Team finds missing immune cells that could fight lethal brain tumors

Glioblastoma brain tumors can have an unusual effect on the body’s immune system, often causing a dramatic drop in the number of circulating T-cells that help drive the body’s defenses.

Where the T-cells go has been unclear, even as immunotherapies are increasingly employed to stimulate the body’s natural ability to fight invasive tumors.

Now researchers at Duke Cancer Institute have tracked the missing T-cells in glioblastoma patients. They found them in abundance in the bone marrow, locked away and unable to function because of a process the brain stimulates in response to glioblastoma, to other tumors that metastasize in the brain and even to injury.

The findings, published online Aug. 13 in the journal Nature Medicine, open a new area of exploration for adjunct cancer drugs that could free trapped T-cells from the bone marrow, potentially improving the effectiveness of existing and new immunotherapies.

“Part of the problem with all these immunotherapies—particularly for glioblastoma and other tumors that have spread to the brain—is that the immune system is shot,” said lead author Peter E. Fecci, M.D., Ph.D., director of the Brain Tumor Immunotherapy Program in Duke’s Department of Neurosurgery. “If the goal is to activate the T-cells and the T-cells aren’t there, you’re simply delivering therapy into a black hole.”

Fecci said the research team began its search for the missing T-cells after observing that many newly diagnosed glioblastoma patients have the equivalent immune systems of people with full-blown AIDS, even before they undergo surgery, chemotherapy and radiation.

Where most people have a CD-4 “helper” T-cell count upwards of 700-1,000, a substantial proportion of untreated glioblastoma patients have counts of 200 or less, marking poor immune function that makes them susceptible to all manner of infections and potentially to progression of their cancer.

Initially, the researchers hunted for the missing T-cells in the spleen, which is known to pathologically harbor the cells in certain disease states. But the spleens were abnormally small, as were the thymus glands—another potential T-cell haven. They decided to check the bone marrow to see if production was somehow stymied and instead found hordes of T-cells.

“It’s totally bizarre—this is not seen in any disease state,” Fecci said. “This appears to be a mechanism that the brain possesses for keeping T-cells out, but it’s being usurped by tumors to limit the immune system’s ability to attack them.”

When examining the stashed T-cells, Fecci and colleagues found that they lacked a receptor on the cell surface called S1P1, which essentially serves as a key that enables them to leave the bone marrow and lymph system. Lacking that key, they instead get locked in, unable to circulate and fight infections, let alone cancer.

Fecci said the research team is now working to learn exactly how the brain triggers the dysfunction of this S1P1 receptor. He said the current theory is that the receptor somehow is signaled to retract from the cell surface into the cell interior.

“Interestingly, when we restore this receptor to T-cells in mice, the T-cells leave the bone marrow and travel to the tumor, so we know this process is reversible,” Fecci said.

His team is collaborating with Duke scientist Robert Lefkowitz, M.D., whose 2012 Nobel Prize in Chemistry honored discovery of the class of receptors to which S1P1 belongs. They are working to develop molecules that would restore the receptors on the cells’ surface.

Source: Read Full Article

Machine-learning system determines the fewest, smallest doses that could still shrink brain tumors

MIT researchers are employing novel machine-learning techniques to improve the quality of life for patients by reducing toxic chemotherapy and radiotherapy dosing for glioblastoma, the most aggressive form of brain cancer.

Glioblastoma is a malignant tumor that appears in the brain or spinal cord, and prognosis for adults is no more than five years. Patients must endure a combination of radiation therapy and multiple drugs taken every month. Medical professionals generally administer maximum safe drug doses to shrink the tumor as much as possible. But these strong pharmaceuticals still cause debilitating side effects in patients.

In a paper being presented next week at the 2018 Machine Learning for Healthcare conference at Stanford University, MIT Media Lab researchers detail a model that could make dosing regimens less toxic but still effective. Powered by a “self-learning” machine-learning technique, the model looks at treatment regimens currently in use, and iteratively adjusts the doses. Eventually, it finds an optimal treatment plan, with the lowest possible potency and frequency of doses that should still reduce tumor sizes to a degree comparable to that of traditional regimens.

