Intraoperative assessment of resection margins by Raman spectroscopy to guide oral cancer surgery

Patients with oral cavity cancer are almost always treated with surgery. The goal is to remove the tumor with a margin of more than 5 mm of surrounding healthy tissue. Unfortunately, this is only achieved in about 15% to 26% of cases. Intraoperative assessment of tumor resection margins (IOARM) can dramatically improve surgical results. However, current methods are laborious, subjective, and logistically demanding. This hinders broad adoption of IOARM, to the detriment of patients. Here we present the development and validation of a high-wavenumber Raman spectroscopic technology, for quick and objective intraoperative measurement of resection margins on fresh specimens. It employs a thin fiber-optic needle probe, which is inserted into the tissue, to measure the distance between a resection surface and the tumor. A tissue classification model was developed to discriminate oral cavity squamous cell carcinoma (OCSCC) from healthy oral tissue, with a sensitivity of 0.85 and a specificity of 0.92. The tissue classification model was then used to develop a margin length prediction model, showing a mean difference between margin length predicted by Raman spectroscopy and histopathology of -0.17 mm.

Development of a near-infrared Raman spectroscopy setup compatible with fluorescence-guided surgery

Near-infrared (NIR) fluorescence imaging using exogenous fluorescent agents provides whole-field images in real-time to assist the surgeon in the excision of a tumor. Although the method has high sensitivity, the specificity can sometimes be lower than expected. Raman spectroscopy can detect tumors with high specificity. Therefore, a combination of both techniques can be advantageous. A complication that must be addressed is that the NIR spectral region is favored by both techniques for (in vivo) tissue analysis. When fluorescence and Raman emissions spectrally overlap, it becomes challenging or impossible to detect the Raman signal. In this paper, by avoiding this overlap, we describe a Raman spectroscopy setup capable of recording high-quality Raman spectra from tissue containing NIR exogenous fluorescent agents. We identify an optimal wavelength interval (900-915 nm) for Raman excitation, which avoids both excitation of fluorescent dyes and Raman signal self-absorption by the tissue. In this way, Raman spectroscopy can be combined with the currently most-used NIR fluorescent dyes. This combined novel setup could pave the way for clinical trials benefiting from both fluorescence imaging and Raman spectroscopy to avoid positive margins in cancer surgery.

Intraoperative Assessment of Resection Margins in Oral Cavity Cancer: This is the Way

The goal of head and neck oncological surgery is complete tumor resection with adequate resection margins while preserving acceptable function and appearance. For oral cavity squamous cell carcinoma (OCSCC), different studies showed that only 15%-26% of all resections are adequate. A major reason for the low number of adequate resections is the lack of information during surgery; the margin status is only available after the final histopathologic assessment, days after surgery. The surgeons and pathologists at the Erasmus MC University Medical Center in Rotterdam started the implementation of specimen-driven intraoperative assessment of resection margins (IOARM) in 2013, which became the standard of care in 2015. This method enables the surgeon to turn an inadequate resection into an adequate resection by performing an additional resection during the initial surgery. Intraoperative assessment is supported by a relocation method procedure that allows accurate identification of inadequate margins (found on the specimen) in the wound bed. The implementation of this protocol resulted in an improvement of adequate resections from 15%-40%. However, the specimen-driven IOARM is not widely adopted because grossing fresh tissue is counter-intuitive for pathologists. The fear exists that grossing fresh tissue will deteriorate the anatomical orientation, shape, and size of the specimen and therefore will affect the final histopathologic assessment. These possible negative effects are countered by the described protocol. Here, the protocol for specimen-driven IOARM is presented in detail, as performed at the institute.

