In vivo detection of changes in cutaneous carotenoids after chemotherapy using shifted excitation resonance Raman difference and fluorescence spectroscopy

Various cutaneous toxicities under chemotherapy indicate a local effect of chemotherapy by secretion after systemic application. Here, changes in the fluorescence and Raman spectral properties of the stratum corneum subsequent to intravenous chemotherapy were assessed.


| INTRODUC TI ON
Cutaneous toxicities count to the most frequent side effects during chemotherapy. 1,2 Previous studies showed that intravenously applied chemotherapeutics can be found within the sweat being secreted to the skin surface. The chemotherapeutics subsequently spread on the skin surface as if topically applied and re-penetrate into the upper skin layers. 3 Here, they can lead to radical formation and inflammatory or toxic skin effects, including development of palmar-plantar erythrodysaesthesia, also known as a hand-foot syndrome. [4][5][6][7][8][9][10] The skin of healthy volunteers, especially the stratum corneum layer, usually contains a high concentration of antioxidants. Among them are carotenoids, vitamins, and enzymes, which form an antioxidant network and serve as a part of the body's protective system against free radicals. Recent studies show that carotenoids serve as marker substances of the entire antioxidant status of the epidermis in vivo 11,12 and the kinetics of their degradation in the skin show the intensity of influencing stress factors. 13 The kinetics of inverse penetration of doxorubicin on the skin surface were described previously. 3 It was found that 30 minutes to 1 hour after systemic administration after chemotherapeutic infusion, fluorescence signals of doxorubicin were detectable on the skin surface. 3 This leads to the conclusion that doxorubicin, like carotenoids and vitamin E, too, is secreted to the skin surface with the sweat, spreads there, and then penetrates into the stratum corneum like topically applied. 14 This result also explains why the dermal side effects associated with systemic administration of doxorubicin occur mainly in the palms of the hands and the soles of the feet. The highest sweat gland density is present at these skin sites 15 so that the proportion of escaping doxorubicin is highest here as well. The horny layer is ten to twenty times thicker on the palms and soles of the feet than on the other areas of the skin, providing an ideal reservoir for the absorption of sweat-derived substances, such as chemotherapeutics in human skin. Depending on the designated chemotherapy schedule and dose, multiple cycles of chemotherapy can cause an accumulation of chemotherapeutic substances 16 resolving in toxic local effects on the skin. However, the specific quantities and dynamics for different chemotherapeutics are not fully understood. 17 Many chemotherapeutics are Raman-active substances, but their direct detection on the skin is hardly possible due to their low concentration and superposition with the skin Raman spectrum. 18,19 In the case of doxorubicin, absorption bands in the range of 440-520 nm 20 and fluorescence in the range of 520-630 nm 21 are known. This means that the excitation around 488 nm resonantly excites not only the carotenoids beta-carotene and lycopene 22 but also a doxorubicin fluorescence signal 23,24 in the skin. Thus, on the one hand, the doxorubicin fluorescent signal acts as a background signal for the Raman signal.
On the other hand, this fluorescent signal makes it possible to detect doxorubicin very sensitively in human skin under in vivo conditions. Noninvasive reflection spectroscopy was used in a previous study to investigate the decrease in cutaneous carotenoids as a result of increased skin radical formation following intravenous administration of chemotherapeutic agents to the patient's palms. The results clearly showed the significant decrease in cutaneous carotenoids in all intravenously administered chemotherapeutic agents. 25 This clear decrease is detectable using even less sensitive techniques such as reflection spectroscopy. 22 Therefore, a novel diagnostic system would be not only of great importance for the direct detection of doxorubicin on the skin but also for indirect detection of a whole range of other chemotherapeutic agents by measuring their influence on cutaneous carotenoids.
In order to be able to quantitatively determine small changes in the Raman signal intensity of cutaneous carotenoids in vivo, a fluorescence background subtraction procedure should be performed. This can be done by taking advantage of the fluorescence photo-bleaching effect by prolonged exposure of the skin with the reference light. 26 However, this method is time-consuming and did not provide complete subtraction of the fluorescence background. 27  Here, changes in the fluorescence and SERRDS signal intensities before and after intravenous chemotherapy were assessed in vivo in cancer patients.
The assessments within this study aimed at determining changes in the carotenoid concentration of the skin as well as detecting different chemotherapeutics by fluorescence changes after intravenous application.

| Measurement system
A miniaturized measurement system based on SERRDS 28 was used for the assessment of changes in cutaneous carotenoids and fluorescence signals on the skin surface in vivo. The system uses a measuring spot diameter of 3 mm and a diode laser-based 488 nm SHG light source providing two excitation wavelengths λ 1 = 487.2 nm and λ 2 = 487.6 nm.
Here, the fluorescence background can be separated from the Raman peaks. The system was calibrated to a detection limit of 0.03 nmol g −1 beta-carotene per gram of skin and was described in detail previously. 28 The carotenoids' signal was recognizable at approx. 1525 cm −1 . 22

| Ethical approval
Prior to initiation of the study, approval by the independent Ethics

| Statistical analysis
The descriptive and statistical analysis of the obtained data was conducted using IBM SPSS vs 22. Analysis of SERRDS values was subject to Mann-Whitney U test, in which P-values of less than .05 were considered to indicate statistical significance.

| Fluorescence analysis
The mean fluorescence signal intensity increased by 1.2 ± 0.3 at T 1 showing an increase in 14 out of the 20 cancer subjects.
As shown in Figure 1, the majority of patients showed an increase in mean fluorescence intensity after the end of chemotherapeutic treatment at T 1 , while the fluorescence intensity in some patients remained almost unchanged (patients 6 and 12) and even decreased in four patients (patients 1, 3, 13, and 15). The highest increases in fluorescence signal were seen in patients receiving doxorubicin, a member of the anthracycline group, which is known for generating its own fluorescence, which was expected to increase the fluorescence of the skin within these investigations. Accordingly, an average increase of 1.3 ± 0.1 was assessed for patients receiving doxorubicin (patients 4, 9, 17, 19, and 20). A

Patient number
Chemotherapeutic substance

| SERRDS analysis: Carotenoids
Healthy subjects showed a mean SERRDS carotenoid intensity at 1136.4 a.u., which was significantly higher (P < .001) than cancer subjects before chemotherapy at T base with a mean SERRDS carotenoid intensity of 435.6 a.u. (Figure 2).
As shown in Figure 3, the intensity of the SERRDS signals, representing the concentration of carotenoids, decreases in the majority of patients (13 out of 20) after administration of chemotherapy.
Patients receiving anthracyclines showed a mean decrease in SERRDS signal intensity of carotenoids of 32.8 a.u., while a mean decrease of SERRDS signal intensity in patients receiving alkaloids was found at 21.6 a.u. (Figure 4). The results obtained show that some of the fluorescence-active chemotherapeutic agents, for example, doxorubicin or epirubicin can be detected due to enhanced skin fluorescence as expected. 31 The direct detection and identification of all fluorescence-free chemotherapeutic agents in the skin using SERRDS was not observed within this study and can be subjected to further investigations.

ACK N OWLED G EM ENTS
Financial support for the investigations was provided by the Einstein Foundation Berlin.

CO N FLI C T O F I NTE R E S T
There is no conflict of interest to declare.