Autoimmunity and Type I Diabetes

Autoimmunity and T~e I Diabetes Jean-FranqoisBach

Insulin-dependentdiabetes mellitus (IDDM) is a T-cell–mediatedautoimmune disease. The efector mechanisms essentiallyinvolve cytokine-mediated inflammationultimatelyleadingto &ell destruction. Severalcandidate autoantigens have been delineatedfor both the pathogenic T-cell response and the nonpathogenic antibody response used for diseaseprediction. Because of antigen spreading,it is not yet clear which of these antigensare involved in the triggeringof the autoimmune response.In any case, this TH1 autoimmune response is amplifiedand perpetuatedby an immune dysregulationinvolving TH2 cells. Both effecter and regulatory mechanisms areplaced under the tightcontrol of majorhistocompatibility complex (A4HC)and non-MHC genes. (TrendsEndocrinol Metab 1997; 8:71–74).O 1997, Elsevier Science Inc.

Insulin-dependent diabetes mellitus (IDDM) is characterized by the progressive and selective destruction of ~ cells in the islets of Langerhans.Evidence has accumulated over the last 2 decades indicating that the disease has an autoimmune pathogenesis (Bach 1994).The investigation of the mechanismsof human IDDM is subject to the obvious limitation that, in general, only patients with already established disease can be studied. This explains the exceptional attention given to two spontaneous animal models of IDDM, the nonobese diabetic (NOD) mouse and the bio breeding (BB) rat, which can be studied before the onset of the autoimmune response and can be submitted to genetic and immunologic manipulation. In spite of some differences, there is indeed considerable analogy between these models and the human disease at the clinical, genetic, and immunologic levels. With the use of these reference models, numerous data have been collected that argue very strongly in favor of the immunologic origin of the disease and more particularly of its T-cell depen-

Jean-Fran$ois Bachis at the INSERMU 25, Htipitd Necker,Paris,France. TEM Vol.8, No. 2, 1997

dency.Thus, (a) grafting allogeneic bone marrow to NOD mice or BB rats prevents diabetes onset (Leiter and Serreze 1991,Nakano et al. 1988),whereas, conversely,grafting NOD bone marrow and thymus to non–diabetes-prone animals induces diabetes onset (Stein et al. 1992),(b) diabetes can be transferred to prediabetic syngeneic recipients by T-cell–purified populations (Bendelac et al. 1987) or even by T-cell clones (Haskins and McDuffie 1990), and (c) the disease is induced in double transgenic mice in which one parent carries a p cell-targeted antigen and the other the corresponding T-cell antigen receptor (TCR) (Ohashi et al. 1991,Oldstoneet al. 1991). The disease can also be induced in transgenic mice expressingin most of their T cells the TCR of diabetogenic T-cell clones (Katz et al. 1993). Several questions remain open: 1. How are the @cells affected? 2. What are the relevant triggering and target &cell autoantigens? 3. What are the factors (genetic and environmental) that lead to the rupture of self-tolerance to P cells and to the emergence of the disease?

I shall try here to answer these questions using the most recent data collected in both murine models and the human disease. The nature of the effecter mechanisms that are at the origin of the ~-cell lesion is also discussed. Last, the main etiologic factors believed to trigger the (3-cell-directed pathogenic autoimmuneresponsewill be presented.

●

Insulin-DependentDiabetes Mellitus As an Autoirnmune Disease

The autoimmune origin of a disease is essentiallyproved by the demonstration that autoreactiveeffecters (either T cells or autoantibodies)mediatethe lesion that causes the clinical manifestationsof the disease. This demonstration requires, first, the identification of the candidate effecter mechanismsand the demonstration of their pathogenic responsibility. This information is not easy to collect in human IDDM, where one has essentially accessonly to circulatingblood. This explainswhy mostof the availabledatahave been derived from the study of the NOD mouse and the BB rat. Insulitis The islets of Langerhans of diabetic patients and animals are heavily infiltrated with mononuclear cells that comprise a majority of T cells and macrophages (insulitis) (Bach 1994). The sequential study of NOD mice has shown that this infiltrate appears several weeks before diabetes onset. T cells include both CD4 and CD8 cells. CD4 cells appear first, and their presence is necessary for the CD8cell migration, as shown by transfer of purified CD4 or CD8 T cells (Thivolet et al. 1991). The CD4 cells synthesize various cytokines, with a predominance of TH1-type cytokines, such as interferon (IFN) -Vand interleukin (IL)-2, as shown by immunohistologic (0’Reilly et al. 1991)and reverse transcriptasepolymerase chain reaction (RT-PCR) (Hancock et al. 1995)evaluation. Identification of Islet-Specific Autoimmune Stigmata Recently diagnosed diabetics produce autoantibodies to a wide variety of autoantigens, including glutamic acid decarboxylase (GAD) (Baekkeskov et al. 1990),IA-2 (a tyrosine phosphatase,also