Raman spectroscopy for guidance of vulvar cancer surgery: a pilot study

For vulvar squamous cell carcinoma (VSCC), the mainstay of treatment is surgical removal with tumour-free margins. Surgeons still operate without objective tools that provide margin-status. This study assesses Raman spectroscopy potentiality for distinguishing ex-vivo VSCC from healthy tissue in 11 patients. Grid-based Raman maps were obtained from processed spectra. Water content and C-H band ratio (2,910-2,966 cm^-1 / 2810-2890 cm^-1) were calculated per spectrum and used as linear discriminant parameters. Healthy tissue was differentiated from VSCC with 0.90 discriminative power, 0.79 sensitivity and 0.86 specificity.This is an important step towards the development of objective tools for VSCC surgical guidance.

Performance of Intraoperative Assessment of Resection Margins in Oral Cancer Surgery: A Review of Literature

Introduction

Achieving adequate resection margins during oral cancer surgery is important to improve patient prognosis. Surgeons have the delicate task of achieving an adequate resection and safeguarding satisfactory remaining function and acceptable physical appearance, while relying on visual inspection, palpation, and preoperative imaging. Intraoperative assessment of resection margins (IOARM) is a multidisciplinary effort, which can guide towards adequate resections. Different forms of IOARM are currently used, but it is unknown how accurate these methods are in predicting margin status. Therefore, this review aims to investigate: 1) the IOARM methods currently used during oral cancer surgery, 2) their performance, and 3) their clinical relevance.

Methods

A literature search was performed in the following databases: Embase, Medline, Web of Science Core Collection, Cochrane Central Register of Controlled Trials, and Google Scholar (from inception to January 23, 2020). IOARM performance was assessed in terms of accuracy, sensitivity, and specificity in predicting margin status, and the reduction of inadequate margins. Clinical relevance (i.e., overall survival, local recurrence, regional recurrence, local recurrence-free survival, disease-specific survival, adjuvant therapy) was recorded if available.

Results

Eighteen studies were included in the review, of which 10 for soft tissue and 8 for bone. For soft tissue, defect-driven IOARM-studies showed the average accuracy, sensitivity, and specificity of 90.9%, 47.6%, and 84.4%, and specimen-driven IOARM-studies showed, 91.5%, 68.4%, and 96.7%, respectively. For bone, specimen-driven IOARM-studies performed better than defect-driven, with an average accuracy, sensitivity, and specificity of 96.6%, 81.8%, and 98%, respectively. For both, soft tissue and bone, IOARM positively impacts patient outcome.

Conclusion

IOARM improves margin-status, especially the specimen-driven IOARM has higher performance compared to defect-driven IOARM. However, this conclusion is limited by the low number of studies reporting performance results for defect-driven IOARM. The current methods suffer from inherent disadvantages, namely their subjective character and the fact that only a small part of the resection surface can be assessed in a short time span, causing sampling errors. Therefore, a solution should be sought in the field of objective techniques that can rapidly assess the whole resection surface.

Is the Depth of Invasion a Marker for Elective Neck Dissection in Early Oral Squamous Cell Carcinoma?

Objective

The depth of invasion (DOI) is considered an independent risk factor for occult lymph node metastasis in oral cavity squamous cell carcinoma (OCSCC). It is used to decide whether an elective neck dissection (END) is indicated in the case of a clinically negative neck for early stage carcinoma (pT1/pT2). However, there is no consensus on the cut-off value of the DOI for performing an END. The aim of this study was to determine a cut-off value for clinical decision making on END, by assessing the association of the DOI and the risk of occult lymph node metastasis in early OCSCC.

Methods

A retrospective cohort study was conducted at the Erasmus MC, University Medical Centre Rotterdam, The Netherlands. Patients surgically treated for pT1/pT2 OCSCC between 2006 and 2012 were included. For all cases, the DOI was measured according to the 8^th edition of the American Joint Committee on Cancer guideline. Patient characteristics, tumor characteristics (pTN, differentiation grade, perineural invasion, and lymphovascular invasion), treatment modality (END or watchful waiting), and 5-year follow-up (local recurrence, regional recurrence, and distant metastasis) were obtained from patient files.

