CELL BIOLOGY - IMMUNOLOGY - PATHOLOGY

Distinct HLA class II alleles determine antibody response tovaccination with hepatitis B surface antigen

SOPHIE CAILLAT-ZUCMAN, JEAN-JACQUES GIMENEZ, FRANCOIS WAMBERGUE, GINETTE ALBOUZE,BERNARD LEBKIRI, CATHERINE NARET, ANNE MOYNOT, PAUL JUNGERS, and JEAN-FRANCOIS BACH

Laboratory of Immunology and Department of Nephrology, Hopital Necker, Paris; Hemodialysis Units of Lille and Chateauroux;Clinique de l’Alma, Centre Edouard Rist and Centre Henri Kuntziger, Paris, France

Distinct HLA class II alleles determine antibody response to vaccina-tion with hepatitis B surface antigen. Major histocompatibility complex(MHC) determinants control antibody production in response to proteinantigens. Vaccination with hepatitis B surface antigen (HBsAg) frequentlyfails in hemodialyzed patients, but the genetic factors that modulatehumoral responsiveness are poorly characterized. We studied the distri-bution of HLA class II alleles in 415 hemodialyzed Caucasian patients whoreceived a full course of HBsAg vaccination, using class II oligotypingafter genomic amplification of the DRB1 and DQB1 loci. Phenotypefrequencies were compared in 114 non responders (anti-HBs antibodies #10 SI units/liter), 301 responders (anti-HBs antibodies .10 units/liter) and471 healthy controls. DRB1*01 (DR1) and DRB1*15 (DR15) frequencieswere lower in nonresponders than in responders and controls (DR1,12.3% vs. 22.9% and 24.8%, respectively; DR15, 14% vs. 22.9% and25.1%), while DRB1*03 (DR3) and DRB1*14 (DR14) frequencies werehigher (DR3, 32.5% vs. 16.6% and 25.3%, respectively; DR14, 9.6% vs.3% and 6.6%). Overall, 44.5% of DR3 or DR14 patients were nonre-sponders, compared to 18.1% of DR1 or DR15 patients (P 5 0.0001). Inconclusion the humoral response to HBsAg vaccine is influenced by classII allelic variants, which differ in their capacity to bind and presentpeptides to T lymphocytes.

The effector mechanisms that underlie the protection conferredby vaccination are not fully understood. Several factors, such asthe type and dose of antigen, and the route of vaccine adminis-tration, may contribute to vaccine unresponsiveness. In addition,genetic factors corresponding to the so-called immune response(Ir) genes of the major histocompatibility complex (MHC) regionhave been shown to control the ability to mount an effectiveresponse to protein antigens [1]. For instance, congenic micestrains respond differently to hepatitis B virus surface antigen(HBsAg) according to their H-2 phenotype [2, 3]. Population-based studies suggest that MHC genes may not only regulateresistance to infectious agents, as in the case of malaria [4] andhuman immunodeficiency virus (HIV) infection [5, 6], but alsodetermine the outcome of infectious diseases such as leprosy [7,

8], leishmaniasis [9, 10], onchocerciasis [11], papillomavirus-induced cervical carcinoma [12], and HIV infection [13, 14]. Thecourse after hepatitis B virus (HBV) infection (clearance of thevirus, persistent infection or chronic active hepatitis) seems to bedetermined by the host’s immune response genes [15–20].

Multiple histocompatibility complex genes may also control theefficiency of the humoral response to vaccination. For example,the DPB1*0501 allele correlates with enhanced antibody re-sponses to vaccination with a malaria sporozoite antigen in Thaisubjects [21]. A significant proportion of HBsAg vaccinees, par-ticularly hemodialyzed patients, fail to respond or develop verylow antibody titers [22, 23]. In Caucasians, poor responsiveness toHBsAg vaccine is associated with the DR3 and/or DR7 alleles[24–28], especially when they are present on the extended haplo-types HLA B8-DR3-SC01 and HLA B44-DR7-FC31. The Bw54-DR4-DQW4 haplotype has been linked to poor vaccine responsesin Japanese subjects [29–31]. However, most studies involved alimited number of alleles and small groups of nonresponders,meaning that the results were rarely statistically significant. Fur-thermore, different HLA alleles have been incriminated in studiesbased on serologic HLA typing, a method that gives only partialinformation as it assigns a large number of alleles to the sameserologic specificity.

We investigated HLA class II-linked mechanisms controllingresponsiveness to HBsAg in a large group of hemodialyzedpatients having received well-defined HBs vaccination protocol.

