75: Photosensitivity and Other Reactions to Light

Sunlight is the most visible and obvious source of comfort in the environment. The sun provides the beneficial effects of warmth and vitamin D synthesis. However, acute and chronic sun exposure also has pathologic consequences. Few effects of sun exposure beyond those affecting the skin have been identified, but cutaneous exposure to sunlight is the major cause of human skin cancer and can have immunosuppressive effects as well.

The sun’s energy reaching the earth’s surface is limited to components of the ultraviolet (UV) spectrum, the visible spectrum, and portions of the infrared spectrum. The cutoff at the short end of the UV spectrum at ~290 nm is due primarily to stratospheric ozone—formed by highly energetic ionizing radiation—that prevents penetration to the earth’s surface of the shorter, more energetic, potentially more harmful wavelengths of solar radiation. Indeed, concern about destruction of the ozone layer by chlorofluorocarbons released into the atmosphere has led to international agreements to reduce production of those chemicals.

Measurements of solar flux showed a twentyfold regional variation in the amount of energy at 300 nm that reaches the earth’s surface. This variability relates to seasonal effects, the path that sunlight traverses through ozone and air, the altitude (a 4% increase for each 300 m of elevation), the latitude (increasing intensity with decreasing latitude), and the amount of cloud cover, fog, and pollution.

The major components of the photobiologic action spectrum that are capable of affecting human skin include the UV and visible wavelengths between 290 and 700 nm. In addition, the wavelengths beyond 700 nm in the infrared spectrum primarily emit heat and in certain circumstances may exacerbate the pathologic effects of energy in the UV and visible spectra.

The UV spectrum reaching the earth represents <10% of total incident solar energy and is arbitrarily divided into two major segments, UV-B and UV-A, which constitute the wavelengths from 290 to 400 nm. UV-B consists of wavelengths between 290 and 320 nm. This portion of the photobiologic action spectrum is the most efficient in producing redness or erythema in human skin and thus is sometimes known as the “sunburn spectrum.” UV-A includes wavelengths between 320 and 400 nm and is ~1000-fold less efficient in producing skin redness than is UV-B.

The wavelengths between 400 and 700 nm are visible to the human eye. The photon energy in the visible spectrum is not capable of damaging human skin in the absence of a photosensitizing chemical. Without the absorption of energy by a molecule, there can be no photosensitivity. Thus, the absorption spectrum of a molecule is defined as the range of wavelengths it absorbs, whereas the action spectrum for an effect of incident radiation is defined as the range of wavelengths that evoke the response.

Photosensitivity occurs when a photon-absorbing chemical (chromophore) present in the skin absorbs incident energy, becomes excited, and transfers the absorbed energy to various structures or to molecular oxygen.

UV RADIATION (UVR) AND SKIN STRUCTURE AND FUNCTION

Human skin consists of two major compartments: the outer epidermis, which is a stratified squamous epithelium, and the underlying dermis, which is rich in matrix proteins such as collagens and elastin. Both compartments are susceptible to damage from sun exposure. The epidermis and the dermis contain several chromophores capable of absorbing incident solar energy, including nucleic acids, proteins, and lipids. The outermost epidermal layer, the stratum corneum, is a major absorber of UV-B, and <10% of incident UV-B wavelengths penetrate through the epidermis to the dermis. Approximately 3% of radiation below 300 nm, 20% of radiation below 360 nm, and 33% of short visible radiation reach the basal cell layer in untanned human skin. In contrast, UV-A readily penetrates to the dermis and is capable of altering structural and matrix proteins that contribute to photoaging of chronically sun-exposed skin, particularly in individuals of light complexion. Thus, longer wavelengths can penetrate more deeply into the skin.

Molecular Targets for UVR-Induced Skin Effects

Epidermal DNA—predominantly in keratinocytes and in Langerhans cells, which are dendritic antigen-presenting cells—absorbs UV-B and undergoes structural changes between adjacent pyrimidine bases (thymine or cytosine), including the formation of cyclobutane dimers and 6,4-photoproducts. These structural changes are potentially mutagenic and are found in most basal cell and squamous cell carcinomas (BCCs and SCCs, respectively). They can be repaired by cellular mechanisms that result in their recognition and excision and the restoration of normal base sequences. The efficient repair of these structural aberrations is crucial, since individuals with defective DNA repair are at high risk for the development of cutaneous cancer. For example, patients with xeroderma pigmentosum, an autosomal recessive disorder, have a variably deficient repair of UV-induced photoproducts. The skin of these patients often shows the dry, leathery appearance of prematurely photoaged skin, and these patients have an increased frequency of skin cancer already in the first two decades of life. Studies in transgenic mice have verified the importance of functional genes that regulate these repair pathways in preventing the development of UV-induced skin cancer. DNA damage in Langerhans cells may also contribute to the known immunosuppressive effects of UV-B (see “Photoimmunology,” below).

