Search

Background: Cell-in-cell structures (caused by cell cannibalistic activity) have been related to prognosis of many cancers. This is the first multi-institutional study to assess the prognostic impact of cell-in-cell structures in a large cohort of early oral tongue squamous cell carcinomas (OTSCC).Methods: A total of 308 cases from five Finnish University Hospitals and from the A.C. Camargo Cancer Center, São Paulo, Brazil, were included in this study. Cell-in-cell structures were evaluated on surgical postoperative sections that stained with hematoxylin and eosin staining.Results: We found that cell-in-cell structures associated with cancer-related mortality in univariable analysis with a hazard ratio (HR) of 2.99 (95%CI 1.52-5.88; P = 0.001). This association was confirmed in multivariable analysis (HR 2.22, 95%CI 1.12-4.44; P = 0.024). In addition, statistically significant associations were observed between the cell-in-cell structures and other adverse histopathologic characteristics including deep invasion (P <  0.001), high index of tumor budding (P = 0.007), worst pattern of invasion (P <  0.001), perineural invasion (P = 0.01), and stroma-rich pattern (P = 0.001).Conclusions: Our findings demonstrate a significant relationship between cell-in-cell formation and aggressive characteristics of early OTSCC. Cell-in-cell structures have a distinct impact as a novel prognostic indicator in early OTSCC and they can be easily assessed during routine pathology practice. [ABSTRACT FROM AUTHOR]

Background: The proper estimate of the risk of recurrences in early-stage oral tongue squamous cell carcinoma (OTSCC) is mandatory for individual treatment-decision making. However, this remains a challenge even for experienced multidisciplinary centers.Objectives: We compared the performance of four machine learning (ML) algorithms for predicting the risk of locoregional recurrences in patients with OTSCC. These algorithms were Support Vector Machine (SVM), Naive Bayes (NB), Boosted Decision Tree (BDT), and Decision Forest (DF).Materials and Methods: The study cohort comprised 311 cases from the five University Hospitals in Finland and A.C. Camargo Cancer Center, São Paulo, Brazil. For comparison of the algorithms, we used the harmonic mean of precision and recall called F1 score, specificity, and accuracy values. These algorithms and their corresponding permutation feature importance (PFI) with the input parameters were externally tested on 59 new cases. Furthermore, we compared the performance of the algorithm that showed the highest prediction accuracy with the prognostic significance of depth of invasion (DOI).Results: The results showed that the average specificity of all the algorithms was 71% . The SVM showed an accuracy of 68% and F1 score of 0.63, NB an accuracy of 70% and F1 score of 0.64, BDT an accuracy of 81% and F1 score of 0.78, and DF an accuracy of 78% and F1 score of 0.70. Additionally, these algorithms outperformed the DOI-based approach, which gave an accuracy of 63%. With PFI-analysis, there was no significant difference in the overall accuracies of three of the algorithms; PFI-BDT accuracy increased to 83.1%, PFI-DF increased to 80%, PFI-SVM decreased to 64.4%, while PFI-NB accuracy increased significantly to 81.4%.Conclusions: Our findings show that the best classification accuracy was achieved with the boosted decision tree algorithm. Additionally, these algorithms outperformed the DOI-based approach. Furthermore, with few parameters identified in the PFI analysis, ML technique still showed the ability to predict locoregional recurrence. The application of boosted decision tree machine learning algorithm can stratify OTSCC patients and thus aid in their individual treatment planning. [ABSTRACT FROM AUTHOR]

Stromal categorization has been used to classify many epithelial cancer types. We assessed the desmoplastic reaction and compared its significance with other stromal characteristics in early (cT1-2N0) oral tongue squamous cell carcinoma (OTSCC). In this multi-institutional study, we included 308 cases treated for early OTSCC at five Finnish university hospitals or at the A.C. Camargo Cancer Center in São Paulo, Brazil. The desmoplastic reaction was classified as immature, intermediate, or mature based on the amount of hyalinized keloid-like collagen and myxoid stroma. We compared the prognostic value of the desmoplastic reaction with a stromal grading system based on tumor-stroma ratio and stromal tumor-infiltrating lymphocytes. We found that a high amount of stroma with a weak infiltration of lymphocytes was associated statistically significantly with a worse disease-free survival with a hazard ratio (HR) of 2.68 (95% CI 1.26-5.69), worse overall survival (HR 2.95, 95% CI 1.69-5.15), and poor disease-specific survival (HR 2.66, 95% CI 1.11-6.33). Tumors having a high amount of stroma with a weak infiltration of lymphocytes were also significantly associated with a high rate of local recurrence (HR 4.13, 95% CI 1.67-10.24), but no significant association was found with lymph node metastasis (HR 1.27, 95% CI 0.37-4.35). Categorization of the stroma based on desmoplastic reaction (immature, intermediate, mature) showed a low prognostic value for early OTSCC in all survival analyses (P > 0.05). In conclusion, categorization of the stroma based on the amount of stroma and its infiltrating lymphocytes shows clinical relevance in early OTSCC superior to categorization based on the maturity of stroma.