Current Treatment of Cutaneous Lupus Erythematosus
Department of Dermatology, University of Pennsylvania
Victoria Werth MD
Dermatology Online Journal 7(1): 2
Cutaneous lupus erythematosus has a variety of clinical presentations, but treatment is similar for the different forms. The antimalarials remain the cornerstone of treatment because of their effectiveness and safety. Optimal use of the antimalarials as well as alternative therapy options are discussed.
Cutaneous lupus erythematosus encompasses at least 10 to 15 different clinical presentations. Some of the common types include chronic cutaneous lupus erythematosus (LE), which includes localized and generalized discoid lupus erythematosus (DLE), hypertrophic LE (a form of DLE), tumid LE, and lupus panniculitis. Subacute cutaneous LE and systemic LE each have several different cutaneous presentations. The treatment options are fairly similar for all the different types of cutaneous LE, although frequently patients with systemic LE require additional therapy for systemic manifestations of the disease.
Ultraviolet light plays a significant precipitating role in most forms. When keratinocytes in culture are irradiated with ultraviolet light, nuclear and intracytoplasmic antigens actually translocate to the keratinocyte membrane, allowing antibody binding. Binding of antibody such as anti-SSA, anti-SSB, or anti-RNP to the cell membrane starts a whole immunologic cascade that culminates in the production of cutaneous lupus lesions. A variety of effects on irradiated keratinocytes have been documented. Adhesion molecules such as ICAM-1 are induced and the production of various cytokines is stimulated by ultraviolet. Elevations in PGE2 and IL-10 can be measured. The apoptotic effects are easily visible. Adhesion molecules are also induced in post capillary venules, facilitating the entry of lymphocytes into the skin. There are a number of genetic polymorphisms that are affected by ultraviolet-induced pro-inflammatory cytokines. In particular, a polymorphism of tumor necrosis factor-alpha (TNF-alpha) may be important in certain photosensitive patients with SCLE.
Complement plays a role in many different aspects of lupus. Various complement deficiencies, such as C2 and C4, as well as deficiencies of other components related to the classical pathway have been associated with SLE and SCLE. In DLE compliment deficiencies are less common, but occasional deficiencies of C1 inhibitor, C1Q, C2 and C4 have been reported. These relationships are not surprising in that complement is important in opsonization of immune complexes which allows clearance. It is easy to imagine how compromise of immune complex clearance could influence lupus development and severity.
Localized DLE represents the most common presentation of cutaneous lupus in the dermatology clinic. Up to 25 to 40 percent of these patients actually remit spontaneously with time and only 5 percent will go on to develop systemic lupus. Scarring is a significant problem and these patients must be treated aggressively to prevent permanent scarring and scarring alopecia. Initially, topical and intralesional therapy is appropriate, but if a rapid response is not obtained or there are numerous areas of involvement, systemic therapy is required. Once the hair is lost, it will not regrow. Tumid lupus may occur in the scalp which may look like alopecia areata, but with more infiltrated, erythematous plaques. It is interesting that about 10% of my patients with lupus also have alopecia areata. These patients regrow their hair well with intralesional steroids. In addition, more diffuse forms of hair loss are seen in lupus, particularly systemic forms. Generalized DLE is more troublesome to treat. These patients typically have lesions both above and below the neck. About 20% of these patients will go on to develop systemic lupus symptoms. These patients should be followed more closely for evidence of systemic disease. The hypertrophic variant of DLE, seen in about 2 percent cutaneous LE, has similar responses to treatment as other forms of DLE. Making the diagnosis of hypertrophic LE may be difficult and histologic examination is very important. Acral DLE is a little more difficult to treat. In general these patients are more disabled by their skin disease and antimalarials are not as effective. Paradoxically, sometimes very low doses of antimalarials (25-50 mg per day) are actually more helpful in these patients. The subacute cutaneous LE variant is more often associated with anti-SSA antibody than typical DLE.
Lupus panniculitis is also often difficult to diagnose as other forms of panniculitis may present similarly. If DLE is present over the panniculitis, making the diagnosis is simplified. Scarring and atrophy are often a serious problem and it is not uncommon on the face.
