Search

Intrahepatic cholangiocarcinoma (ICC) is one of the most lethal liver cancers. Late diagnosis and chemotherapy resistance contribute to the scarce outfit and poor survival. Resistance mechanisms are still poorly understood. Here, we established a Gemcitabine (GEM) resistant model, the MT-CHC01R1.5 cell line, obtained by a GEM gradual exposure (up to 1.5 &micro, M) of the sensitive counterpart, MT-CHC01. GEM resistance was irreversible, even at high doses. The in vitro and in vivo growth was slower than MT-CHC01, and no differences were highlighted in terms of migration and invasion. Drug prediction analysis suggested that Paclitaxel and Doxycycline might overcome GEM resistance. Indeed, in vitro MT-CHC01R1.5 growth was reduced by Paclitaxel and Doxycycline. Importantly, Doxycycline pretreatment at very low doses restored GEM sensitivity. To assess molecular mechanisms underlying the acquisition of GEM resistance, a detailed analysis of the transcriptome in MT-CHC01R1.5 cells versus the corresponding parental counterpart was performed. Transcriptomic analysis showed that most up-regulated genes were involved in cell cycle regulation and in the DNA related process, while most down-regulated genes were involved in the response to stimuli, xenobiotic metabolism, and angiogenesis. Furthermore, additional panels of drug resistance and epithelial to mesenchymal transition genes (n = 168) were tested by qRT-PCR and the expression of 20 genes was affected. Next, based on a comparison between qRT-PCR and microarray data, a list of up-regulated genes in MT-CHC01R1.5 was selected and further confirmed in a primary cell culture obtained from an ICC patient resistant to GEM. In conclusion, we characterized a new GEM resistance ICC model that could be exploited either to study alternative mechanisms of resistance or to explore new therapies.

Hotspot codon 132 mutations (R132xIDH1m) are frequent in intrahepatic cholangiocarcinoma (ICC), are druggable by anti-IDH1m agents, and could represent a marker of disease progression. Developing an assay to identify R132xIDH1m would provide a useful tool to select patients benefitting from targeted treatments. We tested a quantitative real-time allele-specific polymerase chain reaction (qPCR)-based method to detect the main R132xIDH1m in an Italian ICC series (n = 61) of formalin-fixed paraffin-embedded (FFPE) samples, and on circulating-free DNA samples. The outcomes were compared with nested PCR/Sanger sequencing. Reconstitution experiments of plasmids harboring the different R132xIDH1m mixed with wild-type (WT) DNA demonstrated that qPCR is able to detect at least 2% of all mutated allele. High efficiency was also observed on patient-derived mutated DNA mixed with WT DNA (up to 10% and 0.3 ng of mutated template), qPCR detected 16.4% of mutated samples (one R132G, three R132C and six R132L) while nested PCR/Sanger sequencing only 8.2% (four R132L and one R132G). In a single patient with an R132C-mutated tumor, qPCR was also performed on plasma samples collected at four time-points, observing an increase correlating with disease progression. In conclusion, we developed a qPCR assay which could represent a fast, inexpensive and sensitive tool both for detection of R132xIDH1m in ICC samples and monitoring disease progression from liquid biopsy.