Results

A total of 222 patients were included, 117 pT1 and 105 pT2. Occult lymph node metastasis was found in 39 of the 166 patients who received END. Univariate logistic regression analysis showed DOI to be a significant predictor for occult lymph node metastasis (odds ratio (OR) = 1.3 per mm DOI; 95% CI: 1.1-1.5, p = 0.001). At a DOI of 4.3 mm the risk of occult lymph node metastasis was >20% (all subsites combined).

Conclusion

The DOI is a significant predictor for occult lymph node metastasis in early stage oral carcinoma. A NPV of 81% was found at a DOI cut-off value of 4 mm. Therefore, an END should be performed if the DOI is >4 mm.

Intraoperative Assessment of the Resection Specimen Facilitates Achievement of Adequate Margins in Oral Carcinoma

Background

Inadequate resection margins in oral cavity squamous cell carcinoma have an adverse effect on patient outcome. Intraoperative assessment provides immediate feedback enabling the surgeon to achieve adequate resection margins. The goal of this study was to evaluate the value of specimen-driven intraoperative assessment by comparing the margin status in the period before and the period after the introduction of specimen-driven assessment as a standard of care (period 2010-2012 vs period 2013-2017).

Methods

A cohort of patients surgically treated for oral squamous cell carcinoma at the Erasmus MC Cancer Institute, Rotterdam, between 2010-2012 was studied retrospectively and compared to results of a prospectively collected cohort between 2013-2017. The frequency, type and results of intraoperative assessment of resection margins were analyzed.

Results

One hundred seventy-four patients were included from 2010-2012, 241 patients were included from 2013-2017. An increase in the frequency of specimen-driven assessment was seen between the two periods, from 5% in 2010-2012 to 34% in 2013-2017. When performing specimen-driven assessment, 16% tumor-positive resection margins were found in 2013-2017, compared to 43% tumor-positive resection margins overall in 2010-2012. We found a significant reduction of inadequate resection margins for specimen-driven intraoperative assessment (p < 0.001). Also, tumor recurrence significantly decreased, and disease-specific survival improved when performing specimen-driven intraoperative assessment.

Conclusions

Specimen-driven intraoperative assessment improves resection margins and consequently, the outcome of oral cancer patients. We advocate this method as standard of care.

Depth of invasion in early stage oral cavity squamous cell carcinoma: The optimal cut-off value for elective neck dissection

OBJECTIVES

Depth of invasion (DOI) is the most important predictor for lymph node metastasis (LNM) in early stage (T1-T2) oral cancer. The aim of this study is to validate the cut-off value of 4 mm on which the decision to perform an Elective Neck Dissection (END) is made.

MATERIALS AND METHODS

We performed a retrospective study in patients with pathologically proven early stage oral cavity squamous cell carcinoma (OCSCC) without clinical or radiological signs of LNM, who were treated between 2013 and 2018. An END was performed when DOI was ≥ 4 mm and a watchful waiting protocol was applied in patients with DOI < 4 mm.

RESULTS

Three hundred patients were included. END was performed in 77% of patients with DOI ≥ 4 mm, of which 36% had occult LNM (pN+). Patients in the watchful waiting group (48%) developed a regional recurrence in 5.2% for DOI < 4 mm and 24.1% for DOI ≥ 4 mm. For DOI ≥ 4 mm, regional recurrence free survival was higher for patients who were treated with END compared to watchful waiting (p = 0.002). A Receiver-Operator-Curve -analysis showed that a DOI cut-off value of 4.0 mm was the optimal threshold for the prediction of occult LNM (95.1% sensitivity, 52.9% specificity).

CONCLUSION

A DOI of ≥ 4 mm is an accurate cut-off value for performing an END in early stage OCSCC. END results in higher survival rates and lower regional recurrence rates in patients with DOI ≥ 4 mm.

