Introduction

There were ˜83,730 new cases of bladder cancer estimated in 2021, with ˜17,200 predicted to be lethal.¹ Environmental risk factors such as smoking exposure and aromatic amines increase the risk of developing urothelial carcinoma (UC; see Figure).² The accessibility to genomic sequencing and decreased costs of germline testing enhanced the understanding of genetic predisposition to UC. The Nordic Twin Study reported that the familial risk of developing bladder cancer was almost twofold higher in monozygotic (9.9%) compared to dizygotic twins (5.5%).³ Overall, the study reported a substantial ˜30% heritable risk of bladder cancer.

Lynch Syndrome: A Prototype Cancer Susceptibility Syndrome in UC

Lynch syndrome is a cancer syndrome, which also predisposes to upper tract UC as well as bladder cancer. The cumulative risk of UC by 70 years is ˜7% with the highest risk found in MSH2 germline carriers of pathogenic and likely pathogenic (P/LP) variants.⁴ Routine surveillance for UC in individuals with Lynch syndrome is still not recommended, although urinalysis and imaging in conjunction with screening for other malignancies may be employed.

Multi-Genic Approaches: A Promising Arena for New Discoveries of Rare Variants in UC

Recently, the tides have shifted from a high-penetrance single-gene approach toward a multi-genic approach.⁵ Genome-wide association studies (GWAS) center on the combined effect of low-penetrance genetic variants and have been the mainstay of studying associations between candidate variants and risk of cancer. GWAS in bladder cancer identified 13 bladder cancer susceptibility alleles.⁶ Rare cancer risk variants are understudied although important as they 1) provide additional insight to heritability of UC and 2) they are more recent compared to GWAS risk variants and are more likely to be localized geographically.⁷

Figure. Environmental and genetic risk factors predisposing to urothelial carcinoma. Cartoons in figure created with Biorender.com

Table. Prevalence of germline P/LP variants among most frequently altered genes in UC in 3 previously published studies

Recent work by Carlo et al used a targeted sequencing panel of 431 genes and studied a relatively unselected cohort of 586 patients at a single academic center with UC (see Table).⁸ The frequency of P/LP variants in patients with UC was 14% (80/586) with the majority harboring P/LP variants in DNA damage repair genes (DDR, 66/586, 11%) such as BRCA2 (1.5%), MSH2 (1.4%) and BRCA1 (1.4%). Carlo et al also identified enrichment in BRCA2 and MSH2 compared to controls from the Exome Aggregation Consortium.⁸ Importantly, a quarter of patients carried high-penetrance germline variants and would have been overlooked by current guidelines that are based on family history. Similarly, we analyzed 1,038 relatively selected patients with UC who underwent germline testing of up to 83 genes at a single commercial laboratory and found that 24% harbored P/LP variants, among which 20% were in DDR genes (see Table).⁹ We found significant enrichment of P/LP variants in mismatch repair genes (MLH1 and MSH2) and non-mismatch repair DDR genes (ATM, BRCA2) compared to noncancer patients from the Exome Aggregation Consortium. Importantly, 18.6% of the patients carried clinically actionable variants. In another study, whole exome analysis of germline DNA identified deleterious germline variants in 45 of 80 (56%) patients with UC (see Table). In contrast to prior work,^8,9 deleterious variants in new genes were identified with ITGA7, POLQ, KLK6, EPHB6 and CNDP2 being the most frequent genes harboring deleterious germline variants.¹⁰

Somatic Profiling May Guide Germline Evaluation

National Comprehensive Cancer Network^® guidelines recommend germline testing if a somatic pathogenic variant confers germline risks (eg BRCA1, BRCA2, PALB2, microsatellite instability, Lynch syndrome genes [MLH1, MSH2, MSH6, PMS2], EPCAM).¹¹ Importantly, some somatic alterations are common in some genes such as p53 mutations and may not confer germline implications in the absence of personal or family history of cancers.

Future Directions

Expanding testing among underrepresented populations

As we gain more insights into the genetic determinants of UC, future work should concentrate on genotyping of underrepresented populations, developing algorithms to computationally narrow down candidate variants of unknown significance (VUSs) for functional annotation and ensuring paired tumor-normal sequencing is performed to better understand drivers of disease.

Germline studies in UC to date mostly include White patients with minimal representation of other populations. Non-Whites only constituted a small minority of the large cohorts evaluating germline variants in UC.^8,9 Population genetics has shown us that 1) risk variants differ in prevalence across ancestral populations, 2) some high-risk variants only occur in certain ancestries and 3) risk variants that are most relevant to a certain ancestry may be best detected by studying samples from the ancestry itself, rather than extrapolating from other ancestries.⁷ Future work should focus on sampling more non-White patients to better elucidate germline predisposition in other ancestral population and potentially identify high-risk variants that have gone unnoticed.

Keeping a vigilant eye on VUSs

VUSs represent the bulk of variants detected by genetic testing and pose an important challenge in clinical decision making as the functional impact of VUSs cannot be inferred from the sequence information alone. Data from other tumor types suggest enrichment of VUSs in non-White populations compared to Whites. With larger sequencing panels being implemented, more VUSs will be detected with additional enrichment of these VUSs in non-Whites, further compounding the current problem. We thus advocate for 1) more focused sequencing of non-Whites, 2) development of computational algorithms that nominate recurrent VUSs enriched in ancestral populations with cancer compared to cancer-free controls of the same ancestry and 3) functional validation of candidate VUSs.

Towards more paired tumor-normal sequencing

Although recent work is applauded for prioritizing paired germline-somatic sequencing in UC,^8,10 larger and more diverse cohorts are prudent as colocalization of germline and somatic variants allows better understanding of mechanisms that drive cancer development.

Conclusion

In conclusion, studies in germline genetics of UC gained traction. Future studies focused on underrepresented populations, functional approaches to annotating VUSs and paired tumor-normal sequencing will likely improve our understanding.

Somatic genomic profiling is universally performed in metastatic UC to guide systemic therapy. PARP inhibitors are approved in other malignancies (prostate, breast, ovarian, pancreatic) based on germline alterations. Given the incidence of germline P/LP alterations in unselected patients, and the potential implications for therapy and cascade testing, the momentum toward universal germline evaluation is indeed building.