Search

Abstract Background The success of agents that reverse T-cell inhibitory signals, such as anti-PD-1/PD-L1 therapies, has reinvigorated cancer immunotherapy research. However, since only a minority of patients respond to single-agent therapies, methods to test the potential anti-tumor activity of rational combination therapies are still needed. Conventional murine xenograft models have been hampered by their immune-compromised status; thus, we developed a hematopoietic humanized mouse model, hu-CB-BRGS, and used it to study anti-tumor human immune responses to triple-negative breast cancer (TNBC) cell line and patient-derived colorectal cancer (CRC) xenografts (PDX). Methods BALB/c-Rag2nullIl2rγnullSIRPαNOD (BRGS) pups were humanized through transplantation of cord blood (CB)-derived CD34+ cells. Mice were evaluated for human chimerism in the blood and assigned into experimental untreated or nivolumab groups based on chimerism. TNBC cell lines or tumor tissue from established CRC PDX models were implanted into both flanks of humanized mice and treatments ensued once tumors reached a volume of ~150mm3. Tumors were measured twice weekly. At end of study, immune organs and tumors were collected for immunological assessment. Results Humanized PDX models were successfully established with a high frequency of tumor engraftment. Humanized mice treated with anti-PD-1 exhibited increased anti-tumor human T-cell responses coupled with decreased Treg and myeloid populations that correlated with tumor growth inhibition. Combination therapies with anti-PD-1 treatment in TNBC-bearing mice reduced tumor growth in multi-drug cohorts. Finally, as observed in human colorectal patients, anti-PD-1 therapy had a strong response to a microsatellite-high CRC PDX that correlated with a higher number of human CD8+ IFNγ+ T cells in the tumor. Conclusion Hu-CB-BRGS mice represent an in vivo model to study immune checkpoint blockade to human tumors. The human immune system in the mice is inherently suppressed, similar to a tumor microenvironment, and thus allows growth of human tumors. However, the suppression can be released by anti-PD-1 therapies and inhibit tumor growth of some tumors. The model offers ample access to lymph and tumor cells for in-depth immunological analysis. The tumor growth inhibition correlates with increased CD8 IFNγ+ tumor infiltrating T cells. These hu-CB-BRGS mice provide a relevant preclinical animal model to facilitate prioritization of hypothesis-driven combination immunotherapies.

Background: Troxacitabine is a novel L-nucleoside analogue. Preclinical studies showed improved activity with infusions of at least 3 days compared with bolus regimens, especially at concentrations >20 ng/ml. This phase I study tested the feasibility of achieving a troxacitabine steady-state concentration of 20 ng/ml for at least 72 h in patients with solid tumors. Patients and methods: Patients with solid tumors received troxacitabine as a progressively longer infusion on days 1–4 of a 28-day cycle. The initial length of infusion and infusion rate were 48 h and 3 mg/m2/day. Results: Twenty-one patients were treated at infusion lengths that increased from 48 to 72 h and then 96 h. The infusion rate was decreased from 3 to 1.88 mg/m2/day due to toxicity. Dose-limiting toxicities consisted of grade 4 neutropenia (three) and grade 3 constipation (one). The maximum tolerated dose of continuous infusion troxacitabine in patients with solid tumors is 7.5 mg/m2 administered over 96 h. This dose level resulted in steady-state drug concentration of at least 20 ng/ml for 72 h. Conclusions: Administration of troxacitabine by continuous infusion achieved the prospectively defined target plasma concentration. Pharmacokinetics (PK) modeling coupled with real-time PK assessment was an efficient approach to conduct hypothesis-driven phase I trials. [ABSTRACT FROM AUTHOR]