Skip to main content
Download PDF

Targeted gene sequencing reveals disparate genomic mutations between young and older adults in renal cell carcinoma

BMC Cancer volume 24, Article number: 1011 (2024) Cite this article

Abstract

Background

Renal cell carcinoma (RCC) is a type of cancer that can develop at any point in adulthood, spanning the range of age-related changes that occur in the body. However, the specific molecular mechanisms underlying the connections between age and genetic mutations in RCC have not been extensively investigated.

Methods

Clinical and genetic data from patients diagnosed with RCC were collected from two prominent medical centers in China as well as the TCGA dataset. The patients were categorized into two groups based on their prognosticated age: young adults (YAs) and older adults (OAs). Univariate and multivariate analysis were employed to evaluate the relationships between age and genetic mutations. Furthermore, a mediation analysis was conducted to assess the association between age and overall survival, with genetic disparities serving as a mediator.

Results

Our analysis revealed significant differences in clinical presentation between YAs and OAs with RCC, including histopathological types, histopathological tumor stage, and sarcomatoid differentiation. YAs were found to have lower mutation burden and significantly mutated genes (SMGs) of RCC. However, we did not observe any significant differences between the two groups in terms of 10 canonical oncogenic signaling pathways-related genes mutation, telomerase-related genes (TRGs) mutation, copy number changes, and genetic mutations associated with clinically actionable targeted drugs. Importantly, we demonstrate superior survival outcomes in YAs, and we confirmed the mediating effect of genetic disparities on these survival outcome differences between YAs and OAs.

Conclusion

Our findings reveal previously unrecognized associations between age and the molecular underpinnings of RCC. These associations may serve as valuable insights to guide precision diagnostics and treatments for RCC.

Peer Review reports

Introduction

Renal cell carcinoma (RCC) accounts for approximately 4.1% of all new cancer cases, with an estimated 73,750 new cases expected to be diagnosed in the United States in 2020 [1]. The most frequently diagnosed age range for RCC is 65–74 years, while less than 9.0% of new cases occur in young adults under the age of 45 [2]. Aging is a prominent risk factor for RCC [3,4,5]. Previous studies have demonstrated a significant difference in survival between young (YAs) and older adults (OAs), despite variations in the comparison groups used. YAs are more likely to exhibit lower primary tumor classification (T stage) and tend to experience improved progression-free and cancer-specific survival compared to OAs [6]. However, the specific reasons for this result are currently unclear.

The distinct prognosis between YAs and OAs may be attributed to biological heterogeneity. Although age-associated molecular landscapes have been reported in several cancer types [7,8,9], there is limited research on the genetic and clinical differences between YAs and OAs in RCC. Particularly, there is a lack of related studies focusing on the Chinese population. Wang et al. ‘s [10] pan-cancer analysis revealed certain genetic mutations that differ between adolescents and young adults and older adults in RCC, including TERTp mutations and TEF3 mutations. However, it is noteworthy that adolescents included in their study often represent a different histopathologic subtype than those commonly found in adults [11,12,13]. Resources such as The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium have been utilized to study age-associated genetic alterations [14,15,16,17]. However, these resources primarily focus on pan-cancer analysis and lack in-depth analysis of mutations in RCC. Consequently, the age-related genetic landscape of RCC in adult patients remains unclear.

In this study, we aim to conduct a comprehensive examination of age-associated molecular differences in RCC using data from two prominent medical centers in China and TCGA. Our analysis systematically identifies somatic and germline mutations, copy-number alterations (CNAs), and assesses their effects on patient survival. By investigating the genetic and clinical disparities between YAs and OAs with RCC, we aim to enhance our understanding of this disease and shed light on the underlying mechanisms driving age-related differences.

Methods

Data sources and study population

We collected patients with confirmed RCC based on pathology reports from Nanjing Drum Tower Hospital and Jinling Hospital, Clinical School of Medical College, Nanjing University (Nanjing cohort). The enrollment period was from November 1st, 2021, to November 1st, 2022. Fresh tumor samples or formalin-fixed paraffin-embedded (FFPE) DNA were extracted for next-generation sequencing (NGS) testing, while blood leukocyte DNA was sequenced for germline analysis. Prior to NGS testing, we collected demographic, clinical, and pathologic data from medical records, including age, gender, histopathological type and grade, sarcomatoid differentiation, and tumor stage, among other details.

We also obtained somatic mutations and clinical data from the TCGA database (TCGA cohort) (https://gdc.cancer.gov). In this study, individuals aged 44 years old or younger were classified as YAs, while those older than 44 years old were considered OAs. We excluded samples with unknown ages in both cohorts. Additionally, we utilized RCC data from the SEER database (http://www.seer.cancer.gov) to explore the relationship between age and overall survival and cancer-specific survival. To ensure data consistency with the Nanjing cohort and TCGA cohort, we specifically selected clinical and pathological information from the SEER database for patients who underwent surgical treatment.

Next generation sequence

Tumor sample were collected and the next generation sequencing tests of all samples were performed at ChosenMed Technology (Beijing) Co., Ltd., Beijing, China. genetic DNA extraction by QIAamp DNA FFPE Tissue Kit and library preparation with TruSightâ„¢ Oncology 500 (TSO 500) Library Preparation Kit (Illumina, San Diego, CA, United States). The final libraries were sequenced on the Illumina NextSeq 550Dx platform with the paired-end run of 150 base pairs. The BWA-MEM (version 0.7.11) alignment algorithm were used to sequence alignment to the human genome (hg19). SAMtools (version 1.3) was used to do bam-sam conversions. Genome Analysis Toolkit (GATK, version 3.6) module IndelRealigner was used to perform local realignment of indels. CNAs including amplification and deletion were identified by CRAFT copy-number callers from the TSO500 pipeline and fusions having at least 3 unique supporting reads, one of which being a split read crossing the fusion breakpoint, are considered as candidate fusions. The process of SNVs and Indels mutation calling, and reads filtering were performed following the description in the previous study. Germline variants could be filtered out using an in-house built database, and all parameters were set according to the previous workflow.