METHODS

Subjects

A total of 415 hemodialyzed Caucasian patients (218 males and197 females) were recruited in seven French hemodialysis units.The vaccine consisted of either HBs protein purified from pooledplasma of HBsAg-positive subjects (Hevac B, Pasteur Merieux;224 patients) or recombinant HBsAg (GenHevac, PasteurMerieux; 191 patients), the immunogenic potency of both vaccinesbeing very similar [32]. All the subjects received a reinforcedvaccination protocol, with three doses of HBs vaccine adminis-tered at monthly intervals, a fourth injection two months later,and additional injections at two-month intervals, if needed, untilthe anti-HBs titer exceeded 50 SI units/liter. Some patientsreceived up to 10 doses. The anti-HBs humoral response wasmeasured by means of conventional radioimmunoassay following

Key words: humoral response, major histocompatibility complex, HLAclass II, hepatitis B vaccine, immune response, end-stage renal disease,maintenance hemodialysis.

Received for publication October 6, 1997and in revised form December 15, 1997Accepted for publication December 18, 1997

© 1998 by the International Society of Nephrology

Kidney International, Vol. 53 (1998), pp. 1626–1630

1626

the third and fourth injections and after each subsequent injec-tion, and was expressed in standard international units (SIU) ofanti-HBs antibodies per liter. Nonresponse was defined by anantibody titer of 10 SIU/liter of less following the fourth vaccineinjection, whatever the response to additional injections.

The control population consisted of 471 randomly selectedhealthy Caucasian blood donors representative of the FrenchCaucasian population.

Human lymphocyte antigen typing

Human lymphocyte antigen (HLA) class II DNA typing wasperformed by means of hybridization with sequence-specific oli-gonucleotide probes following amplification of the HLA-DRB1and DQB1 genes in the polymerase chain reaction (PCR) aspreviously described [33].

Statistical analysis

Combined chi-square tests were used to determine heteroge-neity values for all alleles detected at each locus. Allele-specificassociations with antibody responses were then identified by usinga 2 3 2 chi-square (X2) test, or Fisher’s exact test if any value wasless than five. Odds ratios (OR), calculated with Woolf’s formulaand given with 95% confidence intervals, reflect the likelihood ofcarrying a particular HLA allele in the non responder groupcompared with the responder group. The level of significance wasset at P , 0.05.

Median anti-HBs antibody responses were compared betweenpatients with a given allele by using the nonparametric Mann-Whitney test.

RESULTS

Characteristics of patients

The mean age of dialysis patients at start of vaccination was59.6 6 1.6 years (mean 6 SEM). All 415 patients were hemodia-lyzed three times per week, with a mean duration of each dialysissession of 4.1 6 0.1 hours, that is, a median weekly dialysis timeof 12 hours. High-flux membranes were used in 45% of patients,and bicarbonate buffer was used in 78% of patients. The meanKt/V of urea was 1.21 6 0.02 and the mean urea reduction ratiowas 0.67 6 0.03. All patients were recommended a proteinintake $ 1 g/kg body wt/day and a caloric intake $ 30 kCal/kgbody wt/day. All were in good nutritional condition, as assessed bya mean body mass index (body weight over square height, kg/m2)of 23.1 6 0.8 in males and 21.8 6 0.9 in females. Meanhemoglobin level was 11.2 6 0.2 g/dl, with a mean corpuscularvolume of 95.5 6 1.1 m3, 63% of patients being treated withrecombinant erythropietin. Serum ferritin was 338 6 35 mg/literand transferrin saturation ratio 24 6 1.8%. Serum albumin was39 6 1 g/liter and serum transferrin 2.2 6 0.1 g/liter. All patientsreceived supplementation with folic acid (5 mg/day) and intrave-nous iron as needed to maintain serum ferritin . 200 mg/liter.They received as a mean 0.25 mg calcitriol per day. None hadconcurrent acute illness at the time of vaccination.

Response to HBsAg vaccine

The humoral response to HBs vaccine was bimodal: 72.5% ofhemodialyzed vaccinees (N 5 301) developed anti-HBs titersexceeding 10 SIU/liter after the fourth injection and were consid-ered as responders; the remaining 27.5% (N 5 114) of patients

had titers of 10 SIU/liter or less after the fourth injection (and insome cases after up to 10 injections) and were considered asnonresponders. No correlation was found between responderstatus and gender, age at the time of vaccination, the type of initialnephropathy leading to hemodialysis, or the type of vaccine(plasma-derived or recombinant).

