Practical Therapeutics Drugs 12 : 69-77 (1976) © ADIS Press 1976
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
D. Alsrcon-Seqovie Department of Immunology and Rheumatology , Instituto Nacional de la Nutrlcien. Mexico
Summary
Drugscapableoftriggering the onset ofsystemic lupus erythematosus may be divided into those that do so by pharmacological properties of their own and those that do so by eliciting allergicreactions which bring about lupus. Drugs in the first group vary in their potency to activate lupus. They all elicit antinuclear antibodies in the majority of patients who receive them but they only cause lupus in a small percentage of patients. This dichotomy suggests that a predisposition is required for the development of lupus upon intake of these drugs. The mechanism whereby these drugs elicit antinuclear antibodies seems to relate to coupling to and/or modification of, nuclear antigens. The patterns ofantinuclear antibodies elicited by these drugs in individuals who receive them correlate well with their known reactivity with variousnuclear antigens.
Development of overt systemic lupus erythematosus (SLE) seems to result from an interplay of genetic and environmental factors (AlarconSegovia, 1969). Thus, patients with SLE, rather than having inherited the disease seem to have inherited a disturbance, perhaps immunological, which upon contact with environmental factors may lead to the disease. Prominent among the environmental agents that may trigger the onset of SLE are ultraviolet radiation and some drugs. It is common practice to make the distinction between spontaneously occurring and drug-induced SLE. Although it seems that spontaneously occurring SLE pre-
dominates 10 to lover the drug-induced disease (Lee et al., 1966; Dubois, 1974) unaccounted drug consumption is so great that this ratio may actually be lower.
1. Classification of Lupus-Activating Drugs
Drugs capable of triggering the onset of SLE may be classified into two groups (AlarconSegovia, 1969). The first group includes those drugs which require prolonged ingestion in order to elicit the
70
Drug- Induced Antinuclear Antibodies and Lupus Syndromes
clinical manifestations of lupus. Other characteristics of the relationship of drugs in this group and SLE are: 1) That no allergic reaction to the drug seems to mediate the onset of lupus 2) That larger numbers of cases have been recorded with drugs in this group, and 3) That these drugs elicit antinuclear antibodies in a large proportion of the individuals who receive them for some time (Alarcon-Segovia and Fishbein, 1975). Indeed, development of antinuclear antibodies while on these drugs is considerably more common than actual development of SLE.
Table I. Drugs reported to induce or activate SLE Group I (by pharmacological action)
Group II (by allergic reaction)
Hydrallazine Procainamide Anticonvulsants mephenytoin phenytoin pr imidone trimethadione ethosuximide carbamazepine pheneturide (phenylethylacety Iurea) Isoniazid Chlorpromazine
Aminosalycilic acid Chlorthalidone' D-penicillamine Griseofulvin Guanoxan Isoquinazepon Levodopa Methyldopa' Methysergide Methylthiouracil Oral contraceptives Oxyphenisatin Penicillin Phenylbutazone Practolol Propylthiouracil Quinidine Reserpine Streptomycin Sulphonamides Tetracycline Tolazamide
Drugs included in this group are listed in table I. The second group of drugs capable of triggering SLE includes those which: 1) May elicit SLE after brief periods of use 2) Often seem to do so by causingallergicreactions which may be quite dramatic (Dornz et al., 1959 ; Rallison et al., 1961) 3) Cause SLE only rarely, and 4) Do not often elicit antinuclear antibodies in patients receiving them (Cetina et al., 1972). Drugs in this second group are also listed in table I. It should be noted that two of these drugs, and perhaps some others, may actually belong to the flrst group. Methyldopa has been reported to cause lupus in a few instances, and to elicit antinuclear antibodies in a larger percentage of patients receiving the drug (Feltkamp et al., 1970) but, has also been found to cause, and this seems to be more frequent , positive Coombs tests (Carstairs et al., 1966), as well as 'full blown' haemolytic anaemia (Worllege et al., 1966). Chlorthalidone has been reported to cause antinuclear antibodies in patients receiving it (Feltkamp et al., 1970) but this has neither been confirmed nor have lupus patients placed on this drug
1
May belong to group I
been noticed to have exacerbations of the disease (Alarcon-Segovia, personal observations). Overt lupus caused by this drug has not been well documented . Another variant of this problem has recently been described: patients on a combination of phenopyrazone, horse-chestnut extract and cardiac glycosides ('Venocuran') for alleged treatment of venous insufficiency have been found to develop a lupus-like illness which is accompanied by mitochondrial antibodies but not by antinuclear antibodies (Grob et al., 1975). 90% of long-term users of 'Venocuran' develop mitochondrial antibodies while only 10% of the whole group develop the full blown syndrome (Grob et al., 1975).
Drug-Induced Antinuclear Antibod ies and Lupus Syndromes
2. The Lupus-Activating Potency of Various Drugs
71
age group. Also to be considered, is the possibility that individuals who develop lupus late in life and only after being exposed to a potent lupus-activatThe relative lupus-activating potency of the ing agent, may have a lesser predisposition to the various drugs in group I can only be inferred from development of lupus. This in turn may lead to a the incidence of overt lupus in patients who milder disease, more likely to remit upon withreceive them. Hydrallazine, which was the first drawal of the offending drug. drug noted to have this side-effect, was found to Another consideration pertinent to anticonvulscause lupus in 10 to 15% of patients who receive it ants , and perhaps chlorpromazine, is that convullong-term for the treatment of hypertension sions or psychosis may be early manifestations of (Dustan et aI., 1954; Perry and Schroeder, 1954; lupus. In such case the drugs would be exacerMuller et aI., 1955). bating rather than eliciting the disease. HyperProcainamide seems to be even more potent for tension or cardiac arrhythmias, as indications for it may cause a lupus-like illness in as many as 30% hydrallazine or procainamide, respectively, are less of patients receiving it (Fakhro et aI., 1967). likely to be early symptoms of SLE. Anticonvulsants, isoniazid and chlorpromazine These facts considered, hydrallazine-activated seem to follow hydrallazine in order of their lupus has been found to be indistinguishable, both lupus-activating potency. clinically (Alarcon-Segovia et al., 1967) and seroThe lupus-activating potential of drugs in group logically from spontaneous SLE (Hahn et aI., II probably resides in their antigenicity and/or 1972). Although a paucity of renal lesions in their capacity of behaving as haptens, l.e. coupling hydrallazine SLE has been mentioned as an arguto proteins or cell membranes. In turn , allergic ment against the indentity of hydrallazine-induced reactions may be more frequent in lupus or lupus- and spontaneously occurring SLE, it may only reprone patients. That immunological challenge may flect the lesser incidence of renal lesions in the activate SLE was shown by Zingale and his elderly lupus patient. Nevertheless, renal lesions co-workers (1963). have been found both clinically and histologically in a few patients with hydrallazine lupus (Dammin et al., 1955; Bendersky and Ramirez, 1960; Alarcon-Segovia et aI., 1967). 3. Clinical Picture ofDrug-Induced Lupus Procainamide-induced lupus-like disease is less In general, the clinical picture of drug-induced similar to spontaneous lupus than its hydrallazine SLE differs little from that of the spontaneously counterpart. Pleural and pleuropulmonary involveoccurring disease. Because hydrallazine, an anti- ment are considerably more frequent in the prohypertensive agent and procainamide, an anti- cainamide disease than in spontaneous lupus arrhythmic drug are usually given to elderly pat- (Blomgren et al., 1972). Indeed, pleurisy, which ients, the mean age of patients with drug-induced was the initial symptom in 30% of 54 cases of lupus is considerably higher than that of the spon- procainamide-induced disease (Alarcon-Segovia, taneously occurring disease. Since spontaneously 1969) occurred initially in only 2.5% of the lupus occurring lupus tends to be less florid and severe in patients in the series of Dubois and Tuffanelli the older adult than in the young (Rupe and (1964) which were mostly of the spontaneously Nickel, 1959) and in adults less than in children occurring form. Serologically there are also impor(Meislin and Rothfield , 1968), the lesser severity tant differences between spontaneous SLE and the of drug-induced lupus correlates well with that of procainamide-induced syndrome. Anti-native DNA the spontaneously occurring disease in the same antibodies do not occur in the procainamide dis-
