- Email: [email protected]
Pemphigus Vulgaris in White Europeans Is Linked with HLA Class II Allele HLA DRB1*1454 but Not DRB1*1401
M Saha et al. HLA Class II Allele Distribution
to a subsequently diminished recruitment of T cells. Tacrolimus could also have a direct effect on CCR10 expression by T cells, as has been shown for other immunosuppressive drugs (Moed et al., 2004; Park et al., 2005; Caproni et al., 2007). However, we were unable to carry out these experiments. In summary, our findings illustrate that topical tacrolimus treatment has the potential to reduce a CD4/CD8 T-cellmediated skin inflammatory process by the downregulation of skin-specific chemokines and chemokine receptors. Informed consent was obtained from parents regarding the use of photographs. Institutional approval is provided to use leftover biological material obtained for diagnostic purposes for related research in line with the Helsinki guidelines. CONFLICT OF INTEREST The authors state no conflict of interest.
ACKNOWLEDGMENTS We thank Anton W. Langerak from the department of Immunology of the Erasmus Medical Centre (Rotterdam, the Netherlands) for his technical assistance. This work was funded by Stichting The Quality of Life gala 2007 and the Dutch Cancer Society (UL 2005-3657).
Claudia M. Faaij1,3, Nicola E. Annels1,3, Geertje Ruigrok1, Mirjam van der Burg2, Lynne M. Ball1, Robbert G. Bredius1, Maarten J. van Tol1 and Arjan C. Lankester1 1
Division of Immunology, Haematology, Oncology Section, Bone Marrow Transplantation and Autoimmune Diseases, Department of Paediatrics, Leiden University Medical Centre, Leiden, The Netherlands and
2 Department of Immunology, Erasmus MC, University Medical Centre Rotterdam, Rotterdam, The Netherlands E-mail: [email protected] 3 These authors contributed equally to this work
REFERENCES Antoine C, Muller S, Cant A, Cavazzana-Calvo M, Veys P, Vossen J et al. (2003) Long-term survival and transplantation of haemopoietic stem cells for immunodeficiencies: report of the European experience 1968–1999. Lancet 361:553–60 Caproni M, Torchia D, Antiga E, Terranova M, Volpi W, del BE et al. (2007) The comparative effects of tacrolimus and hydrocortisone in adult atopic dermatitis: an immunohistochemical study. Br J Dermatol 156:312–9 Carroll CL, Fleischer AB Jr (2004) Tacrolimus ointment: the treatment of atopic dermatitis and other inflammatory cutaneous disease. Expert Opin Pharmacother 5:2127–37 Faaij CM, Lankester AC, Spierings E, Hoogeboom M, Bowman EP, Bierings M et al. (2006) A possible role for CCL27/CTACK-CCR10 interaction in recruiting CD4 T cells to skin in human graft-versus-host disease. Br J Haematol 133:538–49 Harville TO, Adams DM, Howard TA, Ware RE (1997) Oligoclonal expansion of CD45RO+ T lymphocytes in Omenn syndrome. J Clin Immunol 17:322–32 Ho S, Clipstone N, Timmermann L, Northrop J, Graef I, Fiorentino D et al. (1996) The mechanism of action of cyclosporin A and FK506. Clin Immunol Immunopathol 80: S40–5