HLA-DRB1 and DQB1 phenotypes

Phenotype frequencies did not differ between the entire groupof hemodialyzed patients and the control group (combined X2 516.5 for DRB1, P 5 0.17; and X2 5 19.4 for DQB1, P 5 0.11).

Table 1 compares the frequencies of DRB1 and DQB1 allelesin the responders and nonresponders with those in the healthycontrols. Analysis of the combined phenotype frequencies foreach class II locus in the responder and nonresponder groupsshowed that alleles of the DRB1 locus correlated with the vaccineresponse (x2 5 30.84, P 5 0.0021), whereas those of the DQB1locus did not (x2 5 12.5, P 5 0.48).

Frequencies of DRB1*01 (DR1) and DR2 [both DRB1*15(DR15) and DRB1*16] were lower in nonresponders than inresponders (OR 5 0.47, CI 0.25 to 0.87, P 5 0.02; OR 5 0.55, CI0.30 to 0.99, P 5 0.05, respectively). These differences corre-sponded to lower frequencies of DR1 and DR15 phenotypes innonresponders than in controls (DR1, 12.3% in non respondersvs. 24.8% in controls, P 5 0.003; DR15, 11.4% vs. 20.8%, P 50.023).

By contrast, DRB1*03 (DR3) and DRB1*14 (DR14) pheno-type frequencies were higher in nonresponders than in responders(OR 5 2.42, CI 1.46 to 3.96, P 5 0.0007; OR 5 3.46, CI 1.39 to8.6, P 5 0.008, respectively). These differences were related tolower frequencies of DR3 and DR14 in responders than incontrols (DR3, 16.6% vs. 25.3%, P 5 0.005; DR14, 3% vs. 6.6%,P 5 0.03). The frequencies of DR3, DR14, DR15 and DR1homozygous patients were similar in the nonresponder and re-sponder groups, and were in keeping with calculated gene fre-quencies (data not shown).

Overall, a response to HBs vaccine was positively associatedwith the DR1 and DR15 phenotypes, and negatively associatedwith the DR3 and DR14 phenotypes (Table 2). Among DR3and/or DR14 patients, 44.5% were nonresponders, compared toonly 18.1% of DR1 and/or DR15 patients (P 5 0.0001).

Accordingly, the median anti-HBs antibody titer was signifi-cantly lower in DR3 patients (20 SIU/liter) and in DR14 patients(10 SIU/liter) than in DR1 patients (33 SIU/liter, P 5 0.0012 andP 5 0.003, respectively) and in DR15 patients (30 SIU/liter, P 50.0046 and P 5 0.008, Mann-Whitney test).

DISCUSSION

Overall, the DR3 and DR14 phenotypes were associated with apoor humoral response to HBsAg vaccination, whereas the DR1and DR15 phenotypes were associated with a good response. Westudied hemodialyzed patients because of their well-known hypo-responsiveness to HBsAg vaccine [22, 23], giving us the opportu-nity to analyze a population with a high proportion of nonre-sponders. The links we observed were not due to a distortion ofHLA allelic frequencies in hemodialyzed patients, as there was nodifference in the overall phenotype distribution between theentire hemodialysis group and a healthy control group of similargeographic origin.

The weak response of uremic dialysis patients to hepatitis B

Caillat-Zucman et al: HLA and hepatitis B vaccine 1627

vaccine cannot be ascribed to inadequate dialysis or poor nutri-tional status, as parameters of adequate dialysis (Kt/V urea, andurea reduction ratio) and nutrition (serum albumin level and bodymass index) were in the optimal range of values [34].

The main function of HLA class II gene products is to bindantigenic peptides derived from exogenous proteins and topresent them, at the cell surface, to the T cell receptor on CD41T lymphocytes. Allelic differences in HLA molecules can modu-late their ability to bind peptides, and thereby change the natureof T cell recognition. Previous reports have described associationsbetween HLA class II alleles and HBsAg vaccine responses[24–31], but their interpretation has often been controversial. InCaucasians, the HLA A1-B8-DR3 haplotype has been shown toinduce a recessive inability to develop anti-HBs antibodies, par-ticularly in homozygotes. Various immunological mechanisms,most of which have now been rejected, have been proposed toexplain this lack of response:

(a) the absence of dominant Ir genes located in the MHC [24,26, 35];

(b) defective antigen-presenting cell/T cell interactions [36, 37],possibly due to a lack of binding of HBsAg peptides to HLA

molecules or to an absence of a co-stimulatory signal, leading toinefficient presentation to CD41 helper T cells;

(c) elimination of HBsAg-specific helper T cells during thymicor post-thymic repertoire maturation, or inactivation by periph-eral anergy [38];

