Dual-hit strategy for therapeutic targeting of pancreatic cancer in patient-derived xenograft tumors

Paracrine activation of pro-fibrotic sonic hedgehog (sHH) signaling in pancreatic ductal adenocarcinoma (PDAC) results in stromal amplification that compromises tumor drug delivery, efficacy, and patient survival. Interdiction of sHH-mediated tumor-stroma crosstalk with smoothened inhibitors transiently increases tumor permeability and macromolecule delivery. Using diffusion-weighted magnetic resonance (DW-MR) imaging, we demonstrate in multiple patient-derived xenografts (PDX) that responders to short-term sHH inhibitor exposure show increased permeability with proportionate increases in tumor diffusivity. However, a subset of PDXs fails to respond. Responders show co-induction of epithelial-mesenchymal transition (EMT), elevated FGF drive, and distinctly higher nuclear localization of fibroblast growth factor receptor (FGFR1). A pan-FGFR inhibitor (NVP-BGJ398) reverses EMT and nuclear FGFR1 accumulation without compromising the tumor priming effect. This dual-hit strategy of SMO and FGFR inhibition provides a clinically-translatable approach to compromise the profound impermeability of PDAC tumors. Furthermore, clinical deployment of DW-MR imaging could fulfill the essential clinical requirement for patient stratification Total RNA was extracted using RNeasy Minikit (Qiagen, Germantown, MD) according to the manufacturer’s instructions, and included RNase inhibitor (N808-0119, Applied Biosystems, Waltham, MA). RNA-seq libraries from 1μg RNA/sample (low-sample protocol of TruSeq Stranded mRNA Library Prep Kit, Illumina, San Diego, CA) were indexed using an automated IP-Star Compact Liquid Handler (Diagenode, Belgium) amplified, quantified, pooled, and sequenced to yield 25-30M single-end 80BP reads per sample on an Nextseq 500 (Illumina). FASTQ files were de-multiplexed and aligned to pre-compiled reference human (Hg19) and mouse (Mm10) genomes (http://emea.support.illumina.com/sequencing/sequencing_software/igenome.html) using outFilterMultimapNmax= 1 to identify only unique reads. Using the R package Xenofilter, reads were then filtered to generate bam files and mapped to genes with HTSeq to generate gene counts (https://github.com/simon-anders/htseq).