Statistical analysis

Baseline clinical and demographic features were reported as the mean ± standard deviation or percentage. We performed comparisons of clinical and molecular characteristics between YAs and OAs using appropriate statistical tests. Categorical variables were compared using Fisher’s exact test or chi-square test, while continuous variables were compared using either Student’s t-test or Wilcoxon test. To account for potential confounding factors other than age, we constructed a multivariable model. This model controlled for gender, histopathological types, histopathological grade, sarcomatoid differentiation, and tumor stage based on the results of the clinical data analysis following the univariate analysis. Since metastasis and tumor stage were dependent, we excluded metastasis from the multivariable models.

Survival analysis was conducted using the log-rank test, and the results were visualized with Kaplan-Meier curves. Hazard ratios (HRs) were calculated using the Cox proportional hazards model. To assess the association between age and oveall survival through the mediator of genetic disparities, we performed a mediation analysis using the method described by VanderWeele [18].

To address missing data, we employed the random forest method for imputation. Imputations were performed considering the proportion of incomplete cases. For continuous variables, we assessed the imputation accuracy using the normalized root mean squared error. For categorical variables, the evaluation metric employed was the proportion of falsely classified cases.

All statistical analyses were conducted using R v4.2.0 or GraphPad Prism v8.0.

Results

Disparities of clinical characteristics between YAs and OAs with RCC

In order to analyze the clinical characteristics and genetic differences among YAs and OAs with RCC in Chinese population, we first collected samples from 94 patients who were from two major medical centers in Nanjing region (Nanjing cohort). All patients underwent robot-assisted partial nephrectomy or radical nephrectomy, and genetic sequencing was performed on all tumor samples. Cases without precise age information were excluded from the study. YA samples, with an average age of 33.38 ± 6.12, accounted for 44.7% (n = 42). The sex distribution did not differ significantly between YAs and OAs (69.0% in YAs vs. 69.2% in OAs, P = 0.985; percentages are listed in the same order throughout the work). Table 1 presents the clinical and epidemiological characteristics of RCC in both YAs and OAs.

Table 1 Disparities of clinical characteristics between YAs and OAs with RCC
Full size table

We initially examined the histopathological types of RCC in both YAs and OAs. The RCC subtypes included clear cell RCC (ccRCC), non-clear cell RCC (nccRCC) such as papillary RCC (pRCC), chromophobe RCC (chRCC), MiT/TFE family RCC, unclassified RCC, fumarate hydratase-deficient RCC, renal oncocytoma, spindle cell carcinoma, and dedifferentiated liposarcoma. We observed a differential distribution of ccRCC between YAs and OAs (57.1% vs. 84.6%, P = 0.003). The proportions of patients diagnosed with high-grade RCC (G3-4) varied between the two groups in the Nanjing cohort, indicating that YAs have a higher proportion of well-differentiated malignant tumors compared to OAs (23.8% vs. 69.2%, P < 0.001). Additionally, these differences in tumor grade were confirmed by disparities in tumor stage and sarcomatoid dedifferentiation between the two groups, with a higher percentage of YAs in stages I-II compared to OAs (85.7% vs. 55.8%, P = 0.002) and a lower prevalence in sarcomatoid dedifferentiation (2.4% vs. 23.1%, P = 0.004). However, no significant difference was observed in the occurrence of metastasis between the groups (2.4% vs. 7.7%, P = 0.254).

Next, we conducted a comparative analysis of clinical disparities between YAs and OAs in the TCGA cohort. As shown in Table 1, we observed a significant decrease in the prevalence of ccRCC and sarcomatoid dedifferentiation among YAs compared to OAs (48.5% vs. 59.5%, P = 0.031; 29.9% vs. 20.2%, P = 0.036). However, there were no significant differences in other clinical characteristics (refer to Table 1 for specific details).

Genetic mutational landscape of RCC

Figure 1a illustrates the landscape of somatic (left) and germline (middle) alterations of the Nanjing cohort and somatic (right) alterations of the TCGA cohort which generated from 1123 gene (Table S1) panel sequencing data of the Nanjing cohort. The analysis considered only non-synonymous coding mutations and indels in this study.

Fig. 1
figure 1

Landscape of mutations of RCC. a Alteration landscape of the top 20 mutated genes in RCC. left, somatic mutations in Nanjing cohort; middle, germline mutations in Nanjing cohort; right, somatic mutations in TCGA cohort. b Comparison of somatic mutations between Nanjing cohort and TCGA cohort. c Comparison of mutations’ number between somatic mutations and germline mutations in Nanjing cohort. d Comparison of SMGs mutations between Nanjing cohort and TCGA cohort. e Comparison of SMGs mutations between somatic and germline mutations in Nanjing cohort. *p < 0.05, ** p < 0.01, ***p < 0.001, ****p < 0.0001, ns, no significant

Full size image

A total of 329 somatic changes were identified, including 186 nonsynonymous coding mutations and 143 indels mutations in the Nanjing cohort. In the TCGA cohort, we identified 4360 somatic changes, including 4009 nonsynonymous coding mutations and 351 indels mutations. Overall, we observed similar mutation rates between the Nanjing and the TCGA cohorts (Median Nanjing = 3, Median TCGA = 4; Fig. 1b, P = 0.187). We also identified 573 germline mutations in the Nanjing cohort, which includes 530 nonsynonymous coding mutations and 43 indels mutations. The median mutation rate of germline genes was significantly higher than that of somatic genes in the Nanjing cohort (Median somatic = 3, Median germline = 5; Fig. 1c, P = 0.003).

