Introduction

Cancer is one of the major causes of mortality among adults worldwide. A rise in cancer-related deaths is expected, primarily because of an increase in the risk factors for cancer [1]. Oral cancer is a significant health concern, especially in Asian countries, as it shows a great deal of variation worldwide in terms of its incidence rate owing to racial and ethnic differences among the people as well as associated lifestyles [2]. Countries in South and South-East Asia present the most worrisome rate of incidence, according to the International Agency for Research on Cancer, which includes Sri Lanka, India, and Pakistan, among others [3]. These countries, which are said to be developing, lower-income countries, are often more affected by this burden compared with developed countries and are responsible for two-thirds of the oral malignancies worldwide. The Asian population is more prone to oral cancer compared with other communities due to the habit of chewing tobacco and areca nut in betel quid being rampant in these groups of people. The incidence of oral cancer found to occur more commonly in males than in females [4, 5]. With the increase in age, the risk of developing oral cancer increases. Further, the 5-year incidence and mortality rate as per GLOBOCAN (2018) report indicate oral cancer incidence and mortality are higher in males compared with females [6].

The majority of oral cancer incidence is found to occur in people age 50 years and above. In high-risk countries, oral cancer cases among people below the age of 40 have also been reported. A study conducted by Dikshit R et al. has highlighted oral cancer as one of the fatal diseases in Indian men belonging to the age group of 30–69 years, accounting for 22.9% of the cases, which included cancers of the lip, oral cavity, and pharynx, while in women belonging to the same age group, oral cancer accounted for 9.8% of the cancer cases after cervical (17.1%), stomach (14.1%), and breast (10.2%) [1].

Oral cancer could be defined as a non-homogenous group of diseases of the oral cavity, involving the lips, tongue, labial mucosa, buccal mucosa, palate, the floor of the mouth, gingiva, and other anatomical sites. However, literature does not adopt a clear cut definition for the same. Oral cancer has a multifactorial etiology, with tobacco considered a significant risk factor. Apart from smoking or chewing any form of tobacco, along with or without betel quid, excess consumption of alcohol, infection by human papilloma virus (HPV), poor oral hygiene, and poor dentition and even genetic susceptibility are associated with risk factors [7,8,9]. A considerable minority of the cases have also reported that oral cancer can be caused by factors other than the traditional risk factors. Lower economic status and lack of awareness about oral health are also related to increased risk of developing the disease [10]. The level of education is associated with knowledge about oral cancer, and its risk factors were looked into by Elango JK et al. in a survey conducted among a high-risk population. They reported that higher education level, more significant, was the awareness for both oral cancer and its risk factors (p < 0.001), while oral cancer-causing risk habits, which included smoking, chewing pan with/without tobacco, and alcohol, were significantly less prevalent among those with a better level of education [11].

Oral cancers show a low survival rate (less than 50% in most countries) since half of the diagnosed cases would have progressed to an advanced stage. The above progression of the diseases to advanced stage attributed due to the patient’s ignorance or reluctance in obtaining proper medical care and poor acceptance of the treatment [10]. Also, the lack of access to cancer treatment centers may further delay and also result from misdiagnosis and inappropriate treatment. On the contrary, in diagnosis at the stage where the primary tumor is localized, the survival rate is better, suggesting that early diagnosis could lead to better chances of survival. Therefore, detection of oral cancer at an early stage or a precancerous lesion before it becomes malignant is the key to ensure effective treatment and reduced morbidity [12].

In the present review, we have attempted to highlight the potentiality of fluorescence-based spectroscopic tools in the early oral cancer detection/diagnosis, while we also tried to summarize some of the conventional techniques used in oral cancer diagnosis with their limitations, which is routinely used in oral oncology screening.

Conventional diagnostic techniques

The fact is that the oral cavity is accessible for the visual and tactile examination and can be viewed to detect oral cancer or even for its precancerous change [9]. Presently, the assessment of the oral cavity is carried out either by oral examination, invasive approach (oral biopsy), and vital staining techniques (toluidine blue, Lugol’s iodine), or by light-based optical tools such as chemiluminescence techniques.

Oral examination

Conventional oral examination (COE) is carried out with the help of an incandescent light, which is used routinely in the screening of oral cancer. Visual examination to screen patients is advantageous as it is non-invasive and causes no pain and is also well accepted among the clinicians socially. However, studies have highlighted that the method is a bit obscure for screening premalignant or early lesions. On the contrary, few studies report that the COE method has better sensitivity and specificity. Presently there is hardly any test that can reveal an accurate detection of oral lesions when it comes to the screening program; the only method used for screening in the oral malignant or lesion is a conventional oral examination. Some of the earlier studies have reported the COE-based screening for the detection of potential oral malignant lesions or cancers performed by health experts [13, 14].

