Immunologic thrombocytopenic purpura in patients at risk for AIDS

Immunologic Thrombocytopenic in Patients at Risk for AIDS

Purpura

S. Karpatkin S U M M A R Y. HIV-seropositive homosexuals, narcotic addicts and hemophiliacs develop a new syndrome of immunologic thrombocytopenic purpura (ITP) which is clinically indistinguishable from classic autoimmune thrombocytopenic purpura (ATP) with respect to increased megakaryocytes in the bone marrow, peripheral destruction of antibody-coated platelets, negative serology for SLE, response to treatment with prednisone and/or splenectomy. Eleven of 79 homosexual patients have developed AIDS 2 to 43 months after the diagnosis of ITP (mean, 22 months). The mechanism of the ITP appears to be different in homosexuals and narcotic addicts when compared to classic ATP. Homosexuals and narcotic addicts have markedly elevated plateetbound IgG and CX4 (2.54fold ATP levels), PEGprecipitable circulating immune complexes and anti-F(ab’)z antibodies (absent in ATP). There is no inverse relatiouship between platelet-bound IgG and platelet count and platelet antibody is usually not elutable from washed platelets as is the case with classic ATP. Homosexual patients do not have 7s platelet antibody in their sera as do classic ATP patients, but appear to have immune complex deposition on their platelets, possibly due to tbe presence of anti-F(ab’)z antibodies. Narcotic addict patients do have detectable 7s platelet antibody but also appear to have immuue complex deposition on their platelets, possibly due to anti-F(ab’)l antibodies. The anti-F(ab’)z antibodies are of the IgG class. They react with autologous, homologous patient and healthy control F(ab’)z fragments. Some anti-F(ab’)z antibodies have broad reactivity, others are more limited. It is postulated that some of the antiP(ab’)z antibodies may he responsible for the thrombocytopenia.

In 1980 an epidemic of rapidly-progressive Kaposi’s sarcoma was noted in sexually-active homosexual men attending the Bellevue Hospital-New York University Medical School Oncology clinic. This was first reported by Hymes et al’ who suggested that these patients might have a disorder of their immune regulation. Shortly thereafter or about the same time, the Acquired Immune Deficiency Syndrome (AIDS) was recognized in this same group of patients as well as narcotic addicts and individuals having received blood or blood product transfusions.2-5 These patients

S. Kaqmtkia, MD, New York University School of Medicine, 550 First Avenue, New York, NY 10016, USA. Blood Reviews (1987) 1, 119-125 0 1987 Longman Group UK Ltd

had a reversal of their T cell helper/ suppressor ratio, were susceptible to opportunistic infections (particularly Pneumocystis carinii) as well as Kaposi’s sarcoma and other lymphomas, and almost invariably, died of their disease. In November, 1980, sexually-active homosexuals seen at the New York University Medical Center who did not have AIDS began to develop immunologic thrombocytopenic purpura (ITP) which was clinically indistinguishable from classic autoimmune thrombocytopenic purpura seen predominantly in females.6 Eleven severe cases were first described with a mean platelet count of 16000 + 3,000/~1, as well as two mild cases. All patients had increased megakaryocytes in the bone marrow, no splenomegaly, negative anti-

