An Implantable Multilayer Device for Mechanical Sorting of Cancer Cells
Filtering Instead of Killing — A Long-Term Intratumoral Device Based on Flowable Diaphragms
Before a Building
One afternoon I stopped in front of a building. Vertical louvers densely set across its facade, a thin horizontal line at each floor, deep shadows between the columns. An ordinary mid-rise. And yet my eyes stayed too long on it. What if this structure could solve cancer. How would that be possible.
The first word that came to mind was 'prison'. A lattice of closely spaced columns is the archetype of confinement. What happens if cancer is not killed but contained. This one simple question touched several points at once. Metastasis, resistance, toxicity, adaptation — all the problems that conventional cancer therapy has been wrestling with began to look different once the premise 'kill' was replaced.
This essay records the full path of that thought experiment. How one question led to the next, how being stuck forced a return to structure, and how the chain finally arrived at a concept design called the 'multilayer flowable-diaphragm intratumoral sorting device'. Medical and engineering knowledge entered along the way, but the starting point was the geometry of that building alone. Because this document also serves as a prior-art disclosure, the technical specifics are deliberately concrete.
First Shift: What If We Filtered Instead of Killed
Once the word 'confinement' appeared, the notion of a 'sieve' followed naturally. To confine is to not let through. To filter is to let some through while holding others. The difference between the two is not small.
Nearly every modality of modern oncology is built on 'killing'. Chemotherapy kills by cytotoxicity, radiation by DNA damage, immunotherapy by recruiting immune cells to kill. This frame hides three structural limitations.
The first is selectivity. The biochemical differences between cancer cells and normal cells are smaller than one might expect, so drugs that kill almost always harm normal cells as well. Hair falls out, mucosal linings strip, marrow is suppressed. Targeted therapies have improved precision, but they still act across a whole population expressing a given receptor, not on 'cancer cells only'.
The second is resistance evolution. The selective pressure of killing selects for surviving clones that proliferate with resistance. This is Darwinian evolution occurring inside the patient's body. It is why a drug that works well in the first months loses power in the next. And once resistance appears, cross-resistance to other drugs often follows.
The third is heterogeneity. A single tumor contains subclones that differ genetically. Rarely does one drug act uniformly across the whole population. Some die, others survive, and the tumor refills.
Now shift the axis to filtering. Filtering is mechanical sorting. Whether the criterion is size, shape, or stiffness, it is physics rather than chemistry. Physical laws are not overcome by mutation. However many variants a cell accumulates, it cannot dramatically change its own size or cytoskeletal rigidity with one mutation. Filter-based sorting therefore cannot, in principle, develop resistance. Filtering also does not require systemic chemical administration. Place the device in one location, let the sorting happen inside that device, and systemic toxicity is structurally excluded. Finally, filtering is agnostic to heterogeneity. Whether there are three subclones or thirty, as long as they share 'the physical state of being a cancer cell' they are caught by the same sieve.
These three advantages are not incidental. They follow necessarily from the shift of axis — from drug to geometry, from chemistry to physics. Pushing this thought a little further, one question followed. Has medicine already done something like this.
Filtering Is Not Foreign to Medicine
Looking back, filtering turned out to be a concept modern medicine owes a great deal to. It simply had not yet crossed into cancer care.
At the end of the nineteenth century, kidney failure was a lethal chronic disease. In 1924 the German physician Georg Haas made the first attempt at hemodialysis in humans, and in 1945 the Dutch physician Willem Kolff succeeded clinically with a rotating drum dialyzer made of cellophane membranes. The device is not a drug. It is only a semi-permeable membrane. Small wastes pass through, larger proteins are blocked. Filtering in the literal sense. Millions of people worldwide now sustain their lives on this principle. There is still no drug that 'cures' chronic renal failure, yet people live decades on a filtering machine.
In the fourteenth century the Republic of Venice detained ships arriving from the East at sea for forty days. Nobody then understood how plague spread. People filtered out 'the potentially dangerous' as a whole rather than the causative pathogen specifically. Quaranta giorni — forty days. From it the word 'quarantine' came. This too is filtering. When the disease could not be treated, a structure was designed so that it could not spread.
