Photodynamic Therapy in Skin Cancer

Introduction

Photodynamic therapy (PDT) is a non-invasive therapy with proven efficacy in non-melanoma skin cancer (NMSC). PDT involves the activation of a photosensitizing drug by visible light to produce activated oxygen species within target cells, resulting in their destruction. Initially, systemic administration of photosensitizers was required, adding complexity to the procedure and resulting in the complication of generalized photosensitivity that could last several weeks. Subsequently, the use of the topically active agent 5-aminolevulinic acid (5-ALA), a precursor of the endogenous photosensitizer protoporphyrin IX (PpIX), was described, permitting simplification of the treatment process.

Evidence-based guidelines indicate that topical PDT is effective in actinic keratoses (AK), Bowen's disease (squamous cell carcinoma in situ), superficial basal cell carcinomas (BCC) and thin nodular BCC ( Table 45.1 ). Additional evidence exists indicating the potential of topical PDT in treating localized plaques of cutaneous T-cell lymphoma, for epidermal dysplasias in organ transplant recipients (OTR), and as adjunctive therapy in extramammary Paget's disease. The recent publication of studies of long-term response rates of topical PDT in NMSC, as well as evaluation of its potential use in field cancerization and as a preventive therapy for cutaneous malignancy, has stimulated increased interest in this therapy.

Topical PDT has, to date, been approved by regulatory authorities in over 30 countries worldwide, for use in at least one NMSC indication. Two photosensitizing agents are licensed: a formulation of 5-ALA, Levulan® (DUSA Pharmaceuticals, Wilmington, MA, USA), for AK, and an esterified formulation, methyl aminolevulinate (MAL), Metvix®/Metvixia® (Galderma, Paris, France), for AK, Bowen's disease, and superficial and nodular BCCs. Additional formulations are under development.

Topical PDT offers the potential of a practical, non-surgical, outpatient/office therapy in dermatology ( Table 45.2 ). PDT may prove advantageous where size, site or numbers of lesions limit the efficacy and/or acceptability of conventional therapies. Topical PDT studies consistently report a superiority of cosmetic outcome with minimal or no scarring when compared to other standard therapies including cryotherapy and surgery.

In addition to providing a novel therapy, fluorescence emitted by ALA-induced PpIX can be utilized to provide additional information about cutaneous lesions. This permits the delineation of surface tumor margins or recurrent disease where clinical margins are difficult to define and can aid diagnosis of cancer. Refinement of this technique is still required, but offers an additional advantage of using the PDT process for treating skin cancer. Surface tumor delineation can be visualized even using a simple Woods UV lamp prior to illuminating lesions with the PDT lamp.

History

The term ‘photodynamic therapy’ was first used over 100 years ago by von Tappeiner following experiments using eosin and a combination of natural and artificial light in the treatment of skin cancer in 1903. He realized the requirement for oxygen, describing the phenomenon as an oxygen-dependent photosensitization. The subsequent observation of selective localization of porphyrins to tumors and the demonstration of a photodynamic action involving hematoporphyrin in tumors raised interest in PDT. However, large doses of the crude photosensitizer were required and the consequent phototoxicity limited development. Partial purification produced hematoporphyrin derivative (HpD) and re-ignited interest in this modality, and PDT using HpD in human cancer was reported in 1967. The first trial was performed in 1978 in 25 patients with a variety of malignancies, including SCC and BCC. Kennedy et al. reported in 1990 the use of the topically active agent, 5-ALA, which has been the focus of much research activity during the past two decades.

