In 1835, an anatomist at the University of Paris named Jean Cruveilhier recorded the earliest known observation of a tumor that grew along the winding curvature of a nerve. Specifically, he described a mammary tumor that migrated and spread along the facial nerve (1). Over the next several decades, anatomists and pathologists noted the tendency of tumors to grow close to nerves all over the body. Yet, the significance of these discoveries remained largely ignored for more than 150 years. 

Around the turn of the 21^st century, neuroscientists and cancer researchers began to appreciate that finding nerves so close to tumors might not be only a coincidence. Instead, they realized that the nerves could somehow feed the tumor’s progression. In a landmark study in 2013, late stem cell researcher Paul Frenette of the Albert Einstein College of Medicine showed that autonomic nerve fibers were crucial in regulating the growth of prostate tumors in mice (2). Soon after, in 2014, Timothy Wang, a gastric cancer researcher at Columbia University, and his colleagues showed that severing the connection between the vagus nerve and the stomach stopped cancer growth in mice (3). 

“Nerves are the most comprehensive in their ability to communicate with other cells in the body. Of course, that's how the body is designed to have nerves integrating everything,” said Wang. “It turns out that they integrate a lot of cancer biology in interesting ways.”

The role of the nervous system in cancer doesn’t end with nerves. In 2019, independent teams led by Frank Winkler, a neurologist at University Hospital Heidelberg, and Michelle Monje, a neuro-oncologist at Stanford University, showed that tumor cells in the brain formed synapses with neighboring neurons and integrated themselves into neural circuits. The resulting electrical activity drove tumor progression (4,5).

These exciting discoveries led to the birth of a new research field: cancer neuroscience. Nearly two hundred years after Cruveilhier’s first description of tumors growing along nerves, scientists across the globe now investigate the myriad roles of the nervous system in promoting or preventing cancer. Winkler has a hypothesis for why it took so long for the field of cancer neuroscience to emerge: “The idea that something so evil and bad and malignant and awful like cancer is also governed by our most precious thing, the nervous system — maybe this is just something people didn't want to fully appreciate.”

Today, as the profound links between cancer and the nervous system continue to be revealed, many scientists are developing new therapeutics that take advantage of the crosstalk between the brain, nerves, and tumors.

Seeking safety in nerves 

With the connection between nerves and tumors finally cemented, research shifted to investigating why and how nerves are involved in cancer. Jami Saloman, a neuroscientist at the University of Pittsburgh, and her colleagues recently proposed the “safe harbor hypothesis” to explain how a tumor seeks refuge with a nearby nerve (6). They suggested that tumor cells take advantage of a basic evolutionary function in which nerves modulate the immune system’s response to pathogens and ensure that it doesn’t attack its own cells. 

“It's really like the tumor cells hijack this evolutionary protection and turn it into something bad,” said Saloman. “Tumor cells might be more protected from an immune response if they're near these nerves that are shutting down the immune system. It would be like a hideout place where they could keep away from the immune response, or at least, a less severe immune response.” This is particularly true when tumors invade the nerves themselves. Inside of nerves, they gain access to barriers that keep immune cells out, allowing cancer cells to seek shelter from the cells that would usually identify and attack them. Within the safety of nerves, tumor cells also reduce their exposure to cancer drugs like chemotherapy. 

A growing tumor often needs more than just protection, though. “The tumor is, as we see it today, like an organ,” said Winkler. “We know that to form new organs, you need innervation.”

To achieve innervation, tumors must convince nearby nerves that they’re brand-new organs, even if they develop many decades after all of the body’s organs have finished developing. When organs first develop, they make use of signaling molecules that guide them to create the correct tissue patterns and recruit nerves to grow into them. Paola Vermeer, a cancer biologist at Sanford Research, showed in 2018 that tumors can reuse these early developmental signals. Her team revealed that tumor cells release vesicles with an axonal guidance molecule, EphrinB1, that compels nearby sensory nerves to sprout towards the tumor and innervate it just as they would with a developing organ (7). 

“It's a very sophisticated way for the tumor to really have nerves hone in on it, and then they utilize the nerves to induce this immune-suppressive environment and let the tumor grow,” said Vermeer. They also showed that using a pharmacological agent to block the release of the vesicles limited tumor innervation. This suggests a potentially promising new treatment avenue, given that patients with more densely innervated tumors have been found to have higher rates of recurrence and metastasis (8).

