| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Service dImmunologie Clinique (V.J., J.T., J.-P.V.) and Laboratoire Central des Radio-Isotopes (M.I.), Hôpital Necker, 75743 Paris; Service dEndocrinologie (A.P., F.S.) and Service de Maladies Infectieuses et Tropicales (B.H.), Centre Hospitalier Universitaire, 25030 Besançon; Service dImmunologie Clinique, Hôpital Broussais (M.D.K.), 75674 Paris; and Unité de Maladies Infectieuses, Hôpital Saint-Joseph (J.G.), 75674 Paris, France
Address all correspondence and requests for reprints to: Jean-Paul Viard, M.D., Service dImmunologie Clinique, Hôpital Necker, 149 rue de Sèvres, 75743 Paris Cedex 15, France.
| Abstract |
|---|
|
|
|---|
Five patients (median age, 41 yr) were diagnosed with GD after 20 (range, 1422) months on HAART on the basis of clinical and biological hyperthyroidism, diffuse hyperfixation of thyroid scan, and the presence of anti-TSH receptor (anti-TSHR) antibodies (Ab). GD was diagnosed several months after the plasma HIV ribonucleic acid load became undetectable, when the CD4+ cell count had risen from 14 (range, 062) to 340 (range, 163460) x 106 cells/L. Antithyroid peroxidase (anti-TPO) and anti-TSHRAb appeared 14 (range, 918) and 14 (range, 1120) months after starting HAART and 12 (range, 615) and 11 (range, 917) months after the increase in CD4+ cells. In 3 patients, TPOAb preceded TSHRAb by 310 months. No other autoantibodies were detected. Thyroid antibodies were absent in a group of 55 HIV-1-positive patients with comparable response to HAART and no symptoms of hyperthyroidism (cross-sectional study).
Thyroid-specific autoimmunity can occur upon immune restoration with HAART. Our observations suggest a relationship between thymus-dependent immune reconstitution after immunosuppression and autoimmunity and may provide insight into the pathophysiology of GD.
| Introduction |
|---|
|
|
|---|
We recently reported on the delayed occurrence of Graves disease (GD) in three deeply immunocompromised patients after recovery of high CD4+ cell counts on HAART (6). Hereafter, we extended this finding and analyzed the relationship between immune restoration on HAART and the kinetics of thyroid autoantibodies and GD in five such patients.
| Subjects and Methods |
|---|
|
|
|---|
Five patients, four men and one woman, aged 2442 yr, from four distinct HIV units who developed GD, including three cases previously reported (6), were retrospectively investigated after informed consent was obtained. They all met the 1993 revised Centers for Disease Control criteria for acquired immune deficiency syndrome, and their median nadir CD4+ cell count before HAART was 14 x 106 cells/L (range, 062). None had a personal or family history of thyroid or autoimmune disease or had been given drugs known to interfere with thyroid function. As patients developed hyperthyroidism, blood samples were taken at entry for T4, TSH, antithyroid peroxidase (anti-TPO), and anti-TSH receptor (TSHR) autoantibody (Ab) determinations. A diagnostic 123I-labeled thyroid scan was performed. Available frozen plasma was retrospectively tested for TPOAb and TSHRAb.
Control group
We estimated the prevalence of TPOAb and TSHRAb in patients with a baseline CD4+ T cell count below 200 x 106 cells/L and a CD4 increase greater than 100 x 106 cells/L after more than 1 yr on HAART, but with no clinical hyperthyroidism. A cross-sectional study was performed in 55 patients, 45 men and 10 women (median age, 38 yr; range, 23.269.3), on HAART for 34 (range, 1539) months; the pre-HAART CD4+ cell count was 60 x 106/L (range, 0190) and rose to 341 x 106/L (range, 33916).
Lymphocyte phenotype
In all cases, absolute CD4+ cell count was determined by flow cytometry on a FACScan (Becton Dickinson and Co., San Jose, CA) equipped with CellQuest software; briefly, 18 µL TriTEST reagent were added to 50 µL whole blood in a TruCount tube, FACS lysing solution was added after 15 min of staining, and tubes were capped and stored at 28 C until flow analysis within the time limit specified by the manufacturer.