In addition to DNA, molecular oxygen is a target for incident solar UVR, leading to the generation of reactive oxygen species (ROS). These ROS can damage skin components, such as epidermal lipids—either free lipids in the stratum corneum or cell membrane lipids. UVR also can target proteins, leading to increased cross-linking and degradation of matrix proteins in the dermis and accumulation of abnormal dermal elastin leading to photoaging changes known as solar elastosis.

Cutaneous Optics and Chromophores

Chromophores are endogenous or exogenous chemical components that can absorb physical energy. Endogenous chromophores are of two types: (1) normal components of skin, including nucleic acids, proteins, lipids, and 7-dehydrocholesterol (the precursor of vitamin D); and (2) components that are synthesized elsewhere in the body and that circulate in the bloodstream and diffuse into the skin, such as porphyrins. Normally, only trace amounts of porphyrins are present in the skin, but, in selected diseases known as the porphyrias (Chap. 430), porphyrins are released into the circulation in increased amounts from the bone marrow and the liver and are transported to the skin, where they absorb incident energy both in the Soret band (around 400 nm; short visible) and, to a lesser extent, in the red portion of the visible spectrum (580–660 nm). This energy absorption results in the generation of ROS that can mediate structural damage to the skin, manifested as erythema, edema, urticaria, or blister formation. It is of interest that photoexcited porphyrins are currently used in the treatment of nonmelanoma skin cancers and their precursor lesions, actinic keratoses. Known as photodynamic therapy, this modality generates ROS in the skin, leading to cell death. Topical photosensitizers used in photodynamic therapy are the porphyrin precursors 5-aminolevulinic acid and methyl aminolevulinate, which are converted to porphyrins in the skin. It is believed that photodynamic therapy targets tumor cells for destruction more selectively than it targets adjacent nonneoplastic cells. The efficacy of such therapy requires appropriate timing of the application of methyl aminolevulinate or 5-aminolevulinic acid to the affected skin followed by exposure to artificial sources of visible light. High-intensity blue light has been used successfully for the treatment of thin actinic keratoses. Red light has a longer wavelength, penetrates more deeply into the skin, and is more beneficial in the treatment of superficial BCCs.

Acute Effects of Sun Exposure

The acute effects of skin exposure to sunlight include sunburn and vitamin D synthesis.

SUNBURN

This painful skin condition is an acute inflammatory response of the skin, predominantly to UV-B. Generally, an individual’s ability to tolerate sunlight is inversely proportional to that individual’s degree of melanin pigmentation. Melanin, a complex polymer of tyrosine derivatives, is synthesized in specialized epidermal dendritic cells known as melanocytes and is packaged into melanosomes that are transferred via dendritic processes into keratinocytes, thereby providing photoprotection and simultaneously darkening the skin. Sun-induced melanogenesis is a consequence of increased tyrosinase activity in melanocytes. Central to the suntan response is the melanocortin-1 receptor (MC1R), and mutations in this gene contribute to the wide variation in human skin and hair color; individuals with red hair and fair skin typically have low MC1R activity. Genetic studies have revealed additional genes that influence skin color variation in humans, such as the gene for tyrosinase (TYR) and the genes APBA2[OCA2], SLC45A2, and SLC24A5. The human MC1R gene encodes a G protein–coupled receptor that binds α-melanocyte-stimulating hormone, which is secreted in the skin mainly by keratinocytes in response to UVR. The UV-induced expression of this hormone is controlled by the tumor suppressor p53, and absence of functional p53 attenuates the tanning response. Activation of the melanocortin receptor leads to increased intracellular cyclic adenosine 5′-monophosphate (cAMP) and protein kinase A activation, resulting in an increased transcription of the microphthalmia-associated transcription factor (MITF), which stimulates melanogenesis. Since the precursor of α-melanocyte-stimulating hormone, proopiomelanocortin, is also the precursor of β-endorphins, UVR may result in not only increased pigmentation but also in increased β-endorphin production, an effect that has been hypothesized to promote sun-seeking behaviors.

The Fitzpatrick classification of human skin phototypes is based on the efficiency of the epidermal-melanin unit, which usually can be ascertained by asking an individual two questions: (1) Do you burn after sun exposure? (2) Do you tan after sun exposure? The answers to these questions permit division of the population into six skin types, varying from type I (always burn, never tan) to type VI (never burn, always tan) (Table 75-1).

