How Sticky Are the Cancer Cells?

We tend to think the toughest cancer cells are the deadliest, but new research suggests that the weakest, less stickiest ones are the real threat. In a landmark 2025 study, Kane et al. from the University of California San Diego, USA, have identified a simple yet powerful predictor of a tumour’s metastatic potential: adhesion strength. Apparently, the weakest clingers – the cancer cells that barely stick to their surroundings – are the most likely to detach and spread to other areas. Using a cutting-edge device, the team also quantified this cancer cell ‘stickiness’ to predict high-risk tumours with remarkable accuracy. This discovery could reshape how we stratify cancer risk and even inspire new strategies of intercepting metastasis.

Even with all the advances in cancer detection, we still struggle to predict which tumours will spread (metastasise). While many cancers are diagnosed and treated while still localised to their original site, the real danger often emerges later, when a small number of cancer cells break away and form new tumours in other organs. In breast cancer, for example, the bone is the most common site of metastasis, followed by the lungs, liver and brain (Figure 1). However, only a tiny fraction of tumour cells have this ability to metastasise, and one of the biggest challenges in oncology is our current inability to reliably identify these high-risk cells early.

Figure 1. The multistep process of cancer metastasis using breast cancer as an example. (A) Tumour cells escape from the primary site and invade nearby tissue. (B) These cells enter and survive within the bloodstream or lymphatic system, a step known as intravasation. (C) Tumour cells exit the circulation (extravasation) and establish secondary tumours in distant organs. Although breast cancer commonly spreads to the bones, lungs and liver, the brain is shown here to represent the most lethal site of metastasis in breast cancer patients. Source: Riggio et al. (2020), British Journal of Cancer.

Alarmingly, metastatic tumours often appear long after the initial tumour has been removed, catching both patients and doctors off guard. This delayed relapse and spread is called metastatic recurrence, driven by disseminated tumour cells (DTCs) that escape the initial tumour early and lie dormant in distant tissues for years and even decades. During this dormant period, there are often no detectable signs of cancer. Then, without warning, these dormant cells can reawaken and seed metastatic growth. As our current tools struggle to detect minimal residual cancer, it is very difficult to distinguish patients who are truly cancer-free from those still harbouring high-risk cells. As a result, many patients are overtreated with harsh therapies just in case, while others are undertreated and blindsided by metastatic recurrence. Today, metastatic recurrence continues to drive over 90% of cancer-related deaths.

Researchers have poured enormous effort into finding molecular clues (e.g., gene, protein and cell surface markers) that might flag a tumour’s likelihood to metastasise. One major effort has focused on liquid biopsies, which aim to catch circulating tumour cells (CTCs) in the blood. While promising, such approaches still face some challenges. In particular, CTCs are detected only after some tumour cells have detached and entered the bloodstream, by which point it may have been too late for meaningful early intervention. Tumours also shed a minimal number of CTCs, making it challenging to detect them accurately in small blood samples.

Notwithstanding the potential of liquid biopsies, Kane et al. proposed a different approach: looking at how the tumour behaves. Metastasis is a physical process, where cells need to loosen their grip, move and survive in new environments. Hence, efforts have been dedicated to characterising the mechanical phenotype (‘mechanotype’) of cancer cells, such as their adhesiveness or stickiness to their surroundings, which could offer a novel, more universal way to predict which tumours are likely to spread.

In their breakthrough 2025 study, Kane et al. shifted the focus from molecular markers to a physical trait of tumour cells: adhesion strength. Using a custom-built microfluidic device (i.e., essentially a tiny water tunnel for cells), they measured how tightly cancer cells stick to surfaces coated with tissue-like proteins. Specifically, cancer cells are placed on the tunnel’s surface, and the fluid flow is gradually increased to mimic the physical forces cells would face inside the body (Figure 2). Cells that are easily swept away are labelled weakly adherent, while those that stay put under higher flow are considered strongly adherent or sticky.