Early in the twentieth century, Alexis Carrel, the father of vascular surgery, developed the technique of simply suturing arteries and veins. Before him, major vascular injury meant death. Carrel's contribution was not a drug but a technical structure. He opened the door to organ transplantation, and after him came artificial hearts, prosthetic valves, and stents. This is the tradition of managing disease through structure. Closer to our time we have hemodialysis, extracorporeal membrane oxygenation (ECMO), left ventricular assist devices (LVAD), and cardiac pacemakers. Their common trait is 'solving with structure what drugs cannot solve'.
What was striking was that cancer care had not joined this lineage. Every other organ disease had experienced the paradigm of 'filtering and assisting with a machine', but not cancer. The reason was technical. Cancer cells are smaller, more diffuse, and more subtle in their behavior than waste molecules in blood, which made filtering them difficult. But in the 2020s, microfluidic engineering, biodegradable polymers, and viscoelastic fluid control have all reached sufficient maturity. We can now think about a 'filtering device' for cancer as well.
In fact, this shift had already begun a few years earlier. In 2021, Tijore and Sheetz at the National University of Singapore published a paper in Biomaterials introducing the concept of 'mechanoptosis' — mechanically induced apoptosis. Cyclic mechanical stretching triggered influx of calcium through Piezo1 channels in cancer cells specifically, causing selective death while sparing normal cells. Follow-up work in 2024 reproduced the same effect using ultrasound, and selective killing was demonstrated in patient-derived pancreatic tumor organoids. The group founded Mechanobiologics Inc. to commercialize the technology. An earlier US patent from 2013 had already claimed a device using tuned resonant ultrasound to mechanically disrupt cell and nuclear membranes for selective cancer destruction. In other words, the principle of 'mechanically killing only cancer cells, selectively' is no longer a hypothesis. It is a therapeutic grammar with experimental foundations.
That this shift was first enacted through ultrasound and external stretching was technically natural. External energy sources are easier to install and control. But external forces have depth limits, and they fragment treatment into sessions. If the same mechanical sorting could be performed 'continuously, from within the body', a different dimension of possibility would open.
So how should the device look. The question returned to structure.
Then How Should the Device Look
Here, the geometry of the building that had stopped my walk became the starting point of the answer. The vertical louvers of the facade, the thin horizontal line at each floor, the deep shadows between columns — these three elements corresponded directly to the three scales of the device.
The single unit is a cylinder-shaped element. Not a hollow tube but a solid-looking cylindrical structure. Long along the vertical axis, circular in cross-section. Inside, it is segmented into several horizontal layers. The space between layers becomes a 'room' for cells.
Many such units sit side by side in parallel to form a bundle. Between adjacent units there are fixed separating walls, so cells cannot move freely between units. The whole becomes a three-dimensional cell array compartmentalized in all three directions — vertical, horizontal, and depth. Up to this point, the geometry is a scaled-down version of the building facade.
The decisive feature of this structure was the 'horizontal layer'. At first I imagined these layers as fixed walls. A fixed wall makes each layer a cell prison. But long-term therapy requires cells to flow in and out, and fixed walls block flow. So the idea was inverted — what if these layers were 'flowable'.
This single inversion created the heart of the whole design. A flowable diaphragm changes shape when cells approach. It flows in its resting state, hardens locally on cell contact to apply pressure, and returns to flow once the cell passes. A wall that is not a wall. This ambiguous state became the physical implementation of 'sorting'.
As the material for the flowable diaphragm I chose a 'viscoelastic polymer solution with shear-thickening properties'. The same principle underlies modern body armor. It flows like water at rest but stiffens instantly when sudden pressure is applied. Aqueous, composed of biocompatible polymers, stable in vivo for months.
For the vertical outer walls and the separating walls between units, PLGA (poly-lactide-co-glycolide) is chosen. An FDA-approved biodegradable polymer, it has been in clinical use for decades in surgical sutures and implants. The degradation rate can be tuned by composition ratio, making device lifetimes adjustable from months to over a year.
The device is inserted into tumor tissue via image-guided percutaneous puncture or direct surgical placement. The next question then followed — what makes this device run.
Power Comes from the Tumor Itself
I did not want to attach an external energy source to the implant. Batteries need replacement, wires create infection paths, external magnetic fields or light have depth limits. The more complex, the more failure points. So I inverted the question — could the tumor itself be the power source.