The top 10 frequently somatic mutated genes in our cohorts are listed in Fig. 1a. Notably, VHL mutation was the most prominent variation in the Nanjing cohort (43.6%), followed by PBRM1 (19.1%), SETD2, and BAP1 (10.6%). In the TCGA cohort (Fig. 1a), the most prominent mutational gene was VHL (10.6%), followed by PBRM1 (8.3%), SETD2 (5.4%), and KMT2C (4.4%). Figure 1a also showed the top 10 frequently germline mutated genes in the Nanjing cohort. NRG1 mutation was the most prominent variation (14.9%), followed by AR (11.7%), CHEK2 (9.6%), ARID1B (8.5%), and MLH1 (8.5%), indicating different mutated gene profiles in somatic and germline mutations for patients with RCC.

The significantly mutated genes (SMGs) of RCC were reported in previous studies [19], and the mutation frequencies of most putative RCC driver genes were similar in Nanjing and TCGA RCC patients (Fig. 1d). However, the Nanjing cohort had significantly higher mutation frequencies in VHL (43.6% vs. 10.6%, P < 0.0001), PBRM1 (19.1% vs. 8.3%, P = 0.002), BAP1 (10.6% vs. 3.5%, P = 0.004), and KDM5C (5.3% vs. 1.4%, P = 0.021) compared to the TCGA cohort. 10 of the SMGs were germline mutations in the Nanjing cohort, and the mutation rate was significantly lower than that of somatic mutations (18.1% vs. 56.4%, P < 0.0001) (Fig. 1e).

Disparities of genetic mutational landscape between YAs and OAs

To characterize genetic disparities between YAs and OAs with RCC, we first depicted the characteristics of the mutational landscape of the YAs and the OAs group in the two cohorts. As shown in Fig. 2a, in the group of YAs from Nanjing cohort, out of the 115 somatic mutation genes, the most prominent mutation was VHL (35.7%), followed by PBRM (11.9%), AXIN2 (7.14%), KMT2D (7.14%), PHOX2B (7.14%), and SMARCA4 (7.14%). Similarly, in the group of OAs with 214 somatic variations, the most prominent mutation was also VHL (50.0%), followed by PBRM1 (25.0%), SETD2 (19.2%), BPA1 (17.3%), ARID1A (13.5%), and ARID1B (9.6%). Regarding germline mutations in the Nanjing cohort, the most common were NRG1 and POLE mutations (11.9%), followed by CHEK2 (9.5%), MSH6 (9.5%), NOTCH2 (9.5%), PALB2 (9.5%), and TSC2 (9.5%) among YAs. In OAs, NRG1 mutations (17.3%) were dominant, followed by AR (15.4%), ARID1B (13.5%), MLH1 (11.5%), CHEK2 (9.6%), and FANCA (9.6%) (Fig. 2a). In the TCGA cohort, the most prominent mutations among YAs were TP53 (7.3%), followed by PBRM1 (4.6%), VHL (4.6%), ERCC6 (2.8%), KMT2C (2.8%), and KMT2D (2.8%). Among OAs, VHL mutation (11.4%) was the most common, followed by PBRM1 (8.8%), SETD2 (6.1%), KMT2C (4.6%), MACF1 (4.3%), and BAP1 (3.9%).

Fig. 2
figure 2

Landscape of mutation of YAs and OAs. a Top 10 mutated genes in the YAs and OAs. b Comparison of genetic mutations between YAs and OAs. *p < 0.05, ** p < 0.01, ***p < 0.001, ****p < 0.0001, ns, no significant

Full size image

Then, we compared the mutational landscape of somatic and germline genes in the Nanjing cohort, respectively. A total of 115 somatic mutations and 267 germline mutations were identified in YAs, compared to 214 somatic mutations and 306 germline mutations in OAs. We observed significantly different somatic mutations (Median YAs = 2, Median OAs = 4; Fig. 2b, P = 0.016) and comparable germline mutation rates (Median YAs = 6, Median OAs = 7; Fig. 2b, P = 0.849) between YAs and OAs. In the TCGA cohort, there were 250 somatic mutations and 4,110 somatic mutations observed in YAs and OAs, respectively. Similarly, a lower mutational burden was found in YAs of the TCGA cohort (Median YAs = 3, Median OAs = 6; Fig. 2b, P < 0.0001).

Genetic disparities between YAs and OAs

In order to assess genetic differences between YAs and OAs, we initially compared the overall tumor mutation burden (TMB), a characteristic strongly associated with age. Notably, we observed significant disparities between the two groups, with lower TMBs observed in YAs. In the Nanjing cohort, the median TMB for YAs was 2.850, while for OAs it was 4.765 (Fig. 3a, P = 0.0002). Similarly, in the TCGA cohort, the median TMB for YAs was 0.4626, compared to 0.8684 for OAs (Fig. 3a, P < 0.0001).

Fig. 3
figure 3

Mutation disparity of RCC. a Comparison of TMB between YAs and OAs. b Comparison of SMGs mutations between YAs and OAs. c and d Comparison of mutation of 10 canonical oncogenic signaling pathways between YAs and OAs. e Comparison of TRGs mutation between YAs and OAs. *p < 0.05, ** p < 0.01, ***p < 0.001, ****p < 0.0001, ns, no significant

Full size image

We next compared the SMGs mutation between YAs and OAs using Fisher’s exact test. As shown in Fig. 3b), 50.00% YAs and 73.08% OAs had at least one SMGs mutation and there was significant difference between YAs and OAs in Nanjing cohort (P = 0.032). The rate of SMGs mutations was lower in TCGA cohort, 22.94% YA and 37.89% OA patients had at least one SMGs mutation, and there was significant difference between YAs and OAs (P = 0.002, Fig. 3b).

