In recent years, gene expression profiling has revealed that MIBC can be stratified into different molecular subtypes. While most MIBC studies to date have exclusively used messenger RNA (mRNA) expression to differentiate molecular subtypes, the mammalian transcriptome is comprised of a diverse range of coding (mRNA) and non-coding RNAs, including long non-coding RNAs (lncRNAs). In the TCGA study, the lncRNA transcriptome divided the luminal-papillary mRNA subtype into two groups with distinct prognosis, suggesting that lncRNA expression may offer additional resolution for molecular subtypes.
In this study, we aimed to expand these initial TCGA findings, further exploring the utility of lncRNA expression profiling for MIBC subtyping. Therefore, we selected a list of highly variable lncRNAs for consensus clustering and identified a subset of luminal-papillary MIBC patients with favorable prognosis in a cohort (n=223) that was treated by both neo-adjuvant chemotherapy (NAC) and Radical Cystectomy (RC). As a next step, we evaluated clustering with these de novo selected lncRNAs within the TCGA cohort (n=405), revealing that the lncRNA-mediated subdivision was consistent with, though not identical to, the TCGA lncRNA clustering solution. Importantly, we showed that lncRNA profiling identified a biologically distinct MIBC subgroup with a favorable prognosis. These patient tumors were characterized by wild-type p53 activity, FGFR3 activation, higher SHH expression, and lower EMT, all features that are consistent with less-aggressive disease.
To provide clinical utility to these findings, we developed a stringent, single-sample classifier to identify these good prognostic cases, which we named FGFR3+, due to their high FGFR3 activity and enrichment for FGFR3 mutations. After development in the NAC and TCGA cohorts, the classifier was locked and independently validated in two independent cohorts (UTSW, n=95 & PCC, n=255), where the prognostic utility of the lncRNA subgroup was confirmed in the UTSW cohort. Unfortunately, follow-up data from the PCC cohort was not available, so this cohort could only be used for validation of the biology of the FGFR3+ cases.
We hypothesize that in future clinical practice, these FGFR3+ patients may be candidates for less aggressive treatment. Early results from a phase II trial, showed a 40% overall response rate in patients with FGFR-altered, metastatic urothelial cancer after treatment with Erdafitinib, an FGFR inhibitor.1 Consequently, FGFR3+ cases may be candidates for treatment with FGFR inhibitors instead of NAC, as patients with luminal tumors may benefit less from NAC2 while still being exposed to chemotherapy-related toxicity.
In summary, using the lncRNA transcriptome, we identified a subgroup of luminal-papillary MIBC patients that have very good outcomes. These tumors had a distinct biological profile which was consistent with their less aggressive disease presentation. Moreover, these tumors were found to be enriched in FGFR3 mutations, which may make these patients candidates for FGFR3 inhibitor therapy.
Written by: Joep J. de Jong, Department of Urology, Erasmus MC Cancer Institute in Rotterdam, Netherlands; Joost L. Boormans, Department of Pathology, Erasmus MC University Medical Center Rotterdam, Rotterdam, Netherlands; and Ewan. A. Gibb, Decipher Biosciences Inc., Vancouver British Columbia, Canada.
References:
- Loriot, Y., Necchi, A., Park, S. H. et al.: Erdafitinib in Locally Advanced or Metastatic Urothelial Carcinoma. N Engl J Med, 381: 338, 2019.
- Seiler, R., Ashab, H. A. D., Erho, N. et al.: Impact of Molecular Subtypes in Muscle-invasive Bladder Cancer on Predicting Response and Survival after Neoadjuvant Chemotherapy. Eur Urol, 72: 544, 2017