PIIS1043-276O(96)OO271-8 Q 1997,ElsevierScienceInc., 1043-2760/97/$17.00

71

called ICA-512) (Bonifacio et al. 1995), insulin, and gangliosides. These autoantibodies are not considered to contribute to the &cell lesion (which is purely T cell mediated), but they represent remarkable tools for the detection of prediabetics in family members of diabetic patients. Simultaneous presence of GAD, IA-2, and insulin antibodies is associated with a close to 100°/0risk of diabetes onset within 5 years (Verge et al. 1996). Islet-specificT-cellautoimmunityis also found concomitantlywith autoantibodies. Peripheral blood T cells proliferate and producevariouscytokinesin the presence of isletcells,GAD (Atkinsonet al. 1992)or of a 37-kD microsomal antigen recently identified as a mitochondria membrane antigen (Neophytou et al. 1996). The multiplicity of these target autoantigens is frustrating, inasmuch as it does not allow identification of a primary autoantigen.As will be shown later here, in fact, there may not be necessarily a single triggering autoantigen.

tent nude or severe combined immuno- Specific Recognition of Islet Antigenic deficiency (SCID) NOD mice do not be- peptides Under Major Histocompatibility come diabetic (Christiansonet al. 1993). Complex Gene Control Lastly, treatment of NOD mice with The association of human IDDM with T-cell–specific antibodies such as antiHLA DR3 and DR4 is well established TCR, anti-CD4,or anti-CD8prevents di(Bach 1994),and a similar MHC-IDDM abetes onset or induces remission of esassociationis found in NOD mice (Wicktablished diabetes in NOD mice (Hutcher et al. 1995).NOD congenic mice with ings et al. 1992,Chatenoud et al. 1993) non-NOD MHC are protected from diaindicating that the disease depends on betes (Wicker et al. 1992). the presence of T cells and that complete destruction of P cells is preceded by a Amplification or Perpetuation of the phase of reversible T-cell inflammation Reaction by an Immune Dysregulation that is cleared by these various maneuIn the thymus and spleen of prediabetic vers. NOD mice, one may evidenceCD4immunoregulatoryT cells that prevent diabetes onset in cotransfer experiments(Boitard . Etiology of Islet-Specific et al. 1989). These suppressorcells are Autoimmunity wiped out by thymectomyperformed at 3 IDDM is a multifactorial disease due to weeksof age (Dardenneet al. 1989)andby the association of multiple genetic and administrationof cyclophosphamide(Yaenvironmental factors. As far as the ori- sunamiand Bach 1988).Although this is gin of the pathogenic autoimmune re- not yet directly proven, one may assume sponseis concerned, one can distinguish thattheseregulatoryT cellsare of the TH2 type, as suggestedby diabetesprevention three sets of contributing factors. The Nature of the Effecter T Cells by IL-4 (Rapoport et al. 1993) or IL-10 TriggeringMechanism Leading (Pennline et al. 1994)treatment,the abIn humans,IDDM maybe transfen-edto to the Abnormal Expression of (Well sence of diabetes in transgenic mice exrecipients of bone marrow cells when Immunogenic Peptides pressingIL-4 in the ~ cells (N. Sarvetnick, the donor is diabetic (Lampeter et al. 1993,Vialettes and Maraninchi 1993).In Several hypotheses may explain the in- personal communication), and the inthe NOD mouse, diabetes can be trans- creased immunogenicity of the trigger- creased presence of T cells producing ferred by the infusion of spleen cells ing peptides, including increased or ab- these cytokines in NOD mice protected from diabetic mice (CD4 and CD8 cells errant expressionof major histocompat- from diabetes after treatment with oral are necessa~) (Bendelac et al. 1987). ibility complex (MHC) classI, classII, or insulin(Hancock et al. 1995)or complete The difficulties met in attempts to dem- adhesion molecules by ~ cells (Kay et al. Freund’sadjuvant(Shehadehet al. 1994). In addition to these three orders of onstrate directly the existence of &cell1991,Bottazzo et al. 1985).The nature mechanisms, other factors may be inspecific CD8 cytolytic T lymphocytes of the stimuli leading to this abnormalvolved in modulating the disease develhave suggestedthat CD4 cells could play ity is unknown:Is it a virus? [transgenic the predominant role, the more so be- mice expressing in B cells the T antigen opment, such as genes controlling lymphocyte apoptosis (Garchon et al. 1994). cause CD4 clones alone can transfer the of the SV40 virus show insulitis (Adams disease (Haskins and McDuffie 1990); et al. 1987) and double transgenic mice however, CD8 clones can also transfer derived from parents expressing a virus the disease (Wong et al. 1996), leaving glycoprotein in 13cells and the corre- ● Restorationof Self-Tolerance to ~-CeU Antigens open the nature of the cell directly at the sponding TCR become diabetic after Torigin of the B-cell lesion. CD4 cells cell activation by the virus] (Ohashi et al. It is a major clinical challenge to restore could intervene by producing various cy- 1991,Oldstone et al. 1991).Is it a toxic the tolerance to ~-cell antigens whose tokines that induce local inflammation, agent? (low-dose streptozotocin en- loss is at the origin of IDDM. The most where NO synthesiscould play a central hances MHC class II expression by 13 straightforward approach is based on role (Kroncke et al. 1993). cells) (Campbell et al. 1988). the administration of ~-cell candidate The problem is complicated by the autoantigens in young NOD mice. This Prevention of Insulin-Dependent fact that the local inflammation pro- has been successfully achieved with DiabetesMellitus by T-Cell–Directed duced by the initial anti+cell T-cell re- GAD, injected either intravenously Immunointervention sponse may contribute to the triggering (Kaufman et al. 1993)or intrathymically Cyclosporin A, a T-cell–selective immu- of a secondary &cell-specific response at 3 weeks of age (Tisch et al. 1993).This nosuppressive agent, prevents the onset through mechanisms of antigen spread- result does not prove, however,that GAD of diabetes in NOD mice and can induce ing. This spreading is the consequence is the single triggering (and target) aucomplete remission of diabetes in re- of the in situ production of -yIFN in ~ toantigen in the NOD mouse, as unexcently diagnosed human diabetics cells and/orof the release of 13-celldeg- pectedly two other autoantigens have provided the same effect: Tolerance in(Feutren et al. 1986).Immunoincompe- radation products. 72

