April 26, 2021

What is natural language processing?

Before a new medication or treatment arrives on the market, it goes through a lengthy and extensive testing phase known as the clinical trial process. These steps determine whether or not a drug is safe to use and effective in tackling the problem it’s addressing – and they also provide new information concerning any potential side effects that might occur.

2021-04-26 Natural language processing

It’s not cheap to conduct these trials, with the average cost standing at approximately $2 billion per drug. This is because clinical trials involve huge numbers of volunteers – sometimes hundreds of thousands – and take place over the course of many years. Obviously, this involves a Herculean effort on the part of the biotech and pharmaceutical companies, and every precaution needs to be taken to ensure the trial isn’t too risky for the volunteers.

Despite such a huge investment in terms of money, time and effort, only around one in ten clinical trials is actually successful. This means that the one drug that receives go-to-market approval needs to recoup the costs of the nine failed trials – meaning that it will need to earn around $15-20 billion in order to allow the pharmaceutical company to break even. As a result, the actual cost of medications goes up, affecting real-world patients.

So how do we address this problem in a way that’s sustainable, ethical and reliable? One option gaining increasing traction across the field is the use of data analytics to increase clinical trial success – but how does this work?

By using artificial intelligence, it is possible to predict whether or not a trial is likely to be successful based on a number of variables – and the pharmaceutical company can decide whether or not this particular treatment should be pursued, or if they should explore alternatives before engaging in the clinical trial process. As a result, fewer trials are conducted, with only the ones that are likely to be successful actually carried out – saving huge amounts of time and effort.

Using outcome-predicting technology which accurately determines which drugs might be successful, pharmaceutical companies can save money in the long run – reducing the necessary recoupment for the successful treatment and ultimately lowering the cost for the patient. It’s in nobody’s best interests to have unsustainably expensive medications on the market: patients will simply look elsewhere for other solutions, and pharmaceutical companies will end up in the red, with their treatment not helping anyone.