Figure 2. The divergent parallel plate flow chamber used to measure cell stickiness. (A) A digital design. (B) A top-down photograph of the actual working device. As fluid flows from one end to the other, it applies increasing physical stress to see how easily the cells detach from the surface. This setup allows researchers to pinpoint how sticky each cell is. Source: Kane et al. (2025), Cell Reports.

So, what did Kane et al. discover? In short, the less sticky a cancer cell is, the more likely it is to detach, invade nearby tissues and seed metastases. To uncover this, the team carried out a series of innovative experiments, each shedding light on how cancer cell stickiness influences the risk of metastasis.

First, they asked whether tumour cells that break away and spread are physically different from those that remain in the primary tumour. For this purpose, Kane et al. implanted human breast cancer cells into mice to create an animal model of breast cancer. These cells were tagged with light-emitting markers to visualise and track their movement. After several weeks, they discovered that cells that had spread beyond the tumour were less adhesive than those that remained in the breast, indicating that the less sticky cells were more likely to escape.

Next, they tested whether these less sticky cancer cells could lead to full-blown metastasis. They separated tumour cells into weakly and strongly adherent groups and implanted them into another group of mice. Although mice in both groups formed breast tumours of similar size, only the mice injected with less sticky cells developed widespread lung metastases, even after the original breast tumour was surgically removed. Hence, the weaker the cancer cells stick to the breast tissue, the more likely they are to invade and colonise the lungs later.

Interestingly, the less sticky cells were not just physically different; they were also genetically distinct. Tumours that originated from these weakly adherent cells showed a unique pattern of gene activity, especially in genes involved in movement and invasion. When Kane et al. compared these patterns to cancer patient data, they found that higher expression of these same genes was linked to worse survival outcomes. To further validate their findings, Kane et al. tested several human breast cancer cell lines, which are lab-grown models of different cancer types. Some were fast-growing and prone to spread, while others were more benign. Sure enough, the more aggressive the cancer type, the less sticky it is and the faster it moves.

One of the most impressive parts of the study was how well their method predicted metastasis. By measuring the cancer cell stickiness in tissues just outside the tumour, they could forecast whether lung metastases would later appear with 100% specificity and 85% sensitivity, which are remarkably accurate by clinical standards. A 100% specificity means the method never gave a false alarm, i.e., falsely labelling a low-risk tumour as high-risk. An 85% sensitivity means that the method correctly predicted 85 out of every 100 tumours that went on to spread. In essence, their formula is good at catching high-risk tumours before they metastasise about 85% of the time yet never returned a false positive. Avoiding false alarms is crucial because it helps prevent unnecessary treatments for patients with low-risk tumours.

Finally, the team tested their method on human breast tissue samples from 16 patients, which included healthy individuals undergoing breast reduction surgery and breast cancer patients with either early-stage (ductal carcinoma in situ, DCIS) or advanced-stage (invasive ductal carcinoma, IDC) disease. As expected, cells from invasive cancers were the least sticky, followed by those from early-stage cancers. Healthy breast cells from non-cancer patients were the stickiest of all (Figure 3). This real-patient evidence reinforces their central discovery: weak adhesion is not just a laboratory observation but a defining feature of aggressive, metastatic human cancers.

Figure 3. Cancer cells from invasive tumours are less sticky than healthy breast cells. (A) This graph shows how tightly breast cells from different sources stick to a surface when exposed to increasing fluid force (shear stress). Healthy cells from breast reduction surgery (blue) stayed attached longer, while cells from early-stage (DCIS, red) and invasive breast cancer (IDC, green) detached more easily, indicating they were less sticky. (B) This bar chart quantifies the stickiness of the cells using a value called tau-75 (τ₇₅), which represents the force needed to detach 25% of the cells. On average, healthy breast cells were the stickiest, while invasive cancer cells were the least sticky. Source: Kane et al. (2025), Cell Reports.