@ 1997,ElsevierScienceInc., 1043-2760/97/$17.00 PIIS1043-276O(96)OO271-8

TEM vol.8, No. 2, 1997

duction to insulin (Zhang et al. 1991)or to heat shock protein 60 (hsp 60) (Elias et al. 1991) also prevents the onset of diabetes. The only way to explain these data is to infer that tolerance to these autoantigens is dominant, mediated by downregulation of effecter cells. Such suppressorphenomena have been demonstrated in some of the models mentioned earlier here. Thus, tolerance induced by oral administration of insulin (Zhang et al. 1991) is associated with a shift in cytokine pattern in the islets (toward TH2-type cytokines) (Hancock et al. 1995),and tolerance (protection from diabetes) can be broken down by antitransforming growth factor (TGF)~ monoclinal antibody (Zhang et al. 1991). Similarly, diabetes protection induced by hsp60 is associated with the appearance of TH2 cell-dependent IgGl and IgG2b anti-hsp60 antibodieswith in vitro production of TH2-type cytokines in response to hsp 60. One approach to test the hypothesisthat one of the tolerogens has induced central tolerance (deletion of the corresponding T cells) is to produce transgenic mice expressing the autoantigen in the thymus. This has been achieved with hsp60 (Birk et al. 1996) and GAD (our unpublished data). Unfortunately results obtained with hsp60 indicate that the transgenic mice still show dominant tolerance rather than deletion. Final data are not yet available for GAD transgenic mice. A strategy to circumvent the usage of a still putative candidate autoantigen avoiding the risk of undesirable sensitization (disease acceleration) consists in administering anti–T-cell monoclinal antibodies. It has thus been shown that administration of an anti-CD3 monoclinal antibody to recently diagnosed diabetic NOD mice induces long-term remission of the disease (Chatenoud et al. 1993),with, again, a TH1 + TH2 shift.

