How do you build a sensor that blood cannot clog? Copy a cell

Blood destroys sensors. Within minutes of contact, proteins, cells, platelets, and debris coat the sensing surface in a sticky biological film that blinds the device to whatever it was trying to measure. This problem - called biofouling - has been the central obstacle to real-time blood monitoring for decades. Every promising detection technology eventually runs into the same wall: it works beautifully in a clean buffer solution in the lab and fails within minutes in actual blood. The clinical implications are significant. Real-time monitoring of drug levels, hormones, and biomarkers in the bloodstream could transform intensive care, but the biology of blood itself keeps getting in the way.

A team led by La Trobe University in Australia, working with CSIRO (the national science agency), has found a way around the wall. Their approach, published in ACS Sensors, borrows a strategy that living cells have been using for hundreds of millions of years - and it works well enough to track a drug in unprocessed blood for more than 10 consecutive hours.

Stealing a design from biology

Cell surfaces face exactly the same challenge that artificial sensors do. They sit in blood, surrounded by proteins and debris, and yet they maintain the ability to detect specific molecules with exquisite sensitivity. They manage this through a combination of protective coatings that repel unwanted material and specialized receptors embedded within those coatings that capture exactly what the cell needs to sense.

The La Trobe team replicated this architecture with synthetic components. They coated their sensor surface with lubricin, a naturally occurring glycoprotein that prevents unwanted material from sticking to biological surfaces. Lubricin is best known for its role in joint lubrication - it is the molecule that keeps cartilage surfaces slippery inside your knees and hips. Its anti-adhesive properties are so effective that researchers have been exploring its potential for medical device coatings for years. On a sensor surface, it creates a molecular shield that repels the blood components responsible for fouling while remaining thin enough to allow target molecules to pass through.

Within that protective lubricin layer, the researchers embedded aptamers - short, single-stranded DNA molecules that can be engineered to bind specific target molecules with high affinity and selectivity. Unlike traditional antibody-based receptors, aptamers respond quickly, can be synthesized reproducibly, and can be designed for virtually any molecular target. Think of them as programmable molecular hooks, sitting safely beneath an anti-fouling umbrella, grabbing the molecule of interest as it passes by while everything else slides off the surface.

100 million times more sensitive

The detection method sitting beneath the lubricin-aptamer layer is a technique called Surface-Enhanced Raman Scattering, or SERS. It reads molecular fingerprints using laser light, and it is extraordinarily sensitive - capable, in principle, of detecting a single molecule. The technique works by concentrating light at nanoscale metal surfaces, amplifying the faint Raman scattering signal that molecules produce when they interact with light. That extreme sensitivity is also its Achilles' heel in biological applications. SERS surfaces are so reactive that in blood, they get contaminated almost immediately. The signal from the target molecule drowns in noise from everything else.

The lubricin-aptamer combination solved both problems at once. The coating prevented fouling while the aptamers captured target molecules and brought them close enough to the SERS surface for detection. The noise stayed out. The signal got in.

The result was dramatic. The team used their sensor to detect vancomycin, an antibiotic commonly used in hospitals for serious infections including MRSA, in unprocessed blood samples. Not filtered blood. Not diluted blood. Not blood processed to remove proteins and cells. Whole blood, straight from the body, with all its fouling potential intact. The sensor maintained its sensitivity for more than 10 hours of continuous exposure without degradation - a world first for SERS-based detection in blood.

Mingyu Han, a research co-leader from CSIRO, put the performance in perspective: the sensor is 100 million times more sensitive than previous designs that detected vancomycin in biological fluids. That is not an incremental improvement. It represents a qualitative leap into a sensitivity range where clinically relevant concentrations can be tracked in real time rather than through periodic blood draws.

Why vancomycin is the right test case

Vancomycin illustrates a real clinical problem that affects thousands of patients daily. The antibiotic has a notoriously narrow therapeutic window - too little and the infection persists, potentially allowing resistant bacteria to emerge; too much and the drug damages the kidneys, sometimes permanently. Getting the dose right matters, and getting it right requires knowing the drug's concentration in the patient's blood.

Current practice relies on periodic blood draws sent to a hospital laboratory for analysis. The turnaround takes hours, and the results provide only snapshots - static measurements at isolated time points rather than the continuous picture that dynamic drug metabolism actually produces. Between measurements, the drug concentration rises after a dose, peaks, and falls. Clinicians make dosing decisions based on educated guesses about where the concentration is between measurements. For stable patients, this works well enough. For critically ill patients in the ICU, where kidney function can change rapidly, blood volume shifts with fluid resuscitation, and drug metabolism is unpredictable, the guesswork can mean the difference between treatment success and organ damage.

A sensor that monitors vancomycin levels continuously and in real time would allow clinicians to adjust doses dynamically, keeping the drug within its effective range moment to moment rather than dose to dose.

Beyond vancomycin: hormones, toxins, and biomarkers

Vancomycin is the proof of concept, not the endpoint. "Our sensor greatly expands the detection range, allowing us to measure hormones, toxins and other biomarkers that appear only at low concentrations," said Wren Greene, an associate professor at La Trobe who led the research. "This is critical for early disease detection and monitoring the body's response to treatments."

The aptamer component is the key to this versatility. Because aptamers can be engineered for virtually any molecular target, swapping the vancomycin aptamer for one targeting cortisol, troponin, or a cancer biomarker is conceptually straightforward. Each application requires its own validation, but the platform architecture remains the same: lubricin shield, aptamer receptor, SERS detection.

The sensor was built on a platform called NanoMslide, developed by AlleSense, a La Trobe spinout company. Brian Abbey, a La Trobe Distinguished Professor, said AlleSense is establishing clinical-scale manufacturing at the university. The long-term goal is to prototype an inexpensive, mass-produced test strip similar to a blood glucose test - a device that could bring SERS-level sensitivity to point-of-care settings rather than confining it to research laboratories.

What the study does not yet show

The work so far demonstrates detection of a single analyte in a controlled laboratory setting. Clinical validation - testing in actual patients with varying health conditions, medications, ages, and blood compositions - has not been performed. Blood from different patients varies significantly in protein content, hematocrit, lipid levels, and other factors that might affect sensor performance in ways not captured by benchtop experiments using standardized samples.

The 10-hour continuous monitoring window, while impressive and unprecedented for SERS in blood, may not be sufficient for all clinical applications. Some ICU stays last days or weeks. Whether the lubricin coating maintains its anti-fouling properties over longer periods, or whether it gradually degrades under sustained biological exposure, remains to be tested under clinical conditions.

The manufacturing challenge is also nontrivial. Moving from a hand-assembled laboratory prototype to a mass-produced device that works reliably across thousands of clinical settings involves challenges in consistency, quality control, regulatory approval, shelf life, sterilization, and integration with existing clinical monitoring systems. Each of these is a substantial hurdle, and the history of biosensor development is littered with technologies that performed brilliantly in the lab and failed at the manufacturing or regulatory stage.

Still, the core achievement is genuine: a sensing platform that works in whole blood for extended periods without fouling, with sensitivity orders of magnitude beyond previous approaches. If the translation challenges can be met, the clinical applications span drug monitoring, sepsis detection, personalized chemotherapy dosing, hormone tracking, and early disease screening.

The research was conducted with support from the ARC Research Hub for Molecular Biosensors at Point-of-Use (MOBIUS). Lubricin was provided by industry partner Lubris Biopharma.