As 89% of tumors had at least one driver alteration in 10 canonical oncogenic signaling pathways [20], we conducted further investigations on these pathways and discovered that 45.7% of YAs and 61.0% of OAs exhibited at least one mutation in these pathways. However, no significant difference was observed within the Nanjing cohort (Fig. 3c). Similarly, within the TCGA cohort, 14.0% of YAs and 25.8% of OAs had mutations in these pathways, with a similar lack of significant difference (Fig. 3c). Specifically, in the Nanjing cohort, YAs showed lower mutation frequencies compared to OAs in the Cell Cycle, NRF2, and TP53 signaling pathways (9.5% vs. 15.4%; 2.4% vs. 3.8%; and 9.5% vs. 17.3%); while in the HIPPO, MYC, NOTCH, PI3K, RTK/RAS, TGF-β, and WNT signaling pathways, the mutation frequencies were higher in the YAs group compared to OAs (2.4% vs. 0%; 9.5% vs. 7.7%; 21.4% vs. 17.3%; 14.3% vs. 11.5%; 11.9% vs. 11.5%; 4.8% vs. 1.9%; and 11.9% vs. 7.7%). In the TCGA cohort, YAs only had higher mutation frequencies than OAs in the MYC and TP53 signaling pathways (1.8% vs. 1.6% and 6.4% vs. 5.4%) (Fig. 3d).

The activity of telomerase is crucial for maintaining the immortality of cancer cells. Based on this premise, we hypothesized that cancers in YAs are more likely to originate from telomerase-competent cells. To investigate this hypothesis, we examined mutations in telomerase-related genes (TRGs) between YAs and OAs in the both cohorts. As shown in Fig. 3e, we found that 9.52% of YAs and 13.46% of OAs exhibited mutations in at least one TRGs in the Nanjing cohort. In the TCGA cohort, the percentages were 2.75% for YAs and 2.89% for OAs. However, no significant difference was observed between YAs and OAs in either cohort. This pattern persisted when we analyzed individual TRGs, as there were no notable differences between YAs and OAs (Fig. 3e).

Factors affecting genetic disparities between YAs and OAs

To further explore the differences in mutation rates between YAs and OAs in the two cohorts, we identified five significant genes in the Nanjing cohort and TCGA cohort using Fisher’s exact test (Table S2). Wanting to determine if factors other than age influenced these differences, we adjusted for confounders such as gender, histopathological types, histopathological grade, sarcomatoid differentiation, and tumor stage. Multivariate logistic regression analysis revealed that age was not an influencing factor for any gene mutation, except for PBRM1 in the Nanjing cohort. In the TCGA cohort, seven genes showed an association with age after adjusting for confounders (Table S2). When considering the union of the two methods, we identified five genes in the Nanjing cohort and ten genes in the TCGA cohort. Among them, one gene was identified in both methods within the Nanjing cohort and two genes were shared between the methods in the TCGA cohort. Notably, NOTCH2 and AXIN2 were exclusively detected in YAs in the Nanjing cohort, while MAP4K8 was solely found in YAs and SETD2 was exclusively found in OAs in the TCGA cohort (Fig. 4a).

Fig. 4
figure 4

Factors of mutation disparity. a Genes with different mutational rates between YAs and OAs identified by Fisher’s exact test and multivariate logistic regression analysis. Yellow, solely by Fisher’s exact test; Blue, solely by multivariate logistic regression analysis; Red, both by the two methods. b Clinical factors affecting differential mutational rates using multivariate logistic regression models

Full size image

We attempted to analyzed the factors which contribute to the genetic disparity. In the Nanjing cohort, among the three genes identified by the multivariate logistic regression analysis model, PBRM1 mutation was associated with YAs, VHL with histopathological types and BAP1 with gender and histopathological grade (Fig. 4b). Among the 30 genes in the TCGA cohort, 7 were solely explained by the YA status of the patients. For the remaining genes, one or multiple factors, including gender, histopathological types, histopathological grade, sarcomatoid differentiation, and tumor stage, contributed to the mutations, except for age. These results collectively indicate that mutations in certain genes are influenced by age, and pathological factors may also contribute significantly to the genetic landscape of cancer.

Disparities of gene level CNA between YAs and OAs

To evaluate genetic instability, we conducted a comparison of CNA between YAs and OAs in TCGA cohort. The number of CNAs observed across the panel genome was 99 in YAs and 770 in OAs patients. The median number of CNAs per sample was 13, with a range of 3–490. This includes a median of 5 amplifications, ranging from 1 to 156, and a median of 8 deletions, ranging from 1 to 334. The number of CNAs per sample showed no significant difference between YAs and OAs (P = 0.658, Fig. 5a). Similarly, no significant difference was observed between the two groups in terms of amplification (P = 0.815, Fig. 5a) and deletion alterations (P = 0.815, Fig. 5a). We conducted a separate comparison of CNA types between the two groups and found that deletion mutations were more commonly observed in patients with RCC, regardless of age (P = 0.0001 and < 0.0001, Fig. 5b).

Fig. 5
figure 5

CNA level of RCC. a Comparison of CNA between YAs and OAs. b Comparison of CNA types in YAs and OAs. *p < 0.05, ** p < 0.01, ***p < 0.001, ****p < 0.0001, ns, no significant

Full size image

Genetic disparities of clinical actionability between YAs and OAs

Our sought to compare actionable mutations between YAs and OAs in the two cohorts using the OncoKB database, which provides information on targeted drugs in cancer. In the Nanjing cohort, we identified a total of 23 actionable alterations, all at level 4. In the TCGA cohort, there were 2 alterations at level 2, 1 at level 3 A, and 90 at level 4, making a total of 93 actionable alterations. According to Figs. 6 and 26.19% (11 out of 42) of tumors in YAs and 23.08% (12 out of 52) in OAs in the Nanjing cohort exhibited level 4 alterations. There was no significant difference observed between the two groups (P = 0.7270). In the TCGA cohort, 5.5% (6 out of 109) of tumors in YAs had level 4 alterations, and 0.25% (2 out of 797) had level 2 alterations, 0.13% (1 out of 797) had level 3 A alterations, and 11.29% (90 out of 797) had level 4 alterations in OAs. No significant difference was observed between the two groups, too (P = 0.081).

