Screening tool could help oral cancer detection

Aug 2 2012

By Sarah Guy, MedWire Reporter

Indian researchers report the development of a novel, simple screening tool with a high degree of accuracy for the detection of oral premalignancies that can be used in a general health setting without the need for much training or clinical experience.

They hope the tool could help address the current large variation in detection and diagnosis of potentially fatal oral cancer, which results from a lack of definition and diagnostic criteria for the different oral premalignancies.

The Screening Tool for Oral Premalignancies (STOP) scores the presence of four items - white lesions, white lesions with ulcers, mucosal change, and persistent ulcer - and had almost 100% specificity for detecting oral premalignancies compared with specialist examination.

"Oral cancer poses a heavy burden to the health care delivery system, because of high mortality and morbidity and huge expenses incurred for treatment modalities of advanced cases," explain Valsa Thomas (Medical College, Thiruvananthapuram, Kerala) and colleagues.

And despite the presence of visible changes in the oral mucosa before the development of oral cancer, most cancers are detected at a late stage, says the team in Oral Oncology.

The researchers consulted specialists and positively screened oral cancer patients to compile a list of signs and symptoms of oral premalignancy for STOP, as well as patient medical history and personal details including smoking status.

STOP was tested on a group of 255 individuals, 59 of whom had oral premalignancies as determined by a specialist. A total of 135 individuals had oral symptoms, with change of appearance the most common, and 58% of the whole cohort reported smoking.

Thomas and co-authors found that a score of 4 - the highest positive score, indicating presence of all four STOP features - gave the best sensitivity and specificity, at 96.6% and 99.0% in receiver operator characteristic analysis, respectively, compared with scores of individual items.

Furthermore, STOP's positive and negative predictive values were high, at 96.6% and 99.9%, and its reliability coefficient was 0.874.

Of note, almost three-quarters (71%) of participants who had white lesions were smokers, and 88% of those who showed mucosal change were tobacco chewers. The researchers suggest that preventive intervention at this stage, including advice on habit cessation, could bring about symptom regression.

"STOP is recommended for opportunistic screening in the busy general health care setting and staging can give [a] clinical severity risk alert for patients and doctors," they conclude.

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.