120 IMMUNOLOGIC THROMBOCYTOPENIC

PURPURA

nuclear antibody titers and no clinical disorders known to cause thrombocytopenia. Five patients were tested for helper/suppressor T cell ratios; four had decreased ratios that averaged 0.54 + 0.15 (SEM) compared to 1.9 f 0.22 for those of healthy controls. Elevated platelet-bound IgG levels were noted in 8 of 9 patients tested (see below, under Mechanisms of Thrombocytopenia). Elevated circulating immune complexes (PEG method) were noted in 8 of 9 patients (Cfold the normal mean), relative lymphopenia in 6 of 10 patients (OS-fold the normal mean), elevated gamma globulin levels in 4 of 8 patients (1.3fold the normal mean). Both the patient group as well as a control homosexual group (normal complete blood count) had similar increased exposure to herpes simplex virus, cytomegalo virus, Epstein-Barr virus and hepatitis A and B viruses; and similar histories for the use of recreational drugs: marijuana, amyl nitrate and cocaine. HLA-typing for A-, B-, C- and D-loci was performed in eight patients with no significant associations noted. In November, 1982 (2 years later), a similar epidemic of chronic ITP (clinically indistinguishable from classic autoimmune thrombocytopenic purpura seen predominantly in females) was noted in 70 intravenous narcotics addicts treated at Bellevue Hospital (female:male ratio, 1: 3), with mean platelet count of 53 000 + 4,000 (range 13 000 to 140 OOO/ul).’ These patients were or had been chronic intravenous abusers of heroin and/or cocaine for a mean duration of 10 years. Of particular interest was the observation that 33 patients had stopped taking intravenous drugs for an average of 21 months, indicating that the thrombocytopenia was not due to the acute exposure to narcotics. Other possible causes of thrombocytopenia such as hypersplenism, chronic active hepatitis or AIDS had been ruled out. Elevated platelet-bound IgG levels were noted in 66 of 70 patients (see below under Mechanisms of Thrombocytopenia). Elevated circulating immune complexes were noted in 33 of 36 patients tested (7.3-fold the normal mean). In 1983 Ratnoff et al reported on the development of ITP in five patients with hemophilia (AHF deficiency) who were multiply-transfused with lyophilized AHF concentrates for 5-9 years, mean platelet count, 34600 f 29000 (range, 8,000 to 82OOO/ul).s This syndrome was clinically indistinguishable from classic autoimmune thrombocytopenic purpura seen predominantly in females. All five had elevated plateletbound IgG levels (see below under Mechanisms of Thrombocytopenia). Elevated circulating immune complexes (Clq method) were noted in 2 of 3 patients tested. Hypergammaglobulinemia was noted in 3 of 3 patients tested (2.2-fold the normal mean). Relative lymphopenia was noted in 3 of 4 patients (0.8-fold the normal mean). The helper/suppressor T cell ratio was decreased in 4 of 5 patients tested and averaged 0.55 + 0.15 (SEM), compared to 1.94 + 0.47 for healthy controls.

IN PATIENTS AT RISK FROM AIDS

Prognosis and Treatment Practically all patients at risk for AIDS who develop ITP are seropositive for the human immunodeficiency virus (HIV) (Table 1). Some patients have presented with ITP and AIDS, whereas others have developed AIDS after first presenting with ITP. Homosexual patients have been carefully studied with respect to their prognosis by our group9 as well as Abrams et a1.l’ Seventy-nine have been followed for a mean period of 21 months. Eleven patients have developed AIDS 2-43 months after the diagnosis of ITP (Pneumocystis carinii pneumonia, lymphoma, disseminated tuberculosis, cryptococcus neoformans sepsis) (Table 2). Four patients have died 17-46 months after presentation with thrombocytopenia. Nine patients spontaneously reverted to normal platelet counts 5-27 months (median, 10 months) after the diagnosis of thrombocytopenia, in the absence of splenectomy and while not receiving corticosteroids. One of these patients presented with a platelet count of 11 OOO/ul. One patient whose platelet count reverted to normal developed AIDS 13 months later. Thirty-five of 41 patients had a moderate (> SOOOO/ul)to excellent ( > 100 OOO/ul) response while on corticosteroids (30 relapsed following cessation of corticosteroids). Ten of 10 patients in our group had an excellent response to splenectomy which has persisted; 10 of 15 responded in the group of Abrams et al.” Twenty-six patients did not require treatment for their thromboctopenia. Our rationale for treatment has been to use the lowest possible corticosteroid dose capable of raising the platelet count to safe levels, generally >25OOO/ul, with absence of significant purpura. Prednisone is generally given at a dose of 30-40 mg/day for l-2 weeks and then rapidly tapered to a maintenance dose of 1015 mg/day. The patient is then observed for 10 months, which is the median time for spontaneous reversion to a normal platelet count in 18% of our patients. If the patient has not reverted to a safe platelet count by this time, or requires more than lo-15 mgm of prednisone/day to maintain such a platelet count, splenectomy has been performed. We have no evidence that corticosteroid treatment or splenectomy contributes to the developTable 1 Antibody Incidence and Titers Against the Human Immunodeficiency Virus in Thrombocytopenic Homosexuals, Narcotic Addicts and Controls Pas/Total