Hwang ST (2001) Mechanisms of T-cell homing to skin. Adv Dermatol 17:211–41 Mazzolari E, Moshous D, Forino C, De MD, Offer C, Lanfranchi A et al. (2005) Hematopoietic stem cell transplantation in Omenn syndrome: a single-center experience. Bone Marrow Transplant 36:107–14 Moed H, Stoof TJ, Boorsma DM, von Blomberg BM, Gibbs S, Bruynzeel DP et al. (2004) Identification of anti-inflammatory drugs according to their capacity to suppress type-1 and type-2 T cell profiles. Clin Exp Allergy 34:1868–75 Park CW, Lee BH, Han HJ, Lee CH, Ahn HK (2005) Tacrolimus decreases the expression of eotaxin, CCR3, RANTES and interleukin-5 in atopic dermatitis. Br J Dermatol 152: 1173–81 Rego S, Kemp A, Wong M, Knight P (2006) Omenn syndrome: therapeutic effects of cyclosporin. J Paediatr Child Health 42: 319–20 Rieux-Laucat F, Bahadoran P, Brousse N, Selz F, Fischer A, Le DF et al. (1998) Highly restricted human T cell repertoire in peripheral blood and tissue-infiltrating lymphocytes in Omenn’s syndrome. J Clin Invest 102:312–21 Santamaria Babi LF, Perez Soler MT, Hauser C, Blaser K (1995) Skin-homing T cells in human cutaneous allergic inflammation. Immunol Res 14:317–24 Vestergaard C, Johansen C, Christensen U, Just H, Hohwy T, Deleuran M (2004) TARC augments TNF-alpha-induced CTACK production in keratinocytes. Exp Dermatol 13:551–7
Homey B, Alenius H, Muller A, Soto H, Bowman EP, Yuan W et al. (2002) CCL27-CCR10 interactions regulate T cell-mediated skin inflammation. Nat Med 8:157–65
Vestergaard C, Johansen C, Otkjaer K, Deleuran M, Iversen L (2005) Tumor necrosis factor-alphainduced CTACK/CCL27 (cutaneous T-cellattracting chemokine) production in keratinocytes is controlled by nuclear factor kappaB. Cytokine 29:49–55
Hudak S, Hagen M, Liu Y, Catron D, Oldham E, McEvoy LM et al. (2002) Immune surveillance and effector functions of CCR10(+) skin homing T cells. J Immunol 169:1189–96
Villa A, Santagata S, Bozzi F, Imberti L, Notarangelo LD (1999) Omenn syndrome: a disorder of Rag1 and Rag2 genes. J Clin Immunol 19:87–97
Pemphigus Vulgaris in White Europeans Is Linked with HLA Class II Allele HLA DRB1*1454 but Not DRB1*1401 Journal of Investigative Dermatology (2010) 130, 311–314; doi:10.1038/jid.2009.241; published online 22 October 2009
TO THE EDITOR Population studies have consistently shown a link between certain class II
Abbreviation: PV, pemphigus vulgaris
HLA alleles and ethnic groups in patients with pemphigus vulgaris (PV). In particular, HLA DRB1*04 and *14 and
DQB1*0503 and *0302 alleles (Tron et al., 2005) have been shown to be strongly associated with PV. Although previous studies have suggested that PV is associated with DRB1*1401 in a number of populations from Europe and www.jidonline.org
311
M Saha et al. HLA Class II Allele Distribution
Table 1. HLA DRB1 and DQB1 allele frequencies in (a) white Europeans and (b) Indo-Asians with pemphigus vulgaris and in matched controls1 (a) WE allele count Allele
Patients (n=96)
WE allele frequency %
Controls (n=100)
Patients (n=96)
Controls (n=100)
Significance P-value, OR, CI
DRB1 8
29
4.2
14.5
P=0.0009, OR 0.26, CI (0.1, 0.61)
*04
*0301
71
29
37.0
14.5
Po0.00001, OR 3.46, CI (2.06, 5.83)
*0402
52
0
27.1
0.0
*0701
14
29
7.3
14.5
*14
44
8
22.9
4.0
Po0.00001, OR 7.14, CI (3.11, 10.96)
2
0
1.0
0.0
NS
*1404
9
0
5.2
0.0
P=0.0009, OR 9.89, CI (1.27, 210.33)
*1454
32
8
16.7
4.0
P=0.00007, OR 4.8, CI (2.05, 11.65)
8
25
4.2
12.5
*1401
*15
Po0.00001, OR undefined *P=0.034, OR 0.46, CI (0.22, 0.95)
P=0.005, OR 0.3, CI (0.12, 0.73)
DQB1 *02
19
51
9.9
25.5
P=0.0001, OR 0.32, CI (0.17, 0.59)
*0302
63
18
32.8
9.0
Po0.00001, OR 4.94, CI (2.7, 9.11)
*0303
3
13
1.6
6.5
*P=0.027, OR 0.23, CI (0.05, 0.87)
*0501
12
28
6.3
14.0
*P=0.018, OR 0.41, CI (0.19, 0.87)
*0503
43
8
22.4
4.0
Po0.00001, OR 6.93, CI (3.03, 16.48)
*0601
4
1
2.1
0.5
NS
*0602
6
23
3.1
11.5
*P=0.0029, OR 0.25, CI (0.09, 0.66)
