Deep learning-based pathology image analysis predicts cancer progression risk in patients with oral leukoplakia.

BACKGROUND: Oral leukoplakia (OL) is associated with an increased risk for oral cancer (OC) development. Prediction of OL cancer progression may contribute to decreased OC morbidity and mortality by favoring early intervention. Current OL progression risk assessment approaches face large interobserver variability and is weakly prognostic. We hypothesized that convolutional neural networks (CNN)-based histology image analyses could accelerate the discovery of better OC progression risk models. METHODS: Our CNN-based oral mucosa risk stratification model (OMRS) was trained to classify a set of nondysplastic oral mucosa (OM) and a set of OC H&E slides. As a result, the OMRS model could identify abnormal morphological features of the oral epithelium. By applying this model to OL slides, we hypothesized that the extent of OC-like features identified in the OL epithelium would correlate with its progression risk. The OMRS model scored and categorized the OL cohort (n = 62) into high- and low-risk groups. RESULTS: OL patients classified as high-risk (n = 31) were 3.98 (95% CI 1.36-11.7) times more likely to develop OC than low-risk ones (n = 31). Time-to-progression significantly differed between high- and low-risk groups (p = 0.003). The 5-year OC development probability was 21.3% for low-risk and 52.5% for high-risk patients. The predictive power of the OMRS model was sustained even after adjustment for age, OL site, and OL dysplasia grading (HR = 4.52, 1.5-13.7). CONCLUSION: The ORMS model successfully identified OL patients with a high risk of OC development and can potentially benefit OC early diagnosis and prevention policies.

Autofluorescence imaging as a noninvasive tool of risk stratification for malignant transformation of oral leukoplakia: A follow-up cohort study.

BACKGROUND: There is few longitudinal study on oral potentially malignant disorder (OPMDs) focusing on oral leukoplakia (OLK) with clinical endpoints to investigate the role of autofluorescence imaging (AFI) for surveilling the malignant transformation. METHODS: Based on our previous prospective diagnostic study enrolled 517 OPMD patients, 184 OLK patients were retrieved to further investigate the role of AFI using VELscope in predicting malignant transformation. During a median follow-up period of 44 months, 19 (10.33%) developed into oral cancer. RESULTS: OLK patients were divided into loss of autofluorescence (LAF, n = 124) and retention of autofluorescence (RAF, n = 60) groups according to the results of AFI. Interestingly, difference between malignant transformation rate (MTR, 14.52%) of group LAF and overall MTR (10.33%) was not significant, but MTR (1.67%) of group RAF was significantly lower than overall MTR. Kaplan-Meier and Cox-regression analyses revealed that LAF could not directly distinguish the high-risk lesions, but RAF significantly discriminate the low-risk lesions. Importantly, time-dependent ROC curve analysis found that the sensitivity and negative predictive value (NPV) of AFI for the prediction of malignant transformation was 100% and 100% (2-year follow-up), 94.7% and 98.3% (5-year follow-up), respectively. Also, calibration curve and decision curve analyses showed high levels of predictive value and clinical relevance. CONCLUSION: This follow-up cohort study firstly evaluated AFI using VELscope for risk stratification of OLK malignant transformation. Whether conservative management is appropriate for OLK patients with RAF imaging due to minimal rate and risk of malignant transformation and great specificity and NPV is required to be further investigated.

Endoscopic diagnosis value of narrow band imaging Ni classification in vocal fold leukoplakia and early glottic cancer.

