Devoid of nutrients and with limited oxygen, the leptomeninges — the cerebrospinal fluid-filled membranes covering the brain and spinal cord — is a wasteland, making it challenging for foreign cells, such as cancer cells, to survive there. Despite this harsh environment, cancer cells manage to reach and establish themselves there in five to ten percent of patients with solid tumors. Patients with leptomeningeal metastasis currently have a life expectancy of less than six months. 

In a new mouse study published in Science, researchers discovered that breast cancer cells can reach the leptomeninges via the bone marrow (1). Their model suggests that cancer cells crawl along the blood vessels that connect the bone marrow to the leptomeninges, without entering the bloodstream. Once there, the cancer cells trick macrophages in the leptomeninges into secreting a molecule that helps them thrive in this tough territory. If further research shows that cancer cells also spread this way in humans, targeting that macrophage-secreted molecule could offer hope to patients with leptomeningeal metastasis.

Metastasis can sometimes originate from another metastatic site, not just the primary tumor, said Jan Remsik, a cancer immunologist at the VIB-KU Leuven Center for Cancer Biology who did not participate in this work. Yet, studies often model metastasis using a linear approach, they added. “So, what is really amazing about this paper is that [the team] found a way to model this unusual trajectory.”  

Study coauthor Dorothy Sipkins, who is a physician scientist and cancer researcher at Duke University, and her team previously reported that an aggressive form of leukemia traveled through the bone marrow to reach the central nervous system (2). Since later studies showed that benign immune cells also migrate through this corridor under certain conditions, she wondered if other cells — including those shedding from solid tumors — could use it to reach the brain (3).

Breast cancer cells metastasizing to the leptomeninges were a potential candidate. For one, the bone is a common site for breast cancer metastatic spread. Moreover, results from a 1983 autopsy study already suggested that the vertebral bone and its surroundings could be an intermediate site in the pathway from the breast to the brain (4). “That was one of the things that really stuck out in my mind,” said Sipkins. 

Sipkins’ team investigated whether breast cancer cells migrate to the leptomeninges via the bone primarily in two mouse models of breast cancer, in which the tumor spreads to the bone marrow and to the leptomeninges. Results from several experiments and tissue analyses confirmed Sipkins’ suspicions. 

“One of the really cool caught-you-in-action experiments [was] the optical clearing of the intact skull after we removed it from the brain,” explained Sipkins. In that setup, the team engrafted the breast tumor in the mice and sacrificed them three days later — the timepoint at which tumor cells first arrive at the leptomeninges. Before sacrificing the mice, the researchers injected fluorescent reporters that stained the bone marrow, the blood vessels, and the breast cancer cells. When they visualized the tissues, they saw some cancer cells anchored along the outer surface of the vessels connecting the bone marrow and the leptomeninges.  

In their earlier work on leptomeningeal metastasis in leukemia, Sipkins’ team found that cancer cells navigate this corridor by expressing integrin a6, a cell adhesion protein that binds to the extracellular matrix molecule laminin on vessels. Neural progenitor cells use a similar mechanism, expressing integrin a6 to find their way on laminin-rich surfaces during brain development (5). In this study, the researchers found that breast cancer cells exploit this same tactic to reach their target. Mice engrafted with breast cancer cells lacking integrin a6 showed substantially reduced tumor spread to the leptomeninges and lived longer. 

Neurons are so obviously beautifully adapted to the microenvironment in the central nervous system, and so, if you want to make your way in this totally new microenvironment, why don’t you learn from cells, perhaps co-opt pathways that successful cells have to survive in that environment. 
- Dorothy Sipkins, Duke University

While integrin a6 helps cancer cells reach the leptomeninges, it does not ensure survival once there. Sipkins and her colleagues compared gene expression of two integrin a6-expressing cell lines — the first from one of the mouse models, shown to metastasize to the leptomeninges, and another one that failed to proliferate into this site. They discovered that successful cells expressed higher levels of neural cell adhesion molecule (NCAM). In stressful scenarios, neurons express NCAM to induce macrophages to secrete the prosurvival neurotrophin glial-derived neurotrophic factor (GDNF) (6). GDNF helps neurons resist apoptosis in oxygen- and nutrient- deprivation situations, such as during stroke.

Sipkins and her colleagues found that macrophages near the cancer cells in the leptomeninges were indeed secreting GDNF. Furthermore, when the team blocked GDNF with antibodies, mice took longer to develop leptomeningeal disease and survived longer. 

Once again, cancer cells seemed to adopt a neural strategy to thrive in this environment. “When in Rome…” said Sipkins. “Neurons are so obviously beautifully adapted to the microenvironment in the central nervous system, and so, if you want to make your way in this totally new microenvironment, why don’t you learn from cells, perhaps co-opt pathways that successful cells have to survive in that environment.”

Analyses of human biopsies from patients with breast cancer revealed that leptomeningeal metastatic tumors expressed significantly higher levels of integrin a6 compared to primary tumors, bone marrow metastases, and tissues from another form of brain metastasis.

However, Remsik said that a trial looking at these biomarkers is needed to understand the implications for humans. For instance, if integrin a6 proves important for the development of leptomeningeal metastasis, researchers and clinicians could potentially find ways to inhibit it. Overall, Remsik emphasized the need for more human data on these pathways “because we need to know what [cancer types] to focus on if this is going to be a feasible clinical approach.”

Sipkins is hopeful that her team’s research will lead to new ways to treat cancer that has spread to the brain. “Treatment for leptomeningeal metastasis has not really changed in many years,” she noted. “Identifying more new targets for specific therapies ... could improve our care of these patients.”

References

1. Whiteley, A.E. et al. Breast cancer exploits neural signaling pathways for bone-to-meninges metastasis. Science  384, eadh5548 (2024).
2. Yao, H. et al. Leukaemia hijacks a neural mechanism to invade the central nervous system. Nature  560, 55-60 (2018).
3. Herisson, F. et al. Direct vascular channels connect skull bone marrow and the brain surface enabling myeloid cell migration. Nat Neurosci  21, 1209-1217 (2018).
4. Kokkoris, A.E. Leptomeningeal carcinomatosis: How does cancer reach the pia-arachnoid? Cancer  51, 154-60 (1983).
5. Flanagan, L.A. et al. Regulation of human neural precursor cells by laminin and integrins. J Neurosci Res  83, 845-56 (2006).
6. Sariola, H. and Saarma, M. Novel functions and signalling pathways for GDNF. J Cell Sci  116, 3855-62 (2003).