- Email: [email protected]
Relation of human T lymphotropic virus type III antibodies to T lymphocyte subset abnormalities in hemophiliac patients
Relation of Human T Lymphotropic Virus Type III Antibodies to T Lymphocyte Subset Abnormalities in Hemophiliac Patients
JAMES A. HOXIE, M.D. JEROME L. RACKOWSKI ANDREW J. CEDARBAUM SHELLEY HURWITZ, M.S. PATRICIA M. CATALANO, M.D. Philadelphia,
Pennsylvania
From the Hematology-Oncology Section, Department of Medicine, Hospital of the University of Pennsylvania, the Biostatistics Program, University of Pennsylvania Cancer Center, and the Cardeza Foundation Hemophilia Center, Department of Medicine, Jefferson University Hospital, Philadelphia, Pennsylvania. Dr. Hoxie is a Special Fellow of the Leukemia Society of America. This work was supported by a grant from the W. W. Smith Charitable Trust and the Hemophilia Program of the Commonwealth of Pennsylvania. Requests for reprints should be addressed to Dr. James A. Hoxie, Hematology-Oncology Section, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pennsylvania 19104. Manuscript accepted July 10, 1985. A continuing medical education quiz on this ar-ticle (one hour of Category 1 credit) appears on page Al 11 of this issue.
The relationship of T lymphocyte subset abnormalities and the presence of antibodies to the human T lymphotropic virus type Ill (HTLV-III) was evaluated in 66 adult patients with hemophilia. Positive test results for antibodies to HTLV-III were observed in 62 percent of patients with hemophilia A and 9 percent of patients with hemophilia 6. Patients with HTLV-III antibodies had lower percentages and numbers of T helper (T4) cells and increased percentages of T suppressor (16) cells compared with percents in patients without antibodies to HTLV-III. The mean T4/T6 ratio for antibody-negative patients was 1.45 compared with 0.65 for antibody-positive patients (p
August
1966
The American
Journal
of Medicine
Volume
61
201
RELATION
OF HTLV-III
TABLE
I
Analysis
ANTIBODIES
TO T LYMPHOCYTE
of Lymphocyte
Subsets
SUBSET
ABNORMALITIES-HOXIE
in Hemophiliac
Patients
Type of Hemophilia (n i55) Lymphocytes
(n Jll) 1,645
p value
Transfusion History* Factor IX No Transfusions (n = 13) (n = 10)
Factor VIII (n = 40)
NS
1,721 ww NS
183
:83
(556)
ET AL
1,637 (584) NS
0380)
1,512
Control Subjects (n = 46) 2,014 (678)
NS
T3 Percent
p value Number p value
NS
NS
NS
NS
NS
1,342 (839) NS
1,295
1,367
(476)
(885) NS
1,319 (517) NS
1,096 (759) NS
NS 689
<0.0001 512 (349) <0.0001
;‘1, NS 639 (239) NS
NS
1,493
(516)
T4 Percent <0.0001 536 (357) <0.0001
p value Number p value
(232) NS
NS 659
947
VW
(346)
ii
F-G
NS
T8 Percent
(443)
p value Number
<0.0001 759 (531) <0.05 0.86 (0.54) <0.0001
p value T4lT8
ratio p value
NS
$63)
NS
554 (204) NS
<0.0001 799 (566) <0.05
NS
627 (343) NS
451 (311) NS
550
1.32 (0.43) <0.05
0.73 (0.41) <0.0001
1.21 (0.52) <0.005
1.54 (0.59) NS
1.81 (0.51)
(211)
Results are expressed as a mean of percent positive cells or as a mean absolute lymphocyte count/mm3. Numbers in parentheses indicate SD; p values correspond to comparisons with control group. * The factor VIII and factor IX recipient categories do not include the one patient who received both VIII and IX products. Two patients with hemophilia A, whose transfusion records were not available, are omitted from this section.
PATIENTS AND METHODS Patient Population. Blood specimens for lymphocyte subset and serologic analysis were obtained from 66 adult hemophiliac patients from May 1 to September 30, 1984. They included 55 patients with hemophilia A and 11 patients with hemophilia B. The mean age of this study group was 32 (range, 15 to 65); there were 53 Caucasians and 13 blacks. Of the 66 patients studied, one had AIDS compiicated by Prieumocystis carinii pneumonia and three had chronic unexplained lymphadenopathy. The other 62 patients had no symptoms of chronic lymphadenopathy or weight loss. Five had chronic liver disease thought to be related to hepatitis B. One patient was a homosexual, and one had a history of intravenous drug use. Records of blood product transfusion could be obtained for 64 of the 66 patients: two patients with hemophilia A lacked documentation of their transfusion history. During the previous year, 54 of these 64 patients had received factor concentrates. Of these, 41 received factor VIII concentrates (mean number of units transfused, 82,054; range, 1,320 to 225,686), and 14 received factor IX concentrates (mean number of units transfused, 39,143; range, 3,480 to 123,740). The factor IX recipients included four patients
202
August
1986
The American
factor Viii inhibitors and one patient who received both factor VIII and IX concentrates. Among the 54 patients receiving factor concentrates, 19 patients (15 fatittar VIII recipients and four factor IX recipients) received factor concentrates from lots that were subsequently recalled because two of the donors whose blood was used in the preparation of these products demonstrated AIDS foliowing donation. Ten patients had not received factor concentrates during the year prior to this study. Of these 10 patients, three had received factor Viii concentrates from 1982 to 1983, totaling 68,510, 16,260, and 14,860 units. The other seven patients had not received any.factor concentrates. One of these seven patients had received cryoprecipitate (48 bags) in 1983. Additional serum samples obtained for HTLV-III antibody determination included samples from 67 patients with AIDS, 78 male homosexuals with, Chronic unexplained lymphadenopathy or a wasting syndrome, and 22 asymptomatic homosexual men in the Philadelphia area. Testing was also performed on samples from 1 j3 control subjects without AIDS-related conditions including nine multiparous women, 21 patients with systemic lupus erythematosus, 25 with idiopathic thrombocytoperiic purpura, seven with scieroderma, seven with thrombotic thrombocy-topenic purpuwith
Journal
of Medicine
Volume
61
RELATION
OF HTLV-III
ANTIBODIES
ra, nine with pre-eclampsia and thrombocytopenia, and 35 with malignant disease including three with non-AIDS Kaposi’s sarcoma. Serum samples from six patients known to have antibodies to HTLV-I were also evaluated. Control blood samples for lymphocyte subset analysis were obtained from 46 healthy laboratory workers. The mean age of this control group was 36 (range, 23 to 63) and included 22 males and 24 females. Lymphocyte Subset Analysis. Analysis of T lymphocyte subsets was performed using monoclonal antibodies OKT3, OKT4, and OKT8 (Ortho Diagnostic Systems; Raritan, New Jersey). Optimal titrations of antibodies were added directly to fresh samples of whole blood, and, after red blood cells were lysed, lymphocytes were analyzed on a Spectrum-Ill fluorescence celi analyzer (Ortho Diagnostic Systems), as described by Hoffman et al [ 151. HTLV-III Antibody Determlnatlon. Antibodies to HTLV-III were detected by enzyme-linked immunosorbent assay using the Bio-EnzaBead kit provided by Litton Bionetics (Charleston, South Carolina). Serum samples were stored at -7O’C and tested at a 1 to 900 dilution. The optical density at 405 nM of each tested sample was determined with an LBI 30 spectrophotometer (Litton Bionetics) and compared with three negative control samples (provided by Litton ,Bionetics) tested concurrently. Test results were scored as positive when the ratio of the optical density of the tested sample to the, average optical density of the negative control samples was more than 2.0. Selected samples were evaluated by repeating the enzyme-linked immunosorbent assay and by Western blot analysis using purified HTLV-III antigens (performed by Litton Bionetics)