Drug-Induced Ant inuclear Ant ibodies and Lupus Syndromes
72
duals who receive lupus-activating drugs belonging to group I suggests that the potential of these drugs to elicit lupus syndromes resides in their capacity to induce antinuclear antibodies. The dichotomy between the development of antinuclear antibodies in most individuals on these drugs, and the development of clinical disease in only a few may depend on factors predisposing to development of lupus. We once used the term 'lupus diathesis' for this predisposition to the development of lupus which may be revealed by environmental factors such as drugs (Alarcon-Segovia et al, 1965). I now believe that it is more likely to be a predisposition to autoimmune disease of various kinds rather than solely to SLE; a ' rnultipotential immunopathogenic predisposition' . Lupusactivating drugs would not only reveal this predisposition but , by their elicitation of antinuclear antibodies would determine that it is lupus which develops rather than other autoimmune disease. Other drugs, such as methyldopa would primarily elicit Coombs antibody and determine the development of autoimmune haemolytic anaemia rather than that of SLE. The antigenic specificity of the antinuclear antibodies elicited by the various lupus-activating drugs in subjects who have taken them for a long time without developing lupus may be important for the understanding of their mode of action (Alarcon-Segovia and Fishbein, 1975). Thus, antibodies found in isoniazid-treated subjects are primarily directed to soluble nucleoprotein (sNP), an antigen which is physicochemically altered in vitro, and probably also in vivo by isoniazid, which may also bind to it (Fishbein and Alarcon-Segovia, unpublished observations). Antibodies to isoniazid-altered sNP have been found to be present in the serum of nearly 80% of isoniazid-treated 4. Provocation of Antinuclear Antibodies by tuberculosis patients (Alarcon-Segovia et al., 1969) Drugs while only 42% had them to particulate nucleoprotein. Tuberculosis patients on isoniazid do not The fact that there is a considerably higher pre- develop antibodies to DNA, either native or devalence of antinuclear antibodies than that of natured. They may, however, have anti DNA actual lupus erythematosus syndromes in indivi- antibodies if they develop a lupus syndrome
ease (Winfield and Davis, 1974). It has recently been found that patients receiving procainamide develop antibodies to ribonucleoprotein, akin to those found in mixed connective tissue disease (Sharp et al., 1972), following short-term therapy with procainamide (Winfield et al., 1975). It is not unlikely that, contrary to what seems to be the case in hydrallazine-lupus, procainamide-induced syndromes may not be true SLE, but a drug-induced disease closely resembling it. There are no apparent significant differences between the isoniazid-induced and the spontaneously occurring lupus (Delpierre et al., 1971). The observations made on chlorpromazine-induced lupus are too scant, as yet, to evaluate. The original impression that patients who develop lupus while on hydrallazine will go into complete remission upon discontinuation of the drug proved to be false. Persistence of hydrallazine lupus after drug withdrawal has been well documented (Hildreth et al., 1960). Manifestations of lupus have been found present up to 9 years after the eliciting drug was withdrawn (Alarcon-Segovia et al., 1967) . Procainarnide-induced lupus-like syndromes appear to remit better after the drug is stopped. Patients have been recorded, however, whose disease did not remit when the drug was stopped, and who even required corticosteroid treatment (Paine, 1965; Byrd and Schanzer, 1969). Patients who develop SLE while on anticonvulsants seem to be the least likely to remit completely as a result of cessation of the drug (Alarcon-Segovia, personal observations). This may mean that convulsions actually were initial symptoms of SLE.
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
(Alarcon-Segovia et al., 1969). Hydrallazine has been found to bind similarly to, and physicochemically alter, sNP (Tan, 1974), but no data are available on the antigenic specificity of antinuclear antibodies elicited in hypertensive subjects treated with hydrallazine who have not developed a lupus syndrome. Different anticonvulsants also give rise to different patterns of antinuclear antibodies (Alarcon. Segovia et al., 1972). They elicit antibodies to sNP primarily (59%) but may give rise to antibodies to both native (14%) and denatured (22%) DNA. This is particularly true of the hydantoins which are also unique among the lupus-activating drugs in that they elicit antibodies to Sm antigen (AlarconSegovia and Fishbein, 1975). This antibody is thought to be as characteristic of SLE as antidouble stranded DNA antibody (Notman et al., 1975). The antinuclear antibodies found most frequently in patients on chlorpromazine are those directed to denatured DNA (Quismorio et al., 1972; Alarcon-Segovia et al., 1973), a fact which correlates well with the known reactivity of chlorpromazine with denatured DNA (Ohnishi and McConnell, 1965; Kahn and Davis, 1970) . Asymptomatic subjects receiving procainamide have been found to develop antibodies to ribonucleoprotein shortly after they are placed on this drug (Winfield et al., 1975) . Apparently, this antibody tends to disappear and be replaced by antibodies to single-stranded DNA which occurs in 36% of asymptomatic subjects on long-term procainamide and to nucleoprotein which are found in 68% of them (Dubois et al, 1968). It should be noted that the antigenic specificity of the antinuclear antibodies found in asymptomatic subjects on each of the various lupus-activating drugs may be different from that of the antinuclear antibodies found in patients who actually develop lupus while on them. Patients with hydrallazine-activated lupus are known to have antibodies to native DNA (Hahn et al., 1972) which correlates well with the finding
73
that immunisation of rabbits with hydrallazinehuman serum albumin conjugates, results in their development of both anti-native and anti-single. stranded DNA antibodies (Yamauchi et al., 1975) . Patients with procainamide induced lupus-like disease develop antibodies to denatured, but not to native DNA (Winfield and Davis, 1974) and they tend to stop producing antibodies to ribonucleoprotein, an antigen which may be released after myocardial infarction and could interact with procainamide (Winfield et al., 1975) . This interaction would be similar to that which occurs between this drug and DNA, which may explain the development of antibodies to DNA in patients on it (Elredge et al., 1974).
5. Predisposition of Systemic Lupus Erythematosus and Lupus-Activating Drugs: A Partnership The fact that only 10% of patients on highdose, long-term hydrallazine administration develop lupus, suggests that they have a predisposition for the development of lupus that the other 90% do not have (Alarcon-Segovia et al., 1965) . Further evidence of such predisposition can be found in the family and past histories of patients with drug-induced lupus syndromes (Shulman, 1963; Holley, 1964; Alarcon-Segovia et al., 1965; Lappat and Cawein, 1968; Novack and Paine, 1975). That such predisposition may differ quantitatively from one patient to another is suggested by the findings that some patients require higher doses of hydrallazine to develop lupus than others, and that those who develop it after having received lesser amounts of drug, have more premonitory symptoms and more tendency to persist after withdrawal of the drug than those who developed it after larger doses. One important factor influencing the development of antinuclear antibodies, as well as that of the drug-induced lupus syndromes, may reside in
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
the phenotypically determined acetylation rate of certain of these drugs, particularly hydrallazine and isoniazid which are acetylated by the same enzyme. Thus, it has been found that slow acetylators of these drugs are more prone to develop antinuclear antibodies (Perry et aI., 1970 ; Alarcon-Segovia et al., 1971; Price Evans et al., 1972) as well as SLE (Perry et al., 1970) upon their intake. This was originally ascribed to the resulting longer drug action since the drugs are eliminated mostly in the acetylated form. More recently it has been found that patients with spontaneous SLE are also phenotypically slow acetylators (Fishbein and Alarcon-Segovia, 1976; Reidenberg and Martin, 1974), a finding that introduces a characteristic of the lupus patient that may be closely related to the predisposition to develop lupus upon drug intake. The individual's phenotype of drug acetylation would then seem to have two roles: one of augmenting the length of action of the drug or the proportion of reactive metabolite and thereby influencing the development of antinuclear antibodies, albeit modestly; and another, perhaps more important, in the predisposition to develop lupus itself.