(b) IA allele count
IA allele frequency %
Significance
Patients (n=57)
Controls (n=59)
Patients (n=57)
Controls (n=59)
P-value, OR, CI
1
17
0.9
14.4
*04
23
8
20.2
6.8
*P=0.002, OR 4.01, CI (1.54,10.8)
*0402
16
1
14.0
0.8
P=0.0003, OR 19.10, CI (2.59,393)
DRB1 *0301
*0701
P=0.0003, OR 0.05, CI (0.0, 0.38
8
14
7.0
11.7
NS
53
14
46.5
11.9
Po0.00001, OR 6.45, CI (3.16,13.35)
*1401
1
0
0.9
0.0
*1404
48
8
42.1
6.8
*1454
3
5
2.6
4.2
*15
5
25
4.4
21.2
P=0.0003,OR 0.17, CI (0.05, 0.49)
7
30
6.1
25.4
P=0.0001, OR 0.19, CI (0.07, 0.49)
*14
NS Po0.00001, OR 10, CI (4.2, 24.5) NS
DQB1 *02 *0302
18
8
15.8
6.8
*0303
2
2
1.8
1.7
*P=0.049, OR 2.58, CI (1.0, 6.8)
*0501
4
10
3.5
8.5
NS
*0503
54
15
47.4
12.7
Po0.00001, OR 6.18, CI (3.07, 12.57)
*0601
5
18
4.4
15.3
*P=0.011, OR 0.25, CI (0.08, 0.76)
*0602
0
3
0.0
2.5
NS
NS
CI, confidence interval; IA, Indo-Asian; NS, not significant; OR, odds ratio; WE, white European. *P=not significant after Bonferonni correction. 1 HLA DRB1 *0101, *0103, *0401, *0403, *0404, *0405, *0406, *0407,*0408, *0410, *08, *09, *10,*11,*12,*13, *1406/7,*1415,*1428, *16 and DQB1*0301, *04, *0502, *0603, *0604 were also tested, but allele frequencies were not significantly different in either group compared with controls.
312
Journal of Investigative Dermatology (2010), Volume 130
M Saha et al. HLA Class II Allele Distribution
Asia, including Italy (Lombardi et al., 1999), France (Loiseau et al., 2000), and Japan (Miyagawa et al., 1997), recent data suggest that alleles previously classified as DRB1*1401 may actually represent the novel allele DRB1*1454, which was identified while exploring sequence diversity in exon 3 of HLA DRB1 (Horn et al., 2007) and was assigned DRB1*1454 by the World Health Organization Nomenclature Committee in October 2005. In this study, we sought to determine HLA DRB1 (including DRB1*1454) and DQB1 allele frequencies in PV patients living in the United Kingdom from two major ethnic groups:white Europeans and Indo-Asians. A total of 153 patients with PV—96 of white European (WE) descent and 57 Indo-Asians (IAs, from India, Bangladesh, Pakistan, Sri Lanka, and Tanzania)—were recruited from two centers in the United Kingdom. A cohort of normal controls with no history of skin or autoimmune disease were matched on the basis of ethnic origin (100 WE, 59 IA). Medium- to high-resolution HLA class II typing was performed using PCR with sequence-specific primers on DNA samples of patients and controls. In addition, DRB1*14 subtyping was performed on samples positive for HLA-DRB1*14 using the Dynal AllSet SSP DRB1*14 kit (Dynal Biotech, Invitrogen, Wirral, UK). In both ethnic groups, HLA DRB1*04 and *14 and DQB1*0503 and *0302 alleles were significantly associated with PV (Table 1a and b), consistent with the results of previous studies (Tron et al., 2005). In WE patients (Table 1a), HLA DRB1*0402 was the most significantly overexpressed allele, with a frequency of 27.1% in patients compared with 0% in controls (Po0.00001, OR undefined). In contrast to previous studies (Miyagawa et al., 1997; Lombardi et al., 1999; Loiseau et al., 2000), we could not identify any association of DRB1*1401 with PV. However, the DRB1*1454 allele was significantly associated with WE patients (P ¼ 0.00007). Other significantly overexpressed alleles in the WE patient group included DRB1*1404 (P ¼ 0.0009), DQB1*0302 (Po0.00001), and DQB1*0503 (Po0.00001).