Plasma HIV-1 RNA measurement
For patients 1, 2, 4, and 5, an Amplicor HIV Monitor test (Roche, Branchburg, NJ) was used to quantify HIV RNA in plasma (7). The detection threshold was 2.3 log10 HIV RNA copies/mL. The ultrasensitive technique (detection limit, 1.3 or 1.5 log10) was applied on negative plasmas (8). For patient 3, quantitation of HIV RNA in plasma was determined using the branched DNA signal amplification-based hybridization assay (Quantiplex HIV-RNA assay kit, Chiron Corp., Emeryville, CA) according to the manufacturers instructions. The detection threshold was 4 log10 HIV RNA copies/mL in 1996 and 2.5 log10 HIV RNA copies/mL since 1997.
T4 and TSH measurements
The free T4 kit (Immunotech, Marseilles, France) was based on the competition between free T4 in the samples and a biotinylated analog of T4 for a monoclonal 125I-labeled antibody (reference range of the laboratory, 1020 pmol/L). The TSH kit (BRAHMS, Berlin, Germany) was a chemiluminescent immunometric assay using a monoclonal antibody to TSH immobilized on a polystyrene tube and an acridinium ester-conjugated monoclonal anti-TSH antibody (reference range, 0.53. mU/L; functional sensitivity, 0.01 mU/L).
Autoantibodies
Frozen plasma samples used for viral load determination were retrieved for TPOAb and TSHRAb testing. TPOAb were measured using RIA kits: BRAHMS for patients 1 and 3 (test based on the competition between TPOAb in the samples and a monoclonal TPOAb for labeled TPO; reference range, <60 IU/mL), Diagnostic Products (Los Angeles, CA) Immulite for patient 2 (test based on the binding of TPOAb in the samples to immobilized human TPO; reference range, <35 IU/mL), and Biomedical Diagnostics (Croissy Beaubourg, France) for patients 4 and 5 (test based on the binding of TPOAb in the samples to immobilized recombinant TPO; reference range, <50 IU/mL).
RIA kits were used for detection of serum TSHRAb; BRAHMS (Berlin, Germany) for patients 1, 2, and 3 and Sanofi Pharmaceuticals, Inc. (Marnes-la-Coquette, France) for patients 4 and 5 (reference range, <10 IU/L for all patients). These tests are positive in more than 80% of untreated GD cases (to be compared with 95% for the thyroid cell-stimulating immunoglobulin test, unavailable for routine use).
For patients 1, 4, and 5, antinuclear (anti-ANA) were detected by indirect fluorescence on HEp-2 cells (Sanofi Pharmaceuticals, Inc.). Antidouble stranded DNA antibodies were determined by RIA (Farr assay) and enzyme-linked immunosorbent assay according to published procedures (9). Antiglutamic acid decarboxylase and antityrosine phosphatase (IA-2) antibodies were determined by radioligand assay (10, 11).
Descriptive statistics
Values were expressed as medians (range).
| Results |
|---|
|
|
|---|
In the five patients, viral load reached undetectable levels on
HAART, and CD4+ T cell count increased to a
median of 340 x 106 cells/L (range,
163460; Table 1
). They were
admitted for hyperthyroidism 1422 months after they started HAART,
and all presented with progressive weight loss, asthenia, tachycardia,
tremor, anxiety, and hypertension. Laboratory results (Table 2
) revealed elevated free
T4 and suppressed TSH. TPOAb and TSHRAb titers
were markedly positive in all cases. Thyroid scan showed diffuse
hyperfixation characteristic of GD.
|
|
TPOAb and TSHRAb appeared after CD4+ cells
had dramatically risen on HAART, whereas they were repeatedly absent
before institution of HAART. As shown in Table 3
, the occurrence of GD was closely
related to the rise of TSHRAb. In patients 1, 2, and 3, TPOAb rose
before TSHRAb; conversely, in patients 4 and 5, TSHRAb rose before
TPOAb.
|
Screening of patients on HAART for antithyroid antibodies
None of the 55 patients had TPOAb or THSRAb.
| Discussion |
|---|
|
|
|---|
In our patients, autoantibodies were absent before immune restoration, ruling out the exacerbation of a preexisting thyroid autoimmune disease, as reported during interferon therapy (17) or the cure of Cushings syndrome (18). Our observation should therefore be compared with the reported clinical syndromes linked to immune restoration in treated HIV infection (19). Early inflammatory conditions involving opportunistic agents, appearing as early as 6 days and up to 3 weeks after the beginning of HAART, were associated with a predominance of blood memory CD4+ cells, redistributed from lymphoid tissue (1, 20) rather than expanded in the periphery (21, 22, 23), and were probably due to the expansion of CD4+ cells specific for antigens of these agents. In contrast, immune recovery vitritis (24), Hashimotos thyroiditis (25), and GD (6, 26) have been described later than 40 weeks after HAART initiation. In our patients, the median times elapsed from HAART institution to GD (20 months) and from CD4+ T cell rise to GD diagnosis (18 months) are compatible with the second phase of CD4+ T cell increase (1, 4), suggesting the arrival of naive CD4+ T cells from the thymus (5). Although thymus function decreases with age, it was recently demonstrated that it may remain functional in HIV-infected adults (27). Using T cell antigen receptor rearrangement excision circles as a quantification method for cells produced by the thymus, Douek et al. showed a rise in the number of T cell antigen receptor rearrangement excision circles in CD4+ naive T cells of HIV subjects on HAART in whom viral HIV load fell below the detection limit (5). Besides, HAART normalizes the function of progenitor cells in HIV infection and might allow de novo production of T cells (28).