References Adams TE, Alpert S, Hanahan D: 1987. Nontolerance and autoantibodies to a transgenic self antigen expressed in pancreatic 13cells. Nature 325:223-228. AtkinsonMA, Kaufman DL, CarnpbeflL, et af.: 1992. Response of peripheral-bloodmononuclear cells to glutamatedecarboxylasein insulin dependentdiabetes.Lancet339:45&459.

TEM vol.8, No. 2, 1997

Bach JF: 1994.Insulin-dependentdiabetes mellitus as an autoimmune disease. Endocr Rev 4:516-542. BaekkeskovS, Aanstoot HJ, ChristgauS, et al.: 1990. Identification of the 64K’ autoantigen in insulin-dependent diabetes as the GABAsynthesizingenzyme glutamic acid decarboxylase. Nature 347:151–156. Bendelac A, Carnaud C, Boitard C, Bach JF: 1987. Syngeneic transfer to autoimmune diabetes from NOD mice to healthy neonates: requirement for both L3T4+ and Lyt-2+ T cells. J Exp Med 166:823–832. Birk OS, Douek DC, Elias D, et al.: 1996. A role of Hsp60 in autoimmune diabetes: analysis in a transgenic model. Proc Natl Acad Sci USA 93:1032-1037. Boitard C, Yasunami R, Dardenne M, Bach JF: 1989. T cell-mediated inhibition of the transfer of autoimmune diabetes in NOD mice. J Exp Med 169:1669-1680. Bonifacio E, Lampasona y GenoveseS, Ferrari M, Bosi E: 1995. Identificationof protein Vmsine phosphatase-likeIA2 (islet cell antigen 512) as the insulin-depmdentdiabet=-related 37/40K autoantigenand a target of islet-ceff antibodies.J Imrnunol 155:5419–5426. Bottazzo GF, Dean BM, McNaUyJM, MacKay EH, Swift PG, Gamble DR: 1985.In situ chzm acterizationof autoinunune phenomena and expressionof HLAmoleculesin thepancreasin diabeticinsulitis.N Engl J Med 313:353-360. Campbell IL, Oxbrow L, Koulmanda M, Harrison LC: 1988. IFN-y induces islet cell MHC antigensand enhances autoimmune, streptozotocin-induced diabetes in the mouse. J Immunol 140:1111–1116. Chatenoud L, Thervet E, Primo J, Bach JF: 1993. Anti-CD3 antibody induces long-term remission of overt autoimmunity in nonobese diabetic mice. Proc Nad Acad Sci USA 91:123-127. Christianson SW, Shultz LD, Leiter EH: 1993. Adoptive transfer of diabetes into immunodeficient NOD-scidkcid mice: relativecontributions of CD4+ and CD8+ T-cells from diabetic versus prediabetic NOD.NOD-Thy-la donors. Diabetes 42:44-55. Dardenne M, Lepatdt F, Bendelac A, Bach JF: 1989. Acceleration of the onset of diabetes in NOD mice by thymectomy at weaning. Eur J Immunol 19:889-895. Elias D, MarkovitsD, Reshef T,Van Der Zee R, Walker MD, Cohen IR: 1991. Vaccination against autoimmune mouse diabetes with a T-cell epitope of the human 65-kDa heat shock protein. Proc Natl Acad Sci USA 88: 3088-3091. Feutren G, Papoz L, Assan R, et aL: 1986. Cyclosporin increases the rate and length of remissions in insulin-dependent diabetes of recent onset: resultsof a multicentredoubleblind trial. Lancet 2:119-124.

