Machine learning and artificial intelligence have started to show applications around all industries. Healthcare is no exception. In fact, considering the immense amount of data being produced every day in this field, healthcare is a prime area to apply machine learning. Researchers at the University of Washington utilized a new approach to determine which markers to target in leukemia patients.
While there are over a thousand cancer treatments in clinical development in the US alone, choosing the proper treatment remains a challenge. Cancers that appear pathologically similar may react very differently to the same treatment. Hence, methods to more accurately pair patients to treatments are needed. The UW researchers' novel computational technique identifies gene expression markers to reliably target and learns from data on how much certain drivers contribute to the progress of the cancer. This provides a new way to prioritize gene-treatment associations by determining a probability the a certain gene can be reliably used as a marker for various treatments.
This study implemented the new identification technique on a sample of 30 patients and found an improvement in the chance that the gene-treatment association are validated. While the implications of this study are promising, because of the small sample size, additional research will certain need to be conducted to yield more robust data. It is, however, still just as exciting to see how areas like machine learning, predictive analytics, and computational statistics can be applied in the field of medicine.
We demonstrate a promising approach to identify robust molecular markers for targeted treatment of acute myeloid leukemia (AML) by introducing: data from 30 AML patients including genome-wide gene expression profiles and in vitro sensitivity to 160 chemotherapy drugs, a computational method to identify reliable gene expression markers for drug sensitivity by incorporating multi-omic prior information relevant to each gene’s potential to drive cancer. We show that our method outperforms several state-of-the-art approaches in identifying molecular markers replicated in validation data and predicting drug sensitivity accurately. Finally, we identify SMARCA4 as a marker and driver of sensitivity to topoisomerase II inhibitors, mitoxantrone, and etoposide, in AML by showing that cell lines transduced to have high SMARCA4 expression reveal dramatically increased sensitivity to these agents.