The MPT/FLIM technique has numerous applications in dermatology, being suitable for the study of physiological and pathological conditions of the skin, in vivo, on ex vivo samples and on cultured cells [49, 50]. Mesenchymal stem cells undergoing various differentiation stimuli can be monitored in a non-invasive manner evaluating morphology, metabolic activity and oxidative stress [49]. Fibroblast cultures are widely used as an experimental model to study the expression of specific genes or the effect of drugs, especially of new compounds with potential chemotherapeutic activity, and to check the mutagenicity and carcinogenicity of different substances. Using MPT and FLIM, a precise and rapid assessment of the morphologic and metabolic changes which fibroblasts undergo after exposure to various environmental factors can be achieved without the need of cell processing and staining [50].

When studying diseased states of the skin, knowledge about normal morphology is of utmost importance. As regards healthy epidermis, it has been shown that cell and nucleus diameters, cell density and fluorescence lifetime values vary not only according to epidermal cell depth but also depending on skin site [51]. Skin ageing is expressed by variations in shape, size, distribution and morphology of epidermal cells. Moreover, FLIM values at both the upper and lower layers increase, indicating a change in the metabolic activity of epidermal cells in the elderly [52]. These data can be used for the comparison with MPT/FLIM aspects of epidermal cells in pathological conditions.

The application of MPT/FLIM to the field of skin tumours appears promising [25, 35, 52–55].

Employing MPT, a rapid and accurate tumour demarcation from the surrounding noninvolved skin can be achieved [25, 56]. Considering both autofluorescence images and fluorescence lifetime images Galletly NP et al. identified several alterations of cell metabolism and morphology, allowing the discrimination between healthy skin and the tumour region. On the basis of wide-field imaging lifetime information, they demonstrated that BCC exhibits longer fluorescence lifetime than surrounding healthy skin [25]. These observations were confirmed by another working group, which described a shift of the mean lifetime distribution of BCC towards longer values, especially in the intermediate epidermal layers [52].

Another feature revealed by MPT is the alteration of the extracellular matrix in the BCC stroma [46]. In BCC, dermal collagen is replaced by tumour nests of basaloid cells which destroy the surrounding environment by the collagenolytic activity of matrix metalloproteinases. The result consists in a disarrangement of the collagen molecule packing, thus altering the dermal distribution of the fluorophores and decreasing the SHG signal [46, 56, 57]. The increase in the autofluorescence signal in the cancer stroma, in comparison with that of normal dermal stroma, is unclear at this stage [56].

Lin et al. performed a quantitative analysis to discriminate normal dermal stroma from the tumour; in their approach, multiphoton fluorescence (MF) pixel number (a) and SHG pixel number (b) were determined within a region of interest, and the MF/SHG index was defined to be (a  −  b)/(a  +  b). Consequently, MF/SHG reached the maximum value of 1 when only the MF signal was present (corresponding to the presence of only elastic fibres) and decreased with the growth of collagen content (which generates the SHG signal). Whereas in normal dermal stroma MF/SHG was very low, reflecting the high content of collagen bundles, in the cancer stroma it was significantly higher because of the low contribution of undamaged collagen [46].

The epidermis overlying the BCC may be thinned, thickened or ulcerated compared to normal skin [52]; conversely, all the ex vivo specimens investigated by Paoli et al. showed a marked increase in epidermal thickness compared to the corresponding normal perilesional skin [53, 55].

BCC is characterised by clumps of autofluorescent cells in the dermis [46] which show relatively large nuclei, little cytoplasm [55] and a higher nucleus to cytoplasm ratio [46, 56]. Peripheral basaloid cells palisading along the basement membrane [46, 56] are a distinctive histopathologic feature of basal cell carcinoma. They can also be observed in each cancer nest.

Paoli et al. described tumour cells which were monomorphous and palisading at the periphery, with nuclei polarisation corresponding to keratinocytes with elongated nuclei and cytoplasm oriented in the same direction in the X−Y plane [53, 55].

De Giorgi et al. analysed different surgical samples of BCC investigating cell shape and size: compared with healthy skin cells, tumoural ones appeared smaller in dimensions and tightly packed together; they also exhibited a reduced fluorescent contrast between cytoplasm and nucleus [54].

Speckled perinuclear fluorescence was sometimes observed in subcorneal epidermis of superficial BCC but was also present in the corresponding normal perilesional skin [53].