Sunburn erythema is due to vasodilation of dermal blood vessels. There is a lag time (usually 4–12 h) between skin exposure to sunlight and the development of visible redness. The action spectrum for sunburn erythema includes UV-B and UV-A, although UV-B is much more efficient than UV-A in evoking the response. However, UV-A may contribute to sunburn erythema at midday, when much more UV-A than UV-B is present in the solar spectrum. The erythema that accompanies the inflammatory response induced by UVR results from the orchestrated release of cytokines along with growth factors and the generation of ROS. Furthermore, UV-induced activation of nuclear factor κB–dependent gene transcription can augment release of several proinflammatory cytokines and vasoactive mediators. These cytokines and mediators accumulate locally in sunburned skin, providing chemotactic factors that attract neutrophils, macrophages, and T lymphocytes, which promote the inflammatory response. UVR also stimulates infiltration of inflammatory cells through induced expression of adhesion molecules such as E-selectin and intercellular adhesion molecule 1 on endothelial cells and keratinocytes. UVR also has been shown to activate phospholipase A₂, resulting in increases in eicosanoids such as prostaglandin E₂, which is known to be a potent inducer of sunburn erythema. The role of eicosanoids in this reaction has been verified by studies showing that nonsteroidal anti-inflammatory drugs (NSAIDs) can reduce it.

Epidermal changes in sunburn include the induction of “sunburn cells,” which are keratinocytes undergoing p53-dependent apoptosis as a defense, with elimination of cells that harbor UV-B-induced structural DNA damage.

VITAMIN D SYNTHESIS AND PHOTOCHEMISTRY

Cutaneous exposure to UV-B causes photolysis of epidermal 7-dehydrocholesterol, converting it to pre–vitamin D₃, which then undergoes temperature-dependent isomerization to form the stable hormone vitamin D₃. This compound diffuses to the dermal vasculature and circulates to the liver and kidney, where it is converted to the dihydroxylated functional hormone 1,25-dihydroxyvitamin D₃. Vitamin D metabolites from the circulation and those produced in the skin itself can augment epidermal differentiation signaling and inhibit keratinocyte proliferation. These effects are exploited therapeutically in psoriasis with the topical application of synthetic vitamin D analogues. In addition, vitamin D is increasingly thought to have beneficial effects in several other inflammatory conditions, and some evidence suggests that—besides its classic physiologic effects on calcium metabolism and bone homeostasis—it is associated with a reduced risk of various internal malignancies. There is controversy regarding the risk-to-benefit ratio of sun exposure in vitamin D homeostasis. At present, it is important to emphasize that no clear-cut evidence suggests that the use of sunscreens substantially diminishes vitamin D levels. Since aging also substantially decreases the ability of human skin to photocatalytically produce vitamin D₃, the widespread use of sunscreens that filter out UV-B has led to concerns that the elderly might be unduly susceptible to vitamin D deficiency. However, the amount of sunlight needed to produce sufficient vitamin D is small and does not justify the risks of skin cancer and other types of photodamage linked to increased sun exposure or tanning behavior. Nutritional supplementation of vitamin D is a preferable strategy for patients with vitamin D deficiency.

Chronic Effects of Sun Exposure: Nonmalignant

The clinical features of photoaging (dermatoheliosis) consist of wrinkling, blotchiness, and telangiectasia as well as a roughened, irregular, “weather-beaten” leathery appearance.

UVR is important in the pathogenesis of photoaging in human skin, and ROS are likely involved. The dermis and its connective tissue matrix are major targets for sun-associated chronic damage that manifests as solar elastosis, a massive increase in thickened irregular masses of abnormal-appearing elastic fibers. Collagen fibers are also abnormally clumped in the deeper dermis of sun-damaged skin. The chromophore(s), the action spectra, and the specific biochemical events orchestrating these changes are only partially understood, although more deeply penetrating UV-A seems to be primarily involved. Chronologically aged sun-protected skin and photoaged skin share important molecular features, including connective tissue damage and elevated levels of matrix metalloproteinases (MMPs). MMPs are enzymes involved in the degradation of the extracellular matrix. UV-A induces expression of some MMPs, including MMP-1 and MMP-3, leading to increased collagen breakdown. In addition, UV-A reduces type I procollagen mRNA expression. Thus, chronic UVR alters the structure and function of dermal collagen. On the basis of these observations, it is not surprising that high-dose UV-A phototherapy may have beneficial effects in some patients with localized fibrotic diseases of the skin, such as localized scleroderma.

