THE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 6, NO.6, 1993

Selected Summaries271

Clinical response and autoantibodies inidiopathic inflammatory myopathy

Joffe MM, Love LA, Leff RL, Fraser DD, Targoff IN,Hicks JE, Plotz PH, Miller FW. (National Institute ofArthritis, Musculoskeletal and Skin Diseases, Departmentof Rehabilitation Medicine, National Institutes of Health,Bethesda, Maryland, USA and Department of Medicine,University of Oklahoma Health Sciences Center, Oklahoma,USA.) Drug therapy of the idiopathic inflammatorymyopathies: Predictors of response to prednisolone,azathioprine and methotrexate and a comparison of theirefficacy. Am J Med 1993;94':379-87.

SUMMARYThis retrospective cohort study involved 1I3 patients withidiopathic inflammatory myopathy (11M) and evaluated thefactors associated with the response to treatment with pred-nisolone, methotrexate and azathioprine. The data wereobtained from medical records of all patients attending theWarren Grant Magnuson Clinical Centre, National Institutes ofHealth, Bethesda, Maryland, USA, between June 1983 andDecember 1988.The patients included in the study were thosefor whom adequate data were available on therapeutic responseand whose first therapeutic agent was prednisolone alone. Allpatients initially received a trial of prednisolone, and then somereceived further trials of prednisolone, azathioprine ormethotrexate depending on the response to the first trial. Thepatients were categorized on the basis of clinical featuresinto five groups: (i) primary polymyositis (PM), (ii) primarydermatomyositis (DM), (iii) cancer associated myositis (CAM),(iv) connective tissue disease overlap with myositis (CfM) ,and (v) inclusion body myositis (IBM). On the basis of certainmyositis-specificautoantibodies (MSA), the patients were alsodivided into four groups: (i) those with autoantibodies toaminoacyl-tRNA synthetases, (ii) those with autoantibodies tosignal recognition particle proteins (SRP), (iii) those withautoantibodies to a nuclear protein complex of unknownfunction (anti Mi-2), and (iv) those without any of the aboveautoantibodies (MSA negative). The time interval between theonset of disease and diagnosis was recorded. The results showedthat clinical groups, autoantibody profile and time intervalbetween onset to diagnosis influenced the clinical response tothese drugs. The response to prednisolone was best in CAM andCTMfollowed by DM and PM. Patients with IBM had the worstresponse and those with autoantibodies to aminoacyl-tRNAsynthetases or to SRP responded partially to prednisolonewhereas those with antibodies to Mi-2 had complete responses.Patients who presented after a long interval since onset hadpoor responses compared to those with only a short delay(less than 3 months). Those showing good response tothe first trial of prednisolone responded well to an additionaltrial of prednisolone or a trial of azathioprine compared tomethotrexate. Patients responding poorly to the first trial ofprednisolone showed a better response to methotrexate than

another trial of prednisolone or a trial of azathioprine.Azathioprine was more effective than other drugs in OM, PM

without MSA and steroid responsive patients. Methotrexatewas more effective in steroid-resistant cases, men and PMwith autoantibodies to synthetases.

The authors suggest that determining the clinical group,autoantibody status and time from disease onset to diagnosisin patients with myositis is useful for predicting the clinicalresponse to therapy. Methotrexate may be better than eitherazathioprine or re-treatment with steroids in patients who donot respond to a course of prednisolone.

COMMENTThe IIMs are very uncommon with an annual incidenceof 5 to 10 new cases per million persons per year in the USA.'They are a heterogeneous group of systemic rheumaticdiseases characterized by chronic inflammation of theskeletal muscles and occasional involvement of the lung,heart and gastrointestinal tract. Because of the rarity ofIIMs and heterogeneity of clinical groups, there is littleunderstanding of their aetiology, genetics, epidemiology,response to therapy and prognosis.

Bohan and Peter divided 11M into five major clinicalgroups which have been found to be both therapeutically andprognostically useful. 2 Many studies have suggested thatMSA also defines useful groups with different clinical andimmunogenetic features.v' This is the first study examiningthe responses to treatment by prednisolone, azathioprineand methotrexate in various subgroups of 11Mpatients basedon autoantibody status.

The authors have pointed out the limitations of their studywhich include that theirs was a referral centre, the datawere collected retrospectively from hospital records, and theallocation of treatment was not randomized. A prospectiverandomized double blind controlled trial would have beenmore informative. However, keeping in mind the very lowincidence of 11M and difficulties in enrolling an adequatenumber of patients using strict diagnostic criteria, the presentstudy serves a useful purpose, It provides insight into thepredictive responses of various subgroups of 11Mto immuno-

. suppressive agents. Only one blind controlled therapeutictrial in 11Mhas been published, S

What useful inferences can we draw from this study? SinceIBM is uniformly resistant to all forms of therapy-prednisolone, azathioprine and methotrexate, a musclebiopsy at the onset should be done in all cases. In the presentstudy, 50% of patients showed a partial response to pred-nisolone and occasional good responses have been reportedin previous studies as well." Hence patients with IBM shouldbe given a trial of immunosuppressive agents before they arelabelled non-responsive. Patients with CAM and CTMare more likely to show a good response to prednisolone withor without azathioprine and it is, therefore, the optimaltherapy for them.