The answer lay in the physical peculiarity of tumors. A tumor is a mechanically different place from normal tissue. Blood vessels grow in excess but with poor structure, lymphatic drainage is deficient, and interstitial fluid accumulates. The result is that the interstitial pressure in the tumor core is higher than in its periphery. This is the well-established phenomenon of 'elevated interstitial fluid pressure'. Fluid flows slowly from center to edge on its own.
This flow could serve as the power. If the top of the device sits at the tumor core and the bottom at the periphery, interstitial fluid naturally enters from above and exits below. Cells ride this flow through the device. The device borrows, as its own operating energy, the pressure gradient the tumor itself creates. The peculiarity of the disease becomes the energy source of its treatment. A paradox built into the design.
To reinforce the flow, the outer vertical walls carry cancer-cell-attracting signaling molecules. CXCL12 or similar chemokines, slowly released, bias surrounding cells toward the device. Not passive waiting but active collection.
The fluid flow is the operational energy. Even if drug reservoirs empty, even without batteries or external signals, the device continues to function. As long as the structure holds, operation continues. Here another question surfaced — but how does the diaphragm distinguish cancer cells from normal cells.
Two Different Cells
Whether cells can be distinguished physically is an old question. The answer is yes. It is simply a matter of identifying where exactly the difference lies.
The key clue is the mechanical properties of the cell. The proteins that structurally support the nucleus are the lamin family. Lamin A/C forms a mesh along the inner nuclear membrane, maintaining nuclear shape and protecting against external pressure. Normal cells express lamin A/C abundantly, giving them firm nuclei. Cancer cells often show markedly reduced lamin A/C expression. Metastatic cancer cells, in particular, tend to have the lowest lamin A/C. Paradoxically this is what enables metastasis — the lamina must be weak for the cell to squeeze through narrow spaces.
The same property becomes a vulnerability under sustained mechanical compression. A 2016 Science paper by Denais et al. showed that 'nuclear envelope rupture' occurs when cancer cells pass through confined spaces. When the envelope ruptures, DNA is exposed to cytoplasm and nucleocytoplasmic traffic goes out of control. Without repair, DNA damage accumulates and the cell dies.
This was the decisive point. If the diaphragm is flowable, cells are not merely passing through. They are captured and compressed for a period of time. Cancer cells, with weak lamina, deform in the nucleus, and some rupture. Normal cells, with firm lamina, deform temporarily then recover as they descend to the next layer. Sorting becomes principled.
But once I learned this property I immediately wondered. Then why is there not yet a therapy that exploits this vulnerability. Looking for an answer, I met two obstacles.
Obstacles and the Inevitability of Multilayer Structure
The first obstacle was the cell's self-repair machinery. A companion 2016 Science paper by Raab et al. showed that an ESCRT-III protein complex seals nuclear envelope ruptures within about two minutes. Under normal conditions, over 90% of cells survive even repeated nuclear envelope ruptures. Even if a diaphragm ruptures the envelope once, two minutes later it is repaired. Simple compression does not kill.
The second obstacle came from a 2020 Cell Reports paper by Moose et al. When cancer cells are exposed to fluid shear stress, they activate RhoA and mount a mechano-adaptive response. Formin and myosin II reinforce the cytoskeleton and defend against mechanical damage. Cancer cells freshly isolated from mouse and human tumors were, somewhat paradoxically, more resistant to fluid shear than normal epithelial cells. This partially reversed the simple assumption that cancer cells are mechanically fragile.
Two obstacles. And at this precise point, the necessity of a multilayer structure became self-evident.
One layer cannot solve it. A single compression is repaired. A single sorting attempt is evaded by mechano-adaptation. Repetition is needed, and so is repair blockade, adaptation suppression, and post-escape containment. These four functions cannot all fit in one layer. Spatial partition is unavoidable.
The vertical repetition seen in the building facade translated here into therapeutic function. A layer is not mere partition but 'time order inscribed onto space'. As the cell descends through the flow, a program of attack, blockade, re-attack, and final suppression executes in sequence.
Incidentally, the direction of 'mechanical attack combined with mechano-adaptation inhibitor' has been independently validated in recent literature. A 2025 Nano Letters study reported that combining ultrasound-induced mechanoptosis with Cilengitide (an integrin receptor blocker) enhanced tumor killing. Inhibiting the mechano-adaptive pathway while delivering mechanical force is, independently, the correct design direction. The inclusion of a myosin II inhibitor in the fourth layer of this design is the same principle translated into in situ local release.