Chronic Effects of Sun Exposure: Malignant

One of the major known consequences of chronic excessive skin exposure to sunlight is nonmelanoma skin cancer. The two most common types of nonmelanoma skin cancer are BCC and SCC (Chap. 105). A model for skin cancer induction involves three major steps: initiation, promotion, and progression. Exposure of human skin to sunlight results in initiation, a step by which structural (mutagenic) changes in DNA evoke an irreversible alteration in the target cell (keratinocyte) that begins the tumorigenic process. Exposure to a tumor initiator such as UV-B is believed to be a necessary but not a sufficient step in the malignant process, since initiated skin cells not exposed to tumor promoters generally do not develop tumors. The second stage in tumor development is promotion, a multistep process by which chronic exposure to sunlight evokes further changes that culminate in the clonal expansion of initiated cells and cause the development, over many years, of premalignant growths known as actinic keratoses, a minority of which may progress to form SCCs. As a result of extensive studies, it seems clear that UV-B is a complete carcinogen, meaning that it can act as both a tumor initiator and a tumor promoter. The third and final step in the malignant process is malignant conversion of benign precursors into malignant lesions, a process thought to require ­additional genetic alterations.

On a molecular level, skin carcinogenesis results from the accumulation of gene mutations that cause inactivation of tumor suppressors, activation of oncogenes, or reactivation of cellular signaling pathways that normally are expressed only during epidermal embryologic development. Accumulation of mutations in the tumor-suppressor gene p53 secondary to UV-induced DNA damage occurs in both SCCs and BCCs and is important in promoting skin carcinogenesis. Indeed, the majority of both human and murine UV-induced skin cancers have characteristic p53 mutations (C → T and CC → TT transitions). Studies in mice have shown that sunscreens can substantially reduce the frequency of these signature mutations in p53 and inhibit the induction of tumors.

BCCs also harbor inactivating mutations in the tumor-suppressor gene patched, which result in activation of the sonic hedgehog signaling pathway and increased cell proliferation. Thus, these tumors can manifest mutations in tumor suppressors (p53 and patched) or oncogenes (smoothened). New evidence links alterations in the Wnt/β-catenin signaling pathway, which is known to be critical for hair follicle development, to skin cancer as well. Thus interactions between this pathway and the hedgehog signaling pathway appear to be involved in both skin carcinogenesis and embryologic development of the skin and hair follicles.

Clonal analysis in mouse models of BCC revealed that tumor cells arise from long-term resident progenitor cells of the interfollicular epidermis and the upper infundibulum of the hair follicle. These BCC-initiating cells are reprogrammed to resemble embryonic hair follicle progenitors, whose tumor-initiating ability depends on activation of the Wnt/β-catenin signaling pathway.

SCC initiation occurs both in the interfollicular epidermis and in the hair follicle bulge stem cell populations. In mouse models, the combination of mutant K-Ras and p53 is sufficient to induce invasive SCCs from these cell populations.

The transcription factor Myc is important for stem cell maintenance in the skin, and oncogenic activation of Myc has been implicated in the development of BCCs and SCCs. Thus, nonmelanoma skin cancer involves mutations and alterations in multiple genes and pathways that occur as a result of their chronic accumulation driven by exposure to environmental factors such as UVR.

Epidemiologic studies have linked excessive sun exposure to an increased risk of nonmelanoma cancers and melanoma of the skin; the evidence is far more direct for nonmelanoma skin cancers (BCCs and SCCs) than for melanoma. Approximately 80% of nonmelanoma skin cancers develop on sun-exposed body areas, including the face, neck, and hands. Major risk factors include male sex, childhood sun exposures, older age, fair skin, and residence at latitudes relatively close to the equator. Individuals with darker-pigmented skin have a lower risk of skin cancer than do fair-skinned individuals. More than 2 million individuals in the United States develop nonmelanoma skin cancer annually, and the lifetime risk that a fair-skinned individual will develop such a neoplasm is estimated at ~15%. The incidence of nonmelanoma skin cancer in the population is increasing at a rate of 2–3% per year. One potential explanation is the widespread use of indoor tanning. It is estimated that 30 million people tan indoors in the United States annually, including >2 million adolescents.

The relationship of sun exposure to melanoma development is less direct, but strong evidence supports an association. Clear-cut risk factors include a positive family or personal history of melanoma and multiple dysplastic nevi. Melanomas can occur during adolescence; the implication is that the latent period for tumor growth is shorter than that for nonmelanoma skin cancer. For reasons that are only partially understood, melanomas are among the most rapidly increasing human malignancies (Chap. 105). Epidemiologic studies indicate that indoor tanning is a risk factor for melanoma, which may contribute to the increasing incidence of melanoma formation. Furthermore, epidemiologic studies suggest that life in a sunny climate from birth or early childhood may increase the risk of melanoma development. In general, risk does not correlate with cumulative sun exposure but may be related to the duration and extent of exposure in childhood.