Fig. 6
figure 6

Clinical actionability of RCC. *p < 0.05, ** p < 0.01, ***p < 0.001, ****p < 0.0001, ns, no significant

Full size image

Genetic disparities mediated the difference of survival outcomes between YAs and OAs

Finally, we investigated the impact of age on overall survival and cancer-specific survival in patients with RCC. We compared the overall survival of patients using data from the TCGA cohort and the SEER database, and cancer-specific survival using the SEER data (refer to table S3 for clinical and pathological characteristics of the SEER data). As depicted in Fig. 7, YAs exhibited superior overall survival and cancer-specific survival rates. Furthermore, these findings were corroborated in a multivariate analysis that accounted for confounding factors (Fig. 7).

Fig. 7
figure 7

Kaplan–Meier curves and forest plot showing overall and cancer-specific survival of patients in TCGA and SEER cohort

Full size image

To examine whether genetic disparities played a role in mediating the association between age and survival outcomes in RCC patients, we conducted a mediation analysis. Specifically, we assessed the association between YAs and overall survival in the TCGA cohort, considering tumor mutation burden (TMB), mutations in significantly mutated genes (SMGs), and genes with different mutations (GDMs) between YAs and OAs. The outcomes, presented in Table 2, demonstrate the mediation effects of genetic disparities on the association between age and overall survival. After adjusting for genetic disparities, the risk of death was found to be lower for YAs compared to OAs. The mediation analysis revealed that TMB partially mediated this association, with an indirect association hazard ratio (HR) of 1.160 (95% CI, 1.055–1.303). The proportion mediated was 26.7%. Although SMGs mutations and GDMs could directly mediate the association between age and survival outcomes, their contribution may be negligible due to their little proportion mediated (Table 2).

Table 2 Multivariable Cox regression analysis of age for overall survival in patients with RCC and mediation analysis of genomic disparities to age
Full size table

Discussion

At present, there is limited research on the age-related genetic landscape in pateints with RCC, especially in Chinese population. One reason for this is the rarity of YA cases in adult or pediatric cancer projects. Studies that examine the correlation between age and genetics primarily focus on comparing early-onset and late-onset tumors or pan-cancer [10, 14,15,16,17]. Consequently, there is a knowledge gap in understanding YAs with RCC due to the lack of comparative studies. However, the incidence of RCC in YAs is increasing. Therefore, it has become necessary to study the genetic differences related to age in patients with RCC.

In this study, we systematically investigated the clinical, pathological, and genetic characteristics of YAs and OAs with RCC in both the Nanjing and TCGA cohorts. To the best of our knowledge, this is currently the only study on the age-related genetic differences in RCC patients in the Chinese population. We identified significant differences in clinical presentation between the two age groups in both cohorts, including histopathological types, histopathological grade, stage, and sarcomatoid differentiation. However, there were substantial differences observed between the two age groups in the Nanjing cohort, which were not observed in the TCGA cohort, suggesting that these differences may be due to racial disparities. Tumor staging and grading are crucial for assessing the prognosis of tumors. However, even within the same stage and grade of RCC, there remains heterogeneity in clinical prognosis. Therefore, identifying new molecular markers that can predict the progression and prognosis of RCC is of paramount importance. Santorelli and colleagues [21] investigated the association between urinary proteome and the staging and grading of cRCC, identifying a set of proteins that are closely related to staging and grading and are dysregulated. We found that compared to the OAs group, the YAs group had better tumor differentiation and lower stage. This suggests that there may be genomic differences among different stages and grades in RCC, providing the possibility for the development of molecular markers based on genomic differences, which is also one of our future research directions. YAs generally exhibited lower mutation frequencies and mutation load. While there was a higher occurrence of copy number deletions in RCC patients, there were no significant differences in CNA between the two age groups. Accordingly, YAs had fewer mutations in most cancer genes. Further analyses indicated that clinical factors such as gender, histological subtypes, and pathological grading can affect the genetic disparities in the Nanjing cohort, and stage and sarcomatoid differentiation may also be factors affecting genetic disparities in the TCGA cohort. We also observed that OAs generally harbor somatic mutations in significantly mutated genes (SMGs). Although OAs had more somatic mutations in the genes of 10 canonical oncogenic signaling pathways and tumor-related genes (TRGs), there were no statistically significant differences. We also found that genetic disparities can mediate the survival outcomes differences between the two age groups, indirectly highlighting the importance of genetic disparities.

The findings of our study reveal interesting insights into the genetic differences between age groups in two different cohorts, Nanjing and TCGA. These differences could potentially be attributed to racial disparities and the inclusion of germline gene mutations in the Nanjing cohort. However, despite these disparities, our findings align with some previous reports. For instance, we were able to confirm the results reported by Wang et al. [10], regarding the higher mutation rate of FH and the lower mutation rates of PBRM1, SETD2, and VHL in the YAs. This observation was consistent across both the Nanjing and TCGA cohorts. However, it is worth mentioning that previous research on this topic has been contradictory. In a study conducted by Chatsirisupachai et al. [15], no relationship was found between the mutation rates of PBRM1, SETD2, and VHL and age in ccRCC and pRCC. Similarly, Li et al. [14], found a correlation between age and only the mutation rates of PBRM1 and FH in papillary renal cell carcinoma. On the other hand, Shah et al. [16] did not identify any age-related mutation genes in RCC. These discrepancies in the findings may be attributed to differences in the experimental design of these studies.

In addition to genetic differences, other factors such as clinicopathologic features and presentation patterns also contribute to the understanding of age-related characteristics in RCC. Verhoest et al. [22] conducted a multicenter study that explored the relationship between age at diagnosis and clinicopathologic features of RCC. They reported a higher metastatic tumor rate and lower high-grade (G3-4) and ccRCC in patients under 40 years old. However, we did not observe a significant difference in metastasis when analyzing the Nanjing and TCGA data. Furthermore, previous studies have shown that patients commonly present with sarcomatoid differentiation RCC between the ages of 54 and 63 [23,24,25,26,27,28,29]. Interestingly, our data also revealed a lower occurrence of sarcomatoid differentiation in YAs, indicating a potential improvement in early diagnosis among the YAs.

