Nov2023
ESMO 2023: Advances in the fight against cancer through artificial intelligence.
The collaboration between Ariana Pharma, Omicure, and the Léon Bérard Center has led to a breakthrough in identifying survival biomarkers across various cancer types using eXplainable Artificial Intelligence.
In a study, presented at ESMO 2023 [Identification of biomarkers of survival across multiple cancer types using eXplainable Artificial Intelligence], KEM®, a proprietary Ariana Pharma technology, was used to systematically explore association rules in the PROFILER cohort.
The list of candidate biomarkers included drug scores generated by OncoKEM, an AI-transcriptional-based therapeutic recommendation-tool that computes scores for up to 205 drugs based on the drug’s transcriptional signatures and the tumor transcriptional profile. The goal was to demonstrate the biological relevance of OncoKEM® by confronting its results with the findings obtained through a standard pathway analysis.
The analysis identified four pathway dysregulations associated with overall survival, three of which were related to tubulin. Consistently, four OncoKEM® scores were associated with survival, all of which corresponded to microtubule inhibitor drugs: ixabepilone, cabazitaxel, vinflunine, and brentuximab vedotin. The findings demonstrated the biological relevance of OncoKEM® for microtubule inhibitor drugs and paved the way for its use as a prognostic marker for refractory cancers.
Non-small cell lung cancer: Artificial intelligence enables the identification of survival signatures complementary to an Immunologically active gene expression signature involving previous therapies
In a subsequent study presented at ESMO 2023 [Non-small cell lung cancer: Artificial Intelligence enables the preliminary identification of complementary survival signatures that involve previous therapies] KEM Biomarker (IP software from Ariana) was used to increase the performance of the models previously found. The study identified two 4-gene signatures that predict survival with a significant interaction between the number of treatment lines and gene expressions. Patients carrying these signatures and having undergone at least two treatment lines showed improved survival in both models