Relocation of inadequate resection margins in the wound bed during oral cavity oncological surgery: A feasibility study

Background

Specimen‐driven intraoperative assessment of the resection margins provides immediate feedback if an additional excision is needed. However, relocation of an inadequate margin in the wound bed has shown to be difficult. The objective of this study is to assess a reliable method for accurate relocation of inadequate tumor resection margins in the wound bed after intraoperative assessment of the specimen.

Methods

During oral cavity cancer surgery, the surgeon placed numbered tags on both sides of the resection line in a pair‐wise manner. After resection, one tag of each pair remained on the specimen and the other tag in the wound bed. Upon detection of an inadequate margin in the specimen, the tags were used to relocate this margin in the wound bed.

Results

The method was applied during 80 resections for oral cavity cancer. In 31 resections an inadequate margin was detected, and based on the paired tagging an accurate additional resection was achieved.

Conclusion

Paired tagging facilitates a reliable relocation of inadequate margins, enabling an accurate additional resection during the initial surgery.

Raman spectroscopy for cancer detection and cancer surgery guidance: translation to the clinics

Oncological applications of Raman spectroscopy have been contemplated, pursued, and developed at academic level for at least 25 years. Published studies aim to detect pre-malignant lesions, detect cancer in less invasive stages, reduce the number of unnecessary biopsies and guide surgery towards the complete removal of the tumour with adequate tumour resection margins. This review summarizes actual clinical needs in oncology that can be addressed by spontaneous Raman spectroscopy and it provides an overview over the results that have been published between 2007 and 2017. An analysis is made of the current status of translation of these results into clinical practice. Despite many promising results, most of the applications addressed in scientific studies are still far from clinical adoption and commercialization. The main hurdles are identified, which need to be overcome to ensure that in the near future we will see the first Raman spectroscopy-based solutions being used in routine oncologic diagnostic and surgical procedures.

Improving clinical diagnosis of early-stage cutaneous melanoma based on Raman spectroscopy

Background

Clinical diagnosis of early melanoma (Breslow thickness less than 0.8 mm) is crucial to disease-free survival. However, it is subjective and can be exceedingly difficult, leading to missed melanomas, or unnecessary excision of benign pigmented skin lesions. An objective technique is needed to improve the diagnosis of early melanoma.

Methods

We have developed a method to improve diagnosis of (thin) melanoma, based on Raman spectroscopy. In an ex vivo study in a tertiary referral (pigmented lesions) centre, high-wavenumber Raman spectra were collected from 174 freshly excised melanocytic lesions suspicious for melanoma. Measurements were performed on multiple locations within the lesions. A diagnostic model was developed and validated on an independent data set of 96 lesions.

Results

Approximately 60% of the melanomas included in this study were melanomas in situ. The invasive melanomas had an average Breslow thickness of 0.89 mm. The diagnostic model correctly classified all melanomas (including in situ) with a specificity of 43.8%, and showed a potential improvement of the number needed to treat from 6.0 to 2.7, at a sensitivity of 100%.

Conclusion

This work signifies an important step towards accurate and objective clinical diagnosis of melanoma and in particular melanoma with Breslow thickness <0.8 mm.

Raman spectroscopy for assessment of bone resection margins in mandibulectomy for oral cavity squamous cell carcinoma

OBJECTIVES

The aim of this study was to investigate the potential of Raman spectroscopy for detection of oral cavity squamous cell carcinoma (OCSCC) in bone resection surfaces during mandibulectomy.

MATERIALS & METHODS

Raman mapping experiments were performed on fresh mandible resection specimens from patients treated with mandibulectomy for OCSCC. A tumour detection algorithm was created based on water concentration and the high-wavenumber range (2800 cm^-1-3050 cm^-1) of the Raman spectra.

RESULTS

Twenty-six ex vivo Raman mapping experiments were performed on 26 fresh mandible resection specimens obtained from 22 patients. The algorithm was applied on an independent test set and showed an accuracy of 95%, a sensitivity of 95%, and a specificity of 87%.

