Drug-induced photosensitivity refers to the development of cutaneous disease as a result of the combined effects of a chemical and light.[1] Exposure to either the chemical or the light alone is not sufficient to induce the disease; exposure to both is necessary. Treatment therefore involves not only avoidance of the causative agent but also photoprotection, as well as measures for symptomatic relief.[2]
When photoactivation of the chemical occurs, one or more cutaneous manifestations may arise. These include phototoxic and photoallergic reactions,[3] a planus lichenoides reaction, pseudoporphyria, and subacute cutaneous lupus erythematosus (SCLE). Photosensitivity reactions may result from systemic medications and topically applied compounds (see Etiology).[4, 5, 6] Drug-induced photosensitivity is also common with vemurafenib and other targeted cancer therapy[7, 8, 9, 10] and has been reported with immune checkpoint inhibitors and cytotoxic chemotherapeutic agents.[11]
Wavelengths within the UV-A (320-400 nm) range and, for certain compounds, those within the visible range are more likely to cause drug-induced photosensitivity reactions, though occasionally UV-B (290-320 nm) can also be responsible for such effects. UV-B wavelengths are most efficient at causing sunburn and nonmelanoma skin cancer. In patients who present with photosensitivity, it is often difficult to differentiate phototoxic from photoallergic reactions. However, they have a number of distinguishing characteristics (see Pathophysiology).
Phototoxic reactions occur because of the damaging effects of light-activated compounds on cell membranes and, in some instances, DNA. By contrast, photoallergic reactions are cell-mediated immune responses to a light-activated compound. Phototoxic reactions develop in most individuals if they are exposed to sufficient amounts of light and drug. Typically, they appear as an exaggerated sunburn response (see the image below).
View Image
Phototoxic reaction.
Photoallergic reactions resemble allergic contact dermatitis, with a distribution typically limited to sun-exposed areas of the body. When the reactions are severe or prolonged, however, they may extend into covered areas of skin.
Photoallergic reactions develop in only a minority of individuals exposed to the compound and light; they are less prevalent than phototoxic skin reactions. The amount of drug required to elicit photoallergic reactions is considerably smaller than that required for phototoxic reactions. Moreover, photoallergic reactions are a form of cell-mediated immunity; their onset often is delayed by as long as 24-72 hours after exposure to the drug and light. By contrast, phototoxic responses often occur within minutes or hours of light exposure.
Phototoxic reactions result from direct damage to tissue caused by a photoactivated compound. Many compounds have the potential to cause phototoxicity. Most have at least one resonating double bond or an aromatic ring that can absorb radiant energy. Most compounds are activated by wavelengths within the UV-A (320-400 nm) range, though some have a peak absorption within the UV-B or visible range.
In most instances, photoactivation of a compound results in the excitation of electrons from the stable singlet state to an excited triplet state. As excited-state electrons return to a more stable configuration, they transfer their energy to oxygen, leading to the formation of reactive oxygen intermediates. Reactive oxygen intermediates (eg, oxygen singlet, superoxide anion, and hydrogen peroxide) damage cell membranes and DNA. Signal transduction pathways that lead to the production of proinflammatory cytokines and arachidonic acid metabolites are also activated. The result is an inflammatory response that has the clinical appearance of an exaggerated sunburn reaction.
The exception to this mechanism of drug-induced phototoxicity is psoralen-induced phototoxicity. Psoralens intercalate within DNA, forming monofunctional adducts. Exposure to UV-A radiation produces bifunctional adducts within DNA. Exactly how bifunctional adducts cause photosensitivity is unknown.
Photoallergic reactions
Photoallergic reactions are cell-mediated immune responses in which the antigen is a light-activated drug. Photoactivation results in the development of a metabolite that can bind to protein carriers in the skin to form a complete antigen. The reaction then proceeds exactly as other cell-mediated immune responses do.
Specifically, Langerhans cells and other antigen-presenting cells take up the antigen and then migrate to regional lymph nodes. In those locations, the Langerhans cells present the photoallergen to T cells that express antigen-specific receptors. The T cells become activated and proliferate, and they return to the site of photoallergen deposition. In the skin, the T cells orchestrate an inflammatory response that usually has either an eczematous morphology, if the photoallergen is applied topically, or the characteristics of a drug eruption, if the photoallergen is administered systemically.
Although it is frequently difficult to determine whether a patient is experiencing a phototoxic reaction or a photoallergic one, the two reactions have certain characteristics that distinguish one from the other (see Table 1 below).
