1
|
Sari EF, Johnson NW, McCullough MJ, Cirillo N. Prevalence and risk factors of oral potentially malignant disorders in Indonesia: a cross-sectional study undertaken in 5 provinces. Sci Rep 2024; 14:5232. [PMID: 38433259 PMCID: PMC10909850 DOI: 10.1038/s41598-024-54410-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/27/2023] [Accepted: 02/12/2024] [Indexed: 03/05/2024] Open
Abstract
Detection of subjects with oral potentially malignant disorders in a population is key to early detection of oral cancer (OC) with consequent reduction of cancer-related morbidity and mortality. Our aim was to investigate the prevalence and associated risk factors for OPMD in representative provinces of Indonesia. This cross-sectional study was undertaken in five Indonesian provinces: West Java (WJ), Jakarta (JKT), West Papua (WP), West Kalimantan (WK) and Banda Aceh (BA). Respondents answered a previously validated questionnaire including information on ethnicity, occupation, socioeconomic status (SES), oral health practices, and behaviours associated with oral cancer. An oral examination was undertaken using WHO standardized methodology. Data were analysed using ANOVA, Chi-Square, and logistic regression to assess association between risk factors and mucosal disease. A total of 973 respondents between the ages of 17 and 82 years was enrolled (WJ 35.5%,JKT 13.3% WP 18.3%, WK 9%, BA 23.9%). Tobacco smoking (14.8%), Betel quid (BQ) chewing (12.6%) and alcohol drinking (4%) varied geographically. A well-established OPMD was detected in 137 (14.1%) respondents and 2 (0.2%) presented with chronic ulceration later diagnosed as OC. Leukoplakia was the most common OPMD found (9.7%), while the prevalence of oral submucous fibrosis (OSMF), not previously described in the nation, was 2.3%. Poor knowledge of OC risk factors, poor oral hygiene behaviours, low-income SES and ethnicity were significantly associated with the presence of an OPMD. There is a previously under-reported high prevalence of OPMD in Indonesia. Overall, we found a strong correlation between the presence of an OPMD and individual habituation to known risk factors.
Collapse
Affiliation(s)
- Elizabeth Fitriana Sari
- Dentistry and Oral Health Discipline, Department of Rural Clinical Science, La Trobe Rural Health School, Bendigo, 3552, Australia.
- Faculty of Dentistry, Universitas Padjadjaran, 45363, Bandung, Indonesia.
| | - Newell W Johnson
- Menzies Health Institute QueenslandSchool of Medicine and Dentistry, Griffith University, Gold Coast, QLD, Australia
- Faculty of Dentistry Oral and Craniofacial Sciences, King's College London, London, UK
| | - Michael John McCullough
- Melbourne Dental School, The University of Melbourne, 720, Swanston Street, Carlton, VIC, 3053, Australia
| | - Nicola Cirillo
- Melbourne Dental School, The University of Melbourne, 720, Swanston Street, Carlton, VIC, 3053, Australia.
| |
Collapse
|
2
|
Tran Q, Maddineni S, Arnaud EH, Divi V, Megwalu UC, Topf MC, Sunwoo JB. Oral cavity cancer in young, non-smoking, and non-drinking patients: A contemporary review. Crit Rev Oncol Hematol 2023; 190:104112. [PMID: 37633348 PMCID: PMC10530437 DOI: 10.1016/j.critrevonc.2023.104112] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/12/2023] [Revised: 08/11/2023] [Accepted: 08/23/2023] [Indexed: 08/28/2023] Open
Abstract
Oral squamous cell carcinoma (OSCC) in non-smoking and non-drinking (NSND) individuals appears to be distinct from the traditional head and neck squamous cell carcinoma (HNSCC). The incidence of this subset is increasing, as are the number of studies examining its characteristics. NSND OSCC individuals tend to be younger (<45 years) compared to traditional HNSCC patients. The proportion of females in the NSND OSCC cohort is also higher. The tongue is the predominantly affected subsite. Studies have revealed several gene mutations and unique epigenomic profiles but no definitive genetic etiology. Transcriptomic analysis has not found any causative viral agents. Other proposed etiologies include chronic dental trauma, microbiome abnormalities, marijuana consumption, and genetic disorders. There are international efforts to determine the relative prognostic outcome of this unique cohort, but no consensus has been reached. Here, we review the incidence, demographics, subsite, possible etiologies, prognosis, and therapy implications of the NSND OSCC cohort.
