Developing a new drug requires a lot of investment in time, money and human resources. The procedure involves a clinical trial, as well as a test phase. Of course, the new medicine needs the approval from the authorities as well. By tapping the potential of Artificial Intelligence (AI), this process can be sped up considerably. Some phases can even be skipped. This blog informs you about the three ways in which medical data analytics company Keystonemab employs AI to improve and speed up medicine development.
1. Discover new treatments
90% of all clinical studies end in a failure. Yet, these rejected drugs can turn out to be useful to treat other diseases. This could also be the case with medicine used for other treatments. The challenge: it takes a lot of for scientists to dig through all these data, in order to make useful connections.
By employing AI, these possibilities can be mapped in lightning speed. Keystonemab relies on IBM’s Natural Language Understanding (NLU) functionality, in order to go through millions of medical documents about existing and rejected medicine. This provides them with useful data that can be used to develop new treatments. This AI model is a huge time saver: it would take scientists decades (!) to search through the same amount of data for which the data platform only needs several days.
2. Combine drugs to improve existing medicine
Recently, Pfizer announced it won’t allow single-molecule drugs in clinical studies anymore, from 2024 on. This is because cancer’s resistance is growing to single-molecule treatments. The solution: combinations of drugs (and thus of multiple molecules). Keystonemab employs an AI model in order to identify these combinations both accurately and rapidly. Combing drugs doesn’t only have the potential to amplify the effects of a treatment, but it can also reduce side effects.
Keystonemab’s platform automatically collects data to identify meaningful combinations. Think of information on biomarkers (the part of the body that is influenced by the medicine), proteins, drugs and diseases. The AI model doesn’t only display valuable connections, but also indicates their strength. This helps to prioritise when giving useful recommendations.
3. Speed up the medicine development process
Employing AI doesn’t only help to discover new treatments and medicine combinations, but it can also speed up the development process. At first, this is because the data platform can perform several tasks – such as digging through data – a lot faster than humans. Moreover, this way of working means certain phases of the development procedure can be skipped, in case new treatments are possible by combing already existing medicine. These drugs have often passed the expensive safety studies in an earlier stage, which means both a lot of money and time can be saved.
Would you like to read more about Keystonemab’s progressive approach? Then download the reference case.