Table 1. Distinguishing Characteristics of Phototoxic and Photoallergic Reactions
Most phototoxic reactions result from the systemic administration of drugs. Photoallergic reactions can be caused by either topical or systemic administration of the chemical. Compounds that commonly cause phototoxic and/or photoallergic reactions are listed in Table 2 below.
Table 2. Common Photosensitizing Medications
View Table
See Table
Some initial evidence suggests that artificial intelligence (AI) techniques may prove applicable to the prediction of the photosensitizing effects of drugs.[33]
The incidence of drug-induced photosensitivity in the United States is uncertain. Phototoxic reactions are known to be considerably more common than photoallergic reactions. The incidence of drug-induced photosensitivity outside the United States is unknown.
Age-, sex-, and race-related demographics
Drug-induced photosensitivity reactions can occur in persons of any age. Men are more likely to have photoallergic reactions than women. The racial incidence of drug-induced photosensitivity reactions is unknown. Photosensitivity reactions can occur in individuals with heavily pigmented skin.
In most patients, the prognosis is excellent once the offending agent is removed. With some compounds, however, complete resolution of the photosensitivity may take several weeks to months. Occasionally, patients have persistent light reactivity for which the prospects for resolution are poor.
Drug-induced photosensitivity is associated with death only in rare individuals who are exposed to large amounts of sunlight after taking large doses of psoralens. Although mortality is rare, drug-induced photosensitivity can cause significant morbidity in some individuals, who must severely limit their exposure to natural or artificial light.
Voriconazole photosensitivity is associated with a risk of skin cancer.[34, 35] The changes that occur with long-term exposure resemble accelerated photoaging. Acute photosensitivity occurs in 1-2% or more of patients taking voriconazole for more than 12 weeks. It appears to be UV-A induced, but it is not strictly dose-dependent. Cheilitis and facial erythema are typical initial manifestations.
Patients need to be counseled regarding the possible photosensitizing properties of both prescription and nonprescription medications. In most cases, drug-induced photosensitivity reactions can be prevented by taking appropriate measures for sun protection.
Patients with drug-induced photosensitivity often, but not always, note intolerance to sunlight. Whereas most individuals can tolerate minutes or hours of sun exposure, those with drug-induced photosensitivity will exhibit skin lesions of one type or another. In most cases, a sunburn response or dermatitis occurs. Drug-induced photosensitivity may result in phototoxicity, photoallergy, lichenoid reactions, subacute cutaneous lupus erythematosus (SCLE; see the image below), or pseudoporphyria.
View Image
Subacute cutaneous lupus erythematosus exacerbated by terbinafine. Courtesy of Jeffrey P. Callen.
Pseudoporphyria (see the image below) may occur with some medications, the most common of which is naproxen. It is characterized by a bullous reaction that clinically and histologically resembles porphyria cutanea tarda. However, the hypertrichosis and sclerodermoid changes typically seen in porphyria cutanea tarda are not seen in pseudoporphyria. The results of porphyrin studies are normal.
View Image
Pseudoporphyria.
Lichenoid reactions that occur in a photodistribution are often difficult to distinguish from idiopathic lichen planus.[36] These reactions are characterized by violaceous or erythematous papules and plaques that sometimes have Wickham striae. Hydrochlorothiazide, hydroxychloroquine, and captopril are known causes of drug-induced lichenoid reactions.
Drug-induced photosensitivity reactions also may include lupuslike reactions. Drug-induced reactions usually resemble SCLE because of their scaling, annular, and psoriasiform characteristics. Hydrochlorothiazide is the drug most frequently associated with this reaction,[37] but calcium-channel blockers, angiotensin-converting enzyme (ACE) inhibitors, griseofulvin, and terbinafine[38] have also been implicated. The rate of reaction is low for any of these agents. Hydrochlorothiazide is commonly used in many combined antihypertensive agents. Patients with drug-induced reactions commonly have anti-Ro (SS-A) antibodies.
As photodynamic therapy (PDT) becomes a more popular treatment modality for actinic keratoses and nonmelanoma skin cancer, recognition of PDT photosensitizer–induced phototoxicity is important. 5-Aminolevulinic acid (5-ALA) or methyl 5-ALA is applied topically, followed by the use of a blue-light (410-420 nm) or red-light (570-670 nm) PDT illuminator. 5-ALA is a prodrug that enters the heme biosynthetic pathway and is metabolized intracellularly to form the photosensitizing molecule protoporphyrin IX (PpIX). Light activates PpIX to generate free radicals and cytotoxic reactive oxygen species that may cause destruction of malignant and nonmalignant hyperproliferative tissue.
Common adverse effects of PDT include mild-to-moderate local phototoxic reactions that usually resolve in several days.
