Identifying Personalised treatment plan for GBM using Multidimensional Patient Similarity Analytics

Abstract

International Agency for Research on Cancer (IACR) reported an increase in the worldwide cancer rate which is now known to be a major impediment to increasing life expectancy. Glioblastoma multiform, further named as astrocytoma, is a fast-growing truculent type of brain tumour that develops in the cerebral hemispheres, mainly in the frontal and temporal lobes of the brain. According to the National Brain Tumor Society, GBM accounts for 49.1 percent of all primary malignant brain tumors. Despite advances in the available treatment options, there is not much improvement in overall patient survival rate and still ranges from 14.6 to 20.5months. Also, some individuals show adverse drug reactions due to their genetic composition, and the condition is called idiosyncrasy. The proposed work aims to find an effective treatment strategy for GBM patients on the basis of their clinical and genomic factors. The work is presented based on Genomic Data Commons (GDC), cBioportal and Cancer Browser dataset. Here we develop different patient cohorts based on the predictive features using K-means++ algorithm. A test patient acquires the treatment pattern of its most similar neighbour using patient similarity analytics. This is a generalized approach that can be applied to any