As with WE patients, we found significantly higher frequencies of HLA DRB1*0402 (P ¼ 0.0003)* and *1404 (Po 0.00001) and DQB1*0503 (Po0.00001) and *0302 (P ¼ 0.049) alleles in IA PV patients than in the IA control group (Table 1b). Notably, DRB1*1404 was the most significantly overexpressed allele overall in IA patients (42.1% compared with 6.8% in the IA controls). As in WE patients, we could not identify any association with HLA DRB1*1401. However, DRB1*1454 was not observed to a significant degree in IA PV patients. In both ethnic groups, certain DRB1 and DQB1 alleles were in negative association with PV, perhaps suggesting a protective effect. In WE patients, there was a significant decrease in the DQB1*02 allele frequency in PV patients (25.5% in controls, 9.9% in patients, P ¼ 0.0001) and in DRB1*0301 (P ¼ 0.0009) in the WE control group. In IA PV patients, there were also significant reductions in DQB1*02 (P ¼ 0.0001) and DRB1*0301 (P ¼ 0.0003) allele frequencies compared with those in controls. In addition, DRB1*15 also had a lower frequency compared with controls (P ¼ 0.0003). The DRB1 locus is highly polymorphic, with more than 300 alleles. Horn et al. (2007) recently showed that in some samples previously typed as DRB1*1401, a different exon 3 sequence was identified, differing from DRB1*1401 at nucleotide position 51 of exon 3. This results in a change of codon 112 TAC to CAC, resulting in a tyrosine-to-histidine exchange in the transcribed protein. This allele has been named DRB1*1454. Our data clearly show that HLA DRB1*1454 is strongly associated with PV in WE patients but not in IA patients. We only found one patient with HLA DRB1*1401 in the IA patient group and two patients in the WE patient group, despite this allele having been reported numerous times as being associated with PV in a number of populations. It seems likely that all the previously recognized associations of pemphigus with DRB1*1401 are actually with DRB1*1454. Conventional assays by PCR with sequencespecific primers have typed DRB1*14 alleles on the basis of exon 2 and will not have differentiated DRB1*1401 and
DRB1*1454. The use of exon 3 primers now allows for a distinction of these alleles, and we recommend that these be used in future association studies. However, it is unclear whether the amino-acid change associated with DRB1*1454 has an effect on the binding of target antigens in pemphigus; further work is necessary to elucidate this. HLA DRB1*14 and *04 alleles were the most significant alleles associated with PV in both the IA and WE groups. HLA DRB1*0402 was the strongest risk factor in WE patients, whereas HLA DRB1*1404 had the strongest association in IA patients, underscoring the important genetic differences between these groups. Although HLA DQB1* 0503 was significantly associated with PV in both groups, this may well be a reflection of its linkage disequilibrium with DRB1*14 alleles. Similarly, DQB1*0302 is in linkage disequilibrium with DRB1*04. This study represents, to our knowledge, the largest cohort of white European and Indo-Asian patients with PV in whom class II HLA typing has been published to date. We have confirmed previously published associations with the alleles HLA DRB1*0402 and 1404 and DQB1*0302 and 0503, with DRB1*1404 being the strongest risk factor in the IA group and DRB1*0402 being the strongest risk factor in the WE group. We have also shown a significant association with a novel allele, DRB1*1454, in white European individuals with pemphigus vulgaris. Finally, our data indicate that genetic susceptibility differences in ethnic groups are maintained in patients living away from their countries of ethnic origin, underlining the importance of genetic risk factors in this disease. CONFLICT OF INTEREST The authors state no conflict of interest.
ACKNOWLEDGMENTS This work was generously supported by the British Skin Foundation and the Comprehensive Biomedical Research Centre at KCL/GSTT. We thank the PV Network for facilitating the participation of their members and colleagues who provided information on their patients.