A Five-Layer Treatment Program
Layer 1 delivers the first mechanical attack. The flowable diaphragm applies cyclic pulsed compression. Nuclear envelope rupture is induced in cancer cells and the cytoskeleton responds with mechano-adaptation. At this moment cells are not yet dead. They are simply under stress and starting to mount repair.
Layer 2 is the nuclear-envelope-repair blockade zone. Molecules that inhibit ESCRT-III function are released locally. As cells that received damage in Layer 1 pass through, the repair machinery cannot act. Damage accumulates as cells descend.
Layer 3 delivers the second mechanical attack. With repair capacity blocked, the flowable diaphragm applies pressure again. This rupture is not recovered. DNA double-strand breaks accumulate, and some cells die here.
Layer 4 is the zone of DNA-repair blockade and immune ignition. PARP-inhibitor-class molecules are released to prevent intracellular DNA damage from being repaired. Simultaneously, STING agonists are released so that DNA fragments leaking from ruptured nuclei become signals that trigger local immune response. Myosin II inhibitors are added to neutralize mechano-adaptive resistance itself. The majority of cells passing through this layer enter death pathways.
Layer 5 is the zone of containment for any latent escapers. For the rare cell that manages to pass through the previous four layers, CDK4/6-inhibitor-class molecules arrest the cell cycle. Even if alive, the cell cannot divide. It is fixed in a state unable to regrow into a tumor.
The true meaning of these five layers is not 'overlapping attack'. The drugs used — ESCRT inhibitors, PARP inhibitors, STING agonists, myosin II inhibitors, CDK4/6 inhibitors — are difficult to combine systemically because each carries significant toxicity. Marrow suppression, autoimmunity, organ damage, all amplified when combined. Giving all of them to one patient at once intravenously is essentially impossible in the clinic.
But when released locally into spatially partitioned micro-environments by the diaphragms, they act only on cells within that compartment and are metabolized before diffusing out of the device. 'Combination therapy that was theoretically impossible' becomes possible. The therapeutic degree of freedom of the multilayer structure arises here.
Spatial order is temporal order. As the cell rides the flow through Layers 1, 2, 3, 4, 5, the sequential program of attack → repair blockade → re-attack → final blockade → latent containment executes automatically. Intravenous administration cannot control such ordering. Inside the device, space takes the place of time.
Three Rhythms
The device is designed for long-term treatment. Three rhythms operate in superposition: continuous, cyclic, and terminal.
Continuous operation is flow-through sorting. Cells enter from above, descend through each layer of attack, and the process continues for months. Drug reservoirs release slowly, without abrupt depletion.
Cyclic operation is a weekly concentrated strike. In normal state the flow continues, but once a week the fluid flow is halted. Because the viscoelastic properties of the diaphragm can be designed to shift periodically at the material level, this can be configured without any external signal. When flow stops, cells accumulated inside the device contact the diaphragms as a group. A batch attack is delivered, then flow resumes. This rhythm coincides with the cycle pattern of standard chemotherapy — an efficient way to catch heterogeneous populations whose proliferation is not synchronized.
Terminal operation is a large-scale final strike followed by disassembly. When the PLGA structure reaches its designed degradation period, the outer wall weakens and the diaphragms fuse. The fused diaphragm delivers a final attack on any cells remaining inside. Then the whole structure degrades on its own.
This degradation being clean is a hidden strength of the design. PLGA hydrolyzes in vivo into lactic acid and glycolic acid. Both are small organic acids the body metabolizes every day. They are processed in the liver and ultimately converted to carbon dioxide and water, exiting through breath and urine. Thanks to the decades of clinical history of PLGA as surgical sutures, patients do not need a separate procedure to remove the device after treatment. The diaphragm fluid, also composed of biocompatible polymers, is absorbed or excreted after degradation. Normal cells held inside the device during operation are naturally released as the structure comes apart and return to surrounding tissue. Insert once, and it is done. This simplicity is a major advantage for implantable medical devices.
The three scales — layer-by-layer attack on the minute scale, cycle rhythm on the weekly scale, long-term management and clean termination on the monthly scale — mean a single device handles multiple time scales of tumor management at once.