It's really like the tumor cells hijack this evolutionary protection and turn it into something bad. 
– Jami Saloman, University of Pittsburgh

Saloman’s team studies how sensory nerves help create the immune-suppressive environment surrounding tumors. So far, they’ve found that the PD-L1 protein, which helps inactivate immune cells that kill cancer, is expressed not only in immune cells but also in sensory neurons that innervate the pancreas (9). “That's a potential mechanism for how [sensory neurons] could talk to the immune cells and shut them down and prevent them from killing the tumor,” said Saloman.

Now, Saloman and Vermeer aim to identify drugs that target nerve involvement and shut down tumor growth. But Saloman cautioned that treatments that block or alter nerve activity should be designed to only act near the tumor to avoid systemic effects all over the body. 

The treatments will also need to differ based on the specific nerves involved. For example, Wang’s team found that while the vagal nerve promoted tumor growth in gastric cancer, it hindered tumor growth in pancreatic cancer in mice. Giving mice the drug bethanechol, which activates acetylcholine receptors, mimicking vagal nerve activity,  decreased tumorigenesis and extended the lifespans of the mice (10). Based on these results, Wang and his colleagues began a Phase 1 clinical trial assessing the effects of bethanechol in pancreatic cancer patients before surgery.

Becoming the brain

Tumors that grow within the jelly-like substance of the brain may be the most obvious place to investigate how the nervous system plays a role in cancer. Still, cancer neuroscientists like Winkler have been surprised at the magnitude of the relationship between the brain and its infiltrating tumor, and how seamlessly the two can merge into one. 

Twenty years ago, when Winkler was a postdoctoral fellow in Rakesh Jain’s lab at Harvard University, his team observed how incurable glioblastoma tumors grew within mice in real time over weeks and months. They watched as the tumor cells extended long projections outwards from their cell bodies — just as new neurons do during development.

This set Winkler on a path to investigate the complex relationship between the brain and brain tumors in his own lab. He discovered that brain tumors form synapses with neurons and become functioning members of the brain circuitry. “When we discovered this the first time, I talked to Michelle Monje at Stanford and I said, ‘Michelle, we discovered something crazy.’ She said, ‘Well now Frank, we discovered something crazy.’ It took us 10 minutes to find we had discovered the same crazy thing,” said Winkler. Their teams published back to back papers on these findings in 2019 (4,5). 

Since then, emerging evidence has suggested that synapses form not only between neurons and tumor cells in the brain, but also between peripheral neurons and cancer cells elsewhere in the body (11). “A lot of cancer cells seem to hijack this mechanism also. So, the nervous system is willing to support growth, invasion, metastasis, and resilience of these metastatic cancer cells by forming these synapses to cancer,” said Winkler.

On the positive side, the brain’s physical connections with tumors offer unique targets for pharmacological drugs. Winkler collaborates with other researchers in Germany on two ongoing clinical trials that repurpose existing drugs — meclofenamic acid, an anti-inflammatory, and perampanel, an anti-seizure medication — to treat brain cancers by disconnecting tumor cells from neurons.

Another pharmacological option targets the ion channels on the cell membranes, which regulate synaptic functions between the neurons and cancer cells. In a recent paper, Xi Huang, a biologist at The Hospital for Sick Children, and his colleagues engineered a designer peptide that blocks the interaction of two proteins that form a potassium channel complex on the cell’s surface (12). This complex lets potassium ions flow out of the cell and thereby triggers incoming signals from nearby neurons that promote the tumor’s growth and resistance to chemotherapy. Huang’s team found that disrupting the protein-protein interaction in mice with glioblastoma tumors prevented the potassium channel complex formation, inhibiting tumor growth and extending the lives of the mice.

Huang and his colleagues are now working towards moving this work into a clinical trial and possibly setting up a startup biotech company based on the new peptide they engineered. “It’s rare, as a basic science lab, that we are able to have an opportunity to really have something concrete that hopefully could help patients,” said Huang. “That's most exciting to me.”

A new link to mental health

The direct connections between cancer cells, nerves, and neurons may also change the activity of brain circuits, even affecting mental health. Vermeer became interested in why epidemiological studies consistently reveal increased rates of depression in cancer patients, even ten years after they have beaten cancer (13). She wondered if it was possible for a tumor outside of the brain to connect through nerves to a neuronal circuit in the brain, and if so, whether that connection might change the circuit’s activity in a way that ultimately influences mental health.