% Pos

Homosexual controls Homosexual ITP Homosexual ITP + AIDS

4112 13/14 12/12

33 93 100

1:2152 1: 2065 1:1355

Narcotic controls Narcotic ITP

6/10 13115

60 87

1:805 1: 1033

l/36

3

Heterosexual controls

Geometric Titer*


* Mean geometric titer of positive samples. A human T cell line, HUT 78, heavily infected with the AIDS retrovirus was used as antigen in an indirect immunofluorescence assay.

BLOOD REVIEWS Table 2

Change in Clinical Status of 19 Homosexual Patients Presenting with Thrombocytopenia Development of normal platelet count

Development of AIDS

Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

12 1

Months after ITP

Pathology*

2 8 15 16 16 17 22 33 35 37 43

KS, PCP KS, Lym KS PCP, MFLE Crypt0 PCP KS PCP Lym KS, PCP, DTB CMV

Months after splenectomy

Platelet count x 10e3 init-subseq$

Months after ITP

1 _ 10 13 _ _ _ 31 16 35

9

66254

5 9 10 21 27 6 612 12

127-163 60-260 80-262 3@244 11-313

* KS, Kaposi’s sarcoma; PCP, pneumocystis carinii pneumonia; Lym, lymphoma; MFLE, multifocal leukoencephalopathy; DTB, disseminated tuberculosis; Crypto, cryptococcus neoformans sepsis; CMV, cytomegalo inclusion virus infection. 9 Initial thrombocytopenic vs. subsequent normal platelet count

ment of AIDS. Prednisone therapy, however, has been complicated by the development of oral candidiasis as well as activation of latent herpes simplex virus. Our experience with narcotic addicts’ and hemophiliacs has been limited. Twenty-two narcotic addicts have been treated with corticosteroids. Six had a complete response, 13 a moderate response ( > 50 OOO/ul),three did not respond while on prednisone. Two patients responded to splenectomy, one was complete and the other partial. The experience with hemophiliacs is sparse. Ratnoff et al reported a partial response while on prednisone in 4 of 4 patients and a complete response to splenectomy in one of these patients.* Josephson et al” have reported a case with partial response while on prednisone and complete response to splenectomy.

Mechanisms of Thrombocytopenia Homosexuals, narcotic addicts and hemophiliacs are three groups of individuals at risk for the development of AIDS. All three develop the newly described syndromes of immunologic thrombocytopenic purpura in the absence or presence of AIDS. The thrombocytopenia appears to result from exposure to the AIDS virus, because 87-93% of thrombocytopenic patients without AIDS have antibody against the HIV virus, and the syndrome(s) of ITP was first recognized following the epidemic of AIDS. The mechanisms of homosexual and narcotic addict ITP (HSITP and NITP, respectively) have been studied extensively by our g~-oup,~*~~*~~ compared to classic autoimmune thrombocytopenic purpura (ATP) patients, and shown to have similarities as well

as dissimilarities. Both HSITP and NITP patients are similar with respect to the markedly elevated plateletbound IgG and C3C4 (compared to classic ATP patients, Figs. 1 and 2) as well as the presence of circulating immune complexes determined by the PEG method (absent in ATP patients, Tables 3 and 4). Intermediate values are found in non-thrombocy-

>300 ]

COMPLEMENT ATP HSITP .

_ISG ATP HSITP

C

260 220 180

8

9 g 8

403224-

2

16-

. t

L . .

”.

8oFig. 1 Simultaneous measurements of platelet-bound IgG and complement in 23 patients with autoimmune thrombocytopenic purpura (ATP) and 20 homosexual patients with thrombocytopenic purpura (HSITP). Values for IgG are nanograms per 1 O6 platelets, and those for complement are nanograms per 10’ platelets. The shaded area indicates the mean f 2 SD for platelet values from normal subjects. Arrows indicate the mean values for each group.