What the Device Can Actually Do
This device does not claim to be 'the solution for all cancers'. That is a frame the history of oncology has repeatedly offered and repeatedly disappointed. Instead, it secures credibility by focusing on specific indications.
First, prevention of recurrence after resection. For breast cancer, soft tissue sarcoma, brain tumors, ovarian cancer, and other cancers where surgery removes the main mass, this device blocks local recurrence driven by residual microscopic cancer cells. It corresponds to an active dynamic version of the existing Gliadel wafer, placed in the resection cavity after surgery to process residual cells over several months.
Second, interception of metastatic routes. Placed in lymph node areas suspected of micrometastasis, or along blood vessel paths near the tumor, it captures cancer cells trying to circulate. Metastatic cells, with their low lamin A/C, are especially vulnerable to this device. The selectivity 'the most dangerous cells are caught best' is realized naturally.
Third, local management of unresectable tumors. In pancreatic cancer, hepatocellular carcinoma, glioblastoma, and other tumors difficult to resect curatively, the device suppresses tumor growth for months or longer, extending quality of life and survival. A chronic disease management model.
Fourth, a general platform for localizing combination therapy. Arguably this is the greatest value of the device in principle. Beyond mechanical sorting, the diaphragm-based spatial isolation enables the combined use of drugs that cannot be combined systemically due to toxicity. Indications may not be limited to cancer — antibiotic combinations, immunosuppressant combinations, immunostimulant combinations may all benefit.
Nearby Work and the Distance of This Design
Not every component of this concept design is new. Each element exists in previously published research. What is new is the way they are integrated. Examining nearby precedents makes the boundaries clearer.
The most direct precedent is the mechanoptosis field established by the Tijore-Sheetz group at the National University of Singapore. It began with their 2021 Biomaterials paper "Selective killing of transformed cells by mechanical stretch" and expanded into ultrasound-mediated mechanical killing, Piezo1-dependent calcium influx mechanisms, and patient-derived organoid validation. Mechanobiologics Inc. has been founded for commercialization, and the 2013 US patent US20130131432 already claimed selective cancer cell destruction via tuned resonant ultrasound. This design shares the 'mechanical selective cancer cell killing' principle of mechanoptosis. However, the delivery mode and operation form differ fundamentally. Prior work uses external ultrasound or external stretching and operates in sessions. This design uses implanted flowable diaphragms and operates continuously over the long term. It uses tumor interstitial pressure as its energy source without external power, and combines multilayer compartmentalization with spatially sequential combination drug release rather than a single modality. In summary, this design is an implantable, long-term, multilayer-integrated implementation variant within the mechanoptosis field.
The implantable microdevice developed by the Cima lab at MIT since 2016 and reported in Nature Biotechnology in 2022 is an intratumoral multi-drug screening device that delivers different drugs simultaneously into spatially separated regions. It overlaps with this design in being a 'spatially partitioned multi-drug platform'. The differences are two. Cima's device has no mechanical sorting function, and its in vivo residence is limited to 2 to 3 hours. This design incorporates long-term operation over months and mechanical sorting by flowable diaphragms.
The PLG scaffold trap developed by the Shea lab at Northwestern is under evaluation in clinical trial NCT03572673 for metastatic cell capture. The concept of 'implantable cancer cell capture' is shared. The difference is that Shea's scaffold is a passive porous structure without diaphragms. This design performs sorting and processing via active flowable diaphragms.
The nanofluidic drug eluting seed (NDES) developed by Grattoni et al. at Houston Methodist is a 3.5 mm-long cylindrical intratumoral device. The cylindrical form and intratumoral placement are shared, but it remains a single-unit drug-releasing device. This design has a cylinder bundle and horizontal multilayer compartmentalization.
The viscoelastic deformability cytometry (vDC) reported by Rodriguez-Tirado et al. in Science Advances in 2021 implemented viscoelastic-fluid-based cell sorting in an external microfluidic platform. It provides direct evidence that the diaphragm principle of this design actually works. However, vDC is an external analytical device, not therapeutic. This design transplants the same principle into the context of in vivo long-term therapy.
The study by Agustsson et al. published in Scientific Reports in 2020 experimentally demonstrated that tuning fluid viscoelasticity can, within the same structure, either rupture cells or protect them. This directly supports the mode of operation assumed by this design — 'the diaphragm ruptures cancer cells while protecting normal cells'.