Discriminating between photosensitivity diseases and heat-related exacerbation of skin diseases may be difficult for the patient. This issue should be clarified in the history. It is important to assess for symptoms of other diseases that are known to cause photosensitivity and to determine whether a family history of photosensitivity exists.
Establishing whether the photosensitivity can be elicited with exposure to sunlight through window glass may provide information about the wavelengths of light that cause the response. Ultraviolet (UV)-B light does not penetrate window glass, whereas UV-A light and visible light do.
In most patients, the findings of the physical examination will suggest a photosensitivity reaction. The clinician should inquire about any history of intolerance to the sun and should ask patients who report photosensitivity about the medications they are taking and the products they are applying to their skin. Common products such as sunscreens, fragrances, and (in some cases) antibacterial soaps are capable of causing photoallergic reactions when applied to the skin.
Both phototoxic and photoallergic reactions occur in sun-exposed areas of skin, including the face, V of the neck, and dorsa of the hands and forearms. The hair-bearing scalp, postauricular and periorbital areas, and submental portion of the chin are usually spared. A widespread eruption suggests exposure to a systemic photosensitizer, whereas a localized eruption indicates a reaction to a locally applied topical photosensitizer.
Phototoxic reactions in skin
Acute phototoxicity often begins as an exaggerated sunburn reaction with erythema and edema that occurs within minutes to hours of light exposure. (See the image below.) Vesicles and bullae may develop with severe reactions. The lesions often heal with hyperpigmentation, which resolves in a matter of weeks to months. Chronic phototoxicity may also appear as an exaggerated sunburn reaction. However, lichenification often develops because of repeated rubbing and scratching of the photosensitive area. Thus, distinguishing phototoxic reactions from photoallergic reactions strictly on the basis of the physical appearance of the lesions may be difficult.
View Image
Phototoxic reaction.
Other less common skin manifestations of phototoxicity include pigmentary changes. A blue-gray pigmentation is associated with several agents, including amiodarone, chlorpromazine, and some tricyclic antidepressants (TCAs). Reactions to psoralen-containing botanicals (phytophotodermatitis) and drugs may resolve with a brownish discoloration. Frequently, the pigmentary change is preceded by a typical sunburn reaction. If the reaction is not severe, some patients may not notice the erythema.
Photosensitizing drugs may also cause a lichen planus–like eruption in sun-exposed areas. Drugs likely to cause this type of reaction include demeclocycline, hydrochlorothiazide, enalapril, quinine, quinidine, chloroquine, and hydroxychloroquine.
Pseudoporphyria, which involves porphyria cutanea tarda–like changes of skin fragility and subepidermal blisters on the dorsa of hands, may occur after exposure to naproxen, nalidixic acid, tetracycline, sulfonylureas, furosemide, dapsone, amiodarone, bumetanide, and pyridoxine. Frequent use of sun-tanning beds and chronic renal failure are other predisposing factors.
Phototoxic reactions in nails
Photo-onycholysis, or separation of the distal nail plate from the nail bed, is another manifestation of phototoxicity. Photo-onycholysis has been reported with the use of many systemic medications, including tetracycline, psoralen, chloramphenicol, fluoroquinolones, oral contraceptives, quinine, voriconazole,[39] and mercaptopurine. In individuals with heavily pigmented skin, photo-onycholysis may be the only manifestation of phototoxicity.
Photoallergic reactions in skin
Photoallergic reactions typically develop in sensitized individuals 24-48 hours after exposure. The reaction usually manifests as a pruritic eczematous eruption. Erythema and vesiculation are present in the acute phase. More chronic exposure results in erythema, lichenification, and scaling. Hyperpigmentation does not occur in photoallergic reactions. (See the image below.)
Chronic cutaneous effects of repeated phototoxic injury have been evaluated only with psoralen-containing compounds. Premature aging of the skin, lentigines, and skin cancer are common. With respect to skin cancer, increases in incidence are greater for squamous cell carcinoma (SCC) than for basal cell carcinoma (BCC). The incidence of melanoma may also increase over time. The effects of chronic exposure to virtually all other photosensitizing compounds are unknown.
Persistent light reactivity is a form of chronic actinic dermatitis that occurs in patients with photoallergic contact dermatitis. In patients with persistent light reactivity, photosensitivity persists for months or years after the offending agent is eliminated. The disease may involve all sun-exposed areas and may spread to covered areas of skin as well. Initially, persistent light reactivity is commonly misdiagnosed as atopic dermatitis or a lichenoid drug reaction.
