• About UANews
• Sign up for News
• Resources for Media
• Calendar
• Contact Us

Engineering helps bring order to testing thousands of drug combinations.

By Ed Stiles, College of Engineering
April 28, 2008

Combinations of drugs – what doctors call "drug cocktails" – have proven highly effective in treating some diseases, such as AIDS.

But finding just the right drug and dosage combination that provides optimum results can be a daunting task, given the huge number of possible combinations. In some ways it’s like attempting to crack a safe by trying every combination of numbers on the dial.

Pak Kin Wong of The University of Arizona is using engineering optimization strategies to break these codes and bring order to the chaos inherent in sorting through tens of thousands of possibilities.

Wong, who began this work while a graduate student at the University of California, Los Angeles, has developed a testing technology that bypasses the thousands of hours of work required to test individual biological pathways and interactions. Instead, he has designed a closed-loop control scheme and search algorithm that responds to the biological system’s reactions.

In other words, his method is based on a guided search of drug combinations that produces the best and fastest cure in an experimental environment.

The findings from Wong’s UCLA research appeared in the April 1 issue of the journal Proceedings of the National Academy of Sciences.

An Iterative Process

So far, the research has been done with cell-based models. "We stimulate the cells with some drug cocktails and determine how the cells respond," said Wong, an assistant professor in the UA aerospace and mechanical engineering department. "By looking at the response, we use this information to guide the next drug combination to test iteratively.”

In one test case, there were more than 100,000 possible combinations of drug types and concentrations that might inhibit the activity of a particular virus. "We showed that we needed to conduct only 30 to 50 tests to identify the most potent combination from the possibilities," Wong explained. "Instead of doing 100,000 tests, we were able to determine potent combinations by performing only tens of tests and the dosage required can be much smaller than using individual drugs."

The nano-scale biosensors that Wong developed are key to this testing process, and where mechanical engineering interfaces with medicine. "Micro-scale and nano-scale technologies have many important applications, including biomedical ones, which is my specialty research interest," Wong said.

It’s Like Climbing a Mountain

"The way I usually explain our optimization process is to compare it to climbing a mountain," he explained. "When climbers approach a mountain, they don't try to climb all the possible routes to the top. Instead, they select what looks like a promising route, then they climb. When they get to a certain point, they search for the next most likely route from there. Instead of trying all the routes, they select one and narrow down the path to the peak as they climb."

Since coming to the UA, Wong has been refining this testing methodology and working with Donna Zhang, an assistant professor in the department of pharmacology and toxicology, to search for a drug cocktail that will prevent cancer.

"My part of the project is in the screening process," Wong said. "We are developing biosensors to detect cellular activities to help us identify some of the best combinations of drugs with low concentrations for skin cancer chemoprevention."

Wong is developing biosensors that can monitor the response of cells to the various therapies. The sensors are designed to detect specific genes within the cells that have cancer-inhibiting properties.

"They're optical sensors, what we call a molecular probe," he said. These nano-probe molecules are mixed with the cells under study. If the resulting mixture fluoresces, the researchers know the cancer-inhibiting genes are present. The intensity of the fluorescence indicates the chemopreventive activities quantitatively.

Simple Design, Fast Results

"To facilitate high-throughput screening, we are using a microplate reader that can measure many samples semiautomatically,” Wong said. “Because of the simplicity of our design, we can do many tests simultaneously, and our sensor is at least 10 times faster than a conventional sensor."

Currently, Wong and his students also are working on a multiplexed micro-fluidic system that will do this testing automatically, which will further increase the speed and accuracy of the tests.

Wong's work on cancer-preventing drug therapies is being sponsored by the American Cancer Society through the UA Cancer Center.