A hospital-based study was carried out by Chang et al. where a total of 13,878 participants were included in the study. The study aimed to demonstrate the effectiveness of oral visual screening in determining the potential risk of oral cavity cancer in chronic smokers, alcohol consumption, and betel quid chewers. The results clearly show that oral screening is effective and should be used regularly as a routine screening procedure to detect oral cancer at the earliest [15]. In another study, Walsh et al. have also highlighted the diagnostic methods, such as COE, vital stain rinsing, optical tools, and biomarker applications in oral cancer detection. The review mainly includes the sensitivity and specificity of the methods. The study shows that COE has better sensitivity, which was higher than 70%, and the specificity was more than 90% compared with other techniques used in oral cancer screening [16]. Table 1 below highlights the results of a few studies where COE is used as a screening tool in oral cancer.

Table 1 Studies mentioning sensitivity and specificity for the detection of oral abnormalities by conventional oral examination
Full size table

Thus, based on earlier studies, it has been reported that COE could be used to screen the oral lesions and cancers. However, it is not useful in differentiating between the lesions that could progress to cancer and which would be in a benign state. Also, chromosomal aberrations may exist in areas of the mouth appearing to be normal, which could later lead to the development of cancer [22]. Also, certain precancerous lesions get hidden beneath normal-appearing mucosa and which is difficult to detect during COE and could lead to oral malignancies. Therefore, all suspicious lesions should be evaluated histologically to determine their potential in turning into a malignant condition.

Oral biopsy

Oral biopsy is considered an essential clinical tool in the diagnosis of oral lesions ranging from simple lesions such as precancerous lesions to cancerous lesions. The biopsy procedure helps the clinicians to determine the proper diagnosis of the diseases, which eventually will help patients in the timely treatment of the diseases. The suspicious oral lesions are evaluated either by oral brush cytology or by surgical biopsy.

The scalpel or punch biopsy, followed by its histopathological assessment, is considered the gold standard for the diagnosis of oral lesions [23]. However, it is not conceivable to perform a biopsy in all cases, as some of the patients might not give consent as their lesion may be asymptomatic. In some cases, the patients are critically ill due to which the procedure could not be possible to carry out. An alternative to the above, the clinicians go for an oral cytology approach, where they monitor changes in cytology of oral cells by performing a procedure such as brush cytology. The brush cytological study is minimally invasive and well accepted among the patients and gives consent to undergo the procedure, due to which it has been considered an alternative diagnostic option to a more invasive and painful biopsy approach [24]. The technique holds the potential in determining the pathological changes in the oral lesion. The technique mainly involves the collection of oral cells by rinsing the oral cavity after scraping the affected oral cavity area using an oral brush.

Oral computer-assisted brush cytology (OralCDx) is the most standard and routinely brush cytology procedure for assessing the oral lesions [25]. Several studies have reported encouraging results with oral brush cytology in evaluating cancerous lesions. Table 2 illustrates the utilization of oral brush cytology methods highlighting the sensitivity and specificity of the technique.

Table 2 Studies mentioning sensitivity and specificity for the detection of oral abnormalities by oral brush cytology
Full size table

At present, a biopsy followed by a histopathological assessment of a tissue sample suspected of malignancy is considered the gold standard for the diagnosis of oral cancer. Histopathological evaluation of tissues involves a visual analysis of tissues at the cellular level, which enables the clinicians to make decisions on whether the tissue section under evaluation is malignant or not, based on the features like shape and distribution of the cells and the nuclei [31, 32]. Further, during the procedure, the site of the mouth from where the tissue specimen obtained during biopsy should be such that it should be able to represent the most severe changes in the oral cavity. Therefore, the accuracy of diagnosis depends on the clinician’s choice of an appropriate biopsy site and the pathologist’s interpretation based on the histopathological findings [33]. Due to the above problem with visually identifying sites for biopsy, which represents the most malignant transformation, it is subjective and depends entirely on the examiner and, this might lead to selecting an inappropriate lesion which in turn often requires redoing the biopsy, adding to the patient’s distress [34].