The photosensitivity associated with persistent light reactivity can be incapacitating because patients are sensitive not only to UV-A light but also to both UV-B and visible light. Some patients confine themselves to darkened rooms because of their severe photosensitivity. Although systemic drugs (eg, thiazides and quinidine) have been implicated as causes of persistent light reactivity, the most frequent causative agents are sunscreens, halogenated salicylanilides, and musk ambrette.
Treatment of persistent light reactivity involves avoidance of contact with exacerbating agents and photoallergens. Emollients, topical steroids, systemic steroids, and (at times) hydroxychloroquine have been used. Paradoxically, psoralen UV-A (PUVA) and narrowband UV-B have been used in treatment, though relapse is common. Patients who show no signs of improvement may require treatment with an immunosuppressive agent (eg, azathioprine or cyclosporine).
To exclude porphyria cutanea tarda, assessment of urine porphyrin levels is warranted; levels are elevated in porphyria cutanea tarda but within the normal range in pseudoporphyria and drug-induced photosensitivity. Antinuclear antibody (ANA) and anti-Ro (SS-A) antibody levels should also be determined.
Photopatch testing is an important tool in the diagnosis of photoallergic contact dermatitis. Suspected photoallergens are applied to the back in two sets. One set is removed after 24 hours and irradiated with ultraviolet (UV)-A at 5-10 J/cm2. After 48 hours, both sets are evaluated for a positive reaction. Erythema, edema, or vesiculation at an irradiated site indicates a positive reaction. A positive reaction at both sites is interpreted as an allergic contact dermatitis. A positive reaction at the unirradiated site with a stronger one at the irradiated site should be interpreted as both allergic dermatitis and photoallergic contact dermatitis in response to the same compound.
Phototesting with UV-A, UV-B, and sometimes visible light is helpful in diagnosing photosensitivity disorders. This test is performed by exposing small areas of skin on the back or inner aspect of the forearms to gradually increasing doses of light. The minimum dose of light required to produce uniform erythema over the entire irradiated site after 24 hours is called the minimum erythema dose (MED). Patients with phototoxic reactions have a reduced MED to UV-A or, in some instances, UV-B.
In acute phototoxic reactions, necrotic keratinocytes are observed. If the reaction is severe, the necrosis is panepidermal. In addition, epidermal spongiosis with dermal edema and a mixed infiltrate consisting of lymphocytes, macrophages, and neutrophils may be present. Blue-gray pigmentation associated with phototoxic reactions results from increased melanin in the dermis or deposition of the drug or its metabolites within the skin.
The histologic features of a lichen planus–like phototoxic reaction are essentially indistinguishable from those of idiopathic lichen planus. However, increased amounts of spongiosis and necrotic keratinocytes may be present.
The histologic features of a subacute cutaneous lupus erythematosus (SCLE)-like reaction reveal an interface dermatitis that is indistinguishable from non-drug-induced SCLE.
Pseudoporphyria, like porphyria cutanea tarda, causes a subepidermal blister at the level of the lamina lucida. A characteristic feature of both pseudoporphyria and porphyria cutanea tarda is festooning, which refers to the irregular configuration of the dermal papillae in the floor of the bulla.
Histologically, photoallergic reactions are similar to contact dermatitis. Epidermal spongiosis with a dermal lymphocytic infiltrate is a prominent feature. However, the presence of necrotic keratinocytes is suggestive of photoallergy rather than allergic contact dermatitis.
The mainstays of treatment of drug-induced photosensitivity are as follows:
Identification and avoidance of the causative agent
Use of sun protection
Institution of measures for symptomatic relief
Topical corticosteroids and cool compresses may alleviate drug-induced photosensitivity. Systemic corticosteroids should be reserved for the most severe cases.
If sunscreens are not the cause of the photosensitivity, they should be used liberally. It should be noted that the sun protection factor (SPF) may not be a reliable indicator of protection against drug-induced photosensitivity. SPF is a measure of the degree of protection against sunlight-induced sunburn, primarily that caused by ultraviolet (UV)-B. However, most drug-induced photosensitivity reactions are caused by wavelengths in the UV-A range. Therefore, sunscreens that absorb UV-A should be prescribed. Sunscreens containing avobenzone, titanium dioxide, and zinc oxide block UV-A radiation more effectively than sunscreens containing other ingredients.
A case report described successful use of a Polypodium leucotomos extract to treat a severe vandetanib-induced phototoxic reaction that was refractory to topical steroids and unresponsive to sun avoidance.[40] Further research would be required to define this agent's mechanism of action and potential clinical utility.