However, in contrast to nonmelanoma skin cancers, melanoma frequently develops in sun-protected skin, and oncogenic mutations in melanoma may also not be UVR-signature mutations; these observations suggest that UVR-independent factors contribute to melanomagenesis. Low MC1R activity leads to production of the red/yellow pheomelanin pigment in individuals with red hair and fair skin, while high MC1R activity results in increased production of the black/brown eumelanin. Experiments in mice suggest that high pheomelanin content in skin (as in individuals with red hair and fair skin) leads to a UVR-independent increase in the risk of melanoma through a mechanism that involves oxidative damage. Thus, both UVR-dependent and UVR-independent factors are likely to contribute to melanoma formation.

Photoimmunology

Exposure to solar radiation causes both local immunosuppression (through inhibition of immune responses to antigens applied at the irradiated site) and systemic immunosuppression (through inhibition of immune responses to antigens applied at remote, unirradiated sites). For example, human skin exposure to modest doses of UV-B can deplete the epidermal antigen-presenting cells known as Langerhans cells, thereby reducing the degree of allergic sensitization to application of the potent contact allergen dinitrochlorobenzene at the irradiated skin site.

An example of the systemic immunosuppressive effects of higher doses of UVR is the diminished immunologic response to antigens introduced either epicutaneously or intracutaneously at sites distant from the irradiated site. Various immunomodulatory factors and immune cells have been implicated in UVR-induced systemic immunosuppression, including tumor necrosis factor α, interleukin 4, interleukin 10, cis-urocanic acid, and eicosanoids. Experimental evidence suggests that prostaglandin E₂ signaling through prostaglandin E receptor subtype 4 mediates UVR-induced systemic immunosuppression by elevating the number of regulatory T cells, and this effect can be inhibited with NSAIDs.

The major chromophores in the upper epidermis that are known to initiate UV-mediated immunosuppression include DNA, trans-urocanic acid, and membrane components. The action spectrum for UV-induced immunosuppression closely mimics the absorption spectrum of DNA. Pyrimidine dimers in Langerhans cells may inhibit antigen presentation. The absorption spectrum of epidermal urocanic acid closely mimics the action spectrum for UV-B-induced immuno­suppression. Urocanic acid is a metabolic product of the essential amino acid histidine and accumulates in the upper epidermis through breakdown of the histidine-rich protein filaggrin due to the absence of its catabolizing enzyme in keratinocytes. Urocanic acid is synthesized as a trans-isomer, and UV-induced trans-cis isomerization of urocanic acid in the stratum corneum drives immunosuppression. Cis-urocanic acid may exert its immunosuppressive effects through a variety of mechanisms, including inhibition of antigen presentation by Langerhans cells.

One important consequence of chronic sun exposure and associated immunosuppression is an enhanced risk of skin cancer. In part, UV-B activates regulatory T cells that suppress antitumor immune responses via interleukin 10 expression, whereas, in the absence of high UV-B exposure, epidermal Langerhans cells present tumor-associated antigens and induce protective immunity, thereby inhibiting skin tumorigenesis. UV-induced DNA damage is a major molecular trigger of this immunosuppressive effect.

Perhaps the most graphic demonstration of the role of immunosuppression in enhancing the risk of nonmelanoma skin cancer comes from studies of organ transplant recipients who require lifelong immunosuppressive/antirejection drug regimens. More than 50% of organ transplant recipients develop BCCs and SCCs, and these cancers are the most common types of malignancies arising in these patients. Rates of BCC and SCC increase with the duration and degree of immunosuppression. These patients ideally should be screened prior to organ transplantation, be monitored closely thereafter, and adhere to rigorous photoprotection measures, including the use of sunscreens and protective clothing as well as sun avoidance. Notably, immunosuppressive drugs that target the mTOR pathway, such as sirolimus and everolimus, may reduce the risk of nonmelanoma skin cancer in organ transplant recipients from that associated with the use of calcineurin inhibitors (cyclosporine and tacrolimus), which may contribute to nonmelanoma skin cancer formation not only through their immunosuppressive effects but also through suppression of p53-dependent cancer cell senescence pathways independent of host immunity.

PHOTOSENSITIVITY DISEASES

The diagnosis of photosensitivity requires elicitation of a careful history in order to define the duration of signs and symptoms, the length of time between exposure to sunlight and the development of subjective symptoms, and visible changes in the skin. The age of onset can also be a helpful diagnostic clue; for example, the acute photosensitivity of erythropoietic protoporphyria almost always begins in childhood, whereas the chronic photosensitivity of porphyria cutanea tarda (PCT) typically begins in the fourth and fifth decades of life. A patient’s history of exposure to topical and systemic drugs and chemicals may provide important diagnostic clues. Many classes of drugs can cause photosensitivity on the basis of either phototoxicity or photoallergy. Fragrances such as musk ambrette that were previously present in numerous cosmetic products are also potent photosensitizers.