In simulated trials of 50 patients, the machine-learning model designed treatment cycles that reduced the potency to a quarter or half of nearly all the doses while maintaining the same tumor-shrinking potential. Many times, it skipped doses altogether, scheduling administrations only twice a year instead of monthly.

“We kept the goal, where we have to help patients by reducing tumor sizes but, at the same time, we want to make sure the quality of life—the dosing toxicity—doesn’t lead to overwhelming sickness and harmful side effects,” says Pratik Shah, a principal investigator at the Media Lab who supervised this research.

The paper’s first author is Media Lab researcher Gregory Yauney.

Rewarding good choices

The researchers’ model uses a technique called reinforced learning (RL), a method inspired by behavioral psychology, in which a model learns to favor certain behavior that leads to a desired outcome.

The technique comprises artificially intelligent “agents” that complete “actions” in an unpredictable, complex environment to reach a desired “outcome.” Whenever it completes an action, the agent receives a “reward” or “penalty,” depending on whether the action works toward the outcome. Then, the agent adjusts its actions accordingly to achieve that outcome.

Rewards and penalties are basically positive and negative numbers, say +1 or -1. Their values vary by the action taken, calculated by probability of succeeding or failing at the outcome, among other factors. The agent is essentially trying to numerically optimize all actions, based on reward and penalty values, to get to a maximum outcome score for a given task.

The approach was used to train the computer program DeepMind that in 2016 made headlines for beating one of the world’s best human players in the game “Go.” It’s also used to train driverless cars in maneuvers, such as merging into traffic or parking, where the vehicle will practice over and over, adjusting its course, until it gets it right.

The researchers adapted an RL model for glioblastoma treatments that use a combination of the drugs temozolomide (TMZ) and procarbazine, lomustine, and vincristine (PVC), administered over weeks or months.

The model’s agent combs through traditionally administered regimens. These regimens are based on protocols that have been used clinically for decades and are based on animal testing and various clinical trials. Oncologists use these established protocols to predict how much doses to give patients based on weight.

As the model explores the regimen, at each planned dosing interval—say, once a month—it decides on one of several actions. It can, first, either initiate or withhold a dose. If it does administer, it then decides if the entire dose, or only a portion, is necessary. At each action, it pings another clinical model—often used to predict a tumor’s change in size in response to treatments—to see if the action shrinks the mean tumor diameter. If it does, the model receives a reward.

However, the researchers also had to make sure the model doesn’t just dish out a maximum number and potency of doses. Whenever the model chooses to administer all full doses, therefore, it gets penalized, so instead chooses fewer, smaller doses. “If all we want to do is reduce the mean tumor diameter, and let it take whatever actions it wants, it will administer drugs irresponsibly,” Shah says. “Instead, we said, ‘We need to reduce the harmful actions it takes to get to that outcome.'”

This represents an “unorthodox RL model, described in the paper for the first time,” Shah says, that weighs potential negative consequences of actions (doses) against an outcome (tumor reduction). Traditional RL models work toward a single outcome, such as winning a game, and take any and all actions that maximize that outcome. On the other hand, the researchers’ model, at each action, has flexibility to find a dose that doesn’t necessarily solely maximize tumor reduction, but that strikes a perfect balance between maximum tumor reduction and low toxicity. This technique, he adds, has various medical and clinical trial applications, where actions for treating patients must be regulated to prevent harmful side effects.

Optimal regimens

The researchers trained the model on 50 simulated patients, randomly selected from a large database of glioblastoma patients who had previously undergone traditional treatments. For each patient, the model conducted about 20,000 trial-and-error test runs. Once training was complete, the model learned parameters for optimal regimens. When given new patients, the model used those parameters to formulate new regimens based on various constraints the researchers provided.