Patients who experience drug-induced photosensitivity should identify and avoid the causative agent. They should also use a sunscreen, provided that a sunscreen is not the offending agent. Sun protection often prevents photosensitivity reactions.
What is drug-induced photosensitivity?What is the pathophysiology of phototoxicity in drug-induced photosensitivity?What is the pathophysiology of photoallergic reactions in drug-induced photosensitivity?What causes drug-induced photosensitivity?What is the incidence of drug-induced photosensitivity in the US?What is the global incidence of drug-induced photosensitivity?What are the racial predilections of drug-induced photosensitivity?What are the sexual predilections in the prevalence of drug-induced photosensitivity?Which age groups have the highest risk for drug-induced photosensitivity?What is the prognosis of drug-induced photosensitivity?What should be included in patient education about drug-induced photosensitivity?Which clinical history findings are characteristic of drug-induced photosensitivity?What is the role of photodynamic therapy (PDT) in drug-induced photosensitivity?How is drug-induced photosensitivity differentiated for other conditions of photosensitivity?Where do phototoxic and photoallergic reactions typically occur in drug-induced photosensitivity?Which physical findings are characteristic of phototoxic reactions in drug-induced photosensitivity?Which physical findings are characteristic of photo-onycholysis in drug-induced photosensitivity?Which physical findings are characteristic of photoallergic reactions in drug-induced photosensitivity?What are possible complications of drug-induced photosensitivity?What are the differential diagnoses for Drug-Induced Photosensitivity?What is the role of lab testing in the diagnosis of drug-induced photosensitivity?What is the role of photopatch testing in the diagnosis of drug-induced photosensitivity?What is the role of phototesting in the diagnosis of drug-induced photosensitivity?Which histologic findings are characteristic of drug-induced photosensitivity?How is drug-induced photosensitivity treated?How is drug-induced photosensitivity prevented?What is the goal of drug treatment for drug-induced photosensitivity?Which medications in the drug class Corticosteroids, Topical are used in the treatment of Drug-Induced Photosensitivity?
Alexandra Y Zhang, MD, Lucent Dermatology and Skin Surgery Center
Disclosure: Nothing to disclose.
Coauthor(s)
Craig A Elmets, MD, Professor and Chair, Department of Dermatology, Director, Chemoprevention Program Director, Comprehensive Cancer Center, UAB Skin Diseases Research Center, University of Alabama at Birmingham School of Medicine
Disclosure: Serve(d) as a director, officer, partner, employee, advisor, consultant or trustee for: University of Alabama at Birmingham; University of Alabama Health Services Foundation<br/>Serve(d) as a speaker or a member of a speakers bureau for: Ferndale Laboratories<br/>Received research grant from: NIH, Veterans Administration, California Grape Assn<br/>Received consulting fee from Astellas for review panel membership; Received salary from Massachusetts Medical Society for employment; Received salary from UpToDate for employment. for: Astellas.
Specialty Editors
David F Butler, MD, Former Section Chief of Dermatology, Central Texas Veterans Healthcare System; Professor of Dermatology, Texas A&M University College of Medicine; Founding Chair, Department of Dermatology, Scott and White Clinic
Disclosure: Nothing to disclose.
Jeffrey P Callen, MD, Professor of Medicine (Dermatology), Chief, Division of Dermatology, University of Louisville School of Medicine
Disclosure: Received income in an amount equal to or greater than $250 from: Biogen US (Adjudicator for study entry cutaneous lupus erythematosus); Priovant (Adjudicator for entry into a dermatomyositis study); IQVIA (Serono - adjudicator for a study of cutaneous LE) <br/>Received honoraria from UpToDate for author/editor; Received royalty from Elsevier for book author/editor; Received dividends from trust accounts, but I do not control these accounts, and have directed our managers to divest pharmaceutical stocks as is fiscally prudent from Stock holdings in various trust accounts include some pharmaceutical companies and device makers for these trust accounts for: Stocks held in various trust accounts: Allergen; Amgen; Pfizer; 3M; Johnson and Johnson; Merck; Abbott Laboratories; AbbVie; Procter and Gamble;; Celgene; Gilead; CVS; Walgreens; Bristol-Myers Squibb.
Chief Editor
Dirk M Elston, MD, Professor and Chairman, Department of Dermatology and Dermatologic Surgery, Medical University of South Carolina College of Medicine
Disclosure: Nothing to disclose.
Additional Contributors
Abdul-Ghani Kibbi, MD, FACP, Professor and Chair, Department of Dermatology, American University of Beirut Medical Center, Lebanon
){kind=link}
){kind=link}
){kind=link}
){kind=link}