Examination of the skin may offer important clues. Anatomic areas that are naturally protected from direct sunlight, such as the hairy scalp, the upper eyelids, the retroauricular areas, and the infranasal and submental regions, may be spared, whereas exposed areas show characteristic features of the pathologic process. These anatomic localization patterns are often helpful, but not infallible, in making the diagnosis. For example, airborne contact sensitizers that are blown onto the skin may produce dermatitis that can be difficult to distinguish from photosensitivity despite the fact that such material may trigger skin reactivity in areas shielded from direct sunlight.

Many dermatologic conditions may be caused or aggravated by sunlight (Table 75-2). The role of light in evoking these responses may be dependent on genetic abnormalities ranging from well-described defects in DNA repair that occur in xeroderma pigmentosum to the inherited abnormalities in heme synthesis that characterize the porphyrias. The chromophore has been identified in certain photosensitivity diseases, but the energy-absorbing agent remains unknown in the majority.

Polymorphous Light Eruption

A common type of photosensitivity disease is polymorphous light eruption (PMLE). Many affected individuals never seek medical attention because the condition is often transient, becoming manifest in the spring with initial sun exposure but then subsiding spontaneously with continuing exposure, a phenomenon known as “hardening.” The major manifestations of PMLE include (often intensely) pruritic erythematous papules that may coalesce into plaques in a patchy distribution on exposed areas of the trunk and forearms. The face is usually less seriously involved. Whereas the morphologic skin findings remain similar for each patient with subsequent recurrences, significant interindividual variations in skin findings are characteristic (hence the term “polymorphous”).

A skin biopsy and phototest procedures in which skin is exposed to multiple erythemal doses of UV-A and UV-B may aid in the diagnosis. The action spectrum for PMLE is usually within these portions of the solar spectrum.

Whereas the treatment of an acute flare of PMLE may require topical or systemic glucocorticoids, approaches to preventing PMLE are important and include the use of high-SPF and high UVA-protection broad-spectrum sunscreens as well as the induction of “hardening” by the cautious administration of artificial UV-B (broad-band or narrow-band) and/or UV-A radiation or the use of psoralen plus UV-A (PUVA) photochemotherapy for 2–4 weeks before initial sun exposure. Such prophylactic phototherapy or photochemotherapy at the beginning of spring may prevent the occurrence of PMLE throughout the summer.

Phototoxicity and Photoallergy

These photosensitivity disorders are related to the topical or systemic administration of drugs and other chemicals. Both reactions require the absorption of energy by a drug or chemical with consequent production of an excited-state photosensitizer that can transfer its absorbed energy to a bystander molecule or to molecular oxygen, thereby generating tissue-destructive chemical species, including ROS.

Phototoxicity is a nonimmunologic reaction that can be caused by drugs and chemicals, a few of which are listed in Table 75-3. The usual clinical manifestations include erythema resembling a sunburn reaction that quickly desquamates, or “peels,” within several days. In addition, edema, vesicles, and bullae may occur.

Photoallergy is much less common and is distinct in that it is an immunopathologic process. The excited-state photosensitizer may create highly unstable haptenic free radicals that bind covalently to macromolecules to form a functional antigen capable of evoking a delayed-type hypersensitivity response. Some drugs and chemicals that can produce photoallergy are listed in Table 75-4. The clinical manifestations typically differ from those of phototoxicity in that an intensely pruritic eczematous dermatitis tends to predominate and evolves into lichenified, thickened, “leathery” changes in sun-exposed areas. A small subset (perhaps 5–10%) of patients with photoallergy may develop a persistent exquisite hypersensitivity to light even when the offending drug or chemical is identified and eliminated, a condition known as persistent light reaction.

A very uncommon type of persistent photosensitivity is known as chronic actinic dermatitis. The affected patients are typically elderly men with a long history of preexisting allergic contact dermatitis or photosensitivity. These individuals are usually exquisitely sensitive to UV-B, UV-A, and visible wavelengths.

Phototoxicity and photoallergy often can be diagnostically confirmed by phototest procedures. In patients with suspected phototoxicity, determining the minimal erythemal dose (MED) while the patient is exposed to a suspected agent and then repeating the MED after discontinuation of the agent may provide a clue to the causative drug or chemical. Photopatch testing can be performed to confirm the diagnosis of photoallergy. In this simple variant of ordinary patch testing, a series of known photoallergens is applied to the skin in duplicate, and one set is irradiated with a suberythemal dose of UV-A. The development of eczematous changes at sites exposed to sensitizer and light is a positive result. The characteristic abnormality in patients with persistent light reaction is a diminished threshold to erythema evoked by UV-B. Patients with chronic actinic dermatitis usually manifest a broad spectrum of UV hyperresponsiveness and require meticulous photoprotection, including avoidance of sun exposure, use of high-SPF (>30) sunscreens, and, in severe cases, systemic immunosuppression, such as with azathioprine.