122

IMMUNOLOGIC

THROMBOCYTOPENIC

CA: 866

>lOO 80

II .

CAT 0 T N P

me

T P

0

N

T P

T P

.

.

. .

60 -

CA:

:

l

t

PURPURA IN PATIENTS AT RISK FROM AIDS

*

*

l

. 8 .

.

.

I

. :

+.

I

.

l

i

:

s. ; *

El w

w

C3C4

Fig. 2 Simultaneous measurement of platelet-bound IgG, IgM and complement in narcotic addicts with idiopathic thrombocytopenic purpura (NITP) and patients with classic autoimmune thrombocytopenic purpura (ATP). Values for IgG are expressed as rig/l O6 platelets; values for IgM and complement as rig/l 0’ platelets. Arrows refer to mean values. Control values (CON) are expressed as mean + 2 SD by horizontal lines within the open boxes.

topenic HIV seropositive homosexuals and narcotic addicts, Tables 3 and 4. Of particular interest is the presence of IgG anti-F(ab’)a antibodies which correlate with the level of immune complexes’3 (see below, under Anti-F(ab’)z Antibodies). However, they are dissimilar with respect to serum plateletreactive 7S IgG antibody. In our experience, serum platelet-reactive 7S IgG could not be demonstrated in 3 HSITP patients tested (Table 5), whereas it could be demonstrated in 9 of 13 NITP patients as well as 2 of

2 classic ATP patients (Table 5). In addition, antiplatelet IgG could not be eluted from 17 of 18 HSITP patients or 8 of 11 NITP patients, whereas it could be eluted from 20 of 23 classic ATP patients. These data suggest that HSITP may be immune-complex mediated since circulating immune complexes are found in their sera, high levels of IgG and C3C4 are found on their platelets, 7S anti-platelet IgG could not be demonstrated in their sera or on their platelets, and antiplatelet IgG could not be eluted from their platelets. NITP may represent a mixture of immune complex as well as specific 7S antibody mechanisms since high levels of immune complexes can be found in their sera, high levels of IgG, C3C4 and IgM are found on their platelets, but 7S anti-platelet IgG can be found in their sera. Snicker et a1,14 employing a highly-sensitive western blotting technique, have recently reported platelet membrane antibody in the sera of HSITP patients, as well as homosexual controls, with reactivity against a 25000 molecular weight platelet membrane antigen, suggesting the presence of a specific 7S platelet antibody. The pathophysiologic significance of this very low concentration of platelet antibody, found by Stricker et al in HSITP, as well as very high concentration of apparent platelet immune-complex deposition of IgG, C3C4 and IgM noted by our group, in HSITP and NITP patients, remain to be determined. Both groups of patients have increased megakaryocytes in their bone marrow aspirates and decreased platelet survival has recently been demonstrated in HSITP patients. ’ 5 Thus immunoglobulin-mediated peripheral destruction of platelets appears to be the most likely mechanism for the thrombocytopenia. The situation is further complicated by the observation that HSITP as well as AIDS patients have impaired reticula-endothelial function. Perhaps this later observation explains the lack of inverse correlation between platelet count and platelet IgG in both groups of patients, which is generally found in classic ATP patients.

Table 3 Comparison of Idiopathic Thrombocytopenic Purpura (ITP) in Homosexuals with Classic Autoimmune Thrombocytopenic Purpura (ATP)’ Platelet IgG ng/106

Complement ng/106

Serum titer

Immune complexes mg/ml

HSITP Mean + SEM No. high/total§

31.5 * 1.6 29133

1.1+ 1.9 16120

1:64 l/24

0.96 f 0.10 21124

Homosexual controls Mean f SEM No. high/total

5.0 f 0.6 14128

1.9 f 0.08 6128

Classic A TP Mean f SEM No. high/total

8.4 + 1.6 19122

I:32 5119

Controls f 2 SD

1.5 * 1.3

1.9 +0.09 6123 1.5+0.7

0.32 + 0.02 O/5 0.34 * 0.06

* Mean ( f SEM) platelet counts for homosexuals with thrombocytopenic purpura and patients with autoimmune thrombocytopenic purpura were 62000 f 6,000 and 73 000 f 7,W per cubic millimeter, respectively, for measurements of platelet IgG levels. $ Values represent the number of subjects in whom elevated levels were found/the number of subjects studied