CONCLUSION

These results form the basis for further development of a Raman spectroscopy tool as an objective method for intraoperative assessment of bone resection margins.

A Novel Spectroscopically Determined Pharmacodynamic Biomarker for Skin Toxicity in Cancer Patients Treated with Targeted Agents

Raman spectroscopy is a noninvasive and label-free optical technique that provides detailed information about the molecular composition of a sample. In this study, we evaluated the potential of Raman spectroscopy to predict skin toxicity due to tyrosine kinase inhibitors treatment. We acquired Raman spectra of skin of patients undergoing treatment with MEK, EGFR, or BRAF inhibitors, which are known to induce severe skin toxicity; for this pilot study, three patients were included for each inhibitor. Our algorithm, based on partial least squares-discriminant analysis (PLS-DA) and cross-validation by bootstrapping, discriminated to variable degrees spectra from patient suffering and not suffering cutaneous adverse events. For MEK and EGFR inhibitors, discriminative power was more than 90% in the viable epidermis skin layer; whereas for BRAF inhibitors, discriminative power was 71%. There was a 81.5% correlation between blood drug concentration and Raman signature of skin in the case of EGFR inhibitors and viable epidermis skin layer. Our results demonstrate the power of Raman spectroscopy to detect apparition of skin toxicity in patients treated with tyrosine kinase inhibitors at levels not detectable via dermatological inspection and histological evaluation. Cancer Res; 77(2); 557-65. ©2016 AACR.

Water Concentration Analysis by Raman Spectroscopy to Determine the Location of the Tumor Border in Oral Cancer Surgery

Adequate resection of oral cavity squamous cell carcinoma (OCSCC) means complete tumor removal with a clear margin of more than 5 mm. For OCSCC, 85% of the surgical resections appear inadequate. Raman spectroscopy is an objective and fast tool that can provide real-time information about the molecular composition of tissue and has the potential to provide an objective and fast intraoperative assessment of the entire resection surface. A previous study demonstrated that OCSCC can be discriminated from healthy surrounding tissue based on the higher water concentration in tumor. In this study, we investigated how the water concentration changes across the tumor border toward the healthy surrounding tissue on freshly excised specimens from the oral cavity. Experiments were performed on tissue sections from 20 patients undergoing surgery for OCSCC. A transition from a high to a lower water concentration, from tumor (76% ± 8% of water) toward healthy surrounding tissue (54% ± 24% of water), takes place over a distance of about 4 to 6 mm across the tumor border. This was accompanied by an increase of the heterogeneity of the water concentration in the surrounding healthy tissue. The water concentration distributions between the regions were significantly different (P < 0.0001). This new finding highlights the potential of Raman spectroscopy for objective intraoperative assessment of the resection margins. Cancer Res; 76(20); 5945-53. ©2016 AACR.

Raman Spectroscopic Characterization of Melanoma and Benign Melanocytic Lesions Suspected of Melanoma Using High-Wavenumber Raman Spectroscopy

Melanoma is a pigmented type of skin cancer, which has the highest mortality of all skin cancers. Because of the low clinical diagnostic accuracy for melanoma, an objective tool is needed to assist clinical assessment of skin lesions that are suspected of (early) melanoma. The aim of this study was to identify spectral differences in the CH region of HWVN (high-wavenumber) Raman spectra between melanoma and benign melanocytic lesions clinically suspected of melanoma. We used these spectral differences to explore preliminary classification models to distinguish melanoma from benign melanocytic lesions. Data from 82 freshly excised melanocytic lesions clinically suspected of melanoma were measured using an in-house built Raman spectrometer, which has been optimized for measurements on pigmented skin lesions (excitation wavelength 976 nm and a wavelength range of the Raman signal 1340-1540 nm). Clear spectral differences were observed between melanoma and benign melanocytic lesions. These differences can be assigned mainly to the symmetric CH2 stretching vibrations of lipids. Our results show that the Raman bands between 2840 and 2930 cm(-1) have increased intensity for melanoma when compared to benign melanocytic lesions, suggesting an increase in lipid content in melanoma. These results demonstrate that spectroscopic information in the CH-stretching region of HWVN Raman spectra can discriminate melanoma from benign melanocytic lesions that are often clinically misdiagnosed as melanoma and that Raman spectroscopy has the potential to provide an objective clinical tool to improve the clinical diagnostic accuracy of skin lesions suspected of melanoma.