The management of drug photosensitivity involves first and foremost the elimination of exposure to the chemical agents responsible for the reaction and the minimization of sun exposure. The acute symptoms of phototoxicity may be ameliorated by cool moist compresses, topical glucocorticoids, and systemically administered NSAIDs. In severely affected individuals, a rapidly tapered course of systemic glucocorticoids may be useful. Judicious use of analgesics may be necessary.

Photoallergic reactions require a similar management approach. Furthermore, patients with persistent light reaction and chronic actinic dermatitis must be meticulously protected against light exposure. In selected patients to whom chronic systemic high-dose glucocorticoids pose unacceptable risks, it may be necessary to employ an immunosuppressive drug such as azathioprine, cyclophosphamide, cyclosporine, or mycophenolate mofetil.

Porphyria

The porphyrias (Chap. 430) are a group of diseases that have in common inherited or acquired derangements in the synthesis of heme. Heme is an iron-chelated tetrapyrrole or porphyrin, and the nonmetal chelated porphyrins are potent photosensitizers that absorb light intensely in both the short (400–410 nm) and the long (580–650 nm) portions of the visible spectrum.

Heme cannot be reutilized and must be synthesized continuously. The two body compartments with the largest capacity for its production are the bone marrow and the liver. Accordingly, the porphyrias originate in one or the other of these organs, with an end result of excessive endogenous production of potent photosensitizing porphyrins. The porphyrins circulate in the bloodstream and diffuse into the skin, where they absorb solar energy, become photoexcited, generate ROS, and evoke cutaneous photosensitivity. The mechanism of porphyrin photosensitization is known to be photodynamic, or oxygen-dependent, and is mediated by ROS such as singlet oxygen and superoxide anions.

Porphyria cutanea tarda is the most common type of porphyria and is associated with decreased activity of the enzyme uroporphyrinogen decarboxylase. There are two basic types of PCT: (1) the sporadic or acquired type, generally seen in individuals ingesting ethanol or receiving estrogens; and (2) the inherited type, in which there is autosomal dominant transmission of deficient enzyme activity. Both forms are associated with increased hepatic iron stores.

In both types of PCT, the predominant feature is chronic photosensitivity characterized by increased fragility of sun-exposed skin, particularly areas subject to repeated trauma such as the dorsa of the hands, the forearms, the face, and the ears. The predominant skin lesions are vesicles and bullae that rupture, producing moist erosions (often with a hemorrhagic base) that heal slowly, with crusting and purplish discoloration of the affected skin. Hypertrichosis, mottled pigmentary change, and scleroderma-like induration are associated features. The diagnosis can be confirmed biochemically by measurement of urinary porphyrin excretion, plasma porphyrin assay, and assay of erythrocyte and/or hepatic uroporphyrinogen decarboxylase. Multiple mutations of the uroporphyrinogen decarboxylase gene have been identified in human populations. Some patients with PCT have associated mutations in the HFE gene, which is linked to hemochromatosis; these mutations could contribute to the iron overload seen in PCT, although iron status as measured by serum ferritin, iron levels, and transferrin saturation is no different from that in PCT patients without HFE mutations. Prior hepatitis C virus infection appears to be an independent risk factor for PCT.

Treatment of PCT consists of repeated phlebotomies to diminish the excessive hepatic iron stores and/or intermittent low doses of chloroquine and hydroxychloroquine. Long-term remission of the disease can be achieved if the patient eliminates exposure to porphyrinogenic agents and prolonged exposure to sunlight.

Erythropoietic protoporphyria originates in the bone marrow and is due to a decrease in the mitochondrial enzyme ferrochelatase secondary to numerous gene mutations. The major clinical features include acute photosensitivity characterized by subjective burning and stinging of exposed skin that often develops during or just after sun exposure. There may be associated skin swelling and, after repeated episodes, a waxlike scarring.

The diagnosis is confirmed by demonstration of elevated levels of free erythrocyte protoporphyrin. Detection of increased plasma protoporphyrin helps distinguish erythropoietic protoporphyria from lead poisoning and iron-deficiency anemia, in both of which erythrocyte protoporphyrin levels are elevated in the absence of cutaneous photosensitivity and elevated plasma protoporphyrin levels.