Despite advances in cancer detection and treatment, accurately predicting which tumours will metastasise remains a major clinical challenge (Figure 4). Even after the primary tumour is removed or treated, many patients still harbour disseminated tumour cells (DTCs) that can lie dormant in distant tissues for years before reawakening to trigger metastatic recurrence. Moreover, metastasis unfolds differently between patients. So far, no single feature – be it tumour appearance under the microscope or its molecular subtype – can consistently predict who will develop metastatic disease. This unpredictability highlights the pressing need for more reliable tools to assess metastatic risk and guide treatment decisions accordingly.

To this end, the study by Kane et al. offers a novel approach by measuring the stickiness or adhesion strength of cancer cells as a predictor of metastatic risk. Unlike traditional diagnostics that rely on staining techniques or genetic profiling, their method uses a microfluidic device to physically test how tightly cancer cells cling to surfaces. The idea is simple but meaningful: tumour cells that barely stick are more likely to break away and seed metastases. Kane et al. even proposed the possibility of using pharmacological drugs to alter cancer cell stickiness to reduce metastatic risk. Rather than simply killing cancer cells, this strategy aims to reprogram their behaviour, which may reshape how we think about metastasis prevention.

While this idea of altering cancer cell stickiness to prevent metastasis is compelling, most current research has focused on preventing cancer cells from adhering at distant sites. After all, metastasis requires tumour cells to detach from the primary tumour and attach themselves to new tissues. As such, many experimental therapies aim to block this reattachment phase by targeting adhesion molecules. Several antibodies and small-molecule inhibitors have been developed to disrupt these molecules, some of which are undergoing clinical trials. However, none of these agents has yet received FDA approval to treat cancer.

One of our supplemental treatment products, Aeskulap-Modified Citrus Pectin (MCP), also helps intervene at this reattachment phase. MCP is a soluble, indigestible fibre derived from citrus peel and pulp, modified through heat and pH treatment to enhance its absorbability and bioactivity. As MCP is rich in β-galactose, it can bind to galectin-3, a β-galactoside-binding protein that cancer cells exploit to stick to blood vessel walls and distant tissues. By blocking galectin-3, MCP interferes with this crucial adhesive step in metastasis. Several preclinical studies have shown that MCP can reduce the formation of metastatic deposits by preventing circulating tumour cells from reattaching and colonising new organs (Figure 4).

Figure 4. How different compounds affected the ability of prostate (black bars) and breast (white bars) cancer cells to form metastatic deposits in the lungs of mice. Four treatments, i.e., antibodies against TF antigen and galectin-3, and two sugar-based compounds (lactulosyl-L-leucine and Modified Citrus Pectin (MCP), reduced lung metastases by over 90%, highlighting their strong anti-adhesion effect. In contrast, blocking selectin molecules (which are involved in other types of cell adhesion) had little to no effect. This suggests that stopping specific adhesive interactions, especially those involving galectin-3, could be a powerful strategy to prevent metastasis. Source: Glinskii et al. (2005), Neoplasia.

Of course, the study by Kane et al. is not without limitations. The tumour models used did not always mimic real-world clinical scenarios, such as patients who have undergone surgery or chemotherapy. While their results in patient samples were compelling, the dataset was small and lacked diversity. Moreover, the study focused exclusively on breast cancer, raising questions about generalisability to other cancer types. Nonetheless, as the basic mechanism of metastasis is shared across cancer types, the study provides a compelling case for further research into cancer cell stickiness as a behavioural marker of cancer aggressiveness.

“What we were able to show in this trial is that the physical property of how adhesive tumour cells could be a key metric to sort patients into more or less aggressive cancers,” said the study’s senior author Adam Engler, who is a professor at the UC San Diego Jacobs School of Engineering. “If we can improve diagnostic capabilities with this method, we could better personalise treatment plans based on the tumours that patients have.” Another senior author, Anne Wallace, a professor at UC San Diego Health’s Moores Cancer Centre, added, “We don’t want to over-treat with aggressive surgery, medicines and radiation if not necessary, but we need to utilise those when the cancer has higher invasive potential.”

In short, Kane et al. have introduced an ingenious way to stratify cancer risk through the physics of how cells move and behave. By reading a tumour’s mechanical signature early on, clinicians may one day predict and even prevent metastasis before it begins.