Pharmacogenomic profiling of intra-tumor heterogeneity by a large organoid biobank of liver cancer
Description
Inter- and intra-tumor heterogeneity is a major hurdle in primary liver cancer (PLC) precision therapy. Here, we established a PLC biobank, consisting of 399 organoids derived from 144 patients, which recapitulated histopathology and genomic landscape of parental tumors, and were reliable for drug sensitivity screening, evidenced by both in vivo models and patient response. Integrative analysis dissected PLC heterogeneity, regarding genomic, transcriptomic characteristics and sensitivity to seven clinically-relevant drugs, as well as clinical associations. Pharmacogenomic analysis identified and validated multi-gene expression signatures predicting drug response for better patient stratification. Furthermore, c-Jun was revealed as a major mediator to Lenvatinib resistance, through JNK and β-catenin signaling. A new compound (PKUF-01) comprising moieties of Lenvatinib and Veratramine (c-Jun inhibitor) was synthesized and screened, exhibiting a marked synergistic effect. Together, our study characterized the landscape of PLC heterogeneity, developed predictive biomarker panels, and revealed a novel Lenvatinib resistant mechanism for combinatory therapy.
Files
Steps to reproduce
For bulk RNA-seq data analysis, we adopted cutadapt v2.5 and trimmomatic v0.39 to filter adapters and low quality sequences, where QC by using fastqc v0.11.9. Clean data was then aligned to the reference human genome hg19 under default STAR v2.7.3a aligner pipeline and saved the BAM output with the parameter "--quantMode TranscriptomeSAM". Then we extracted the TPM value from RSEM quantification with the genome annotation file gencode.v19.annotation.gtf. We used the log2 ( TPM + 1 ) format as the gene expression profile. Based on the gene classes, all ~20k protein-coding genes were extracted for further analysis. Here, "RNA.rds" is the gene expression profile saved as data.frame in R, and "RNA.meta.txt" marks the sample name in this study, whose clinical, genomic and/or phenotypic data were provided in publication.