News
July 11, 2024. The DeepRx team published a new framework of genome-informed precision oncology in Nature Machine Intelligence. This framework overcomes the limitation of the one-mutation-to-one-drug approach of the current genome-informed precision oncology. The DeepRx framework transforms somatic genome alteration information into a representation of the cellular state of cancer cells. It utilizes such information to predict drug sensitivities of the cancer cells. This framework enables the prediction of cancer cell responses to both molecularly targeted and chemotherapy drugs, significantly expanding the scope of genome-informed precision oncology. Read the paper for more details.
June 18, 2024. Drs. Xinghua Lu and Lujia Chen's team is awarded a $1.5 million grant by the National Library of Medicine. The grant supports further improvement and validate the COLOXIS model.
April 16, 2024. DeepRx team is awarded an NSF I-Corp grant supporting the market analyses and business development.
Feb 6, 2024. Our article on developing and validating DeepRx COLOXIS signature for predicting oxaliplatin benefit is published by the Journal of Clinical Oncology: https://ascopubs.org/doi/10.1200/JCO.23.01080. The NRG Oncology, a national clinical trial organization for managing large clinical trials, also published a press release highlighting this impactful work. As the first predictive AI model for oxaliplatin benefit, the COLOXIS signature will significantly improve the accuracy of colon cancer (CC) adjuvant therapies, as shown in the diagram. It will double the response rate to oxaliplatin in the COLOXIS-positive group patients and, on the other hand, save half of patients from unnecessary oxaliplatin-induced neural toxicity. This will impact close to a million patients worldwide.
May, 2022. The abstract of our study of using DeepRx COLOXIS system to guide regimen selection for adjuvant therapy of colon cancer is published in ASCO 2022. The significance of this work is two-fold: First, if applied in practice, the system potentially can improve clinical outcomes of over a million colorectal cancer patients worldwide. Second, this study provides the first evidence that AI methods can be used to predict efficacy of chemotherapy and biological drugs in a real clinical setting. Since chemo drugs remain the backbone of contemporary oncology, this study opens the new direction of developing more AI models to guide application of more chemotherapy agents to treat different cancers, which will benefit millions of patients.
