Background
UCS is a common entity and usually managed similarly to pure UC but requires focused study. Limited data are available on the genomic profile of UCS. We hypothesized that UCS would have a distinct genomic landscape compared to UC.
Methods
In this IRB-approved retrospective study, pts) with advanced UCS and UC undergoing tumor based comprehensive genomic profiling from a CLIA-certified laboratory were included. An independent genitourinary pathologist reviewed all cases. Pts were determined to have UCS based on presence of any component of squamous differentiation. Pts with UC having any other secondary histology variant were excluded. Genes with alterations (GA) in less than 5% of pts and variants of unknown significance were excluded from the analysis. test was used to compare gene aberration frequency and the p-values were adjusted for false using Benjamini-Hochberg (BH) correction.
Results
87 pts were eligible and included. UCS (n=31) vs UC (n=56): median age, 66 vs 68 years; male vs female: 21/10 vs 48/8Genomic aberrations in both groups are listed in the . GA in KMT2D was found to be significantly enriched in UCS vs UC (15/31 vs 0/56, FDR <0.001, p value = <0.001). GA in CUL4A was numerically higher in UCS vs UC (4/31 vs 1/56, FDR= 0.444, p-value = 0.03). Tumor mutation burden the frequency of genomic aberrations per pt were not significantly different in the groups.
Conclusions
While tumor GA in only KMT2D was significantly enriched in the UCS cohort, it is interesting to note that both KMT2D and CUL4A (higher frequency in UCS) are involved in epigenetic regulation. Identification of underlying molecular targets and biomarkers can guide further drug development in this population. Limitations include small sample size, selection and confounding biases. These hypothesis-generating data need external validation.