br Abbreviations AUC area under
Abbreviations: AUC, area under the curve; BC, Talaporfin sodium (ME2906) cancer; NMIBC, non-muscle invasive bladder cancer; NR-PFBC, non-recurrent control patients in follow-up for BC; SN, sensitivity.
Montalbo et al
76 Montalbo et al
and June 2016 in the different centers. External centers were asked to collect and prepare the urine samples for final processing at the Hospital Clinic.
After excluding nonevaluable samples (see nCounter Elements gene expression analysis section), 546 samples were included in the study (Table I). The grade and stage of the tumors were determined accord-ing to WHO criteria12 and Tumor Node Metastasis (TNM) classification,13 respectively. Tumors were classified into 3 categories according to their risk: non-high-risk NMIBC, high-risk NMIBC and muscle invasive BC. Bacillus Calmette Guerin (BCG) treat-ments were applied following European Association of Urology guidelines.1
Urine cytology. Urine cytologies were performed according to Papanicolau staining and were evaluated by expert pathologists in each participating center blinded to the patient’s clinical history. The results were either considered as positive, negative, or suspi-cious. Suspicious cytology was defined as those sam-ples that contained cells with morphologies that could not be clearly classified as tumor cells or normal cells.
Urine processing and RNA isolation. Around 50 100 mL of voided urine was collected from all patients of the series. Urine samples were processed as previously described.14 Briefly, urine samples were mixed with 1 of 25 volumes of 0.5 M EDTA, pH 8.0, stored at 4˚C and processed within the next 24 hours. Urines were centrifuged at 1000 £ g for 10 minutes, the cell pellets were resuspended in 1 mL of TRIzol reagent (Invitro-gen, Carlsbad, California) and frozen at ¡80˚C until Ribonucleic Acid (RNA) extraction. RNAs from the urinary cell pellets were extracted following manufac-turer’s instructions and quantified with a Nano-Drop1000 (NanoDrop Technologies, Wilmington, Delaware). In the discovery phase, RNA integrity was assessed with Agilent Bionalyzer by using Eukaryote Total RNA Nano kit (mean Integrity Number (RIN) value was 2.5; range: 0 8.9).
Library preparation and sequencing method. RNA sequencing of 20,000 RefSeqs was performed in the dis-covery phase. Ion AmpliSeq Transcriptome Human Gene Expression Kit (Thermo Fisher Scientific, P/N A26325) was used for library preparation. Briefly, cDNA was synthesised from total RNA by using the SuperScript VILO cDNA Synthesis kit (Thermo Fisher Scientific) from 10 ng of RNA. Then, cDNA was amplified using Ion AmpliSeq technology. Finally, after a partial diges-tion of the primer sequence with FUPA reagent, ligation of the barcoded adapters, and purification by Agencourt AMPure XP Reagent of the amplified cDNA, the library was quantified with Ion Library TaqMan Quantitation Kit (Thermo Fisher Scientific). Translational Research June 2019
An input concentration of 8 pooled libraries copy/ Ion Sphere Particles (ISPs) was added to the emulsion Polymerase Chanin Reaction (PCR) master mix and the emulsion was generated using the Ion Chef Instru-ment (Thermo Fisher Scientific) using the Ion PI Hi-Q Chef Kit (Thermo Fisher Scientific). Template-positive Ion Sphere Particles were enriched, and sequencing was undertaken using an Ion PI v3 (Thermo Fisher Sci-entific) on the Ion Proton sequencer (Thermo Fisher Scientific) using the Ion PI Hi-Q Sequencing 200 Kit (Thermo Fisher Scientific).
Read alignment and differential gene expression analysis. Partek Flow 6.0 (http://www.partek.com/) was used to analyze AmpliSeq transcriptome data. Briefly, primary read alignment for AmpliSeq sequencing data of all samples was performed using the Torrent Mapping Alignment Program. After quantification, features with a minimum 1.0 were excluded. Between-sample normal-ization at gene level was performed using the trimmed mean method followed by quantile normalization. Gene-specific analysis was used to identify a statistical model that is the best for a specific transcript, and then the best model was used to test for differential expression.