[1’31. Data An&&. Group comparisons for continuous variables included the t test or Wilcoxon rank sum test, and analysis of variance with the Sheffe procedure. Categorical comparisons used the chi-square test with the Yates’ correction. Variable independence was tested by Pearson product moment or Spearman rank correlation when appropriate. All tests were two-tailed, and resulting p values above 0.05 were considered not statistically significant. Data are expressed as the mean f standard deviation (SD). RESULTS Analysis of T Lymphocyte Subsets. Patients with hemophilia A were found to have lower T4/T8 ratios than those in normal control subjects (Table I). Significant decreases were found in both the percent and absolute number of T4 cells, and significant increases were present in both the percent and absolute number of T8 cells. No differences were observed in the total number of T lymphocytes as determined by the OKT3-reactive cells. As shown in Table I, these alterations in T4 and T8 cells were more apparent in patients receiving factor concentrate transfusions, particularly factor VIII. Among the 44 patients with hemophilia A who had received transfusions within the previous year, the mean T4/T8 ratio was 0.73 f 0,43, compared with 1.59 f 0.60 for the nine patients with hemophilia A who had not received transfusions
August
TO T LYMPHOCYTE
0
2
SUBSET
4
ABNORMALITIES-HOXIE
6
OD 405
8
ET AL
IO
12
RATIO
Figure 1. Frequency histogram of HTLV-III antibody resuits in hemophiliac patients. Results are expressed as the optical density ratio at 405 nM (00405) of the patient sample result to the average result of three negative control samples.
within the previous year (p <0.0005). Abnormalities in lymphocyte subsets were less apparent in patients with hemophilia 8. T4/T8 ratios were below 1.0 in 38 of 55 patients (69 percent) with hemophilia A and two of 11 patients (18 percent) with hemophilia B, whereas T4/T8 ratios were above 1.0 in all of the 46 normal control subjects. HTLV-III Antibody Determination. Antibodies to HTLVIll were detected using the Bio-EnzaBead enzyme-linked immunosorbent assay, which employs purified viral antigens coupled to magnetic beads [ 171. The results were designated “positive” for antibodies to HTLV-III if the optical density at 405 nM was more than two times the average of three negative control samples tested concurrently. A bimodal distribution was seen among hemophiliacs (Figure 1). Positive values were observed in 41 of 66 (62 percent) of the patients with hemophilia, including 40 of 55 (73 percent) with hemophilia A and one of 11 (9 percent) with hemophilia B. The mean ratio of the optical density value in patient serum samples relative to the average optical density value of the negative control samples was 7.9 f 2.5 for the 41 HTLV-Ill-positive patients and 1.0 f 0.1 for the 25 HTLV-Ill-negative patients. Positive results for HTLV-III antibodies were obtained in each of the three patients with generalized lymphadenopathy, the single patient with AIDS, and the six patients with chronic liver disease. To further validate the enzyme-linked immunosorbent assay test, serum samples from the two HTLV-Ill-positive patients with the lowest optical density ratios (2.94 and 2.62; Figure 1) were tested by Western blot analysis and found to be reactive with the viral p41 and p24 bands, thus confirming the positive results of the immunosorbent assay. Similarly, serum samples from three patients with negative results for HTLV-III antibodies and optical density ratios of 1.29, 1.29, and 1.10 were also found to have
1986
The American
Journal
of Medicine
Volume
81
203
RELATION
OF HTLV-III
TABLE II
ANTIBODIES
Analysis
TO T LYMPHOCYTE
of Lymphocyte
Subsets
SUBSET
ABNORMALITIES-HOXIE
in Hemophiliac
Patients
T3 Lymphocytes HTLV-Ill-negative (n = 25) HTLV-Ill-positive (n = 41) p value Results SD.
Percent
August
1988
Antibodies
T4 Number
to HTLV-III
T8
Percent
Number
Percent
Number
T4rfB Ratio
(1,053)
76 (9)
1,412
(940)
43 (7)
765 (402)
32 (7)
596 (469)
1.45 (0.47)
1,632
(820)
78 (9)
1,286
(686)
27 (8)
438 (227)
48 (12)
804 (502)
0.63
as a mean
NS of oercent
NS
oositive
cells
<0.0005
or as a mean
negative results on Western blot testing. In addition, samples from 11 asymptomatic hemophiliac patients with positive immunosorbent assay results for HTLV-III anti,bodies all gave positive results on Western blot analysis. Tests for HTLV-III antibodies were also performed in non-hemophiliac patients with AIDS and in groups without AIDS. Positive immunosorbent assay results were found in all of the 67 patients with AIDS, 77 of 78 (99 percent) of the patients with AIDS-related complex, and in 10 of 22 (45 percent) of the asymptomatic homosexual men. Antibody testing was also performed on 113 control subjects without AIDS-related conditions including multiparous women and patients with systemic lupus erythematosus, idiopathic thrombocytopenic purpura, thrombotic thrombocytopenic purpura, scleroderma, preeclampsia with thrombocytopenia, or malignant disease. Of these 113 samples, three gave positive results in the initial enzymelinked immunosorbent assay. One patient had undergone bone marrow transplantation for acute myelogenous leukemia, and this result was confirmed as a true-positive by Western blot testing. The other two immunosorbent assay results with optical density ratios above 2.0 (3.01 and 2.20) were subsequently negative on repeated enzymelinked immunosorbent assay testing. Six patients with antibodies to HTLV-I had negative results by enzymelinked immunosorbent assay for anti-HTLV-Ill antibodies. Additional studies have shown that of approximately 12,000 healthy blood/plasma donors tested, 0.44 percent have repeatedly positive results by the HTLV-I!I BioEnzaBead assay and negative results by Western blot testing (i.e., false-positive) (personal communication, T. Fitzharris, Litton Bionetics). Relation of HTLV-III Antibody Levels with T-Lymphocyte Subsets. When patients who had positive immunosorbent assay results for HTLV-III antibodies were compared with patients who had negative assay results, significant differences were apparent in T lymphocyte subsets (Table II). This was true for both percents and absolute numbers of T4 cells and for percents of T8 cells. The mean T4/T8 ratio of patients with HTLV-Iii antibodies was 0.63 f 0.31, compared with 1.45 f 0.47 for patients with negative HTLV-III results (p <0.0005). The relation of the optical density ratios to the T4/T8
204
with and wlthout
1,806
NS
are exoressed
ET AL
The American
Journal
of Medicine
Volume
absolute
<0.005 lymphocyte
<0.0005 cou?t/mm3.
