Applicability of preoperative nuclear morphometry to evaluating risk for cervical lymph node metastasis in oral squamous cell carcinoma

A preoperative nomogram model for the prediction of lymph node metastasis in buccal mucosa cancer

OBJECTIVES

We sought to construct a nomogram model predicting lymph node metastasis (LNM) in patients with squamous cell carcinoma of the buccal mucosa based on preoperative clinical characteristics.

METHODS

Patients who underwent radical resection of a primary tumor in the buccal mucosa with neck dissection were enrolled. Clinical characteristics independently associated with LNM in multivariate analyses were adopted to build the model. Patients at low risk of LNM were defined by a predicted probability of LNM of less than 5%.

RESULTS

Patients who underwent surgery in an earlier period (January 2015-November 2019) were defined as the model development cohort (n = 325), and those who underwent surgery later (November 2019-March 2021) were defined as the validation cohort (n = 140). Age, tumor differentiation, tumor thickness, and clinical N stage assessed by computed tomography/magnetic resonance imaging (cN) were independent predictors of LNM. The nomogram model based on these four predictors showed good discrimination accuracy in both the model development and validation cohorts, with areas under the receiver-operating characteristic curve (AUC) of 0.814 and 0.828, respectively. LNM prediction by the nomogram model was superior to cN in AUC comparisons (0.815 vs. 0.753) and decision curve analysis of the whole cohort. Seventy-one patients were defined as having a low risk of LNM, among whom the actual metastasis rate was only 1.4%.

CONCLUSIONS

A robust nomogram model for preoperative LNM prediction is built.

Quantitative nuclear phenotype signatures predict nodal disease in oral squamous cell carcinoma

BACKGROUND

Early-stage oral squamous cell carcinoma (OSCC) patients have a one-in-four risk of regional metastasis (LN+), which is also the most significant prognostic factor for survival. As there are no validated biomarkers for predicting LN+ in early-stage OSCC, elective neck dissection often leads to over-treatment and under-treatment. We present a machine-learning-based model using the quantitative nuclear phenotype of cancer cells from the primary tumor to predict the risk of nodal disease.

METHODS AND FINDINGS

Tumor specimens were obtained from 35 patients diagnosed with primary OSCC and received surgery with curative intent. Of the 35 patients, 29 had well (G1) or moderately (G2) differentiated tumors, and six had poorly differentiated tumors. From each, two consecutive sections were stained for hematoxylin & eosin and Feulgen-thionin staining. The slides were scanned, and images were processed to curate nuclear morphometric features for each nucleus, measuring nuclear morphology, DNA amount, and chromatin texture/organization. The nuclei (n = 384,041) from 15 G1 and 14 G2 tumors were randomly split into 80% training and 20% test set to build the predictive model by using Random Forest (RF) analysis which give each tumor cell a score, NRS. The area under ROC curve (AUC) was 99.6% and 90.7% for the training and test sets, respectively. At the cutoff score of 0.5 as the median NRS of each region of interest (n = 481), the AUC was 95.1%. We then developed a patient-level model based on the percentage of cells with an NRS ≥ 0.5. The prediction performance showed AUC of 97.7% among the 80% (n = 23 patient) training set and with the cutoff of 61% positive cells achieved 100% sensitivity and 91.7% specificity. When applying the 61% cutoff to the 20% test set patients, the model achieved 100% accuracy.

CONCLUSIONS

Our findings may have a clinical impact with an easy, accurate, and objective biomarker from routine pathology tissue, providing an unprecedented opportunity to improve neck management decisions in early-stage OSCC patients.

Role of nuclear morphometry in diagnosis of salivary gland neoplasms

BACKGROUND

Fine needle aspiration cytology is an important diagnostic tool in salivary gland cytopathology. However, cytology of salivary gland neoplasms shows diversity with morphological overlap, heterogeneity and subjective variability, posing diagnostic challenge. Nuclear cytomorphometric image analysis turns the subjective criteria into quantitative parameters, which may aid not only in differentiating between benign and malignant tumours but also can aid in categorising challenging cases with morphological overlap.

