• HA-1 minor histocompatibility antigen;
  • stem cell transplantation;
  • acute graft-versus-host disease;
  • RSCA;
  • chronic graft-versus-host disease


  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Disparity for the minor histocompatibility antigen HA-1 between patient and donor has been associated with an increased risk of acute graft-versus-host disease (GvHD) after allogeneic human leucocyte antigen (HLA)-identical sibling donor stem cell transplantation (SCT). However, no data concerning the impact of such disparity on chronic GvHD, relapse or overall survival are available. A retrospective multicentre study was performed on 215 HLA-A2-positive patients who received an HLA-identical sibling SCT, in order to determine the differences in acute and chronic GvHD incidence on the basis of the presence or absence of the HA-1 antigen mismatch. Disease-free survival and overall survival were also analysed. We detected 34 patient–donor pairs mismatched for HA-1 antigen (15·8%). Grades II–IV acute GvHD occurred in 51·6% of the HA-1-mismatched pairs compared with 37·1% of the non-mismatched. The multivariate logistic regression model showed statistical significance (P: 0·035, OR: 2·96, 95% CI: 1·07–8·14). No differences were observed between the two groups for grades III–IV acute GvHD, chronic GvHD, disease-free survival or overall survival. These results confirmed the association between HA-1 mismatch and risk of mild acute GvHD, but HA-1 mismatch was not associated with an increased incidence of chronic GvHD and did not affect relapse or overall survival.

Minor histocompatibility antigens (mHA) are polymorphic small peptides presented to T lymphocytes restricted by the major histocompatibility complex (MHC) molecule (Perreault et al, 1990). These peptides are generated by the degradation of endogenous proteins that are coded by genes located outside the Mhc and the MHC molecule presents them on the cell surface. This MHC-restricted presentation may produce an allogeneic immune recognition by T lymphocytes after HLA-identical sibling donor stem cell transplantation (SCT) if there are minor histocompatibility antigen disparities between patient and donor (Goulmy, 1997).

The HA-1 antigen is a nonapeptide produced by the natural processing of an unknown endogenous protein. The presentation of this minor histocompatibility antigen to the T cells is restricted by the HLA-A*0201 molecule. The tissue distribution of the HA-1 antigen is restricted to the haematopoietic lineage (Warren et al, 1998). This antigen has two allelic variants, the HA-1H allele, characterized by the presence of histidine at position 3 of the nonapeptide (VLHDDLLEA), and the HA-1R allele, which encodes arginine at the same position (VLRDDLLEA). The correlation between the HA-1-positive phenotype and the HA-1H allele has been shown. This is as a result of the greater affinity of the HA-1H peptide than the HA-1R peptide to the HLA-A*0201 molecule. For this reason, HA-1 mismatch is defined by the presence of HA-1H allele in the host but not in the donor (den Haan et al, 1998).

The Leiden group (Goulmy et al, 1996) studied the impact of several mHAs on acute graft-versus-host disease (GvHD) after allogeneic HLA-identical sibling donor SCT. They detected an increased risk of acute GvHD for adult patient–donor pairs mismatched at the mHA HA-1. The disparity in more than one mHA was slightly more predictive of GvHD than the HA-1 antigen mismatch alone, suggesting immunodominant behaviour of this antigen. More recently, another study (Tseng et al, 1999) confirmed the relevance of the mHA HA-1 mismatch for acute GvHD.

Recently, we applied the reference strand-mediated conformation analysis (RSCA) for HA-1 genomic typing, based on the difference at the nucleotide sequence between the two alleles (Aróstegui et al, 2000). We used this method to correlate the HA-1 disparity with acute and chronic GvHD, relapse and overall survival after allogeneic HLA-identical sibling donor SCT.

Patients and methods

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Patients DNA samples from 215 patient–donor pairs were collected by 14 Spanish transplant teams. All the patients were more than 15 years old and received a myeloablative conditioning regimen followed by infusion of stem cells from an HLA-identical sibling donor, between March 1991 and February 2000. All the patient–donor pairs were positive for HLA-A2. The stem cell source was granulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood in 100 cases and bone marrow in the remaining 125 patients. T-cell depletion was performed in 52 cases (24·1%). Table I shows the characteristics of the patients included in the study. GvHD was graded according to the standard criteria (Glucksberg et al, 1974).