Novocure's US patent US20220096854A1 claims implantable devices with multiple stimulation zones providing tumor treating fields. Implantation and multi-zone architecture are shared, but it is electric-field based. This design performs sorting via flowable diaphragms rather than electric fields.
The 2025 Nature Biomedical Engineering paper by Wang et al. applied magnetically driven biohybrid blood hydrogel fibers to intracranial tumor therapy. It shows the recent trajectory of implantable cancer therapy devices but is a different approach from diaphragm-based cell sorting. This design uses no external energy and relies on tumor interstitial pressure.
As this comparison makes clear, each precedent shares one specific element of this design. Mechanoptosis provided the principle of 'mechanical selective killing'. Cima, Shea, and Grattoni offered various forms of 'implantable intratumoral cancer therapy devices'. vDC and viscoelastic fluid studies demonstrated the physical plausibility of 'viscoelastic diaphragm-based cell sorting'. However, the integrated combination — 'implantable + cylindrical unit bundle + horizontal multilayer compartmentalization + viscoelastic flowable diaphragm + mechanical cell sorting + spatially sequential combination drug release + long-term operation + external-energy-free operation by interstitial pressure' — has not been identified in currently published literature and patents. The novelty of this design rests entirely in this integration.
Next Steps and the Meaning of Public Disclosure
Concrete implementation of this concept will require several experimental validations. First, in vitro experiments to finalize the specific composition of the diaphragm fluid. Then microfluidic validation measuring cell sorting efficiency at the single-unit level, and pharmacokinetic experiments to measure drug release profiles in multilayer structures. Organ-on-chip validation of operation within simulated tumor tissue follows, then safety and efficacy evaluation in small-animal models.
On the materials side, the degradation rate of PLGA, protein-adsorption-resistant surface treatment of the diaphragm fluid, and the sustained-release kinetics of the attracting signal molecules are the key engineering tasks. Because each component technology is already established in the medical device field, the engineering difficulty of the integrated design is estimated to be moderate. On the indication side, prevention of local recurrence after breast cancer resection may be the fastest clinical entry route. The indication has the precedent of the Gliadel wafer, offers good surgical accessibility, and has a relatively clear clinical trial design.
The purpose of this public disclosure is not exclusive appropriation by any specific group. The opposite. The core elements described here — horizontal multilayer compartmentalization of cylindrical unit bundles, mechanical cell sorting by viscoelastic flowable diaphragms, spatially sequential multilayer drug release, external-energy-free operation based on interstitial pressure, biodegradable terminal design — are disclosed from the date of this document as prior art. Interested researchers and companies are free to examine and implement them in their own ways. This document functions as a record against future monopolistic patent claims on these elements. For formal research collaboration inquiries, contact is possible through wonbrand.co.kr.
Back at That Building
Let us return to the facade that stopped my walk. The reason the exterior of an ordinary mid-rise could be translated into a cancer therapy concept is that its geometry is the ancient archetype of 'filtering'. Sieves, meshes, lattices, filters, dialysis membranes. Humans have filtered water, filtered air, filtered blood. Only the filtering of cells has been postponed for a long time.
'Killing cancer' has been the basic grammar of the last hundred years of medicine. This grammar produced achievements. It also left unsolved problems — resistance, toxicity, heterogeneity, metastasis. Some parts of these problems may come from the grammar itself. It is time to try a different grammar.
In truth, the attempt has already begun. Mechanoptosis, mechanical cell sorting, implantable intratumoral devices — scattered studies have each been writing a little of the language of 'filtering' in their own corners. If there is something new in this design, it is not that the premise is new, but that these scattered languages have been bound together into one integrated device.
Filter. Do not kill, filter. Not with drug but with structure. Not systemically but locally. Not once but over months.
This grammar does not replace the older one. The two combine. Surgery, radiation, chemotherapy, immunotherapy all remain valid tools. A 'filtering device' is simply added to them. But the difference that one addition can make may resolve some of the problems the hundred-year grammar could not.
A building met on a walk led to the thought of a device. Whether this device will ever change the life of one person is another question. But the question is now raised. Can we bring the grammar of filtering — which has already begun — into the body, over the long term, as an integrated system. To share this question is the first purpose of this writing.
