By DDW Editor Reece Armstrong.
If there’s one thing I would take from the Summer 2023 issue of Drug Discovery World, it would be that the application of technologies and processes throughout the pharmaceutical industry are changing how scientists pursue the development of therapies.
Artificial intelligence (AI), gene editing, synthetic biology and more, are beginning to prove their usefulness in drug design and development, taking therapies away from the one-size-fits-all approach, towards more definable targets that operate more effectively. Think how cell and gene therapies or monoclonal antibodies have become targeted medicines for cancer and you’ll get an idea of the shifting pharmaceutical market.
These targeted approaches are being enabled by technologies that allow scientists to better understand human biology than ever before, or to make development decisions that help cut down both costs and the length of time it takes to get to market.
There are myriad ways that this is being enabled. AI has long been thought of as a tool that can help drive us towards a kind of personalised medicine ‘revolution’. Its convergence with genomics makes sense and we have now multiple examples of AI being used to inform prescribing, diagnostics and drug discovery/ development.
For instance, machine learning has been used to accurately predict how cancer patients will respond to a variety of chemotherapies. A team of scientists in Atlanta, Georgia1 used a machine learning algorithm to predict how 175 cancer patients would respond to a range of chemotherapies. The study predicted patient responses with more than 80% accuracy and the authors noted that its success held promise for further clinical approaches. In particular, for patients failing first-line therapies, the approach could be useful in identifying second-line treatments that work in patients.
The team utilised genomic data from a database consisting of over 11,000 patients and which represented 33 types of human cancers. This was done to assess the accuracy of the team’s machine-learning algorithms. The algorithms work by inputting the gene expression profiles of cancers cells to predict how each individual cancer will respond to chemotherapeutic drugs.
The work is a great example of how the availability of data can first inform a project’s success but also how technology can now be used to improve patient care and make it more targeted. Knowing how a patient’s tumour will respond to a certain therapy can drive improvements in cancer care. The challenge now is doing this not on a reactive basis with small cohorts of patients, but conducting much more widespread studies that gather enough data so that clinicians know what therapies they need to use on patients’ once their cancer has been identified.
This kind of more personal approach to healthcare now extends to drug discovery and development. The rise of synthetic biology has helped create some of the most exciting drugs in recent years, with CAR-T therapies acting as potential cures in some cases for blood cancers.
More so, the success of mRNA vaccines throughout the Covid-19 pandemic has seen pharma take those approaches and apply them to other disease areas, including oncology and HIV. Some of these will no doubt see more success than others. For instance, this year saw Janssen discontinue a Phase III HIV clinical trial after the vaccine was shown to be non- effective in preventing HIV transmission. Disappointing, but work is continuing.
Last year, the National Institutes of Health (NIH) launched a Phase I clinical trial assessing three experimental HIV vaccines based on an mRNA platform. Hopefully the success we saw from mRNA-based Covid-19 vaccines will be translated into HIV and other disease areas soon. Either way, targeted approaches represent promising methods that scientists from across the world are using to enhance healthcare.
The new issue of DDW is available to view now.
Reference:
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6219522/