Another recent study by the same authors has reported astrong association between the autoantibody profile andHLA phenotype in patients with PMlDM. A higher incidence

272

of DRw52 was seen in patients with antibodies to theaminoacyl-tRNA synthetases and anti-SRP autoantibodiesand of DRw53 in patients with Mi-2 and MAS autoantibodies.'The DRw52 positive patients having associated auto-antibodies against tRNA synthetases and SRP are lessresponsive to prednisolone, have more severe myositis, aworse prognosis and show better response to methotrexate.Patients with DRw53 and autoantibodies against Mi-2and MAS autoantigens have less severe myositis, a lowermortality rate and show a better response to prednisolone.Such patients should be treated with azathioprine andprednisolone. Patients with no MSA behave as Mi-2 positivepatients.' Thus, in patients with PM and DM, the determina-tion of HLA and MSA may serve as useful additional toolsfor choosing appropriate immunosuppressive agents.

Finally, the time period between onset and diagnosisappears to have an important bearing on disease outcome.The longer this time gap, the poorer is the response toimmunosuppressive drugs. Hence, these patients should bereferred early to specialized centres.

Autoimmunity and gestational diabetes

McEvoy RC, Franklin B, Ginsberg-Fellner F. (Departmentsof Pediatrics and Cell Biology/Anatomy, Mount Sinai Schoolof Medicine, New York, USA.) Gestational diabetesmellitus: Evidence for autoimmunity against the pancreaticbeta cells. Diabetologia 1991;34:507-10.

SUMMARYThe authors studied the role of autoimmunity in gestationaldiabetes mellitus (GDM) in 312 women between 18 and28 weeks of gestation who attended the prenatal clinic at theMount Sinai Hospital, New York, during the period 1983-84.These women were predominantly Hispanic (65%) and Black(32%), and the majority were receiving public monetary assis-tance. All were evaluated for the presence of glucose intolerance.Of the 312 women who had an abnormal result on a 50 g glucosescreening examination (1 hour value exceeding 150 mg/dl), 254were subjected to a formal glucose tolerance test, the remainderbeing lost to follow up. GDM was detected in 144 subjects(57%); the remaining women, who had positive screening and anormal glucose tolerance test, were termed 'suspect controlsubjects' (Se). The authors examined the sera of these subjectsfor the presence of islet cell surface antibodies (ICSA) andinsulin autoantibodies (IAA) being unaware of the relevantclinical data. The presence of ICSA was determined by acomplement-mediated cytotoxic assay [using rat insulinomacells (RIN5F) as the target 1 and IAA was determined by a liquidphase displacement assay. In addition to the study and SCgroup, sera of 160 normal blood donors (control group) werealso examined by immunological tests.

THE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 6, NO.6, 1993

REFERENCESPlotz PH, Dalakas M, Leff RL, Love LA, Miller FW, Cronin ME. Currentconcepts in the idiopathic inflammatory myopathies: Polymyositis, der-matomyositis and related disorders. Ann Intern Med 1989;111:14:>--57.

2 Bohan A, Peter JB. Polymyositis and dermatomyositis (parts 1 and 2). N Eng!J Med 1975 ;292:344-7, 40:>--7.

3 Hochberg MC, Feldman D, Stevens MB, Arnett FC, Reichlin M. Antibodies toJo-I in polymyositis/dermatomyositis: Association with interstitial pulmonarydisease. J Rheumato/1984;11:66:>--5.

4 Targoff IN, Reichlin M. The association between Mi-2 antibodies and der-matomyositis. Arthritis Rheum 1985;28:796-803.

5 Bunch TW, Worthington JW, Combs JJ, I1strup DM, Engel AG. Azathioprinewith prednisolone for polymyositis. A controlled clinical trial. Ann Intern Med1980;92:365-9.

6 Cohen MR, Sulaiman AR, Garancis JC, Wortmann RL. Clinical heterogeneityand treatment response in inclusion body myositis. Arthritis Rheum 1989;32:734-40.

7 Love LA, Leff RL, Fraser DD, et al. A new approach to the classification ofidiopathic inflammatory myopathy: Myositis-specific autoantibodies defineuseful homogeneous patient groups. Medicine (Baltimore) 1991 ;70:36G-74.

DEEPAK GUPTASITA NAIK

Department of ImmunologySanjay Gandhi Postgraduate Institute of Medical Sciences

LucknowUttar Pradesh

Whereas 45 of the 144 sera (31%) of GDM subjects werepositive for ICSA, only 9 of 108 (8%) SC subjects had sero-logical evidence of autoimmunity (p<O.OOOOl).The prevalenceof ICSA was significantly greater in both the GDM and SCgroups compared to the normal blood donor control population(3 of 160,1.8% v. SC: p<0.05, v. GDM: p=O.OOOOI).

Eight sera from the GDM patients and one from a controlsubject were positive for IAA. Subsequent examination of theirmedical records revealed that all of them had had insulin therapypreviously, making it impossible to comment on the autoimmunenature of IAA. The prevalence of ICSA was significantly higherin the GDM group requiring insulin treatment (48%) than in theone managed with dietary changes alone (7%). All subjectswere followed up till delivery to observe whether they neededinsulin therapy and to assess their glycaemic status. The authorssuggest that a high proportion of pregnant women with glucoseintolerance develop an autoimmune response against the.pancreatic islets in spite of the relative immune tolerance duringpregnancy and that the presence of ICSA predicts the onset ofgestational diabetes requiring insulin therapy.

COMMENTPregnancy is a diabetogenic condition, a consequence of thealtered carbohydrate metabolism resulting from the action ofa multitude of hormones and placental products.' While thisensures a steady supply of nutrients to the foetus, 2% to 3%of pregnant women become glucose intolerant in GDM. Theunfavourable metabolic status consequent to hyper-glycaemia leads to adverse effects on maternal and foetalwell-being, necessitating early diagnosis and prompt andappropriate therapeutic intervention.

Similar to non-insulin dependent diabetes mellitus, theimpairment of glucose tolerance in GDM is widely believedto be the result of peripheral insulin resistance, pre-