6. Experimental Drug-Induced Lupus
Although it was originally thought that hydrallazine administration could produce SLE in dogs (Comens, 1956), this could not be confirmed by numerous authors (reviewed by Alarcon-Segovia et al., 1967). Antinuclear antibodies and LE cells have been elicited by means of administration of drugs included in group I to rats (Monier, 1960), mice (Cannat and Seligmann, 1968; Ten Veen and Feltkamp, 1972) and rabbits (Yamauchi et aI., 1975). Differences have been found in the incidence and titres of antibodies that appear in different strains of mice. Highly inbred strains show greatest antibody production (Cannat and Seligrnann, 1968). Hydrallazine coupled to human serum
74
albumin elicit the production of anti-DNA antibodies in rabbits, while when coupled to rabbit serum albumin it does not (Yamauchi et aI., 1975). This suggests an important role of the antigenicity of the carrier protein. These studies, however, have not provided a clear cut animal model of drug-induced SLE. 7. Lupus Erythematosus and Drugs: Some Practical Considerations
Patients with systemic lupus erythematosus may be more prone to have allergic drug reactions. These, in turn may exacerbate lupus. It is therefore important that drug intake in lupus patients be kept to an essential minimum. This is more so because it is unlikely that we have detected all drugs with the pharmacological property of activating or exacerbating lupus. The question of lupus patients requiring hydrallazine for the control of hypertension, procainamide for the control of cardiac arrhythmias, isoniazid for the prevention or treatment of tuberculosis, or anticonvulsant therapy, often arises. Although these drugs are best avoided in lupus patients , if possible, preliminary data indicates that they can be used in low doses in patients who are receiving corticosteroid or immunosuppressive therapy (Reza et aI., 1975). However, caution should still be exercised in the use of these drugs in lupus patients . This is particularly true of hydrallazine, procainamide and the hydantoins. I have witnessed and reported (Alarcon-Segovia, 1972) a marked rise in titre of antinuclear antibodies and a fall in complement, coincident with exacerbation of renal disease, shortly after the onset of hydrallazine treatment to a patient who seemingly did not have SLE, but had chronic glomerulonephritis with antinuclear antibodies at a low titre. The phenomenon was attributed to an anamnestic response for the production of antinuclear antibodies triggered by hydrallazine.
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
References Alarcon-Segovia, D.: Drug-induced lupus synd romes . Mayo Clinic Proceedings 44: 664-681 (1969). Alarcon-Segovia, D.: Acute immunologic hydralazine reaction. Clinical and serological changes after hyd ralazine admin istration to a patient with circulating antinuclear antibodies. Revista de Investigacion Clinica, Mexico 24: 167-174 (1972). Alarcon-Segovia, D. and Fishbein, E.: Patterns of antinuclear antibodies and lupus-activating drugs. Journal of Rheumatology 2: 167-171 (1975). Alarcon-Segovia, D.; Fishbein, E. and Alcala, H.: Isoniazid acetylation rate and development of antinuclear antibodies upon isoniazid treatment. Arthritis and Rheumatism 14: 748-752 (1971). Alarcon-Segovia, D.; Fishbein, E. and Betancourt, V.M.: Antibodies to nucleoprotein and to hydrazide-altered soluble nucleoprotein in tuberculous patients receiving isoniazid . Clinical and Experimental Immunology 5: 429-437 (1969) . Alarcon-Segovia, D.; Fishbein, E.; Cetina , J.A.; Raya , R.J. and Barrera, E.: Antigen ic specificity of chlorpromazine-indu ced antinuclear antibodies. Clinical and Experimental Immunology 15: 543 -548 (1973) . Alarcon-Segovia, D.; Fishb ein, E.; Reyes , P.A.; Dies, H. and Schwadsky , S.: Antinuclear antibodies in patients on ant iconvulsant therapy. Clinical and Experimental Immunology 12: 39-47 (1972) . Alarcon-Segovia, D.; Wakim, K.G.; Worthington, J.W. and Ward, L.E.: Clinical and experimental studies on the hydralazine syndrome and its relationship to systemic lupus erythematosus. Medicine (Baltimore) 46 : 1-33 (1967). Alarcon-Segovia, D.; Worthington, J .W.; Ward, L.E. and Wakim,K.G.: Lupus diathesis and the hydralazine syndrome . New England Journal of Medicine 272: 462466 (1965) . Bendersky, G. and Ramirez, C.: Hydralazine poisoning. Review of the literature and autopsy study of person with massive intestinal bleeding. Journal of the American Medical Association 173: 1789-1794 (1960) . Blomgren, S.E.; Condemi, J.1. and Vaughan, J.H.: Procainamide induced lupus erythematosus. Clinical and laboratory observations. American Journal of Medicine 52: 338-348 (1972). Byrd, R.B. and Schanzer, B.: Pulmonary sequelae in procaine amide induced lupus-like syndrome. Diseases of the Chest 55 : 170-172 (1969) . Cannat, A. and Seligmann, M.: Induction by isoniazid and hydralazine of antinuclear factors in mice. Clinical and Experimental Immunology 3: 99-105 (1968).