BLOOD REVIEWS Table 4

Comparison of Idiopathic Thrombocytopenic Addicts with Classic Autoimmune Purpura (ATP)*

123

Purpura (ITP) in Narcotic

Platelet ng/106

Complement ng/106

Serum titer

Immune complexes mgiml

48.9 + 9.6 66170

2.86 f 0.51 45166

1:28 11/18

1.54* 0.20 33136

10.7 + 2.2 S/l0

1.11 kO.18 318

18.8 &-2.6 24125

1.20+0.10 9125

1:32 5/19

0.21 k 0.06 O/8

IgG NITP

Mean i SEM No. high/total Narcotic

controls

Mean + SEM No. high/total Classic ATP

Mean k SEM No. high/total

0.91 f 0.26

2.9 k 1.3

Controls + 2 SD

0.21 + 0.20

* All values given as mean + SE except those for controls which are + 2 SD. The mean platelet count of patients with narcotic idiopathic or autoimmune thrombocytopenic purpura was 53 000 & 4,000/mm3 (SE) and 51000 f 6,500/mm3, respectively. Narcotic control patients represented addicts with normal platelet counts, averaging 205 000 k 16 000/mm3

Anti-F(ab’)z Antibodies

Table 5 Comparison of the Effect of Whole Serum with that of 7s IgG Fraction on Binding to Platelets* ITP in three homosexuals

Classic ATP

PLAI

Serum titer IgG titer

I:128 0

1:128 > 1:32

1:128 >1:32

Serum titer IgG titer

I:128 0

z I:32 > 1:32

Serum titer IgG titer

1:64 0

Preliminary studies designed to analyze platelets and immune complexes for viral antigens suggested the presence of ‘anti-antibodies’. Fixed washed platelets or the ether eluates of six HSITP patients revealed absence of EBV, CMV, HSV or ADE (Adenovirus) viral antigens on their platelets, despite the presence of viral antibodies for these antigens in their sera. Similar findings were noted in four fixed immune complexes of these patients. However, viral antibodies against EBV, CMV, HSV and rubeola virus were

* IgG was separated by gel filtration. ITP denotes idiopathic thrombocytopenic purpura, ATP autoimmune thrombocytopenic purpura, and PLAI isoimmune antiplatelet antibody

Table 6 Viral Antibody Titers in Sera and Immune Complexes of Homosexual Thrombocytopenic Patients and Control Subjects* Patients Virus

Pl

Controls P2

P3

P4

Cl

c2

c3

HSV

Serum Complex

128 128

CMV Serum Complex

~8 ~8

512 ~8

128 128

512 ~8

32 -=8

128 ~8

2,048 2,048

256 ~8

256 128

128 128

32 ~8

~8 ~8

~8 ~8

1,280 2,560

5,120 20 480

10240 20 480

640

1.280

320 <20

320 <20

80 <20

160 320

160 1,280

320 640

80 320

80 <5

80 <5

80 <5

64 256

256 512

64 128

32 8

EBV( VCA)

Serum Complex EBV(EBNA)

Serum Complex Rubeola

Serum Complex

ND

ND

ND

VCA, viral capsid antibody; EBNA, Epstein Barr nuclear antigen; ND, not done * Reciprocal viral titers are given for patients (PI-P4) and controls (ClC3) in serum samples as well as immune complexes obtained from the same serum samples. Immune complex precipitates were diluted to the same volume of serum from which they were obtained

124 IMMUNOLOGIC THROMBOCYTOPENIC

PURPURA IN PATIENTS AT RISK FROM AIDS

Table 7 Binding Ratios of Homosexual IgG to F(ab’)2 Fragments of Homosexual Patients and Control Individuals Compared to Binding of Control IgG to Control F(ab’)s Fragments* Purified IgG Immunoglobulin of Homosexuals F(ab% fragments

