New light shines on melanoma detection

Melanin is the characteristic chromophore (a molecule that exhibits color) of human skin with several proposed functions including protection from solar radiation, antioxidant defense, and camouflage.¹ Melanin is also involved in several conditions and diseases such as malignant melanoma, an aggressive skin cancer with high metastatic potential. Despite its importance, melanin is poorly understood. Attempts to elucidate melanin structure are hampered by the lack of effective physicochemical methods because, in contrast to other biopolymers such as proteins and nucleic acids, melanin resists chemical analysis due to the strong non-hydrolyzable carbon-carbon bonds linking its monomers.²

A better understanding of the structure and function of melanin is crucial to the assessment of skin color and for the study of associated pathological conditions such as albinism, vitiligo, melasma, lentigo, and melanoma. Of particular importance is the investigation of melanoma and its potential precursor conditions which may include melanocytic nevi (common skin moles), dysplastic or atypical nevi, and melanoma in situ (where the atypical melanocytes are confined between the dermal and epidermal layers). In light of reports of altered melanin production in melanoma and dysplastic nevi, the characterization and detection of melanocytic lesions through the study of melanin is of particular relevance.

Figure 1. Melanin absorption values, k_m, for malignant melanoma, melanoma in situ, and dysplastic nevi for four different independent studies (A, B, C, D). The k_m values from all four studies illustrate clear and consistent differences among the different skin lesion types.

One way to non-invasively characterize melanin in vivo is through its optical properties. The absorption spectrum of melanin has previously been investigated using diffuse reflectance spectroscopy. We have recently performed a more detailed quantitative study of melanin absorption using a similar technique.³ The study included both dysplastic nevi and malignant melanoma lesions. We found that melanin absorption depends on the wavelength and that a single parameter can describe the absorption curve: the exponential decay slope, k_m.

In addition to our own data, we have analyzed average diffuse-reflectance spectra from three similar, independent, previously published studies. The results of the analysis of the data from all four studies were in agreement, confirming our own observations (see Figure 1 ).

In all studies, melanoma is characterized by lower absorption values, k_m, compared to skin nevi, while the values for melanomas in situ fall in between the melanoma and nevi values.

Our results may be summarized in four key findings. First, melanin absorption can be measured directly in vivo using diffuse reflectance spectroscopy. Secondly, in vivo melanin absorption exhibits an exponential dependence on wavelength, an observation consistent with in vitro results. Therefore, the melanin absorption spectral curve can be described by the exponential decay slope. Thirdly, our results on the study of melanoma and dysplastic nevi indicate intrinsic differences in melanin absorption values among these skin lesion types. Finally, intrinsic changes associated with the histological transition from dysplastic nevi to melanoma in situ to melanoma are reflected in the optical absorption spectrum of melanin.

We expect that the study of melanin absorption in vivo will open new ways for the early detection of melanoma as well for the non-invasive characterization of a wide variety of pigmented skin lesions. In addition, the technique offers promise for the understanding of melanin and melanin structure in its native environment, i.e., within living human skin. We plan to explore the utility of our findings in the aforementioned directions.