NS Numbers
in parentheses
(0.31)
<0.0005 indicate
ratios for the entire hemophiliac population is shown in Figure 2. Of the 45 patients with T4/T8 ratios of 1.1 or less, 39 had positive results for HTLV-III antibodies. In contrast, only two of 18 patients with ratios of more than 1.1 had positive HTLV-III antibody results (p <0.0002). No significant differences were found in total lymphocyte number or numbers of T cells as determined by OKT3positive lymphocytes. Four patients with abnormal T4/T8 ratios of 0.69, 0.74,0.73, and 0.77 had negative immunosorbent assay results for HTLV-III antibodies. These were confirmed as true-negative results both by repeated enzyme-linked immunosorbent assay and by Western blot analysis. Relation of HTLV-III Antibody Status to Transfusion History. Among the 54 patients who had received factor concentrates during the previous year, the mean number of units transfused for patients who proved to be HTLV-III antibody-positive was 91,001 (range, 3,130 to 225,886) compared with 29,568 (range, 1,320 to 108,830) for antibody-negative patients (p <0.0005). Only one of the 10 patients who had not received a transfusion in the previous year was antibody-positive, and this patient had received factor VIII concentrates in 1982. Of the 13 patients who received factor IX concentrates exclusively during the previous year (mean number units transfused, 39,888; range, 3,480 to 123,740) only two (15 percent) were HTLV-III antibody-positive. One patient had hemophilia B, and one was a patient with hemophilia A and a factor VIII inhibitor who had received VIII concentrates in previous years. Among the 41 factor VIII users including the one patient who received both VIII and IX products, 36 (88 percent) were HTLV-III antibody-positive. The difference in the presence of HTLV-III antibodies between factor VIII and IX recipients was highly significant (p <0.00005). Among factor Viii recipients, the mean T4/T8 ratio was 0.64 f 0.32 for the 41 patients with antibodies to HTLVIll and 1.34 f 0.43 for the five patients without antibodies to HTLV-III (p <0.005). For these patients, the mean number of factor VIII units transfused was 9 1,807 (range, 3,130 to 225,886) for HTLV-III antibody-positive patients and 21,517 (range, 1,320 to 75,586) for antibody-negative patients (p
81
RELATION
OF HTLV-III
ANTIBODIES
TO T LYMPHOCYTE
SUBSET
ABNORMALITIES-HOXIE
ET AL
. l .
Figure 2. Comparison of T4/T8 ratios and HTLV-III antibody optical density (00405) ratios from the immunosorbent assay in the hemophiliac patients. Points above the dotted line indicate positive results for HTL V-ill antibodies.
0.0
.
l
l
l
. l
COMMENTS This study demonstrates a strong association between the presence of positive results for HTLV-III antibodies and T lymphocyte subset abnormalities in 66 adults with hemophilia. As shown in previous reports [g-14], alterations in both the relative and absolute numbers of T4 and T8 lymphocytes are common in hemophiliac patients. In this study, they were more apparent in patients who had received factor concentrates than in those who had not. We also detected antibodies to HTLV-III in 41 of 66 patients (62 percent), confirming the reports of antibody in 58 to 92 percent of asymptomatic hemophiliac patients receiving factor concentrates produced in the United States [5-81. As in these reports, antibodies to HTLV-III were found more frequently in patients who had received factor concentrates during the previous year, particularly if they had received factor VIII concentrate. When results of lymphocyte subset analysis were com-
August
.
l.
.
015
transfused and the value for the T4/T8 ratio were not significantly correlated in either antibody-positive or antibody-negative patients. No significant differences in antibody levels were found in factor VIII recipients who had received recalled lots (13 of 15 antibody-positive) compared with those who had not received recalled lots (24 of 36 antibody-positive). Two of four patients who received recalled IX products had positive results. A small but significant difference was found in serum glutamic oxaloacetic transaminase levels (normal range, 7 to 28 IU/ml) of patients with HTLV-III antibodies (mean, 77 f 64) compared with antibody-negative patients (mean, 42 f 23; p <0.05). No differences were found in serum glutamic pyruvic transaminase or bilirubin levels or in the presence of hepatitis B surface antigen or antibody. HTLV-III antibody status was also unrelated to age or race.
l
.
110
115
T4/T8
RATIO
210
215
pared with HTLV-III antibody measurements, striking associations were observed. Patients with antibodies to HTLV-III as determined by enzyme-linked immunosorbent assay had lower relative and absolute numbers of T4 cells and higher percents of T8 cells compared with antibodynegative hemophiliac patients (Table II). Antibody-positive patients had T4/T8 ratios that were less than half those observed in antibody-negative patients (p <0.0005). Abnormalities in parameters of cellular immunity have been noted in other reports on asymptomatic hemophiliac patients, including T lymphocyte subsets, delayed hypersensitivity, and in vitro lymphocyte mitogenesis [8-141. However, the relationship of these changes to AIDS has been questioned in view of the detection of similar changes in patients with hemophilia who received factor concentrates obtained from donor populations with a low prevalence of AIDS [ 12- 141. An early study of 25 asymptomatic hemophiliac patients found no association between T lymphocyte subset abnormalities and the presence of HTLV-III antibodies [5]. The striking association between the presence of HTLV-III antibodies and T4/T8 ratio abnormalities in our study may relate to the larger sample size, the sensitivity of the method used to detect HTLV-III antibodies, or the time of testing in relation to viral exposure. Although the presence of lymphocyte subset abnormalities in hemophiliac patients with HTLV-III antibodies may be the result of viral infection, other interpretations are possible. It is conceivable that these immunologic alterations preceded or even predisposed to subsequent viral infection. A direct cause-and-effect relationship cannot be determined from this study. However, the strong association of immune abnormalities and seroreactivity to this virus does suggest that infection by HTLV-III has occurred, rather than a passive exposure to viral antigens contained in factor concentrates. Studies are in progress
1966
The American
Journal
of Medicine
Volume
61
206
RELATION
OF HTLV-III
ANTIBODIES
TO T LYMPHOCYTE
SUBSET
ABNORMALITIES-HOXIE
at several centers to determine the frequency with which HTLV-III can be cultured from seropositive hemophiliac patients [6]. These studies may help to resolve this issue, as well as determine how many of these patients if any, are persistently infected with this virus. The clinical significance of this finding is uncertain, particularly in view of the prevalence of HTLV-III antibodies, and presumably infection, in healthy members of groups at risk for AIDS, including intravenous drug-users [ 181 and asymptomatic homosexual men [ 19-2 11. Transient reductions in T4/T8 ratios for as long as several months have been shown to occur following a variety of viral infections [22,23]. The mechanism by which HTLVIll produces immunodeficiency as well as other factors that may potentiate this process are, at present, unknown. A prospective evaluation of this hemophiliac population as well as other seropositive persons should help to answer these crucial questions and better define the natural history of HTLV-III infection.