A

B

C

D

I

R

S

T

U

V

E

F

Control G

A B C D E F G Control H Control Q Control

6.6 6.6 4.4 3.2 3.1 7.0 3.4 6.2 -

2.6 2.5 1.9 1.7 1.7 1.7 2.5 2.9 -

1.8 1.8 1.8 1.3 1.5 1.4 2.0 1.6 -

4.0 2.8 9.4 2.2 2.0 6.4 1.2 6.0 -

1.4 1.2 1.6 1.0 2.5 2.5 2.1 -

0.8 1.5 1.3 1.2 -

1.1 1.5 0.7 1.0 1.6 0.9

0.8 1.1 0.6 0.9

0.9 1.1 0.6 0.9

0.9 1.1 0.8 1.0

1.5 1.3 1.3 1.4 1.3 1.2 1.0 1.3 -

1.5 1.3 1.6 1.7 2.0 1.4 1.1 1.6 -

1.1 0.9 0.9 1.0 0.8 1.3 1.0 1.0 -

* F(ab’)z fragments (0.2 ug) of sexually active homosexual patients and control subjects (G,H,Q) were adsorbed to microtiter plates that were then reacted with purified IgG of patients or control subjects at 12 serial dilutions of IgG, starting at 0.2 pg. r2%protein A was then added and assayed for radioactivity. The cpm obtained for each serial dilution was divided by the cpm obtained from a simultaneously run control experiment in which control F(ab’)r fragments were reacted with autologous control IgG at serial dilution. The numbers given refer to the mean of the highest ratios obtained for a minimum of two consecutive dilutions. A ratio > 1.3 is considered positive. It was determined from the two SD measurement obtained from six replicate determinations with six different control subjects. Each ratio is the mean of 2-5 experiments. F(ab’)r fragments were passed through a protein A column and shown to be free of IgG as determined by both SDS-PAGE and radioimmunoassay. -, not done

TaMe 8 Binding Ratios of Narcotic Addict IgG to F(ab’)l Fragments of Patients and Control Individuals Compared to Binding of Control IgG to Control F(ab’)r Fragments* Purified IgG immunoglobulin of narcotic addicts F(ab’)z fragments

AA

BB

CC

DD

EE

A Z V E F W X H Control Q Control

4.9 0.8 1.5 1.0 2.4 1.0 0.9 1.4 0.9

4.8 0.8 1.0 1.0 2.0 1.0 0.8 I.1 1.0

4.9 0.8 1.0 1.0 1.8 1.2 0.8 0.9 -

5.4(9.0)-$ 0.7 1.1 1.2 1.6 1.5 0.8 1.0 0.9

6.0 0.8 1.1 2.1 1.8 1.2 0.8 1.0 0.9

* Methodology same as in Table 7. Patients A,Z,V have Kaposi’s sarcoma and ITP; patients E and F have Kaposi’s sarcoma, patients W and X have narcotic addict ITP, subjects H and Q are controls. 8 When adsorbed to solid-phase F(ab’)s of patient A and eluted. the ratio was 9.0

noted, and correlated with their presence in sera (Table 6). No such correlation was noted with preparations from three healthy control subjects whose immune complex preparations were negative in the presence of positive serum titers. Following this lead, anti-F(ab’)z antibodies were sought and found in 10 of 16 homosexual patients (Table 7) and 6 of 6 narcotic addicts (Table 8) which correlated with immune complex levels Seven HSITP patients had anti-F(ab’)z antibodies of moderate to marked titer with broad reactivity against autologous, homologous patient and normal control F(ab’)z fragments. Three others demonstrated limited reactivity against one or two F(ab’)z fragments. Five of 6 narcotic addict patients had limited reactivity, one other had broad reactivity. In contrast, anti-F(ab’)z

