Transscleral visible/near-infrared spectroscopy for quantitative assessment of melanin in a uveal melanoma phantom of ex vivo porcine eyes
Introduction
Ultrasonography plays an important role in the characterization of intraocular tumors, including uveal melanomas, which typically demonstrate medium to low internal reflectivity and characteristic features like acoustic hollowness and choroidal excavation (Sobottka et al., 1998). Computed tomography and magnetic resonance imaging (MRI) have proved to be accurate in determining the location and size of intraocular tumors and evaluating possible extraocular tumor extension, but lesions with a thickness of less than 2–3 mm are generally poorly visualized by these imaging techniques (Peyman and Mafee, 1987). Owing to the paramagnetic properties of melanin, uveal melanomas often appear hyperintense on T1-weighted and hypointense on T2-weighted MRI (Mafee et al., 1986). However, the predictive value of MRI is limited by overlap of signal intensities between uveal melanomas and intraocular hemorrhages, metastases and other solid tumors (Ferris et al., 1993, Lemke et al., 2001). In general, current imaging modalities provide precise information on the topography of intraocular lesions, but there is still a need to further refine the analysis of tumor tissue and morphology by ancillary diagnostic methods.
Visible/near-infrared spectroscopy (Vis/NIRS) is an optical non-invasive or minimally invasive method that offers qualitative and quantitative analysis of tissues (Richards-Kortum and Sevick-Muraca, 1996). It relies on measuring the absorption imprint by different tissue chromophores, e.g., oxy- and deoxyhemoglobin, melanin, carotenes, lipids, and water. The near-infrared part of the spectrum (700–1000 nm) has a good penetration depth in tissue. However, in this wavelength range, scattering dominates heavily over absorption. The scattering originates from refractive-index mismatches within the cellular or extracellular tissue structures, whereas absorption is caused by electronic transitions within the chromophore molecules. The interpretation of the resulting absorption and scattering spectra is often combined with models of light propagation and statistical regression models to predict the composition of unknown tissue samples. Based on spectral differences associated with parameters like edema, fibrosis, vascularization, and oxygenation, Vis/NIRS has been successfully used to detect and characterize cancer and precancerous lesions in the breast (Tromberg et al., 2005), prostate (Svensson et al., 2007), lung (Bard et al., 2006), female reproductive organs (Hornung et al., 1999), and gastrointestinal tract (Kondepati et al., 2007). Because optical spectroscopy provides quantitative information about specific chromophores such as melanin and hemoglobin, the method has also been comprehensively investigated for the evaluation of melanocytic skin tumors (McIntosh et al., 2001, Murphy et al., 2005). The spectroscopic identification of melanin is difficult because it has no distinct absorption bands. The light absorption increases linearly in the wavelength range between 720 and 620 nm and then exponentially toward shorter wavelengths. According to Kollias and Baqer, 1985, Kollias and Baqer, 1987, the gradient of this absorption spectrum is a sensitive indicator for the total melanin content in the skin. Similarly, Dwyer et al. (1998) found that the absorbance difference between 400 nm and 420 nm makes it possible to estimate the density of cutaneous melanin. Moreover, it has been shown that laser-induced fluorescence of melanin could serve as a sensitive approach for early diagnosis of melanomas (Schneider et al., 2005, Eichhorn et al., 2009).
Transpupillary reflectance spectroscopy has been widely applied to study pathophysiological aspects of pigmentation and vascularization in the ocular fundus (Hammer and Schweitzer, 2002, Hardarson et al., 2006, van de Kraats et al., 2008). For the purpose of analyzing choroidal tumors, the usefulness of this approach is limited by media opacities, the retinal pigment epithelium (RPE), and possible retinal hemorrhages or metaplasia that may obscure the real nature of the lesions. The relative homogeneous structure of the sclera, the location of various choroidal lesions immediately underneath it, and the fact that melanin is a spectroscopically quantifiable marker suggest that optical spectroscopy via a transscleral approach could be of value in the differentiation of choroidal tumors. Herein, we report the feasibility and accuracy of using transscleral Vis/NIRS to estimate the content of natural melanin in a novel uveal melanoma phantom of ex vivo porcine eyes.
Section snippets
Porcine eyes
Thirty eyes from domestic pigs (Norwegian Landrace) of both sexes, each with a live weight of about 75 kg and an age of 6–7 months, were obtained from a local abattoir. The enucleations were carried out within 12 h post-mortem, and the eyes were stored at 4 °C in a moist chamber until preparation. All experiments were performed within 3–5 days after the death of the animals. The study adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.
Preparation of injection suspensions
A freshly prepared
Uveal melanoma phantoms
The injection procedure was successfully performed in all eyes. Ultrasonography showed that the suspension was localized within a limited area of the suprachoroidal space without any penetration through the retina or into the vitreous cavity. In all cases, the phantom presented as a solitary dome-shaped tumor with medium internal reflectivity (Fig. 2A). In the melanin concentration groups of 1 mg/ml (n = 10), 2 mg/ml (n = 10), and 3 mg/ml (n = 10), the largest basal phantom diameters
Discussion
Within the field of ocular oncology, the technical complexity and potential risks associated with intraocular biopsies have led to a demand for ancillary diagnostic tests. The present study demonstrates that transscleral Vis/NIRS is a possible diagnostic method for predicting the content of melanin in choroidal tumors. To the best of our knowledge, the feasibility and utility of using this method to evaluate choroidal lesions have not been previously reported. The reasons for deviating from a
Acknowledgments
The study was supported by grants from the Western Norway Regional Health Authority and a Linnaeus grant for the Lund Laser Centre.
References (40)
- et al.
Optical spectroscopy for the classification of malignant lesions of the bronchial tree
Chest
(2006) - et al.
Spectroscopic characteristics of human melanin in vivo
J. Invest. Dermatol.
(1985) - et al.
Absorption mechanisms of human melanin in the visible, 400–720 nm
J. Invest. Dermatol.
(1987) - et al.
Towards non-invasive screening of skin lesions by near-infrared spectroscopy
J. Invest. Dermatol.
(2001) - et al.
Human sclera: thickness and surface area
Am. J. Ophthalmol.
(1998) - et al.
Uveal melanoma and similar lesions: the role of magnetic resonance imaging and computed tomography
Radiol. Clin. North Am.
(1987) - et al.
Quantitative scattering of melanin solutions
Biophys. J.
(2006) - et al.
Optical properties of melanin in the skin and skin-like phantoms
Proc. SPIE
(2000) - et al.
The use of spectrophotometry to estimate melanin density in Caucasians
Cancer Epidemiol. Biomarkers Prev.
(1998) - et al.
Early diagnosis of melanotic melanoma based on laser-induced melanin fluorescence
J. Biomed. Opt.
(2009)