Florian Büttner

Survival Prediction of Clear Cell Renal Cell Carcinoma Based on Gene Expression Similarity to the Proximal Tubule of the Nephron

UNLABELLED There is evidence that molecular features support subclassification of tumours, thereby improving prediction of patient outcome. Currently, two gene expression signatures (ccA/ccB and ClearCode34) have been established to classify clear cell renal cell carcinoma (ccRCC). Because RCC arises from nephron cell types, we aimed to explore its heterogeneity on a molecular level by comparing gene expression between tumour tissue and nephron regions. Based on genes that differ in expression between nephron regions, expression data of 479 ccRCCs and 212 papillary and 66 chromophobe RCCs from The Cancer Genome Atlas were correlated to those of nephron cell types. Cancer-specific survival (CSS) of ccRCC patients was significantly associated with gene expression similarity to the proximal tubules. Subsequently, a ccRCC risk score (S3-score) was established. Survival analyses indicated that the S3-score was significantly associated with CSS considering all cases of ccRCC, as well as metastatic and nonmetastatic ccRCC. Results could be validated in an independent cohort. The S3-score significantly improved the predictive ability of the ccA/ccB and ClearCode34 signatures, and the clinicopathologic-based stage, size, grade, and necrosis score (p [chi-square] = 1.56E-04). Intratumour heterogeneity of the S3-score was observed in 6 of 10 ccRCCs. In summary, the nephron-based S3-score enables prognostic risk stratification for ccRCC. Further studies are needed to evaluate its clinical utility. PATIENT SUMMARY We developed a novel risk score for clear cell renal cell carcinoma to identify patients at risk of worse outcome that may improve patient care in the future.

Methylomes of renal cell lines and tumors or metastases differ significantly with impact on pharmacogenes

Current therapies for metastatic clear cell renal cell carcinoma (ccRCC) show limited efficacy. Drug efficacy, typically investigated in preclinical cell line models during drug development, is influenced by pharmacogenes involved in targeting and disposition of drugs. Here we show through genome-wide DNA methylation profiling, that methylation patterns are concordant between primary ccRCC and macro-metastases irrespective of metastatic sites (rs ≥ 0.92). However, 195,038 (41%) of all investigated CpG sites, including sites within pharmacogenes, were differentially methylated (adjusted P < 0.05) in five established RCC cell lines compared to primary tumors, resulting in altered transcriptional expression. Exemplarily, gene-specific analyses of DNA methylation, mRNA and protein expression demonstrate lack of expression of the clinically important drug transporter OCT2 (encoded by SLC22A2) in cell lines due to hypermethylation compared to tumors or metastases. Our findings provide evidence that RCC cell lines are of limited benefit for prediction of drug effects due to epigenetic alterations. Similar epigenetic landscape of ccRCC-metastases and tumors opens new avenue for future therapeutic strategies.

Comprehensive Metabolomic and Lipidomic Profiling of Human Kidney Tissue: A Platform Comparison

Metabolite profiling of tissue samples is a promising approach for the characterization of cancer pathways and tumor classification based on metabolic features. Here, we present an analytical method for nontargeted metabolomics of kidney tissue. Capitalizing on different chemical properties of metabolites allowed us to extract a broad range of molecules covering small polar molecules and less polar lipid classes that were analyzed by LC-QTOF-MS after HILIC and RP chromatographic separation, respectively. More than 1000 features could be reproducibly extracted and analyzed (CV < 30%) in porcine and human kidney tissue, which were used as surrogate matrices for method development. To further assess assay performance, cross-validation of the nontargeted metabolomics platform to a targeted metabolomics approach was carried out. Strikingly, from 102 metabolites that could be detected on both platforms, the majority (>90%) revealed Spearmans correlation coefficients ≥0.3, indicating that quantitative results from the nontargeted assay are largely comparable to data derived from classical targeted assays. Finally, as proof of concept, the method was applied to human kidney tissue where a clear differentiation between kidney cancer and nontumorous material could be demonstrated on the basis of unsupervised statistical analysis.

Metabolic and Lipidomic Reprogramming in Renal Cell Carcinoma Subtypes Reflects Regions of Tumor Origin

