eTumorType, An Algorithm of DiscriminatingCancer Types for Circulating Tumor Cells orCell-free DNAs in Blood

With the technology development on detecting circulating tumor cells(CTCs) and cellfree DNAs(cf DNAs) in blood, serum, and plasma, non-invasive diagnosis of cancer becomes promising. A few studies reported good correlations between signals from tumor tissues and CTCs or cf DNAs, making it possible to detect cancers using CTCs and cf DNAs. However, the detection cannot tell which cancer types the person has. To meet these challenges, we developed an algorithm,e Tumor Type, to identify cancer types based on copy number variations(CNVs) of the cancer founding clone. e Tumor Type integrates cancer hallmark concepts and a few computational techniques such as stochastic gradient boosting, voting, centroid, and leading patterns. e Tumor Type has been trained and validated on a large dataset including 18 common cancer types and 5327 tumor samples. e Tumor Type produced high accuracies(0.86–0.96) and high recall rates(0.79–0.92) for predicting colon, brain, prostate, and kidney cancers. In addition, relatively high accuracies(0.78–0.92)and recall rates(0.58–0.95) have also been achieved for predicting ovarian, breast luminal, lung, endometrial, stomach, head and neck, leukemia, and skin cancers. These results suggest that e Tumor Type could be used for non-invasive diagnosis to determine cancer types based on CNVs of CTCs and cf DNAs.

Network Analysis Reveals A Signaling RegulatoryLoop in PIK3CA-mutated Breast Cancer Predicting Survival Outcome

Mutated genes are rarely common even in the same pathological type between cancer patients and as such, it has been very challenging to interpret genome sequencing data and difficult to predict clinical outcomes. PIK3 CA is one of a few genes whose mutations are relatively popular in tumors. For example, more than 46.6% of luminal-A breast cancer samples have PIK3 CA mutated, whereas only 35.5% of all breast cancer samples contain PIK3 CA mutations. To understand the function of PIK3 CA mutations in luminal A breast cancer, we applied our recentlyproposed Cancer Hallmark Network Framework to investigate the network motifs in the PIK3CA-mutated luminal A tumors. We found that more than 70% of the PIK3CA-mutated luminal A tumors contain a positive regulatory loop where a master regulator(PDGF-D), a second regulator(FLT1) and an output node(SHC1) work together. Importantly, we found the luminal A breast cancer patients harboring the PIK3 CA mutation and this positive regulatory loop in their tumors have significantly longer survival than those harboring PIK3 CA mutation only in their tumors. These findings suggest that the underlying molecular mechanism of PIK3 CA mutations in luminal A patients can participate in a positive regulatory loop, and furthermore the positive regulatory loop(PDGF-D/FLT1/SHC1) has a predictive power for the survival of the PIK3 CAmutated luminal A patients.