OBJECTIVES: To explore the diagnostic value and the correlation between histological diagnosis and the Ni classification under narrow band imaging (NBI) for vocal fold leukoplakia (VFL) and early glottic cancer. METHODS: A total of 91 patients with 119 vocal fold lesions were selected from January 2017 to May 2020. All these patients were subsequently examined by white light imaging (WLI) and NBI endoscopy, and then all lesions were classified by the Ni classification according to the characteristics of intraepithelial papillary capillary loop (IPCL) observed. The gold standard of diagnosis was histopathological results. Eventually, the chi-square and kappa test were applied, respectively, to evaluate the diagnostic value of NBI endoscopy and the consistency of Ni classification and pathological results. RESULTS: The accuracy and sensitivity of NBI endoscopy were significantly higher than that of WLI endoscopy (P < 0.05). For the diagnosis of precancerous lesions under the NBI, the sensitivity, specificity, positive and negative predictive value were 69.6% (16/23), 90.6% (87/96), 64.0% (16/25) and 92.6% (87/94), which for malignant lesions were 84.4% (65/77), 92.9% (39/42), 95.6% (65/68) and 76.5% (39/51). Moreover, for patients with low-grade intraepithelial neoplasia (mild and moderate dysplasia), type IV lesions accounted for the most (69.6 vs 30.4%; χ2 = 36.961, P < 0.001). For high-grade intraepithelial neoplasia or carcinoma in situ, type Va lesions were predominant (χ2 = 30.526, P < 0.001), while type Vb and Vc lesions were dominant in invasive carcinoma (χ2 = 64.373, P < 0.001). Besides, the kappa test revealed that there was a high consistency between Ni classification and pathological diagnosis (Kappa = 0.667, P < 0.001). CONCLUSIONS: The Ni classification can improve the diagnosis accuracy of vocal fold lesions which enables clear visualization of mucosal microvasculature. This is essential for the early diagnosis of VFL and early glottic cancer during routine endoscopic examination.

Theoretical Background to Automated Diagnosing of Oral Leukoplakia: A Preliminary Report.

Oral leukoplakia represents the most common oral potentially malignant disorder, so early diagnosis of leukoplakia is important. The aim of this study is to propose an effective texture analysis algorithm for oral leukoplakia diagnosis. Thirty-five patients affected by leukoplakia were included in this study. Intraoral photography of normal oral mucosa and leukoplakia were taken and processed for texture analysis. Two features of texture, run length matrix and co-occurrence matrix, were analyzed. Difference was checked by ANOVA. Factor analysis and classification by the artificial neural network were performed. Results revealed easy possible differentiation leukoplakia from normal mucosa (p < 0.05). Neural network discrimination shows full leukoplakia recognition (sensitivity 100%) and specificity 97%. This objective analysis in the neural network revealed that involving 3 textural features into optical analysis of the oral mucosa leads to proper diagnosis of leukoplakia. Application of texture analysis for leukoplakia is a promising diagnostic method.

[Application of intraoral ultrasonic imaging in diagnosis and treatment of 18 patients with oral leukoplakia in non-masticatory mucosa].

PURPOSE: To explore the sonographic appearance of leukoplakia in non-masticatory oral mucosa, classifying mucosal leukoplakia according to the characteristics of sonogram, and providing reference for clinical diagnosis and treatment. METHODS: Eighteen patients (24 lesions) were diagnosed as oral leukoplakia at the Department of Oral Mucosal Diseases, Shanghai Ninth People's Hospital. The lesions were located in the tongue, floor of mouth, buccal mucosa and libial mucosa. Before the biopsy was taken, intra-oral path ultrasound was performed at the Department of Ultrasound to observe the lesion's extent, continuity, presence or absence of keratinization, the thickness of each layer in the epithelium, and color doppler flow imaging of the lesions. Quantitative analysis software 'Qontraxt' was used to randomly measure the relative echo intensity of the mucosal surface in leukoplakia areas, and summarize the keratinization type. SPSS 25.0 software package was used for statistical analysis of the data, and paired t test was used for inter-group comparison of the data. RESULTS: Oral leukoplakia sonograms showed that the epithelial layer appeared keratinization, the epithelial was thickened, and the echo was enhanced. The stratum intermedium showed a low echo thickening band, and the echo of partial lesions' surface decreased or the blood flow signal in oral mucosa increased. The hyperechoic band in the leukoplakia area was significantly thickened (P<0.001), and the echo was enhanced, with the tongue and buccal mucosa being the most significant. The hypoechoic band was significantly thicker (P<0.001), with the buccal mucosa and labial mucosa being the most significant. The surface and stratum corneum echo intensity values were determined by Qontraxt quantitative analysis software to determine whether there were keratinization and the keratinization types. The echo intensity values was 43.28±9.33 in non-OLK area, 92.88±3.12 in OLK with orthokeratosis, and 84.75±5.76 in OLK with parakeratosis. CONCLUSIONS: Ultrasound imaging can effectively define mucosal leukoplakia and measure the thickness of each layer in the epithelium. In addition, special adjoint changes such as ulcers, infections and cancerous changes can be detected. Intraoral ultrasonic imaging can provide imaging evidence for clinical diagnosis, treatment planning and post-treatment follow-up and contribute to avoid unnecessary mucosal iatrogenic injury or recurrence of disease after treatment.