Table I.   Clinical characteristics of the patient/donor pairs included in the study (n = 215).
  1. M, male; F, female; BM, bone marrow; PB, peripheral blood; CSA, cyclosporin A; MTX, methotrexate; PDR, prednisone; TBI, total body irradiation.

Patient's age (median, range)36·5 (15–59)
Patient's sex (M:F)133:79
Diagnoses:68 chronic myeloid leukaemia
54 acute myelogenous leukaemia
28 acute lymphoblastic leukaemia
20 non-Hodgkin Lymphoma
17 myelodysplastic syndrome
11 severe aplastic anaemia
8 multiple myeloma
6 other
Disease stage (early: advanced)154:47 (11 missing data)
Patient's CMV status (+ve: –ve)159:42 (11 missing data)
Donor's age (median, range)34 (8–73)
Sex combination (patient:donor)M:M 64
M:F 66
F:F 34
F:M 48
Donor with previous pregnancies35
HA-1 mismatch34 (15·8%)
Source of stem cells (PB:BM)92:123
T-cell depletion52 (24·2%)
Pharmacological GvHD prophylaxis140 CSA + MTX
45 CSA
24 CSA + PDR
4 missing data
Conditioning regimen98 busulphan + cyclophosphamide
81 cyclophosphamide + TBI
33 other
3 missing data

HA-1 genotyping To obtain HA-1 genomic typing, we applied the RSCA technique, which has been used to explore polymorphic genetic systems, such as HLA or cystic fibrosis mutations on the CFTR gene (Argüello et al, 1998a). RSCA allows the study of polymorphic genetic systems and is based on three steps: (1) locus-specific polymerase chain reaction (PCR), (2) hybridization with a fluorescent-labelled reference (FLR), and (3) polyacrylamide gel electrophoresis in an automated DNA sequencer.

The HA-1 locus-specific amplification of the samples was performed with the primers and PCR conditions described previously (Aróstegui et al, 2000). To obtain the fluorescent-labelled references (FLRs), we selected two homozygous samples (HA-1R/R and HA-1H/H), identified by the PCR-SSP (PCR-single-stranded polymorphism) typing method described by the Leiden group (Wilke et al, 1998) and confirmed by sequence analysis. The FLR was the PCR product obtained when amplifying a known homozygous sample for the studied fragment. This PCR was performed with the same locus-specific primers and PCR conditions as the studied samples. The only difference was that, in the case of the FLRs, the forward primer was labelled with a fluorophor (Cy5), which confers fluorescent properties on this PCR product. We gave the name FLR-H to the FLR containing only the HA-1H allele and FLR-R to the FLR containing only the HA-1R allele.

After amplification of the sample, the PCR product was hybridized separately with both FLRs by mixing at a sample:FLR ratio of 3:1, denaturing for 5 min at 95°C and cooling for 5 min at 55°C and 5 min at 15°C. The second step facilitated the random hybridization between sense and antisense strands. This process generates the formation of homoduplexes and heteroduplexes. Only the combinations containing the sense strand of the FLR have fluorescent properties: the homoduplex and one heteroduplex for each allele present in the PCR product of the sample.

The hybridization product (2 μl) and 6× Ficoll loading buffer were loaded in a 6% non-denaturing Long Ranger gel (FMC Bioproducts, Rockland, Maine, USA). Electrophoresis was performed with 1× Tris-borate-EDTA (TBE) running buffer, on an ALFexpress II automated sequencer (Amersham Pharmacia Biotech, Uppsala, Sweden) at 30 W constant power. The running time for an 8-cm long gel was 100 min.

The heteroduplexes ran slower than the homoduplex and the detection of the fluorescent molecules by the laser located at the end of the gel enabled the identification of each fluorescent-labelled duplex. One sample with genotype HA-1R/R will generate only one heteroduplex when using the FLR-H, and none when using the FLR-R. If the typing corresponds to the HA-1H/H genotype, the pattern will be inverted. The detection of one heteroduplex with both FLRs corresponds to a HA-1H/R heterozygous sample.