Overview

The photodynamic reaction concerns the activation of photosensitizers in target tissue following absorption of light of an appropriate wavelength. For skin tumors, the topical route of application of 5-ALA permits an additional method of tumor selectivity, with selective uptake through altered epidermis. 5-ALA is a precursor in the heme biosynthesis pathway of protoporphyrin IX (PpIX), an endogenous photosensitizer not normally present within tissue in therapeutically useful concentrations. Exogenous administration of 5-ALA can increase the intracellular concentration of PpIX ( Fig. 45.1 ), as 5-ALA is the first precursor of heme after the feedback control point and the conversion of PpIX to heme is relatively slow. Local application of 5-ALA is possible due to its increased passage, when in aqueous solution, through an abnormal epidermis, thus restricting the photosensitization primarily to the tumor/dysplastic cells. Proliferating, relatively iron-deficient tumor cells preferentially accumulate PpIX as iron is required for the final conversion of PpIX into heme. This tissue selectivity in 5-ALA photodynamic therapy can be demonstrated by the detection of PpIX-induced fluorescence. There is a relatively greater specificity with the methyl ester of 5-ALA (MAL) in comparison with 5-ALA due to increased lipophilicity, with a ratio of 9:1 compared with 2:1 for BCC compared with normal skin.

Szeimies et al. demonstrated a homogenous distribution of PpIX fluorescence of nodular and superficial (but not morpheic) BCC, including tumor lobules in deep dermis, 12 hours after 10% ALA application. Roberts et al. reported that PpIX distribution in BCC was most intense in those regions of tumor immediately adjacent to the dermis following application of 20% 5-ALA for 4 hours to Bowen's disease and superficial BCC.

As 5-ALA is hydrophilic, to facilitate penetration, most studies report the use of a 20% concentration in an oil-in-water emulsion. Further enhancement of efficacy has been attempted using the penetration enhancer dimethylsulfoxide, and the iron chelators desferrioxamine and ethylenediaminetetra-acetic acid disodium. As MAL is more lipophilic, Peng et al. found that MAL penetrated to a 2 mm depth in BCC, contrasting with more limited penetration with ALA. However, using similar protocols, ALA is reported to result in higher PpIX levels than MAL, but with less selectivity for the diseased compared with healthy tissue and in AK. A randomized double-blind study compared ALA and MAL for the treatment of extensive scalp AK. MAL was applied for 3 hours, but ALA for 5 hours. No significant difference in mean lesion count reduction was observed 1 month after treatment, although pain was more intense on the ALA side. Direct comparison of these agents remains limited, but the current literature, while indicating both to be effective, indicates reduced specificity with high efficacy in superficial lesions with ALA, but greater lesion specificity and tissue penetration with MAL. This supports the impressive clearance rates for MAL-PDT in nodular BCC as well as with superficial lesions.

Mechanism of action

The rationale for PDT is based on the cytotoxic action of products generated by excited photosensitizers. When a photosensitizer absorbs light of the appropriate wavelength, it is converted from a stable ground state to a short-lived singlet state that may undergo conversion to a longer-lived excited triplet state. This is the photo-active species responsible for the generation of cytotoxic products. This may either directly react with substrate by hydrogen atom or electron transfer to form radicals (type I reaction), or the triplet state can transfer its energy to oxygen directly to produce singlet oxygen. Singlet oxygen is highly reactive in biological systems (type II reaction), causing photo- oxidation and cell death ( Fig. 45.2 ). The complete process, with excitation of photosensitizer, transfer of energy through intersystem crossings, to excitation of oxygen from its triplet state and the subsequent quenching of singlet oxygen through cytotoxic mechanisms, takes place in a time scale of microseconds.

The effect of PDT on cells depends on the concentration and localization of the sensitizer and its efficiency in that environment, the light dose reaching the cell, and the oxygen supply. As the diffusion distance of singlet oxygen in cells is estimated to be only 0.1 mm, cell damage is likely to be close to the site of its generation. For lipophilic photosensitizers, including protoporphyrin IX, inhibition of mitochondrial enzymes may be the key event in PDT cell death.

The predominant mechanism of action of PDT is presumed to be direct tumor cell kill. Immunologic effects such as the elimination of small foci of cancer cells that have escaped PDT-induced cytotoxicity may also contribute to the success of PDT, although the importance of this remains undetermined. Tumor-sensitized immune cells and the anti-tumor activity of inflammatory cells probably both contribute to this immune response.

