Background: A study group composed of researchers from across the United States undertook a study to evaluate the sensitivity and specificity of OralCDx (OralScan Laboratories Inc.), a computer-assisted method of analysis of the oral brush biopsy, in the detection of precancerous and cancerous lesions of the oral mucosa.

Methods: The study group conducted a multicenter double-blind study comparing results of OralCDx analysis with those of scalpel biopsy of suspicious oral lesions, as well as using OralCDx on oral lesions that appeared benign clinically.

Results: In 945 patients, OralCDx independently detected every case of histologically confirmed oral dysplasia and carcinoma (sensitivity = 100 percent, false-negative rate = 0 percent). Every OralCDx "positive" result was subsequently confirmed by histology as dysplasia or carcinoma. The specificity for the OralCDx "positive" result was 100 percent, while the specificity for the OralCDx "atypical" results was 92.9 percent. In 4.5 percent of clinically benign-appearing lesions that would not have received additional testing or attention other than clinical follow-up, OralCDx uncovered dysplasia or carcinoma (statistical sensitivity > 96 percent, P < .05, n = 131; statistical specificity for the OralCDx "positive" result > 97 percent and for the "atypical" result > 90 percent, P < .05, n = 196).

Conclusions: The authors propose that this multicenter trial demonstrates that OralCDx is a highly accurate method of detecting oral precancerous and cancerous lesions. OralCDx can aid in confirming the nature of apparently benign oral lesions and, more significantly, revealing those that are precancerous and cancerous when they are not clinically suspected of being so. All OralCDx "atypical" and "positive" results should be referred for scalpel biopsy and histology to completely characterize the lesion.

Clinical implications: Given the difficulty in clinically differentiating premalignant and malignant lesions from benign lesions with a similar appearance, OralCDx appears to determine the significance of an oral lesion definitively and detect innocuous-appearing oral cancers at early, curable stages.