The Dielectrophoresis Network

at the University of Surrey

Graphic1

Detecting Oral Cancer by DEP

One of our long-terms goals is the detection of oral cancer using dielectrophoresis, most recently using the 3DEP system.  Oral cancer is one of the top ten most common cancers in the UK – twice as common than cervical cancer, for example.  Annual diagnoses exceed 6000, annual deaths are currently around 2000.  Furthermore, there is a trend towards increasing incidence of the disease, particularly among the young.  Worldwide it is more significant, with annual deaths in India estimated in excess of 45,000, for example.

 

Despite the fact that oral cancers can be seen in the oral cavity, reticence on the part of dentists to refer patients for the gold-standard excisional biopsy to confirm diagnosis means that the majority (70%) of cancers are diagnosed in late stage.  This requires extensive revisional surgery and chemotherapy to treat, but still reduces life expectancy to 50%.  Where the cancer is caught early, life expectancy and quality of life are improved significantly.  However, there are no reliable methods allowing dentists and doctors to make an early diagnosis; the gold standard remains the highly invasive excisional biopsy, but dentists are generally loath to refer patients for this.  

 

Given the ready availability of cells from the tumour site (which can be collected painlessly on a Rovers brush, see picture), we have explored the possibility of using DEP as an early diagnostic method.  Samples have been collected from the Eastman Dental Institute and Bradford Royal Infirmary and analysed using 3DEP systems; sample were posted using conventional surface mail.  

 

Rather than determine the electrical properties of cells as a biomarker (though we have explored this), we have devised a simpler assay that does not require the measuring of cellular dimensions, examining instead the key spectrum events.  The differences in the spectra between normal and oral squmous cell carcinoma (OSCC) samples are clearly shown in the figure on the left; as can be seen, there are substantial differences between cancerous cells (red) and healthy cells (blue).  More broadly, we have established sensitivity and specificity rates above 82%, and are investigating improved sample preparation methods to increase this significantly.  If effective it will provide a low-cost method of cancer screening, without the need to buy a reader in every dentist’s.

 

Key papers in the archive: 37, 53, 60

oscc2 tongue