Further, the interpretation of microscopic sections includes the presence of artifacts that are common in biopsy specimens of the oral mucosa, which can result in a change in morphology and cytology, which will render the specimen useless and their presence depends on the type of biopsy technique used.

Staining techniques

The conventional vital staining techniques such as toluidine blue, Lugol’s iodine staining, and Rose Bengal staining are commonly used in oral cancer diagnosis [35]. The toluidine blue technique is the most accepted staining technique and coupled with other modalities to visualize the oral lesions. The staining technique helps in the diagnosis of oral cancer and has been used to detect both premalignant and malignant conditions. Toluidine blue is a vital stain used to visualize changes in the mucosa of the oral cavity as it is capable of staining nucleic acids. The dye binds DNA and RNA, which is seen to be higher side during the onset of malignancy [36], while Lugol’s iodine binds to the glycogen content present in the epithelium cells. The stain has been used in discriminating against the inflammatory or cancerous oral lesion from normal epithelium as they contain lower glycogen contents [37].

Various studies have reported the application of these stains in oral cancer detection; some of these studies are highlighted in Table 3 above. A study conducted by Chainani W et al. has reported the sensitivity and specificity of toluidine blue to the detection of high-risk lesions (carcinoma in situ or OSCCA). Of the 100 lesions evaluated from 43 patients, 77 lesions were biopsied and included for analysis. The results of the toluidine blue technique showed a sensitivity of 94% in the case of high-risk lesions, while that for OSCCA was 100%. However, the authors pointed out the limitations in the study where they found the specificity of 45% and 39% for the above cases, respectively [43]. To increase the sensitivity and specificity and overcome the limitations of the staining techniques, some authors have combined two different staining techniques to detect abnormal oral lesions. A study was carried out by Nagaraju K et al. where the authors determined the diagnostic applicability and reliability of toluidine blue with Lugol’s staining. The results show the combined staining technique can assess and discriminate between the premalignant and malignant oral lesions with a sensitivity of 92.7% and specificity of 60%, respectively [41]. The staining of the oral lesion by toluidine blue is shown in Fig. 1.

Table 3 Studies mentioning sensitivity and specificity for the detection of oral abnormalities by toluidine staining and Lugol’s iodine
Full size table
Fig. 1
figure 1

TBlue staining of leukoplakia, a before staining and b retention of TBlue dye by the white patches, reprinted from Awan K. et al. [36] with permission of John Wiley and Sons

Full size image

Although the staining techniques are quite a sensitive tool, however, they possess some limitations such as erroneous retention of stain in regions that show inflammation and ulcers and inability to detect deeper-lying dysplasia in the epithelium and the hindrance in dye-uptake lesion area caused by the mucous [38].

Chemiluminescence

Chemiluminescence is a light-based screening technique used to examine the oral cavity. The method principally depends on alterations in the refractive index of the mucosa corresponding to the alterations in the metabolism and structure occurring due to malignancy. Normal mucosa absorbs the blue-white light, while the cells which are dysplastic and neoplastic reflect it. The chemiluminescence-based technique is marketed under the names of Vizilite plus and MicroLux DL. Vizilite Plus is a disposable light packet, while MicroLux DL is reusable. Vizilite Plus tool is a combination of chemiluminescence and toluidine blue (TB). The toluidine blue (TB) stain specifically stains the DNA. The procedure mainly involves 1-min rinsing with a 1% acetic acid solution, followed by the clinician observation of the rinsed lesion sites using the tool which emits light in a wavelength from 430 to 580 nm (as shown in Fig. 2) [44, 45]. The sensitivity of Vizilite was reported by Chaudhry A et al. where the authors claimed Vizilite-based detection showed better sensitivity in identifying leukoplakia [46]. Further, several studies have reported the application of the chemiluminescence tool in screening the oral lesions by clinicians as shown in Table 4 below.

Fig. 2
figure 2

Oral lesion at buccal mucosa under a incandescent light b stained with toluidine blue and c Aceto-white delineation of oral lesion on Vizilite plus, reprinted from Jain N. et al. [44] with permission of Elsevier

Full size image
Table 4 Reports on Vizilite-based detection of oral lesions
Full size table

Although the above technique possesses better application compared with other conventional methods, however, they possess some limitations such as interference of saliva in visualizing lesions, inability to identify erythroplakia, and to differentiate non-dysplasia from dysplasia has been reported.

