Advancing Precision Medicine for Neuro-Oncology Patients: Q&A with David Solomon, MD, PhD

UC San Francisco neuropathologist and molecular neuro-oncologist David Solomon, MD, PhD has reached a unique and impressive milestone: the publication of his team's 50th neuro-oncology research study stemming from the UCSF500 Cancer Gene Panel, a molecular diagnostic test that identifies genetic changes in the DNA of a patient's cancer. Below, Solomon reflects on the scientific impact of these 50 studies and their implications for present and future neuro-oncology treatment. 

Congratulations on the publication of your 50th neuro-oncology research study stemming from the UCSF500 Cancer Gene Panel! What is the focus of this recent paper?

Our latest study identifies that prospective genomically guided identification of "early/evolving glioblastoma" not only improves diagnostic classification for affected patients, but critically leads to improved clinical outcomes by enabling more appropriate therapy compared to microscopic impression alone. These “early/evolving glioblastoma” are a group of diffuse gliomas occurring in adults that would have been previously diagnosed as WHO grade 2 or 3 based on their lower-grade microscopic features, but the recognition that they have molecular features of glioblastoma and lack the IDH gene mutation that defines lower-grade gliomas with more favorable prognosis has led to their reclassification as “Glioblastoma, IDH-wildtype, CNS WHO grade 4” in the latest 2021 World Health Organization (WHO) Classification.

We found that prospective treatment of these patients based on a molecularly integrated diagnosis of glioblastoma resulted in a median survival of approximately 24 months, whereas a biologically matched control cohort of patients where genomic profiling testing was performed on a retrospective research basis only with clinical management that was predicated entirely upon the lower-grade microscopic diagnosis had a median survival of approximately 16 months.

What do you see as the major scientific accomplishments among these 50 published studies?

We have identified and further defined multiple new molecularly distinct brain tumor types including “Myxoid glioneuronal tumor, PDGFRA p.K385-mutant,” “Intracranial mesenchymal tumor, FET::CREB-fused,” “Diffuse midline glioma, EGFR-mutant,” and “CNS tumor with BCOR internal tandem duplication.” We have identified the responsible genetic driver for existing brain tumor types previously defined based on microscopic criteria, including PRKCA p.D463H mutation in all chordoid gliomas, MAP2K1 exon 2 mutations in multinodular and vacuolating neuronal tumor, and diverse oncogenic alterations in the MAP kinase pathway in gangliogliomas.

We have revealed that a surprising number of brain tumors are not sporadic, but rather arise as part of several different inherited tumor predisposition syndromes (e.g. Lynch syndrome, Li-Fraumeni syndrome, neurofibromatosis type 1 and 2). We have defined the molecular landscape and clinicopathologic features of multiple different brain tumor predisposition syndromes and identified that syndrome-associated brain tumors are often not equivalent to their sporadic counterparts. We identified that gliomas arising years later after radiation therapy for a childhood malignancy (so-called “radiation-induced gliomas”) are molecularly distinct and have a unique genomic pattern compared to their spontaneous counterparts.

We have revealed that certain genetic alterations (e.g. BRAF mutation/fusion, FGFR1 mutation/rearrangement, TP53 mutation) are actually promiscuous and can be seen across multiple different brain tumor types with divergent outcomes, and that the most accurate diagnostic classification for many brain tumors is actually DNA methylation-based rather than genetic mutation-based. We have identified that microscopically identical gliomas occurring in infants vs. school-aged children vs. teenagers vs. young adults vs. older adults are genetically and prognostically different, setting the stage for precision diagnosis based on molecular profile rather than microscopic features.

How has this body of work impacted patient care at UCSF and beyond?

These advances in our understanding of the molecular pathogenesis of CNS tumors have refined the way we classify tumors beyond just how they visually appear under the microscope (i.e. conventional histopathology). In other words, we are now able to more accurately classify and treat brain tumors when histopathology is combined with genomic and epigenomic profiling, which have now become standard-of-care for the diagnosis and management of neuro-oncology patients. All patients who undergo resection of a brain tumor at UCSF Medical Center now undergo UCSF500 genomic profiling on a prospective clinical basis to enable accurate diagnostic classification and identify any potential actionable genetic alterations that may serve as therapeutic vulnerabilities.