In a recent preprint, Vermeer’s team reported studying mice with head and neck tumors that have recruited nearby nerves to innervate them (14). They found that the nerves within the tumor connected to the trigeminal ganglion, which sends sensory information from the face and jaw and connects to circuitry in the brainstem and cortex. Vermeer’s team showed that nerve endings embedded in the tumor functionally changed activity in the circuit, such that neurons in the brain exhibited higher sensitivity to pain stimuli and elevated electrical activity. These changes likely resulted from the nerve endings in the tumor being exposed to an acidic environment deprived of oxygen, which is typical of tumor microenvironments. “These nerves are exposed to things that they're not normally exposed to, and that changes them. And this change goes all the way from the tumor to the ganglion and into the brain,” said Vermeer. 

Ultimately, Vermeer’s team found that these changes in brain activity led to behavioral indications of depression and anxiety in the mice. When they studied mice with tumors that did not have nerves innervating them, the researchers did not observe the same behaviors, suggesting that the depressive and anxiety-like behaviors resulted from the nerves in the tumor bed changing the activity of the circuits in the brain.

“Now we can start to understand that cancer patients are not just depressed because they have the knowledge that they have cancer; they're depressed because there's a biological process happening here,” said Vermeer. “We're starting to shed light on this previously black box about the connection between having a cancer and your mental health.”

Vermeer now plans to test whether existing FDA-approved drugs for epilepsy or Parkinson’s disease could be used to treat mental health symptoms by dampening the electrical activity between the tumor, nerve, and the circuit in the brain. “We want to see if we essentially short-circuit this connection to the brain, can we restore normal activity so that after patients undergo their treatments, they can really have a normal mental health status and really enjoy, and really live as survivors, and not just exist as survivors,” said Vermeer. “We need to do better for our patients.”

Using the brain to detect cancer in the body

Beyond the direct effects that the nervous system and cancer exert on each other, complex body-wide cancer interactions also influence the brain. These lead to changes in neural activity that can cause dysregulation between the brain and bodily processes that might cause well-known cancer symptoms. Evidence of this phenomenon comes from Jeremy Borniger, a neuroscientist at Cold Spring Harbor Laboratory. While attending talks by doctors and nurses earlier in his career, Borniger learned that nearly all patients with cancer experience chronic fatigue and sleep disruptions. “Everyone seemed to assume that, well, of course, if you have cancer, you're not going to sleep well because you're stressed about having cancer or taking these toxic chemotherapy drugs or [getting] radiotherapy,” said Borniger. “But we wanted to see, can cancer by itself influence sleep and arousal?”

Borniger set out to answer this question by studying a mouse model of non-metastatic breast cancer and monitoring sleep patterns as tumors grew. “I was surprised that no one had done that experiment,” he said.

His team noticed that over time, sleep became more fragmented, and the metabolic hormones also changed (15). That directed them to look at the hypocretin/orexin neurons in the hypothalamus that regulate sleep and arousal. They found hyperactivity in these neurons. When they gave the mice a drug to block the activity of hypocretin/orexin neurons, the sleep problems improved, and metabolic functions resumed. 

“That was really interesting to me, because it was, at least on the surface, initial evidence that a tumor in the body that is nowhere near the brain can influence processes that are controlled by the brain,” said Borniger.

Now, Borniger studies how the brain detects cancer-related changes in the body, and whether the crosstalk between the brain and body sets up feedback loops that cause abnormal symptoms or prevent treatments from working. “A lot of the neurons that are sensitive to these changes in the body, they don't just sense the changes; they elicit changes in the body too. So, that is a really exciting area that I think is going to explode in the next couple of years,” said Borniger.

Deciphering how the brain detects cancer in the body could also lead to new diagnostic tests to catch cancer earlier. Borniger currently collaborates with researchers at Weill Cornell Medicine to train machine learning algorithms to recognize signatures of cancer in brain waves. With research like this, patients may one day be able to get a noninvasive EEG recording in their doctor’s office and find out if they have cancer before they notice any symptoms. The technology could also be used to decode the brain to track cancer’s progression. Recently, researchers led by Manuel Valiente of the Spanish National Cancer Research Center successfully used a machine learning algorithm to predict different types of cancer metastasis in the brains of mice based on electrical brain activity alone (16). 

“Using the brain to read what's going on in the body is in its infancy, but it's not impossible. It's something that I think we just need more data to train models,” said Borniger.

Combining cancer, neuroscience, and more

As the field of cancer neuroscience takes flight, many researchers are enthusiastic about uncovering new tactics to treat cancer. At the same time, they emphasize that the goal should be to combine neuro-focused therapeutics with other therapies for a multipronged approach. “In and of itself, it’s not a silver bullet,” said Vermeer.