HLA-A2 subtyping To detect the patient/donor pairs with HA-1 mismatch that expressed the HLA-A*0201 allele, we performed a high-resolution typing by RSCA, as described previously (Argüello et al, 1998b). The cell line SPO-010 was used as the HLA-A*0201 positive control to identify the mobility pattern of this allele.

Statistical analysis Homogeneity between HA-1 antigen-mismatched donor–patient pairs and the non-mismatched pairs was performed using the chi-squared test for qualitative variables and Student's t-test for continuous variables. Univariate and multivariate logistic regression models were used to analyse the association between risk factors and the probability of acute GvHD, chronic GvHD, relapse or mortality. Actuarial probability of acute GvHD, disease-free survival and overall survival were calculated using the method of Kaplan–Meier (Kaplan & Meier, 1958). Comparison of curves was calculated using the log-rank test.


  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Thirty-four of the 215 patient–donor pairs had a mHA HA-1 disparity (15·8%). HLA-A2 subtyping of these 34 pairs confirmed the presence of HLA-A*0201 in 31 cases. The remaining three cases expressed the HLA-A*0205 allele and were excluded from the statistical analysis.

Table II shows the comparison between the pairs carrying the HA-1-antigen mismatch and those non-carriers. No statistical differences were detected for demographic data or risk factors of acute GvHD.

Table II.   Comparison between the HA-1 mismatched group and the HA-1 matched donor/patient pairs (n = 212).
 HA-1 mismatch (n = 31)No mismatch (n:181)P
  1. M, male; F, female; BM, bone marrow; CMV, cytomegalovirus; PB, peripheral blood.

Patient's age (mean ± S.D.)36·4 ± 12·735·9 ± 11·6N.S.
Donor's age (mean ± S.D.)35·3 ± 12·834·4 ± 12·7N.S.
Patient's sexM: 22 (71%)M: 111 (61·3%)N.S.
F: 9 (29%)F: 70 (38·7%) 
Patient male/donor female11 (35·5%)55 (30·4%)N.S.
Advanced disease6 (19·4%)41 (22·6%)N.S.
Patient CMV positive25 (80·6%)134 (74·0%)N.S.
Donor CMV positive23 (74·1%)132 (72·9%)N.S.
Donor with previous pregnancies5 (16·1%)31 (17·1%)N.S.
Source of stem cellsBM: 16 (51·6%)BM: 106 (58·6%)N.S.
PB: 15 (48·4%)PB: 75 (41·4%) 
T-cell depletion10 (32·3%)42 (23·2%)N.S.

Four patients could not be evaluated for acute GvHD. Sixteen patients within the HA-1-mismatched group (51·6%) developed grades II–IV acute GvHD, compared with 67 in the non-mismatched group (37%). This difference was not significant in a univariate logistic regression model (P: 0·129; OR: 1·81, 95% CI: 0·84–3·89), but showed statistical significance in the multivariate logistic regression model (P: 0·035; OR: 2·96, 95% CI: 1·07–8·14). Table III shows the results obtained in the multivariate analysis for grades II–IV acute GvHD: only HA-1 mismatch and donor women sensitized by previous pregnancies were identified as risk factors. Figure 1 shows the probability of grades II–IV acute GvHD on the basis of the presence or absence of mHA HA-1 mismatch. The acute GvHD incidence was higher in the HA-1-mismatched group, but comparison of the actuarial curves was not significant (P: 0·134).

Table III.   Multivariate logistic regression model for GvHD grades II–IV (n = 212 pairs).
 POR (95% CI)
Donor with previous pregnancies0·0202·87 (1·17–7·00)
Mismatch HA-10·0352·96 (1·07–8·14)
Prophylaxis with cyclosporin A + methotrexate0·0642·14 (0·95–4·81)
Donor CMV positive0·0692·35 (0·93–5·95)
T cell depletion0·443
Patient's age > 30 years0·233
Source of stem cells0·293
Patient CMV positive0·924
Advanced disease0·903
Donor female versus host male0·451

Figure 1.  Probability of grades II–IV acute GvHD on the basis of the presence or absence of mHA HA-1 mismatch. (A) Grades II–IV GvHD probability when all the patients in this study are included. (B) The incidence of grades II–IV acute GvHD when only the patients receiving a non-T cell-depleted graft are included.

