Aristolochic acid mutational signature defines the low-risk subtype in upper tract urothelial carcinoma

Abstract

Rationale: Dietary exposure to aristolochic acids and similar compounds (collectively, AA) is a significant risk factor for nephropathy and subsequent upper tract urothelial carcinoma (UTUC). East Asian populations, who have a high prevalence of UTUC, have an unusual genome-wide AA-induced mutational pattern (COSMIC signature 22). Integrating mutational signature analysis with clinicopathological information may demonstrate great potential for risk ranking this UTUC subtype. Methods: We performed whole-genome sequencing (WGS) on 90 UTUC Chinese patients to extract mutational signatures. Genome sequencing data for urinary cell-free DNA from 26 UTUC patients were utilized to noninvasively identify the mutational signatures. Genome sequencing for primary tumors on 8 out of 26 patients was also performed. Metastasis-free survival (MFS) and cancer-specific survival (CSS) were measured using Kaplan-Meier methods. Results: Data analysis showed that a substantial proportion of patients harbored the AA mutational signature and were associated with AA-containing herbal drug intake, female gender, poor renal function, and multifocality. Field cancerization was found to partially contribute to multifocality. Nevertheless, AA Sig subtype UTUC patients exhibited favorable outcomes of CSS and MFS compared to the No-AA Sig subtype. Additionally, AA Sig subtype patients showed a higher tumor mutation burden, higher numbers of predicted neoantigens, and infiltrating lymphocytes, suggesting the potential for immunotherapy. We also confirmed the AA signature in AA-treated human renal tubular HK-2 cells. Notably, the AA subtype could be ascertained using a clinically applicable sequencing strategy (low coverage) in both primary tumors and urinary cell-free DNA as a basis for therapy selection. Conclusion: The AA mutational signature as a screening tool defines low-risk UTUC with therapeutic relevance. The AA mutational signature, as a molecular prognostic marker using either ureteroscopy and/or urinary cell-free DNA, is especially useful for diagnostic uncertainty when kidney-sparing treatment and/or immune checkpoint inhibitor therapy were considered.

Keywords: aristolochic acids; clinical outcome; mutational signature; upper tract urothelial carcinoma; whole-genome sequencing.

Figure 1

The AA mutational signature defines etiologically distinct subgroups with favorable outcomes. (A) Bar plot of the number of SNVs attributable to 10 merged signatures in each of the 90 tumors, sorted by hierarchical clustering (dendrogram at top), revealing AA Signature-related (AA Sig, yellow) and non-AA Signature-related (No-AA Sig, orange). Selected clinical features are represented in the bottom tracks. Frozen samples are labeled as T001-T049 and FFPE samples are labeled as T050-T106. (B) The box plot shows the mutation counts of signature 22 mutations in tumors within each subtype. Statistical significance was determined by the Wilcoxon rank test, ***P< 0.001. (C) The bar graph shows the association between the two subtypes and clinicopathologic features. Statistical significance was determined by the Kruskal-Wallis test, *P<0.05; **P<0.01; ***P< 0.001. (D)-(G), Kaplan-Meier survival curves showed that the mutational signature subtypes can predict both CSS and MFS for the whole cohort, as well as for muscle-invasive UTUC patients. CSS: cancer-specific survival. MFS: metastasis-free survival. P-values were calculated by the log-rank test. n, the number of cases.

Figure 2

Field cancerization may contribute to malignant transformation, especially for the AA Sig subtype. (A) Spatial locations of core biopsies of the multifocal AA patient. (B) Trinucleotide contexts for somatic mutations in biopsies from the multifocal patient of the AA Sig subtype. (C) Copy number plots of the core biopsies from the multifocal patient. (D) Phylogenetic relationships of the six samples from the multifocal patient were deciphered using mrbayes_3.2.2. Branch lengths are proportional to the number of somatic mutations separating the branching points. (E) Trinucleotide contexts for somatic mutations in biopsies of another two AA Sig subtype patients. (F) Copy number profiles of another two AA Sig subtype patients.

Figure 3

High neoantigen burden and heavy tumor-infiltrating lymphocytes in the AA Sig subtype. (A) Neoantigen burden was significantly higher in the AA group. Statistical significance was determined by the Wilcoxon rank test (***P<0.001). (B) Positive correlation of the percentage of stromal tumor-infiltrating mononuclear cells (TIMCs) and the number of CD3⁺ lymphocytes in 76 UTUC patients in our cohort (n_{AA Sig}=23; n_{No-AA Sig}=53). (C-D) The percentages of stromal TIMCs (C) and CD3⁺ lymphocytes (D) are shown in each subtype of patients. Statistical significance was determined by the Wilcoxon rank test (**P<0.01, ***P<0.001). HP represents a high-power field. (E-F) Images of TIMCs and CD3⁺ lymphocytes of a representative patient from the AA Sig subtype (E) and the No-AA Sig subtype (F) Triangle highlighting the TIMCs or CD3⁺ lymphocytes in the stromal tumour region. The arrow highlights the TIMCs or CD3⁺ lymphocytes in the intratumor region.

Figure 4

AA mutational signature as “molecular fingerprints” for inferring previous AA exposure and AA Sig subtype patients by urinary cell-free DNA (A) The AA killing curve of the HK-2 and SV-HUC-1 cells. (B) Trinucleotide contexts for mutations in HK-2 cells and AA-treated HK-2 cells. The mutations in AA-treated HK-2 cells were further filtered by untreated HK-2 cells. Trinucleotide contexts for the filtered mutations in AA-treated HK-2 cells are shown in the bottom panel. (C) The box plot shows the ratio of 10 merged signatures in the cell-free DNA at low coverage. (D) The box plot shows the ratio of 12 merged signatures in the selected primary tumors of matched urinary cell-free DNA samples.