Article In Brief
In a mouse model of methotrexate chemotherapy-induced dysfunction, researchers found that oligodendrocyte precursor cells (OPCs) were depleted in white matter, leading to a persistent deficit in myelination and cognitive deficits. Feeding the methotrexate-exposed mice chow enriched with an investigational drug that inhibits colony-stimulating factor 1 receptor depleted the number of microglia, normalized the OPC population, restored normal myelin sheath thickness, and rescued cognitive performance.
A cascade of alterations to neuroglia have been observed in postmortem brain samples of both children and mice exposed to chemotherapy, offering one of the first mechanistic explanations — and a potential treatment target — for the cognitive dysfunction sometimes associated with such treatment.
In five children and one young adolescent exposed to multi-agent chemotherapy for the treatment of cancer, the density of white-matter oligodendrocyte precursor cells (OPCs) in postmortem frontal lobe tissue was more than 70 percent lower than in age-matched youths who died of other causes, a paper published in Cell on January 10 reported.
In subsequent experiments involving mice exposed to high-dose methotrexate, the paper described not only a similar depletion of OPCs but also thinner myelin sheaths and a reduction in the number of mature oligodendrocytes in subcortical white matter. What's more, these changes were associated with the presence of inflammatory microglia, as well as with astrocytes in a state of neurotoxic activation.
The paper reported that feeding the methotrexate-exposed mice chow enriched with an investigational drug that inhibits colony-stimulating factor 1 receptor (CSF1R) depleted the number of microglia, normalized the OPC population, restored normal myelin sheath thickness, and rescued cognitive performance.
The study, led by researchers at Stanford University's department of neurology and neurological sciences, drew varying reactions. Some neurologists called it a “brilliant” and “exciting” advance in understanding the underlying neural mechanisms by which chemotherapy may cause so-called chemo fog or chemo brain. Others, however, noted that the postmortem brain samples involved relatively few subjects, all of them young, and that the mice studies involved only methotrexate. They pointed out that other factors can also contribute to cognitive dysfunction after cancer therapy, including endocrine effects of the therapy, the stress of diagnosis and treatment, and effects of the cancer itself.
“Chemo brain is probably better called ‘multi-factorial brain dysfunction,’” said Priya U. Kumthekar, MD, assistant professor of neurology and oncology at Northwestern University Feinberg School of Medicine, who was not involved with the study. “Chemotherapy is certainly a component of that, as this Cell paper makes clear. But there are other factors involved in how people respond cognitively.”
The paper in Cell described multiple experiments, beginning with an examination of preserved postmortem samples, collected over decades, from three children and one young adult, each of whom had died of cancer following treatment with a variety of chemotherapies. (None had undergone cerebral radiation.) OPC cells were significantly depleted in the chemotherapy-treated samples (4,150 ± 876 cells/mm3) compared with the four controls who died of other causes (14,893 ± 2,063 cells/mm3), the researchers found. The decrease was seen primarily in subcortical white matter rather than in gray matter.
Not all of the chemotherapy-treated children had received methotrexate. But a postmortem sample from a fifth child, who had recently died at the age of 3 following treatment with high-dose methotrexate, revealed a “stark depletion in OPCs in the subcortical white matter,” said the senior author of the paper, Michelle Monje, MD, associate professor of pediatric neurology at Stanford.
Study Design, Observations
To test the effect further, Dr. Monje's group developed a juvenile mouse model of methotrexate-induced neurological dysfunction. They observed a depletion of white matter OPCs similar to that seen in the human samples, increased but incomplete OPC differentiation, and a persistent deficit in myelination.
“The OPCs were going down the maturation pathway but were getting stuck,” Dr. Monje told Neurology Today. “We then wanted to know whether this reflected an intrinsic lesion to the OPCs caused by exposure to methotrexate, or whether it reflected disturbances to the microenvironment in which they exist.”
Dr. Monje's group sought to answer that question by observing what happened when they transplanted OPCs from methotrexate-naive mice into the previously methotrexate-exposed brains. They found that these OPCs likewise exhibited aberrant differentiation.
Seeking next to discern what in the microenvironment was causing the OPCs to grow abnormally, the team examined the microglia and found they became inflamed in response to methotrexate.
The Stanford group then exposed the astrocytes to methotrexate, finding no direct effect. But when exposed to the inflamed microglia, the astrocytes became activated into a neurotoxic state.
Even six months later, all three impairments — on OPCs, astrocytes, and microglia — were still in evidence, leading to what the authors of the Cell paper called “persistent tri-glial dysregulation.”
In a final experiment, Dr. Monje's group tested whether depleting the inflammatory microglia might normalize the cognitive test results of the methotrexate-treated mice. They used a small-molecule investigational agent, PLX5622, which previously was shown to eliminate microglia and improve cognitive function in mouse models. As hoped, PLX5622 corrected the tri-glial dysregulation as well as the mice's performance on the novel object recognition test of cognition.