Collapse
Affiliation(s)
- Quan Tran
- Department of Medicine, Stanford University School of Medicine, Palo Alto, CA, USA
| | - Sainiteesh Maddineni
- Department of Otolaryngology - Head and Neck Surgery, Stanford University, Palo Alto, CA, USA
| | - Ethan Hunter Arnaud
- Department of Otolaryngology - Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Vasu Divi
- Department of Otolaryngology - Head and Neck Surgery, Stanford University, Palo Alto, CA, USA
| | - Uchechukwu C Megwalu
- Department of Otolaryngology - Head and Neck Surgery, Stanford University, Palo Alto, CA, USA
| | - Michael C Topf
- Department of Otolaryngology - Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, TN, USA.
| | - John B Sunwoo
- Department of Otolaryngology - Head and Neck Surgery, Stanford University, Palo Alto, CA, USA.
| |
Collapse
|
3
|
Adeoye J, Zheng LW, Thomson P, Choi SW, Su YX. Explainable ensemble learning model improves identification of candidates for oral cancer screening. Oral Oncol 2023; 136:106278. [PMID: 36525782 DOI: 10.1016/j.oraloncology.2022.106278] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/04/2022] [Revised: 11/26/2022] [Accepted: 12/06/2022] [Indexed: 12/15/2022]
Abstract
OBJECTIVES Artificial intelligence could enhance the use of disparate risk factors (crude method) for better stratification of patients to be screened for oral cancer. This study aims to construct a meta-classifier that considers diverse risk factors to identify patients at risk of oral cancer and other suspicious oral diseases for targeted screening. MATERIALS AND METHODS A retrospective dataset from a community oral cancer screening program was used to construct and train the novel voting meta-classifier. Comprehensive risk factor information from this dataset was used as input features for eleven supervised learning algorithms which served as base learners and provided predicted probabilities that are weighted and aggregated by the meta-classifier. Training dataset was augmented using SMOTE-ENN. Additionally, Shapley additive explanations (SHAP) values were generated to implement the explainability of the model and display the important risk factors. RESULTS Our meta-classifier had an internal validation recall, specificity, and AUROC of 0.83, 0.86, and 0.85 for identifying the risk of oral cancer and 0.92, 0.60, and 0.76 for identifying suspicious oral mucosal disease respectively. Upon external validation, the meta-classifier had a significantly higher AUROC than the crude/current method used for identifying the risk of oral cancer (0.78 vs 0.46; p = 0.001) Also, the meta-classifier had better recall than the crude method for predicting the risk of suspicious oral mucosal diseases (0.78 vs 0.47). CONCLUSION Overall, these findings showcase that our approach optimizes the use of risk factors in identifying patients for oral screening which suggests potential clinical application.
Collapse
Affiliation(s)
- John Adeoye
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, University of Hong Kong, Hong Kong, China
| | - Li-Wu Zheng
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, University of Hong Kong, Hong Kong, China
| | - Peter Thomson
- College of Medicine and Dentistry, James Cook University, Cairns, Queensland, Australia
| | - Siu-Wai Choi
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, University of Hong Kong, Hong Kong, China
| | - Yu-Xiong Su
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, University of Hong Kong, Hong Kong, China.
| |
Collapse
|
4
|
Huang Q, Guo Y, Shen Y, Hsueh CY, Tao L, Zhang M, Wu C, Gong H, Zhou L. Epidemiological, Clinical, and Oncological Outcomes of non-Alcohol Drinking and non-Smoking Laryngeal Squamous Cell Carcinoma Patients: A Distinct Entity. Technol Cancer Res Treat 2022; 21:15330338221133690. [PMID: 36259221 PMCID: PMC9583220 DOI: 10.1177/15330338221133690] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/08/2022] Open
Abstract
Purpose: To explore the discrepancy in clinicopathological and prognostic features between smoking and alcohol drinking (SA) and non-smoking and non-alcohol drinking (NSNA) patients with laryngeal squamous cell carcinoma (LSCC). Methods: This retrospective study including 1735 patients with LSCC was conducted from January 2005 to December 2010, which were categorized into 4 groups, NSNA group, smoking only group, alcohol-drinking only group, and SA group. We compared overall survival (OS) and disease-free survival (DFS) using the Kaplan-Meier method and indicated clinicopathological features by Cox proportional hazards regression models before and after propensity score matching (PSM). Results: A total of 415 patients (23.92%) were identified as NSNA. The SA group was predominantly patients ≤60 years old (46.63%) while the NSNA group was more older (58.07%). NSNA group was more likely to present at earlier disease stage and more female. No significant difference in OS (P = .685) and DFS (P = .976) was found between the 2 groups. In addition to age and recurrence and metastasis being common independent prognostic factors in terms of OS in both groups of patients, NSNA group also exhibited other factors, namely tumor area >3.7 cm2 and positive resection margin. For DFS, N + stage, tumor size >3.7 cm2, and positive resection margin were prognostic features specific to NSNA group. Conclusion: The outcome is similar in LSCC patients with and without SA. NSNA group shows a distinct profile from that found in SA group. Clinicopathological features from NSNA group should be considered for LSCC management.
