'Turbocharged artificial intelligence' could personalize combination therapy in pediatric leukemia

December 13, 2016 by Brianna Aldrich
Charts showing how the technology identifies the drug ratios required to achieve optimal treatment outcomes for children with leukemia. Credit: Dong-Keun Lee/Dean Ho Lab

A team of UCLA bioengineers has demonstrated that its technology may go a long way toward overcoming the challenges of treatment for acute lymphoblastic leukemia, among the most common types of cancer in children, and has the potential to help doctors personalize drug doses.

The five-year survival rate for individuals with pediatric is about 85 percent, however those who experience a recurrence generally have a poor prognosis and a bone marrow transplant is their only option for a permanent cure.

Conventional treatment for this leukemia includes a combination of drugs, which come with short- and long-term side effects. Two of these drugs, 6-mercaptopurine and methotrexate, can cause liver disease and other life-threatening infections. During the maintenance phase of treatment, which aims to keep individuals in remission, dosing for these two drugs is frequently adjusted through a system of trial and error, which is not always accurate.

The UCLA team from the Schools of Dentistry, Medicine and Engineering showed that its digital health , called Phenotypic Personalized Medicine, or PPM, can surmount the treatment challenges for this leukemia. The platform is based on the team's stunning discovery that a patient's physical response to drug treatment, such as tumor size or bacterial and viral levels in the blood, could be visually represented in the shape of a parabola, or U-shaped line. The graphs plot the drug dose along the horizontal axis and the patient's response to treatment on the vertical axis.

The technology has the ability to accurately identify a person's optimal drug and dose combinations throughout an entire course of treatment. In addition, the technology platform does not require any complex and expensive analysis of a patient's genetic information or the biological basis of the disease, greatly accelerating the ability to optimize and personalize care.

The team's findings appear online in the peer-reviewed journal SLAS Technology, which features innovations in technology for drug development and diagnostics.

"Phenotypic Personalized Medicine is like turbocharged artificial intelligence. It personalizes combination therapy to optimize efficacy and safety," said Dean Ho, co-corresponding author of the study and professor of oral biology and medicine. "The ability for our technology to continuously pinpoint the proper dosages of multiple drugs from such a large pool of possible combinations overcomes a challenge that is substantially more difficult than finding a needle in a haystack," added Ho, who is also the co-director of the Weintraub Center for Reconstructive Biotechnology at the UCLA School of Dentistry.

In this study, patient records were obtained on the dosing of 6-mercaptopurine and methotrexate, as well as on the corresponding absolute neutrophil count, or levels of a subset of white blood cells called neutrophils that are vital for staving off potentially life-threatening infections. These records showed multiple instances where conventional chemotherapy doses caused deviations from acceptable neutrophil levels. Using the personalized medicine technology, individualized three-dimensional maps were generated to determine the optimal 6-mercaptopurine/methotrexate drug ratios.

The team members found that their technology-suggested drug dosages were as much as 40 percent lower compared to clinical chemotherapy dosages, while still maintaining target neutrophil levels. The parabolas showed that markedly different dosages of each drug were required to maintain normal white blood cell counts for each patient. Their results demonstrated a clear need to personalize acute lymphoblastic leukemia treatment, and will serve as a foundation for a pending clinical trial to optimize multi-drug chemotherapy.

"PPM has the ability to personalize combination therapy for a wide spectrum of diseases, making it a broadly applicable technology," said Chih-Ming Ho, distinguished research professor of mechanical and aerospace engineering and co-corresponding author of the study. "The fact that we don't need any information pertaining to a disease's biological process in order to optimize and personalize treatment is a revolutionary advance. We're at the interface of digital health and cancer treatment."

Dr. Vivian Chang, co-first author of the study and assistant professor of pediatrics and hematology and oncology, said, "Optimizing for [pediatric leukemia] on a patient-specific level would be a game-changer for the way that this cancer, as well as many other cancers, is addressed. Reducing side effects while maintaining or even enhancing efficacy could also improve the long-term outcomes of our patients." Chang is also co-director of the Pediatric Cancer Predisposition Clinic at UCLA.

The research team is planning to recruit patients for a prospective trial within the next year. The technology is approved for additional infectious disease and oncology studies.

Explore further: New platform optimizes drug doses to prevent organ rejection for transplant patients

Related Stories

New platform optimizes drug doses to prevent organ rejection for transplant patients

April 7, 2016
For decades, doctors and scientists have predicted that personalized medicine—tailoring drug doses and combinations to people's specific diseases and body chemistry—would be the future of health care.

Leukemia drug combo is encouraging in early Phase I clinical trial

December 5, 2016
Researchers from Columbia University Medical Center and NewYork-Presbyterian reported that 8 out of 12 patients with relapsed and/or chemotherapy refractory acute myeloid leukemia (AML) or other blood cancers responded to ...

Study determines efficacy of two drugs to treat a form of leukemia

October 24, 2016
Researchers have determined that two Phase 1 drugs (CX-4945 and JQ1) can work together to efficiently kill T-cell acute lymphoblastic leukemia cells while having minimal impact on normal blood cells.

Shedding light on the mutational landscape of the most common pediatric cancer

September 22, 2016
A group of researchers from Columbia University, Rutgers University, and institutions in Europe and Japan have identified genomic alterations in pediatric relapsed acute lymphoblastic leukemia (ALL) that cause both therapy ...

Cutting-edge technology optimizes cancer therapy with nanomedicine drug combinations

February 24, 2015
In greater than 90 percent of cases in which treatment for metastatic cancer fails, the reason is that the cancer is resistant to the drugs being used. To treat drug-resistant tumors, doctors typically use multiple drugs ...

More chemo drugs don't improve treatment of rare bone cancer

August 26, 2016
Osteosarcoma patients with tumors that haven't responded well to the standard chemotherapy regimen have unimproved outcomes and more side effects when given two additional drugs, a large international trial has found.

Recommended for you

Researchers find novel mechanism of resistance to anti-cancer drugs

October 17, 2017
The targeted anti-cancer therapies cetuximab and panitumumab are mainstays of treatment for advanced colorectal cancer, the second leading cause of cancer-related deaths in the United States. However, many patients have tumors ...

Biology of childhood brain tumor subtypes offers clues to precision treatments

October 17, 2017
Researchers investigating pediatric low-grade gliomas (PLGG), the most common type of brain tumor in children, have discovered key biological differences in how mutated genes combine with other genes to drive this childhood ...

New assay may boost targeted treatment of non-Hodgkin lymphoma

October 17, 2017
Diffuse large B-cell lymphoma (DLBCL) is an aggressive cancer and the most frequently diagnosed non-Hodgkin lymphoma worldwide (nearly 40% of cases). Recent advancements indicate that both the prognosis and choice of treatment ...

Bolstering fat cells offers potential new leukemia treatment

October 16, 2017
Killing cancer cells indirectly by powering up fat cells in the bone marrow could help acute myeloid leukemia patients, according to a new study from McMaster University.

Study reveals complex biology, gender differences, in kidney cancer

October 13, 2017
A new study is believed to be the first to describe the unique role of androgens in kidney cancer, and it suggests that a new approach to treatment, targeting the androgen receptor (AR), is worth further investigation.

Cholesterol byproduct hijacks immune cells, lets breast cancer spread

October 12, 2017
High cholesterol levels have been associated with breast cancer spreading to other sites in the body, but doctors and researchers don't know the cause for the link. A new study by University of Illinois researchers found ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.