Recent advances in optical diagnosis of oral cancers: Review and future perspectives

Flavinated SDHA underlies the change in intrinsic optical properties of oral cancers

The molecular basis of reduced autofluorescence in oral squamous cell carcinoma (OSCC) cells relative to normal cells has been speculated to be due to lower levels of free flavin adenine dinucleotide (FAD). This speculation, along with differences in the intrinsic optical properties of extracellular collagen, lies at the foundation of the design of currently-used clinical optical detection devices. Here, we report that free FAD levels may not account for differences in autofluorescence of OSCC cells, but that the differences relate to FAD as a co-factor for flavination. Autofluorescence from a 70 kDa flavoprotein, succinate dehydrogenase A (SDHA), was found to be responsible for changes in optical properties within the FAD spectral region, with lower levels of flavinated SDHA in OSCC cells. Since flavinated SDHA is required for functional complexation with succinate dehydrogenase B (SDHB), decreased SDHB levels were observed in human OSCC tissue relative to normal tissues. Accordingly, the metabolism of OSCC cells was found to be significantly altered relative to normal cells, revealing vulnerabilities for both diagnosis and targeted therapy. Optimizing non-invasive tools based on optical and metabolic signatures of cancers will enable more precise and early diagnosis leading to improved outcomes in patients.

A combined autofluorescence and diffuse reflectance spectroscopy for mucosa tissue diagnosis: Dual-distance system and data-driven decision

Combined autofluorescence (AF) and diffuse reflectance (DR) spectroscopies have been expected to offer enhanced diagnostic accuracies for noninvasive early detection of mucosa lesions, that is, oral cavity carcinoma and cervical carcinoma. This work reports on a hybrid AF and DR spectroscopic system that is developed for quantification and diagnosis of mucosa abnormalities. The system stability and reliability are firstly assessed by phantom experiments, showing a measurement variation lower than 1% within 20 min. In vitro and in vivo validations are then conducted for tissue identification and lesion differentiation. For enhanced decision, a data-driven diagnosis algorithm is explored in pilot under different experimental configurations. The results conclude a promising accuracy of >96% for the in vivo classification as well as an excellent sensitivity of >88% for the in vitro mucosa lesions detection, and demonstrate sound potential of the system in early detection of mucosa lesions.

Flavinated SDHA Underlies the Change in Intrinsic Optical Properties of Oral Cancers

The molecular basis of reduced autofluorescence in oral squamous cell carcinoma (OSCC) cells relative to normal cells has been speculated to be due to lower levels of free flavin adenine dinucleotide (FAD). This speculation, along with differences in the intrinsic optical properties of extracellular collagen, lie at the foundation of the design of currently-used clinical optical detection devices. Here, we report that free FAD levels may not account for differences in autofluorescence of OSCC cells, but that the differences relate to FAD as a co-factor for flavination. Autofluorescence from a 70 kDa flavoprotein, succinate dehydrogenase A (SDHA), was found to be responsible for changes in optical properties within the FAD spectral region with lower levels of flavinated SDHA in OSCC cells. Since flavinated SDHA is required for functional complexation with succinate dehydrogenase B (SDHB), decreased SDHB levels were observed in human OSCC tissue relative to normal tissues. Accordingly, the metabolism of OSCC cells was found to be significantly altered relative to normal cells, revealing vulnerabilities for both diagnosis and targeted therapy. Optimizing non-invasive tools based on optical and metabolic signatures of cancers will enable more precise and early diagnosis leading to improved outcomes in patients.

Label-Free Optical Spectroscopy for Early Detection of Oral Cancer

Oral cancer is the 16th most common cancer worldwide. It commonly arises from painless white or red plaques within the oral cavity. Clinical outcome is highly related to the stage when diagnosed. However, early diagnosis is complex owing to the impracticality of biopsying every potentially premalignant intraoral lesion. Therefore, there is a need to develop a non-invasive cost-effective diagnostic technique to differentiate non-malignant and early-stage malignant lesions. Optical spectroscopy may provide an appropriate solution to facilitate early detection of these lesions. It has many advantages over traditional approaches including cost, speed, objectivity, sensitivity, painlessness, and ease-of use in clinical setting for real-time diagnosis. This review consists of a comprehensive overview of optical spectroscopy for oral cancer diagnosis, epidemiology, and recent improvements in this field for diagnostic purposes. It summarizes major developments in label-free optical spectroscopy, including Raman, fluorescence, and diffuse reflectance spectroscopy during recent years. Among the wide range of optical techniques available, we chose these three for this review because they have the ability to provide biochemical information and show great potential for real-time deep-tissue point-based in vivo analysis. This review also highlights the importance of saliva-based potential biomarkers for non-invasive early-stage diagnosis. It concludes with the discussion on the scope of development and future demands from a clinical point of view.

