Analysis of DNA methylation and copy number variation in renal cell carcinoma.

摘要

Renal cell carcinoma (RCC) is the 10th most commonly diagnosed cancer in the United States and its incidence is increasing. It is difficult to detect early and is relatively non-responsive to traditional radiation and chemotherapies. Previous work demonstrates the value of measuring copy number variation (CNV) and DNA methylation changes in RCC, but diagnostic biomarkers and additional treatment options in the clinic are still needed. This study examined genome-wide DNA methylation and CNV in RCC tumor and benign adjacent kidney tissues from 101 clinically-annotated patients. Analyses revealed widespread DNA methylation and copy number differences between tumor and benign tissues, particularly in immune, G-protein coupled receptor, and metabolism-related genes. Also reported is a panel of DNA methylation biomarkers that successfully distinguish tumor from benign tissue and validation of these markers in The Cancer Genome Atlas RCC datasets. Additionally, variable DNA methylation and copy number profiles effectively classified the patients into distinct risk of recurrence groups for potential assessment of alternative treatment therapies. Overall, increased global DNA methylation is strongly correlated with increased incidence of recurrence, and there is a non-monotonic relationship between recurrence and global levels of copy number events. RCC is composed of several subtypes defined by morphological differences, and more recently, by genetic variation. This study also examined DNA methylation and CNV differences in five subtypes of RCC, and found widespread differences between them. These alterations could be used in addition to pathological analyses for the diagnosis of subtype specific RCC, or to inform additional chemotherapeutic targets. Finally, the mismatch repair (MMR) pathway, responsible for the detection and repair of nucleotide mismatches during DNA replication, had widespread DNA methylation and copy number variation in the clinically-annotated patients. To investigate functional consequences, microsatellites (which are prone to instability when MMR genes are not functioning properly) were examined. Instability was detected in some patients, but did not correlate with DNA methylation or CNV in MMR genes. By examining DNA methylation and CNV in clinically-annotated patients, this study offers powerful insights into the etiology and recurrence of RCC, as well as provides clinically applicable biomarkers for RCC diagnosis.