Collapse
Affiliation(s)
- Qiang Huang
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Yang Guo
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Yujie Shen
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Chi-Yao Hsueh
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Lei Tao
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Ming Zhang
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Chunping Wu
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China
| | - Hongli Gong
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China,Hongli Gong, MD, Department of
Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai 200031,
China.
| | - Liang Zhou
- Department of Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai, China,Liang Zhou, MD, Department of
Otorhinolaryngology, Eye & ENT Hospital, Fudan University, Shanghai 200031,
China.
| |
Collapse
|
5
|
Association between smoking habits and dental care utilization and cost using administrative claims database and specific medical check-up data. BMC Oral Health 2022; 22:372. [PMID: 36056338 PMCID: PMC9440590 DOI: 10.1186/s12903-022-02397-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/26/2022] [Accepted: 08/16/2022] [Indexed: 11/10/2022] Open
Abstract
BACKGROUND This study aims to evaluate the association between smoking habits and dental care utilization and cost in individuals registered with the Japan Health Insurance Association, Osaka branch. METHODS We used the administrative claims database and specific medical check-up data and included 226,359 participants, who visited dental institutions, underwent dental examinations, and underwent specific medical checkups, with smoking data from April 2016 to March 2017. We calculated propensity scores with age, gender, exercise, eating habits, alcohol intake, and sleep. We also compared dental care utilization with the total cost of each procedure. RESULTS According to propensity score matching, 62,692 participants were selected for each group. Compared to non-smokers, smokers were younger, and a higher proportion were men. Smokers tended to skip breakfast, have dinner just before bed, and drink alcohol. After adjusting for potential confounding factors with propensity score matching, the mean annual dental cost among smokers was significantly higher than non-smokers. The prevalence of pulpitis, missing teeth, and apical periodontitis were higher among smokers than non-smokers, while inlay detachment, caries, and dentine hypersensitivity were higher among non-smokers. CONCLUSION This study suggests that smokers have higher dental cost consisted of progressive dental caries, missing teeth, and uncontrolled acute inflammation that necessitated the use of medications. It is suggested that smokers tend to visit the dentist after their symptoms become severe.
Collapse
|
6
|
Matthes R, Jablonowski L, Pitchika V, Holtfreter B, Eberhard C, Seifert L, Gerling T, Vilardell Scholten L, Schlüter R, Kocher T. Efficiency of biofilm removal by combination of water jet and cold plasma: an in-vitro study. BMC Oral Health 2022; 22:157. [PMID: 35524324 PMCID: PMC9074283 DOI: 10.1186/s12903-022-02195-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/10/2022] [Accepted: 04/25/2022] [Indexed: 11/10/2022] Open
Abstract
Background Peri-implantitis therapy is a major problem in implantology. Because of challenging rough implant surface and implant geometry, microorganisms can hide and survive in implant microstructures and impede debridement. We developed a new water jet (WJ) device and a new cold atmospheric pressure plasma (CAP) device to overcome these problems and investigated aspects of efficacy in vitro and safety with the aim to create the prerequisites for a clinical pilot study with these medical devices. Methods We compared the efficiency of a single treatment with a WJ or curette and cotton swab (CC) without or with adjunctive use of CAP (WJ + CAP, CC + CAP) to remove biofilm in vitro from rough titanium discs. Treatment efficacy was evaluated by measuring turbidity up to 72 h for bacterial re-growth or spreading of osteoblast-like cells (MG-63) after 5 days with scanning electron microscopy. With respect to application safety, the WJ and CAP instruments were examined according to basic regulations for medical devices. Results After 96 h of incubation all WJ and CC treated disks were turbid but 67% of WJ + CAP and 46% CC + CAP treated specimens were still clear. The increase in turbidity after WJ treatment was delayed by about 20 h compared to CC treatment. In combination with CAP the cell coverage significantly increased to 82% (WJ + CAP) or 72% (CC + CAP), compared to single treatment 11% (WJ) or 10% (CC). Conclusion The newly developed water jet device effectively removes biofilm from rough titanium surfaces in vitro and, in combination with the new CAP device, biologically acceptable surfaces allow osteoblasts to grow. WJ in combination with CAP leads to cleaner surfaces than the usage of curette and cotton swabs with or without subsequent plasma treatment. Our next step will be a clinical pilot study with these new devices to assess the clinical healing process. Supplementary Information The online version contains supplementary material available at 10.1186/s12903-022-02195-1.