Development and validation of Raman spectroscopic classification models to discriminate tongue squamous cell carcinoma from non-tumorous tissue

BACKGROUND

Currently, up to 85% of the oral resection specimens have inadequate resection margins, of which the majority is located in the deeper soft tissue layers. The prognosis of patients with oral cavity squamous cell carcinoma (OCSCC) of the tongue is negatively affected by these inadequate surgical resections. Raman spectroscopy, an optical technique, can potentially be used for intra-operative evaluation of resection margins.

OBJECTIVE

To develop in vitro Raman spectroscopy-based tissue classification models that discriminate OCSCC of the tongue from (subepithelial) non-tumorous tissue.

MATERIALS AND METHODS

Tissue classification models were developed using Principal Components Analysis (PCA) followed by (hierarchical) Linear Discriminant Analysis ((h)LDA). The models were based on a training set of 720 histopathologically annotated Raman spectra, obtained from 25 tongue samples (11 OCSCC and 14 normal) of 10 patients, and were validated by means of an independent validation set of 367 spectra, obtained from 19 tongue samples (6 OCSCC and 13 normal) of 11 patients.

RESULTS

A PCA-LDA tissue classification model 'tumor' versus 'non-tumorous tissue' (i.e. surface squamous epithelium, connective tissue, muscle, adipose tissue, gland and nerve) showed an accuracy of 86% (sensitivity: 100%, specificity: 66%). A two-step PCA-hLDA tissue classification model 'tumor' versus 'non-tumorous tissue' showed an accuracy of 91% (sensitivity: 100%, specificity: 78%).

CONCLUSION

An accurate PCA-hLDA Raman spectroscopy-based tissue classification model for discrimination between OCSCC and (especially the subepithelial) non-tumorous tongue tissue was developed and validated. This model with high sensitivity and specificity may prove to be very helpful to detect tumor in the resection margins.

Investigation of the potential of Raman spectroscopy for oral cancer detection in surgical margins

The poor prognosis of oral cavity squamous cell carcinoma (OCSCC) patients is associated with residual tumor after surgery. Raman spectroscopy has the potential to provide an objective intra-operative evaluation of the surgical margins. Our aim was to understand the discriminatory basis of Raman spectroscopy at a histological level. In total, 127 pseudo-color Raman images were generated from unstained thin tissue sections of 25 samples (11 OCSCC and 14 healthy) of 10 patients. These images were clearly linked to the histopathological evaluation of the same sections after hematoxylin and eosin-staining. In this way, Raman spectra were annotated as OCSCC or as a surrounding healthy tissue structure (i.e., squamous epithelium, connective tissue (CT), adipose tissue, muscle, gland, or nerve). These annotated spectra were used as input for linear discriminant analysis (LDA) models to discriminate between OCSCC spectra and healthy tissue spectra. A database was acquired with 88 spectra of OCSCC and 632 spectra of healthy tissue. The LDA models could distinguish OCSCC spectra from the spectra of adipose tissue, nerve, muscle, gland, CT, and squamous epithelium in 100%, 100%, 97%, 94%, 93%, and 75% of the cases, respectively. More specifically, the structures that were most often confused with OCSCC were dysplastic epithelium, basal layers of epithelium, inflammation- and capillary-rich CT, and connective and glandular tissue close to OCSCC. Our study shows how well Raman spectroscopy enables discrimination between OCSCC and surrounding healthy tissue structures. This knowledge supports the development of robust and reliable classification algorithms for future implementation of Raman spectroscopy in clinical practice.

