News briefs

Artificial intelligence could boost success of clinical trials

Artificial intelligence (AI) could boost the success rate of clinical trials, according to a review by researchers from IBM Research‐Australia and MIT published in Trends in Pharmacological Sciences. Big Pharma and other drug developers are grappling with a dilemma: the era of blockbuster drugs is coming to an end. At the same time, adding new drugs to their portfolios is slow and expensive. It takes on average 10–15 years and $1.5–2 billion to bring a new drug to market; approximately half of this time and investment is devoted to clinical trials. Although AI has not yet had a significant impact on clinical trials, AI‐based models are helping trial design, AI‐based techniques are being used in patient recruitment, and AI‐based monitoring systems aim to boost study adherence and reduce dropout rates. The researchers found that AI can boost the success rate of clinical trials by efficiently measuring biomarkers that reflect the effectiveness of the drug being tested; identifying and characterising patient subpopulations best suited for specific drugs. Fewer than one‐third of all phase 2 compounds advance to phase 3 testing, and one in three phase 3 trials fail, not because the drug is ineffective or dangerous, but because the number or types of patients are inadequate. Start‐ups, large corporations, regulatory bodies, and governments are all exploring the use of AI for improving clinical trial design. The authors also identify several areas in which AI may be of greatest benefit for patients: AI‐enabled systems could allow patients more access to and control over their personal data; coaching by AI‐based apps could be undertaken before and during trials; AI could assist continuous monitoring of individual patients’ adherence to protocols; and AI techniques could help guide patients to trials of which they may not have been aware. Finally, AI could support precision medicine; for instance, by improving the efficiency and accuracy with which professionals diagnose, treat and manage neurological diseases.

https://www.cell.com/trends/pharmacological-sciences/fulltext/S0165-6147(19)30130-0

Daily coffee does not affect cancer risk

Researchers at the QIMR Berghofer Medical Research Institute have found that drinking coffee does not change the risk of being diagnosed with or dying of cancer, as reported in a study published in the International Journal of Epidemiology. The large Mendelian randomisation study, which examined data for more than 300 000 people, found that drinking coffee every day neither reduced nor increased a person's risk of developing any cancer. Coffee contains a complex mixture of bioactive ingredients, including caffeine and kahweol, that have shown anti‐tumour effects in animal studies. However, anti‐cancer effects in humans have not been established, with conflicting study findings regarding the influence of coffee on overall cancer risk and on risks for individual cancers. The QIMR Berghofer study analysed cancer data from the UK Biobank cohort for more than 46 000 people who had been diagnosed with the most invasive cancer types, including about 7000 people who died of the cancer. Senior author, Associate Professor Stuart MacGregor said “Our two‐pronged research looked at whether cancer rates differed among people with different levels of self‐reported coffee consumption, and whether the same trend was seen when we replaced self‐reported consumption with genetic predisposition towards coffee consumption. We found there was no real relationship between how many cups of coffee a person had a day and if they developed any particular cancers.” The genetic and preference information from the people with cancer was compared with data for more than 270 000 people never diagnosed with cancer. The study also examined data for some common individual cancers, including breast, ovarian, lung and prostate cancers, and found that drinking coffee did not increase or decrease their incidence.

https://academic.oup.com/ije/advance-article-abstract/doi/10.1093/ije/dyz144/5532583?redirectedFrom=fulltext