75
Carstairs, K.C.; Breckenridge, A.; Dollerey , C.T. and Worlledge, S.M.: Incidence of a positive direct Coombs test in patients on a-Methyldopa. Lancet 2: 133-135 (1966). Cetina, J .A.; Fishbein , E. and Alarcon-Segovia, D.: Antinuclear antibodies and propylthiouracil therapy. Journal of the American Medical Association 220:1012 (1972). Comens, P.: Experimental hydralazine disease and its similarity to disseminated lupus erythematosus. Journal of Laboratory and Clinical Medicine 47: 444 -454 (1956). Dammin , GJ.; Nora, J .R. and Reardan, J .B.: Hydralazine reaction : case with LE cells antemortem and postmortem study and pulmonary, renal , splenic and muscular lesions of disseminated lupus erythematosus. Journal of Laboratory and Clinical Medicine 46: 806 (1955). Delepierre, F.; Rulliere, R. and Coury , C.H.: Syndromes lupiques et apparentes: complication reputee rare des tra itements prologes par l'isoniazide et autres medicaments antituberculeux (a propos de deux nouvelles observations). Revue de Tuberculose et de Pneumologie 35: 397-412 (1971). Domz, C.A.; McNamara, D.H. and Holzpfel, H.F.: Tetracycline provocation in lupus erythematosus. Annals of Internal Medicine 50: 1217-1226 (1959) . Dubois, E.L. : The clinical picture of systemic lupus erythematosus chapter 9, supplement 1; in Dupois (Ed) Lupus Erythematosus. A Review of the Current Status of Discoid and Systemic Lupus Erythematosus and their Variants, p. 38-438 (University of Southern California Press, Los Angeles 1974). Dubois, E.L.; Molina, J.; Bilitch, M. and Friou, G.J.: Procainamide-induced serologic changes in asymptomatic patients. Arthritis and Rheumatism 11: 477-478 (1968). Dubois, E.L. and Tuffanelli, D.L.: Clinical manifestations of systemic lupus erythematosus: Computer analysis of 520 cases. Journal of the American Medical Association 190: 104-111 (1964). Dustan, H.P.; Taylor, R.D.; Corcoran, A.C. and Page, I.H.: Rheumatic and febrile syndrome during prolonged hydralazine treatment. Journal of the American Medical Association 154 : 23-29 (1954) . Eldredge, N.T.; Robertson, W. Van B. and Miller, J.J. III: The interaction of lupus-inducing drugs with deoxyribonucleic acid.Clinicallmmunology and Immunopathology 3: 263-271 (1974) . Fakhro, A.M.; Ritchie, R.F . and Lown, B.: Lupus-like syndrome induced by procainamide. American Journal of Cardiology 20: 367-373 (1967). Feltkamp, T.E.W.; Dorhout Mees, E.J. and Nieuwenhuis,
Drug-Induced Antinuclear Antibod ies and Lupus Syndromes
M.G.: Auto antibodies related to treatment with chlorthalidone and a-methyldopa. Acta Medica Scandinavica 187: 219-223 (1970). Fishbein, E. and Alarcon-Segovia, D.: Actividad de acetil transferasa hepatica. Un marcador genetico del lupus eritematoso generalizado. Memorias VII Congreso Mexicano de Reumatologfa, Mexico City, Mexico, p.32 (1976) . Grob, P.J.; Mtiller-Schoop, J.W.; Hacki, M.A. and JollerJemelka , H.I.: Drug-induced pseudolupus. Lancet 2: 144-148 (1975) . Hahn, B.H.; Sharp, G.C.; Irvin, W.S.; Kantor , O.S.; Gardner, C.A.; Bagby, M.K.; Perry, H.M. and Osterland, C.K.: Immune responses to hydralazine and nuclear antigens in hydralazine-induced lupus erythematosus. American Journal of Medicine 76: 365-374 (1972). Hildreth, E.A.; Biro, C.E. and McCreary, T.A. : Persistence of the 'hydralazine syndrome' : a follow-up study of eleven cases. Journal of the American Medical Association 173: 657-660 (1960) . Holley, H.L.: Evidence for a predisposition to rheumat ic diseases in families of patients developing druginduced systemic lupus erythematosus. Arthritis and Rheumatism 7: 684-686 (1964). Kahn, G. and Davis, B.P.: In vitro studies on longwave ultraviolet light-dependent reactions of the skin photosensitizer chlorpromazine with nucleic acids, purines and pyrimidines. Journal of Investigative Dermatology 55: 47-52 (1970). Lappat, E.J. and Cawein, M.J.: A familial study of procainamide-induced systemic lupus erythematosus: a question of pharmacogenetic polymorphism. American Journal of Medicine 45 : 846-852 (1968). Lee, S.L.; Rivero, I. and Siegel, M.: Activation of systemic lupus erythematosus by drugs. Archives of Internal Medicine 117: 620-626 (1966) . Meislin, A.G. and Rothfield, N.: Systemic lupus erythematosus in childhood: analysis of 42 cases with comparative data in 200 adult cases followed concurrently. Pediatrics 42: 37-49 (1968). Monier, J.C.: Collagenose posthydralazinique chez le rat. Algerie Medicale 64: 421-426 (1960) . Muller, J.C.; Rast, C.L. Jr .; Pryor, W.W. and Orgain, E.S.: Late systemic complication of hydralazine (Apresoline) therapy. Journal of the American Medical Association 157: 894-899 (1955). Notwan, D.D.; Kurata, N. and Tan, E.M.: Profiles of antinuclear antibodies in systemic rheumatic diseases. Annals of Internal Medicine 83: 464-469 (1975). Novack, M.A. and Paine, R.: Procainamide-induced lupus syndrome in siblings. Journal of the American Medical Association 232: 1269 (1975).
76
Ohnishi, S. and McConnel, H.M.: Interaction of the radical ion of chlorpromazine with deoxyribonucleic acid. Journal of the American Chemical Society 87: 2293-2296 (1965). Paine, R.: Procainamide hydrochloride and lupus erythematosus. Journal of the American Medical Association 194: 23-26 (1965). Perry, H.M. Jr. and Schroeder, H.A.: Syndrome simulating collagen disease caused by hydralazine (Apresoline). Journal of the American Medical Association 154 : 670-673 (1954). Perry, H.M. Jr. ; Tan, E.M.; Carmody, S. and Sakamoto , A.: Relationship of acetil transferase activity to antinuclear antibodies and toxic symptoms in hypertensive pat ients treated with hydrallazine, Journal of Laboratory and Clinical Medicine 76 : 114-125 (1970). Price Evans, D.A.; Bullen, M.F.; Houston, J.; Hopkins, C.A. and Vetters, J.M.: Antinuclear factors in rapid and slow acetylators patients treated with isoniazid. Journal of Medical Genetics 9: 53-56 (1972). Quismorio, F.P.; Bjarnasson, D.F.; Dubois, E.L. and Friou, G.J.: Chlorpromazine-induced antinuclear antibodies. Arthritis and Rheumatism 15: 451 (1972) . Rallison, M.L.; O'Brien, J. and Good, R.A.: Severe reactions to long-acting sulfonamides: erythema multiforme exudativum and lupus erythematosus following administration of sulfamethoxypyridazine and sulfadimethoxine. Pediatrics 28: 908-917 (1961) . Reidenberg, M.M. and Martin, J.H.: The acetylator phenotype of patients with systemic lupus erythematosus. Drug Metabolism and Disposition 2: 71-73 (1974) . Reza, M.D.; Dornfeld, L. and Goldberg, L.S.: Hydralazine therapy in hypertensive patients with idiopathic systemic lupus erythematosus. Arthritis and Rheumatism 18: 335-338 (1975). Rupe, C.E. and Nickel, S.N.: New clinical concept of systemic lupus erythematosus: analysis of 100 cases. Journal of the American Medical Association 171 : 1055-1061 (1959). Sharp, G.C.; Irvin, W.S.; Tan, E.M. and Holman, H.R.: Mixedconnective tissue disease - An apparently distinct rheumatic disease syndrome associated with a specific antibody to an extractable nuclear antigen (ENA). American Journal of Medicine 52: 148-159 (1972) . Shulman, L.E.: Inducing agents and relationship to other diseases. Arthritis and Rheumatism 6 (Suppl.): 558-571 (1963). Tan, E.M.: Drug-induced autoimmune disease. Federation Proceedings 32: 1894-1897 (1974) . TenVeen, J.H. and Feltkamp, T.E.W.: Studies on druginduced lupus erythematosus in mice. I. Drug induced
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
antinuclear antibodies (ANA). Clinical and Experimentallmmunology 11: 265-276 (1972). Winfeld, J.B. and Davis, J.S.IV.: Anti-DNA antibody in procainamide-induced lupus erythematosus. Determinations using DNA fractioned by methylated albumin -Kieselguhr chromatography. Arthritis and Rheumatism 17: 97-110 (1974). Winfield, J.B.; Koffler, D. and Kunkel , H.G.: Development of antibodies to ribonucleoprotein following short-term therapy with procainamide. Arthritis and Rheuma tism 18: 531-534 (1975). Worlledge, S.M.; Carstairs, K.C. and Dacie, LV .: Autoimmune haemolytic anaemia associated with ornethyldopa therapy. Lancet 2: 135-139 (1966) . Yamauchi, Y.; Litwin , A.; Adams, L.; Zimmer , H. and Hess, E.V.: Induction of antibodies to nuclear antigens in rabbits by immunization with hydralazine-human
77
serum albumin conjugates. Journal of Clinical Investigation 56: 958-969 (1975). Zingale, S.B.; Sanchez Avalos, J.C.; Andrada, J.A.; String , S.G. and Manni, J.A. : Appearance of anticoagulant factors and certain 'autoimmune' antibodies following antigenic stimulation with blood groups substances in patients with systemic lupus erythematosus. Arthritis and Rheumatism 6: 581-598 (1963).
Author's address : Prof. Donato Alarcon-Segovia, Department of Immunology and Rheumatology, Instituto Nacional de la Nutricion, Mexico 22, DF (Mexico).