In a study on 98 BCCs, Seidenari et al. identified three epidermal and seven tumour descriptors for BCC [58]. Compared to normal epidermal cells, those overlying the BCC exhibited irregular cellular contours, had lost the normal cohesion and were disposed in a random order; moreover, intercellular spaces were larger and irregular (Fig. 18.1). BCC cells appeared monomorphous with an elongated shape and nuclei; they were aligned along the same direction and tightly packed together (Fig. 18.2a); sometimes a double alignment was observable with divergent sheets of cells (Fig. 18.2b). At the edge of the nodule, a palisading phenomenon was often visible (Fig. 18.2c). By an excitation wavelength of 760 nm, basaloid nodules were observable as cell aggregates surrounded by fibres (Fig. 18.3). Whereas BCC cells exhibited a bluish colour corresponding to a high fluorescence decay time, the collagen fibres appeared in red due to very low fluorescence lifetimes (Fig. 18.3). When shifting the excitation wavelength to 800–820 nm, to explore the extracellular matrix, BCC cells disappeared and only red fibres surrounding black spaces were visible (phantom islands) (Fig. 18.4). Finally, BCC cells showed mean fluorescence lifetime values which were significantly higher with respect to normal epidermal cells, indicating that FLIM may also provide quantitative data for the identification of single tumour cells.

The diagnostic aid provided by MPT can sometimes be limited by the deep location of the tumour nests, especially in the nodular BCC type [47] or by epidermal thickening and hyperkeratosis [55] which makes the exploration of the tumour mass difficult.

MPT can also be very useful for the investigation of surgically removed samples before pathologist’s examination on ex vivo specimens [47] or as an alternative to Mohs’ surgery.

Skala and colleagues investigated low- and high-grade epithelial precancers compared with normal epithelial tissues. A significant decrease in the contribution of protein-bound NADH and a significantly increased variability in the redox ratio and NADH fluorescence lifetime were observed in precancerous cells [35]. Precancerous and cancerous samples generally show increased keratin thickness, increased epithelial thickness, decreased nuclear density ratio and decreased keratin layer fluorescence [59]. Moreover, cellular fluorescence becomes progressively weaker and more perinuclear in precancerous and cancerous tissue when compared with normal tissue [60].

Paoli et al. investigated five squamous cell carcinoma specimens. In these samples the stratum corneum was abnormally thick compared to the corresponding perilesional skin. In the same specimens, fluorescent nuclei compartments within the corneocytes were observed. This morphological and fluorescent feature had been described earlier in the presence of hyperkeratosis [20]. In two hyperkeratotic lesions, large, rounded bundles of keratin were observed within the SC, corresponding to the so-called keratin pearls.

In all but one lesion, the keratinocytes within the SG, the SS and the SB were irregularly distributed, reflecting a loss of cell polarity. Dimly fluorescent and widened intercellular spaces were present in three of these four lesions. Signs of bowenoid dysplasia, including pleomorphism and substantial variation of nuclei size, were observed within the epidermis of four lesions. Large, multinucleated cells were noticed in all five lesions. Keratinocytes with brighter cytoplasmic fluorescence compared to surrounding cells in the SG and the SS were discerned in two lesions, possibly corresponding to dyskeratosis. Four lesions presented speckled perinuclear fluorescence in the SG and/or the SS, whereas this feature was only observed in one corresponding normal perilesional skin sample.

Melanocytic lesions are easily investigated when considering that melanin is a fluorophore characterised by emission lifetimes different from those of other endogenous fluorophores. Dimitrow et al. investigated some procedures of selective melanin imaging and spectral fluorescence lifetime imaging in combination with high-resolution multiphoton laser tomography by analysing 46 melanocytic lesions of human skin [23]. Remarkable differences in lifetime behaviour of keratinocytes in contrast to melanocytes were observed. The fluorescence lifetime distribution was found in correlation with the intracellular amount of melanin. Examining the unique light absorption behaviour of melanin, they found a selective fluorescence of melanin-containing cells at an excitation wavelength of 800 nm. Their results revealed that keratinocyte lifetime values correspond to NAD(P)H and melanocyte lifetime values to the endogenous fluorophore melanin. Thus, whereas FLIM on single data points allowed a clear separation of different melanised cell types (e.g. keratinocytes, melanocytes), it seemed to be unsuitable for a classification of benign versus malignant melanocytic skin lesions. Both melanocytic nevi and MM were characterised by enhanced cell density of melanocytes primarily in the basal epidermal layer. Being the alteration of the upper epidermal layers a unique characteristic of malignant proliferation, the crucial point of differentiation was identified in the presence of ascending melanocytes. Spectral analysis of MM revealed a main fluorescence peak around 470 nm in combination with an additional peak close to 550 nm throughout all epidermal layers. From this they concluded that morphological alterations, in combination with a selective excitability at 800 nm, a short medium lifetime and a distinct peak around 550 nm in suprabasal epidermal layers, are highly suggestive of MM.