Download figure to PowerPoint

As T-cell depletion is the most useful method for acute GvHD prevention, we decided to analyse the patients receiving a non-T cell-depleted graft (n = 160) separately. We detected grades II–IV acute GvHD in 12 patients in the HA-1 antigen-mismatched group (57·1%) compared with 51 (37·5%) in the non-mismatched group. This difference was statistically significant both in the comparison of actuarial curves (p: 0·05) and in the multivariate logistic regression model (P: 0·014; OR: 4·60, 95% CI: 1·35–15·69).

The incidence of grades III–IV acute GvHD was similar in both groups: 16·1% in the HA-1 mismatched pairs compared with 15·2% in the non-mismatched pairs (P: 0·891). The results did not change when the analysis was performed only on those patients receiving a non-T cell-depleted graft (23·8% versus 16·2%; P: 0·391).

The incidence of extensive chronic GvHD was slightly less (19·2%) in the mismatched cases than in the non-mismatched group (27%), but it did not reach statistical significance (P: 0·402).

The percentage of the patients mismatched for HA-1 who relapsed was 22·6%, which was similar to the incidence in the non-mismatched group (27·1%). Figure 2 shows the probability of relapse and disease-free survival on the basis of the HA-1 mismatch. The mHA HA-1 disparity did not affect the relapse rate even when the analysis was stratified according to the phase (early or advanced) of the disease or to the haematological disease [acute leukaemia or chronic myeloid leukaemia (CML)].


Figure 2.  (A) Probability of relapse on the basis of the HA-1 mismatch. (B) Disease-free survival on the basis of the HA-1 mismatch.

Download figure to PowerPoint

Finally, mortality was slightly higher (45·2% versus 38·1%) in the HA-1 mismatched group, but comparison between the actuarial curves did not show statistical significance (Fig 3).


Figure 3.  Actuarial survival for patients carrying the HA-1 mismatch versus those matched for the HA-1 antigen.

Download figure to PowerPoint

We did not find statistically significant differences in chronic GvHD, disease-free-survival or overall survival when the analysis was performed only on those patients receiving a non-T cell-depleted graft.


  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Our data confirm the results from previous studies about the relationship between HA-1 antigen disparity and acute GvHD (Goulmy et al, 1996; Tseng et al, 1999). However, we found a positive association of the HA-1 mismatch with grades II–IV acute GvHD, but not with grades III–IV, suggesting that the alloreactive response against mHA HA-1 produces only mild GvHD that usually responds well to the treatment.

HA-1 antigen expression is restricted to the haematopoietic cells. One hypothesis to explain the increased risk of acute GvHD for patients with HA-1 mismatch could be the expression of this antigen on the host antigen-presenting cells (Kupffer cells in the liver, Langerhans cells in the skin or small bowel macrophages).

Surprisingly, we did not find an increased risk of chronic GvHD in the HA-1-mismatched group. Development of acute GvHD is the main risk factor for developing chronic GvHD (Atkinson et al, 1990) and it seems clear that patients with HA-1 mismatch more frequently develop acute GvHD, but chronic GvHD incidence was even lower than in non-mismatched patients. One hypothesis to justify these results could be the ability of the HA-1H peptide to stimulate donor CD8+ T lymphocytes producing an inflammatory response through direct cytotoxicity and liberation of proinflammatory cytokines (Th1 profile), without inducing a Th2 response, characteristic of chronic GvHD (Krenger & Ferrara, 1996).

We found no difference in relapse-free survival between patients carrying the mHA HA-1 mismatch and those non-mismatched. As previously mentioned, HA-1 expression is restricted to the haematopoietic lineage and for this reason the HA-1 antigen has been considered a potential target for immunotherapy approaches. Recently, cytotoxic T lymphocytes (CTLs) against HA-1 have been found and quantified in patients with GvHD using tetrameric HLA complexes (Mutis et al, 1999a), and so demonstrated the existence of a specific alloresponse. The same authors showed the feasibility of generating ex vivo anti-HA1-specific CTLs as immunotherapy against haematological malignancies (Mutis et al, 1999b).

Our results indicate that, despite the expression of HA-1H on the host haematopoietic cells and the generation of a graft-versus-host reaction, the alloreactive response does not have any greater effect on the eradication of the minimal residual disease than the non-mismatched group. To detect confounding factors, we performed separate disease-free survival curves according to the phase of the disease at transplantation. No significant differences were detected.

