Abstract
Rb-pathway disruption is of great clinical interest, as it has been shown to predict outcomes in multiple cancers. We sought to develop a transcriptomic signature for detecting biallelic
loss (RBS) that could be used to assess the clinical implications of
loss on a pan-cancer scale.
We utilized data from the Cancer Cell Line Encyclopedia (
= 995) to develop the first pan-cancer transcriptomic signature for predicting biallelic
loss (RBS). Model accuracy was validated using The Cancer Genome Atlas (TCGA) Pan-Cancer dataset (
= 11,007). RBS was then used to assess the clinical relevance of biallelic
loss in TCGA Pan-Cancer and in an additional metastatic castration-resistant prostate cancer (mCRPC) cohort.
RBS outperformed the leading existing signature for detecting
biallelic loss across all cancer types in TCGA Pan-Cancer (AUC, 0.89 vs. 0.66). High RBS (
biallelic loss) was associated with promoter hypermethylation (
= 0.008) and gene body hypomethylation (
= 0.002), suggesting RBS could detect epigenetic gene silencing. TCGA Pan-Cancer clinical analyses revealed that high RBS was associated with short progression-free (
< 0.00001), overall (
= 0.0004), and disease-specific (
< 0.00001) survival. On multivariable analyses, high RBS was predictive of shorter progression-free survival in TCGA Pan-Cancer (
= 0.03) and of shorter overall survival in mCRPC (
= 0.004) independently of the number of DNA alterations in
.
Our study provides the first validated tool to assess
biallelic loss across cancer types based on gene expression. RBS can be useful for analyzing datasets with or without DNA-sequencing results to investigate the emerging prognostic and treatment implications of Rb-pathway disruption.
.
