Application of AI to make medicine: Unbelievably fast, unbelievably profitable

The application of AI in drug research is considered by many experts to be a great step forward in the pharmaceutical industry. This measure not only shortens a lot of time, but can also save tens of billions of dollars for pharmaceutical companies around the world.

The pharmaceutical industry can save tens of billions of dollars thanks to AI applications

Usually, to research a drug will take years to find the right drug molecules, then create a compound that is analyzed, screened, tested for safety and effectiveness.

But, with just an experimental drug bought for $4 billion from a Boston startup, Takeda Pharmaceuticals of Japan succeeded in developing a psoriasis-curing compound within six months. through the flexible use of artificial intelligence.

The drug, now known as TAK-279, has undergone two successful phases of human trials. TAK-279 was studied for the treatment of psoriasis and other common conditions such as digestive disorders.

It is expected that next month, the compound selected by AI and algorithms from thousands of potential molecules will move to the final stage of clinical trials. If successful, it would be one of the first drugs to be developed with the direct assistance of AI.

AI brings unbelievable profits

Analysts say that, if successful, it is estimated that the dose will bring Takeda 500 billion yen ($3.7 billion) in annual revenue.

According to a forecast by Morgan Stanley Bank, in the next decade, the use of AI in drug development will not only accelerate the speed of production, but also could create 50 more different therapies equivalent to 50 billion dollars in revenue.

Research firm Deep Pharma Intelligence estimates that investments in AI drug research companies have tripled in the past four years, reaching $24.6 billion by 2022.

Pharmaceutical manufacturing and trading enterprises are also gradually showing interest in this intelligent application of intelligence. For example, in January 2022, Sanofi Pharmaceutical Company agreed to pay Exscientia Plc (AI drug development unit) $100 million in advance for research and development of new drugs.

At the same time, Sanofi asked Exscientia Plc to develop up to 15 drugs with cancer and immunology capabilities, for important transactions urgently needed to research new drugs, Sanofi is willing to pay an additional 5, 2 billion USD.

Currently, global pharmaceutical companies are trying to approach AI by contracting startups with a digital background combined with a lot of data researched by researchers. science. The purpose of these businesses is to be able to cut production costs as well as speed up the time to complete and bring the product to market.

The preparation time for the potion is greatly shortened

For traditional methods, bringing a new drug to market would cost nearly $3 billion, and about 90% of experimental drugs fail. Therefore, it can be seen that applying AI not only helps bring unimaginable profits, but also accelerates the production process.

Using technology to select the right drug molecules would shorten the hundreds of years of experiments that could take years to complete.

Before that, Big Pharma Pharmaceutical Company spent serious time investing in AI and Machine Learning. In 2018, Google DeepMind, Google’s artificial intelligence unit, successfully produced and operated an AI program called AlphaFold that predicts the 3-D shapes that proteins can fold into (the fundamental element that determines the shape of proteins). disease determination).

As a result, one of the most difficult problems in biology has been solved, helping scientists narrow down the molecules selected to create a drug that inhibits diseases. Eric Topol, a founder and director of the Scripps Research Institute in California, said that confirming the structure of a protein now takes only a few seconds using AlphaFold technology.

Another proof that AI helps shorten time is during the ongoing Covid-19 pandemic. At this moment, the pharmaceutical industry is rushing to develop weapons against an unknown virus that is spreading at breakneck speed by the hour.

It was at this stage that Pfizer Inc partnered with BioNTech SE and decided to turn to AI to develop the Comirnaty vaccine. The company also expanded its partnership with XtalPi Inc, an AI drug research company based in Shenzhen (China), to accelerate the chemical formulation of the Covid-19 drug called Paxlovid.

Both drugs and vaccines were approved by the U.S. Food and Drug Administration in less than two years, much faster than the 10 years it takes most conventional drugs to be released. market.

Ai: “Premium” but still have worries

Although AI can be of great help in drug development, scientists still need to perform traditional tasks such as analysis, screening, testing, studying selected molecules.

The psoriasis compound that Takeda has developed is much faster than conventional methods, but still needs more time for clinical trials on humans. And yet, AI still has many other limitations, such as the inability to predict the effectiveness and side effects of individual compounds.

Therefore, although it is being applied flexibly in many industries because of its superior features, it is still a long way for artificial intelligence to completely replace the role of humans. long road.