One of the biggest challenges in modern medicine is diagnosis. This is particularly important in cancer, where early detection can dramatically improve the chance of survival.
Diagnosis is difficult enough when you have access to fully-equipped laboratories and reliable technology. But what happens in developing countries where even basic infrastructure like refrigeration may not be available?
The answer is shockingly clear. On average, 70% of people in developing countries do not get diagnosed until a late stage, when treatment is no longer effective. As a result of this, there is a big international drive to develop simple and accurate diagnostic technology.
A lot of research in early cancer diagnosis has focused on biomarkers. These are molecules, cells or any other measurable biological characteristic that can be used as an objective way to detect disease. Biomarkers are particularly useful since they do not rely on symptoms perceived by the patient, which can be ambiguous or may not appear until the disease has progressed substantially.
For detecting cancer (and in fact many other diseases), a useful class of biomarkers are the glycoproteins. These molecules are proteins that are bound to one or more carbohydrate chains (sugars). They are found throughout the body, in blood, mucus, saliva and even sperm and they can be a signal for disease. If a suitable detection device is developed, a simple sample, for example of blood or urine, could be all that is needed to detect a cancer in its early stages.
Crucially, the biological function of a glycoprotein depends on its chemical structure. What is more important for disease diagnosis is that the chemical structure of a glycoprotein can be changed by the disease. This makes them ideal as a target for accurately detecting specific diseases.
Despite this promise, there are currently two big problems with using glycoproteins to diagnose cancer. The first problem is the technology. Existing methods rely on antibodies to detect glycoproteins. Antibodies are large proteins produced by our immune systems. They can be generated for use in devices, but this is challenging. They generally also need to be stored in refrigerators and they are readily degraded by UV light.
The second problem with glycoproteins is accuracy. Current detection methods rely on the fact that many individuals with cancer produce higher levels of glycoproteins. They do not account for healthy people who may naturally have high levels of certain glycoproteins. Some cancer patients may also have anomalously low levels of glycoproteins. Detection of glycoprotein levels with antibodies can therefore lead to a large number of false positive and negative results. This is distressing for the patient and can also waste crucial resources if a healthy person is sent for further testing or unnecessary treatment.
If early cancer diagnosis is to become available across the world, we need new technologies that are easy to transport and store. There also need to be substantial advances in accuracy.
Researchers at the University of Birmingham have come up with some clever chemistry that might solve both of these problems. The team has approached the problem from a different direction. Rather than just detecting levels of glycoproteins, they have created a method which can identify specific glycoproteins for specific diseases. They do this by detecting both the shape of the glycoprotein and its particular chemical fingerprint.
Lock and key
The method works by taking a disease biomarker (in this case, a glycoprotein associated with prostate cancer) and essentially taking a cast of it. The prostate cancer glycoprotein is tethered to a surface and detection molecules are assembled around it. When the glycoprotein is removed, it leaves behind a perfect chemical “cast”. The group have essentially made a lock and the only key that will fit is the specific prostate cancer glycoprotein. Other glycoproteins might be the right size, but they won’t be able to bind to the very specific molecules inside the lock.
This new work could pave the way to fast and accurate cancer diagnosis. What is particularly exciting is that the method does not rely on antibodies. This means that there is no need for special storage, which could make a big difference for early cancer diagnosis in the developing world.
Another exciting feature of this work is the scope for diagnosing other problems. The prostate cancer glycoprotein they have used is just a template. In theory, they should be able to make a device using any glycoprotein. This would mean the technology could be adapted according to what disease you want to detect, not just cancers, but also immune deficiencies, neurodegenerative diseases and cardiovascular disorders.
To keep up with the Agenda subscribe to our weekly newsletter.
Image: A researcher extracts a protein sample. REUTERS/Francois Lenoir.