Monika Saha1,4, Karen Harman2,4, Neil J. Mortimer2, Valentina Binda3, Martin M. Black1, Ellie Kondeatis3, www.jidonline.org
313
P Terheyden et al. Imatinib Therapy of Melanoma
Robert Vaughan3 and Richard W. Groves1 1
Department of Immunodermatology, St John’s Institute of Dermatology, King’s College London, London, UK; 2Department of Dermatology, University Hospitals, Leicester, UK and 3Clinical Transplantation Laboratory, Guy’s Hospital, King’s College London, London, UK E-mail: [email protected] 4 These authors contributed equally to this work.
human leukocyte antigen alleles in pemphigus vulgaris and pemphigus foliaceus Italian patients. J Invest Dermatol 113:107–10
REFERENCES Horn PA, bis-Camps M, Verboom M, Bunce M, Yousaf K, Williams S et al. (2007) The nature of diversity of HLA-DRB1 exon 3. Tissue Antigens 70:335–7 Loiseau P, Lecleach L, Prost C, Lepage V, Busson M, Bastuji-Garin S et al. (2000) HLA class II polymorphism contributes to specify desmoglein derived peptides in pemphigus vulgaris and pemphigus foliaceus. J Autoimmun 15:67–73 Lombardi ML, Mercuro O, Ruocco V, Lo SA, Lombari V, Guerrera V et al. (1999) Common
Miyagawa S, Higashimine I, Iida T, Yamashina Y, Fukumoto T, Shirai T. (1997) HLA-DRB1*04 and DRB1*14 alleles are associated with susceptibility to pemphigus among Japanese. J Invest Dermatol 109:615–8 Tron
F, Gilbert D, Mouquet H, Joly P, Drouot L, Makni S et al. (2005) Genetic factors in pemphigus. J Autoimmun 24:319–28
Response to Imatinib Mesylate Depends on the Presence of the V559A-Mutated KIT Oncogene Journal of Investigative Dermatology (2010) 130, 314–316; doi:10.1038/jid.2009.197; published online 8 October 2009
TO THE EDITOR Identification of markers that predict the response to a targeted therapy is a key goal for the treatment of melanoma. A major advance was the observation that certain subtypes, such as mucous membrane melanoma and acral lentiginous melanoma, are frequently associated with activating mutations in the KIT oncogene (Curtin et al., 2006). Imatinib mesylate, a small-molecule inhibitor of protein tyrosine kinases, has been used with considerable success in the treatment of gastrointestinal stromal tumors, which are induced by abnormal activation and overexpression of the KIT receptor in 90% of cases (Sleifjer et al., 2008). Recent reports have described the usefulness of this molecule in the treatment of rectal (Hodi et al., 2008) and acral lentiginous melanoma (Kim et al., 2008). Here, we report a patient with a rapid response to imatinib in a lung metastasis from an acral lentiginous melanoma. Surprisingly, this was paralleled by progressive disease of locoregional lymph node metastasis. A 64-year-old woman was admitted because of progressive severe dyspnea at rest. A pulmonary mass obstructing the left main bronchus was revealed by computed tomographic scanning (Figure 1a). Grayish pigmentation on
her left big toe revealed an acral lentiginous melanoma. Computed tomographic imaging of the abdomen and pelvis excluded visceral metastases but revealed lymph node metastases in the left ilioinguinal region (Figure 1b). Biopsy specimens were excised from the skin and lymph nodes and by incision via endoluminal access from the pulmonary mass. Probes were processed for routine histology and immunohistochemistry and preserved for molecular biology. Our studies were conducted after obtaining written informed consent and institutional review board approval and were in compliance with the Declaration of Helsinki Principles. Histopathology of the skin pigmentation confirmed the diagnosis and revealed parts of an acral lentiginous melanoma (not shown). Histopathology of both the lung (Figure 1c) and lymph node (Figure 1d) biopsies showed highly proliferative atypical epithelioid cells. Expression analysis by immunohistochemistry using a specific monoclonal mouse antibody against KIT (CD117) revealed that both metastases showed the presence of KIT; however, the lung specimen (Figure 1e) expressed the protein more brightly and uniformly than did the lymph node metastasis (Figure 1f). Furthermore, S100, Melan-A,