Validity of narrow band imaging in the detection of oral and oropharyngeal malignant lesions: A systematic review and meta-analysis.

BACKGROUND: This meta-analysis describes the validity of narrow band imaging (NBI) in the assessment of suspicious oral lesions. METHODS: Medline, EMBASE, and Scopus were searched for trial studies comparing NBI with conventional modalities in the oral cavity (OC) and oropharynx (OP) for the detection of dysplastic and malignant change. RESULTS: Seven studies were found, and generally supported the utility of NBI in different clinical settings, although there were exceptions. Pooled data from 4 studies of NBI validity demonstrated high summary specificity and sensitivity for a wide range of suspicious lesions of the OC or OP (75.7% with 95% CI 65.1%-83.9%, and 91.5% with 95% CI 81.8%-96.3%, respectively). Summary positive likelihood ratio (LR+) was 8.91 (95% CI 4.1-19.6) and 0.27 (95% CI 0.18-0.39), respectively. CONCLUSIONS: NBI is a promising diagnostic and surveillance tool for suspicious lesions in the OC or OP; however, higher powered studies will define precise NBI criterion and clinical recommendations.

Noninvasive diagnostic adjuncts for the evaluation of potentially premalignant oral epithelial lesions: current limitations and future directions.

Potentially premalignant oral epithelial lesions (PPOELs) are a group of clinically suspicious conditions, of which a small percentage will undergo malignant transformation. PPOELs are suboptimally diagnosed and managed under the current standard of care. Dysplasia is the most well-established marker to distinguish high-risk PPOELs from low-risk PPOELs, and performing a biopsy to establish dysplasia is the diagnostic gold standard. However, a biopsy is limited by morbidity, resource requirements, and the potential for underdiagnosis. Diagnostic adjuncts may help clinicians better evaluate PPOELs before definitive biopsy, but existing adjuncts, such as toluidine blue, acetowhitening, and autofluorescence imaging, have poor accuracy and are not generally recommended. Recently, in vivo microscopy technologies, such as high-resolution microendoscopy, optical coherence tomography, reflectance confocal microscopy, and multiphoton imaging, have shown promise for improving PPOEL patient care. These technologies allow clinicians to visualize many of the same microscopic features used for histopathologic assessment at the point of care.

[Diagnostic value of optical coherence tomography for oral leukoplakia assessment]. / Diagnosticheskaia tsennost' metoda kogerentnoi tomografii u bol'nykh s leikoplakiei slizistoi obolochki rta.

The article presents a comparative analysis of the results of in vivo scanning of oral mucosa leukoplakia with optical coherence tomography (OCT) method and pathological studies in 15 patients (9 females and 6 males aged 27-82). The study revealed specific OCT signs of leukoplakia: even disorder of tissue layering and less clear visualization of lamina propria. OCT is an important stage in the differential diagnosis of leukoplakia of the oral mucosa and significantly complements the clinical and pathomorphological study.

Correlation analysis of oral lesion sizes by various standardized criteria.