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
D. Alsrcon-Seqovie Department of Immunology and Rheumatology , Instituto Nacional de la Nutrlcien. Mexico
Summary
Drugscapableoftriggering the onset ofsystemic lupus erythematosus may be divided into those that do so by pharmacological properties of their own and those that do so by eliciting allergicreactions which bring about lupus. Drugs in the first group vary in their potency to activate lupus. They all elicit antinuclear antibodies in the majority of patients who receive them but they only cause lupus in a small percentage of patients. This dichotomy suggests that a predisposition is required for the development of lupus upon intake of these drugs. The mechanism whereby these drugs elicit antinuclear antibodies seems to relate to coupling to and/or modification of, nuclear antigens. The patterns ofantinuclear antibodies elicited by these drugs in individuals who receive them correlate well with their known reactivity with variousnuclear antigens.
Development of overt systemic lupus erythematosus (SLE) seems to result from an interplay of genetic and environmental factors (AlarconSegovia, 1969). Thus, patients with SLE, rather than having inherited the disease seem to have inherited a disturbance, perhaps immunological, which upon contact with environmental factors may lead to the disease. Prominent among the environmental agents that may trigger the onset of SLE are ultraviolet radiation and some drugs. It is common practice to make the distinction between spontaneously occurring and drug-induced SLE. Although it seems that spontaneously occurring SLE pre-
dominates 10 to lover the drug-induced disease (Lee et al., 1966; Dubois, 1974) unaccounted drug consumption is so great that this ratio may actually be lower.
1. Classification of Lupus-Activating Drugs
Drugs capable of triggering the onset of SLE may be classified into two groups (AlarconSegovia, 1969). The first group includes those drugs which require prolonged ingestion in order to elicit the
70
Drug- Induced Antinuclear Antibodies and Lupus Syndromes
clinical manifestations of lupus. Other characteristics of the relationship of drugs in this group and SLE are: 1) That no allergic reaction to the drug seems to mediate the onset of lupus 2) That larger numbers of cases have been recorded with drugs in this group, and 3) That these drugs elicit antinuclear antibodies in a large proportion of the individuals who receive them for some time (Alarcon-Segovia and Fishbein, 1975). Indeed, development of antinuclear antibodies while on these drugs is considerably more common than actual development of SLE.
Table I. Drugs reported to induce or activate SLE Group I (by pharmacological action)
Group II (by allergic reaction)
Hydrallazine Procainamide Anticonvulsants mephenytoin phenytoin pr imidone trimethadione ethosuximide carbamazepine pheneturide (phenylethylacety Iurea) Isoniazid Chlorpromazine
Aminosalycilic acid Chlorthalidone' D-penicillamine Griseofulvin Guanoxan Isoquinazepon Levodopa Methyldopa' Methysergide Methylthiouracil Oral contraceptives Oxyphenisatin Penicillin Phenylbutazone Practolol Propylthiouracil Quinidine Reserpine Streptomycin Sulphonamides Tetracycline Tolazamide
Drugs included in this group are listed in table I. The second group of drugs capable of triggering SLE includes those which: 1) May elicit SLE after brief periods of use 2) Often seem to do so by causingallergicreactions which may be quite dramatic (Dornz et al., 1959 ; Rallison et al., 1961) 3) Cause SLE only rarely, and 4) Do not often elicit antinuclear antibodies in patients receiving them (Cetina et al., 1972). Drugs in this second group are also listed in table I. It should be noted that two of these drugs, and perhaps some others, may actually belong to the flrst group. Methyldopa has been reported to cause lupus in a few instances, and to elicit antinuclear antibodies in a larger percentage of patients receiving the drug (Feltkamp et al., 1970) but, has also been found to cause, and this seems to be more frequent , positive Coombs tests (Carstairs et al., 1966), as well as 'full blown' haemolytic anaemia (Worllege et al., 1966). Chlorthalidone has been reported to cause antinuclear antibodies in patients receiving it (Feltkamp et al., 1970) but this has neither been confirmed nor have lupus patients placed on this drug
1
May belong to group I
been noticed to have exacerbations of the disease (Alarcon-Segovia, personal observations). Overt lupus caused by this drug has not been well documented . Another variant of this problem has recently been described: patients on a combination of phenopyrazone, horse-chestnut extract and cardiac glycosides ('Venocuran') for alleged treatment of venous insufficiency have been found to develop a lupus-like illness which is accompanied by mitochondrial antibodies but not by antinuclear antibodies (Grob et al., 1975). 90% of long-term users of 'Venocuran' develop mitochondrial antibodies while only 10% of the whole group develop the full blown syndrome (Grob et al., 1975).
Drug-Induced Antinuclear Antibod ies and Lupus Syndromes
2. The Lupus-Activating Potency of Various Drugs
71
age group. Also to be considered, is the possibility that individuals who develop lupus late in life and only after being exposed to a potent lupus-activatThe relative lupus-activating potency of the ing agent, may have a lesser predisposition to the various drugs in group I can only be inferred from development of lupus. This in turn may lead to a the incidence of overt lupus in patients who milder disease, more likely to remit upon withreceive them. Hydrallazine, which was the first drawal of the offending drug. drug noted to have this side-effect, was found to Another consideration pertinent to anticonvulscause lupus in 10 to 15% of patients who receive it ants , and perhaps chlorpromazine, is that convullong-term for the treatment of hypertension sions or psychosis may be early manifestations of (Dustan et aI., 1954; Perry and Schroeder, 1954; lupus. In such case the drugs would be exacerMuller et aI., 1955). bating rather than eliciting the disease. HyperProcainamide seems to be even more potent for tension or cardiac arrhythmias, as indications for it may cause a lupus-like illness in as many as 30% hydrallazine or procainamide, respectively, are less of patients receiving it (Fakhro et aI., 1967). likely to be early symptoms of SLE. Anticonvulsants, isoniazid and chlorpromazine These facts considered, hydrallazine-activated seem to follow hydrallazine in order of their lupus has been found to be indistinguishable, both lupus-activating potency. clinically (Alarcon-Segovia et al., 1967) and seroThe lupus-activating potential of drugs in group logically from spontaneous SLE (Hahn et aI., II probably resides in their antigenicity and/or 1972). Although a paucity of renal lesions in their capacity of behaving as haptens, l.e. coupling hydrallazine SLE has been mentioned as an arguto proteins or cell membranes. In turn , allergic ment against the indentity of hydrallazine-induced reactions may be more frequent in lupus or lupus- and spontaneously occurring SLE, it may only reprone patients. That immunological challenge may flect the lesser incidence of renal lesions in the activate SLE was shown by Zingale and his elderly lupus patient. Nevertheless, renal lesions co-workers (1963). have been found both clinically and histologically in a few patients with hydrallazine lupus (Dammin et al., 1955; Bendersky and Ramirez, 1960; Alarcon-Segovia et aI., 1967). 3. Clinical Picture ofDrug-Induced Lupus Procainamide-induced lupus-like disease is less In general, the clinical picture of drug-induced similar to spontaneous lupus than its hydrallazine SLE differs little from that of the spontaneously counterpart. Pleural and pleuropulmonary involveoccurring disease. Because hydrallazine, an anti- ment are considerably more frequent in the prohypertensive agent and procainamide, an anti- cainamide disease than in spontaneous lupus arrhythmic drug are usually given to elderly pat- (Blomgren et al., 1972). Indeed, pleurisy, which ients, the mean age of patients with drug-induced was the initial symptom in 30% of 54 cases of lupus is considerably higher than that of the spon- procainamide-induced disease (Alarcon-Segovia, taneously occurring disease. Since spontaneously 1969) occurred initially in only 2.5% of the lupus occurring lupus tends to be less florid and severe in patients in the series of Dubois and Tuffanelli the older adult than in the young (Rupe and (1964) which were mostly of the spontaneously Nickel, 1959) and in adults less than in children occurring form. Serologically there are also impor(Meislin and Rothfield , 1968), the lesser severity tant differences between spontaneous SLE and the of drug-induced lupus correlates well with that of procainamide-induced syndrome. Anti-native DNA the spontaneously occurring disease in the same antibodies do not occur in the procainamide dis-