ET AL
ADDENDUM Since submission of this article, several reports have appeared that confirm the strong correlation of antibodies to HTLV-III with T lymphocyte subset abnormalities in patients with hemophilia [24-281.
ACKNOWLEDGMENT We gratefully acknowledge T. Fitzharris and J. Holper of Litton Bionetics, Charleston, South Carolina, for providing testing kits for HTLV-III antibody determination and N. Pomato of the Litton Institute of Applied Biotechnology, Rockville, Maryland, for performing Western blot analysis of serum samples. We also thank D. Cines, G. Henle, W. Henle, and E. Davila for providing serum samples, D. Matthews and B. Haggarty for technical assistance, G. Coffey for secretarial assistance, and R. A. Cooper for reviewing this manuscript.
REFERENCES 1.
2.
3.
7.
8.
9.
10.
11.
206
Centers for Disease Control: Update: acquired immunodeficiency syndrome (AIDS) in persons with hemophilia. Morbid Mortal Weekly Rep 1984; 33: 589-591. Gallo RC, Salahuddin SZ, Popovic M, et al: Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science 1984; 22: 500-503. Barre’-Sinoussi F, Chermann JC, Rey F, et al: Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immunodeficiency syndrome (AIDS). Science 1983; 220: 868-871. Levy JA, Hoffman AD, Kramer SM, et al: Isolation of lymphocytopathic retroviruses from San Francisco patients with AIDS. Science 1984; 225: 840-842. Ramsey RB, Palmer EL, McDougal JS, et al: Antibody to lymphadenopathy-associated virus in hemophiliacs with and without AIDS. Lancet 1984; II: 397-398. Goedert JJ, Sarngadharan MG, Eyster E, et al: Antibodies reactive with human T cell leukemia viruses in the serum of hemophiliacs receiving factor VIII concentrate. Blood 1985; 65: 492-495. Kitchen LW, Barin F, Sullivan JL, et al: Aetiology of AIDSantibodies to human T-cell leukemia virus (type Ill) in hemophiliacs. Nature 1984; 312: 367-369. Tsoukas C, Gervais F, Shuster J, et al: Association of HTLVIll antibodies and cellular immune status of hemophiliacs. N Engl J Med 1984; 311: 1514-1515. Lederman MM, Ratnoff OD, Scillian JJ, et al: Impaired cell mediated immunity in patients with classic hemophilia. N Engl J Med 1983; 308: 79-83. Menitove JE, Aster RH, Casper JT, et al: T-lymphocyte subpopulations in patients with classic hemophilia treated with cryoprecipitate and lyophylized concentrates. N Engl J Med 1983; 308: 83-86. Weintraub PS, Koerper MA, Addiego JE, et al: Immunologic abnormalities in patients with hemophilia A. J Pediatr
August
1966
The
American
Journal
of Medicine
Volume
12.
13.
14.
15.
18.
17.
18.
19.
20.
21.
61
1983; 103: 692-695. Froebel KS, Madhok R, Forbes CD, et al: Immunological abnormalities in hemophilia: are they caused by American factor VIII concentrate? Br Med J 1983; 287: 1091-1093. Rickard KA, Joshua DE, Campbell J, et al: Absence of AIDS in Australia treated from an entirely voluntary blood donor system. Lancet 1983; II: 50-5 1. Carr R, Veitch SE, Edmond E, et al: Abnormalities of circulating lymphocyte subsets in haemophiliacs in an AIDSfree population. Lancet 1984; I: 1431-1434. Hoffman RA, Kung PL, Hansen WP, et al: Simple and rapid measurement of human T-lymphocytes and their subclasses in peripheral blood. Proc Natl Acad Sci USA 1980; 77: 4914-4917. Safai B, Sarngadharan MG, Groopman JE, et al: Seroepidemiological studies of human T-lymphotropic retrovirus type Ill in acquired immunodeficiency syndrome. Lancet 1984; I: 1438-1441. Smith KO, Gehle WD: Magnetic transfer devices for use in solid phase radioimmunoassays and enzyme linked immunosorbent assays. J Infect Dis 1977; 136 (suppl): 5329-5336. Spira TJ, DesJarlais DC, Marmor M, et al: Prevalence of antibody to lymphadenopathy-associated virus among drug-detoxification patients in New York. N Engl J Med 1984; 311: 467-468. Melbye M, Biggar RJ, Ebbesen P, et al: Seroepidemiology of HTLV-III antibody in Danish homosexual men: prevalence, transmission, and disease outcome. Br Med J 1984; 289: 573-577. Weiss SH, Goedert JJ, Sarngadharan MG, et al: Screening test for HTLV-III (AIDS agent) antibodies: specificity, sensitivity, and applications. JAMA 1985; 253: 221-225. Goedert JJ, Sarngadharan MG, Biggar RJ, et al: Determinants of retrovirus (HTLV-III) antibody and immunodefi-
RELATION
22.
23.
24.
25.
OF HTLV-III
ANTIBODIES
ciency conditions in homosexual men. Lancet 1984; II: 711-718. DeWaele M, Thielmans C, Van Camp BKG: Characterization of immunoregulatory T cells in EBV-induced infectious: mononucleosis by monoclonal antibodies. N Engl J Med 1981; 304: 460-462. Carney WP, Rubin RH, Hoffman RA, et al: Analysis of Tlymphocyte subsets in cytomegalovirus mononucleosis. J lmmunol 1981; 126: 2114-2116. Eyster EM, Goedert JJ, Sarngadharan MG, et al: Development and early natural history of HTLV-III antibodies in persons with hemophilia. JAMA 1985; 253: 22192223. Gjerset GF, McGrady G, Counts RB, et al: Lymphadenopathy-
August
26.
27.
28.
1988
TO T LYMPHOCYTE
SUBSET
ABNORMALITIES-HOXIE
ET AL
associated virus antibodies and T cells in hemophiliacs treated with cryoprecipitate or concentrate. Blood 1985; 66: 718-720. Lederman MM, Ratnoff OD, Evatt BL, et al: Acquisition of antibody to lymphadenopathy-associated virus in patients with classic hemophilia (factor VIII deficiency). Ann Intern Med 1985; 102: 753-757. Kreiss JK, Kitchen LW, Prince HE, et al: Human T cell leukemia virus type Ill antibody, lymphadenopathy, and acquired immune deficiency syndrome in hemophiliac subjects. Am J Med 1986; 80: 345-350. AIDS-Hemophilia French Study Group: Natural history of primary infection with LAV in multitransfused patients. Blood 1986; 68: 89-94.