VHL, PBRM1, SETD2, and BAP1 have been identified as the most common and extensively studied genes mutated in RCC in previous studies [30]. Our study found that YAs exhibited fewer mutations in these tumor-suppressor genes. This can be explained by the frequent loss of VHL, which is often accompanied by the loss of neighboring tumor suppressor genes such as BAP1, PBRM1, and SETD2 [31]. Wang et al. [10] reported high mutation rates of epigenetic regulators in YA cancers. As you mentioned, several genes with epigenetic regulatory functions, including AXIN2, ATXN7, FH, and PIGU, were found to have higher mutation frequencies in YAs in our study. Interestingly, we also observed that many of these genes are associated with hereditary diseases, suggesting that hereditary factors may play a significant role in RCC in YAs. For instance, mutations in the FH gene have been linked to hereditary leiomyomatosis and renal cell cancer (HLRCC), with approximately 70% of HLRCC patients diagnosed with RCC before the age of 20 and the youngest diagnosed at 11 years old and the presence of type 2 papillary renal cell carcinoma before the age of 40 is one of its diagnostic criteria [32, 33].

A multicenter study including large international RCC cohort was conducted to examine the influence of age on the prognosis of RCC at various stages. The findings demonstrated that younger age is an independent prognostic factor in women with RCC [34]. However, the other retrospective study indicated that YAs had similar 5-year overall survival and cancer-specific survival rates compared to OAs [35]. As a result, the overall impact of age on RCC prognosis remains inconclusive. In our study, we utilized data from the TCGA and SEER to verify that YAs exhibited better overall survival and cancer-specific survival. Subsequent mediation analyses revealed that the association between age and RCC prognosis was indirectly mediated through TMB. Notably, when we specifically analyzed the mediating effects of differentially mutated genes in the two age groups on the association between age and RCC prognosis, although they all directly mediated the association between age and RCC prognosis, the Proportion mediated were little. This outcome suggests a potential conclusion: the cumulative effect of gene mutations has a stronger impact on the prognosis of RCC compared to the effect of individual gene mutations.

An increasing body of evidence suggests that tumor and immune reprogramming, occurring during targeted or immune therapy, serves as a mechanism for systemic drug resistance in RCC [36, 37]. As a result, several confounding factors, including clinical and molecular characteristics, may impact the tumor mutation profiles, posing a significant challenge when comparing genetic disparities between YAs and OAs. In our model, we controlled for cancer histopathology characteristics and stages because of the differences exhibited between YAs and OAs. Additionally, we examined the influence of gender, considering the substantial differences in mutation density and gender bias in the frequency of mutation in specific genes [38,39,40]. However, further testing is necessary once such data becomes available.

While multivariable regression models offer the advantage of controlling for confounders, univariable models serve a useful and necessary purpose because determining the most relevant or exhaustive list of confounders can be challenging. Additionally, adding a variable to a multivariable model can decrease its statistical power in detecting the effect of YAs. Univariable models, on the other hand, utilize the entire dataset, thereby providing a conservative and reliable approach to identifying genetic disparities between YAs and OAs. From a clinical standpoint, the over representation of a mutation in YAs can be informative for basket trials and molecular diagnoses, regardless of other clinical and molecular parameters.

An area that merits further exploration to deepen our understanding of the disparities between YAs and OAs is germline mutations. In this study, we examined germline gene mutations in the Nanjing cohort. However, due to data availability, germline gene mutations were not included in the TCGA data, which might contribute to the disparities observed in mutation profiles between the two cohorts. Germline mutations might have a disproportionate impact on young cancer patients [41]. Previous studies have also revealed a higher prevalence of germline gene mutations among young patients with RCC [42].

This study still has some limitations. First, we only conducted targeted sequencing on RCC patients in the Nanjing region, which may not represent the entire Chinese population. Second, using targeted sequencing instead of whole exome sequencing may overlook some genomic differences between the YAs and OAs groups. Additionally, the lack of data on germline mutations in the TCGA database makes it impossible for us to compare germline mutation differences between different ethnic groups. Finally, our study has a small sample size, which may introduce bias, and future studies will require larger sample sizes for validation. Nevertheless, our research can still provide guidance for the clinical treatment of RCC and offer direction for future studies.

Conclusion

In summary, based on the analysis of panel sequencing data, we examined the age-related genetic features of RCC. Further studies that investigate the interplay between germline and somatic mutations in YAs with RCC, using larger sample sizes, will enable the development of a comprehensive approach that integrates genetic counseling and multi-modal treatments. This integrated approach aims to mitigate the mortality and morbidity associated with RCC in YAs.

Data availability

The raw data supporting are available in the National Genomics Data Cencer (https://ngdc.cncb.ac.cn) with the accession number：HRA008160 and and is publicly available.

Abbreviations

chRCC:

Chromophobe RCC

CNAs:

Copy-number alterations

FFPE:

Formalin-fixed paraffin-embedded

GDMs:

Genes with different mutations between YAs and OAs

HLRCC:

Hereditary leiomyomatosis and renal cell cancer

HRs:

Hazard ratios, ccRCC, clear cell RCC, nccRCC, non-clear cell RCC

NGS:

Next-generation sequencing

OAs:

Older adults

pRCC:

Papillary RCC

RCC:

Renal cell carcinoma

SMGs:

The significantly mutated genes of RCC

TCGA:

The Cancer Genome Atlas

TMB:

Tumor mutation burden

TRGs:

Telomerase-related genes

YAs:

Young adults

References

  1. Rosiello G, Larcher A, Montorsi F, Capitanio U. Renal cancer: overdiagnosis and overtreatment. World J Urol. 2021;39(8):2821–3.

    Article  PubMed  Google Scholar 

  2. Palumbo C, Pecoraro A, Rosiello G, Luzzago S, Deuker M, Stolzenbach F, et al. Renal cell carcinoma incidence rates and trends in young adults aged 20–39 years. Cancer Epidemiol. 2020;67:101762.

    Article  PubMed  Google Scholar 

  3. Lotan Y, Karam JA, Shariat SF, Gupta A, Roupret M, Bensalah K, Margulis V. Renal-cell carcinoma risk estimates based on participants in the prostate, lung, colorectal, and ovarian cancer screening trial and national lung screening trial. Urol Oncol. 2016;34(4):e1679–16.