METHODS

A retrospective analysis of Papanicolaou stained fine needle aspirate smears from 50 patients with salivary gland neoplasms, with available follow-up histopathology, was performed. The aspirates were categorised in to three neoplastic categories of Milan system for reporting salivary gland cytology (MSRSGC) that is benign neoplasm, salivary gland neoplasm with uncertain malignant potential, and malignant. The smears were subjected to cytomorphometric image analysis using Image Pro Software (version 10). Each nucleus was analysed for 13 parameters including aspect ratio, intensity, diameter, perimeter, roundness, area, Feret diameter, circularity, radii, Fournier description and chromatin texture characteristics such as margination, heterogeneity and clumpiness. The data obtained was compared through relevant statistical methods.

RESULTS

The comparative morphometric data obtained for parameters using ANOVA, for the three categories showed statistically significant P values (<0.001). The post hoc test (Games-Howel) test also showed significant P values (<.001).

CONCLUSION

Cytomorphometric image analysis is an important tool in diagnosis of salivary gland neoplasms and can improve the diagnostic accuracy for better treatment and improved prognosis. However, more studies with larger sample size may be needed in future for further affirming its role.

Ordinal logistic regression model describing factors associated with extent of nodal involvement in oral cancer patients and its prospective validation

Background

Oral cancer is the most common cancer among Indian men, and has strong tendency of metastatic spread to neck lymph node which strongly influences prognosis especially 5 year survival-rate and also guides the related managements more effectively. Therefore, a reliable and accurate means of preoperative evaluation of extent of nodal involvement becomes crucial. However, earlier researchers have preferred to address mainly its dichotomous form (involved/not-involved) instead of ordinal form while dealing with epidemiology of nodal involvement. As a matter of fact, consideration of ordinal form appropriately may increase not only the efficiency of the developed model but also accuracy in the results and related implications. Hence, to develop a model describing factors associated with ordinal form of nodal involvement was major focus of this study.

Methods

The data for model building were taken from the Department of Surgical Oncology, Dr.BRA-IRCH, AIIMS, New Delhi, India. All the OSCC patients (duly operated including neck dissection) and confirmed histopathologically from 1995 to 2013 were included. Further, another data of 204 patients collected prospectively from 2014 to 2015 was considered for the validation of the developed model. To assess the factors associated with extent of nodal involvement, as a first attempt in the field of OSCC, stepwise multivariable regression procedure was used and results are presented as odds-ratio and corresponding 95% confidence interval (CI). For appropriate accounting of ordinal form, the ordinal models were assessed and compared. Also, performance of the developed model was validated on a prospectively collected another data.

Results

Under multivariable proportional odds model, pain at the time of presentation, sub mucous fibrosis, palpable neck node, oral site and degree of differentiation were found to be significantly associated factors with extent of nodal involvement. In addition, tumor size also emerged to be significant under partial-proportional odds model.

Conclusions

The analytical results under the present study reveal that in case of ordinal form of the outcome, appropriate ordinal regression may be a preferred choice. Present data suggest that, pain, sub mucous fibrosis, palpable neck node, oral site, degree of differentiation and tumor size are the most probable associated factors with extent of nodal involvement.

Clinical Spectrum, Pattern, and Level-Wise Nodal Involvement Among Oral Squamous Cell Carcinoma Patients - Audit of 945 Oral Cancer Patient Data

Oral Squamous cell carcinoma (OSCC) is a locoregionally aggressive malignancy. Timely management of neck node dissemination, an important prognostic factor, impacts survival. The aim of the current study was to obtain comprehensive data on patterns or level-wise involvement of neck nodes to optimize neck management in OSCC. It was a retrospective analysis of a prospectively maintained database in a hospital-based setting. The current study evaluated patterns of spread to neck nodes in 945 pathologically proven OSCC patients who underwent neck dissection between 1995 and 2013. Clinical, surgical, pathological, level-wise information of neck nodes was available, and records of these patients were analyzed in relation to the pattern of involvement. Absolute/relative frequency distribution was used to describe the distribution of categorical variables. Continuous measures were organized as mean (standard deviation) and/or median (range). Buccal mucosa (28.78%) was the most common, whereas lip (5.08%) was the least common oral subsite. Modified neck dissection (69.75%) was the most common type of neck dissection. Pathological node positivity was documented in 39.8% patients and Level I(62.54%) and level II(57.33%) are the most common neck levels for nodal involvement. Involvement of Level III to V was seen less often (7.17%). There was no significant association between node positivity among different subsites of oral cancer. Neck level I and II are the most commonly involved levels. Sensitivity and specificity of clinical assessment are 83.51% and 30.05%, respectively. In view of this void in clinical assessment and a predictable nodal spread, alternate node assessment methodology must be explored.

