Comprehensive analysis of targetable alterations in urachal cancer by NGS


David Joseph Benjamin, Tolulope Tosin Adeyelu, Andrew Elliott, Sourat Darabi, Thomas Lee, Rana R. McKay, Matthew James Oberley, Arash Rezazadeh

Background: Urachal cancer (UrC) is a rare genitourinary cancer with molecular drivers commonly found in colorectal cancer and harbors fewer actionable alterations. Although prior studies have identified genomic alterations associated with UrC, these lack comprehensive genomic profiling, often using targeted panels, and lack RNA interrogation. This study presents detailed characterization of molecular profiles and the tumor microenvironment of real-world UrC patient samples.

Methods: UrC samples (n = 42) were profiled using next-generation sequencing (NGS) of DNA and RNA. Prevalence was calculated for pathogenic SNVs/Indels and copy number amplifications, dMMR/MSI status assessed by IHC and NGS, PD-L1 expression measured by IHC, and TMB-H defined as $ 10 mut/Mb. Immune cell fractions of TME were estimated using RNA deconvolution. Mann-Whitney U, chi-square, and Fisher exact tests were applied where appropriate, with p-values adjusted for multiple comparisons. Histological features were reviewed by a genitourinary pathologist.

Results: UrC samples were collected from 19 males and 23 females with a median age at sample collection was 58.5 years (range: 24- 86 years). Pathology review revealed the majority were adenocarcinomas (n=41/42, 97.6%), including mucinous (n=17/41, 41.5%) and enteric adenocarcinomas (n=10/41, 24.4%). Pathogenic variants were most commonly observed in TP53 (n = 35, 83.3%), KRAS (n=15, 36%), GNAS (n=5, 12%), SMAD4 (n=4, 10%), PIK3CA (n=4, 10%), STK11 (n=3, 7%), BRCA2 (n=3, 7%), ARID1A (n=2, 5%), PTEN (n=2, 5%), ERBB2 (n=2, 5%), BRAF (n=2, 5%), and APC (n=2, 5%). MAPK pathway alterations were present in 52.3% (n=22) overall, with increased MAPK pathway activation in MAPK-altered UrC (median MPAS: 0.77 vs -0.61, p = 0.066). Loss of heterozygosity was prevalent (n = 9, 29%). While no UrC tumors had dMMR/MSI, 5% (n=2) were TMB-H and 9% (n=3) were PD-L1+. The mean fraction of T cells – CD8 (0.4%) and monocytes (0.0%) represented a smaller fraction UrC TME composition compared to macrophages M1 (5.19%) and neutrophil (6.4%). The median NK cell fraction was significantly higher in MAPK-altered UrC patients (3.5 vs 2.7%, p = 0.022), yet a transcriptional signature of response to immunotherapy (IFN-g) was similar between MAPK-altered and MAPK-WT UrC patients (median IFN: -0.366 vs -0.384, p = 0.052).

Conclusions: This study provides a comprehensive molecular characterization of UrC and the associated immune landscape. While MAPK and DNA repair gene alterations were common, UrC tumors rarely harbored predictive markers of response to immunotherapy, suggesting limited efficacy in this patient population. However, the recurrence of MAPK alterations and associated pathway activation in UrC warrants further investigation of MAPK-targeted therapies in prospective clinical trials.