The T cell lymphocyte repertoire, which is significantly skewed in immunocompromised patients, is partially restored by HAART (3). As our patients had very low CD4+ cell nadirs, the time interval before GD was compatible with thymic production of naive CD4+ cells, and autoantibodies were absent before HAART, we hypothesize that autoimmunity resulted from abnormal reconstitution of a severely altered T cell repertoire.
Endocrine organ-specific autoimmune disorders mediated by autoreactive CD4+ T cell clones are observed after neonatal thymectomy in genetically susceptible mice or rats (29), adult thymectomy and cyclophosphamide-induced lymphopenia (30), or neonatal administration of cyclosporin A (31). These models and the clinical condition of our patients are characterized by profound lymphopenia and disrupted thymic maturation (32). Peripheral tolerance relies on mature peripheral T cells with suppressive functions (30, 31), and homing of peripheral CD4+ T cells to the thymus allows deletion of autoreactive clones (33). Thus, lymphopenia and disruption of thymic maturation may conjugate to allow tolerance disruption and autoimmunity. We suggest that immune restoration on HAART, occurring on a markedly altered T cell repertoire and thymic microenvironment, may result in autoimmunity through disruption of peripheral tolerance.
Our observations may also provide insight into the pathophysiology of GD. First, in HIV infection, the cytokine profile of activated CD4+ cells is switched to T helper cell type 2 (Th2) as immunodeficiency develops, and immune recovery after HAART restores a T helper cell type 1 (Th1) cytokine profile (34). Th1 CD4+ T cells are generally involved in cellular autoimmunity, but this has not been clearly established in GD (35, 36); the occurrence of GD after immune restoration may favor the involvement of Th1 cytokines in its pathogenesis. Second, the kinetics of appearance of autoantibodies to thyroid antigens in GD before clinical symptoms are unknown; in three of the patients described here, TPOAb antibodies preceded TSHRAb, and the opposite was observed in two subjects, suggesting a mechanism of epitope spreading (37). In these patients, we were unable to find markers of other organ-specific or nonorgan-specific autoimmune diseases, ruling out that autoimmunity resulted from polyclonal B cell activation. It is also noteworthy that none of the three tested patients bore the classical DR3 susceptibility allele for GD (38).
None of the 55 patients in whom HAART allowed a quantitative increase in the CD4+ T cell count had positive results for TPOAb or TSHRAb screening, which probably indicates a low incidence of such autoimmune disorders, but larger prospective studies are still warranted.
In the patients described here, thyroid-specific autoimmunity occurred late during immune reconstitution on HAART. We hypothesize that cellular immunodepression associated with thymic environment disturbances generates a situation comparable to that observed in lymphopenic animal models of autoimmunity. The recent description of GD in patients with multiple sclerosis treated with a T cell- depleting anti-CD52 monoclonal antibody, a condition that mimics that of immunocompromised patients, adds weight to this hypothesis (39).
| Acknowledgments |
|---|
Received April 8, 2000.
Revised July 7, 2000.
Accepted July 14, 2000.
| References |
|---|
|
|
|---|
interferon in
chronic hepatitis C. Hepatology. 26(Suppl 1):112S121S.
This article has been cited by other articles:
![]() |
K. A. Hagen, C. T. Moses, E. F. Drasler, K. M. Podetz-Pedersen, S. C. Jameson, and A. Khoruts A Role for CD28 in Lymphopenia-Induced Proliferation of CD4 T Cells J. Immunol., September 15, 2004; 173(6): 3909 - 3915. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. T. Moses, K. M. Thorstenson, S. C. Jameson, and A. Khoruts Competition for self ligands restrains homeostatic proliferation of naive CD4 T cells PNAS, February 4, 2003; 100(3): 1185 - 1190. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |