Genomic insights of bladder cancer obtained from rapid autopsy

From the Hsieh lab, Human Biology Division

Inter- and intra-patient tumor heterogeneity is a challenging and urgent issue in oncology. This heterogeneiety poses a significant obstacle to precision medicine and its promise of individually tailored treatments exploiting tumor vulnerablilities.  Dr. Hsieh (Human Biology Division) is investigating this issue for urothelial carcinoma (UC), in collaboration with Dr. Lam (University of Washington), Dr. Wright (Public Health Sciences Division), and Dr. Montgomery (University of Washington). “We have developed the first bladder cancer focused rapid autopsy program in the world”, explains Dr. Hsieh, “with an extensive genomic database of metastatic bladder cancer tumors, available to the research community worldwide. This has been possible due to an unusual level of team work and a shared vision across UW, SCCA, and FHCRC.” Their work was recently published in JCI Insight.

Most urothelial carcinomas arise from the lower tract (LTUC) involving the bladder, but tumors can also arise from the renal pelvis and ureter (10% of the time), known as upper tract urothelial carcinoma (UTUC). Advanced LTUC and UTUC are both characterized by a poor overall survival, and a better understanding of the mechanism of tumor progression is necessary. The rapid autopsy program established by Dr. Hsieh and colleagues allows the collection of abundant, longitudinal, high quality samples, allowing a unique opportunity to have corresponding metastatic and primary tumor specimens. For this study, tissues were collected from three UTUC and four LTUC patients with whole exome sequencing performed on a total of 37 tumors including primaries, germline, and multiple metastatic samples per patient.

Plot showing the increased mutational burden in LTUC compared to UTUC.
Plot showing the increased mutational burden in LTUC compared to UTUC. Illustration from publication

When analyzing the total number of single nucleotide mutations, Dr. Winters, the first author of the study, and his colleagues observed that both primary and metastatic tumors of UTUC patients displayed a lower mutational burden compared to their LTUC counterparts. They also found that the mutational landscape was mostly conserved between primary and metastatic tumor samples within patients. However, UTUC metastases were found to have more private mutations, implying greater genomic diversity in terms of single nucleotide variations compared to LTUC metastases, suggesting different drivers in these disparate diseases. Importantly, in one patient metastatic samples displayed mutations that were identified  at a low level in primary tumor. This demonstrates that “metastases may be seeded by small clonal populations present in the primary tumor, and not necessarily the predominant tumor population of the primary tumor,” says Dr. Hsieh.

The authors also analyzed differences in structural genomic changes by way of copy number variation. Overall, hierarchical clustering of gene-restricted copy number profile revealed that the genomic variation between metastases and the primary tumor was higher in UTUC than in LTUC tumors. In other words, “upper tract urothelial carcinoma possesses significantly more structural derangements at the genomic level than lower tract urothelial carcinoma,” clarified Dr. Hsieh. Interestingly, MDM2, encoding for an essential negative regulator of the p53 tumor suppressor, was exclusively amplified in UTUC but not LTUC. This may be a mechanism through which genomic instability drives UTUC progression, while LTUC may be preferentially driven  by single nucleotide mutations.

Mapping of genomic copy variation events in UTUC and LTUC tumors. Note the exclusive amplification of MDM2 and FGF receptor ligands in UTUC tumors.
Mapping of genomic copy variation events in UTUC and LTUC tumors. Note the exclusive amplification of MDM2 and FGF receptor ligands in UTUC tumors. Illustration from publication.

Finally, Winters et al. focused on genomic alterations of druggable target genes, and noticed that 70% of the mutations identified were present in all metastatic samples from the same patient. This suggests that personalized targeting of metastases-specific drivers is relevant and promising in urothelial cancer patients.

The researchers have acknowledged, these preliminary findings come from a small patient sample: “We have now increased the size of our rapid autopsy program to more than nineteen individuals. Our plan is to conduct more intricate sequencing studies into the methylome, transcriptome, and beyond.”


This work was supported by the National Institute of Health, the Seattle Translation Tumor Research Program in bladder cancer, the Howard J Cohen Bladder Cancer Foundation and the Stinchcomb Memorial Funds.

Fred Hutch/UW Cancer Consortium member Drs. Ha, Nelson, Montgomery, Wright, Lam and Hsieh contributed to this research.

Winters BR, De Sarkar N, Arora S, Bolouri H, Jana S, Vakar-Lopez F, Cheng HH, Schweizer MT, Yu E, Grivas P, Lee JK, Kollath L, Holt SK, McFerrin L, Ha G, Nelson PS, MontgomeryRB, Wright JL, Lam HM, Hsieh AC. Genomic distinctions between metastatic lower and upper tract urothelial carcinoma revealed through rapid autopsy. JCI Insight. 2019.