Abbreviation: CT, computed tomography
314
Journal of Investigative Dermatology (2010), Volume 130
and HMB-45 were positively stained in both probes (Figures 1g and h). We initiated monotherapy with imatinib mesylate (Glivec, Novartis Pharma Nu¨rnberg, Germany) 400 mg once daily. After 6 weeks, radiological reassessment of the chest showed regression of the pulmonary mass; the largest diameter had decreased from 3.2 to 2.0 cm. Moreover, the atelectasis of the left upper lobe had reversed completely (Figure 1i). The performance status was remarkably improved. However, the lymph node metastasis in the left groin showed marked progression; the largest diameter had increased from 3.0 to 5.8 cm in the inguinal node (Figure 1j) and from 2 to 3 cm in the iliacal node (not shown). In an effort to understand this differential biological behavior of the two tumor manifestations, we investigated the KIT mutation status of the primary tumor, the lymph node, and the lung metastases. To this end, genomic DNA was extracted and exons 11, 13, 17, and 18 of the KIT gene, which harbor the vast majority of mutations identified in human tumors, were analyzed by PCR amplification followed by direct sequencing. Interestingly, in both the primary tumor and the lymph node metastasis, we detected only wild-type sequences, whereas in the lung metastasis, a homo- or hemizygous nucleotide
to a subsequently diminished recruitment of T cells. Tacrolimus could also have a direct effect on CCR10 expression by T cells, as has been shown for other immunosuppressive drugs (Moed et al., 2004; Park et al., 2005; Caproni et al., 2007). However, we were unable to carry out these experiments. In summary, our findings illustrate that topical tacrolimus treatment has the potential to reduce a CD4/CD8 T-cellmediated skin inflammatory process by the downregulation of skin-specific chemokines and chemokine receptors. Informed consent was obtained from parents regarding the use of photographs. Institutional approval is provided to use leftover biological material obtained for diagnostic purposes for related research in line with the Helsinki guidelines. CONFLICT OF INTEREST The authors state no conflict of interest.
ACKNOWLEDGMENTS We thank Anton W. Langerak from the department of Immunology of the Erasmus Medical Centre (Rotterdam, the Netherlands) for his technical assistance. This work was funded by Stichting The Quality of Life gala 2007 and the Dutch Cancer Society (UL 2005-3657).
Claudia M. Faaij1,3, Nicola E. Annels1,3, Geertje Ruigrok1, Mirjam van der Burg2, Lynne M. Ball1, Robbert G. Bredius1, Maarten J. van Tol1 and Arjan C. Lankester1 1
Division of Immunology, Haematology, Oncology Section, Bone Marrow Transplantation and Autoimmune Diseases, Department of Paediatrics, Leiden University Medical Centre, Leiden, The Netherlands and
2 Department of Immunology, Erasmus MC, University Medical Centre Rotterdam, Rotterdam, The Netherlands E-mail: [email protected] 3 These authors contributed equally to this work