BACKGROUND Renal cell carcinoma (RCC) consists of prognostic distinct subtypes derived from different cells of origin (eg, clear cell RCC [ccRCC], papillary RCC [papRCC], and chromophobe RCC [chRCC]). ccRCC is characterized by lipid accumulation and metabolic alterations, whereas data on metabolic alterations in non-ccRCC are limited. OBJECTIVE We assessed metabolic alterations and the lipid composition of RCC subtypes and ccRCC-derived metastases. Moreover, we elucidated the potential of metabolites/lipids for subtype classification and identification of therapeutic targets. DESIGN, SETTING, AND PARTICIPANTS Metabolomic/lipidomic profiles were quantified in ccRCC (n=58), chRCC (n=19), papRCC (n=14), corresponding nontumor tissues, and metastases (n=9) through a targeted metabolomic approach. Transcriptome profiling was performed in corresponding samples and compared with expression data of The Cancer Genome Atlas cohorts (patients with ccRCC, n=452; patients with papRCC, n=260; and patients with chRCC, n=59). OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS In addition to cluster analyses, metabolomic/transcriptomic data were analyzed to evaluate metabolic differences of ccRCC and chRCC using Welchs t test or paired t test as appropriate. Where indicated, p values were adjusted for multiple testing using Bonferroni or Benjamini-Hochberg correction. RESULTS AND LIMITATIONS Based on their metabolic profiles, RCC subtypes clustered into two groups separating ccRCC and papRCC from chRCC, which mainly reflected the different cells of origin. ccRCC-derived metastases clustered with primary ccRCCs. In addition to differences in certain lipids (lysophosphatidylcholines and sphingomyelins), the coregulation network of lipids differed between ccRCC and chRCC. Consideration of metabolic gene expression indicated, for example, alterations of the polyamine pathway at metabolite and transcript levels. In vitro treatment of RCC cells with the ornithine-decarboxylase inhibitor difluoromethylornithine resulted in reduced cell viability and mitochondrial activity. Further evaluation of clinical utility was limited by the retrospective study design and cohort size. CONCLUSIONS In summary, we provide novel insight into the metabolic profiles of ccRCC and non-ccRCC, thereby confirming the different ontogeny of RCC subtypes. Quantification of differentially regulated metabolites/lipids improves classification of RCC with an impact on the identification of novel therapeutic targets. PATIENT SUMMARY Several subtypes of renal cell carcinoma (RCC) with different metastatic potentials and prognoses exist. In the present study, we provide novel insight into the metabolism of these different subtypes, which improves classification of subtypes and helps identify novel targets for RCC therapy.

Profiles of miRNAs matched to biology in aromatase inhibitor resistant breast cancer

Aromatase inhibitor (AI) resistance during breast cancer treatment is mimicked by MCF-7:5C (5C) and MCF-7:2A (2A) cell lines that grow spontaneously. Survival signaling is reconfigured but cells are vulnerable to estradiol (E2)-inducible apoptosis. These model systems have alterations of stress related pathways including the accumulation of endoplasmic reticulum, oxidative, and inflammatory stress that occur prior to E2-induced apoptosis. We investigated miRNA expression profiles of 5C and 2A to characterize their AI resistance phenotypes. Affymetrix GeneChip miRNA2.0 arrays identified 184 miRNAs differentially expressed in 2A and 5C compared to E2-free wild-type MCF-7:WS8. In 5C, 34 miRNAs of the DLK1-DIO3 locus and miR-31 were overexpressed, whereas miR-222 was low. TCGA data revealed poor and favorable overall survival for low miR-31 and miR-222 levels, respectively (HR=3.0, 95% CI:1.9-4.8; HR=0.3, 95% CI:0.1-0.6). Targets of deregulated miRNAs were identified using CLIP-confirmed TargetScan predictions. KEGG enrichment analyses for 5C- and 2A-specific target gene sets revealed pathways associated with cell proliferation including insulin, mTOR, and ErbB signaling as well as immune response and metabolism. Key genes overrepresented in 5C- and 2A-specific pathway interaction networks including EGFR, IGF1R and PIK3R1 had lower protein levels in 5C compared to 2A and were found to be differentially modulated by respective miRNA sets. Distinct up-regulated miRNAs from the DLK1-DIO3 locus may cause these attenuative effects as they are predicted to interact with corresponding 3′ untranslated regions. These new miRNA profiles become an important regulatory database to explore E2-induced apoptotic mechanisms of clinical relevance for the treatment of resistant breast cancer.

Characterization of the breast cancer resistance protein (BCRP/ABCG2) in clear cell renal cell carcinoma: BCRP/ABCG2 in clear cell renal cell carcinoma

The efflux transporter breast cancer resistance protein BCRP/ABCG2 is well‐known for its contribution to multi‐drug resistance in cancer. Its relevance in cancer biology independent from drug efflux remains largely elusive. Our study aimed at elucidating the biological relevance and regulatory mechanisms of BCRP/ABCG2 in clear cell renal cell carcinoma (ccRCC) and disease progression. Two independent ccRCC‐cohorts [Cohort 1 (KIRC/TCGA): n = 453, Cohort 2: n = 64] were investigated to elucidate BCRP/ABCG2 mRNA and protein expression and their association with survival. The impact of BCRP/ABCG2 on response to sunitinib treatment was investigated in two independent sunitinib‐treated ccRCC‐cohorts based on mRNA levels. Moreover, underlying regulatory mechanisms for interindividual variability of BCRP/ABCG2 expression were systematically assessed. Owing to redundant functional properties, mRNA and protein expression of the multidrug resistance protein MDR1/ABCB1 were additionally evaluated in these cohorts. In independent ccRCC‐cohorts, low BCRP/ABCG2 and MDR1/ABCB1 mRNA and protein expression were associated with severity (e.g., tumor stage) of ccRCC and poor cancer‐specific survival. BCRP/ABCG2 and MDR1/ABCB1 mRNA expression were linked to decreased progression‐free survival after sunitinib treatment. Germline and somatic variants influenced interindividual variability of BCRP/ABCG2 expression only moderately. miR‐212‐3p and miR‐132‐3p were identified to regulate BCRP/ABCG2 posttranscriptionally by interaction with the ABCG2 3′UTR as confirmed through reporter gene assays in RCC cell lines. In summary, BCRP/ABCG2 expression in ccRCC underlies considerable interindividual variability with impact on patient survival and response to sunitinib treatment. While germline or somatic genetic variants and DNA methylation cannot explain aberrant BCRP/ABCG2 expression, miR‐212‐3p and miR‐132‐3p were identified to contribute to posttranscriptional regulation of BCRP/ABCG2.