Collapse
Affiliation(s)
- Rutger Matthes
- Department of Restorative Dentistry, Periodontology, Endodontology, Preventive Dentistry and Pedodontics, Dental School, University Medicine Greifswald, Fleischmannstr. 42, 17475, Greifswald, Germany
| | - Lukasz Jablonowski
- Department of Restorative Dentistry, Periodontology, Endodontology, Preventive Dentistry and Pedodontics, Dental School, University Medicine Greifswald, Fleischmannstr. 42, 17475, Greifswald, Germany
| | - Vinay Pitchika
- Department of Restorative Dentistry, Periodontology, Endodontology, Preventive Dentistry and Pedodontics, Dental School, University Medicine Greifswald, Fleischmannstr. 42, 17475, Greifswald, Germany
| | - Birte Holtfreter
- Department of Restorative Dentistry, Periodontology, Endodontology, Preventive Dentistry and Pedodontics, Dental School, University Medicine Greifswald, Fleischmannstr. 42, 17475, Greifswald, Germany
| | | | - Leo Seifert
- Sirona Dental Systems GmbH, Bensheim, Germany
| | - Torsten Gerling
- ZIK Plasmatis, Leibniz-Institute for Plasma Science and Technology e.V. (INP), Greifswald, Germany
| | - Laura Vilardell Scholten
- ZIK Plasmatis, Leibniz-Institute for Plasma Science and Technology e.V. (INP), Greifswald, Germany
| | - Rabea Schlüter
- Imaging Center of the Department of Biology, University of Greifswald, Greifswald, Germany
| | - Thomas Kocher
- Department of Restorative Dentistry, Periodontology, Endodontology, Preventive Dentistry and Pedodontics, Dental School, University Medicine Greifswald, Fleischmannstr. 42, 17475, Greifswald, Germany.
| |
Collapse
|
7
|
Uddin S, Singh A, Mishra V, Agrawal N, Gooi Z, Izumchenko E. Molecular drivers of oral cavity squamous cell carcinoma in non-smoking and non-drinking patients: what do we know so far? Oncol Rev 2022; 16:549. [PMID: 35340886 PMCID: PMC8941340 DOI: 10.4081/oncol.2022.549] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/27/2021] [Accepted: 11/09/2021] [Indexed: 11/23/2022] Open
Abstract
Oral cavity squamous cell carcinoma (OCSCC) is one of the most common head and neck cancers worldwide. It is well known that risk factors for OCSCC include tobacco and excess alcohol consumption. However, in recent years, OCSCC incidence has been increasing in patients without these traditional risk factors. The cause of this increase is unclear and various genetic, environmental, and infectious factors have been hypothesized to play a role. Additionally, there are expert opinions that oral cancer in non-smoking, non-drinking (NSND) patients have a distinct phenotype resulting in more aggressive disease presentation and poorer prognosis. In this review, we summarize the current state of knowledge for oral cavity cancer in patients without traditional risk factors.