Garchon HJ, Luan JJ, Eloy L, B6dossa P, Bach JF: 1994. Genetic analysis of immune dysfunction in non-obese diabetic (NOD) mice: mapping of a susceptibilitylocus close to the Bc1-2 gene correlates with increased resistance of NOD T cells to apoptosis induction. Eur J Immunol 24:380-384. Hancock WW, Polanski M, Zhang J, Blogg N, Weiner HL: 1995. Suppression of insulitis in non-obese diabetic (NOD) mice by ord insu-

linadministration isassociated withselective expressionof interleukin-4 and -10, transforming growth factor-~, and prostaglandin-E. Am J Pathol 147:1193-1199. Haskins K, McDuffie M: 1990. Acceleration of diabetes in young NOD mice with a CD4+ islet-specific T cell clone. Science 249:1433– 1436. Hutchings P,O’ReillyL, Parish NM, Waldmann H, Cooke A: 1992. The use of a non-depleting anti-CD4 monoclinal antibody to re-establish tolerance to beta cells in NOD mice. Eur J Immunol 22:1913-1918. Katz JD,Wang B, HaskinsK, Benoist C, Mathis D: 1993. Following a diabetogenic T cell from genesis through pathogenesis. Cell 74:1089-

1100. Kaufman DL, Clare-Salzler M, Tian J, et al.: 1993. Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 366:69-72. Kay TW, Campbell IL, Oxbrow L, Harrison LC: 1991. Overexpressionof class I major histocompatibility complex accompanies insulitis in the non-obese diabetic mouse and is prevented by anti-interferon-gamma antibody. Diabetologia 34:779-785. Kroncke KD, Brenner HH, Rodriguez ML, et aL: 1993. Pancreatic islet cells are highly susceptible towards the cytotoxic effects of chemically generated nitric oxide. Biochim Biophys Acta 1182:221-229. Lampeter EF, Homberg M, Quabeck K, et al.: 1993. Transferof insulin-dependentdiabetes between HLA-identicalsiblings by bone marrow transplantation.Lancet 341:1243–1244. LeiterEH, Serreze DV: 1991. Autoimmune diabetes in the non-obese diabetic mouse: suppression of immune defects by bone marrow transplantationand implications for therapy. Clin Immunol Immunopathol 59:323-334. Nakano K, MoralesJP,Handler ES, GreinerDL, Rossini AA: 1988. Role of host immune system in BB/Wor rat:predisposition to diabetes resides in bone marrow. Diabetes 37:52& 525. Neophytou PI, Roep BO, Arden SD, et aL: 1996. T-cell epitope analysis using subtracted expression libraries(REASEL): application to a 38-kDa autoantigen recognized by T cells from an insulin-dependent diabetic patient. Proc Natl Acad Sci USA 93:2014-2018. Ohashi PS, Oehen S, Buerki K, et al.: 1991. Ablation of “tolerance” and induction of dia-

@ 1997,ElsevierScienceInc., 1043-2760/97/$17.00 PIIS1043-276O(96)OO271-8

73

betes by virus infection in viral antigentransgenic mice. Cell 65:305-317.

rapid onset of diabetes in NOD mice in the absence of CD4 cells. J Exp Med 183:67–76.

Oldstone MBA, NerenbergM, SouthernP,Price J, Lewicki H: 1991. Wus infection triggers insulin-dependent diabetes mellitus in a transgenic model : role of anti-self (virus) immune response. Cell 65:319–331.

Yasunami R, Bach JF: 1988. Anti-suppressor effect of cyclophosphamide on the development of spontaneous diabetes in NOD mice. Eur J Immunol 18:481-484.

Zhang ZJ, Davidson L, Eisenbarth G, Weiner HL: 1991. Suppression of diabetes in nonobese diabetic mice by oraf administrationof porcine insulin. Proc Natl Acad Sci USA 88: 10,252-10,256. TEM

