Encyclopedia
Please note this is an old version of this entry, which may differ significantly from the current revision.
Subjects:
Primary Health Care
Oral squamous cell carcinoma (OSCC) is one of the most-prevalent cancer types worldwide, and it poses a serious threat to public health due to its high mortality and morbidity rates. OSCC typically has a poor prognosis, significantly reducing the chances of patient survival.
- oral squamous cell carcinoma
- metabolites’ salivary biomarkers
- intelligent diagnostic support system
1. Introduction
Oral squamous cell carcinoma (OSCC) is among the ten most prevalent cancer categories worldwide, associated with high mortality and morbidity rates, representing a global public health problem. In 2020, according to statistics from the World Health Organization’s Global Cancer Observatory, lip and oral cavity cancers accounted for more than 377,700 cases worldwide [1]. About 90% of oral malignancies correspond to squamous cell carcinoma [2,3].
OSCC is a malignant tumor that develops from the squamous epithelial cells lining the oral cavity, encompassing the lips, tongue, buccal mucosa, upper and lower gums, and palate, among other areas. Understanding the pathophysiology of OSCC is crucial for early detection, prevention, and effective management of this aggressive cancer. The main risk factors for OSCC include tobacco and alcohol consumption and human papillomavirus (HPV) infection present in most patients diagnosed with oral and oropharyngeal cancer, making them critical etiological factors [4]. Moreover, chronic irritation, such as from sharp teeth, ill-fitting dentures, and so forth, and a poor diet in terms of fruits and vegetables are factors. Additionally, socioeconomic factors, such as poor oral hygiene and limited access to healthcare services, also impact early detection and treatment outcomes [5].
The pathophysiology involves a multi-step process of genetic and molecular changes that lead to uncontrolled cell growth and the formation of malignant tumors. As the tumor advances, it can infiltrate nearby tissues, extend to regional lymph nodes, and metastasize to distant organs. This progression may lead to the temporary or permanent compromise of crucial functions such as speech, voice, swallowing, and mastication and may also be influenced by the chosen treatment approach [6]. In summary, OSCC involves the accumulation of genetic mutations, dysregulation of signaling pathways, invasion, and metastasis. Early detection and intervention are crucial for improving the prognosis and quality of life for individuals with OSCC. Preventive measures, such as avoiding tobacco and alcohol and receiving the HPV vaccine, can also reduce the risk of developing this devastating disease.
Management of all oral cavity cancers must occur in a Multidisciplinary Head and Neck Oncology Team due to the functional and aesthetic implications of the treatment, and the diagnosis of OSCC is traditionally performed by biopsy. OSCC’s 5-year survival rates are around 50%, and most of these patients survive a short time after diagnosis [7]. This scenario is because most tumors are identified late, compromising treatment, prognosis, and patient survival [8]. Given its high mortality, early and accurate diagnosis is extremely important [9].
To better understand the disease and the contribution of early diagnosis, disseminating OSCC signs and symptoms and implementing actions for early detection are essential. In this sense, tools and solutions that help diagnosis and prognosis in the early stages of the disease can be of great value, especially for patients. With early identification or an early prognosis, an increase in survival rates due to initiating treatment in the early stages is expected. In addition, a reduction in the mutilating impact of the treatment can also be cited, benefiting the patient and ensuring a better quality of life. These tools help medical professionals decide on the best treatment [10].
Studies indicate that salivary biomarkers can be used for OSCC’s early recognition [11]. Furthermore, salivary collection can be performed in a simple and non-invasive way, and its potential to aid in OSCC diagnosing has been studied by several authors [12,13,14,15,16,17,18].
Similarly, Artificial-Intelligence-guided approaches have shown considerable benefits for early diagnosis and prognosis of oral cancers [19], presenting computational machine learning solutions with improved accuracy of cancer sensitivity, recurrence, and survival predictions [20] and providing a consequent improvement in patient care and survival [21,22,23,24,25].
These approaches have been gaining more and more space in the medical field, with models based on artificial neural networks successfully used in several applications. For example, classification, prognosis, diagnosis, or identification of lung diseases [26,27], chronic kidney diseases [28], macular degeneration and diabetic retinopathy [29], Parkinson’s disease [30], and oral cancer and periodontitis [31,32,33] are among these applications, to name a few.