Collapse
Affiliation(s)
| | - Alka Singh
- Department of Medicine, Section of Hematology and Oncology
| | - Vasudha Mishra
- Department of Medicine, Section of Hematology and Oncology
| | - Nishant Agrawal
- Department of Surgery, Section of Otolaryngology-Head and Neck Surgery, University of Chicago, USA
| | | | | |
Collapse
|
8
|
Adeoye J, Koohi-Moghadam M, Lo AWI, Tsang RKY, Chow VLY, Zheng LW, Choi SW, Thomson P, Su YX. Deep Learning Predicts the Malignant-Transformation-Free Survival of Oral Potentially Malignant Disorders. Cancers (Basel) 2021; 13:cancers13236054. [PMID: 34885164 PMCID: PMC8657223 DOI: 10.3390/cancers13236054] [Citation(s) in RCA: 22] [Impact Index Per Article: 7.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/08/2021] [Revised: 11/26/2021] [Accepted: 11/26/2021] [Indexed: 12/12/2022] Open
Abstract
Simple Summary Mouth cancer is the most common malignancy in the head-and-neck region. Usually, these tumors develop from white lesions in the mouth that appear long before cancer diagnosis. However, platforms that can estimate the time-factored risk of cancer occurring from these diseases and guide treatment and monitoring approaches are elusive. To this end, our study presents time-to-event models that are based on machine learning for prediction of the risk of malignancy from oral white lesions following pathological diagnosis as a function of time. These models displayed very satisfactory discrimination and calibration after multiple tests. To facilitate their preliminary use in clinical practice and further validation, we created a website supporting the use of these models to aid decision making. Abstract Machine-intelligence platforms for the prediction of the probability of malignant transformation of oral potentially malignant disorders are required as adjunctive decision-making platforms in contemporary clinical practice. This study utilized time-to-event learning models to predict malignant transformation in oral leukoplakia and oral lichenoid lesions. A total of 1098 patients with oral white lesions from two institutions were included in this study. In all, 26 features available from electronic health records were used to train four learning algorithms—Cox-Time, DeepHit, DeepSurv, random survival forest (RSF)—and one standard statistical method—Cox proportional hazards model. Discriminatory performance, calibration of survival estimates, and model stability were assessed using a concordance index (c-index), integrated Brier score (IBS), and standard deviation of the averaged c-index and IBS following training cross-validation. This study found that DeepSurv (c-index: 0.95, IBS: 0.04) and RSF (c-index: 0.91, IBS: 0.03) were the two outperforming models based on discrimination and calibration following internal validation. However, DeepSurv was more stable than RSF upon cross-validation. External validation confirmed the utility of DeepSurv for discrimination (c-index—0.82 vs. 0.73) and RSF for individual survival estimates (0.18 vs. 0.03). We deployed the DeepSurv model to encourage incipient application in clinical practice. Overall, time-to-event models are successful in predicting the malignant transformation of oral leukoplakia and oral lichenoid lesions.
Collapse
Affiliation(s)
- John Adeoye
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong 999077, China; (J.A.); (L.-W.Z.); (S.-W.C.)
| | - Mohamad Koohi-Moghadam
- Division of Applied Oral Sciences and Community Dental Care, Faculty of Dentistry, The University of Hong Kong, Hong Kong 999077, China;
| | | | - Raymond King-Yin Tsang
- Division of Otorhinolaryngology, Department of Surgery, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong 999077, China;
| | - Velda Ling Yu Chow
- Division of Head and Neck Surgery, Department of Surgery, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong 999077, China;
| | - Li-Wu Zheng
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong 999077, China; (J.A.); (L.-W.Z.); (S.-W.C.)
| | - Siu-Wai Choi
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong 999077, China; (J.A.); (L.-W.Z.); (S.-W.C.)
| | - Peter Thomson
- College of Medicine and Dentistry, James Cook University, Cairns, QLD 4870, Australia
- Correspondence: (P.T.); (Y.-X.S.)
| | - Yu-Xiong Su
- Division of Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong 999077, China; (J.A.); (L.-W.Z.); (S.-W.C.)