SCLE may be seen in about 10% of our LE patients. The cutaneous lesions usually take the form of psoriasiform plaques or annular polycyclic plaques. Both of these forms are typically non-scarring as opposed to DLE lesions. SCLE patients are usually very photosensitive. About 50% of the time, these patients will meet the criteria for SLE and occasionally these patients will also show acute cutaneous LE lesions. Arthritis, arthralgias, malaise, myalgias and fever are not uncommon in this subset. However, central nervous system and renal disease are much less common in SCLE than in SLE.
Drug-induced SLE and SCLE must not be forgotten. LE patients should always be questioned about their medications. In particular, thiazide diuretics, penicillamine, glyburide and calcium channel blockers (diltiazem, felodipine) have been reported to cause SCLE. We recently had a case of furosemide-induced SCLE, which recurred upon rechallenge. It is important to remember that stopping the drug may not stop the disease process.
Other types of drug eruptions may mimic LE. One patient developed a lichenoid eruption associated with captopril. Subsequent administration of enalopril produced an extensive erythematous eruption, that looked very much like SCLE. These types of drug eruptions may continue for as long as a year after the offending drug is discontinued.
Acute cutaneous LE generally involves photo-distributed erythema, particularly facial or malar erythema. A more erosive or even TEN-like eruption may also occur. Biopsy is extremely important in these patients. Furthermore, periorbital edema may be associated with both SLE and dermatomyositis and may be a presenting sign. Bullous SLE is quite rare. To diagnose this, the patient must meet the SLE criteria. Bullae are most prominent on sun-exposed skin, but are often on other areas as well. Neutrophils, not lymphocytes will be most prominent in histologic sections. Direct immunofluorescence will show a linear pattern of immunoglobulin and complement at the dermo-epidermal junction.
Neonatal LE generally becomes apparent after sun exposure. The infants usually clear within six months when the transplacentally derived maternal SSA and SSB antibodies have disappeared. Though the skin disease is the most common component of neonatal LE, it is a systemic, multi-system disease. Congenital heart block is the most worrisome complication. Mothers who have a history of having a prior baby with neonatal LE are followed very closely during subsequent pregnancies as these babies have a 25% chance of having neonatal LE. Particular attention is paid to diagnosing heart block early and treating the mothers to prevent complications. Forty percent of mothers of infants with neonatal LE have symptoms of Sjogren's syndrome or LE at the time of delivery. The remaining 60% will get some symptoms of LE with time. When Ro positive women are studied, however, the incidence of neonatal LE in their offspring is only about 5%. To treat the babies' cutaneous lesions, generally only hydrocortisone cream is required.
When we begin treating patients with cutaneous LE, it is essential to evaluate them for signs of systemic disease. I work closely with the rheumatologists when significant systemic disease is present. The initial evaluation should include a history, physical exam, CBC, sedimentation rate, ANA and urinalysis. If ANA positive, or if suspicious for SLE, but ANA negative, I get a panel of tests that includes anti-SSA, anti-SSB, anti-RNP, anti-double stranded DNA and anti-Smith antibodies. Complement studies are also useful. Those with positive tests will be followed more closely. In particular, those with high titer anti-dsDNA antibody are at more risk for renal problems and following urinalyses and creatinine levels closely is advisable. Anti-Smith antibody is demonstrated in just 20% of SLE patients, but is very specific for the disease. Anti-SSA is a marker for SCLE and Sjogren's syndrome. About 40% of LE patients will have antiphospholipid antibodies, which are associated with thrombosis and spontaneous abortion. I recommend low dose aspirin therapy for patients with anticardiolipin antibody. If they have actually had any thrombotic events, prolonged treatment with coumadin is appropriate.
