The frequency of participants with an event and censoring reasons will be presented.

Clinical benefit rate (CBR)

CBR will be defined as the proportion of patients achieving a CR or PR per RECIST 1.1 criteria at 9 weeks or stable disease at 9 weeks accompanied by a 1.5 or greater reduction in hrCRP. CBR will be reported along with a corresponding two-sided 95% Wilson CI and p-value will be reported from Binomial exact test.

High-sensitivity C-reactive protein (hsCRP)

CRP response rate will be defined as achieving a ≥ 1.5 fold reduction in CRP at 9 weeks. The CRP response rate will be reported along with a corresponding two-sided 95% Wilson CI and p-value will be reported from Binomial exact test. Linear mixed models will also be fitted for comparisons between patients achieving and not achieving an objective response and between patients achieving and not achieving clinical benefit. Random effects with autoregressive model of order 1 will be incorporated to describe the correlation of repeated measures. Univariable PH Cox regression models for PFS and OS will also be fitted with hsCRP levels either at baseline or as a time-dependent continuous covariate.

Cytokines and chemokines

Linear mixed models will also be used for comparisons between patients achieving and not achieving an objective response and between patients achieving and not achieving clinical benefit. Random effects with autoregressive model of order 1 will be incorporated to describe the correlation of repeated measures. Residual plots will be used to check model assumptions. Univariable PH Cox regression models for PFS and OS will also be fitted with cytokine and chemokine levels either at baseline or as time-dependent continuous covariates. Schoenfeld residuals will be used to check PH assumption. P-values will be adjusted by false discovery rate of 5%.

Multiplex immunohistochemistry markers

A paired t-test will be used to test whether there are a reduction of SPP1+ MoMacs and an increase in intra-tumoral T cells from baseline to on-treatment time points. Univariable logistic regression models will be fitted for ORR and CBR as response variables with either marker levels at baseline or change in marker levels as continuous covariates. PH Cox regression models for PFS and OS will also be fitted with marker levels either at baseline or as time-dependent continuous covariates. Schoenfeld residuals will be used to check PH assumption. P-values will be adjusted by false discovery rate of 5%.