REFERENCES Antoine C, Muller S, Cant A, Cavazzana-Calvo M, Veys P, Vossen J et al. (2003) Long-term survival and transplantation of haemopoietic stem cells for immunodeficiencies: report of the European experience 1968–1999. Lancet 361:553–60 Caproni M, Torchia D, Antiga E, Terranova M, Volpi W, del BE et al. (2007) The comparative effects of tacrolimus and hydrocortisone in adult atopic dermatitis: an immunohistochemical study. Br J Dermatol 156:312–9 Carroll CL, Fleischer AB Jr (2004) Tacrolimus ointment: the treatment of atopic dermatitis and other inflammatory cutaneous disease. Expert Opin Pharmacother 5:2127–37 Faaij CM, Lankester AC, Spierings E, Hoogeboom M, Bowman EP, Bierings M et al. (2006) A possible role for CCL27/CTACK-CCR10 interaction in recruiting CD4 T cells to skin in human graft-versus-host disease. Br J Haematol 133:538–49 Harville TO, Adams DM, Howard TA, Ware RE (1997) Oligoclonal expansion of CD45RO+ T lymphocytes in Omenn syndrome. J Clin Immunol 17:322–32 Ho S, Clipstone N, Timmermann L, Northrop J, Graef I, Fiorentino D et al. (1996) The mechanism of action of cyclosporin A and FK506. Clin Immunol Immunopathol 80: S40–5
Hwang ST (2001) Mechanisms of T-cell homing to skin. Adv Dermatol 17:211–41 Mazzolari E, Moshous D, Forino C, De MD, Offer C, Lanfranchi A et al. (2005) Hematopoietic stem cell transplantation in Omenn syndrome: a single-center experience. Bone Marrow Transplant 36:107–14 Moed H, Stoof TJ, Boorsma DM, von Blomberg BM, Gibbs S, Bruynzeel DP et al. (2004) Identification of anti-inflammatory drugs according to their capacity to suppress type-1 and type-2 T cell profiles. Clin Exp Allergy 34:1868–75 Park CW, Lee BH, Han HJ, Lee CH, Ahn HK (2005) Tacrolimus decreases the expression of eotaxin, CCR3, RANTES and interleukin-5 in atopic dermatitis. Br J Dermatol 152: 1173–81 Rego S, Kemp A, Wong M, Knight P (2006) Omenn syndrome: therapeutic effects of cyclosporin. J Paediatr Child Health 42: 319–20 Rieux-Laucat F, Bahadoran P, Brousse N, Selz F, Fischer A, Le DF et al. (1998) Highly restricted human T cell repertoire in peripheral blood and tissue-infiltrating lymphocytes in Omenn’s syndrome. J Clin Invest 102:312–21 Santamaria Babi LF, Perez Soler MT, Hauser C, Blaser K (1995) Skin-homing T cells in human cutaneous allergic inflammation. Immunol Res 14:317–24 Vestergaard C, Johansen C, Christensen U, Just H, Hohwy T, Deleuran M (2004) TARC augments TNF-alpha-induced CTACK production in keratinocytes. Exp Dermatol 13:551–7
Homey B, Alenius H, Muller A, Soto H, Bowman EP, Yuan W et al. (2002) CCL27-CCR10 interactions regulate T cell-mediated skin inflammation. Nat Med 8:157–65
Vestergaard C, Johansen C, Otkjaer K, Deleuran M, Iversen L (2005) Tumor necrosis factor-alphainduced CTACK/CCL27 (cutaneous T-cellattracting chemokine) production in keratinocytes is controlled by nuclear factor kappaB. Cytokine 29:49–55
Hudak S, Hagen M, Liu Y, Catron D, Oldham E, McEvoy LM et al. (2002) Immune surveillance and effector functions of CCR10(+) skin homing T cells. J Immunol 169:1189–96
Villa A, Santagata S, Bozzi F, Imberti L, Notarangelo LD (1999) Omenn syndrome: a disorder of Rag1 and Rag2 genes. J Clin Immunol 19:87–97
Pemphigus Vulgaris in White Europeans Is Linked with HLA Class II Allele HLA DRB1*1454 but Not DRB1*1401 Journal of Investigative Dermatology (2010) 130, 311–314; doi:10.1038/jid.2009.241; published online 22 October 2009
TO THE EDITOR Population studies have consistently shown a link between certain class II
Abbreviation: PV, pemphigus vulgaris
HLA alleles and ethnic groups in patients with pemphigus vulgaris (PV). In particular, HLA DRB1*04 and *14 and
DQB1*0503 and *0302 alleles (Tron et al., 2005) have been shown to be strongly associated with PV. Although previous studies have suggested that PV is associated with DRB1*1401 in a number of populations from Europe and www.jidonline.org
311
M Saha et al. HLA Class II Allele Distribution
Table 1. HLA DRB1 and DQB1 allele frequencies in (a) white Europeans and (b) Indo-Asians with pemphigus vulgaris and in matched controls1 (a) WE allele count Allele
Patients (n=96)
WE allele frequency %
Controls (n=100)
Patients (n=96)
Controls (n=100)
Significance P-value, OR, CI