Lynne P. Taylor, MD, FAAN, who holds the Alexander M. Spence endowed chair in neuro-oncology and serves as co-director of the Alvord Brain Tumor Center at the University of Washington in Seattle, called the paper a “brilliant beginning in helping us understand methotrexate-induced cognitive dysfunction in our primary CNS lymphoma patients and solid-tumor patients with leptomeningeal disease exposed to methotrexate.”
“Methotrexate-induced demyelination is quite clinically severe and striking,” Dr. Taylor said, “and we have been trying to understand it for some time. The Monje paper offers an elegant basic-science explanation for what we observe.”
Patrick Y. Wen, MD, FAAN, professor of neurology and director of the Center for Neuro-Oncology at Dana-Farber Cancer Institute, concurred.
“Having such an improved understanding of the mechanism of cognitive dysfunction from methotrexate treatment is a really important advance,” Dr. Wen said. “And the fact that Dr. Monje has found a potential strategy to reduce this toxicity, using the CSF1R inhibitor to decrease microglial activity, is really exciting. Whether this will translate into benefits for patients is not clear, but it's something that really should be tested.”
Kerstin Hermelink, PhD, a senior psychologist at the University of Munich, Germany, who has published a series of papers examining the role of chemotherapy in causing cognitive dysfunction in breast-cancer patients, had a different view of the paper, however.
“Cognitive function is particularly well-studied in breast cancer patients,” Dr. Hermelink said in an email to Neurology Today. “In most studies, about a third of patients show some measurable decline, which is generally very mild and far from being debilitating, while cognitive function is not affected in the majority of patients. Moreover, chemotherapy does not have a singular role in the causation of cancer-related cognitive deficits; in many studies, deficits were observed already before the start of treatment and in cancer patients who did not receive chemotherapy.”
In a 2017 paper in the Journal of the National Cancer Institute, Dr. Hermelink and colleagues compared cognitive functioning in breast cancer patients before and after receiving chemotherapy and in comparison to breast cancer patients who did not receive chemotherapy, as well as to healthy controls. The study found only mild cognitive impairment in the breast cancer patients, largely irrespective of whether they had received chemotherapy. Instead, the effect was seen prior to any treatment, mediated primarily by clinically diagnosed symptoms of post-traumatic stress caused by the cancer.
Other recent papers have reached similar conclusions about the limited effects of chemotherapy on cognition, including a 2015 meta-analysis of breast cancer patients and a prospective longitudinal study of colorectal patients. Both concluded that the cancer diagnosis itself, rather than treatment with chemotherapy, is the cause of cognitive impairment.
Dr. Wen stated, however, that the cognitive effects of methotrexate are known to be uniquely severe, far more so than other chemotherapeutic agents.
“Every neurologist knows that methotrexate is the most neurotoxic type of chemotherapy,” he said. “The neurocognitive effects are so severe, I don't think anyone thinks it's like ordinary ‘chemo brain,’ where some of it could be due to stress.”
Dr. Kumthekar agreed that Dr. Monje's paper offers convincing proof of a biological pathway by which methotrexate causes its cognitive effects.
“It shows us a nice, concrete pathophysiology,” she said. “But despite that really clean paradigm, I still believe that chemotherapy is not the only thing that impacts cognition in cancer patients. The psychological effects of post-traumatic stress disorder [PTSD], depression, anxiety and adjustment disorder are still there.”
Lisa R. Rogers, DO, FAAN, professor of neurology and medical director of the neuro-oncology program at the University of Cleveland Medical Center, semphasized that methotrexate is a drug that carries a unique risk of severe neurotoxicity, resulting in cognitive impairment. Systemic methotrexate is the backbone of therapy for CNS lymphoma and intraCSF methotrexate is commonly used to prevent or treat leptomeningeal metastasis of cancer. Thus, understanding the pathophysiology of methotrexate neurotoxicity, and potentially developing preventive strategies for this, is of considerable clinical importance.”
“Michelle Monje has demonstrated, in an elegant research model, that methotrexate produces a lasting dysfunction of oligodendrocyte lineage cells, astrocytes and microglia. Importantly, microglial activation is necessary for this dysregulation, indicating a possible therapeutic target,” she said.
“Some other chemotherapies that are administered to cancer patients are also associated with cognitive impairment,” she continued. “In this setting, a number of other clinical factors may contribute to “chemobrain,” such as baseline cognitive impairment, treatment-related fatigue, disrupted sleep patterns, mood abnormalities, hormonal therapy, and PTSD, as demonstrated by the study by Dr. Hermelink and colleagues.”
Drs. Monje, Kumthekar, Rogers, and Wen disclosed no relevant conflicts of interest.
Dr. Taylor received royalties from Oxford press for the publication of Navigating Life with a Brain Tumor.