(d) selective killing of HBsAg-specific B cells by MHC-re-stricted cytotoxic T lymphocytes [39];

(e) the presence of dominant immune suppression (Is) genes, apossibility raised by the induction of Ag-specific suppressor T cellsin Japanese subjects with the extended Bw54-DR4-DRw53-DQw4haplotype [29, 30];

(f) one may also propose another mechanism, given recentinsights. Protein antigens tend to elicit predominantly one of twoT-helper cell (Th) subsets that have different patterns of cytokineproduction [reviewed in 40]. Following stimulation, antigen-spe-cific T cells progress from naıve cells to Th0 cells that furtherdifferentiate into Th1 or Th2 cells. Th1 cells mainly produceinterleukin (IL)-2, tumor necrosis factor (TNF) and interferon-g(IFNg), thereby favoring cellular immune responses such asmacrophage activation and delayed-type hypersensitivity (DTH).Th2 cells mainly produce IL-4, IL-5, IL-6 and IL-13, cytokinesthat not only stimulate humoral responses but can also suppressTh1 cellular responses. Th1 and Th2-type responses are thusantagonistic.

Microbial pathogens frequently elicit polarized responses ofeither the Th1 or Th2 type, probably because they alter thecytokine milieu during initial antigen priming. Other factors mayalso influence which T cell subset dominates after infection [41].One may imagine that some alleles, such as DR3 and DR14,preferentially induce a Th1 (cellular) response, which wouldinhibit the Th2 (humoral) response and limit the production ofanti-HBsAg antibodies. Conversely, the response would beskewed towards Th2 responsiveness in subjects with DR1 andDR2 alleles. This hypothesis fits with the low DR3 and DR14frequencies in responders, and the low DR1 and DR2 frequenciesin nonresponders, and can reconcile discrepancies among previ-ous studies, which have been attributed to differences in ethnicgroup or in the route of HBsAg vaccination. Similar MHC-linked

Table 1. DRB1 and DQB1 phenotype frequencies in control subjects, and in responders and nonresponders to HBsAg vaccine

DRB1*

Controls(N 5 471)

Nonresponders(N 5 114)

Responders(N 5 301) ORa (95%

confidence interval) PN % N % N % (¶)

01 117 24.8 14 12.3b 69 22.9 0.47 (0.25–0.87) 0.0202 (15116) 118 25.1 16 14.0c 69 22.9 0.55 (0.30–0.99) 0.0515 98 20.8 13 11.4d 52 17.316 20 4.2 3 2.6 17 5.603 119 25.3 37 32.5 50 16.6e 2.42 (1.46–3.96) 0.000704 93 19.7 26 22.8 78 25.911 107 22.7 32 28.1 85 28.212 16 3.4 3 2.6 9 3.013 110 23.4 29 25.4 73 24.314 31 6.6 11 9.6 9 3.0f 3.46 (1.39–8.6) 0.00807 121 25.7 26 22.8 70 23.308 29 6.2 4 3.5 20 6.609 9 1.9 1 0.9 8 2.710 13 2.8 7 6.1 9 3.0

a Only significant risks are indicated.b P 5 0.003, c P 5 0.012, d P 5 0.23, x2 test for heterogeneity between nonresponders and controls.e P 5 0.005, f P 5 0.03, x2 test for heterogeneity between responders and controls.

Table 2. AntiHBsAg antibody response according to DR phenotype

DRB1* phenotype N

Anti-HBs antibody titer

P.10 IU/liter

N (%),10 IU/liter

N (%)

DRB1*03 87 50 (57.5) 37 (42.5)non-DRB1*03 328 251 (76.5) 77 (23.5) 5 0.0007DRB1*14 21 10 (47.6) 11 (52.4)non-DRB1*14 394 291 (73.8) 103 (26.1) 5 0.02DRB1*01 83 69 (83.1) 14 (16.9)non-DRB1*01 332 232 (69.9) 100 (30.1) 5 0.018DRB1*15 66 53 (80.3) 13 (19.7)non-DRB1*15 349 248 (71) 101 (28.9) 5 0.13, NSDRB1*3 or DRB1*14 108 60 (55.5) 48 (44.4)DRB1*1 or DRB1*15 149 122 (81.9) 27 (18.1) , 0.0001

NS is not significant.

