Practice Update: Haematology & Oncology

CNS/BRAIN

9

EXPERT OPINION DR STEVEN TOMS Metastatic processes of glioblastoma Dr Farzanna Haffizulla speaks to Dr Steven Toms, Director of Neurosurgery at the Geisinger Health System in Danville Pennsylvania, on the latest in the metastatic process of glioblastoma.

has not changed very much from the primary site to the secondary cancer. But if we look at the proteins, they do change quite a bit. It appears that most of these changes occur in the epigenetic phenomena; so, it’s not the genes themselves changing, but, is in fact, things like methylation and acetylation that can turn on and off genes. So, different parts of genes are transcribed, and, remember, after transcription into RNA, they are then translated into protein, and the other thing that is changing is the microRNA environ- ment realm. The microRNAs have been big in scientific news in the last 10 years because we find they’re regulatory elements that help turn on and off genes. Dr Haffizulla: And they’re packaged in extracellular vesicles, right? Dr Toms: They can be. Dr Haffizulla: And there is a commu- nication that happens between, not just the tumour cells, but the normal brain parenchyma, and there is still a lot of research to be done to under- stand that communication pattern. Dr Toms: Exactly, because the cancer itself is made up of many different cells. You have to think of cancer as a whole organ, like you would a liver or a heart; so, you have the cancer stem cells, which are the ones that are dividing and are so resistant to our chemotherapy, the cancer stromal cells, and then there are fibroblasts. There are endothelial cells. There are the immune cells that are trying to fight against the cancer, all in this tiny little package, and there’s an in- credible amount of communication among those different cells. We find that some of the cells, like fibroblasts in and around tumours, are enablers of the cancer themselves and com- municate signals that help the cancer survive and maintain itself.

Dr Haffizulla: What new data have come to light helping us to un- derstand the metastatic process of glioblastoma? Dr Toms: Certainly, glioblastoma is not one of those cancers that typically metastasise outside of the central nervous system. So, most commonly, we have to deal with the migration or the crawling of the glioblastoma cells throughout the central nerv- ous system that get them away from our therapeutics. Most of the therapeutics that we use in cancer treatment right now for glioma are local therapeutics. I, as one of the barbarian surgeons, can go in and cut out the disease; however, when I cut everything out and can’t see anything left on MRI, there are still tens of thousands, if not millions, of cancer cells left behind. Because what happens in a glioblastoma is that it sort of reverts back to what it used to be. If we go back to thinking about brain development, originally all the cells that populate the brain are born right next to the ventricle fluid sac in the brain, and then they had to crawl away from that. So, when we get a cancer that derives from one of the glial cells, they start to revert to what they did when they were young. They crawl, and they crawl away from the edges of the area that we operate, the edges of what we see on MRI. In our laboratory, led by Dr Nikolaos Tapinos, who’s a wonderful MD/ PhD working in our system, Dr Tap- inos and his partner Dr Atom Sarkar have been able to show, one, that we’ve found a new molecule that can chemo-attract, or bring these tu- mour cells back toward where they

seems to inhibit the migration pro- cess, and, interestingly enough, the entire process seems to occur very locally, right down in these little foot processes, again called pseudopodia. We found a local translation mecha- nism that occurs down in there that is dependent on an elongation factor called eIF4, which allows this to oc- cur, and, when we inhibit that pro- cess, the pseudopods stop forming and the migration is settled down. So, again, we can keep the cells from wandering away from us, and keep them in the neighborhood where we can treat them more effectively. Dr Haffizulla: Just to backtrack a little bit, you’re looking at animal models to look at the milieu sanctuary, as it were, as a glioblastoma is growing. The sex of the cells being used, both male and female? Dr Toms: Both male and female, yes. Dr Haffizulla: Now, of course, sex- and gender-specific medicine, more precision medicine, and person- alised targeted therapy, is just sort 56.8% across all four cohorts. Although first-line (adjuvant) radiation is clearly beneficial for completely resected EPN-PFA, a substantial proportion of patients with EPN-PFB can be cured with surgery alone, and patients with relapsed EPN-PFB can often be treated successfully with delayed external- beam irradiation. CONCLUSION The most impactful bio- marker for posterior fossa ependymo- ma is molecular subgroup affiliation, independent of other demographic or treatment variables. However, both EPN-PFA and EPN-PFB still benefit from increased extent of resection, with the survival rates being particularly poor for subtotally resected EPN-PFA, even with adjuvant radiation therapy. Patients with EPN-PFB who undergo gross total re- section are at lower risk for relapse and should be considered for inclusion in a randomised clinical trial of observation alone with radiation reserved for those who experience recurrence. Therapeutic impact of cytoreductive surgery and irradiation of posterior fossa ependymoma in the molecular era: A retrospective multicohort anal- ysis. J Clin Oncol 2016 Jun 06;[Epub ahead of print], Ramaswamy V, Hielscher T, Mack SC, et al.