The American
Journal
of Medicine
Volume
81
207
JAMES A. HOXIE, M.D. JEROME L. RACKOWSKI ANDREW J. CEDARBAUM SHELLEY HURWITZ, M.S. PATRICIA M. CATALANO, M.D. Philadelphia,
Pennsylvania
From the Hematology-Oncology Section, Department of Medicine, Hospital of the University of Pennsylvania, the Biostatistics Program, University of Pennsylvania Cancer Center, and the Cardeza Foundation Hemophilia Center, Department of Medicine, Jefferson University Hospital, Philadelphia, Pennsylvania. Dr. Hoxie is a Special Fellow of the Leukemia Society of America. This work was supported by a grant from the W. W. Smith Charitable Trust and the Hemophilia Program of the Commonwealth of Pennsylvania. Requests for reprints should be addressed to Dr. James A. Hoxie, Hematology-Oncology Section, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pennsylvania 19104. Manuscript accepted July 10, 1985. A continuing medical education quiz on this ar-ticle (one hour of Category 1 credit) appears on page Al 11 of this issue.
The relationship of T lymphocyte subset abnormalities and the presence of antibodies to the human T lymphotropic virus type Ill (HTLV-III) was evaluated in 66 adult patients with hemophilia. Positive test results for antibodies to HTLV-III were observed in 62 percent of patients with hemophilia A and 9 percent of patients with hemophilia 6. Patients with HTLV-III antibodies had lower percentages and numbers of T helper (T4) cells and increased percentages of T suppressor (16) cells compared with percents in patients without antibodies to HTLV-III. The mean T4/T6 ratio for antibody-negative patients was 1.45 compared with 0.65 for antibody-positive patients (p
August
1966
The American
Journal
of Medicine
Volume
61
201
RELATION
OF HTLV-III
TABLE
I
Analysis
ANTIBODIES
TO T LYMPHOCYTE
of Lymphocyte
Subsets
SUBSET
ABNORMALITIES-HOXIE
in Hemophiliac
Patients
Type of Hemophilia (n i55) Lymphocytes
(n Jll) 1,645
p value
Transfusion History* Factor IX No Transfusions (n = 13) (n = 10)
Factor VIII (n = 40)
NS
1,721 ww NS
183
:83
(556)
ET AL
1,637 (584) NS
0380)
1,512
Control Subjects (n = 46) 2,014 (678)
NS
T3 Percent
p value Number p value
NS
NS
NS
NS
NS
1,342 (839) NS
1,295
1,367
(476)
(885) NS
1,319 (517) NS
1,096 (759) NS
NS 689
<0.0001 512 (349) <0.0001
;‘1, NS 639 (239) NS
NS
1,493
(516)
T4 Percent <0.0001 536 (357) <0.0001
p value Number p value
(232) NS
NS 659
947
VW
(346)
ii
F-G
NS
T8 Percent
(443)
p value Number
<0.0001 759 (531) <0.05 0.86 (0.54) <0.0001
p value T4lT8
ratio p value
NS
$63)
NS
554 (204) NS
<0.0001 799 (566) <0.05
NS
627 (343) NS
451 (311) NS
550
1.32 (0.43) <0.05
0.73 (0.41) <0.0001
1.21 (0.52) <0.005
1.54 (0.59) NS
1.81 (0.51)
(211)
Results are expressed as a mean of percent positive cells or as a mean absolute lymphocyte count/mm3. Numbers in parentheses indicate SD; p values correspond to comparisons with control group. * The factor VIII and factor IX recipient categories do not include the one patient who received both VIII and IX products. Two patients with hemophilia A, whose transfusion records were not available, are omitted from this section.
PATIENTS AND METHODS Patient Population. Blood specimens for lymphocyte subset and serologic analysis were obtained from 66 adult hemophiliac patients from May 1 to September 30, 1984. They included 55 patients with hemophilia A and 11 patients with hemophilia B. The mean age of this study group was 32 (range, 15 to 65); there were 53 Caucasians and 13 blacks. Of the 66 patients studied, one had AIDS compiicated by Prieumocystis carinii pneumonia and three had chronic unexplained lymphadenopathy. The other 62 patients had no symptoms of chronic lymphadenopathy or weight loss. Five had chronic liver disease thought to be related to hepatitis B. One patient was a homosexual, and one had a history of intravenous drug use. Records of blood product transfusion could be obtained for 64 of the 66 patients: two patients with hemophilia A lacked documentation of their transfusion history. During the previous year, 54 of these 64 patients had received factor concentrates. Of these, 41 received factor VIII concentrates (mean number of units transfused, 82,054; range, 1,320 to 225,686), and 14 received factor IX concentrates (mean number of units transfused, 39,143; range, 3,480 to 123,740). The factor IX recipients included four patients
202
August
1986
The American
factor Viii inhibitors and one patient who received both factor VIII and IX concentrates. Among the 54 patients receiving factor concentrates, 19 patients (15 fatittar VIII recipients and four factor IX recipients) received factor concentrates from lots that were subsequently recalled because two of the donors whose blood was used in the preparation of these products demonstrated AIDS foliowing donation. Ten patients had not received factor concentrates during the year prior to this study. Of these 10 patients, three had received factor Viii concentrates from 1982 to 1983, totaling 68,510, 16,260, and 14,860 units. The other seven patients had not received any.factor concentrates. One of these seven patients had received cryoprecipitate (48 bags) in 1983. Additional serum samples obtained for HTLV-III antibody determination included samples from 67 patients with AIDS, 78 male homosexuals with, Chronic unexplained lymphadenopathy or a wasting syndrome, and 22 asymptomatic homosexual men in the Philadelphia area. Testing was also performed on samples from 1 j3 control subjects without AIDS-related conditions including nine multiparous women, 21 patients with systemic lupus erythematosus, 25 with idiopathic thrombocytoperiic purpura, seven with scieroderma, seven with thrombotic thrombocy-topenic purpuwith
Journal
of Medicine
Volume
61
RELATION
OF HTLV-III
ANTIBODIES
ra, nine with pre-eclampsia and thrombocytopenia, and 35 with malignant disease including three with non-AIDS Kaposi’s sarcoma. Serum samples from six patients known to have antibodies to HTLV-I were also evaluated. Control blood samples for lymphocyte subset analysis were obtained from 46 healthy laboratory workers. The mean age of this control group was 36 (range, 23 to 63) and included 22 males and 24 females. Lymphocyte Subset Analysis. Analysis of T lymphocyte subsets was performed using monoclonal antibodies OKT3, OKT4, and OKT8 (Ortho Diagnostic Systems; Raritan, New Jersey). Optimal titrations of antibodies were added directly to fresh samples of whole blood, and, after red blood cells were lysed, lymphocytes were analyzed on a Spectrum-Ill fluorescence celi analyzer (Ortho Diagnostic Systems), as described by Hoffman et al [ 151. HTLV-III Antibody Determlnatlon. Antibodies to HTLV-III were detected by enzyme-linked immunosorbent assay using the Bio-EnzaBead kit provided by Litton Bionetics (Charleston, South Carolina). Serum samples were stored at -7O’C and tested at a 1 to 900 dilution. The optical density at 405 nM of each tested sample was determined with an LBI 30 spectrophotometer (Litton Bionetics) and compared with three negative control samples (provided by Litton ,Bionetics) tested concurrently. Test results were scored as positive when the ratio of the optical density of the tested sample to the, average optical density of the negative control samples was more than 2.0. Selected samples were evaluated by repeating the enzyme-linked immunosorbent assay and by Western blot analysis using purified HTLV-III antigens (performed by Litton Bionetics)