Caillat-Zucman et al: HLA and hepatitis B vaccine1628

mechanisms might regulate susceptibility to various infections.Indeed, the type of Th cell response to certain pathogens candiffer according to the MHC haplotype of mouse strains [42]. Inhumans, the course after HBV infection is determined by theHLA phenotype. Association between the A1-B8 haplotype andclearance of HBV in hemodialyzed patients has long been de-scribed [15, 16], and suggests that this haplotype, which usuallybears the DR3 allele due to linkage disequilibrium, might beassociated with high Th1 response against HBsAg. The frequencyof DR3 is increased in patients with HBV chronic active hepatitisin comparison with healthy carriers [17, 18]. In Gambia, theDRB1*1302 allele is associated with protection against persistentHBV infection [20], whereas DR7, but not DR2 subjects fre-quently develop chronic HBV infection in Katar [19]. The clinicaldisease induced by Mycobacterium leprae is also influenced byHLA class II alleles: the DR3 and DR2 haplotypes confer a highrisk of polar tuberculoid leprosy, which is associated with strongDTH reactions and low antibody levels (Th1 response), whereasother alleles are associated with the multibacillary lepromatousform characterized by excessive production of nonprotectiveantibodies (Th2 response) [7, 8, 43, 44]. Susceptibility to muco-cutaneous leishmaniasis, a strong DTH reaction caused by Leish-mania braziliensis infection, is associated with a higher frequencyof the DR3-TNF-b2 haplotype and a lower frequency of the DR2haplotype [45]. The A1 B8 DR3 haplotype is associated with rapiddisease progression in HIV-seropositive individuals, whereas re-sistance to infection among frequently exposed prostitutes isassociated with DR13 [13, 14]. DR2 and DR5 are associated withan IgE allergic response (Th2) to Aspergillus fumigatus in bron-chomulmonary aspergillosis [46]. Finally, known associations be-tween HLA and susceptibility to autoimmune diseases are also inkeeping with this hypothesis. In an animal model of collagenIV-induced arthritis, a single peptide can induce a Th1 or a Th2response depending on the MHC genotype [47]. The DR3haplotype, which predisposes humans to insulin-dependent dia-betes mellitus, is associated with a cellular response (Th1) leadingto immune destruction of pancreatic b cells, whereas the stronglyprotective DR15 haplotype might antagonize this Th1 response.

Crystallography of MHC class II molecules has revealed thatpeptides are accomodated in a polymorphic binding groove [48].Allelic differences in amino acid residues located within the

groove directly affect binding affinity for particular peptides. Itappears that high-affinity ligands shift the lymphokine-secretionpattern towards a Th1-like response [39, 47]. The products ofHLA class II alleles associated with strong and weak responses toHBsAg differ structurally, with nonconservative amino acid sub-stitutions occurring mainly in the first hypervariable region,between residues 9 to 13. Based on the three-dimensional modelof the class II antigen-binding cleft [48], these positions arethought to influence interactions with peptide residues. Knowl-edge of the relative binding affinities of HBsAg peptides for classII molecules associated with strong and weak responses to thevaccine might lead to strategies for controlling the type of helperT cells activated in response to defined antigens.

ACKNOWLEDGMENTS

This work was supported by the “Contrat de Recherche Clinique CRC94026-Delegation a la Recherche Publique” from Assistance Publique-Hopitaux de Paris. We thank E. Audran, P. Przednowed and I. Texier fortheir excellent technical assistance.

Reprint request to Dr. Sophie Caillat-Zucman, Laboratoire d’Immun-ologie, Hopital Necker, 161 Rue de Sevres, 75015 Paris, France.E-mail: caillat@necker.fr

APPENDIX

Abbreviations used in this article are: DR1, DRB*01 phenotype; DR3,DRB*03 phenotype; DR14, DRB*14 phenotype; DR15, DRB1*15 phe-notype; DTH, delayed-type hypersensitivity; HBsAg, hepatitis B surfaceantigen; HBV, hepatitis B virus; HIV, human immunodeficiency virus;HLA, human lymphocyte antigen; IFNg, interferon-g; IL, interleukin; Is,immune suppression; MHC, multiple histocompatibility complex; OR,odds ratio; PCR, polymerase chain reaction; SIU, standard internationalunits; Th, T helper cell; TNF, tumor necrosis factor.