Non-invasive fluorescence-based tools in oral diagnosis

When an experienced practitioner performs a visual examination, he limits to differentiate between premalignant lesions from the conditions that are inflammatory and one that can lead to malignancy. Also, in many of the cases, the patient would get diagnosed at a later stage of cancer by the time they reach the clinician, followed by their histopathological assessment, which is very subjective. Impediments such as these, presented by conventional diagnostic methods in early detection of oral cancer, have led to the need for the development of alternative diagnostic modalities. The optical tools based on autofluorescence are effective in discriminating dysplastic lesions from early neoplastic and also provide information regarding changes at the biochemical and molecular levels. Autofluorescence is a phenomenon that takes place when an intrinsic fluorophore molecule gets excited by a light source of a particular wavelength, and emits fluorescence upon relaxation. On absorbing incident light of suitable wavelength, the electrons in the fluorophores get excited from the ground vibrational states to the higher energy states. The electrons then relax to the 1st excited state of the fluorophore through non-radiative relaxations followed by its transition to the ground vibrational levels through radiative relaxations known as fluorescence and thereby releasing the excitation energy of the fluorophores at higher wavelength compared with the excitation wavelength of the fluorophore as shown in Fig. 3 [52, 53].

Fig. 3
figure 3

Block diagram depicting the generation of fluorescence along with its underlying phenomenon

Full size image

Tissues possess endogenous fluorophores, which include aromatic amino acids, collagen, elastin, porphyrins, and metabolic products such as nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotides (FAD). During the neoplastic and pre-neoplastic stages, these endogenous fluorophores demonstrate alterations in their metabolism and structures. Thus, a light of a particular wavelength is excited upon these endogenous fluorophores present in the tissues. They display different optical properties (absorbance, reflectance, and fluorescence) which are sensitive to their microenvironment changes, which in turn get reflected in their corresponding wavelength and emission profiles [54]. The observed fluorescence from tissues under pathological conditions truly reflects the change in biochemical and tissue architecture. Abnormal conditions such as dysplasia may lead to disturbance in the fluorophore distribution and quantity, showing corresponding loss of tissue autofluorescence as compared with the healthy tissues. The NADH and FAD are important fluorophores in the epithelium of the oral mucosa, while in the stroma, it is the collagen that is dominant. Any change in the levels of these fluorophores is indicative of the metabolic status of the cells. Hence, change in spectral intensity from disease tissues could be contributed by a change in the concentrations of fluorophores [55,56,57].

Autofluorescence-based tools are non-invasive, easy to handle, and also minimize unnecessary surgical biopsy interventions. Due to these properties, the techniques can be used as a potential tool to monitor the malignant area in medically compromised patients. The autofluorescence-based diagnostic tools for oral cancer include the following:

  • Visual autofluorescence (visually enhanced lesion scope) VELscope

  • Autofluorescence spectroscopy

VELscope

Over the last few decades, tissue fluorescence has been exploited much to detect oral lesions by involving a variety of devices. VELscope is one such device that is based on fluorescence emitted by the natural fluorophores of the oral tissues upon suitable light excitations in the region 400–600 nm. It is a handheld device that consists of an arc lamp as an excitation light source to shine into the oral cavity and a series of reflectors and filters to view the emitted fluorescence through them. The device avoids procedures like rinsing of the oral cavity with chemical agents [57, 58]. The diagnostic ability of VELscope depends on the differences in the fluorescence emitted by the diseased and healthy tissues. Pale green is seen in the case of normal mucosa, while dark coloration is seen in dysplastic tissues, as shown in Fig. 4 [59]. Various studies have reported using VELscope in the diagnosis of oral lesions which are listed in Table 5. Further, some studies have highlighted the limitations of using VELscope alone in the diagnosis of potentially malignant disorders or oral squamous cell carcinoma [65].

Fig. 4
figure 4

a Clinical oral examination of the tongue and b loss of fluorescence at the region of oral lesion observed under VELscope, reprinted from Cicciù M. et al. [59] with creative copyright Licensee MDPI, Basel, Switzerland

Full size image
Table 5 Reports on VELscope-based detection of oral lesion
Full size table