antibodies were not detectable in six classic ATP patients against autologous, homologous ATP patient or normal control subject’s F(ab’)z fragments. The IgG antibody found in the patient was not rheumatoid factor (IgM against IgG-Fc), nor IgM anti-IgG, but rather IgG anti-F(ab’)z. Similar antiF(ab’)z antibodies have recently been reported in the sera of patients with rheumatoid arthritis’6-‘g and systemic lupus erythematosus,16*1g-21 disorders of immune regulation that are associated with the presence of circulating immune complexes as well as autoantibodies. The markedly elevated plateletbound IgG and C3C4 in homosexual and narcotic addict ITP patients has suggested a role for immune complexes in the induction of thrombocytopenia. It is likely that some of the platelet-bound IgG and C3C4 represent immune complexes and that some of the immune complexes may be composed of IgG antiF(ab’)z complexes. It is possible that some of these anti-F(ab’)z antibodies might be anti-idiotypes. Attempts to demonstrate specificity were unsuccessful; however this was for technical reasons. If these represent anti-idiotype antibodies against HIV neutralizing antibody, then it could represent an immune-regulatory mechanism in which enhanced anti-idiotype antibody (against neutralizing anti-HIV antibody) results in lack of resistance to HIV infection. Indeed, IgG anti-F(ab’)z antibodies with blocking activity against DNA antibody2’s2 ’ as well as tetanus toxoid antibody’l have recently been reported in patients with SLE. In addition, individuals immunized with KLH have antiF(ab’)z antibodies against KLH antibody in their circulating immune complexes.” The etiology of the circulating immune complexes, as well as autoimmune and alloimmune anti-F(ab’)z antibodies in these patients at risk from AIDS, or

BLOOD REVIEWS

with AIDS, is not readily apparent. Experimental anti-antibodies have been produced after immunization of rabbits with antigen as well as antigenantibody complexes. 23-2s It has been postulated that immune complex formation results in unfolding of sites on the autologous IgG molecule that could expose buried antigenic determinants not recognized by the host. Repeated antigenic stimulation in patients at risk from AIDS (a disorder of immune regulation with polyclonal antibody stimulation) could lead to the very high level of circulating immune complexes noted. These in turn may contribute to the formation of anti-F(ab’)z antibodies, thus perpetuating immune complex formation, immune complex deposition on platelets, resultant thrombocytopenia, and possibly, down regulation of anti-HIV neutralizing antibody.

10.

11.

12.

13.

14.

References 1. Hymes K, Cheung T, Greene J B, Prose N S, Marcus A, Ballard H. William D C. Laubenstein L J 1981 Kauosi’s sarcoma in homosexual men: a report of 8 cases. Lancet ii: 598-600 2. Friedman-Kien A, Laubenstein L, Marmor M, et al 1981 Kaposi’s sarcoma and pneumocystis pneumonia among homosexual men-New York City and California. Morbidity and Mortality Weekly Report 30: 305-308 3. Gottlieb M S, Schroff R, Schanker H, Weisman J D, Thim Fan P, Wolf R A, Saxon A 1981 Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men. Evidence of a new acquired cellular immunodeficiency. New England Journal of Medicine 305: 1425-1431 4. Masur H, Michelis M A, Greene J B, Gnorato I, Stouwe Vande R A, Holxman R S, Wormser G, Brettman L, Lange M, Murray HW, Cunningham-Rundles S 1981 An outbreak of community acquired pnemnocystis carinii pneumonia. Initial manifestation of cellular immune dysfunction. New England Journal of Medicine 385: 1431-1438 5. Siegal F P, Lopez C, Hammer G S, Brown A E, Kornfeld S & Gold J, Hassett J, Hirschman S Z, CunninghamRundles C. Adelsbera B R. Parham D M. Sieeel M. Cunningham-Rum& S, Armstrong D 1981 S&ere acquired immunodeficiency in male homosexuals, manifested by chronic perianal -ulcerative herpes simplex lesions. New Enaland Journal of Medicine 305: 1439-1444 6. M&is L, Distenfeld A, Amorosi E, Karpatkin S 1982 Autoimmune thrombocytopenic purpura in homosexual men. Annals of Internal Medicine 96: 714-717 7. Savona S, Nardi M A, Lennette E T, Karpatkin S 1985 Thrombocytopenic purpura in narcotics addicts. Annals of Internal Medicine 102: 737-741 8. Ratnoff 0 D, Menitove J E, Aster R H, Lederman M M 1982 Coincident classic hemonbilia and ‘idionathic’ thrombocytopenic purpura in-patients under-treatment with concentrates of antihemophilic factor (factor VIII). New England Journal of Medicine Xl& 439-442 9. Walsh C, Krigel R, Lennette E, Karpatkin S 1985 Thrombocytopenia in homosexual patients. Prognosis,