Raman Spectroscopy: A Potential Diagnostic Tool for Oral Diseases

Oral diseases impose a major health burden worldwide and have a profound effect on general health. Dental caries, periodontal diseases, and oral cancers are the most common oral health conditions. Their occurrence and development are related to oral microbes, and effective measures for their prevention and the promotion of oral health are urgently needed. Raman spectroscopy detects molecular vibration information by collecting inelastic scattering light, allowing a “fingerprint” of a sample to be acquired. It provides the advantages of rapid, sensitive, accurate, and minimally invasive detection as well as minimal interference from water in the “fingerprint region.” Owing to these characteristics, Raman spectroscopy has been used in medical detection in various fields to assist diagnosis and evaluate prognosis, such as detecting and differentiating between bacteria or between neoplastic and normal brain tissues. Many oral diseases are related to oral microbial dysbiosis, and their lesions differ from normal tissues in essential components. The colonization of keystone pathogens, such as Porphyromonas gingivalis, resulting in microbial dysbiosis in subgingival plaque, is the main cause of periodontitis. Moreover, the components in gingival crevicular fluid, such as infiltrating inflammatory cells and tissue degradation products, are markedly different between individuals with and without periodontitis. Regarding dental caries, the compositions of decayed teeth are transformed, accompanied by an increase in acid-producing bacteria. In oral cancers, the compositions and structures of lesions and normal tissues are different. Thus, the changes in bacteria and the components of saliva and tissue can be used in examinations as special markers for these oral diseases, and Raman spectroscopy has been acknowledged as a promising measure for detecting these markers. This review summarizes and discusses key research and remaining problems in this area. Based on this, suggestions for further study are proposed.

Detection of inaccessible head and neck lesions using human saliva and fluorescence spectroscopy

Head and neck cancer detection using fluorescence spectroscopy from human saliva is reported here. This study has been conducted on squamous cell carcinoma (SCC), and dysplastic (precancer) and control (normal) groups using an in-house developed compact set-up. Fluorescence set-up consists of a 375-nm laser diode and optical components. Spectral bands of flavin adenine dinucleotide (FAD), porphyrins, and Raman are observed in the spectral range of 400 to 800 nm. Presence of FAD and porphyrin bands in human saliva is confirmed by the liquid phantoms of FAD and porphyrin. Significant differences in fluorescence intensities among all the three groups are observed. Three spectral ranges from 455 to 600, 605 to 770, and 400 to 800 nm are selected for each group and area values under each spectral range are computed. To differentiate among the groups, receiver operating characteristic (ROC) analysis is employed on the area values. ROC differentiates among the groups with accuracies of 98%, 92.85%, and 81.13% respectively in the spectral ranges of 400 to 800 nm. However, in other two spectral ranges (455 to 600 and 605 to 770 nm), low accuracy values are found. Obtained accuracy values indicate that selection of human saliva for head and neck cancer detection may be a good alternative.

Bimodal multispectral imaging system with cloud-based machine learning algorithm for real-time screening and detection of oral potentially malignant lesions and biopsy guidance

SIGNIFICANCE

Screening and early detection of oral potentially malignant lesions (OPMLs) are of great significance in reducing the mortality rates associated with head and neck malignancies. Intra-oral multispectral optical imaging of tissues in conjunction with cloud-based machine learning (CBML) can be used to detect oral precancers at the point-of-care (POC) and guide the clinician to the most malignant site for biopsy.

AIM

Develop a bimodal multispectral imaging system (BMIS) combining tissue autofluorescence and diffuse reflectance (DR) for mapping changes in oxygenated hemoglobin (HbO2) absorption in the oral mucosa, quantifying tissue abnormalities, and guiding biopsies.

