Texas Tech Researchers Develop Models For Targeted Cancer Therapy
Released: 8-Jun-2015 10:05 AM EDT
Source Newsroom: Texas Tech University
[COMMENT: While not directly related to Rhabdomyosarcoma, this is a major breakthrough that should be transferable. Congrats to Drs. Ranadip Pal and Charles Keller and doctoral student Noah Berlow for their incredible work on this model. (ARE)]
Texas Tech Researchers Develop Models For Targeted Cancer Therapy
Newswise — The results of a recent study on targeted therapy of a specific type of brain cancer were published by Nature Medicine showing specific progress in cancer treatment.
Diffuse intrinsic pontine glioma (DIPG) may be one of the lesser-known forms of cancer, yet may be one of the most diabolical.
DIPG is a cancer of the brain stem that affects mostly children, 200-400 per year in the United States, and has less than a 1 percent survival rate five years after diagnosis. Radiation treatment provides only temporary relief.
Thanks to a team of scientists, including Texas Tech University associate professor Ranadip Pal in the Department of Electrical & Computer Engineering and electrical engineering doctoral student Noah Berlow, the tide may be turning.
Pal was one of four senior authors and Berlow was a contributing author on the study, funded partly by the National Science Foundation, investigating the effectiveness of targeted drug therapy on DIPG cancer cells. The results of the study were recently published by Nature Medicine in an article entitled “Functionally Defined Therapeutic Targets in Diffuse Intrinsic Pontine Glioma.”
“We treated the tumor like an engineering system, like a computer,” Pal said. “You can push different components inside a system and block different paths and see how a computer reacts. This study is similar to that where we can try different drugs to look at the output and how much sensitivity there is to that drug and generate a model for the tumor.”
One study revealed the FDA-approved drug Panobinostat showed prolonged survival in mice with DIPG. However, some DIPG samples showed resistance to the drug, meaning Panobinostat combinations or new drug combinations will need to be identified for some patients.
Using a method invented in Pal’s laboratory in the last three years called Probabilistic Target Inhibition Maps (PTIM), Pal and Berlow generated mathematical models for anti-cancer drug sensitivity prediction from drug screen and gene expression data. This allowed researchers to determine which drugs were most effective, either by themselves or in combination, in killing or limiting the growth of cancer cells.
“This article reports the result of an international collaborative initiative of medical and computational modeling research laboratories working with families and foundations to study DIPG and suggest potential new treatments using targeted drugs,” Pal said.
Long time coming
Pal, along with Dr. Charles Keller at the Children’s Cancer Therapy Development Institute in Fort Collins, Colorado, Dr. Michelle Monje at Stanford Medical School and Dr. Jacques Grill at Universite Paris-Sud, directed the study. Pal said he first met Keller in the mid-2000s when Keller was at the University of Texas Health Sciences Center in San Antonio and Pal was working on his doctoral degree at Texas A&M.
Pal came to Texas Tech in 2007, and a few years later the two began discussing the work Keller’s lab at the Oregon Health & Science Center was doing on targeted drug therapy. That’s where the idea of the PTIM was developed.
“If we know the targets of the drugs and we know their sensitivity, from that we can come up with a model,” Pal said. “We apply individual drugs to the tumor and look at the output, then come up with a mathematical model that is predictive of combining two drugs and seeing whether they work.”
Pal was involved in the design and interpretation of the experiments and the results related to the computational modeling aspects of the project. Berlow, who will earn his doctorate in June, analyzed the data and implementation of the PITM model for the data set to generate the predictions.
“Chemotherapy is the nuclear option because it hits everything. But that doesn’t mean it kills all the cancer cells,” Berlow said. “If chemo doesn’t work then you have stronger chemo-resistant cancer cells ready to grow again. And using chemo is extremely damaging to the human body. The idea with targeted therapy is to do significantly less damage. The ability to precisely target cancer cells with the right kind of drugs is the next step, and you’d have a hard time finding anyone who wouldn’t want that.”
The current study is an unprecedented collaborative effort of scientists from 13 institutions working toward a common goal of identifying promising therapies for DIPG, a fatal brain tumor with limited therapeutic progress.
“The result is highly significant as there are limited therapeutic options for children with DIPG,” Pal said, “and the current research shows that Panobinostat alone or in combination with other drugs could have substantial therapeutic activity.”
The research results paved the way for a clinical trial currently being designed for single-drug therapy with Panobinostat, which will likely begin enrolling patients later this year. However, the observed resistance to Panobinostat in some DIPG cell lines suggests a combination of two or more drugs might be required to effectively tackle DIPG.
Electrical engineering and cancer research might not seem like they go together, but Berlow said there is a great need for technology to gather and analyze data as well as developing methods for doing so.
That’s where software, algorithms and statistical analysis come into play, and why the Pal lab was able to play a key role in the DIPG drug therapy research. The ability to generate models from the large amount of data collected compliments the biological work done in terms of experiments based on the data generated.
Berlow said their focus wasn’t specifically on designing drugs or drug combinations, but rather discovering the underlying functional biological drivers of cancer cells. This leads to the discovery of specific pathways with common targets, and specific drugs can be used or developed that attack those pathways, therefore killing the cancer cells and inhibiting their ability to develop resistance to drugs.
This information can be used to develop personalized drug combinations to treat patients on a case-by-case basis.
“In the end, that’s what we can target, and that’s what we can treat,” Berlow said. “If you can offer lower toxicity, better outcomes, happier patients and longer lives, there’s nothing bad about that at all. Most of the clinicians and oncologists I meet got into this because they want to help people. Having to watch patients dying and not being able to do anything about it is painful. Having something like our approach to aid in personalized targeted therapy for a patient is really important.”
Pal likened the collaboration between engineering and medicine to the development of a magnetic resonance imaging (MRI) machine, which has become a common, necessary diagnostic tool for doctors. Pal sees the generation of computer models in targeted drug therapy eventually becoming just as big a part of everyday medicine.
“There are so many similarities between medicine and engineering,” Pal said. “This may not be a cure, but it is the first step in defeating this deadly disease and we can go from there. This is a very good start, a very promising start that might excite people to look more into this area. From our perspective, this case is a validation of our overall framework, and we’d definitely like to apply it to other forms of cancer.”