Oral leukoplakia is a potentially malignant lesion of the oral cavity, for which no effective treatment is available. We investigated the effectiveness of curcumin, a potent inhibitor of NF-κB/COX-2, molecules perturbed in oral carcinogenesis, to treat leukoplakia. Subjects with oral leukoplakia (n = 223) were randomized (1:1 ratio) to receive orally, either 3.6 g/day of curcumin (n = 111) or placebo (n = 112), for 6 months. The primary endpoint was clinical response obtained by bi-dimensional measurement of leukoplakia size at recruitment and 6 months. Histologic response, combined clinical and histologic response, durability and effect of long-term therapy for an additional six months in partial responders, safety and compliance were the secondary endpoints. Clinical response was observed in 75 (67.5%) subjects [95% confidence interval (CI), 58.4–75.6] in the curcumin and 62 (55.3%; 95% CI, 46.1–64.2) in placebo arm (P = 0.03). This response was durable, with 16 of the 18 (88.9%; 95% CI, 67.2–96.9) subjects with complete response in curcumin and 7 of 8 subjects (87.5%) in placebo arm, demonstrating no relapse after 6 months follow-up. Difference in histologic response between curcumin and placebo was not significant (HR, 0.88, 95% CI, 0.45–1.71; P = 0.71). Combined clinical and histologic response assessment indicated a significantly better response with curcumin (HR, 0.50; 95% CI, 0.27–0.92; P = 0.02). Continued therapy, in subjects with partial response at 6 months, did not yield additional benefit. The treatment did not raise any safety concerns. Treatment of oral leukoplakia with curcumin (3.6 g for six months), thus was well tolerated and demonstrated significant and durable clinical response for 6 months. Cancer Prev Res; 9(8); 683–91. ©2016 AACR.

Early detection of oral cancer necessitates a minimally invasive, tissue-specific diagnostic tool that facilitates screening/surveillance. Brush biopsy, though minimally invasive, demands skilled cyto-pathologist expertise. In this study, we explored the clinical utility/efficacy of a tele-cytology system in combination with Artificial Neural Network (ANN) based risk-stratification model for early detection of oral potentially malignant (OPML)/malignant lesion. A portable, automated tablet-based tele-cytology platform capable of digitization of cytology slides was evaluated for its efficacy in the detection of OPML/malignant lesions (n = 82) in comparison with conventional cytology and histology. Then, an image pre-processing algorithm was established to segregate cells, ANN was trained with images (n = 11,981) and a risk-stratification model developed. The specificity, sensitivity and accuracy of platform/ stratification model were computed, and agreement was examined using Kappa statistics. The tele-cytology platform, Cellscope, showed an overall accuracy of 84–86% with no difference between tele-cytology and conventional cytology in detection of oral lesions (kappa, 0.67–0.72). However, OPML could be detected with low sensitivity (18%) in accordance with the limitations of conventional cytology. The integration of image processing and development of an ANN-based risk stratification model improved the detection sensitivity of malignant lesions (93%) and high grade OPML (73%), thereby increasing the overall accuracy by 30%. Tele-cytology integrated with the risk stratification model, a novel strategy established in this study, can be an invaluable Point-of-Care (PoC) tool for early detection/screening in oral cancer. This study hence establishes the applicability of tele-cytology for accurate, remote diagnosis and use of automated ANN-based analysis in improving its efficacy.

Deep learning has become a powerful tool in solving image classification problems¹ and has been widely used in medical image analysis.² Convolutional neural networks (CNNs) show the potential for automated classification of different cancer lesions, such as lung cancer,³ skin cancer,⁴ and breast cancer.⁵ But the lack of interpretability and explainability makes CNNs less understandable.⁶ To push the envelope of CNN-based medical diagnostic methods, more rigorous neural network models and training methods must be developed to provide both visually interpretable and explainable analysis.^{7 – 9}