“We hope maybe by really targeting these crucial neuro-cancer interactions or neuro-cancer networks, we might be able then to really find crucial vulnerabilities of cancers, maybe not as a monotherapy, but at least to make the existing therapies more efficient,” said Winkler. For example, in a 2020 Phase 2 clinical trial, researchers at the Peter MacCallum Cancer Centre gave patients a beta blocker, which blocks the activity of two neurotransmitters, noradrenaline and adrenaline, for a week before their breast cancer surgeries (17). They found that the pretreatment decreased the sympathetic nervous system activity triggered by the stress of impending surgery; as a result, they saw reduced gene expression and immune cell markers associated with cancer recurrence (17). 

 “We're in a similar space as cancer immunotherapy was in the late ‘80s when it was clear that it was important, but it wasn't clear exactly how it worked and how we could take advantage of it. But I think for the next 10 years that it'll be the same story for cancer neuroscience,” said Borniger.

Challenges remain for this nascent field as researchers continue to develop and refine the interdisciplinary tools and techniques they need to advance. But with more researchers joining the field each year, Winkler is certain that there will be no shortage of interesting discoveries. “The deeper we dig, the more crazy stuff we find,” he said. “We need to prepare for getting some really crazy insights in the next 10 years.” 

References

1. Cruveilhier, J. Anatomie pathologique du corps humain, ou descriptions, avec figures lithographiées et coloriées, des diverses altérations morbides dont le corps humain est susceptible. (Chez J.B. Baillière, 1835).
2. Magnon, C. et al. Autonomic nerve development contributes to prostate cancer progression. Science  341, 1236361 (2013).
3. Zhao, C.-M. et al. Denervation suppresses gastric tumorigenesis. Sci Transl Med  6, 250ra115 (2014).
4. Venkataramani, V. et al. Glutamatergic synaptic input to glioma cells drives brain tumour progression. Nature  573, 532–538 (2019).
5. Venkatesh, H. S. et al. Electrical and synaptic integration of glioma into neural circuits. Nature  573, 539–545 (2019).
6. Saloman, J. L., Scheff, N. N. & Davis, B. M. in Cancer Neuroscience (eds. Amit, M. & Scheff, N. N.) 185–200 (Springer International Publishing, 2023).
7. Madeo, M. et al. Cancer exosomes induce tumor innervation. Nat Commun  9, 4284 (2018).
8. Reavis, H. D., Chen, H. I. & Drapkin, R. Tumor Innervation: Cancer Has Some Nerve. Trends Cancer  6, 1059–1067 (2020).
9. Meerschaert, K. A. et al. Neuronally expressed PDL1, not PD1, suppresses acute nociception. Brain Behav Immun  106, 233–246 (2022).
10. Renz, B. W. et al. Cholinergic Signaling via Muscarinic Receptors Directly and Indirectly Suppresses Pancreatic Tumorigenesis and Cancer Stemness. Cancer Discov  8, 1458–1473 (2018).
11. Schmitt, A. et al. Functional synapses between small cell lung cancer and glutamatergic neurons. Preprint at https://doi.org/10.1101/2023.01.19.524045
12. Dong, W. et al. A designer peptide against the EAG2-Kvβ2 potassium channel targets the interaction of cancer cells and neurons to treat glioblastoma. Nat Cancer  4, 1418–1436 (2023).
13. Götze, H. et al. Depression and anxiety in long-term survivors 5 and 10 years after cancer diagnosis. Support Care Cancer  28, 211–220 (2020).
14. Barr, J. et al. Tumor-infiltrating nerves functionally alter brain circuits and modulate behavior in a male mouse model of head-and-neck cancer. Preprint at https://doi.org/10.1101/2023.10.18.562990
15. Borniger, J. C. et al. A Role for Hypocretin/Orexin in Metabolic and Sleep Abnormalities in a Mouse Model of Non-metastatic Breast Cancer. Cell Metab  28, 118-129.e5 (2018).
16. Sanchez-Aguilera, A. et al. Machine learning identifies experimental brain metastasis subtypes based on their influence on neural circuits. Cancer Cell  41, 1637-1649.e11 (2023).
17. Hiller, J. G. et al. Preoperative β-Blockade with Propranolol Reduces Biomarkers of Metastasis in Breast Cancer: A Phase II Randomized Trial. Clin Cancer Res  26, 1803–1811 (2020).