Somatic mutation profiles of muscle-invasive bladder cancer in Uganda: A cross-sectional study utilising whole-exome sequencing
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Muscle-invasive bladder cancer (MIBC) is a significant cause of cancer-related mortality in sub-Saharan Africa, yet its genomic landscape remains poorly characterised. In Uganda, nearly 60% of urothelial carcinoma cases are muscle-invasive (MIUC), with a near-equal sex distribution, in contrast to global trends. We aimed to define the somatic mutational landscape of Ugandan MIUC using whole-exome sequencing (WES).
Methods
In this descriptive cross-sectional study, 57 formalin-fixed paraffin-embedded MIUC tissue blocks were consecutively selected. DNA extraction and exome library preparation were performed for all; 47 passed quality control and 33 remained in the final dataset. Sequencing was performed on the Illumina NovaSeq 6000 platform. Reads were aligned to hg19/GRCh37, and variants called using GATK Mutect2 in tumor-only mode with stringent artefact filters. Variants were annotated with COSMIC, ClinVar, REVEL, and classified according to ACMG/AMP guidelines. Eighteen samples were resequenced on the Aviti platform to validate the low frequency of TP53 mutations. Clinical and histopathological data were also reviewed.
Results
All patients presented with gross hematuria, with a median age of 56 years; 17/33 (51.5%) were male. Histological variants occurred in 23/33 (69.7%), most commonly the squamous, microcystic, and nested types (each accounting for 5/23; 21.7%). LVI was present in 28/33 (84.8%). A total of 305 coding mutations were identified, averaging 9.2 ± 6.5 per tumor. Squamous and microcystic variants showed the highest burdens (16.4% and 14.4%). At the nucleotide level, 200/331 (60.4%) substitutions were C>T or C>G, consistent with APOBEC mutagenesis. Missense variants predominated (274/305; 89.8%), followed by frameshifts (31/305; 10.2%). PDE4DIP was the most frequent alteration (8/33; 24.2%), although it was not a canonical driver. Canonical recurrent drivers included FGFR3 (18.2%), FOXA1 (15.2%), BAP1 (15.2%), KMT2C (15.2%), and ERCC2 (12.1%). TP53 mutations were rare (2/33; 6.1%), and Aviti resequencing confirmed the same two variants, with additional low-confidence calls excluded. Mutations were broadly distributed, enriched on chromosomes 1, 16, 7, 19, and 17.
Conclusion
Ugandan MIUC displays a heterogeneous mutational profile, marked by frequent missense mutations, a high VUS burden, low TP53 frequency, and APOBEC signatures, with PDE4DIP a non-canonical gene emerging as the most frequently mutated. These findings underscore the need for African-specific genomic references to guide precision oncology.