Towards oncological application of Raman spectroscopy

As the possibilities in the treatment of cancer continue to evolve, its early detection and correct diagnosis are becoming increasingly important. From the early detection of cancer to the guidance of oncosurgical procedures new sensitive in vivo diagnostic tools are much needed. Many studies report the Raman spectroscopic detection of malignant and premalignant tissues in different sites of the body with high sensitivities. The great appeal of this technique lies in its potential for in vivo clinical implementation. We present an overview of the in vitro and in vivo work on the oncological application of Raman spectroscopy and discuss its potential as a new tool in the clinico-oncological practice. Opportunities for integration of Raman spectroscopy in oncological cure and care as a real-time guidance tool during diagnostic (i.e. biopsy) and therapeutic (surgical resection) modalities as well as technical shortcomings are discussed from a clinician's point of view.

Discriminating basal cell carcinoma from perilesional skin using high wave-number Raman spectroscopy

An expanding body of literature suggests Raman spectroscopy is a promising tool for skin cancer diagnosis and in-vivo tumor border demarcation. The development of an in-vivo diagnostic tool is, however, hampered by the fact that construction of fiber optic probes suitable for Raman spectroscopy in the so-called fingerprint region is complicated. In contrast, the use of the high wave-number region allows for fiber optic probes with a very simple design. We investigate whether high wave-number Raman spectroscopy (2800 to 3125 cm(-1)) is able to provide sufficient information for noninvasive discrimination between basal cell carcinoma (BCC) and noninvolved skin. Using a simple fiber optic probe, Raman spectra are obtained from 19 BCC biopsy specimens and 9 biopsy specimens of perilesional skin. A linear discriminant analysis (LDA)-based tissue classification model is developed, which discriminates between BCC and noninvolved skin with high accuracy. This is a crucial step in the development of clinical dermatological applications based on fiber optic Raman spectroscopy.

Discrimination between Nontumor Bladder Tissue and Tumor by Raman Spectroscopy

We have applied Raman spectroscopy to discriminate between nontumor and tumor bladder tissue and to determine the biochemical differences therein. Tissue samples from 15 patients were collected, and frozen sections were made for Raman spectroscopy and histology. Twenty-five pseudocolor Raman maps were created in which each color represents a cluster of spectra measured on tissue areas of similar biochemical composition. For each cluster, the cluster-averaged spectrum (CAS) was calculated and classified as tumor and nontumor in accordance to pathohistology. Unguided hierarchical clustering was applied to display heterogeneity between and within groups of nontumor and tumor CAS. A linear discriminant analysis model was developed to discriminate between CAS from tumor and nontumor. The model was tested by a leave-one-patient-out validation, 84 of the 90 CAS (93%) were correctly classified with 94% sensitivity and 92% specificity. Biochemical differences between tumor and nontumor CAS areas were analyzed by fitting spectra of pure compounds to the CAS. Nontumor CAS showed higher collagen content while tumor CAS were characterized by higher lipid, nucleic acid, protein, and glycogen content. Raman spectroscopy enabled effective discrimination between tumor and nontumor bladder tissue based on characterized biochemical differences, despite heterogeneity expressed in both tumor and nontumor CAS.