Drug-Induced Ant inuclear Ant ibodies and Lupus Syndromes
72
duals who receive lupus-activating drugs belonging to group I suggests that the potential of these drugs to elicit lupus syndromes resides in their capacity to induce antinuclear antibodies. The dichotomy between the development of antinuclear antibodies in most individuals on these drugs, and the development of clinical disease in only a few may depend on factors predisposing to development of lupus. We once used the term 'lupus diathesis' for this predisposition to the development of lupus which may be revealed by environmental factors such as drugs (Alarcon-Segovia et al, 1965). I now believe that it is more likely to be a predisposition to autoimmune disease of various kinds rather than solely to SLE; a ' rnultipotential immunopathogenic predisposition' . Lupusactivating drugs would not only reveal this predisposition but , by their elicitation of antinuclear antibodies would determine that it is lupus which develops rather than other autoimmune disease. Other drugs, such as methyldopa would primarily elicit Coombs antibody and determine the development of autoimmune haemolytic anaemia rather than that of SLE. The antigenic specificity of the antinuclear antibodies elicited by the various lupus-activating drugs in subjects who have taken them for a long time without developing lupus may be important for the understanding of their mode of action (Alarcon-Segovia and Fishbein, 1975). Thus, antibodies found in isoniazid-treated subjects are primarily directed to soluble nucleoprotein (sNP), an antigen which is physicochemically altered in vitro, and probably also in vivo by isoniazid, which may also bind to it (Fishbein and Alarcon-Segovia, unpublished observations). Antibodies to isoniazid-altered sNP have been found to be present in the serum of nearly 80% of isoniazid-treated 4. Provocation of Antinuclear Antibodies by tuberculosis patients (Alarcon-Segovia et al., 1969) Drugs while only 42% had them to particulate nucleoprotein. Tuberculosis patients on isoniazid do not The fact that there is a considerably higher pre- develop antibodies to DNA, either native or devalence of antinuclear antibodies than that of natured. They may, however, have anti DNA actual lupus erythematosus syndromes in indivi- antibodies if they develop a lupus syndrome
ease (Winfield and Davis, 1974). It has recently been found that patients receiving procainamide develop antibodies to ribonucleoprotein, akin to those found in mixed connective tissue disease (Sharp et al., 1972), following short-term therapy with procainamide (Winfield et al., 1975). It is not unlikely that, contrary to what seems to be the case in hydrallazine-lupus, procainamide-induced syndromes may not be true SLE, but a drug-induced disease closely resembling it. There are no apparent significant differences between the isoniazid-induced and the spontaneously occurring lupus (Delpierre et al., 1971). The observations made on chlorpromazine-induced lupus are too scant, as yet, to evaluate. The original impression that patients who develop lupus while on hydrallazine will go into complete remission upon discontinuation of the drug proved to be false. Persistence of hydrallazine lupus after drug withdrawal has been well documented (Hildreth et al., 1960). Manifestations of lupus have been found present up to 9 years after the eliciting drug was withdrawn (Alarcon-Segovia et al., 1967) . Procainarnide-induced lupus-like syndromes appear to remit better after the drug is stopped. Patients have been recorded, however, whose disease did not remit when the drug was stopped, and who even required corticosteroid treatment (Paine, 1965; Byrd and Schanzer, 1969). Patients who develop SLE while on anticonvulsants seem to be the least likely to remit completely as a result of cessation of the drug (Alarcon-Segovia, personal observations). This may mean that convulsions actually were initial symptoms of SLE.
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
(Alarcon-Segovia et al., 1969). Hydrallazine has been found to bind similarly to, and physicochemically alter, sNP (Tan, 1974), but no data are available on the antigenic specificity of antinuclear antibodies elicited in hypertensive subjects treated with hydrallazine who have not developed a lupus syndrome. Different anticonvulsants also give rise to different patterns of antinuclear antibodies (Alarcon. Segovia et al., 1972). They elicit antibodies to sNP primarily (59%) but may give rise to antibodies to both native (14%) and denatured (22%) DNA. This is particularly true of the hydantoins which are also unique among the lupus-activating drugs in that they elicit antibodies to Sm antigen (AlarconSegovia and Fishbein, 1975). This antibody is thought to be as characteristic of SLE as antidouble stranded DNA antibody (Notman et al., 1975). The antinuclear antibodies found most frequently in patients on chlorpromazine are those directed to denatured DNA (Quismorio et al., 1972; Alarcon-Segovia et al., 1973), a fact which correlates well with the known reactivity of chlorpromazine with denatured DNA (Ohnishi and McConnell, 1965; Kahn and Davis, 1970) . Asymptomatic subjects receiving procainamide have been found to develop antibodies to ribonucleoprotein shortly after they are placed on this drug (Winfield et al., 1975) . Apparently, this antibody tends to disappear and be replaced by antibodies to single-stranded DNA which occurs in 36% of asymptomatic subjects on long-term procainamide and to nucleoprotein which are found in 68% of them (Dubois et al, 1968). It should be noted that the antigenic specificity of the antinuclear antibodies found in asymptomatic subjects on each of the various lupus-activating drugs may be different from that of the antinuclear antibodies found in patients who actually develop lupus while on them. Patients with hydrallazine-activated lupus are known to have antibodies to native DNA (Hahn et al., 1972) which correlates well with the finding
73
that immunisation of rabbits with hydrallazinehuman serum albumin conjugates, results in their development of both anti-native and anti-single. stranded DNA antibodies (Yamauchi et al., 1975) . Patients with procainamide induced lupus-like disease develop antibodies to denatured, but not to native DNA (Winfield and Davis, 1974) and they tend to stop producing antibodies to ribonucleoprotein, an antigen which may be released after myocardial infarction and could interact with procainamide (Winfield et al., 1975) . This interaction would be similar to that which occurs between this drug and DNA, which may explain the development of antibodies to DNA in patients on it (Elredge et al., 1974).
5. Predisposition of Systemic Lupus Erythematosus and Lupus-Activating Drugs: A Partnership The fact that only 10% of patients on highdose, long-term hydrallazine administration develop lupus, suggests that they have a predisposition for the development of lupus that the other 90% do not have (Alarcon-Segovia et al., 1965) . Further evidence of such predisposition can be found in the family and past histories of patients with drug-induced lupus syndromes (Shulman, 1963; Holley, 1964; Alarcon-Segovia et al., 1965; Lappat and Cawein, 1968; Novack and Paine, 1975). That such predisposition may differ quantitatively from one patient to another is suggested by the findings that some patients require higher doses of hydrallazine to develop lupus than others, and that those who develop it after having received lesser amounts of drug, have more premonitory symptoms and more tendency to persist after withdrawal of the drug than those who developed it after larger doses. One important factor influencing the development of antinuclear antibodies, as well as that of the drug-induced lupus syndromes, may reside in
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
the phenotypically determined acetylation rate of certain of these drugs, particularly hydrallazine and isoniazid which are acetylated by the same enzyme. Thus, it has been found that slow acetylators of these drugs are more prone to develop antinuclear antibodies (Perry et aI., 1970 ; Alarcon-Segovia et al., 1971; Price Evans et al., 1972) as well as SLE (Perry et al., 1970) upon their intake. This was originally ascribed to the resulting longer drug action since the drugs are eliminated mostly in the acetylated form. More recently it has been found that patients with spontaneous SLE are also phenotypically slow acetylators (Fishbein and Alarcon-Segovia, 1976; Reidenberg and Martin, 1974), a finding that introduces a characteristic of the lupus patient that may be closely related to the predisposition to develop lupus upon drug intake. The individual's phenotype of drug acetylation would then seem to have two roles: one of augmenting the length of action of the drug or the proportion of reactive metabolite and thereby influencing the development of antinuclear antibodies, albeit modestly; and another, perhaps more important, in the predisposition to develop lupus itself.