The researchers then tested the model on 50 new simulated patients and compared the results to those of a conventional regimen using both TMZ and PVC. When given no dosage penalty, the model designed nearly identical regimens to human experts. Given small and large dosing penalties, however, it substantially cut the doses’ frequency and potency, while reducing tumor sizes.

The researchers also designed the model to treat each patient individually, as well as in a single cohort, and achieved similar results (medical data for each patient was available to the researchers). Traditionally, a same dosing regimen is applied to groups of patients, but differences in tumor size, medical histories, genetic profiles, and biomarkers can all change how a patient is treated. These variables are not considered during traditional clinical trial designs and other treatments, often leading to poor responses to therapy in large populations, Shah says.

Source: Read Full Article

Analysis of prostate tumors reveals clues to cancer’s aggressiveness

Using genetic sequencing, scientists have revealed the complete DNA makeup of more than 100 aggressive prostate tumors, pinpointing important genetic errors these deadly tumors have in common. The study lays the foundation for finding new ways to treat prostate cancer, particularly for the most aggressive forms of the disease.

The multicenter study, which examined the genomes of tumors that grew and spread quickly, was led by Washington University School of Medicine in St. Louis and the University of California, San Francisco. The research appears July 19 in the journal Cell.

“This study could aid the search for better therapies to treat aggressive prostate cancer,” said co-first author Christopher A. Maher, Ph.D., an associate professor of medicine and an assistant director at The McDonnell Genome Institute at Washington University School of Medicine. “More immediately, the new information could help doctors find ways to identify which patients may develop aggressive tumors, and help guide their treatment decisions.”

More than 160,000 cases of prostate cancer are diagnosed each year in the U.S. While some 80 percent of prostate cancer patients have tumors that are slow-growing and have effective treatment options, about 20 percent of such patients develop the most aggressive forms of the disease—the focus of the new study.

Most genetic studies of prostate cancer have focused on parts of the genome that control what proteins a tumor manufactures. Proteins act like the machinery of cells. When they function properly, proteins perform cellular tasks required for good health. But when proteins don’t work properly, disease, including cancer, can result.

Still, genes that make proteins represent only 1 to 2 percent of the entire genome. The new analysis is the first large-scale study of the whole genomes—all of the DNA, including all of each tumor’s genes—of metastatic prostate tumors, and reveals that many of these tumors have problems in the sections of the genome that tell protein-coding genes what to do.

“Protein-coding genes are important, but when you focus only on them you can miss mutations in regions of the genome that regulate those genes,” Maher said.

The researchers were surprised to find that about 80 percent of the aggressive tumors studied had the same genetic alterations in a region of the genome that controls the androgen receptor, Maher said. This genetic error dialed up levels of androgen receptor on prostate cancer cells. Such receptors bind to male hormones such as testosterone and drive tumor growth.

“This was one of the most surprising findings,” said Maher, also a research member of Siteman Cancer Center at Barnes-Jewish Hospital and Washington University School of Medicine. “We saw too many repeated copies of DNA in this region of the genome. In some of these patients, the androgen receptor looks totally normal. But they have too much androgen receptor because the receptor’s regulatory region is dialed up, which would be missed by the protein-coding focused sequencing studies.”

A common treatment for prostate cancer, beyond the traditional options of surgery, chemotherapy and radiation, involves androgen deprivation therapy, in which drugs are used to block testosterone from binding to the androgen receptor. Since prostate tumors are often hormone-driven cancers, blocking testosterone from binding this receptor slows tumor growth.

All the men in this study had tumors that developed resistance to androgen deprivation therapy, meaning the androgen receptor is always switched on, fueling the tumor, whether testosterone is present or not. Patients in this situation have no effective treatment options. The researchers showed that more than 80 percent of these patients had mutations that help explain the aggressiveness of their cancers; these genetic errors activated the androgen receptor.

Source: Read Full Article