Scientists from Department of Translational Medicine, Center for Biomedical Big Data Research, Medical College of Xiamen University discovered smoking-related two subsets of KRAS-mutant lung adenocarcinoma and potential drug candidates for each subset.
KRAS mutations occur in ~30% of lung adenocarcinomas. Oncogenic KRAS is known to be “undruggable”; therefore, major efforts to combat KRAS-driven cancer focus on either inhibiting KRAS downstream targets or screening for molecules that exhibit synthetic lethal interactions with oncogenic KRAS. However, rare common hits have been identified among these screens. Fruitful results obtained from synthetically lethal screening suffered from poor reproducibility. This lack of reproducibility is likely due to the heterogeneous background of both cell lines and patients harboring KRAS mutations. Therefore, it is critical to identify KRAS subtypes, characterize their features and further explore these features to predict of potential effective drugs.
The study was published online, Sep 27th 2018, in EBioMedicine, https://www.ebiomedicine.com/article/S2352-3964(18)30392-X/fulltext.
KRAS-mutant lung cancer is associated with smoking activity and smoking can cause genomic and epigenomic changes. However, it is not clear that whether the smoking-related genomic and epigenomic changes contribute to the heterogeneity of KRAS-mutant lung adenocarcinomas.
For this new study, researchers took advantage of the aforementioned databases and identified two major subtypes of KRAS-mutant lung adenocarcinoma in patients by integrating multiplatform datatypes. They found that the heterogeneity of KRAS-mutant lung adenocarcinomas was associated with smoking and could be interpreted at many levels, from genomic and epigenomic to transcriptomic and immunogenic levels.
“In this study, we identified five important features of the heterogeneity of KRAS-mutant tumors and cell lines, including smoking-induced two processes. Subtype 1 had an enhanced smoking-related mutational signature, while subtype 2 had increased smoking-related methylation signature although the reported smoking history were not different between the subtypes.” said lead author Ying Zhou, a research scientist at Department of Translational Medicine, Center for Biomedical Big Data Research, Medical College of Xiamen University, China.
“After we obtained two subsets of patients and corresponding cell lines, we reanalyzed the publicly available drug sensitivity data including GDSC, CTRP and CCLE and found several promising drugs for each subtype. Interestingly, drug sensitivity was significantly associated with one or more oncogenic features of the KRAS subtypes.” said co-lead author Qiyuan Li，professor at Department of Translational Medicine, Center for Biomedical Big Data Research, Medical College of Xiamen University.
“We further validated the patient subsets on lung cancer cell lines and found consistent features in cell line subtypes. We are excited to find these useful surrogates of patient for in silico drug screening.” said first author Ke Liu, a graduate student at Department of Translational Medicine, Center for Biomedical Big Data Research, Medical College of Xiamen University, China.”
Support for this study came from the Young Scientists Fund of the National Natural Science Foundation of China (81802823), the Natural Science Foundation of Fujian Province, China (2018J01054) and the Education and Research Foundation for Young Scholars of Education Department of Fujian Province, China (JAT170004).
Link to the paper：https://www.ebiomedicine.com/article/S2352-3964(18)30392-X/fulltext