Treatment includes reduction of sun exposure and oral administration of the carotenoid β-carotene, which is an effective scavenger of free radicals. This drug increases tolerance to sun exposure in some affected individuals, although it has no effect on deficient ferrochelatase.

An algorithm for managing patients with photosensitivity is presented in Fig. 75-1.

Since photosensitivity of the skin results from exposure to sunlight, it follows that absolute avoidance of sunlight will eliminate these disorders. However, contemporary lifestyles make this approach impractical for most individuals. Thus better approaches to photoprotection have been sought.

Natural photoprotection is provided by structural proteins in the epidermis, particularly keratins and melanin. The amount of melanin and its distribution in cells are genetically regulated, and individuals of darker complexion (skin types IV–VI) are at decreased risk for the development of acute sunburn and cutaneous malignancy.

Other forms of photoprotection include clothing and sunscreens. Clothing constructed of tightly woven sun-protective fabrics, irrespective of color, affords substantial protection. Wide-brimmed hats, long sleeves, and trousers all reduce direct exposure. Sunscreens are now considered over-the-counter drugs, and a monograph from the U.S. Food and Drug Administration (FDA) has recognized category I ingredients as safe and effective. Those ingredients are listed in Table 75-5. Sunscreens are rated for their photoprotective effect by their sun protection factor (SPF). The SPF is simply a ratio of the time required to produce sunburn erythema with and without sunscreen application. The SPF of most sunscreens reflects protection from UV-B but not from UV-A. The FDA monograph stipulates that sunscreens must be rated on a scale ranging from minimal (SPF ≥2 and <12) to moderate (SPF ≥12 and <30) to high (SPF ≥30, labeled as 30+).

Broad-spectrum sunscreens contain both UV-B-absorbing and UV-A-absorbing chemicals, the latter including avobenzone and ecamsule (terephthalylidene dicamphor sulfonic acid). These chemicals absorb UVR and transfer the absorbed energy to surrounding cells. In contrast, physical UV blockers (zinc oxide and titanium dioxide) scatter or reflect UVR.

In addition to light absorption, a critical determinant of the sustained photoprotective effect of sunscreens is their water resistance. The FDA monograph has defined strict testing criteria for sunscreens that claim to possess a high degree of water resistance.

Some degree of photoprotection can be achieved by limiting the time of sun exposure during the day. Since a large part of an individual’s total lifetime sun exposure may occur by age 18, it is important to educate parents and young children about the hazards of sunlight. Simply eliminating exposure at midday will substantially reduce lifetime UVR exposure.

UVR can be used therapeutically. The administration of UV-B alone or in combination with topically applied agents can induce remissions of many dermatologic diseases, including psoriasis and atopic dermatitis. In particular, narrow-band UV-B treatments (with fluorescent bulbs emitting radiation at ~311 nm) have enhanced efficacy over that obtained with broad-band UV-B in the treatment of psoriasis.

Photochemotherapy in which topically applied or systemically administered psoralens are combined with UV-A (PUVA) is effective in treating psoriasis and the early stages of cutaneous T cell lymphoma and vitiligo. Psoralens are tricyclic furocoumarins that, when intercalated into DNA and exposed to UV-A, form adducts with pyrimidine bases and eventually form DNA cross-links. These structural changes are thought to decrease DNA synthesis and to be related to the amelioration of psoriasis. Why PUVA photochemotherapy is effective in cutaneous T cell lymphoma is only partially understood, but it has been shown to induce apoptosis of atypical T lymphocyte populations in the skin. Consequently, direct treatment of circulating atypical lymphocytes by extracorporeal photochemotherapy (photopheresis) has been used in Sézary syndrome as well as in other severe systemic diseases with circulating atypical lymphocytes, such as graft-versus-host disease.

In addition to its effects on DNA, PUVA photochemotherapy stimulates epidermal thickening and melanin synthesis; the latter property, together with its anti-inflammatory effects, provides the rationale for use of PUVA in the depigmenting disease vitiligo. Oral 8-methoxypsoralen and UV-A appear to be most effective in this regard, but as many as 100 treatments extending over 12–18 months may be required for satisfactory repigmentation.

Not surprisingly, the major side effects of long-term UV-B photo­therapy and PUVA photochemotherapy mimic those seen in individuals with chronic sun exposure and include skin dryness, actinic keratoses, and an increased risk of skin cancer. Despite these risks, the therapeutic index of these modalities continues to be excellent. It is important to choose the most appropriate phototherapeutic approach for a specific dermatologic disease. For example, narrow-band UV-B has been reported in several studies to be as effective as PUVA photochemotherapy in the treatment of psoriasis but to pose a lower risk of skin cancer development than PUVA.