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

125

response to therapy, and prevalence of antibody to the retrovirus associated with the acquired immunodeflciency syndrome. Annals of Internal Medicine 103: 542-545 Abrams D I, Kiprov D D, Goedert J J, Sarngadharan M G, Gal10 R L, Volberding P A 1986 Antibodies to human T lymphotropic virus type III and development of the acquired immunodeficiency syndrome in homosexual men presenting with immune thrombocytopenia. Annals of Internal Medicine 104: 47-50 Josephson M A, Vokes E E, Vogelzang N J, Ash M, Telfer M C. Baron J M 1987 Acauired immune thrombocvtonenia in an HTLV-III antibody positive patient with hemophilia A: A case report with platelet immune studies. (Submitted for publication) Walsh C M, Nardi M A, Karpatkin S 1984 On the mechanism of thrombocytopenic purpura in sexually active homosexual men. New England Journal of Medicine 311: 635-639 Yu, Ji-Ren, Lennette E T, Karpatkin S 1986 Anti-F(ab’)l antibodies in thrombocytopenic patients at risk for acquired immunodeficiency syndrome. Journal of Clinical Investigation 77:.17561761 Stricker R. Abrams D. Corash L. Shuman M A 1985 Tareet platelet antigen in homosexual men with immune thrombocytopenia. New England Journal of Medicine 313: 1375-l 380 Siegel R S, Rae J L, Kessler C M 1986 Immune thrombocytopcnic purpura in HTLV-III men. Blood 68 (Supplement 1): 134a Heimer R, Wolfe L D, Abruxzo J L 1982 IgM and IgG antiF(ab’), antibodies in rheumatoid arthritis and systemic lupus erythematosus. Arthritis and Rheumatism 25: 1298-1306 Birdsall H H, Lidsky M D, Rossen R D 1983 Anti-F(ab’)t antibodies in rheumatoid arthritis. Arthritis and Rheumatism 26: 1481-1492 Persselin J E, Stevens R H 1985 Anti-Fab antibodies in humans. Predominance of minor immunoglobulin G subclasses in rheumatoid arthritis. Journal of Clinical Investigation 76: 723-730 Nasu H, Chia D S, Knutson D W, Bamett E V 1980 Naturally occurring human antibodies to the F(ab’)l portion of IgG. Clinical and Experimental Immunology 42: 378-386 Nasu H, Chia D S, Taniguchi 0, Bamett E V 1982 Characterization of anti-F(ab’)a antibodies in SLE patients. Evidence for cross-reacting auto-anti-idiotypic antibodies. Clinical Immunology and Immunopathology 25: 80-90 Silvestris F, Bankhurst A D, Searles R P, Williams R C Jr 1984 Studies of anti-F(ab’)z antibodies and possible immunologic control mechanisms in systemic lupus erythematosus. Arthritis and Rheumatism 27: 1387-1396 Birdsall H H, Rossen R D 1983 Characterization of antiF(ab’), antibodies in human sera; identification of soluble immune complexes that contain hidden anti-KLH and blocking anti-immunoglobulins following immunization with keyhole limpet haemocyanin. Clinical Experimental Immunology 53: 497-504 Williams R C Jr, Kunkel H G 1963 Antibodies to rabbit gamma globulin after immunizing with various preparations of autologous gamma-globulin. Proceedings of the Society of Experimental Biology and Medicine 112: 554-561 Abruzxo J L, Christian C L 1961 The induction of rheumatoid factor-like substance in rabbits. Journal of Experimental Medicine 114: 791-806 Aho K, Wagner 0 1961 Production of ‘anti-antibodies’ in rabbits. 1961 Annales Medicinae Experimentalis et Biologiae Fenniae 39: 79-87