REFERENCES1. BENACERRAF B, MCDEVITT HO: Histocompatibility-linked immune

response genes. Science 175:273–279, 19722. MILICH DR, CHISARI FV: Genetic regulation of the immune response

to the hepatitis B surface antigen (HBsAg). I. H-2 restriction of themurine humoral response to the a and d determinants of HBsAg.J Immunol 129:320–325, 1982

3. MILICH DR, LEROUX-ROELS GG, LOUIE RE, CHISARI FV: Geneticregulation of the immune response to hepatitis B surface antigen(HBSAg). IV. Distinct H-2 linked Ir genes control antibody responsesto different HBsAg determinants on the same molecule and map tothe I-A and I-C regions. J Exp Med 159:41–56, 1984

4. HILL AV, ALLSOPP CE, KWIATKOWSKI D, ANSTEY NM, TWUMASI P,ROWE PA, BENNETT S, BREWSTER D, MCMICHAEL AJ, GREENWOODBM: Common west African HLA antigens are associated with pro-tection from severe malaria. Nature 352:595–600, 1991

5. GOZLAN M: Update on HIV transmission and pathogenesis. Lancet346:1290, 1995

6. WESTBY M, MANCA F, DALGLEISH AG: The role of host immuneresponses in determining the outcome of HIV infection. ImmunolToday 17:120–126, 1996

7. DE VRIES RR, MEHRA NK, VAIDYA MC, GUPTE MD, MEERA KHAN P,VAN ROOD JJ: HLA-linked control of susceptibility to tuberculoidleprosy and association with HLA-DR types. Tissue Antigens 16:294–304, 1980

8. VAN EDEN W, GONZALEZ NM, DE VRIES RR, CONVIT J, VAN ROOD JJ:HLA-linked control of predisposition to lepromatous leprosy. J InfectDis 151:9–14, 1985

9. LARA ML, LAYRISSE Z, SCORZA JV, GARCIA E, STOIKOW Z, GRANA-DOS J, BIAS W: Immunogenetics of human American cutaneousleishmaniasis. Study of HLA haplotypes in 24 families from Venezu-ela. Hum Immunol 30:129–135, 1991

10. PETZL-ERLER ML, BELICH MP, QUEIROZ-TELLES F: Association ofmucosal leishmaniasis with HLA. Hum Immunol 32:254–260, 1991

DQB1*

Controls(N 5 471)

Nonresponders(N 5 110)

Responders(N 5 277)

N % N % N %

0501 120 25.5 20 17.5 78 25.90502 26 5.5 5 4.4 16 5.30503 25 5.3 9 7.9 14 4.l70601 8 1.7 1 0.9 2 0.70602 79 16.8 11 9.6 44 14.60603 55 11.7 17 14.9 35 11.60604 29 6.2 7 6.1 17 5.60615 3 0.6 6 5.3 8 2.70201 195 41.4 49 43.0 103 34.20301 160 34.0 43 37.7 202 33.90302 57 12.1 15 13.2 51 16.90303 33 7.0 4 3.5 13 4.30401 4 0.8 0 0 4 1.30402 24 5.1 5 4.4 13 4.3

Caillat-Zucman et al: HLA and hepatitis B vaccine 1629

11. MEYER CG, GALLIN M, ERTTMANN KD, BRATTIG N, SCHNITTGER L,GELHAUS A, TANNICH E, BEGOVICH AB, ERLICH HA, HORSTMANNRD: HLA-D alleles associated with generalized disease, localizeddisease, and putative immunity in Onchocerca volvulus infection. ProcNatl Acad Sci USA 91:7515–7519, 1994

12. APPLE RJ, ERLICH HA, KLITZ W, MANOS MM, BECKER TM,WHEELER CM HLA DR-DQ associations with cervical carcinomashow papillomavirus-type specificity. (abstract) Nature Genet 6:157,1994

13. STEEL CM, LUDLAM CA, BEATSON D, PEUTHERER JF, CUTHBERT RJ,SIMMONDS P, MORRISON H, JONES M: HLA haplotype A1 B8 DR3 asa risk factor for HIV-related disease. Lancet i:1185–188, 1988

14. KASLOW RA, CARRINGTON M, APPLE R, PARK L, MUNOZ A, SAAH AJ,GOEDERT JJ, WINKLER C, O’BRIEN SJ, RINALDO C, DETELS R,BLATTNER W, PHAIR J, ERLICH H, MANN DL: Influence of combina-tions of human major histocompatibiliy complex genes on the courseof HIV-1 infection. Nature Med 2:405–411, 1996

15. BACH JF, ZINGRAFF J, DESCAMPS B, NARET C, JUNGERS P: HL-A1,8phenotype and Hbs antigenaemia in haemodialysis patients. (letter)Lancet 2:707, 1975

16. DESCAMPS B, JUNGERS P, NARET C, DEGOTT C, ZINGRAFF J, BACH JF:HLA-A1, B8 phenotype association and HBs antigenemia evolutionin 440 hemodialyzed patients. Digestion 15:271–277, 1977