[1’31. Data An&&. Group comparisons for continuous variables included the t test or Wilcoxon rank sum test, and analysis of variance with the Sheffe procedure. Categorical comparisons used the chi-square test with the Yates’ correction. Variable independence was tested by Pearson product moment or Spearman rank correlation when appropriate. All tests were two-tailed, and resulting p values above 0.05 were considered not statistically significant. Data are expressed as the mean f standard deviation (SD). RESULTS Analysis of T Lymphocyte Subsets. Patients with hemophilia A were found to have lower T4/T8 ratios than those in normal control subjects (Table I). Significant decreases were found in both the percent and absolute number of T4 cells, and significant increases were present in both the percent and absolute number of T8 cells. No differences were observed in the total number of T lymphocytes as determined by the OKT3-reactive cells. As shown in Table I, these alterations in T4 and T8 cells were more apparent in patients receiving factor concentrate transfusions, particularly factor VIII. Among the 44 patients with hemophilia A who had received transfusions within the previous year, the mean T4/T8 ratio was 0.73 f 0,43, compared with 1.59 f 0.60 for the nine patients with hemophilia A who had not received transfusions
August
TO T LYMPHOCYTE
0
2
SUBSET
4
ABNORMALITIES-HOXIE
6
OD 405
8
ET AL
IO
12
RATIO
Figure 1. Frequency histogram of HTLV-III antibody resuits in hemophiliac patients. Results are expressed as the optical density ratio at 405 nM (00405) of the patient sample result to the average result of three negative control samples.
within the previous year (p <0.0005). Abnormalities in lymphocyte subsets were less apparent in patients with hemophilia 8. T4/T8 ratios were below 1.0 in 38 of 55 patients (69 percent) with hemophilia A and two of 11 patients (18 percent) with hemophilia B, whereas T4/T8 ratios were above 1.0 in all of the 46 normal control subjects. HTLV-III Antibody Determination. Antibodies to HTLVIll were detected using the Bio-EnzaBead enzyme-linked immunosorbent assay, which employs purified viral antigens coupled to magnetic beads [ 171. The results were designated “positive” for antibodies to HTLV-III if the optical density at 405 nM was more than two times the average of three negative control samples tested concurrently. A bimodal distribution was seen among hemophiliacs (Figure 1). Positive values were observed in 41 of 66 (62 percent) of the patients with hemophilia, including 40 of 55 (73 percent) with hemophilia A and one of 11 (9 percent) with hemophilia B. The mean ratio of the optical density value in patient serum samples relative to the average optical density value of the negative control samples was 7.9 f 2.5 for the 41 HTLV-Ill-positive patients and 1.0 f 0.1 for the 25 HTLV-Ill-negative patients. Positive results for HTLV-III antibodies were obtained in each of the three patients with generalized lymphadenopathy, the single patient with AIDS, and the six patients with chronic liver disease. To further validate the enzyme-linked immunosorbent assay test, serum samples from the two HTLV-Ill-positive patients with the lowest optical density ratios (2.94 and 2.62; Figure 1) were tested by Western blot analysis and found to be reactive with the viral p41 and p24 bands, thus confirming the positive results of the immunosorbent assay. Similarly, serum samples from three patients with negative results for HTLV-III antibodies and optical density ratios of 1.29, 1.29, and 1.10 were also found to have
1986
The American
Journal
of Medicine
Volume
81
203
RELATION
OF HTLV-III
TABLE II
ANTIBODIES
Analysis
TO T LYMPHOCYTE
of Lymphocyte
Subsets
SUBSET
ABNORMALITIES-HOXIE
in Hemophiliac
Patients
T3 Lymphocytes HTLV-Ill-negative (n = 25) HTLV-Ill-positive (n = 41) p value Results SD.
Percent
August
1988
Antibodies
T4 Number
to HTLV-III
T8
Percent
Number
Percent
Number
T4rfB Ratio
(1,053)
76 (9)
1,412
(940)
43 (7)
765 (402)
32 (7)
596 (469)
1.45 (0.47)
1,632
(820)
78 (9)
1,286
(686)
27 (8)
438 (227)
48 (12)
804 (502)
0.63
as a mean
NS of oercent
NS
oositive
cells
<0.0005
or as a mean
negative results on Western blot testing. In addition, samples from 11 asymptomatic hemophiliac patients with positive immunosorbent assay results for HTLV-III anti,bodies all gave positive results on Western blot analysis. Tests for HTLV-III antibodies were also performed in non-hemophiliac patients with AIDS and in groups without AIDS. Positive immunosorbent assay results were found in all of the 67 patients with AIDS, 77 of 78 (99 percent) of the patients with AIDS-related complex, and in 10 of 22 (45 percent) of the asymptomatic homosexual men. Antibody testing was also performed on 113 control subjects without AIDS-related conditions including multiparous women and patients with systemic lupus erythematosus, idiopathic thrombocytopenic purpura, thrombotic thrombocytopenic purpura, scleroderma, preeclampsia with thrombocytopenia, or malignant disease. Of these 113 samples, three gave positive results in the initial enzymelinked immunosorbent assay. One patient had undergone bone marrow transplantation for acute myelogenous leukemia, and this result was confirmed as a true-positive by Western blot testing. The other two immunosorbent assay results with optical density ratios above 2.0 (3.01 and 2.20) were subsequently negative on repeated enzymelinked immunosorbent assay testing. Six patients with antibodies to HTLV-I had negative results by enzymelinked immunosorbent assay for anti-HTLV-Ill antibodies. Additional studies have shown that of approximately 12,000 healthy blood/plasma donors tested, 0.44 percent have repeatedly positive results by the HTLV-I!I BioEnzaBead assay and negative results by Western blot testing (i.e., false-positive) (personal communication, T. Fitzharris, Litton Bionetics). Relation of HTLV-III Antibody Levels with T-Lymphocyte Subsets. When patients who had positive immunosorbent assay results for HTLV-III antibodies were compared with patients who had negative assay results, significant differences were apparent in T lymphocyte subsets (Table II). This was true for both percents and absolute numbers of T4 cells and for percents of T8 cells. The mean T4/T8 ratio of patients with HTLV-Iii antibodies was 0.63 f 0.31, compared with 1.45 f 0.47 for patients with negative HTLV-III results (p <0.0005). The relation of the optical density ratios to the T4/T8
204
with and wlthout
1,806
NS
are exoressed
ET AL
The American
Journal
of Medicine
Volume
absolute
<0.005 lymphocyte
<0.0005 cou?t/mm3.