APPROACH

The hand-held widefield BMIS consisting of LEDs emitting at 405, 545, 575, and 610 nm, 5MPx monochrome camera, and proprietary Windows-based software was developed for image capture, processing, and analytics. The DR image ratio (R610/R545) was compared with pathologic classification to develop a CBML algorithm for real-time assessment of tissue status at the POC.

RESULTS

Sensitivity of 97.5% and specificity of 92.5% were achieved for discrimination of OPML from patient normal in 40 sites, whereas 82% sensitivity and 96.6% specificity were obtained for discrimination of abnormal (OPML + SCC) in 89 sites. Site-specific algorithms derived for buccal mucosa (27 sites) showed improved sensitivity and specificity of 96.3% for discrimination of OPML from normal.

CONCLUSIONS

Assessment of oral cancer risk is possible by mapping of HbO2 absorption in tissues, and the BMIS system developed appears to be suitable for biopsy guidance and early detection of oral cancers.

Biomedical applications of vibrational spectroscopy: Oral cancer diagnostics

Vibrational spectroscopy, based on either infrared absorption or Raman scattering, has attracted increasing attention for biomedical applications. Proof of concept explorations for diagnosis of oral potentially malignant disorders and cancer are reviewed, and recent advances critically appraised. Specific examples of applications of Raman microspectroscopy for analysis of histological, cytological and saliva samples are presented for illustrative purposes, and the future prospects, ultimately for routine, chairside in vivo screening are discussed.

Fourier Transform Infrared Spectroscopy in Oral Cancer Diagnosis

Oral cancer is one of the most common cancers worldwide. Despite easy access to the oral cavity and significant advances in treatment, the morbidity and mortality rates for oral cancer patients are still very high, mainly due to late-stage diagnosis when treatment is less successful. Oral cancer has also been found to be the most expensive cancer to treat in the United States. Early diagnosis of oral cancer can significantly improve patient survival rate and reduce medical costs. There is an urgent unmet need for an accurate and sensitive molecular-based diagnostic tool for early oral cancer detection. Fourier transform infrared spectroscopy has gained increasing attention in cancer research due to its ability to elucidate qualitative and quantitative information of biochemical content and molecular-level structural changes in complex biological systems. The diagnosis of a disease is based on biochemical changes underlying the disease pathology rather than morphological changes of the tissue. It is a versatile method that can work with tissues, cells, or body fluids. In this review article, we aim to summarize the studies of infrared spectroscopy in oral cancer research and detection. It provides early evidence to support the potential application of infrared spectroscopy as a diagnostic tool for oral potentially malignant and malignant lesions. The challenges and opportunities in clinical translation are also discussed.

Raman exfoliative cytology for prognosis prediction in oral cancers: A proof of concept study

Oral cancer is associated with high rates of recurrence, attributable to field cancerization. Early detection of advanced field changes that can potentially progress to carcinoma can facilitate timely intervention and can lead to improved prognosis. Previous in vivo studies have successfully detected advanced field effects in oral cancers. Raman exfoliative cytology has previously shown to differentiate normal, oral pre-cancer and cancers. The present study explores Raman-exfoliative-cytology-based detection of field effects. Exfoliated cells were collected from tumor (n = 16) and contralateral-normal appearing mucosa (n = 16) of oral cancer patients, and healthy tobacco habitués (n = 20). After spectral acquisition, specimens were Pap-stained for cytological evaluation. Data analysis, by Principal Component Analysis and Principal Component-Linear Discriminant Analysis, indicate several spectral-misclassifications between contralateral normal and tumor, which were investigated and correlated with spectral, cytological and clinical outcomes. A qualitative analysis by grouping patients with number of misclassifications with tumor (Group 1: 0, Group 2: 1 and Group 3: >1) was explored. Group 3 with highest misclassifications showed spectral and cytological similarity to tumor group - one patient was a case of early inoperable residual disease, despite clear margins on histopathology. Thus, these misclassifications could be indicative of cancer field changes, and can prospectively help to identify patients susceptible to recurrences .