6. Experimental Drug-Induced Lupus
Although it was originally thought that hydrallazine administration could produce SLE in dogs (Comens, 1956), this could not be confirmed by numerous authors (reviewed by Alarcon-Segovia et al., 1967). Antinuclear antibodies and LE cells have been elicited by means of administration of drugs included in group I to rats (Monier, 1960), mice (Cannat and Seligmann, 1968; Ten Veen and Feltkamp, 1972) and rabbits (Yamauchi et aI., 1975). Differences have been found in the incidence and titres of antibodies that appear in different strains of mice. Highly inbred strains show greatest antibody production (Cannat and Seligrnann, 1968). Hydrallazine coupled to human serum
74
albumin elicit the production of anti-DNA antibodies in rabbits, while when coupled to rabbit serum albumin it does not (Yamauchi et aI., 1975). This suggests an important role of the antigenicity of the carrier protein. These studies, however, have not provided a clear cut animal model of drug-induced SLE. 7. Lupus Erythematosus and Drugs: Some Practical Considerations
Patients with systemic lupus erythematosus may be more prone to have allergic drug reactions. These, in turn may exacerbate lupus. It is therefore important that drug intake in lupus patients be kept to an essential minimum. This is more so because it is unlikely that we have detected all drugs with the pharmacological property of activating or exacerbating lupus. The question of lupus patients requiring hydrallazine for the control of hypertension, procainamide for the control of cardiac arrhythmias, isoniazid for the prevention or treatment of tuberculosis, or anticonvulsant therapy, often arises. Although these drugs are best avoided in lupus patients , if possible, preliminary data indicates that they can be used in low doses in patients who are receiving corticosteroid or immunosuppressive therapy (Reza et aI., 1975). However, caution should still be exercised in the use of these drugs in lupus patients . This is particularly true of hydrallazine, procainamide and the hydantoins. I have witnessed and reported (Alarcon-Segovia, 1972) a marked rise in titre of antinuclear antibodies and a fall in complement, coincident with exacerbation of renal disease, shortly after the onset of hydrallazine treatment to a patient who seemingly did not have SLE, but had chronic glomerulonephritis with antinuclear antibodies at a low titre. The phenomenon was attributed to an anamnestic response for the production of antinuclear antibodies triggered by hydrallazine.
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
References Alarcon-Segovia, D.: Drug-induced lupus synd romes . Mayo Clinic Proceedings 44: 664-681 (1969). Alarcon-Segovia, D.: Acute immunologic hydralazine reaction. Clinical and serological changes after hyd ralazine admin istration to a patient with circulating antinuclear antibodies. Revista de Investigacion Clinica, Mexico 24: 167-174 (1972). Alarcon-Segovia, D. and Fishbein, E.: Patterns of antinuclear antibodies and lupus-activating drugs. Journal of Rheumatology 2: 167-171 (1975). Alarcon-Segovia, D.; Fishbein, E. and Alcala, H.: Isoniazid acetylation rate and development of antinuclear antibodies upon isoniazid treatment. Arthritis and Rheumatism 14: 748-752 (1971). Alarcon-Segovia, D.; Fishbein, E. and Betancourt, V.M.: Antibodies to nucleoprotein and to hydrazide-altered soluble nucleoprotein in tuberculous patients receiving isoniazid . Clinical and Experimental Immunology 5: 429-437 (1969) . Alarcon-Segovia, D.; Fishbein, E.; Cetina , J.A.; Raya , R.J. and Barrera, E.: Antigen ic specificity of chlorpromazine-indu ced antinuclear antibodies. Clinical and Experimental Immunology 15: 543 -548 (1973) . Alarcon-Segovia, D.; Fishb ein, E.; Reyes , P.A.; Dies, H. and Schwadsky , S.: Antinuclear antibodies in patients on ant iconvulsant therapy. Clinical and Experimental Immunology 12: 39-47 (1972) . Alarcon-Segovia, D.; Wakim, K.G.; Worthington, J.W. and Ward, L.E.: Clinical and experimental studies on the hydralazine syndrome and its relationship to systemic lupus erythematosus. Medicine (Baltimore) 46 : 1-33 (1967). Alarcon-Segovia, D.; Worthington, J .W.; Ward, L.E. and Wakim,K.G.: Lupus diathesis and the hydralazine syndrome . New England Journal of Medicine 272: 462466 (1965) . Bendersky, G. and Ramirez, C.: Hydralazine poisoning. Review of the literature and autopsy study of person with massive intestinal bleeding. Journal of the American Medical Association 173: 1789-1794 (1960) . Blomgren, S.E.; Condemi, J.1. and Vaughan, J.H.: Procainamide induced lupus erythematosus. Clinical and laboratory observations. American Journal of Medicine 52: 338-348 (1972). Byrd, R.B. and Schanzer, B.: Pulmonary sequelae in procaine amide induced lupus-like syndrome. Diseases of the Chest 55 : 170-172 (1969) . Cannat, A. and Seligmann, M.: Induction by isoniazid and hydralazine of antinuclear factors in mice. Clinical and Experimental Immunology 3: 99-105 (1968).
75
Carstairs, K.C.; Breckenridge, A.; Dollerey , C.T. and Worlledge, S.M.: Incidence of a positive direct Coombs test in patients on a-Methyldopa. Lancet 2: 133-135 (1966). Cetina, J .A.; Fishbein , E. and Alarcon-Segovia, D.: Antinuclear antibodies and propylthiouracil therapy. Journal of the American Medical Association 220:1012 (1972). Comens, P.: Experimental hydralazine disease and its similarity to disseminated lupus erythematosus. Journal of Laboratory and Clinical Medicine 47: 444 -454 (1956). Dammin , GJ.; Nora, J .R. and Reardan, J .B.: Hydralazine reaction : case with LE cells antemortem and postmortem study and pulmonary, renal , splenic and muscular lesions of disseminated lupus erythematosus. Journal of Laboratory and Clinical Medicine 46: 806 (1955). Delepierre, F.; Rulliere, R. and Coury , C.H.: Syndromes lupiques et apparentes: complication reputee rare des tra itements prologes par l'isoniazide et autres medicaments antituberculeux (a propos de deux nouvelles observations). Revue de Tuberculose et de Pneumologie 35: 397-412 (1971). Domz, C.A.; McNamara, D.H. and Holzpfel, H.F.: Tetracycline provocation in lupus erythematosus. Annals of Internal Medicine 50: 1217-1226 (1959) . Dubois, E.L. : The clinical picture of systemic lupus erythematosus chapter 9, supplement 1; in Dupois (Ed) Lupus Erythematosus. A Review of the Current Status of Discoid and Systemic Lupus Erythematosus and their Variants, p. 38-438 (University of Southern California Press, Los Angeles 1974). Dubois, E.L.; Molina, J.; Bilitch, M. and Friou, G.J.: Procainamide-induced serologic changes in asymptomatic patients. Arthritis and Rheumatism 11: 477-478 (1968). Dubois, E.L. and Tuffanelli, D.L.: Clinical manifestations of systemic lupus erythematosus: Computer analysis of 520 cases. Journal of the American Medical Association 190: 104-111 (1964). Dustan, H.P.; Taylor, R.D.; Corcoran, A.C. and Page, I.H.: Rheumatic and febrile syndrome during prolonged hydralazine treatment. Journal of the American Medical Association 154 : 23-29 (1954) . Eldredge, N.T.; Robertson, W. Van B. and Miller, J.J. III: The interaction of lupus-inducing drugs with deoxyribonucleic acid.Clinicallmmunology and Immunopathology 3: 263-271 (1974) . Fakhro, A.M.; Ritchie, R.F . and Lown, B.: Lupus-like syndrome induced by procainamide. American Journal of Cardiology 20: 367-373 (1967). Feltkamp, T.E.W.; Dorhout Mees, E.J. and Nieuwenhuis,