Nuclear PKM2 promotes the progression of oral squamous cell carcinoma by inducing EMT and post-translationally repressing TGIF2

Pyruvate kinase M2 (PKM2), a glycolytic enzyme, acts as a metabolic function leading to an energy production critical for cancer progression, known as Warburg effect. In this study we showed a pivotal role of PKM2 acting as a non-metabolic function to promote cancer cell progression in human oral squamous cell carcinoma (OSCC) through the induction of epithelial-mesenchymal transition (EMT), which is crucial for the potential in cancer cell invasion, and post-translational TGIF2 degradation. PKM2 immunoreaction was strong in the cytoplasm of invasive cancer cells, and distinct in the nucleus of spindle-shaped cancer cells with EMT characteristics. TGIF2 nuclear immunoreaction was seen in dysplastic epithelial cells but was repressed in cancer cells. In vitro analyses, cytoplasmic expression of PKM2 was translocated into the nucleus in human OSCC derived HSC-4 and SAS cells when EMT was stimulated. In addition, nuclear expression of TGIF2 was distinctively repressed in EMT induced HSC-4 and SAS cells. We recognized a mismatch in TGIF2 protein and mRNA expression in EMT induced HSC-4 and SAS cells and found that TGIF2 protein was post-translationally degraded through a ubiquitin proteasome system by an MG132 proteasome inhibition assay. Finally, promotion of HSC-4 and SAS cell progression by PKM2 was recognized in PKM2 knockdown assays. Thus, we clarified a new mechanism of non-metabolic function of PKM2 to promote the progression of OSCC through PKM2 nuclear translocation, subsequently induced EMT, and post-translationally repressed TGIF2 expression by a ubiquitin proteasome system.

Cytological study of DNA content and nuclear morphometric analysis for aid in the diagnosis of high-grade dysplasia within oral leukoplakia

OBJECTIVE

To quantitatively examine the DNA content and nuclear morphometric status of oral leukoplakia (OL) and investigate its association with the degree of dysplasia in a cytologic study.

STUDY DESIGN

Oral cytobrush biopsy was carried out to obtain exfoliative epithelial cells from lesions before scalpel biopsy at the same location in a blinded series of 70 patients with OL. Analysis of nuclear morphometry and DNA content status using image cytometry was performed with oral smears stained with the Feulgen-thionin method.

RESULTS

Nuclear morphometric analysis revealed significant differences in DNA content amount, DNA index, nuclear area, nuclear radius, nuclear intensity, sphericity, entropy, and fractal dimension (all P < .01) between low-grade and high-grade dysplasia. DNA content analysis identified 34 patients with OL (48.6%) with DNA content abnormality. Nonhomogeneous lesion (P = .018) and high-grade dysplasia (P = .008) were significantly associated with abnormal DNA content. Importantly, the positive correlation between the degree of oral dysplasia and DNA content status was significant (P = .004, correlation coefficient = 0.342).

CONCLUSION

Cytology analysis of DNA content and nuclear morphometric status using image cytometry may support their use as a screening and monitoring tool for OL progression.

Advances toward fully automated in vivo assessment of oral epithelial dysplasia by nuclear endomicroscopy-A pilot study

BACKGROUND

Uncertainties in detection of oral epithelial dysplasia (OED) frequently result from sampling error especially in inflammatory oral lesions. Endomicroscopy allows non-invasive, "en face" imaging of upper oral epithelium, but parameters of OED are unknown.