Feasibility of a Video-Mosaicking Approach to Extend the Field-of-View For Reflectance Confocal Microscopy in the Oral Cavity In Vivo

BACKGROUND

Reflectance confocal microscopy (RCM) is a developing approach for noninvasive detection of oral lesions with label-free contrast and cellular-level resolution. For access into the oral cavity, confocal microscopes are being configured with small-diameter telescopic probes and small objective lenses. However, a small probe and objective lens allows for a rather small field-of-view relative to the large areas of tissue that must be examined for diagnosis. To extend the field-of-view for intraoral RCM imaging, we are investigating a video-mosaicking approach.

METHODS

A relay telescope and objective lens were adapted to an existing confocal microscope for access into the oral cavity. Imaging was performed using metal three-dimensional-printed objective lens front-end caps with coverslip windows to contact and stabilize the tissue and set depth. Four healthy volunteers (normal oral mucosa), one patient (with an amalgam tattoo) in a clinical setting, and 20 anesthetized patients (with oral squamous cell carcinoma [OSCC]) in a surgical setting were imaged. Instead of the usual still RCM images, videos were recorded and then processed into video-mosaics. Thirty video-mosaics were read and qualitatively assessed by an expert reader of RCM images of the oral mucosa.

RESULTS

Whereas the objective lens' native field-of-view is 0.75 mm × 0.75 mm, the video-mosaics display larger areas, ranging from 2 mm × 2 mm to 4 mm × 2 mm, with resolution, morphologic detail, and image quality that is preserved relative to that observed in the original videos (individual images). Video-mosaics in healthy volunteers' and the patients' images showed cellular morphologic patterns in the lower epithelium and at the epithelial junction, and connective tissue along with capillary loops and blood flow in the deeper lamina propria. In OSCC, tumor nests could be observed along with normal looking mucosa in margin areas.

CONCLUSIONS

Video-mosaicking is a reasonably quick and efficient approach for extending the field-of-view of RCM imaging, which can, to some extent, overcome the inherent limitation of an intraoral probe's small field-of-view. Reading video-mosaics can mimic the procedure for examining pathology: initial visualization of the spatial cellular and morphologic patterns of the tumor and the spread of tumor margins over larger areas of the lesion, followed by digitally zooming (magnifying) for closer inspection of suspicious areas. However, faster processing of videos into video-mosaics will be necessary, to allow examination of video-mosaics in real-time at the bedside. Lasers Surg. Med. 51:439-451, 2019. © 2019 Wiley Periodicals, Inc.

Raman spectroscopic analysis of the molecular composition of oral cavity squamous cell carcinoma and healthy tongue tissue

A Raman tissue spectrum is a quantitative representation of the overall molecular composition of that tissue. Raman spectra are often used as tissue fingerprints without further interpretation of the specific information that they contain about the tissue's molecular composition. In this study, we analyzed the differences in molecular composition between oral cavity squamous cell carcinoma (OCSCC) and healthy tissue structures in tongue, based on their Raman spectra. A total of 1087 histopathologically annotated spectra (142 OCSCC, 202 surface squamous epithelium, 61 muscle, 65 adipose tissue, 581 connective tissue, 26 gland, and 10 nerve) were obtained from Raman maps of 44 tongue samples from 21 patients. A characteristic, average spectrum of each tissue structure was fitted with a set of 55 pure-compound reference spectra, to define the best library of fit-spectra. Reference spectra represented proteins, lipids, nucleic acids, carbohydrates, amino acids and other miscellaneous molecules. A non-negative least-squares algorithm was used for fitting. Individual spectra per histopathological annotation were then fitted with this selected library in order to determine the molecular composition per tissue structure. The spectral contribution per chemical class was calculated. The results show that all characteristic tissue-type spectra could be fitted with a low residual of <4.82%. The content of carbohydrates, proteins and amino acids was the strongest discriminator between OCSCC and healthy tissue. The combination of carbohydrates, proteins and amino acids was used for a classification model of 'tumor' versus 'healthy tissue'. Validation of this model on an independent dataset showed a specificity of 93% at a sensitivity of 100%.

Chemometric analysis of integrated FTIR and Raman spectra obtained by non-invasive exfoliative cytology for the screening of oral cancer

FTIR spectroscopy and Raman spectroscopy of biological analytes are increasingly explored as screening tools for early detection of cancer.