As demonstrated by the experience with donor lymphocyte infusion, CML is a model of good response to allogeneic immunotherapy, while other haematological malignancies such as acute lymphoblastic leukaemia (ALL) do not perform so well (Kolb et al, 1995; Collins et al, 1997, 2000). For this reason we studied whether there was any difference in disease-free survival between the HA-1-mismatched cases and the HA-1-matched group according to the underlying disease. We could not find any difference between the two groups for acute leukaemia or CML.

The reason for the absence of an increased graft-versus-leukaemia despite HA-1-antigen mismatch could be the downregulation of the expression of the HLA-A*0201 molecules or the lack of expression of co-stimulatory molecules by leukaemic cells, as described for other malignancies (Garrido et al, 1993; Dorfman et al, 1997). Further studies of the feasibility of HA-1-based immunotherapy are needed before conclusions can be reached.

The overall survival of the HA-1-mismatched group is slightly lower than the matched group, without reaching statistical significance. This could be attributable to the increased incidence of acute GvHD observed. The higher mortality may be associated with the immunosuppression caused by the GvHD treatment, as severe acute GvHD incidence is similar in both groups.

In conclusion, mHA HA-1 disparity is a significant risk factor of acute GvHD after HLA-identical sibling donor SCT. Nevertheless, the incidence of severe acute GvHD or chronic GvHD is not affected by HA-1-antigen mismatch between patient and donor. Disease-free survival or overall survival are not influenced by the presence of a HA-1 mismatch. Further studies are needed to ascertain the real impact of HA-1 disparity on relapse and overall survival after HLA-identical sibling donor stem cell transplantation.


  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The Alloreactivity Unit worked in co-operation with the José Carreras Leukämie Stiftüng e.V. Dr J. I. Aróstegui is a recipient of a grant from the Fundació Catalana de Transplantament. Dr M. Rodríguez-Luaces is a recipient of grant BEFI 99/9136, from the Health Ministry of Spain. The authors thank Mrs. Mayte Encuentra for her assistance in the statistical analysis.


