The treatments of many illnesses require not one, but a combination of drugs. Usually, you’ll see patients taking blood pressure medication with anti-depressants or Advil with multivitamins. And of course, most chemotherapy treatments will go hand in hand with anti-nausea pills. While combining drugs is nothing new, researchers are still not sure exactly how these things work together, do they work better together and if they do, does this improvement change for different patients.
To answer at least some of these questions, a team of scientists has published a paper that talks about how to best model the variables of drug combinations to determine the ideal dose and efficacy. In it, the team states they hope that their work will guide “the determination of algorithms and their parameters when we optimize drug combinations for patients.”
The team found that the common method of modeling the relationship between the efficacy and dose of the drugs based on experimental observations does work, but only up to a point – as the number of variables increases, the interaction prediction becomes significantly less accurate. In other words, with more variables, the algorithm is less able to separate out random errors.
After experimentation with eight chemotherapeutic drugs and a total of 59 combinations tested, the researchers found that models with only observations for two variables work best – this is especially true in combinations with more than six drugs.
“These algorithms can be used to optimize almost all kinds of drug combinations, even combinations of biological molecules like growth factors,” said one of the researchers, Prof. Boqian Wang.
Source:
Phys.org (http://phys.org/news/2017-01-problem-potential-solution-combining-drugs.html)
The Wonderful World of Algorithms 