METHODS

Mucosal nuclei were imaged in 34 toluidine blue-stained oral lesions with a commercial endomicroscopy. Histopathological diagnosis showed four biopsies in "dys-/neoplastic," 23 in "inflammatory," and seven in "others" disease groups. Strength of different assessment strategies of nuclear scoring, nuclear count, and automated nuclear analysis were measured by area under ROC curve (AUC) to identify histopathological "dys-/neoplastic" group. Nuclear objects from automated image analysis were visually corrected.

RESULTS

Best-performing parameters of nuclear-to-image ratios were the count of large nuclei (AUC=0.986) and 6-nearest neighborhood relation (AUC=0.896), and best parameters of nuclear polymorphism were the count of atypical nuclei (AUC=0.996) and compactness of nuclei (AUC=0.922). Excluding low-grade OED, nuclear scoring and count reached 100% sensitivity and 98% specificity for detection of dys-/neoplastic lesions. In automated analysis, combination of parameters enhanced diagnostic strength. Sensitivity of 100% and specificity of 87% were seen for distances of 6-nearest neighbors and aspect ratios even in uncorrected objects. Correction improved measures of nuclear polymorphism only. The hue of background color was stronger than nuclear density (AUC=0.779 vs 0.687) to detect dys-/neoplastic group indicating that macroscopic aspect is biased.

CONCLUSIONS

Nuclear-to-image ratios are applicable for automated optical in vivo diagnostics for oral potentially malignant disorders. Nuclear endomicroscopy may promote non-invasive, early detection of dys-/neoplastic lesions by reducing sampling error.

Diagnostic accuracy of oral cancer cytology in a pilot study

BACKGROUND

Recently, cytology has been applied to the diagnosis of oral lesions. We aimed to explore the diagnostic accuracy of oral cytology based on the histological diagnosis.

METHODS

Histological diagnoses of 327 cases were classified as Negative, Borderline lesion -, Borderline lesion +, oral intraepithelial neoplasia/carcinoma in situ (OIN/CIS), or Positive. Cytological diagnoses were classified as NILM (negative for intraepithelial lesion or malignancy), LSIL (low-grade squamous intraepithelial lesion), HSIL (high-grade squamous intraepithelial lesion), or SCC (squamous cell carcinoma). The cytology slides were evaluated by 10 raters and the results were compared with the histology results.

RESULTS

In 142 cases that were histologically negative, the number of NILM, LSIL, HSIL, and SCC and other malignancy was 77 (54.2%), 47 (34.3%), 8 (5.6%), and 10 (7.0%), respectively. Among 32 cases of Borderline lesion -, the number of NILM, LSIL, HSIL, and SCC and other malignancy was 11 (34.3%), 11 (34.3%), 9 (28.1%), and 1 (3.1%), respectively. Also, in 4 cases of Borderline lesion +, the number of NILM, LSIL, HSIL, and SCC and other malignancy was 2 (50.0%), 0 (0.0%), 0 (0.0%), 2 (50.0%), respectively. Among 12 cases of OIN/CIS, the number of NILM, LSIL, HSIL, and SCC and other malignancy was 1 (8.3%), 2 (16.7%), 4 (33.3%), and 5 cases (41.7%), respectively. Among 137 cases with a histological diagnosis of Positive, the number of NILM, LSIL, HSIL, and SCC and other malignancy was 7 (5.1%), 22 (16.1%), 19 (13.9%), and 89 (65.0%), respectively. Sensitivity, specificity, and positive predictive and negative predictive values were 93.5, 50.6, 62.4, and 89.8%, respectively, when the cytological diagnosis of Negative was assumed to be NILM; they were 77.8, 83.9, 81.0 and 81.1%, respectively, if the cytological diagnosis of Negative was assumed to be NILM and LSIL. The number of false-positive and false-negative diagnosis affected cases with LSIL and HSIL may indicate the difficulty in the cytological diagnosis of borderline lesions. While the negative predictive value was relatively high (89.8%) when cytological Negative was assumed to be NILM only.

CONCLUSION

Histopathological examination should be recommended in cases with cytological diagnoses of LSIL, HSIL, and SCC.