17. FORZANI B, ACTIS GC, VERME G, AMOROSO A, BORELLI I, CURTONIES, RUMI MG, PICCIOTTO A, MARINUCCI G, FRENI MA: HLA-DRantigens in HBsAg-positive chronic active liver disease with andwithout association with delta infection. Hepatology 4:1107–1110, 1984

18. YANG PM, SUNG JL, CHEN DS: HLA-A, B, C, DR antigens in chronichepatitis B viral infection. Hepatogastroenterology 36:363–366, 1989

19. ALMARRI A, BATCHELOR JR: HLA and hepatitis B infection. Lancet344:1194–1195, 1994

20. THURSZ MR, KWIATKOWSKI D, ALLSOPPC, GREENWOOD BM, THOMASHC, HILL A: Association between an MHC class II allele andclearance of hepatitis B virus in the Gambia. N Engl J Med 332:1065–1069, 1995

21. STEPHENS HA, BROWN AE, CHANDANAYINGYONG D, WEBSTER HK,SIRIKONG M, LONGTA P, VANGSERATTHANA R, GORDON D, LEKMAK S,RUNGRUANG E: The presence of the HLA class II allele DPB1*0501in ethnic Thais corelates with an enhanced vaccine-induced antibodyresponse to a malaria sporozoite antigen. Eur J Immunol 25:3142–3147, 1995

22. CROSNIER J, JUNGERS P, COUROUCE AM, LAPLANCHE A, BENHAMOUE, DEGOS F, LACOUR B, PRUNET P, CERISIER Y, GUESRY P: Random-ised placebo-controlled trial of hepatitis B surface antigen vaccine inFrench haemodialysis units: II, Haemodialysis patients. Lancet 1:797–800, 1981

23. STEVENS CE, ALTER HJ, TAYLOR PE, ZANG EA, HARLEY EJ, SZMU-NESS W, THE DIALYSIS VACCINE TRIAL STUDY GROUP: Hepatitis Bvaccine in patients receiving hemodialysis. Immunogenicity and effi-cacy. N Engl J Med 311:496–501, 1984

24. CRAVEN DE, AWDEH ZL, KUNCHES LM, YUNIS EJ, DIENSTAG JL,WERNER BG, POLK BF, SNYDMAN DR, PLATT R, CRUMPACKER CS,GRADY GF, ALPER CA: Nonresponsiveness to hepatitis B vaccine inhealth care workers. Ann Intern Med 105:356–360, 1986

25. WEISSMAN JY, TSUCHIYOSE MM, TONG MJ, CO R, CHIN K, ETTENGERRB: Lack of response to recombinant Hepatitis B vaccine in nonre-sponders to the plasma vaccine. JAMA 260:1734–1738, 1988

26. ALPER CA, KRUSKALL MS, MARCUS-BAGLEY D, CRAVEN DE, KATZAJ, BRINK SJ, DIENSTAG JL, AWDEH Z, YUNIS EJ Genetic predictionof nonresponse to hepatitis B vaccine. N Engl J Med 321:708–712,1989

27. POL S, LEGENDRE C, MATTLINGER B, BERTHELOT P, KREIS H: Geneticbasis of nonresponse to hepatitis B vaccine in hemodialyzed patients.J Hepatol 11:385–387, 1990

28. VARLA-LEFTHERIOTI M, PAPANICOLAOU M, SPYROPOULOU M, VAL-LINDRA H, TSIROYIANNI P, TASSOPOULOS N, KAPASOURI H, STAVRO-POULOS-GIOKAS C: HLA-associated non-responsiveness to hepatitis Bvaccine. Tissue Antigens 35:60–63, 1990

29. WATANABE H, MATSUCHITA S, KAMIKAWAJI N, HIRAYAMA K, OKU-MURA M, SASAZUKI T: Immune suppression gene on HLA-Bw54-DR4-DRw53 haplotype controls nonresponsiveness in humans to

hepatitis B surface antigen via CD81 suppressor T cells. HumImmunol 22:9–17, 1988

30. WATANABE H, OKUMURA M, HIRAYAMA K, SASAZUKI T: HLA-Bw54-DR4-DRw53-DQw4 haplotype controls nonresponsiveness to hepati-tis-B surface antigen via CD8-positive suppressor T cells. TissueAntigens 36:69–74, 1990

31. HATAE K, KIMURA A, OKUBO R, WATANABE H, ERLICH HA, UEDA K,NISHIMURA Y, SASAZUKI T: Genetic control of nonresponsiveness tohepatitis B virus vaccine by an extended HLA haplotype. Eur J Im-munol 22:1899–1905, 1992