Light of appropriate wavelength for activation of the photosensitizer is required in the target tissue. While 635 nm light may penetrate up to 6 mm (compared with 1–2 mm for light at 400–500 nm), the therapeutically effective maximum depth of PDT will depend on sufficient light dose being delivered to tissue that also has sufficient photosensitizer to achieve a photodynamic reaction. The therapeutically effective depth of PDT in the skin is therefore likely to be less, at 1–3 mm at 635 nm, depending on the tissue. ¹ 5-ALA-induced photosensitivity has a porphyrin-like spectrum with maximum excitation at 410 nm and additional smaller peaks at 510, 545, 580 and 635 nm ( Fig. 45.3 ). Using shorter wavelength light could thus achieve more efficient activation of PpIX, but at the expense of depth of therapeutic effect. In Europe, most clinical applications of PDT have used red light around 630–635 nm to achieve adequate penetration, although in the US, blue light that activates the 410 nm peak is commonly used when treating thin/moderate thickness AKs.

Several light sources have been used in clinical PDT studies for cutaneous applications, including lasers, xenon arc/discharge lamps, incandescent filament lamps, and solid-state light-emitting diodes (LEDs). Coherent light is not required for PDT. The development of energy-efficient LED sources has facilitated the development of large-area, yet portable, red-light sources that have become the most frequently used lights in clinical practice.

For superficial disease such as non-hyperkeratotic AK, short wavelength blue light is a more efficient wavelength to use. This is near the maximum absorption wavelength of PpIX at 410 nm and may reduce the stinging/pain of PDT. In a randomized comparison of green with red light in Bowen's disease, however, green light was significantly inferior, suggesting deeper light penetration with longer wavelength red light is necessary to treat the entire thickness of disease (including skin appendages) in Bowen's disease.

Fractionation (discontinuous illumination) may improve tumor responsiveness by permitting tissue re-oxygenation during ‘dark’ periods. Recent studies support superiority of the fractionation approach over conventional single illumimations in BCC, but not in Bowen's disease, although direct comparison with approved protocols is required. MAL-PDT is approved on the basis of a double treatment 1 week apart. This ‘long-interval’ fractionation is presumed to target residual disease before re-epithelialization and re-growth occur.

Recent studies have suggested that pulsed light therapy may be useful in topical PDT, including a randomized trial comparing a variable pulse light with LED-PDT in AK. Efficacy was equivalent, while pain was significantly less, with the pulsed light source. However, a study performed in healthy human skin in vivo following microdermabrasion and acetone scrub showed that two pulsed light sources previously reported in PDT, the pulse dye laser (PDL) and a broadband flashlamp filtered intense pulsed light (IPL), produced evidence of minimal activation of photosensitizer, with a dramatically smaller photodynamic reaction than seen with a conventional continuous wave broadband source.

Current interest is focusing on developing small, potentially disposable, organic LED light devices that can be worn by patients, hence achieving ‘ambulatory PDT’. Initial studies are promising, with reduced pain over conventional therapy.

Ambient light as the source for ALA-PDT for AK has also been explored with a randomized ambient light-controlled study using 5-ALA demonstrating no significant effect on lesion ablation. However, a randomized right/left intra-patient comparison of conventional MAL-PDT delivered with an LED device versus daylight (for 2.5 hours) for the treatment of AK of face and scalp, showed an equivalent reduction in AK and significantly less pain with daylight, recently confirmed in a further study. Although daylight exposure might achieve a therapeutically effective dose in certain circumstances, it is unlikely to offer a consistent, practical approach to the delivery of PDT.

Choice of light source requires consideration of their relative efficiency for the photosensitizer used and for the indications proposed. If clinical practice is limited to AK, then blue light remains a good option, while red LED devices increase the therapeutic range of office PDT.