Israeli algorithm may predict if cancer patients will respond to treatment

Original source here

Only between 20% and 40% of cancer patients respond to immunotherapy treatments, according to the latest research published in the medical journal Cell. Now, an Israeli company says it has developed a method to analyze and predict individual patient responsiveness to therapy and discover new targets to overcome resistance, which could save money on treatment costs and, most importantly, improve patient outcomes.

It could also spare patients from undue suffering, since immunotherapy treatments can trigger severe auto-immune reactions.

“Profiling of host response to immunotherapy addresses a major challenge in oncology today – understanding why treatment helps some patients but not others,” said Dr. Ofer Sharon, CEO of OncoHost. “By combining OncoHost’s proprietary technology with proteomic analysis (the large-scale study of proteins), we are able to predict how individual patients may respond to treatment. This is the first step towards developing successful personalized cancer treatment plans.”

OncoHost – founded by Prof. Yuval Shaked, a professor of cell biology and cancer science at the Technion-Israel Institute of Technology – last week opened a state-of-the-art proteomics laboratory for host response analysis in Binyamina, Israel. The lab is one of only a few specialized proteomics laboratories around the world and is the first industrial-level lab aimed at human host response in Israel.

The lab can create proteomic signatures by analyzing more than 1,000 proteins utilizing a very low volume of plasma. Using its “PROphet” platform, the lab will then help identify key biological processes and proteins that drive host response in patients undergoing cancer treatment.

The lab is operating in research mode, and OncoHost already has an active partner lab in Atlanta.
Until recently, according to Sharon, most cancer treatments focused on the interaction between treatment and the tumor, but few looked at the complexity of the individual patient’s body and how it was impacting or being impacted by the treatment.

“When you are looking at two patients who look clinically the same – the same cancer type and stage, the same background diseases, the same medications – yet one is responding and one is not, the difference is probably the way the body is responding to the cancer treatment,” Sharon said.

He said some patients’ bodies actually support tumors against treatment.

With the company’s solution, a physician could collect blood samples on premise and provide them to an OncoHost lab, which would then analyze the sample and tell the clinician where the patient falls in terms of his or her response to treatment.

“Today, when a physician treats a patient, he has no idea if the patient is going to respond or not,” Sharon shared. He noted that in cases of lung cancer, for example, it is predicted that only about 20%-30% of patients will respond to standard treatment. “Physicians know that only two or three out of 10 patients will respond to treatment. But they try the treatment and then have to assess its success after three or six months.

“What happens if the doctor doesn’t see a response?” Sharon asked. “Should he stop treatment? Choose another treatment? If so, which one?”

Generally, the answer is determined by a combination of clinical experience and guess work.

“Our platform will assess in a more educated way whether the patient is likely to ever respond,” Sharon explained. If the doctors know the patient is likely to respond, they might wait an additional six months before taking the patient off the standard treatment protocol. On the other hand, if doctors are already expecting the patient to be resistant to treatment, they would most likely look for alternatives, like clinical trials.

The next stage, which OncoHost is also working on, would be determining which alternative treatment the clinician should explore. By understanding the key biological pathways that drive patient response, the doctor can better discuss strategies for the next steps.

“There are hundreds of clinical trials out there; there are so many options,” Sharon continued. “We can also provide the doctor with clinical insights as to what needs to be done to improve the patient’s chances of response.”

Currently, OncoHost is running two large-scale clinical trials in the United States, Israel and Europe, focusing on validating the capability of its algorithm in cases of lung cancer and melanoma. If those trials prove effective, the company will begin the process of achieving FDA approval.