OBJECTIVE: Requirements of an NCI contract examining a novel treatment for leukoplakia were to compare standard bi-dimensional measurement of oral lesions to examine for correlation with Response Evaluation Criteria in Solid Tumors (RECIST) criteria, and to examine the feasibility of digital image analysis for automated measurements. STUDY DESIGN: Retrospective review. METHODS: We examined 13 patients by bi-dimensional measurement and compared these measurements to 1) RECIST criteria, 2) scalar digital measurements using a standardized measuring device within the photograph, and 3) pixel number. RESULTS: RECIST criteria correlated (r-squared=0.8535, p<0.0001) with bi-dimensional measurements. Digitized measures in photographs correlated with bi-dimensional measurements (r-squared=0.6661, p=0.0007), but were time consuming. There was minimal to no correlation between pixel number in Adobe Photoshop and the other measures. CONCLUSION: Bi-dimensional measurement of oral leukoplakia and RECIST criteria are highly correlated. Digital photography measurements, though highly correlative, are very cumbersome. We recommend bi-dimensional or longest length measurement and a simple photograph as standard of documentation for leukoplakia lesions.

Assessment of oral mucosal lesions with autofluorescence imaging and reflectance spectroscopy.

BACKGROUND: The aim of this prospective study was to evaluate the efficacy of a new form of autofluorescence imaging and tissue reflectance spectroscopy (Identafi, DentalEZ) in examining patients with oral mucosal lesions. METHODS: The authors examined 88 patients with 231 oral mucosal lesions by conventional oral examination (COE) using white-light illumination and ×2.5 magnification loupes, followed by examination using Identafi. The authors noted fluorescence visualization loss, the presence of blanching, and diffuseness of vasculature. They performed incisional biopsies to provide definitive histopathologic diagnosis. RESULTS: Identafi's white light produced lesion visibility and border distinctness equivalent to COE. Identafi's violet light displayed a sensitivity of 12.5% and specificity of 85.4% for detection of oral epithelial dysplasia (OED). The authors noted visible vasculature using the green-amber light in 40.9% of lesions. CONCLUSIONS: Identafi's intraoral white light provided detailed visualization of oral mucosal lesions comparable with examination using an extraoral white-light source with magnification. A high level of clinical experience is required to interpret the results of autofluorescence examination as the violet light displayed low sensitivity for detection of OED. The green-amber light provided additional clinical information in relation to underlying vasculature and inflammation of lesions. PRACTICAL IMPLICATIONS: Examination using Identafi can provide clinicians with more clinical data than a standard COE with yellow incandescent light, but the clinical and optical findings should be interpreted as a whole and not in isolation. Clinicians should use the light features of Identafi in a sequential and differential manner.

Computer aided morphometric analysis of oral leukoplakia and oral squamous cell carcinoma.

We compared the changes in the cells in the basal layer of normal mucosa, oral leukoplakia with dysplasia and different grades of oral squamous cell carcinoma (OSCC) using computer aided image analysis of tissue sections. We investigated three morphometric parameters: nuclear area (NA), cell area (CA) and their ratio (NA:CA). NA and NA:CA ratio showed a statistically significant increase from dysplasia to increasing grades of OSCC. Nuclear size was useful for differentiating normal tissue, potentially malignant leukoplakia and OSCC.

Exophytic oral verrucous hyperplasia: a new entity.

OBJECTIVE: Exophytic oral verrucous hyperplasia (OVH) is a new entity described by an expert working group from South Asia. First reported in Taiwan, there are no reports so far from an Indian population. The aim was to use the microscopic features described by the expert group to differentiate OVH from other oral verruco-papillary lesions in an Indian archive. MATERIALS AND METHODS: In a retrospective multicentre study, using pathology archives, 188 verruco-papillary lesions were retrieved from pathology archives. A proforma listing histopathological criteria for OVH based on published guidelines (Annals of Dentistry, University of Malaya, 2013) was used. Patients' demographic and clinical data were transcribed from patient charts. The Pearson chi-square test was used to determine associations between clinical and histopathological features. RESULTS: Of 188 oral verruco-papillary lesions that were evaluated, based on microscopic features the cases were reclassified as OVH (57), verrucous carcinoma (VC) (84), oral squamous cell carcinoma (16), and other verruco-papillary lesions (31). Both OVH (70%) and VC (60%) showed male predominance and commonly affected buccal mucosa (OVH 74% and VC 57%). Absence of downward growth of the hyperplastic epithelium into lamina propria when compared with the level of the basement membrane of the adjacent normal epithelium was a distinct feature in OVH. Keratin plugging, epithelial dysplasia and subepithelial lymphocytic infiltration were found to be significantly different (P < 0.05) in OVH versus VC. The sample size of other verruco-papillary lesions was insufficient for statistical comparison. CONCLUSION: Apart from the absence of an endophytic growth pattern in OVH, we noted the presence of dysplasia in OVH. This significant observation does institute a debate as to whether this enigmatic lesion could possibly be a precedent of oral squamous or verrucous carcinoma. We propose OVH is a distinct entity in our Indian population and should be considered in the classification of oral potentially malignant disorders.