Drug-Induced Antinuclear Antibod ies and Lupus Syndromes
M.G.: Auto antibodies related to treatment with chlorthalidone and a-methyldopa. Acta Medica Scandinavica 187: 219-223 (1970). Fishbein, E. and Alarcon-Segovia, D.: Actividad de acetil transferasa hepatica. Un marcador genetico del lupus eritematoso generalizado. Memorias VII Congreso Mexicano de Reumatologfa, Mexico City, Mexico, p.32 (1976) . Grob, P.J.; Mtiller-Schoop, J.W.; Hacki, M.A. and JollerJemelka , H.I.: Drug-induced pseudolupus. Lancet 2: 144-148 (1975) . Hahn, B.H.; Sharp, G.C.; Irvin, W.S.; Kantor , O.S.; Gardner, C.A.; Bagby, M.K.; Perry, H.M. and Osterland, C.K.: Immune responses to hydralazine and nuclear antigens in hydralazine-induced lupus erythematosus. American Journal of Medicine 76: 365-374 (1972). Hildreth, E.A.; Biro, C.E. and McCreary, T.A. : Persistence of the 'hydralazine syndrome' : a follow-up study of eleven cases. Journal of the American Medical Association 173: 657-660 (1960) . Holley, H.L.: Evidence for a predisposition to rheumat ic diseases in families of patients developing druginduced systemic lupus erythematosus. Arthritis and Rheumatism 7: 684-686 (1964). Kahn, G. and Davis, B.P.: In vitro studies on longwave ultraviolet light-dependent reactions of the skin photosensitizer chlorpromazine with nucleic acids, purines and pyrimidines. Journal of Investigative Dermatology 55: 47-52 (1970). Lappat, E.J. and Cawein, M.J.: A familial study of procainamide-induced systemic lupus erythematosus: a question of pharmacogenetic polymorphism. American Journal of Medicine 45 : 846-852 (1968). Lee, S.L.; Rivero, I. and Siegel, M.: Activation of systemic lupus erythematosus by drugs. Archives of Internal Medicine 117: 620-626 (1966) . Meislin, A.G. and Rothfield, N.: Systemic lupus erythematosus in childhood: analysis of 42 cases with comparative data in 200 adult cases followed concurrently. Pediatrics 42: 37-49 (1968). Monier, J.C.: Collagenose posthydralazinique chez le rat. Algerie Medicale 64: 421-426 (1960) . Muller, J.C.; Rast, C.L. Jr .; Pryor, W.W. and Orgain, E.S.: Late systemic complication of hydralazine (Apresoline) therapy. Journal of the American Medical Association 157: 894-899 (1955). Notwan, D.D.; Kurata, N. and Tan, E.M.: Profiles of antinuclear antibodies in systemic rheumatic diseases. Annals of Internal Medicine 83: 464-469 (1975). Novack, M.A. and Paine, R.: Procainamide-induced lupus syndrome in siblings. Journal of the American Medical Association 232: 1269 (1975).
76
Ohnishi, S. and McConnel, H.M.: Interaction of the radical ion of chlorpromazine with deoxyribonucleic acid. Journal of the American Chemical Society 87: 2293-2296 (1965). Paine, R.: Procainamide hydrochloride and lupus erythematosus. Journal of the American Medical Association 194: 23-26 (1965). Perry, H.M. Jr. and Schroeder, H.A.: Syndrome simulating collagen disease caused by hydralazine (Apresoline). Journal of the American Medical Association 154 : 670-673 (1954). Perry, H.M. Jr. ; Tan, E.M.; Carmody, S. and Sakamoto , A.: Relationship of acetil transferase activity to antinuclear antibodies and toxic symptoms in hypertensive pat ients treated with hydrallazine, Journal of Laboratory and Clinical Medicine 76 : 114-125 (1970). Price Evans, D.A.; Bullen, M.F.; Houston, J.; Hopkins, C.A. and Vetters, J.M.: Antinuclear factors in rapid and slow acetylators patients treated with isoniazid. Journal of Medical Genetics 9: 53-56 (1972). Quismorio, F.P.; Bjarnasson, D.F.; Dubois, E.L. and Friou, G.J.: Chlorpromazine-induced antinuclear antibodies. Arthritis and Rheumatism 15: 451 (1972) . Rallison, M.L.; O'Brien, J. and Good, R.A.: Severe reactions to long-acting sulfonamides: erythema multiforme exudativum and lupus erythematosus following administration of sulfamethoxypyridazine and sulfadimethoxine. Pediatrics 28: 908-917 (1961) . Reidenberg, M.M. and Martin, J.H.: The acetylator phenotype of patients with systemic lupus erythematosus. Drug Metabolism and Disposition 2: 71-73 (1974) . Reza, M.D.; Dornfeld, L. and Goldberg, L.S.: Hydralazine therapy in hypertensive patients with idiopathic systemic lupus erythematosus. Arthritis and Rheumatism 18: 335-338 (1975). Rupe, C.E. and Nickel, S.N.: New clinical concept of systemic lupus erythematosus: analysis of 100 cases. Journal of the American Medical Association 171 : 1055-1061 (1959). Sharp, G.C.; Irvin, W.S.; Tan, E.M. and Holman, H.R.: Mixedconnective tissue disease - An apparently distinct rheumatic disease syndrome associated with a specific antibody to an extractable nuclear antigen (ENA). American Journal of Medicine 52: 148-159 (1972) . Shulman, L.E.: Inducing agents and relationship to other diseases. Arthritis and Rheumatism 6 (Suppl.): 558-571 (1963). Tan, E.M.: Drug-induced autoimmune disease. Federation Proceedings 32: 1894-1897 (1974) . TenVeen, J.H. and Feltkamp, T.E.W.: Studies on druginduced lupus erythematosus in mice. I. Drug induced
Drug-Induced Antinuclear Antibodies and Lupus Syndromes
antinuclear antibodies (ANA). Clinical and Experimentallmmunology 11: 265-276 (1972). Winfeld, J.B. and Davis, J.S.IV.: Anti-DNA antibody in procainamide-induced lupus erythematosus. Determinations using DNA fractioned by methylated albumin -Kieselguhr chromatography. Arthritis and Rheumatism 17: 97-110 (1974). Winfield, J.B.; Koffler, D. and Kunkel , H.G.: Development of antibodies to ribonucleoprotein following short-term therapy with procainamide. Arthritis and Rheuma tism 18: 531-534 (1975). Worlledge, S.M.; Carstairs, K.C. and Dacie, LV .: Autoimmune haemolytic anaemia associated with ornethyldopa therapy. Lancet 2: 135-139 (1966) . Yamauchi, Y.; Litwin , A.; Adams, L.; Zimmer , H. and Hess, E.V.: Induction of antibodies to nuclear antigens in rabbits by immunization with hydralazine-human
77
serum albumin conjugates. Journal of Clinical Investigation 56: 958-969 (1975). Zingale, S.B.; Sanchez Avalos, J.C.; Andrada, J.A.; String , S.G. and Manni, J.A. : Appearance of anticoagulant factors and certain 'autoimmune' antibodies following antigenic stimulation with blood groups substances in patients with systemic lupus erythematosus. Arthritis and Rheumatism 6: 581-598 (1963).
Author's address : Prof. Donato Alarcon-Segovia, Department of Immunology and Rheumatology, Instituto Nacional de la Nutricion, Mexico 22, DF (Mexico).
