RNA sequencing reveals PNN and KCNQ1OT1 as predictive biomarkers of clinical outcome in stage III colorectal cancer patients treated with adjuvant chemotherapy

• Published in: International journal of cancer 2019-11, Vol.145 (9), p.2580-2593
• Main Authors: Mini, Enrico, Lapucci, Andrea, Perrone, Gabriele, D'Aurizio, Romina, Napoli, Cristina, Brugia, Marco, Landini, Ida, Tassi, Renato, Picariello, Lucia, Simi, Lisa, Mancini, Irene, Messerini, Luca, Magi, Alberto, Pinzani, Pamela, Mazzei, Teresita, Tonelli, Francesco, Nobili, Stefania
• Format: Article
• Language: English
• Subjects: adjuvant chemotherapy; Aged; Bioindicators; Biomarkers; Biomarkers, Tumor - genetics; Cancer; Cell Adhesion Molecules - genetics; Chemotherapy; Chemotherapy, Adjuvant - methods; Class I Phosphatidylinositol 3-Kinases - genetics; Cohort Studies; Colorectal cancer; Colorectal carcinoma; Colorectal Neoplasms - genetics; Colorectal Neoplasms - pathology; Disease-Free Survival; Down-Regulation - genetics; Explants; Female; Gene Expression Profiling - methods; Genes; Humans; Immune response; KCNQ1OT; KCNQ1OT1 protein; Male; Medical prognosis; Medical research; Middle Aged; miRNA; Multivariate analysis; Mutation - genetics; Neoplasm Staging - methods; Nuclear Proteins - genetics; PNN; Potassium channels (voltage-gated); Potassium Channels, Voltage-Gated - genetics; predictive biomarkers; Prognosis; Pseudogenes; RNA sequencing; Sequence Analysis, RNA - methods; Signal Transduction - genetics; Survival; Transcriptome - genetics; Tumors; Up-Regulation - genetics
• ISSN: 0020-7136; 1097-0215
• DOI: 10.1002/ijc.32326

Five‐year overall survival of stage III colorectal cancer (CRC) patients treated with standard adjuvant chemotherapy (ACHT) is highly variable. Genomic biomarkers and/or transcriptomic profiles identified lack of adequate validation. Aim of our study was to identify and validate molecular biomarkers predictive of ACHT response in stage III CRC patients by a transcriptomic approach. From a series of CRC patients who received ACHT, two stage III extreme cohorts (unfavorable vs. favorable prognosis) were selected. RNA‐sequencing was performed from fresh frozen explants. Tumors were characterized for somatic mutations. Validation was performed in stage III CRC patients extracted from two GEO datasets. According to disease‐free survival (DFS), 108 differentially expressed genes (104/4 up/downregulated in the unfavorable prognosis group) were identified. Among 104 upregulated genes, 42 belonged to olfactory signaling pathways, 62 were classified as pseudogenes (n = 17), uncharacterized noncoding RNA (n = 10), immune response genes (n = 4), microRNA (n = 1), cancer‐related genes (n = 14) and cancer‐unrelated genes (n = 16). Three out of four down‐regulated genes were cancer‐related. Mutational status (i.e., RAS, BRAF, PIK3CA) did not differ among the cohorts. In the validation cohort, multivariate analysis showed high PNN and KCNQ1OT1 expression predictive of shorter DFS in ACHT treated patients (p = 0.018 and p = 0.014, respectively); no difference was observed in untreated patients. This is the first study that identifies by a transcriptomic approach and validates PNN and KCNQ1OT1 as molecular biomarkers predictive of chemotherapy response in stage III CRC patients. After a further validation in an independent cohort, PNN and KCNQ1OT1 evaluation could be proposed to prospectively identify stage III CRC patients benefiting from ACHT. What's new? Survival rates among stage III colorectal cancer (CRC) patients treated with adjuvant chemotherapy vary significantly, due in large part to substantial biological heterogeneity in CRC. Treatment of these patients is potentially facilitated by the use of markers that predict treatment response, though such markers currently are lacking. Here, the authors describe two genes, PNN and KCNQ1OT1, identified from a set of 108 differentially expressed genes in an unfavorable and in a favorable stage III CRC prognosis cohorts, that are associated specifically with survival in patients treated with fluoropyrimidine‐based adjuvant