Toluidine blue aids in detection of dysplasia and carcinoma in suspicious oral lesions.

OBJECTIVES: Accurate clinical identification of 'higher-risk' oral premalignant lesions or 'higher-risk' areas within lesions is important. Assessment methods that predict their presence have great utility. SUBJECTS AND METHODS: A cross-sectional, observational study enrolled a consecutive sample of consenting patients diagnosed with oral leukoplakia, erythroleukoplakia, or erythroplakia. Medical history, visual oral examination, ViziLite(®) examination, toluidine blue staining (TBlue(®) ), and finally a biopsy were completed in a single clinic visit. Seventy-seven of 100 examined lesions in 43 patients were biopsied. Sensitivity, specificity, and positive and negative predictive values were computed for visual examination, ViziLite(®) , and TBlue(®) using biopsy results as the gold standard. RESULTS: The sensitivity of TBlue(®) in detecting high-risk lesions (carcinoma in situ or carcinoma) was 94 (71-100, P < 0.0003) and specificity 45 (32-58, P < 0.53), while for carcinoma, sensitivity was 100 (54-100, P < 0.032) and specificity 39 (28-52, P < 0.097). The results of ViziLite(®) testing either by itself or in combination with the information from toluidine blue testing revealed low sensitivity for the detection of high-risk lesions. CONCLUSIONS: Clinical examination of leukoplakia, erythroplakia, or erythroleukoplakia lesions combined with toluidine blue staining may aid in the identification of severe dysplasia (carcinoma in situ) or carcinoma. This may help in determining whether, when, and where (the site within a lesion) a biopsy should be taken.

Doppler sonography of the deep lingual artery.

PURPOSE: To clarify the Doppler sonographic features of the lingual artery in normal subjects and to evaluate those of patients with cancer of the tongue. MATERIAL AND METHODS: Sixty-seven volunteers and 12 patients with cancer and/or leukoplakia of the tongue were examined with an intraoral sonographic probe. The visibility of the deep lingual artery was determined on transverse and anteroposterior images. On the transverse images, the vascular index, which was defined as the number of colored pixels, was measured on bilateral lingual arteries. Thereafter, the degree of symmetry was evaluated for normal subjects and patients. RESULTS: In normal subjects, between younger and older volunteers, there were no significant differences in visibility of the trunk but differences were found between the two groups for the dorsal branches. The vascular indices of the right and left sides were not different. The characteristic Doppler sonographic feature was vasculature in and around the tumors in the patients with cancer of the tongue. The symmetry indices of the cancer patients were significantly different from those of normal subjects. CONCLUSION: Doppler sonography should be an important procedure for evaluation of tongue neoplasms.

[Study of cellular kinetics in oral leukoplakia]. / Studio della cinetica cellulare delle leukcoplachie orali.

The Authors aim was to study the Labeling Indices (LI) of 12 subjects with oral leukoplakia, using an in vitro labeling technique with tritiated thymidine (Silvestrini-Kit, Ribbon spa, Milan). Twelve volunteers of our Dental Clinic were enrolled in this study as a control group. The kinetic parameter was investigated by counting cells from autoradiographs on histological selections of the biopsies. By a Wilcoxon test, significant differences were found between the LI of the two groups: p = 0.00003. Finally, a comparison has also been made between LI of these lesions and the LI of oral squamous cell carcinomas.