32. JUNGERS P, CHAUVEAU P, COUROUCE AM, DEVILLIER P, EXCLER JL,BAILLEUX F, SALIOU P: Imminogenicity of the recombinant GenHevacB Pasteur vaccine against hepatitis B in chronic uremic patients.J Infec Dis 169:399–402, 1994

33. CAILLAT-ZUCMAN S, GARCHON HJ, TIMSIT J, ASSAN R, BOITARD C,DJILALI-SAIAH I, BOUGNERES P, BACH JF: HLA genetic heterogeneityof insulin dependent diabetes mellitus. Indication for an age depen-dent susceptibility gradient. J Clin Invest 90:2242–2250, 1992

34. OWEN WF, LEW NL, LIN Y, LOWRIE EG, LAZARUS JM: The ureareduction ratio and serum labumin concentration as predictors ofmortality in patients undergoing hemodialysis. N Engl J Med 329:1001–1006, 1993

35. KRUSKALL MS, ALPER CA, AWDEH Z, YUNIS EJ, MARCUS-BAGLEY D:The immune response to hepatitis B vaccine in humans: Inheritancepatterns in families. J Exp Med 175:495–502, 1992

36. MILICH DR: Genetic and molecular basis for T and B cell recognitionof hepatitis B viral antigens. Immunol Rev 99:72–103, 1987

37. EGEA E, IGLESIAS A, SALAZAR M, MORIMOTO C, KRUSKALL MS,AWDEH Z, SCHLOSSMAN, SF, ALPER CA, YUNIS EJ: The cellular basisfor lack of antibody response to hepatitis B vaccine in humans. J ExpMed 173:531–538, 1991

38. SALAZAR M, DEULOFEUT H, GRANJA C, DEULOFEUT R, YUNIS DE,MARCUS-BAGLEY D, AWDEH Z, ALPER CA, YUNIS EJ: Normal HBsAgpresentation, T-cell defect in the immune response of nonresponders.Immunogenetics 41:366–374, 1995

39. BARNABA V, FRANCO A, ALBERTI A, BENVENUTO R, BALSANO F:Selective killing of hepatitis B envelope antigen-specific B cells byclass I-restricted, exogenous antigen-specific T lymphocytes. Nature345:258–260, 1990

40. ABBAS A, MURPHY KM, SHER A: Functional diversity of helper Tlymphocytes. Nature 383:787–793, 1996

41. KUMAR V, BHARDWAJ V, SOARES L, ALEXANDER J, SETTE A, SERCARZE: MHC binding affinity of an antigenic determinant is crucial for thedifferential secretion of interleukin 4/5 or interferon g by T cells. ProcNatl Acad Sci USA 92:9510–9514, 1995

42. HEINZEL FP, SADICK MD, HOLADAY BJ, COFFMAN RL, LOCKSLEYRM: Reciprocal expression of interferon gamma or interleukin 4during the resolution or progression of murine leishmaniasis. Evi-dence for expansion of distinct helper T cell subsets. J Exp Med169:59–72, 1989

43. SCHAUF V, RYAN S, SCOLLARD D, JONASSON O, BROWN A, NELSON K,SMITH T, VITHAYASAI V: Leprosy associated with HLA-DR2 andDQw1 in the population of northern Thailand. Tissue Antigens26:243–247, 1985

44. DESSOUKEY MW, EL-SHIEMY S, SALLAM T: HLA and leprosy: Segre-gation and linkage study. Int J Dermatol 35:257–264, 1996

45. CABRERA M, SHAW MA, SHARPLES C, WILLIAMS H, CASTES M,CONVIT J, BLACKWELL JM: Polymorphism in tumor necrosis factorgenes associated with mucocutaneous leishmaniasis. J Exp Med 182:1259–1264, 1995

46. CHAUHAN B, KNUTSEN AP, HUTCHESON PS, SLAVIN RG, BELLONE CJ:T cell subsets, Epitope mapping, and HLA restriction in patients withbronchopulmonary aspergillosis. J Clin Invest 10:2324–2331, 1996

47. PFEIFFER C, STEIN J, SOUTHWOOD S, KETELAAR H, SETTE A, BOT-TOMLY K: Altered peptide ligands can control CD4 T lymphocytedifferenciation in vivo. J Exp Med 181:1569–1574, 1995

48. BROWN JH, JARDETZKY TS, GORGA JC, STERN LJ, URBAN RG,STROMINGER JL, WILEY DC: Three-dimensional structure of thehuman class II histocompatibility antigen HLA-DR1. Nature 364:33–39, 1993

Caillat-Zucman et al: HLA and hepatitis B vaccine1630