# Single base focal hypermutation cooccurs with structural variation as an early event in advanced prostate tumourigenesis with ancestry specific independence: a multi-ancestral observational study

## TL;DR

This study explores how focal hypermutations in prostate cancer genomes are linked to cancer progression and differ across ancestral groups, particularly highlighting African ancestry.

## Contribution

The study reveals ancestry-specific patterns of kataegis in prostate cancer and its association with clinical outcomes across multiple ancestral groups.

## Key findings

- Kataegis is significantly associated with genomic instability and poor clinical outcomes across all ancestral groups.
- SV-independent and subclonal kataegis show specific associations with African ancestry.
- Kataegis-positive tumors are linked to higher prostate-specific antigen levels in African patients and increased metastatic risk in European patients.

## Abstract

Kataegis, the focal hypermutation of single base positions in tumour genomes, has received little attention with regards to prostate cancer (PCa) molecular features, tumour evolution and associated clinical presentation. Most notably, the impact of this phenomenon is yet to be explored across ancestral lineages representing the extremities of PCa presentation and outcomes, with men of African ancestry disproportionately disadvantaged. The purpose of this study is to address the knowledge gap through African inclusive multi-ancestral interrogation.

We assessed for ancestrally shared and unique molecular, evolutionary and clinical features of kataegis in 669 multi-ancestral whole PCa genomes. Access to raw whole-genome sequenced data allowed for direct single-pipeline comparative analysis between 109 southern African and 57 European derived treatment naïve high-risk-biased primary tumours (74% and 88%) with paired blood samples, further assessed against publicly available 207 Asian high-risk-leaning comparative (65%) and 296 European low-risk-biased alternative (79%) resources. Comparisons between ancestries and risk groups were through Wilcoxon’s rank sum test and Fisher’s exact tests, with P values adjusted by false discovery rate.

Confirming relatively low burdens, we found kataegis to be significantly associated with genomic instability, cancer drivers, and clinical adversity across ancestries (false discovery rate = 7 × 10−6 - 0.04). Notably, kataegis-postive tumours were associated with elevated prostate-specific antigen levels at presentation in African (false discovery rate = 0.002) and higher risk for metastatic progression in European patients (Kaplan-Meier estimator, P=0.03). Enrichment of APOBEC’s context preferences showed more attribution from APOBEC3B than APOBEC3A. Further through analyses of evolution and structural variant (SV) cooccurrence, commonly the ancestry agnostic SV-associated kataegis predominated in the clonal evolutionary state, while the less common the SV-independent kataegis (P=0.002) and subclonal kataegis (P=0.03) showed African specificity.

We found kataegis-positivity to be associated with poor PCa presentation and prognosis, irrespective of patient ancestry. Kataegis-related genomic instability occurring early and late during African derived tumourigenesis, may partly explain the heightened tumour and clinical heterogeneity observed for patients of African ancestry.