    Article  Google Scholar 

  4. Hunt JD, van der Hel OL, McMillan GP, Boffetta P, Brennan P. Renal cell carcinoma in relation to cigarette smoking: meta-analysis of 24 studies. Int J Cancer. 2005;114(1):101–8.

    Article  CAS  PubMed  Google Scholar 

  5. Rossi SH, Klatte T, Usher-Smith J, Stewart GD. Epidemiology and screening for renal cancer. World J Urol. 2018;36(9):1341–53.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Thompson RH, Ordonez MA, Iasonos A, Secin FP, Guillonneau B, Russo P, Touijer K. Renal cell carcinoma in young and old patients–is there a difference? J Urol. 2008;180(4):1262–6. discussion 6.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Brennan CW, Verhaak RG, McKenna A, Campos B, Noushmehr H, Salama SR, et al. The somatic genomic landscape of glioblastoma. Cell. 2013;155(2):462–77.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Gerhauser C, Favero F, Risch T, Simon R, Feuerbach L, Assenov Y, et al. Molecular evolution of early-onset prostate cancer identifies molecular risk markers and clinical trajectories. Cancer Cell. 2018;34(6):996–e10118.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Liao S, Hartmaier RJ, McGuire KP, Puhalla SL, Luthra S, Chandran UR, et al. The molecular landscape of premenopausal breast cancer. Breast Cancer Res. 2015;17:104.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Wang X, Langevin AM, Houghton PJ, Zheng S. Genomic disparities between cancers in adolescent and young adults and in older adults. Nat Commun. 2022;13(1):7223.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Ray S, Jones R, Pritchard-Jones K, Dzhuma K, van den Heuvel-Eibrink M, Tytgat G, et al. Pediatric and young adult renal cell carcinoma. Pediatr Blood Cancer. 2020;67(11):e28675.

    Article  PubMed  Google Scholar 

  12. Craig KM, Poppas DP, Akhavan A. Pediatric renal cell carcinoma. Curr Opin Urol. 2019;29(5):500–4.

    Article  PubMed  Google Scholar 

  13. Young EE, Brown CT, Merguerian PA, Akhavan A. Pediatric and adolescent renal cell carcinoma. Urol Oncol. 2016;34(1):42–9.

    Article  PubMed  Google Scholar 

  14. Li CH, Haider S, Boutros PC. Age influences on the molecular presentation of tumours. Nat Commun. 2022;13(1):208.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Chatsirisupachai K, Lesluyes T, Paraoan L, Van Loo P, de Magalhaes JP. An integrative analysis of the age-associated multi-omic landscape across cancers. Nat Commun. 2021;12(1):2345.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Shah Y, Verma A, Marderstein AR, White J, Bhinder B, Garcia Medina JS, Elemento O. Pan-cancer analysis reveals molecular patterns associated with age. Cell Rep. 2021;37(10):110100.

    Article  CAS  PubMed  Google Scholar 

  17. Lee W, Wang Z, Saffern M, Jun T, Huang KL. Genomic and molecular features distinguish young adult cancer from later-onset cancer. Cell Rep. 2021;37(7):110005.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. VanderWeele TJ. Causal mediation analysis with survival data. Epidemiol (Cambridge Mass). 2011;22(4):582–5.

    Article  Google Scholar 

  19. Ricketts CJ, De Cubas AA, Fan H, Smith CC, Lang M, Reznik E, et al. The cancer genome atlas comprehensive molecular characterization of renal cell carcinoma. Cell Rep. 2018;23(1):313–e265.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Sanchez-Vega F, Mina M, Armenia J, Chatila WK, Luna A, La KC, et al. Oncogenic signaling pathways in the cancer genome atlas. Cell. 2018;173(2):321–e3710.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Santorelli L, Stella M, Chinello C, Capitoli G, Piga I, Smith A et al. Does urinary proteome reflect ccRCC stage grade progression? Diagnostics (Basel). 2021;11(12).

  22. Verhoest G, Veillard D, Guillé F, De La Taille A, Salomon L, Abbou CC, et al. Relationship between age at diagnosis and clinicopathologic features of renal cell carcinoma. Eur Urol. 2007;51(5):1298–304. discussion 304-5.

    Article  PubMed  Google Scholar 

  23. Cangiano T, Liao J, Naitoh J, Dorey F, Figlin R, Belldegrun A. Sarcomatoid renal cell carcinoma: biologic behavior, prognosis, and response to combined surgical resection and immunotherapy. J Clin Oncology: Official J Am Soc Clin Oncol. 1999;17(2):523–8.

    Article  CAS  Google Scholar 

  24. de Peralta-Venturina M, Moch H, Amin M, Tamboli P, Hailemariam S, Mihatsch M, et al. Sarcomatoid differentiation in renal cell carcinoma: a study of 101 cases. Am J Surg Pathol. 2001;25(3):275–84.

    Article  PubMed  Google Scholar 

  25. Gu L, Li H, Wang H, Ma X, Wang L, Chen L, et al. Presence of sarcomatoid differentiation as a prognostic indicator for survival in surgically treated metastatic renal cell carcinoma. J Cancer Res Clin Oncol. 2017;143(3):499–508.

    Article  CAS  PubMed  Google Scholar 

  26. Korenbaum C, Pierard L, Thiéry A, Story F, Lindner V, Lang H, et al. Treatments, outcomes, and validity of prognostic scores in patients with sarcomatoid renal cell carcinoma: a 20-year single-institution experience. Clin Genitourin Cancer. 2018;16(3):e577–86.

    Article  PubMed  Google Scholar 

  27. Klatte T, Rossi SH, Stewart GD. Prognostic factors and prognostic models for renal cell carcinoma: a literature review. World J Urol. 2018;36(12):1943–52.