Treatment must start with avoidance of precipitating factors such as heat, certain drugs and, of course sunlight. The cornerstone of treatment is sunscreen and sun avoidance. Provocative phototesting has shown that, as expected, SCLE patients are the most photosensitive. After one week of irradiation, 64% will develop skin lesions in areas of irradiation within three weeks. In comparison, 42% of chronic cutaneous LE patients and 25% of SLE patients developed lesions. UVB is the most common precipitating part of the ultraviolet spectrum, but half of the LE patients appear to respond to UVA as well as UVB. Fourteen percent respond only to UVA and not UVB. However, it must be mentioned that this testing was done with UVA doses of 100 Joules per sq. cm, much higher than physiologic doses. Nevertheless, it seems reasonable to make an effort to effectively block UVA as well as UVB in LE patients. Remember that UVA through a car windshield may be a stimulating factor. Many patients, who claim that they are not photosensitive, actually had abnormal photo-provocative testing. In these patients there is often a delayed reaction, as much as two to three weeks. Hence, protection is crucial even in those patients who do not recognize photosensitivity. I insist that my patients wear a sunscreen with an SPF of 30 with Parsol 1789 (avobenzone) as a UVA screening agent.
Topical corticosteroids are the mainstay of treatment in DLE patients. I usually start with a Class I steroid in severe DLE patients. Patients are given a high, medium and low potency steroid. They start with the most potent and are instructed that as they improve, they are to taper to the medium and low potency preparations in sequence. Intralesional steroids are particularly useful in the scalp. I usually use Kenalog at 5 mg per ml. A longer acting steroid may also be utilized alone or mixed together with the Kenalog for a more prolonged effect.
Antimalarials are crucial to the treatment of cutaneous LE. Hydroxychloroquine, chloroquine and quinacrine are the three that are used. For most patients, hydroxychloroquine is started at 200 mg per day to assess gastrointestinal tolerance. If the patients are not having diarrhea or GI upset, the dose is doubled to 200 mg twice daily. Hydroxychloroquine should always be dosed at less than 6.5 mg per kilogram per day. In a thin patient, this may actually be only 300 mg per day. Problems with retinopathy are vanishingly small if doses are kept within this range. Wait six to eight weeks to evaluate effectiveness. Jeff Callen has reported that 70% of patients will respond to hydroxychloroquine alone. If monotherapy is not successful, add quinacrine at a dose of 100 mg per day. Of course, quinacrine is not marketed. Quinacrine powder (Sigma, veterinary grade) can be put into capsules. For six years dermatologists and rheumatologists have been using these preparations without problems. Panorama Pharmacy in California will prepare this and actually perform an additional purification step in the process. Many patients who have not been adequately controlled with hydroxychloroquine alone will respond within six to eight weeks of the addition of quinacrine. If chloroquine is used instead of hydroxychloroquine, one should use less that 3.5 mg per kilogram per day. Hydroxychloroquine and chloroquine should not be used together as was done in the early 70's with the introduction of the drug, Triquin. With this combination, the risk of retinopathy is too great. Bernstein studied the eye toxicity and concluded that both hydroxychloroquine and chloroquine are safe if used with the recommended guidelines. He claims that there have been no published cases of hydroxychloroquine retinopathy when the maintenance dose was less that 6.5 mg per kg per day, treatment duration was less that 10 years and there was no renal impairment. Nevertheless, I still have my patients monitored by an ophthalmologist every six months for hydroxychloroquine and every four months for chloroquine after a baseline exam in the first month of treatment. This approach, however, is considered unnecessary by many if the strict guidelines are followed. My approach is to start with hydroxychloroquine and add quinacrine if needed. However, if after 6-8 weeks on this combination the result is still unsatisfactory, hydroxychloroquine will be replaced by chloroquine while continuing quinacrine. Some patients will do better with this regimen. I also find that some patients can be taken off their antimalarials during the winter when ultraviolet exposure is significantly less.
Besides eye toxicity, some other adverse effects of antimalarials are sometimes seen. Skin effects include blue-gray hyperpigmentation, which warrants warning the patient at the outset, and urticaria. Gastrointestinal upset includes cramping, distention, nausea, diarrhea and heartburn. Musculoskeletal, flu-like symptoms are occasionally noted. Rare central nervous system effects include headache, nervousness, insomnia, psychosis and seizures. Hydroxychloroquine may lower the seizure threshold in patients who are already prone to seizures. Central nervous system LE can present with psychosis, which could be difficult to distinguish from that caused by hydroxychloroquine.