grew up. We’re working along with Robert Langer from MIT on some hydrogels and some treatments that we hope to bring to clinical trial in the next few years, so that we can, after surgery, put a hydrogel with these molecules in that will draw all the tumour cells back where we can treat them more effectively. Then also, the Tapinos Lab has figured out some of the very basic mechanisms of how these glioma stem cells extend little foot processes called pseudopodia. These pseudopodia grab onto the axons and dendrites around them, and it’s almost like watching someone crawl up a rope. When we look at them under the microscope we can see them extend, grab hold, and move themselves along, much the way their ancestors did when we were born – you know, when the radial glia had to migrate away to form what becomes our brain. The laboratory has found a molecule called LCK, from lymphocyte-spe- cific protein tyrosine kinase, which

of extrapolating from that animal research. I want to see if...you know, as you’re now translate it to human research, what kinds of gaps do you think might be happening. Just to sort of think out of the box a little bit. Dr Toms: Exactly. So, I think that gaps are there, of course; yet, this is but one part of the process that keeps gliomas evading us. You know, we and many other labs across the world have been studying this tumour be- cause it’s one of the most malignant cancers we deal with. When I started in this field 25 years ago, at meet- ings such as I’m attending today, the Society for Neuro-Oncology would be 50 or 100 people getting together, and now we have nearly 2000 people getting together every day. So, this is but one part of the story. The story that we’re finding here, likely, will be a story that will translate to other disease entities. Metastasis in other cancers, for example, do re- quire cell motility. It’s very likely that the processes we see in the glioma are preserved in other parts of the body and other cancer motility, but we haven’t tested that hypothesis yet. Dr Haffizulla: Now, as the tumour itself metastasises, there is communication happening from the main lesion to the micro-metastatic lesions. Have we been able to decide that yet? Dr Toms: Well, certainly, there are changes that occur from the primary tumour to the secondary, or metastat- ic, sites. For a long time, we thought these were basic genetic changes; but, honestly, if we look at the cells, phe- notypically or as a pathologist would see, they look the same. If we look at the genotype of the cell, it really

Dr Steven A Toms is Associate Chief Medical Officer, Director of Neurosurgery, and Co-Director of the Neurosciences

JOURNAL SCAN Benefit of cytoreductive surgery and irradiation in posterior fossa ependymoma Journal of Clinical Oncology Take-home message

Institute at Geisinger Clinic in Danville, Pennsylvania. Prior to coming to Geisinger, he was the head of the Section of Metastatic Disease for the Brain Tumor Institute at the Cleveland Clinic, Director of Neurosurgical Oncology at Oregon Health and Sciences University, and Director of Neurosurgical Oncology and the Skull Base Center at Vanderbilt.

• In a retrospective study of posterior fossa ependymomas (n = 820), the therapeutic value of cytoreductive surgery and radiation therapy was evalu- ated for the two molecular variants EPN-PFA and EPN-PFB. Incompletely resected ependymomas of either molecular variant conferred poorer prognosis. Patients with the EPN-PFA molecular variant with incomplete resection had a 5-year progression-free survival rate ranging from 26.1% to 56.8%. Patients with completely resected EPN-PFA benefited from first-line radiation while surgery alone was sufficient to cure a large proportion of patients with EPN-PFB. External-beam irradiation was successful in patients with EPN-PFB who had relapsed. • Patients with incompletely resected ependymomas of the EPN-PFA variant have poor survival rates. Increased extent of resection improves clinical outcomes in patients with EPN-PFA and patients with EPN-PFB epend- ymoma variants.

Key new trial data in neuro-oncology INTERVIEW WITH DR PATRICK Y. WEN

PURPOSE Posterior fossa ependymoma comprises two distinct molecular vari- ants termed EPN-PFA and EPN-PFB that have a distinct biology and natural his- tory. The therapeutic value of cytore- ductive surgery and radiation therapy for posterior fossa ependymoma after accounting for molecular subgroup is not known. METHODS Four independent nonover- lapping retrospective cohorts of pos- terior fossa ependymomas (n = 820) were profiled using genome-wide

methylation arrays. Risk stratification models were designed based on known clinical and newly described molecular biomarkers identified by multivariable Cox proportional hazards analyses. RESULTS Molecular subgroup is a pow- erful independent predictor of outcome even when accounting for age or treat- ment regimen. Incompletely resected EPN-PFA ependymomas have a dismal prognosis, with a 5-year progression- free survival ranging from 26.1% to

Dr Patrick Wen, Director of the Center of Neuro Oncology at the Dana Farber Cancer Institute and Director of the Division of Cancer Neurology at Brigham Women’s

Hospital, shares his perspective on the key take- aways in neuro- oncology from ASCO 2016.

VOL. 1 • No. 2 • 2016