NS Numbers
in parentheses
(0.31)
<0.0005 indicate
ratios for the entire hemophiliac population is shown in Figure 2. Of the 45 patients with T4/T8 ratios of 1.1 or less, 39 had positive results for HTLV-III antibodies. In contrast, only two of 18 patients with ratios of more than 1.1 had positive HTLV-III antibody results (p <0.0002). No significant differences were found in total lymphocyte number or numbers of T cells as determined by OKT3positive lymphocytes. Four patients with abnormal T4/T8 ratios of 0.69, 0.74,0.73, and 0.77 had negative immunosorbent assay results for HTLV-III antibodies. These were confirmed as true-negative results both by repeated enzyme-linked immunosorbent assay and by Western blot analysis. Relation of HTLV-III Antibody Status to Transfusion History. Among the 54 patients who had received factor concentrates during the previous year, the mean number of units transfused for patients who proved to be HTLV-III antibody-positive was 91,001 (range, 3,130 to 225,886) compared with 29,568 (range, 1,320 to 108,830) for antibody-negative patients (p <0.0005). Only one of the 10 patients who had not received a transfusion in the previous year was antibody-positive, and this patient had received factor VIII concentrates in 1982. Of the 13 patients who received factor IX concentrates exclusively during the previous year (mean number units transfused, 39,888; range, 3,480 to 123,740) only two (15 percent) were HTLV-III antibody-positive. One patient had hemophilia B, and one was a patient with hemophilia A and a factor VIII inhibitor who had received VIII concentrates in previous years. Among the 41 factor VIII users including the one patient who received both VIII and IX products, 36 (88 percent) were HTLV-III antibody-positive. The difference in the presence of HTLV-III antibodies between factor VIII and IX recipients was highly significant (p <0.00005). Among factor Viii recipients, the mean T4/T8 ratio was 0.64 f 0.32 for the 41 patients with antibodies to HTLVIll and 1.34 f 0.43 for the five patients without antibodies to HTLV-III (p <0.005). For these patients, the mean number of factor VIII units transfused was 9 1,807 (range, 3,130 to 225,886) for HTLV-III antibody-positive patients and 21,517 (range, 1,320 to 75,586) for antibody-negative patients (p
81
RELATION
OF HTLV-III
ANTIBODIES
TO T LYMPHOCYTE
SUBSET
ABNORMALITIES-HOXIE
ET AL
. l .
Figure 2. Comparison of T4/T8 ratios and HTLV-III antibody optical density (00405) ratios from the immunosorbent assay in the hemophiliac patients. Points above the dotted line indicate positive results for HTL V-ill antibodies.
0.0
.
l
l
l
. l
COMMENTS This study demonstrates a strong association between the presence of positive results for HTLV-III antibodies and T lymphocyte subset abnormalities in 66 adults with hemophilia. As shown in previous reports [g-14], alterations in both the relative and absolute numbers of T4 and T8 lymphocytes are common in hemophiliac patients. In this study, they were more apparent in patients who had received factor concentrates than in those who had not. We also detected antibodies to HTLV-III in 41 of 66 patients (62 percent), confirming the reports of antibody in 58 to 92 percent of asymptomatic hemophiliac patients receiving factor concentrates produced in the United States [5-81. As in these reports, antibodies to HTLV-III were found more frequently in patients who had received factor concentrates during the previous year, particularly if they had received factor VIII concentrate. When results of lymphocyte subset analysis were com-
August
.
l.
.
015
transfused and the value for the T4/T8 ratio were not significantly correlated in either antibody-positive or antibody-negative patients. No significant differences in antibody levels were found in factor VIII recipients who had received recalled lots (13 of 15 antibody-positive) compared with those who had not received recalled lots (24 of 36 antibody-positive). Two of four patients who received recalled IX products had positive results. A small but significant difference was found in serum glutamic oxaloacetic transaminase levels (normal range, 7 to 28 IU/ml) of patients with HTLV-III antibodies (mean, 77 f 64) compared with antibody-negative patients (mean, 42 f 23; p <0.05). No differences were found in serum glutamic pyruvic transaminase or bilirubin levels or in the presence of hepatitis B surface antigen or antibody. HTLV-III antibody status was also unrelated to age or race.
l
.
110
115
T4/T8
RATIO
210
215
pared with HTLV-III antibody measurements, striking associations were observed. Patients with antibodies to HTLV-III as determined by enzyme-linked immunosorbent assay had lower relative and absolute numbers of T4 cells and higher percents of T8 cells compared with antibodynegative hemophiliac patients (Table II). Antibody-positive patients had T4/T8 ratios that were less than half those observed in antibody-negative patients (p <0.0005). Abnormalities in parameters of cellular immunity have been noted in other reports on asymptomatic hemophiliac patients, including T lymphocyte subsets, delayed hypersensitivity, and in vitro lymphocyte mitogenesis [8-141. However, the relationship of these changes to AIDS has been questioned in view of the detection of similar changes in patients with hemophilia who received factor concentrates obtained from donor populations with a low prevalence of AIDS [ 12- 141. An early study of 25 asymptomatic hemophiliac patients found no association between T lymphocyte subset abnormalities and the presence of HTLV-III antibodies [5]. The striking association between the presence of HTLV-III antibodies and T4/T8 ratio abnormalities in our study may relate to the larger sample size, the sensitivity of the method used to detect HTLV-III antibodies, or the time of testing in relation to viral exposure. Although the presence of lymphocyte subset abnormalities in hemophiliac patients with HTLV-III antibodies may be the result of viral infection, other interpretations are possible. It is conceivable that these immunologic alterations preceded or even predisposed to subsequent viral infection. A direct cause-and-effect relationship cannot be determined from this study. However, the strong association of immune abnormalities and seroreactivity to this virus does suggest that infection by HTLV-III has occurred, rather than a passive exposure to viral antigens contained in factor concentrates. Studies are in progress
1966
The American
Journal
of Medicine
Volume
61
206
RELATION
OF HTLV-III
ANTIBODIES
TO T LYMPHOCYTE
SUBSET
ABNORMALITIES-HOXIE
at several centers to determine the frequency with which HTLV-III can be cultured from seropositive hemophiliac patients [6]. These studies may help to resolve this issue, as well as determine how many of these patients if any, are persistently infected with this virus. The clinical significance of this finding is uncertain, particularly in view of the prevalence of HTLV-III antibodies, and presumably infection, in healthy members of groups at risk for AIDS, including intravenous drug-users [ 181 and asymptomatic homosexual men [ 19-2 11. Transient reductions in T4/T8 ratios for as long as several months have been shown to occur following a variety of viral infections [22,23]. The mechanism by which HTLVIll produces immunodeficiency as well as other factors that may potentiate this process are, at present, unknown. A prospective evaluation of this hemophiliac population as well as other seropositive persons should help to answer these crucial questions and better define the natural history of HTLV-III infection.
