Paul Mischel, MD: Paul Mischel, MD
In a study published in Cancer Cell , Paul Mischel, MD, and his team worked with colleagues at The Scripps Research Institute to identify a metabolic vulnerability in the incurable brain cancer glioblastoma (GBM) and show how it might be exploited for therapy (2016;30(5):683-93). They found that the cholesterol GBM tumors consume in large amounts is almost exclusively produced by astrocytes and imported by the cancer cells. The researchers described how GBM cells re-engineer existing metabolic pathways to ramp up cholesterol import and retention, and demonstrated that these mechanisms can be undermined by an experimental drug for metabolic disease named LXR-623. This drug accumulates in the mouse brain, and the researchers showed it causes dramatic GBM cell death while sparing non-cancerous cells, including astrocytes. They also report the drug dramatically shrank GBM tumors from human patients that were implanted in mice, significantly prolonging their survival. Mischel told Oncology Times more about the research and how it might suggest a new approach to the development of cancer therapies.
1 What turned your attention to cholesterol import and metabolism as a potential vulnerability of GBM?
“Glioblastoma is one of the most sequenced of human cancers. We know more about the mutations that cause the disease than those, perhaps, of any other cancer, since GBM was the first tumor analyzed by The Cancer Genome Atlas. But, so far, we haven't been able to use that information for the benefit of patients. One reason is that the drugs that target those mutations have a difficult time crossing the blood-brain barrier and hardly make it into the tumor.
“We'd been thinking about this problem and wanted to essentially flip it on its head, approaching it by looking not at the oncogenes themselves but at how oncogenes change the way normal, unmutated enzymes are used by the cell. Cancer cells rely heavily on some of those changes for survival, which is why they're referred to as oncogene-induced co-dependencies. Based on previous observations, we suspected that GBM oncogenes radically alter the way GBM cells take up and use lipids like cholesterol. Further, the tumor's location in the brain would also influence how they process cholesterol. We hypothesized that the GBM cell's metabolic co-dependencies and the tumor's location would together create vulnerabilities very specific to GBM cells and targetable with drugs, or experimental drugs, that can get into the brain.”
2 Are there additional oncogene-induced co-dependencies or metabolic vulnerabilities in GBM and other cancers that might be ripe for targeting?
“Yes, I think so. We already have evidence that some transcription factors, though they aren't themselves mutated in any way, are used very differently by tumor cells than by normal cells, creating targetable co-dependencies. Tumor cells also process nutrients for glucose metabolism, lipid metabolism, and protein metabolism in very different ways, and these co-dependencies too may create vulnerabilities. This is probably true for cancers in general.
“Currently, the field tends to focus on targeting oncogene products themselves, and most of them are kinases whose enzymatic activity is blocked by the drugs. But many such drugs have trouble accessing their targets, especially in the brain. Yet there are many other drugs that target other critical enzymes and some are highly brain-penetrant. These could be used to treat cancers that either start in the brain or metastasize to the organ. So this whole idea of targeting those enzymes with drugs that aren't from cancer portfolios or pipelines may represent a very viable strategy, because co-dependencies stem from both the tumor itself and from the micro-environment in which the tumor sits.”
3 What are the prospects that this strategy or others like it might be evaluated in clinical trials for GBM?
“There are really two aspects to that. One is access to the drugs themselves for testing, and the other is the design of the clinical trial. As to the first point, part of the message of this paper is that drugs that come from outside the cancer therapy pipeline and even drugs that might have failed in the clinic could turn out to be very useful for cancer patients, including brain cancer patients. As we've shown in this paper, the CNS side effects of an experimental cardiovascular disease drug may actually turn out to be a benefit in brain cancer therapy because it suggests the drug is actually getting in there. This would mean really re-examining and repurposing drugs that have failed in trials for other indications.
“The other aspect is that it's very difficult to design clinical trials for cancers of the brain, since they stratify into a whole bunch of different diseases based on their molecular profiles. One of the fascinating aspects of our recent work is that the metabolic co-dependency we identify seems to be there in the vast majority of GBMs, as well as in cancers that have metastasized to the brain. So this raises the possibility of doing really interesting types of clinical trials—for example, global adaptive clinical trials like the recently announced GBM AGILE trial. In such trials, different therapies and strategies can be tested on patients, and the treatments themselves altered, based on the molecular profiles of tumors and new information. Such approaches, which are also being taken for other cancers, will accelerate the development of innovative therapies like the targeting of cancer cell metabolism and other co-dependencies.”