Quinacrine has somewhat different side effects. Eye toxicity is not a concern, but headache, dizziness and gastrointestinal symptoms may be a problem. In Vietnam, it was used for malaria prophylaxis in doses twice as high as we use in LE. At those doses, aplastic anemia was rarely seen, often preceded by a lichenoid eruption. At the recommended dose for LE of 100 mg per day or less, I have never seen a case of aplastic anemia. Several skin adverse effects have been reported. Yellow discoloration of the skin is common and especially noticeable in light skinned persons. Blue-black discoloration can also be seen similar to that seen with the use of hydroxychloroquine and chloroquine. Other skin eruptions with quinacrine may be eczematous, lichenoid or exfoliative in nature.
Pregnancy is not advised while using antimalarials. We attempt to discontinue them in our patients if pregnancy is anticipated and prefer to have them off the drugs for six months prior to conception due to the long storage of the drugs. There have been reports of pregnancies that have proceed without problems while the mother was taking hydroxychloroquine, but it is thought that quinacrine is more dangerous to the fetus.
The antimalarials have several mechanisms of action. First, they raise intracellular vacuolar pH. An acidic pH is necessary for antigen processing and presentation by dendritic cells. Hence with pH alteration, the inhibition of antigen processing and presentation causes down-regulation of the immune response against autoantigens. In addition, antimalarials inhibit the release by monocytes of pro-inflammatory cytokines such as IL-1, IL-6 and TNF-alpha. Inhibition of granulocyte migration and Phospholipase A2 activity have also been reported. Therefore, the antimalarials may be acting in a variety of ways and it is unclear which mechanism is the most important. Other desirable effects of hydroxychloroquine include the ability to block platelet aggregation and adhesion, which would reduce the size of thrombi without prolonging bleeding time. Michelle Petrie has found that hydroxychloroquine decreases the cholesterol level which is useful in SLE patients taking steroids that elevate these levels. In Britain, hydroxychloroquine is used from thrombosis prophylaxis. This antithrombotic effect may be very beneficial in LE patients with antiphospholipid antibodies and thrombotic problems. There are recommendations in the rheumatologic literature to keep LE patients on hydroxychloroquine even after they are clinically clear because of the long term benefit. In addition, fatigue, fever, headache, arthralgias, arthritis, pleuritis and pericardial inflammation may be improved with the use of antimalarials. Many of these symptoms will recur when antimalarials are withdrawn. Quinacrine, in particular, has been useful for fatigue.
Dapsone and retinoids (isotretinoin, etretinate) are somewhat less useful in cutaneous LE. Some SCLE patients will respond favorably to dapsone and it may be used in antimalarial failures. Dapsone is particularly helpful in bullous LE and may be the treatment of choice in this condition. Dapsone is also a good choice in LE patients with vasculitis. Retinoids have been used in DLE with success in a few cases, but does not seem to be a panacea. Azathioprine may be helpful in several types of LE including DLE. Gold has been reported to be beneficial, but I have no experience with it. Thalildomide has a role in the treatment of resistant LE patients.
Methotrexate may be a helpful adjunct in patients unresponsive to antimalarials. LE patients seem to have some trouble tolerating oral methotrexate and the subcutaneous route is sometimes preferred. It is not clear if there are differences in incidence of methotrexate-related liver disease in rheumatologic patients as opposed to psoriasis patients. I generally follow the same liver biopsy guidelines as recommended for psoriasis patients.
Systemic corticosteroids must be mentioned, but I am so familiar with the adverse effects in my bullous disease patients that I prefer not to use them in LE. There are good alternative therapies for most cutaneous LE patients. Obviously, LE patients with severe systemic disease will be more likely to require corticosteroids. However, patients with LE are more prone to getting avascular necrosis and also have an increased tendency to form thrombi which also make the use of systemic steroids less desirable. Occasionally, in patients who are rapidly losing hair, it is reasonable to give a short course of steroids at the very beginning of antimalarial therapy before they have had time to become efficacious.
© 2001 Dermatology Online Journal