References
Tijore, A. et al. (2021). Selective killing of transformed cells by mechanical stretch. Biomaterials 275, 120866.
Tijore, A. et al. (2024). Ultrasound-mediated mechanical forces activate selective tumor cell apoptosis. Advanced Healthcare Materials (published online 2024).
Saigaonkar, S. et al. (2025). Nanoscale Ligand Spacing Regulates Mechanical Force-Induced Cancer Cell Killing. Nano Letters 25, 2418–2425.
Sankar, G. et al. (2026). Selective killing of cancer-associated fibroblasts by ultrasound-mediated mechanical forces. Biomaterials 328, 123844.
Patent: Tijore, A., Yao, M., Sheetz, M. et al. (2013). Selective Destruction of Cancer Cells via Tuned Ultrasound Resonance. US 20130131432.
Raab, M. et al. (2016). ESCRT III repairs nuclear envelope ruptures during cell migration to limit DNA damage and cell death. Science 352, 359–362.
Denais, C. M. et al. (2016). Nuclear envelope rupture and repair during cancer cell migration. Science 352, 353–358.
Moose, D. L. et al. (2020). Cancer Cells Resist Mechanical Destruction in Circulation via RhoA/Actomyosin-Dependent Mechano-Adaptation. Cell Reports 30, 3864–3874.
Hatch, E. M., Hetzer, M. W. (2016). Nuclear envelope rupture is induced by actin-based nucleus confinement. Journal of Cell Biology 215, 27–36.
Rodriguez-Tirado, C. et al. (2021). Real-time viscoelastic deformability cytometry: High-throughput mechanical phenotyping of liquid and solid biopsies. Science Advances 7, eabj1133.
Agustsson, A. S. et al. (2020). Improving viability of leukemia cells by tailoring shell fluid rheology in constricted microcapillary. Scientific Reports 10, 11469.
Tatarova, Z. et al. (2022). A multiplex implantable microdevice assay identifies synergistic combinations of cancer immunotherapies and conventional drugs. Nature Biotechnology 40, 1823–1833.
Jonas, O. et al. (2015). An implantable microdevice to perform high-throughput in vivo drug sensitivity testing in tumors. Science Translational Medicine 7, 284ra57.
Rao, S. S. et al. (2016). Enhanced Survival with Implantable Scaffolds That Capture Metastatic Breast Cancer Cells In Vivo. Cancer Research 76, 5209–5218.
Chen, J. et al. (2012). Microfluidics separation reveals the stem-cell-like deformability of tumor-initiating cells. PNAS 109, 18707–18712.
Wang, B. et al. (2025). Magnetically driven biohybrid blood hydrogel fibres for personalized intracranial tumour therapy under fluoroscopic tracking. Nature Biomedical Engineering 9, 1471–1485.
Kazim, M., Pal, A., Goswami, D. (2025). Mechanical Metamaterials for Bioengineering: In Vitro, Wearable, and Implantable Applications. Advanced Engineering Materials 27, 2401806.
Grattoni, A. et al. (2020). Nanofluidic-based drug eluting seed: Advanced implantable drug delivery technologies for chronic diseases. Drug Delivery and Translational Research 10, 1274–1296.
Stylianopoulos, T., Jain, R. K. (2013). Combining two strategies to improve perfusion and drug delivery in solid tumors. PNAS 110, 18632–18637.
Jain, R. K. (1987). Transport of molecules across tumor vasculature. Cancer and Metastasis Reviews 6, 559–593.
Patent: Novocure Ltd. (2022). Implantable Arrays For Providing Tumor Treating Fields. US20220096854A1.
Patent: Giner Life Sciences Inc. (2023). Implantable Cell Chamber Device And Uses Thereof. EP 4120958 A1.
Clinical trial: An Implantable Device for Early Detection of Metastasis of Breast Cancer. ClinicalTrials.gov Identifier NCT03572673.
Historical reference: Kolff, W. J., Berk, H. T. J. (1944). The artificial kidney: a dialyser with a great area. Acta Medica Scandinavica 117, 121–134.
Historical reference: Tognotti, E. (2013). Lessons from the History of Quarantine, from Plague to Influenza A. Emerging Infectious Diseases 19, 254–259.
An Seungwon / Wonbrand / https://wonbrand.co.kr