2. Oral Squamous Cell Carcinoma
Artificial Intelligence has been successfully used to assist in the diagnosis and prognosis of OSCC cases by analyzing hyperspectral clinical, pathological, and radiographic images. Halicek et al. [34,35] used hyperspectral imaging and convolutional neural networks to diagnose aerodigestive tract tumors with a better sensitivity of 81%. Ariji et al. [36] used computerized tomography (CT) images to detect cervical lymph node metastasis and infer disease-free survival. Fujima et al. [37] similarly used F-fluorodeoxyglucose (FDG) uptake by positron emission tomography (PET/CT) and achieved 80% accuracy in predicting survival. Pathological images were used by Folmsbee et al. [38] to identify oral cavity cancerous tissues among six other tissue classes with an accuracy of 96.37%, while Das et al. [39] classified biopsy images with similar performance to the pathological classification system in [40], achieving 97.51% accuracy.
Although the importance of clinical images in treating OSCC and of Artificial Intelligence solutions to aid in the diagnosis and prognosis of this disease is undeniable, image acquisition is not trivial, requiring highly complex human and material resources, while Artificial Intelligence applications that model the training steps need proper data preparation and labeling in an expensive and time-consuming process. Therefore, we must consider other data types, such as gene expression data, spectral data, autofluorescence, and saliva metabolites.
Thus, specifically concerning chemical methods, salivary biomarkers are an excellent alternative for analyzing and monitoring OSCC cases. Salivary biomarkers offer good accuracy in the early recognition of this disease. In addition, collecting salivary samples is a simple, non-invasive process with lower costs than clinical imaging.
Hu et al. [18] demonstrated that protein biomarkers of the human salivary proteome have a high capacity to discriminate OSCC patients from healthy patients, indicating that the combination of some salivary proteins reaches a sensitivity of up to 90% in this task and showing a promising approach for the search for OSCC biomarkers.
Using autofluorescence spectroscopy, Yuvaraj et al. [15] performed a pilot study demonstrating marked differences in the spectral signatures between healthy and OSCC patients, obtaining 85.7% sensitivity when using the fluorescence emission spectrum at a 405nm excitation. The authors further suggested that the reasons are linked to endogenous porphyrin, nicotinamide adenine dinucleotide (phosphate), reduced (NAD(P)H), and flavin adenine dinucleotide (FAD) in exfoliated saliva cells.
Sridharan et al. [41] used quadrupole time-of-flight (Q-TOF) liquid chromatography–mass spectrometry to assess the metabolomic profile of OSCC, oral leukoplakia (OLK), and control groups. The authors found a significant increase in 1-methylhistidine, inositol 1,3,4-triphosphate, d-glycerate-2-phosphate, 4-nitroquinoline-1-oxide, 2-oxoarginine, norcocaine nitroxide, sphinganine-1-phosphate, and pseudouridine in OLK and OSCC in comparison to control groups.
Song et al. [12] used conductive polymer spray–mass spectrometry (CPSI–MS) to quantify saliva samples from OSCC, patients with premalignant lesions, and healthy individuals. The authors showed that using CPSI–MS with machine learning (ML) can be a viable tool to distinguish OSCC and premalignant lesions from healthy conditions in real-time, with an achieved accuracy of 86.7%.
Ishikawa et al. [42] developed a multiple logistic regression model to differentiate OSCC from oral lichen planus. They realized a comprehensive analysis of fourteen hydrophilic metabolites in the saliva samples by capillary electrophoresis–mass spectrometry, showing high discrimination of the combination of indole-3-acetate and ethanolamine phosphate metabolites, resulting in 85.6% of the area under the curve (AUC).
Recently, Kouznetsova et al. [32] studied metabolic pathways related to oral cancer and periodontitis and, using a deep-learning neural network achieved the best accuracy of 79.54% in classifying these cases.
Finally, Costa et al. [43] analyzed saliva samples using gas chromatography–mass spectrometry with a data-mining approach to find the best biomarkers to diagnose OSCC. Using feature selection classification and random forest classification algorithms, they reported an AUC of 91% with only three metabolites versus 76% when using all 51 metabolites analyzed.
This entry is adapted from the peer-reviewed paper 10.3390/healthcare11192675
This entry is offline, you can click here to edit this entry!