O’Reilly LA, Hutchings PR, Crocker PR, et al.: 1991. Characterizationof pancreatic isletcell infiltratesin NOD mice: effect of cell transfer and transgene expression. Eur J Immunol 21:1171-1180. Pennline KJ, Roque-Gaffney E, Monahan M: 1994. Recombinant human IL-10 prevents the onset of diabetesin the nonobese diabetic mouse. Clin Immunol Immunopathol 71: 169-175. Rapoport MJ, JaramilloA, Zipris D, et al.: 1993. Interleukin4 reversesT cell proliferativeunresponsiveness and preventsthe onset of diabetes in nonobese diabetic mice. J Exp Med 178:87-99. Shehadeh N, Calcinaro F,BradleyBJ,Bruchlim I, Vardi P, Lafferty KJ: 1994.“Effect of adjuvant therapy on development of diabetes in mouse and man. Lancet 343:706–707. Stein PH, Rees MA, Singer A: 1992. Reconstitution of (BALB/cxB6)Fl normal mice with stem cells and thymus from nonobese diabetic mice resultsin autoimmune insulitis of the normal hosts’ pancreases. Transplantation 53:1347–1352. ThivoletC, BendelacA, Bedossa P,Bach W, Carnaud C: 1991.CD8+ T cell horningto the pam creasin the nonobese diabetic mouse is CD4+ T-celldependent.J Immunol 146:85–88. Tisch R, Yang D, Singer SM, Liblau RS, Fugger L, McDevitt HO: 1993. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 366:72-75. Verge CF, Gianani R, Kawasaki E, et al.: 1997. Number of autoantibodies (against insulin, GAD or ICA512/IA2) rather than particular autoantibody specificitiesdetermines risk of type I diabetes. J Autoimmun (in press). Viafettes B, Maraninchi D: 1993. Transfer of insulin-dependent diabetes between HLAidentical siblings by bone marrow transplantation. Lancet 342:174. Wicker L, Appe] MC, Dotta F, et al: 1992. Autoimmune syndromes in major histocompatibility complex (MHC) congenic strains of nonobese diabetic (NOD) mice: the NOD MHC is dominant for insulitis and cyclophosphamide-induced diabetes. J Exp Med 176:67-77. Wicker LS, Todd JA,PetersonLB: 1995.Genetic control of autoimmune diabetes in the NOD mouse. Annu Rev Immunol 13:179–200. Wong FS, Visintin I, Wen L, Flavell RA, Janeway CA: 1996. CD8 T cell clones from young nonobese diabetic (NOD) islets can transfer

74

Autoimmune Hypophysitis Shereen Ezzat and Robert G. Jesse

Autoimmune (lymphocytic) hypophysitis has emerged as a distinct and specific clinical and pathological disease entity. Although relativelyrare compared with other autoimmune endocn”ne diseases, nearly a hundred cases have been described. The condition is much more common in females (9:1) and appears to have a particular predilection for the pregnant and postpartum states. The anterior pituitary, and less often the neurohypophysis, appear to be the targetfo~ inflammatory autoimmune destruction. During the evolution of the disease process, pituitaq hyperfiunction (usually hypeqwolactinernia) has been noted. This disease should now be included in the differential diagnosis of pituitary disorders, especially in females presenting with pituitary enlargement, pa~ticularlyif symptoms occur in tempo~al relationship to pregnancy. The disease may form part of the spectrum of the polyglandula~ autoimmune endocrine disorders. (TrendsEndocrinol Metab 1997;8:7480). 01997, Elsevie~ Science Inc.

Many endocrineglands,eithersinglyor in combination,are subjectto autoimmune injury.The basic mechanismsunderlying the endocrine dysfunctionare thoughtto resultfrom similar aberrationsof the immune system.Lymphocytichypophysitis, a relatively rare example of autoimmune endocrine disease,was first described in 1962(Goudie and Pinkerton 1962).Since that time, nearly a hundred cases have been reported. The condition may be more prevalent,and althoughthere are a few distinctiveclinicalclues,thereare currently no laboratory markers and definitive diagnosiswithout surgicalbiopsy has

Shereen Ezzat is at the Department of Medicine, Division of Endocrinology, WellesleyHospital, Toronto, Ontario, Canada M4Y-1J3; and Robert G. Jesse is at the Division of Endocrinology, St. Michael’s Hospital, University of Toronto, Toronto, Ontario M5B-1W8, Canada.

been problematic.The conditionhasbeen linkedwith severalother autoimmuneendocrinedisorders(Thodou et al. 1995,Asa et al. 1981,Pholsena et al. 1994,Ozawa and Shishiba 1993, Bevan et al. 1992, Jesse 1990). Lymphocytic hypophysitis may have an acuteonsetand occasionally resultsin severe complications,even a lethal outcome.In addition,the clinicaland radiographic presenting features may simulate those of a pituitary adenoma. The diagnosis has been most commonly confirmed by histologic examination.In many endocrine autoimmune diseases, serum autoantibodiesact as serological markersof disease;however,antipituitary antibody measurementis difficult to perform and not routinely available. Moreove~ it remainsunclearwhether the pituitary antibodies have any real diagnostic or predictivevalue,as there is still no definite correlation between the presence of

@ 1997,ElsevierScienceInc., 1043-2760/97/$17.00 PIIS1043-2760(96)O027 O-6

TEM Vol. 8, No. 2, 1997