Helping anemia patients with their optimal dosage

September 26, 2018

Headshot of Dr. Tamer InancFinding the right dosage for a patient can be difficult. Dr. Tamer Inanc who recently received a grant from the National Science Foundation for $286K, believes that advances in engineering and computational intelligence can help.

Inanc and his team are first turning their attention to patients suffering from anemia, which due to the nature of the disease affords a unique test-bed.

“In the short term, we will apply the proposed research to anemia patients,” said Inanc. “In the long run, we hope to be able to apply the proposed methodology to other diseases which require careful monitoring of dosing of drug(s).”

Working with a former student, Inanc developed a new optimal control algorithm. The purpose of the algorithm is to quantify the best path based on the available criteria. For example, this method is employed if you want determine the best route from one point to another or the least energy intensive.

“Optimal control takes into account this objective function,” said Inanc. “We are trying to maximize the benefits for the patient while trying to minimize the amount of drug(s) through a personalized, patient specific optimal drug dose regimen.”

The National Kidney Foundation offers recommendations on how to maintain appropriate hemoglobin levels for anemics, but no guidelines on how to balance that with the strategies in dialysis facilities who have determined their own Anemia Management Protocols. The price associated with outpatient erythropoiesis stimulating agents has steadily increased since 1994, according to the United States Renal Data System.

That value is not only the literal pricing of the medication, but economy of time and effectiveness as well. To satisfy these interests, Inanc will have to walk a narrow path that balances each.

“The cost is going to be trying to keep the patient's hemoglobin levels at the desired output levels,” said Inanc. “At the same time, when you do that, you cannot just maximize the amount of the drug. They have a very narrow therapeutic range, and side effects, so you’re trying to avoid those as well.”

Additionally, Inanc and his team help to offset the associated morbidity and mortality costs connected with poor anemia management stemming from optimal dosing. Physicians often rely on trial-and-error to establish appropriate dosing, which can lead to sub-therapeutic results.

To combat this, Inanc is investigating an approach to individualized drug dosing using a new modern control method based on radial basis functions and robust patient-specific models.

Unlike classical dosage identification techniques for patients, Inanc’s research uses a dynamic modeling system that employs noisy patient-specific clinic data.

“Robust system identification takes into account system uncertainties, unmodeled dynamics and model complexity, i.e., there is no assumption on the model order, uncertainties and noise affecting data,” said Inanc. “Based on these predictive individualized models, we propose development of a patient-specific dosing approach using a novel radial basis function-based controller to generate optimal dosing sequence and expected response adapting to temporal changes in patients’ dose-response characteristics.”

Inanc is aided by Drs. Michael Brier, Adam Gaweda and Alfred Jacobs are from School of Medicine, and Dr. Jacek Zurada from the Department of Electrical and Computer Engineering. Inanc and Zurada are collaborating to develop the proposed methodology in coordination with their co-principal (co-PI) investigators from the medical school.

While Inanc’s approach focuses on a narrow path to optimize patient treatment, he is optimistic that these techniques will be applicable to other drug regimens.

“This is the beginning. This approach, we are going to be applying it to anemia patients, but it has the potential that it can be applicable to other patients and other drugs,” said Inanc. “If the drugs they use have a narrow therapeutic range, side effects, this methodology could be applicable to those. I would love to see this applied to other fields, especially cancer or diabetic patients.”

The grant began on August 15, 2018 and is estimated to end on July 31, 2020.