- Correspondence: (P.T.); (Y.-X.S.)
| |
Collapse
|
9
|
Adeoye J, Hui L, Koohi-Moghadam M, Tan JY, Choi SW, Thomson P. Comparison of time-to-event machine learning models in predicting oral cavity cancer prognosis. Int J Med Inform 2021; 157:104635. [PMID: 34800847 DOI: 10.1016/j.ijmedinf.2021.104635] [Citation(s) in RCA: 11] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2021] [Revised: 10/28/2021] [Accepted: 10/29/2021] [Indexed: 12/13/2022]
Abstract
BACKGROUND Applying machine learning to predicting oral cavity cancer prognosis is important in selecting candidates for aggressive treatment following diagnosis. However, models proposed so far have only considered cancer survival as discrete rather than dynamic outcomes. OBJECTIVES To compare the model performance of different machine learning-based algorithms that incorporate time-to-event data. These algorithms included DeepSurv, DeepHit, neural net-extended time-dependent cox model (Cox-Time), and random survival forest (RSF). MATERIALS AND METHODS Retrospective cohort of 313 oral cavity cancer patients were obtained from electronic health records. Models were trained on patient data following preprocessing. Predictors were based on demographic, clinicopathologic, and treatment information of the cases. Outcomes were the disease-specific and overall survival. Multivariable analyses were conducted to select significant prognostic features associated with tumor prognosis. Two models were generated per algorithm based on all-prognostic features and significant-prognostic features following statistical analysis. Concordance index (c-index) and integrated Brier scores were used as performance evaluators and model stability was assessed using intraclass correlation coefficients (ICC) calculated from these measures obtained from the cross-validation folds. RESULTS While all models were satisfactory, better discriminatory performance and calibration was observed for disease-specific than overall survival (mean c-index: 0.85 vs 0.74; mean integrated Brier score: 0.12 vs 0.17). DeepSurv performed best in terms of discrimination for both outcomes (c-indices: 0.76 -0.89) while RSF produced better calibrated survival estimates (integrated Brier score: 0.06 -0.09). Model stability of the algorithms varied with the outcomes as Cox-Time had the best intraclass correlation coefficient (mean ICC: 1.00) for disease-specific survival while DeepSurv was most stable for overall survival prediction (mean ICC: 0.99). CONCLUSIONS Machine learning algorithms based on time-to-event outcomes are successful in predicting oral cavity cancer prognosis with DeepSurv and RSF producing the best discriminative performance and calibration.
Collapse
Affiliation(s)
- John Adeoye
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China; Oral Cancer Research Group, Faculty of Dentistry, University of Hong Kong, Hong Kong, China.
| | - Liuling Hui
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China
| | - Mohamad Koohi-Moghadam
- Applied Oral Sciences and Community Dental Care, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China
| | - Jia Yan Tan
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China; Oral Cancer Research Group, Faculty of Dentistry, University of Hong Kong, Hong Kong, China
| | - Siu-Wai Choi
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China; Oral Cancer Research Group, Faculty of Dentistry, University of Hong Kong, Hong Kong, China
| | - Peter Thomson
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong, China; Oral Cancer Research Group, Faculty of Dentistry, University of Hong Kong, Hong Kong, China; College of Medicine and Dentistry, James Cook University, Queensland, Australia.
| |
Collapse
|
10
|
Adeoye J, Tan JY, Ip CM, Choi SW, Thomson P. "Fact or fiction?": Oral cavity cancer in nonsmoking, nonalcohol drinking patients as a distinct entity-Scoping review. Head Neck 2021; 43:3662-3680. [PMID: 34313348 DOI: 10.1002/hed.26824] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/06/2021] [Revised: 07/11/2021] [Accepted: 07/19/2021] [Indexed: 12/20/2022] Open
Abstract
Oral cavity cancer is often described as a lifestyle-related malignancy due to its strong associations with habitual factors, including tobacco use, heavy alcohol consumption, and betel nut chewing. However, patients with no genetically predisposing conditions who do not indulge in these risk habits are still being encountered, albeit less commonly. The aim of this review is to summarize contemporaneous reports on these nonsmoking, nonalcohol drinking (NSND) patients. We performed database searching to identify relevant studies from January 1, 2000 to March 31, 2021. Twenty-six articles from 20 studies were included in this study. We found that these individuals were mostly females in their eighth decade with tumors involving the tongue and gingivobuccal mucosa. This review also observed that these patients were likely diagnosed with early stage tumors with overexpression of programmed death-ligand 1 (PD-L1) and increased intensity of tumor infiltrating lymphocytes. Treatment response and disease-specific prognosis were largely comparable between NSND and smoking/drinking patients.
Collapse
Affiliation(s)
- John Adeoye
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China
| | - Jia Yan Tan
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China
| | - Cheuk Man Ip
- Department of Anesthesia, Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, China
| | - Siu-Wai Choi
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China
| | - Peter Thomson
- Oral and Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China
| |
Collapse
|