Abstract 1309: Distinct molecular subtypes and a high diagnostic urinary biomarker of upper urinary tract urothelial carcinoma

2020 
[Backgrounds] Upper urinary tract urothelial carcinomas (UTUCs) are rare malignancies with a histological similarity to urothelial bladder carcinomas (UBCs). Unlike UBCs, little is known about the genetic basis of UTUCs, for which no useful biomarkers are currently available. [Material & Methods] Surgical specimens and matched normal samples were obtained from 209 UTUC patients with various stages and were subjected to integrated molecular analysis including whole exome/transcriptome sequencing. We also investigated a possibility of UTUC diagnosis using urinary sediment-derived DNA. [Result] Hypermutation (≥20/Mb) was observed in 15 samples (6.7%), of which 12 harbored biallelic mismatch repair genes alterations suggestive of Lynch syndrome. The most frequently affected driver genes included TERT (promoter) (50%), followed by KMT2D (46%), FGFR3 (44%), CDKN2A (44%), and TP53 (36%). Mutually exclusive patterns were observed among alterations in FGFR3, RAS (H-/K-/NRAS), TP53, and MDM2, which covered 93.8% of the entire cohort. Approximately 40% of the samples showed highly complexed karyotype (CK) with frequent aneuploidy and chromothripsis. Based on these findings, UTUC cases were classified into 5 genetic subgroups: hypermutated, FGFR3-mutated, RAS-mutated, TP53-mutated with CK, and other cases with none of these features, which were significantly associated distinct clinical outcomes. While hypermutated and FGFR3 subtypes showed favorable prognosis, the TP53/CK subtype exhibited an aggressive clinical course and a shorter survival. Expression-based molecular classification identified unique gene expression subtypes: clusters 1-5, which showed a strong correlation with the mutation-based classification. Of interest, cluster 3 was highly enriched for TP53/CK and showed high expression of PDL1 and PDL2, suggesting a possible role of checkpoint inhibitions. The sequencing analysis of urinary sediment-derived DNA successfully captured those mutations and copy number alterations (CNAs) that had been detected in the corresponding preoperative samples in 84.4% of the cases (38/45), which was significantly higher than the positive diagnosis in urine cytology (class 4 and 5, 31.1%). Failure of detecting mutations were found in 5 tumors that were associated with severe hydronephrosis, which might prevented tumor-containing urinary flow. None of the post operative (0/34) and control (0/19) samples showed mutations/CNAs, including two with positive cytology (class4), both of which were diagnosed as benign pelvis tumors after surgery. [Conclusion] UTUC are classified into 5 distinct genetic subtypes characterized by unique profiles of gene mutations, expression and clinical outcomes, which help an optimal choice of therapeutics, including novel molecular-targeted drugs and checkpoint-inhibitors. Sequencing urinary sediments is a useful, non-invasive biomarker for precise diagnosis and proper treatment. Citation Format: Yoichi Fujii, Yusuke Sato, Hiromichi Suzuki, Tetsuichi Yoshizato, Kenichi Yoshida, Yuichi Shiraishi, Taketo Kawai, Tohru Nakagawa, Hiroaki Nishimatsu, Toshikazu Nishimatsu, Masashi Sanada, Hideki Makishima, Satoru Miyano, Haruki Kume, Seishi Ogawa. Distinct molecular subtypes and a high diagnostic urinary biomarker of upper urinary tract urothelial carcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1309.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []