Cancer of the oral cavity is one of the most common malignancies in Asia. In Singapore, it is the eighth most common cancer in the Indian population and overall accounts for over 400 new cases every year. The 5-year survival rates remain poor despite improvements in surgical techniques and advances in adjuvant and neoadjuvant treatment modalities. This is largely due to the fact that majority of patients present with advanced disease.

Early detection of pre-malignant or early stage cancer will allow curative treatment that will significantly improve survival rates and secure optimum organ preservation. However small and flat mucosal neoplasm, like dysplasia and carcinoma in situ, are difficult to identify using typical morphological criteria by white light endoscopy, and are frequently overlooked in routine examinations.

Several studies assessed the applicability of vital staining with agents such as Lugol’s iodine and toluidine blue for improving detection of early oral neoplasia. But these methods have not found their clinical usefulness in the head and neck clinic due to high false-positive or false-negative rates. Therefore, there is an urgent need to develop rapid, non-invasive techniques for clinical diagnosis and evaluation of early oral malignancy in vivo.

Recently, light-induced fluorescence (LIF) spectroscopy has shown potential of improving diagnostic accuracy and efficacy for detecting cancer in various organs including the oral cavity. The principle of LIF technique is based on detecting tissue fluorescence from endogenous fluorophores in tissue or exogenous fluorescence from photosensitisers accumulated in tumour tissue.

One of the most promising photosensitisers for oral cancer diagnosis is 5-aminolevulinic acid (ALA) induced protoporphyrin IX (PPIX). Topical or systemic administration of 5-ALA leads to selective accumulation of PPIX in neoplastic tissues, resulting in a high contrast as compared to surrounding normal tissue. ALA induced PPIX has a further advantage of being excreted from human tissue in less than six hours and avoiding side effects of skin phototoxicity encountered when using approved photosensitisers, such as hematoporphyrin derivatives (HpD) or Photofrin™.

More recently, there is an increased interest in applying 5-ALA fluorescence-based technique to clinical oncology. Several groups reported promising results on ALA-induced fluorescence endoscopic imaging for detecting pre-malignant and malignant oral mucosa with a high degree of sensitivity of 95%-100%.

However, benign oral lesions are still misdiagnosed as positive, resulting in a lower diagnostic specificity during ALA-mediated fluorescence endoscopy. Most of the ALA-mediated fluorescence imaging systems cannot be used to quantify fluorescence images, thus hampering the ability to explore the relationship between fluorescence intensity and histopathology of diseased tissues.

To address this problem, we applied digital image processing methods in our studies to quantify florescence intensity of ALA fluorescence images to improve diagnostic accuracy to a higher level of sensitivity (98%) and specificity (92%). The initial clinical results on non-invasive diagnosis of early oral cancer in vivo at NCC using a digitised 5-ALA PPIX fluorescence endoscopic imaging system has shown to be promising.

A/Prof Malini Olivo
Principal Investigator
Lab of Photodynamic Diagnosis and Treatment