226P - Prognosis stratification of immunotherapy by a mutational signature in urothelial carcinoma

Background

Urothelial carcinoma (UC) has a highly complex genomic landscape. With the spread of immunotherapy, accurate stratification strategies are needed. As cancer tissues are now frequently screened for specific sets of mutations, a large number of samples has become available for analysis. Classification of patients with similar mutation profiles may help identifying subgroups of patients outcomes. However, classification based on somatic mutations is challenging due to the sparseness and heterogeneity of the data.

Methods

A retrospective study was performed to identify the prognosis-related somatic mutations from 192 UC in The Cancer Genome Atlas (TCGA) database. Cox regression were performed to screen out prognostic genes.

Results

A total of 176 genes were related with immuno-survival. Then, a stepwise multivariate Cox regression analysis was performed, and 14 gene were selected to establish a predictive model. Compared with the wildtype group, the patients with mutated signature had unfavorable to prognosis (p<0.001). ROC curve analysis demonstrated the predictive ability for 1-, 3-, 5-, and 10-year OS, with areas under the curve (AUCs) of 0.7684, 0.667,0.619 and 0.647, respectively. In mutated signature cohorts, the five most frequently mutated genes were TP53 (50%), KMT2D (43%), LRP1B (33%), PER1 (33%), and RNF213 (33%). Dissimilarly, the five most frequently mutated genes were TP53 58%), ARID1A (29%), KMT2D (29%), RNF213 (27%), and KDM6A (26%) in wildtype cohorts, which may imply that different characteristic states have different molecular mechanisms.

Conclusions

These data underline the potential value of using somatic mutations to accurately stratify UC patients into clinically actionable subgroups. This model could reduce overtreatment in UC patients with mutations.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.