Brain Tumor Evolution & Therapy Laboratory

The Laboratory of Brain Tumor Evolution & Therapy, under the direction of Baoli Hu, PhD, is interested in the genetic and epigenetic events contributing to the evolution of brain tumors. The long-term goal of the lab is to achieve a better understanding of brain tumor biology and to develop more effective diagnoses and therapeutic strategies for the treatment of brain cancer.

Cancer is increasingly being viewed as an ecosystem where the cancer cells dynamically evolve and spatiotemporally communicate with surrounding cells and environmental factors. Deciphering this evolutionary complexity allows us to better understand brain tumor initiation, progression, recurrence, and drug resistance. The Brain Tumor Evolution & Therapy Lab is focusing on glioma and medulloblastoma, the most common malignant brain tumors in adults and children, respectively. Specific projects are as follows

1. Modeling the evolution and diversity of brain tumors using human-in-mouse system. Intratumor genetic heterogeneity and phenotypic diversity are the hallmarks of glioma and medulloblastoma, which predict the risk of tumor development, progression and response to treatment. To delineate crosstalk mechanisms of these factors, we have been developing human-in-mouse model systems based on malignant transformation of human neural/cerebellar stem cells driven by subtype-specific genetic/epigenetic alterations. These models can faithfully recapitulate the molecular diversity, cellular heterogeneity, and histology seen in patient tumors. In addition, these models enable precise system-level comparisons of premalignant and malignant states of these stem cells, which deepens our understanding of tumor evolutionary dynamics in the molecular and cellular level. The key regulators in this process are validated as diagnostic biomarkers and therapeutic targets for clinical application.

2. Interrogating consequences of stem cells plasticity within brain tumor microenvironment. Emerging evidence suggests that glioma/medulloblastoma stem cells may contribute to tumor evolution and anti-therapy. We previously found that glioblastoma stem cells (GSCs) differentiate into endothelial-like cells (GdECs), which recruit host endothelial cells (ECs) to form an invasive niche, resulting in tumor invasiveness and recurrence. We are continuing our efforts to gain a better understanding of the molecular mechanisms of these cancer stem cells, and how they communicate with their surrounding cells (e.g. endothelial cells, microglia/macrophages, astrocytes, etc.), which allows us to develop novel and more effective therapies by targeting critical components of the tumor microenvironment.

The invasive niche is comprised of GSCs, GdECs, ECs, and other types of cells in GBM patient tumor.  Representative images with IHC double-staining and cell segmentation obtained from Caliper InForm analysis software show the close proximity of GdEC (yellow) and host ECs (green) compared with GSCs (red). Cell-cell distance was assessed by an automated quantitative pathology imaging system. Scale bar, 20 mm. (Hu, et al., Cell 167(5),1281-1295, 2016).

3. Illuminating mechanisms governing cancer cells invasion and dissemination in brain. The major challenge in the clinical management of glioblastoma is that cancer cells extensively infiltrate into the surrounding tissue, leading to nearly universal recurrence. Group 3 medulloblastoma is characterized by frequent metastasis at diagnosis and the worst prognosis among all the subgroups. We aim to elucidate molecular mechanisms of de novo invasion and treatment-induced invasion (e.g. TMZ, bevacizumab, etc.), which enables us to identify “drivers” mediating cancer cells invasion and to dissemination and to aid in the development of new therapies.

lab scans

Suppression of tumor invasiveness by depleting WNT5A-mediated GdECs using HSVTK/ganciclovir (GCV) cell ablation system.  Representative images for tumor appearance (left, scale bar, 2,000 μm) and peritumoral satellite lesions (right, scale bar, 200 μm) in PDX mouse model. Tumor cells are labeled by staining human specific antigen TRA-1-85 (red). (Hu, et al., Cell 167(5),1281-1295, 2016).

The Hu Lab is looking for talented and highly-motivated postdocs, undergraduate and graduate students who are interested in our research and joining us. Interested candidates should directly contact Dr. Hu at