  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  • Argüello, R., Little, A.M., Pay, A.L., Gallardo, D., Rojas, I., Marsh, S.G.E., Goldman, J.M., Madrigal, J.A. (1998a) Mutation detection and typing of polymorphic loci through double-strand conformation analysis. Nature Genetics, 18, 192194.
  • Argüello, R., Little, A.M., Bohan, E., Goldman, J.M., Marsh, S.G.E., Madrigal, A. (1998b) High resolution HLA class I typing by reference strand mediated conformation analysis (RSCA). Tissue Antigens, 52, 5766.
  • Aróstegui, J.I., Gallardo, D., Rodríguez-Luaces, M., Querol, S., Madrigal, J.A., García-López, J., Grañena, A. (2000) Genomic typing of minor histocompatibility antigen HA-1 by Reference Strand Mediated Conformation Analysis (RSCA). Tissue Antigens, 56, 6976.DOI: 10.1034/j.1399-0039.2000.560109.x
  • Atkinson, K., Horowitz, M.M., Gale, R.P., Van Bekkum, D.W., Gluckman, E., Good, R.A., Jacobsen, N., Kolb, H.J., Rimm, A.A., Ringden, O. (1990) Risk factors for chronic graft-versus-host disease after HLA-identical sibling bone marrow transplantation. Blood, 75, 24592464.
  • Collins, R.H., Shpilberg, O., Drobyski, W.R., Porter, D.L., Giralt, S., Champlin, R., Goodman, S.A., Wolff, S.N., Hu, W., Verfaillie, C., List, A., Dalton, W., Ognoskie, N., Chetrit, A., Antin, J.H., Nemunaitis, J. (1997) Donor leukocyte transfusions in 140 patients with relapsed malignancy after allogeneic bone marrow transplantation. Journal of Clinical Oncology, 15, 433444.
  • Collins, R.H., Goldstein, S., Giralt, S., Levine, J., Porter, D., Drobyski, W., Barret, J., Johnson, M., Kirk, A., Horowitz, M., Parker, P. (2000) Donor leukocyte infusions in acute lymphocytic leukaemia. Bone Marrow Transplantation, 26, 511516.
  • Den Haan, J.M., Meadows, L.M., Wang, W., Pool, J., Blokland, E., Bishop, T.L., Reinhardus, C., Shabanowitz, J., Offringa, R., Hunt, D.F., Engelhard, V.H., Goulmy, E. (1998) The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science, 279, 10541057.DOI: 10.1126/science.279.5353.1054
  • Dorfman, D.M., Schultze, J.L., Shahsafaei, A., Michalak, S., Gribben, J.G., Freeman, G.J., Pinkus, G.S., Nadler, L.M. (1997) In vivo expression of B7–1 and B7–2 on follicular lymphoma is not sufficient to induce significant T cell proliferation but is sufficient to prevent T cell anergy in vitro. Blood, 90, 42974306.
  • Garrido, F., Cabrera, T., Concha, A., Glew, S., Ruiz-Cabello, F., Stern, P.L. (1993) Natural history of HLA expression during tumor development. Immunology Today, 14, 491499.
  • Glucksberg, H., Storb, R., Fefer, A., Buckner, C.D., Neiman, P.E., Clift, R.A., Lerner, K.G., Thomas, E.D. (1974) Clinical manifestation of graft-versus-host disease in human recipients of marrow from HLA matched sibling donors. Transplantation, 18, 295304.
  • Goulmy, E. (1997) Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunology Reviews, 157, 125140.
  • Goulmy, E., Schipper, R., Pool, J., Blokland, E., Falkenburg, F., Vossen, J., Gratwohl, A., Vogelsang, G.B., Van Houwelingen, H.C., Van Rood, J.J. (1996) Mismatches of minor histocompatibility antigens between HLA-identical donors and recipients and the development of graft-versus-host disease after bone marrow transplantation. New England Journal of Medicine, 334, 281285.
  • Kaplan, E. & Meier, P. (1958) Non-parametric estimation from incomplete observations. Journal of the American Statistical Association, 53, 457481.
  • Kolb, H.J., Schattenberg, A., Goldman, J.M., Herstentein, B., Jacobsen, N., Arcese, W., Ljungman, P., Ferrant, A., Verdonck, L., Niederwieser, D. (1995) Graft-versus-leukaemia effect of donor lymphocyte transfusions in marrow grafted patients. Blood, 86, 20412050.
  • Krenger, W. & Ferrara, J.L. (1996) Graft-versus-host disease and the Th1/Th2 paradigm. Immunology Research, 15, 5073.
  • Mutis, T., Gillespie, G., Schrama, E., Falkenburg, J.H., Moss, P., Goulmy, E. (1999a) Tetrameric HLA class I-minor histocompatibility antigen peptide complexes demonstrate minor histocompatibility antigen-specific cytotoxic T lymphocytes in patients with graft-versus-host disease. Nature Medicine, 5, 839842.DOI: 10.1038/10563
  • Mutis, T., Verdijk, R., Schrama, E., Esendam, B., Brand, A., Goulmy, E. (1999b) Feasibility of immunotherapy of relapsed leukaemia with ex vivo-generated cytotoxic T lymphocytes specific for hematopoietic system-restricted minor histocompatibility antigens. Blood, 93, 23362341.
  • Perreault, C., Decary, F., Brochu, S., Gyger, M., Belanger, R., Roy, D. (1990) Minor histocompatibility antigens. Blood, 76, 12691280.
  • Tseng, L.H., Lin, M.T., Hansen, J.A., Gooley, T., Pei, J., Smith, A.G., Martin, E.G., Petersdorf, E.W., Martin, P.J. (1999) Correlation between disparity for the minor histocompatibility antigen HA-1 and the development of acute graft-versus-host disease after allogeneic marrow transplantation. Blood, 94, 29112914.
  • Warren, E.H., Greenberg, P.D., Riddell, S.R. (1998) Cytotoxic T-lymphocyte-defined human minor histocompatibility antigens with a restricted tissue distribution. Blood, 91, 21972207.
  • Wilke, M., Pool, J., Den Haan, J.M., Goulmy, E. (1998) Genomic identification of the minor histocompatibility antigen HA-1 locus by allele-specific PCR. Tissue Antigens, 52, 312317.