ET AL
ADDENDUM Since submission of this article, several reports have appeared that confirm the strong correlation of antibodies to HTLV-III with T lymphocyte subset abnormalities in patients with hemophilia [24-281.
ACKNOWLEDGMENT We gratefully acknowledge T. Fitzharris and J. Holper of Litton Bionetics, Charleston, South Carolina, for providing testing kits for HTLV-III antibody determination and N. Pomato of the Litton Institute of Applied Biotechnology, Rockville, Maryland, for performing Western blot analysis of serum samples. We also thank D. Cines, G. Henle, W. Henle, and E. Davila for providing serum samples, D. Matthews and B. Haggarty for technical assistance, G. Coffey for secretarial assistance, and R. A. Cooper for reviewing this manuscript.
REFERENCES 1.
2.
3.
7.
8.
9.
10.
11.
206
Centers for Disease Control: Update: acquired immunodeficiency syndrome (AIDS) in persons with hemophilia. Morbid Mortal Weekly Rep 1984; 33: 589-591. Gallo RC, Salahuddin SZ, Popovic M, et al: Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science 1984; 22: 500-503. Barre’-Sinoussi F, Chermann JC, Rey F, et al: Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immunodeficiency syndrome (AIDS). Science 1983; 220: 868-871. Levy JA, Hoffman AD, Kramer SM, et al: Isolation of lymphocytopathic retroviruses from San Francisco patients with AIDS. Science 1984; 225: 840-842. Ramsey RB, Palmer EL, McDougal JS, et al: Antibody to lymphadenopathy-associated virus in hemophiliacs with and without AIDS. Lancet 1984; II: 397-398. Goedert JJ, Sarngadharan MG, Eyster E, et al: Antibodies reactive with human T cell leukemia viruses in the serum of hemophiliacs receiving factor VIII concentrate. Blood 1985; 65: 492-495. Kitchen LW, Barin F, Sullivan JL, et al: Aetiology of AIDSantibodies to human T-cell leukemia virus (type Ill) in hemophiliacs. Nature 1984; 312: 367-369. Tsoukas C, Gervais F, Shuster J, et al: Association of HTLVIll antibodies and cellular immune status of hemophiliacs. N Engl J Med 1984; 311: 1514-1515. Lederman MM, Ratnoff OD, Scillian JJ, et al: Impaired cell mediated immunity in patients with classic hemophilia. N Engl J Med 1983; 308: 79-83. Menitove JE, Aster RH, Casper JT, et al: T-lymphocyte subpopulations in patients with classic hemophilia treated with cryoprecipitate and lyophylized concentrates. N Engl J Med 1983; 308: 83-86. Weintraub PS, Koerper MA, Addiego JE, et al: Immunologic abnormalities in patients with hemophilia A. J Pediatr
August
1966
The
American
Journal
of Medicine
Volume
12.
13.
14.
15.
18.
17.
18.
19.
20.
21.
61
1983; 103: 692-695. Froebel KS, Madhok R, Forbes CD, et al: Immunological abnormalities in hemophilia: are they caused by American factor VIII concentrate? Br Med J 1983; 287: 1091-1093. Rickard KA, Joshua DE, Campbell J, et al: Absence of AIDS in Australia treated from an entirely voluntary blood donor system. Lancet 1983; II: 50-5 1. Carr R, Veitch SE, Edmond E, et al: Abnormalities of circulating lymphocyte subsets in haemophiliacs in an AIDSfree population. Lancet 1984; I: 1431-1434. Hoffman RA, Kung PL, Hansen WP, et al: Simple and rapid measurement of human T-lymphocytes and their subclasses in peripheral blood. Proc Natl Acad Sci USA 1980; 77: 4914-4917. Safai B, Sarngadharan MG, Groopman JE, et al: Seroepidemiological studies of human T-lymphotropic retrovirus type Ill in acquired immunodeficiency syndrome. Lancet 1984; I: 1438-1441. Smith KO, Gehle WD: Magnetic transfer devices for use in solid phase radioimmunoassays and enzyme linked immunosorbent assays. J Infect Dis 1977; 136 (suppl): 5329-5336. Spira TJ, DesJarlais DC, Marmor M, et al: Prevalence of antibody to lymphadenopathy-associated virus among drug-detoxification patients in New York. N Engl J Med 1984; 311: 467-468. Melbye M, Biggar RJ, Ebbesen P, et al: Seroepidemiology of HTLV-III antibody in Danish homosexual men: prevalence, transmission, and disease outcome. Br Med J 1984; 289: 573-577. Weiss SH, Goedert JJ, Sarngadharan MG, et al: Screening test for HTLV-III (AIDS agent) antibodies: specificity, sensitivity, and applications. JAMA 1985; 253: 221-225. Goedert JJ, Sarngadharan MG, Biggar RJ, et al: Determinants of retrovirus (HTLV-III) antibody and immunodefi-
RELATION
22.
23.
24.
25.
OF HTLV-III
ANTIBODIES
ciency conditions in homosexual men. Lancet 1984; II: 711-718. DeWaele M, Thielmans C, Van Camp BKG: Characterization of immunoregulatory T cells in EBV-induced infectious: mononucleosis by monoclonal antibodies. N Engl J Med 1981; 304: 460-462. Carney WP, Rubin RH, Hoffman RA, et al: Analysis of Tlymphocyte subsets in cytomegalovirus mononucleosis. J lmmunol 1981; 126: 2114-2116. Eyster EM, Goedert JJ, Sarngadharan MG, et al: Development and early natural history of HTLV-III antibodies in persons with hemophilia. JAMA 1985; 253: 22192223. Gjerset GF, McGrady G, Counts RB, et al: Lymphadenopathy-
August
26.
27.
28.
1988
TO T LYMPHOCYTE
SUBSET
ABNORMALITIES-HOXIE
ET AL
associated virus antibodies and T cells in hemophiliacs treated with cryoprecipitate or concentrate. Blood 1985; 66: 718-720. Lederman MM, Ratnoff OD, Evatt BL, et al: Acquisition of antibody to lymphadenopathy-associated virus in patients with classic hemophilia (factor VIII deficiency). Ann Intern Med 1985; 102: 753-757. Kreiss JK, Kitchen LW, Prince HE, et al: Human T cell leukemia virus type Ill antibody, lymphadenopathy, and acquired immune deficiency syndrome in hemophiliac subjects. Am J Med 1986; 80: 345-350. AIDS-Hemophilia French Study Group: Natural history of primary infection with LAV in multitransfused patients. Blood 1986; 68: 89-94.
The American
Journal
of Medicine
Volume
81
207