Raman microspectroscopic mapping studies of human bronchial tissue

Characterization of the biochemical composition of normal bronchial tissue is a prerequisite for understanding the biochemical changes that accompany histological changes during lung cancer development. In this study, 12 Raman microspectroscopic mapping experiments are performed on frozen sections of normal bronchial tissue. Pseudocolor Raman images are constructed using principal component analysis and K-means cluster analysis. Subsequent comparison of Raman images with histologic evaluation of stained sections enables the identification of the morphologic origin (e.g., bronchial mucus, epithelium, fibrocollagenous stroma, smooth muscle, glandular tissue, and cartilage) of the spectral features. Raman spectra collected from the basal side of epithelium consistently show higher DNA contributions and lower lipid contributions when compared with superficial epithelium spectra. Spectra of bronchial mucus reveal a strong signal contribution of lipids, predominantly triolein. These spectra are almost identical to the spectra obtained from submucosal glands, which suggests that the bronchial mucus is mainly composed of gland secretions. Different parts of fibrocollagenous tissue are distinguished by differences in spectral contributions from collagen and actin/myosin. Cartilage is identified by spectral contributions of glycosaminoglycans and collagen. As demonstrated here, in situ analysis of the molecular composition of histologic structures by Raman microspectroscopic mapping creates powerful opportunities for increasing our fundamental understanding of tissue organization and function. Moreover, it provides a firm basis for further in vitro and in vivo investigations of the biochemical changes that accompany pathologic transformation of tissue.

Discriminating vital tumor from necrotic tissue in human glioblastoma tissue samples by Raman spectroscopy

Vital and necrotic glioblastoma tissues were studied by Raman microspectroscopy to identify possibilities for the development of an in vivo Raman method for real-time intraoperative brain biopsy guidance. The histologic malignancy grade of gliomas depends on the presence of parameters such as endothelial proliferation and necrosis, which are often not evenly distributed within the tumor. Because tissue samples obtained by stereotactic surgery are relatively small, sampling errors may easily occur by missing these crucial features. Although necrosis is important for grading, specimens containing only necrosis are diagnostically useless. Raman microspectroscopic mapping experiments were performed on unfixed cryosections of glioblastoma, obtained from 20 patients. After spectral acquisition, a clustering analysis was performed, resulting in groups of similar spectra. Each cluster was assigned a color, and pseudo-color Raman maps of the tissue sections were constructed. After the Raman experiments, the tissue sections were stained for histopathologic analysis, enabling identification of the histologic origin of the Raman spectra and assignment of the Raman spectral clusters to either vital or necrotic tissue. A classification model for discrimination between vital and necrotic tumor tissue based on linear discriminant analysis was developed. The classification model was evaluated using independent Raman data obtained from nine other tissue sections and yielded 100% accuracy. Information about the biochemical differences between necrosis and vital tumor was obtained by the analysis of difference spectra. Necrotic tissue was found to consistently contain higher levels of cholesterol (-esters). This in vitro result indicates that Raman spectra contain the information to distinguish vital glioblastoma from necrosis and makes Raman spectroscopy a powerful candidate for guidance of stereotactic brain biopsy.

Discriminating basal cell carcinoma from its surrounding tissue by Raman spectroscopy

The objective of this in vitro study was to explore the applicability of Raman spectroscopy to distinguish basal cell carcinoma from its surrounding noncancerous tissue; therefore, identifying possibilities for the development of an in vivo diagnostic technique for tumor border demarcation. Raman spectra were obtained in a two-dimensional grid from unstained frozen sections of 15 basal cell carcinoma specimens. Pseudo-color Raman images were generated by multivariate statistical analysis and clustering analysis of spectra and compared with histopathology. In this way a direct link between histologically identifiable skin layers and structures and their Raman spectra was made. A tissue classification model was developed, which discriminates between basal cell carcinoma and surrounding nontumorous tissue, based on Raman spectra. The logistic regression model, shows a 100% sensitivity and 93% selectivity for basal cell carcinoma. The Raman spectra were, furthermore, used to obtain information about the differences in molecular composition between different skin layers and structures. An interesting finding was that in four samples of nodular basal cell carcinoma, the collagen signal contribution in spectra of dermis close to a basal cell carcinoma, was markedly reduced. The study demonstrates the sensitivity of Raman spectroscopy to biochemical changes in tissue accompanying malignancy, resulting in a high accuracy when discriminating between basal cell carcinoma and noncancerous tissue.