Fully convolutional networks in multimodal nonlinear microscopy images for automated detection of head and neck carcinoma: Pilot study

BACKGROUND

A fully convolutional neural networks (FCN)-based automated image analysis algorithm to discriminate between head and neck cancer and noncancerous epithelium based on nonlinear microscopic images was developed.

METHODS

Head and neck cancer sections were used for standard histopathology and co-registered with multimodal images from the same sections using the combination of coherent anti-Stokes Raman scattering, two-photon excited fluorescence, and second harmonic generation microscopy. The images analyzed with semantic segmentation using a FCN for four classes: cancer, normal epithelium, background, and other tissue types.

RESULTS

A total of 114 images of 12 patients were analyzed. Using a patch score aggregation, the average recognition rate and an overall recognition rate or the four classes were 88.9% and 86.7%, respectively. A total of 113 seconds were needed to process a whole-slice image in the dataset.

CONCLUSION

Multimodal nonlinear microscopy in combination with automated image analysis using FCN seems to be a promising technique for objective differentiation between head and neck cancer and noncancerous epithelium.

In vivo detection of drug-induced apoptosis in tumors using Raman spectroscopy

We describe a label-free approach based on Raman spectroscopy, to study drug-induced apoptosis in vivo. Spectral-shifts at wavenumbers associated with DNA, proteins, lipids, and collagen have been identified on breast and melanoma tumor tissues. These findings may enable a new analytical method for rapid readout of drug-therapy with miniaturized probes.

Label-free Molecular Imaging and Analysis by Raman Spectroscopy

Raman scattering of a cell conveys the intrinsic information inherent to chemical structures of biomolecules. The spectroscopy of Raman scattering, or Raman spectroscopy, allows label-free and quantitative molecular sensing of a biological sample in situ without disruption. For the last five decades Raman spectroscopy has been widely utilized in biological research fields. However, it is just within the latest decade that molecular imaging and discrimination of living cells and tissues have become practically available. Here we overview recent progress in Raman spectroscopy and its application to life sciences. We discuss imaging of functional molecules in living cells and tissues; e.g., cancer cells and ischemic or infarcted hearts, together with a number of studies in the biomedical fields. We further explore comprehensive understandings of a complex spectrum by multivariate analysis for, e.g., accurate peripheral nerve detection, and characterization of the histological differences in the healing process of myocardial infarct. Although limitations still remain, e.g., weakness of the scattering intensity and practical difficulty in comprehensive molecular analysis, continuous progress in related technologies will allow wider use of Raman spectroscopy for biomedical applications.

Clinical applications of infrared and Raman spectroscopy: state of play and future challenges

This review examines the state-of-the-art of clinical applications of infrared absorption and Raman spectroscopy, outstanding challenges, and progress towards translation.

Label-free characterization of ultra violet-radiation-induced changes in skin fibroblasts with Raman spectroscopy and quantitative phase microscopy

Minimizing morbidities and mortalities associated with skin cancers requires sustained research with the goal of obtaining fresh insights into disease onset and progression under specific stimuli, particularly the influence of ultraviolet rays. In the present study, label-free profiling of skin fibroblasts exposed to time-bound ultra-violet radiation has been performed using quantitative phase imaging and Raman spectroscopy. Statistically significant differences in quantifiable biophysical parameters, such as matter density and cell dry mass, were observed with phase imaging. Accurate estimation of changes in the biochemical constituents, notably nucleic acids and proteins, was demonstrated through a combination of Raman spectroscopy and multivariate analysis of spectral patterns. Overall, the findings of this study demonstrate the promise of these non-perturbative optical modalities in accurately identifying cellular phenotypes and responses to external stimuli by combining molecular and biophysical information.

Superpixel-based spectral classification for the detection of head and neck cancer with hyperspectral imaging

Hyperspectral imaging (HSI) is an emerging imaging modality for medical applications. HSI acquires two dimensional images at various wavelengths. The combination of both spectral and spatial information provides quantitative information for cancer detection and diagnosis. This paper proposes using superpixels, principal component analysis (PCA), and support vector machine (SVM) to distinguish regions of tumor from healthy tissue. The classification method uses 2 principal components decomposed from hyperspectral images and obtains an average sensitivity of 93% and an average specificity of 85% for 11 mice. The hyperspectral imaging technology and classification method can have various applications in cancer research and management.