As a recognized leader in brain cancer research, the MCW Neuro-Oncology Program brings together experts across various specialties to improve detection, diagnosis, and treatment outcomes for patients in Wisconsin. However, driving the program to the forefront of the field is the one-of-a-kind MCW Neuro-Oncology Brain Bank that houses more than 150 whole brain samples, nearly 100 of which are from patients with glioblastoma—the most common primary malignant brain tumor in adults. After more than a decade in the making, the Brain Bank is rapidly evolving into a hub for neuro-oncology-focused research that equips MCW scientists (and many national and global collaborators) with imaging expertise, unique equipment, and an expansive database to accelerate breakthroughs and bring hope to patients with brain cancer.
“What started as a conversation with Drs. Jennifer Connelly and Elizabeth Cochran 14 years ago has transformed into what we believe is the largest brain cancer-specific, whole brain tissue bank in the world,” said Peter LaViolette, PhD, MS, Professor of Radiology and Director of the MCW Cancer Center Translational Metabolomics Shared Resource. “A goal of my lab is to create and validate imaging techniques that improve both patient outcomes and treatment efficacy. To ensure our imaging techniques are accurate, we need to compare them with real tissue samples; studying whole brains is particularly helpful because brain tumors change over time with treatment.”
“The Brain Bank is an asset that has really transformed neuro-oncology research at MCW. So many studies have been made possible by having whole brain samples, not just tissue blocks. However, what’s truly unique about this resource is the database that contains each patient’s clinical images, medical histories, and digital tissue samples. This tool may help accelerate discoveries by providing researchers with critical information to improve diagnosis, treatment, and understanding of brain tumors,” said Dr. LaViolette.
A pioneer in medical imaging techniques, the LaViolette lab was the first in the field to develop and implement radio-pathomic imaging, a novel technique that uses advanced imaging methods and machine learning to provide scientists with a deeper, more accurate understanding of tumor presence and behavior. This unique expertise, coupled with the cultivation of a robust Brain Bank, has been pivotal in advancing brain cancer research, and has also helped the institution build a global presence that positions the MCW Neuro-Oncology Program as one that pushes the boundaries of knowledge and practice.
“As a member of the prestigious Glioma Longitudinal AnalySiS (GLASS) Consortium, our neuro-oncology team collaborates with neuropathologists, clinicians, scientists, and bioinformaticians from top institutions across the globe to accelerate research on how glioma tumors change over time, and to find weaknesses in these tumors that can be targeted with treatments,” said Jennifer Connelly, MD, Professor of Neurology and Neuro-Oncology. “While all the member institutions have tissue samples, MCW is the only one that has tissue from its first original surgery, its redo surgery, and post-mortem. When we were asked to join this group, it was eye-opening to them that we had these resources. It really speaks to the value of our Brain Bank and institution.”
“The MCW Neuro-Oncology Program is truly translational. It took years to get to this point and involved multi-disciplinary collaboration. Dr. LaViolette couldn’t do what he does without neuro-oncologists in the clinic. I wouldn’t be able to offer additional research opportunities without Dr. LaViolette and his lab. Now we have other experts, like Dr. Max Krucoff, who are involved on a surgical level and moving from retrospective to prospective analysis. It’s incredible to see the teamwork that has helped this program grow,” said Dr. Connelly.
MCW Brain Bank Accelerates Cancer Breakthroughs in Wisconsin and Beyond
The Brain Bank has been crucial in several high-impact studies conducted at MCW and in collaboration with other top institutions. Its resources have also been leveraged by centers that are designated by the National Cancer Institute, including MD Anderson Cancer Center and the University of Southern California Norris Comprehensive Cancer Center. But recently, with the Brain Bank at arm’s reach, it’s MCW trainees who are making significant strides to advance brain cancer detection and treatment.For example, a new study recently published in Neurosurgery found that 41.5% of patients with glioma had tumor areas that didn’t show up with conventional imaging such as Magnetic Resonance Imaging( MRI), which linked to worse survival outcomes. MCW researchers developed a new method, radio-pathomic tumor probability mapping, that successfully found hidden tumors that traditional scans missed in 69% of patients before surgery; these unseen tumors were also linked to reduced survival. Samuel Bobholz, PhD, post-doctoral fellow in the LaViolette lab and first author of the study, said that by being able to forecast tumor growth, researchers can add further information to critical treatment decisions, as well as better understand the tumor’s biological response to treatment, particularly those that impact the appearance of tumors on imaging.
“These results may help improve treatment direction in the clinical setting. We’ve talked to clinicians internally and externally who have been excited to use this technique for identifying non-enhancing tumor to increase the amount of tumor that can be removed during surgery, as well as to define post-surgical radiation dose maps to target invisible tumor while sparing healthy or non-tumor tissue from unnecessary dosage. Through clinical trials affirming the efficacy of these approaches, we can hopefully improve patient survival for a particularly dismal diagnosis,” said Dr. Bobholz.
In another study, recently published in the Journal of Neuro-Oncology, Dr. Bobholz and the MCW team demonstrated that radio-pathomic mapping could identify specific tumor characteristics that predict better responses to a targeted cancer treatment—bevacizumab—in patients with glioblastoma. They discovered that patients who had less dense tumors (identified through the mapping technique) showed better treatment outcomes and a significant reduction in tumor size. “We have found preliminary evidence to suggest that different subgroups of tumors exist depending on the appearance of the area just outside the known tumor volume that our maps are sensitive enough to distinguish. These subgroups could be used to help aid in assessing clinical prognosis and can help determine whether patients would be good candidates for receiving anti-angiogenic treatment,” said Dr. Bobholz.
Bobholz noted that the MCW Brain Bank has been critical to advancing research in the LaViolette lab, as it has a one-of-a-kind focus on tying pathological information at autopsy to non-invasive imaging, such as MRI.
“This dataset could not happen without extensive, excellent communication and collaboration between neuro-oncologists, pathologists, neurosurgeons, radiologists, radiation oncologists, and a range of research personnel from the undergraduate to faculty level. Our group has developed both the physical and human infrastructure to not only build the MCW Brain Bank as a resource but ensure that it is put to use to develop innovative, clinic-ready tools that can hopefully translate to real patient impact,” said Dr. Bobholz.
But this is just the beginning for the research team, as the LaViolette lab recently received a fundable score on a National Institutes of Health RO1 grant to support multi-site clinical validation of its radio-pathomic mapping technique in the surgical setting. The $4 million project, “Radiopathomic Modeling of Glioma Heterogeneity Throughout a Patient’s Disease Trajectory,” will be conducted in collaboration with the University of California San Francisco (UCSF), allowing both research teams to merge their models and create a hybrid model that accounts for various treatments.
“UCSF has been developing their biopsy-based radio-pathomic maps in parallel to our autopsy-based maps for several years, and we’ve been fortunate enough to develop a collaborative relationship focused on trying a range of different algorithms, experimental designs, and data sets to improve each model to the best of its ability. In addition to symmetrically working to merge the biopsy and autopsy based models, they will implement the maps at surgery at their local site to validate the combined map in the first line clinical setting,” said Dr. Bobholz.