Drug responses are conserved across patient-derived xenograft models of melanoma leading to identification of novel drug combination therapies.

BACKGROUND: Patient-derived xenograft (PDX) mouse tumour models can predict response to therapy in patients. Predictions made from PDX cultures (PDXC) would allow for more rapid and comprehensive evaluation of potential treatment options for patients, including drug combinations. METHODS: We developed a PDX library of BRAF-mutant metastatic melanoma, and a high-throughput drug-screening (HTDS) platform utilising clinically relevant drug exposures. We then evaluated 34 antitumor agents across eight melanoma PDXCs, compared drug response to BRAF and MEK inhibitors alone or in combination with PDXC and the corresponding PDX, and investigated novel drug combinations targeting BRAF inhibitor-resistant melanoma. RESULTS: The concordance of cancer-driving mutations across patient, matched PDX and subsequent PDX generations increases as variant allele frequency (VAF) increases. There was a high correlation in the magnitude of response to BRAF and MEK inhibitors between PDXCs and corresponding PDXs. PDXCs and corresponding PDXs from metastatic melanoma patients that progressed on standard-of-care therapy demonstrated similar resistance patterns to BRAF and MEK inhibitor therapy. Importantly, HTDS identified novel drug combinations to target BRAF-resistant melanoma. CONCLUSIONS: The biological consistency observed between PDXCs and PDXs suggests that PDXCs may allow for a rapid and comprehensive identification of treatments for aggressive cancers, including combination therapies.

Dual Targeting of EGFR and MTOR Pathways Inhibits Glioblastoma Growth by Modulating the Tumor Microenvironment.

Glioblastoma's (GBM) aggressive growth is driven by redundant activation of a myriad of signaling pathways and genomic alterations in tyrosine kinase receptors, such as epidermal growth factor receptor (EGFR), which is altered in over 50% of cases. Single agents targeting EGFR have not proven effective against GBM. In this study, we aimed to identify an effective anti-tumor regimen using pharmacogenomic testing of patient-derived GBM samples, in culture and in vivo. High-throughput pharmacological screens of ten EGFR-driven GBM samples identified the combination of erlotinib (EGFRi) and MLN0128 (a mammalian target of rapamycin inhibitor, or MTORi) as the most effective at inhibiting tumor cell viability. The anti-tumor activity of erlonitib+MLN0128 was synergistic and produced inhibition of the p-EGFR, mitogen-activated protein kinase (MAPK), and Phosphoinositide 3-kinase (PI3K) pathways in culture. Using an orthotopic murine model of GBM, we show that erlotinib+MLN0128 inhibited tumor growth in vivo and significantly prolonged the survival of tumor-bearing mice. Expression profiling of tumor tissues from treated mice revealed a unique gene signature induced by erlotinib+MLN0128, consisting of downregulation of immunosuppressive chemokines in the tumor microenvironment, including C-C motif chemokine ligand 2 (CCL2) and periostin. Lower periostin levels resulted in the inhibition of Iba1+ (tumor-promoting) macrophage infiltration of GBM xenografts. Taken together, our results demonstrate that pharmacological co-targeting of EGFR and MTOR using clinically available drugs represents an effective treatment paradigm for EGFR-driven GBMs, acting both by inhibiting tumor cell growth and modulating the immune tumor microenvironment.

Pan-Cancer Pharmacogenomic Analysis of Patient-Derived Tumor Cells Using Clinically Relevant Drug Exposures.

As a result of tumor heterogeneity and solid cancers harboring multiple molecular defects, precision medicine platforms in oncology are most effective when both genetic and pharmacologic determinants of a tumor are evaluated. Expandable patient-derived xenograft (PDX) mouse tumor and corresponding PDX culture (PDXC) models recapitulate many of the biological and genetic characteristics of the original patient tumor, allowing for a comprehensive pharmacogenomic analysis. Here, the somatic mutations of 23 matched patient tumor and PDX samples encompassing four cancers were first evaluated using next-generation sequencing (NGS). 19 antitumor agents were evaluated across 78 patient-derived tumor cultures using clinically relevant drug exposures. A binarization threshold sensitivity classification determined in culture (PDXC) was used to identify tumors that best respond to drug in vivo (PDX). Using this sensitivity classification, logic models of DNA mutations were developed for 19 antitumor agents to predict drug response. We determined that the concordance of somatic mutations across patient and corresponding PDX samples increased as variant allele frequency increased. Notable individual PDXC responses to specific drugs, as well as lineage-specific drug responses were identified. Robust responses identified in PDXC were recapitulated in vivo in PDX-bearing mice and logic modeling determined somatic gene mutation(s) defining response to specific antitumor agents. In conclusion, combining NGS of primary patient tumors, high-throughput drug screen using clinically relevant doses, and logic modeling, can provide a platform for understanding response to therapeutic drugs targeting cancer.

Pitfalls and Rewards of Setting Up a Liquid Biopsy Approach for the Detection of Driver Mutations in Circulating Tumor DNAs: Our Institutional Experience.

We describe our institutional experience of developing a liquid biopsy approach using circulating tumor DNA (ctDNA) analysis for personalized medicine in cancer patients, focusing on the hurdles encountered during the multistep process in order to benefit other investigators wishing to set up this type of study in their institution. Blood samples were collected at the time of cancer surgery from 209 patients with one of nine different cancer types. Extracted tumor DNA and circulating cell-free DNA were sequenced using cancer-specific panels and the Illumina MiSeq machine. Almost half of the pairs investigated were uninformative, mostly because there was no trackable pathogenic mutation detected in the original tumor. The pairs with interpretable data corresponded to 107 patients. Analysis of 48 gene sequences common to both panels was performed and revealed that about 40% of these pairs contained at least one driver mutation detected in the DNA extracted from plasma. Here, we describe the choice of our overall approach, the selection of the cancer panels, and the difficulties encountered during the multistep process, including the use of several tumor types and in the data analysis. We also describe some case reports using longitudinal samples, illustrating the potential advantages and rewards in performing ctDNA sequencing to monitor tumor burden or guide treatment for cancer patients.