This NGS platform that we have established is also currently being used to guide genomically tailored targeted therapy combinations in two ongoing precision medicine clinical trials: one in adults with recurrent glioblastoma led by Dr. Jennifer Clarke as part of the UCSF Glioblastoma Precision Medicine Program (GPMP), and the other in children with newly diagnosed high-grade gliomas led by Dr. Sabine Mueller as part of the Pacific Pediatric Neuro-Oncology Consortium (PNOC).

Your research has identified new brain tumor types that were incorporated into the latest 2021 World Health Organization (WHO) Classification of Central Nervous System Tumors. What is the significance of this inclusion?

The foundation of precision medicine for cancer starts with an accurate pathologic diagnosis for each patient. Recognizing and accurately classifying patients’ tumors into each of these new distinct brain tumor entities means that affected patients will now be given the correct diagnosis and prognosis, enabling them to be treated based on the most precise tumor classification. Separating out these newly recognized entities also means that the existing tumor entities in the WHO Classification are becoming more and more homogenous, rather than a hodgepodge of neoplasms with divergent biology and clinical outcomes. The end result is that clinical trials will be able to better focus on targeting and improving outcomes for individual homogenous tumor entities, recognizing that therapies which work for one tumor type may likely be ineffective for another. The hope and expectation is that improved tumor classification will result in improved patient outcomes.

Looking back to your early work involving brain tumor profiling, what drew you to this field of research? How has your work evolved over time?

Starting early in my education, I became fascinated by human disease that results from mutations in our genetic code. It is truly amazing to me that a single mutation in our genetic blueprint consisting of approximately 3 billion nucleotides can result in muscular dystrophy, multiple cancers, or neurodegenerative disease, dependent entirely on which nucleotide in the genome is affected. I decided to focus my career to the study and treatment of cancer based on my fascination with cancer’s genetic underpinnings, and I have devoted a substantial portion of the first decade of my career here at UCSF working to advance precision medicine for neuro-oncology patients.

What's on the horizon for genomic profiling as it relates to brain cancer treatment? What problem do you hope to have solved in the next decade?

Brain tumor research is such an exciting field to be working in right now. Now that genomic profiling has entered routine standard-of-care for our neuro-oncology patients under UCSF Health Genomics Services, I have since refocused my efforts on working together with our neuro-oncology research team to develop the next generation of technologies that will lead to breakthroughs in glioma research and treatment, which include DNA methylation profiling, single cell sequencing, digital spatial profiling, proteogenomic and metabolomic profiling, and liquid biopsy.

These technologies are further advancing our understanding of the molecular pathogenesis of brain tumors, deciphering their cellular composition at single cell resolution, and unraveling the tumor microenvironment including how the immune system interfaces with tumor cells. Additionally, my research lab has been focusing our efforts on developing cell culture and mouse models of the many different brain tumor types to provide a deeper mechanistic understanding of brain tumor biology and enable the development of new effective targeted therapies.

Are there any particular collaborators or partners you'd like to acknowledge?

This body of work over the past seven years has been an amazing collaborative team effort involving numerous faculty, trainees, and staff in Neurosurgery, Neuro-Oncology, Neuropathology, Neuroradiology, Radiation Oncology, and the Clinical Cancer Genomics Laboratory. These studies have been made possible by generous support from the Departments of Pathology and Neurosurgery, the UCSF Glioblastoma Precision Medicine Program sponsored by the Sandler Foundation, the Panattoni Family, the UCSF Brain Tumor SPORE grant, the UCSF Brain Tumor Center's T32 training grant, the Morgan Adams Foundation, the Yuvaan Tiwari Foundation, the NIH Director's Early Independence Award, UCSF School of Medicine Strategic Funds, the UCSF Physician-Scientist Scholar Program, and several other sources.