    Article  PubMed  Google Scholar 

  28. Alevizakos M, Gaitanidis A, Nasioudis D, Msaouel P, Appleman LJ. Sarcomatoid renal cell carcinoma: population-based study of 879 patients. Clin Genitourin Cancer. 2019;17(3):e447–53.

    Article  PubMed  Google Scholar 

  29. Blum KA, Gupta S, Tickoo SK, Chan TA, Russo P, Motzer RJ, et al. Sarcomatoid renal cell carcinoma: biology, natural history and management. Nat Reviews Urol. 2020;17(12):659–78.

    Article  CAS  Google Scholar 

  30. Bui TO, Dao VT, Nguyen VT, Feugeas JP, Pamoukdjian F, Bousquet G. Genomics of clear-cell renal cell carcinoma: a systematic review and meta-analysis. Eur Urol. 2022;81(4):349–61.

    Article  CAS  PubMed  Google Scholar 

  31. Peña-Llopis S, Vega-Rubín-de-Celis S, Liao A, Leng N, Pavía-Jiménez A, Wang S, et al. BAP1 loss defines a new class of renal cell carcinoma. Nat Genet. 2012;44(7):751–9.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Smit DL, Mensenkamp AR, Badeloe S, Breuning MH, Simon ME, van Spaendonck KY, et al. Hereditary leiomyomatosis and renal cell cancer in families referred for fumarate hydratase germline mutation analysis. Clin Genet. 2011;79(1):49–59.

    Article  CAS  PubMed  Google Scholar 

  33. Wong MH, Tan CS, Lee SC, Yong Y, Ooi AS, Ngeow J, Tan MH. Potential genetic anticipation in hereditary leiomyomatosis-renal cell cancer (HLRCC). Fam Cancer. 2014;13(2):281–9.

    Article  CAS  PubMed  Google Scholar 

  34. Rampersaud EN, Klatte T, Bass G, Patard JJ, Bensaleh K, Böhm M, et al. The effect of gender and age on kidney cancer survival: younger age is an independent prognostic factor in women with renal cell carcinoma. Urol Oncol. 2014;32(1):e309–13.

    Article  Google Scholar 

  35. Pal DK, Maurya AK, Jana D. Comparative study of renal cell carcinoma in patients less than 40 years of age and older age patients: a retrospective single-center study. Indian J Cancer. 2018;55(3):297–300.

    Article  PubMed  Google Scholar 

  36. Makhov P, Joshi S, Ghatalia P, Kutikov A, Uzzo RG, Kolenko VM. Resistance to systemic therapies in clear cell renal cell carcinoma: mechanisms and management strategies. Mol Cancer Ther. 2018;17(7):1355–64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Bi K, He MX, Bakouny Z, Kanodia A, Napolitano S, Wu J, et al. Tumor and immune reprogramming during immunotherapy in advanced renal cell carcinoma. Cancer Cell. 2021;39(5):649–e615.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Claus EB, Cannataro VL, Gaffney SG, Townsend JP. Environmental and sex-specific molecular signatures of glioma causation. Neurooncology. 2022;24(1):29–36.

    CAS  Google Scholar 

  39. Gadek M, Sherr EH, Floor SN. The variant landscape and function of DDX3X in cancer and neurodevelopmental disorders. Trends Mol Med. 2023;29(9):726–39.

    Article  CAS  PubMed  Google Scholar 

  40. Li CH, Haider S, Shiah YJ, Thai K, Boutros PC. Sex differences in cancer driver genes and biomarkers. Cancer Res. 2018;78(19):5527–37.

    Article  CAS  PubMed  Google Scholar 

  41. Lalloo F, Varley J, Moran A, Ellis D, O’Dair L, Pharoah P, et al. BRCA1, BRCA2 and TP53 mutations in very early-onset breast cancer with associated risks to relatives. Eur J cancer (Oxford England: 1990). 2006;42(8):1143–50.

    Article  CAS  Google Scholar 

  42. Kong W, Yang T, Wen X, Mu Z, Zhao C, Han S, et al. Germline mutation landscape and associated clinical characteristics in chinese patients with renal cell carcinoma. Front Oncol. 2021;11:737547.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This study was supported by National Natural Science Foundation of China (82172777).

Author information

Author notes

  1. Baochao Zhang, Tianlei Xie and Hao Li contributed equally to this work.

Authors and Affiliations

  1. Department of Urology, Nanjing Drum Tower Hospital, Affiliated Hospital of Nanjing University Medical School, Nanjing, China

    Baochao Zhang, Meng Ding, Changwei Ji & Hongqian Guo

  2. Department of Urology, Nanjing Drum Tower Hospital Clinical College of Nanjing Medical University, Nanjing, China

    Tianlei Xie, Hao Li & Changwei Ji

  3. Department of Urology, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China

    Xiaoming Yi & Song Xue

Authors
  1. Baochao Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Tianlei Xie
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Hao Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Xiaoming Yi
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Meng Ding
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Song Xue
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Changwei Ji
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Hongqian Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

BZ designed the study. BZ, TX, and SX performed the statistical analysis. XY, HL and MD collected samples and clinical information in individual studies. CJ and HG verified the underlying data. BZ wrote the manuscript. All authors reviewed and revised the manuscript and approved the final version, and are fully responsible for all content and editorial decisions.

Corresponding authors

Correspondence to Song Xue, Changwei Ji or Hongqian Guo.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the ethics committee of Nanjing Drum Tower Hospital and Jinling Hospital, Clinical School of Medical College, Nanjing University and was conducted strictly following the Declaration of Helsinki. The need for written informed consent was waived by the ethics committee of Nanjing Drum Tower Hospital and Jinling Hospital, Clinical School of Medical College, Nanjing University due to the retrospective nature of the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, B., Xie, T., Li, H. et al. Targeted gene sequencing reveals disparate genomic mutations between young and older adults in renal cell carcinoma. BMC Cancer 24, 1011 (2024). https://doi.org/10.1186/s12885-024-12785-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12885-024-12785-7

Keywords

BMC Cancer

ISSN: 1471-2407

Contact us