Autofluorescence spectroscopy

Fluorescence spectroscopy is a very sensitive technique that can detect minor metabolic or pathological changes taking place in the cells and tissues by illuminating the concerned fluorophores with appropriate monochromatic sources of light [66]. The fluorescence properties can differentiate abnormal tissues from the normal with higher sensitivity and specificity and thus providing an opportunity to use in diagnostic decision-making, including the detection of premalignant conditions and thereby bringing down the chances of repeated biopsies [67]. There are several reports of using fluorescence spectroscopy in the past to distinguish malignant tissues of different organs from their normal healthy counterparts in vitro/in vivo. The technique has shown its ability to detect even benign tissues from normal along with malignant tissues [68, 69]. The application of this technique in oral cancer detection is very well documented, and some studies are listed in Table 6. The fluorescence spectroscopy, when used along with suitable statistical analyses, has shown very high sensitivity and specificity in the discrimination of pathological conditions. The study by Jayanthi J L et al. applied linear discriminate analysis on the autofluorescence spectra recorded from 15 healthy people and 34 oral cancer patients. The results demonstrated clear differentiation of premalignant dysplasia from squamous cell carcinoma, as well as benign hyperplasia and hyperplasia from normal with a sensitivity of 86%, 78%, and 92%, and specificities of 90%, 100%, and 100%, respectively [75]. Similarly, in another study carried out by Sudha K D et al. the authors performed MATLAB-based PCA and k-means nearest neighbor (k-NN) analysis on 142 fluorescence spectra obtained from pathologically clinical samples recorded at 325-nm excitation using laser-induced fluorescence as a tool. The various features such as mean, median, maximum intensity, energy, spectral residuals, and standard deviation extracted from the fluorescence spectra of the normal, premalignant, and malignant oral tissue samples. In the study, the authors were able to show a robust application of PCA and k-NN analysis by which it can discriminate and also able to classify the unknown sample into a diagnostic group, which helps in reducing the misclassification as shown in Fig. 5. The authors were able to achieve sensitivity, specificity, and accuracy of 100, 94.5, and 96.17%, respectively. Based on the above study, the authors have tried to show that the laser-induced fluorescence technique can effectively be used to discriminate between normal, premalignant, and malignant conditions. Further, the time required to acquire and analyze the data is very less, and thus it holds very potentially as an alternative tool in the diagnosis of diseases [76].

Table 6 Studies mentioning sensitivity and specificity for the detection of oral abnormalities by fluorescence spectroscopy
Full size table
Fig. 5
figure 5

The normalized fluorescence spectra of oral tissue. a Normal and b malignant and c scattered plot of 143 spectra from normal, premalignant, and malignant samples (NC, normal calibration; MC, malignant calibration; PC, premalignant calibration; NT, normal test; MT, malignant test; and PT, premalignant test), reprinted from Sudha KD. et al. [76] with permission of SPIE

Full size image

While a study carried out by Yuvaraj M et al. monitored the spectral variation of porphyrin fluorophore in saliva samples collected using autofluorescence spectroscopy from oral squamous cell carcinoma (OSCC) patients and healthy individuals, the results demonstrate predominant emission patterns in the case of cancer patients as compared with the normal at 405-nm excitations. They observed additional spectral peaks at 625 and 687 nm in the cancer samples as compared with control due to the porphyrins. However, both of these peaks were absent in normal saliva samples [77]. The above observations have clearly demonstrated the potential of identifying porphyrins in saliva samples as a biomarker for oral cancer diagnosis using autofluorescence non-invasively. The above study reflects the potential use of fluorescence spectroscopy in early detection of premalignant lesions, even before when there are no visible clinical symptoms. The technique records fluorescence spectra qualitatively and is relatively inexpensive, portable, and easy to operate even by a technician with minimum training. These impeccable qualities make the fluorescence spectroscopy a promising tool to screen people with minimal complexity. They can provide a powerful and robust diagnostic platform for disease screening and subsequent treatment opportunities.

Conclusion

Oral cancer is preventable, and yet the present-day incidence and mortality rates remain to be high. Despite the existence of numerous conventional diagnostic techniques, many of which are invasive, and most of the cases are detected after they would have progressed to an advanced stage. The prevention entails in averting the progression of a premalignant disorder into full-blown cancer, by focusing on early detection strategies by using fluorescence-based spectroscopic tools, highlighted in this article. The fluorescence technique, explicitly fluorescence spectroscopy, holds the potential to identify oral precancerous and cancerous lesions with high sensitivity at an early stage, even when external morphological changes may not be visible. Besides, their cost-effectiveness, speed, and ability to provide real-time analysis of diagnosis objectively make them ideal diagnostic tools. Further, the technique holds immense potential in providing the molecular information generated from a fluorophore upon excitation. Therefore, it could increase the chances of survival, decrease morbidity, reduce the mortality rates, and ensure effective treatment, post-treatment, and a better quality of life for the patient.