DRB1 8
29
4.2
14.5
P=0.0009, OR 0.26, CI (0.1, 0.61)
*04
*0301
71
29
37.0
14.5
Po0.00001, OR 3.46, CI (2.06, 5.83)
*0402
52
0
27.1
0.0
*0701
14
29
7.3
14.5
*14
44
8
22.9
4.0
Po0.00001, OR 7.14, CI (3.11, 10.96)
2
0
1.0
0.0
NS
*1404
9
0
5.2
0.0
P=0.0009, OR 9.89, CI (1.27, 210.33)
*1454
32
8
16.7
4.0
P=0.00007, OR 4.8, CI (2.05, 11.65)
8
25
4.2
12.5
*1401
*15
Po0.00001, OR undefined *P=0.034, OR 0.46, CI (0.22, 0.95)
P=0.005, OR 0.3, CI (0.12, 0.73)
DQB1 *02
19
51
9.9
25.5
P=0.0001, OR 0.32, CI (0.17, 0.59)
*0302
63
18
32.8
9.0
Po0.00001, OR 4.94, CI (2.7, 9.11)
*0303
3
13
1.6
6.5
*P=0.027, OR 0.23, CI (0.05, 0.87)
*0501
12
28
6.3
14.0
*P=0.018, OR 0.41, CI (0.19, 0.87)
*0503
43
8
22.4
4.0
Po0.00001, OR 6.93, CI (3.03, 16.48)
*0601
4
1
2.1
0.5
NS
*0602
6
23
3.1
11.5
*P=0.0029, OR 0.25, CI (0.09, 0.66)
(b) IA allele count
IA allele frequency %
Significance
Patients (n=57)
Controls (n=59)
Patients (n=57)
Controls (n=59)
P-value, OR, CI
1
17
0.9
14.4
*04
23
8
20.2
6.8
*P=0.002, OR 4.01, CI (1.54,10.8)
*0402
16
1
14.0
0.8
P=0.0003, OR 19.10, CI (2.59,393)
DRB1 *0301
*0701
P=0.0003, OR 0.05, CI (0.0, 0.38
8
14
7.0
11.7
NS
53
14
46.5
11.9
Po0.00001, OR 6.45, CI (3.16,13.35)
*1401
1
0
0.9
0.0
*1404
48
8
42.1
6.8
*1454
3
5
2.6
4.2
*15
5
25
4.4
21.2
P=0.0003,OR 0.17, CI (0.05, 0.49)
7
30
6.1
25.4
P=0.0001, OR 0.19, CI (0.07, 0.49)
*14
NS Po0.00001, OR 10, CI (4.2, 24.5) NS
DQB1 *02 *0302
18
8
15.8
6.8
*0303
2
2
1.8
1.7
*P=0.049, OR 2.58, CI (1.0, 6.8)
*0501
4
10
3.5
8.5
NS
*0503
54
15
47.4
12.7
Po0.00001, OR 6.18, CI (3.07, 12.57)
*0601
5
18
4.4
15.3
*P=0.011, OR 0.25, CI (0.08, 0.76)
*0602
0
3
0.0
2.5
NS
NS
CI, confidence interval; IA, Indo-Asian; NS, not significant; OR, odds ratio; WE, white European. *P=not significant after Bonferonni correction. 1 HLA DRB1 *0101, *0103, *0401, *0403, *0404, *0405, *0406, *0407,*0408, *0410, *08, *09, *10,*11,*12,*13, *1406/7,*1415,*1428, *16 and DQB1*0301, *04, *0502, *0603, *0604 were also tested, but allele frequencies were not significantly different in either group compared with controls.
312
Journal of Investigative Dermatology (2010), Volume 130
M Saha et al. HLA Class II Allele Distribution
Asia, including Italy (Lombardi et al., 1999), France (Loiseau et al., 2000), and Japan (Miyagawa et al., 1997), recent data suggest that alleles previously classified as DRB1*1401 may actually represent the novel allele DRB1*1454, which was identified while exploring sequence diversity in exon 3 of HLA DRB1 (Horn et al., 2007) and was assigned DRB1*1454 by the World Health Organization Nomenclature Committee in October 2005. In this study, we sought to determine HLA DRB1 (including DRB1*1454) and DQB1 allele frequencies in PV patients living in the United Kingdom from two major ethnic groups:white Europeans and Indo-Asians. A total of 153 patients with PV—96 of white European (WE) descent and 57 Indo-Asians (IAs, from India, Bangladesh, Pakistan, Sri Lanka, and Tanzania)—were recruited from two centers in the United Kingdom. A cohort of normal controls with no history of skin or autoimmune disease were matched on the basis of ethnic origin (100 WE, 59 IA). Medium- to high-resolution HLA class II typing was performed using PCR with sequence-specific primers on DNA samples of patients and controls. In addition, DRB1*14 subtyping was performed on samples positive for HLA-DRB1*14 using the Dynal AllSet SSP DRB1*14 kit (Dynal Biotech, Invitrogen, Wirral, UK). In both ethnic groups, HLA DRB1*04 and *14 and DQB1*0503 and *0302 alleles were significantly associated with PV (Table 1a and b), consistent with the results of previous studies (Tron et al., 2005). In WE patients (Table 1a), HLA DRB1*0402 was the most significantly overexpressed allele, with a frequency of 27.1% in patients compared with 0% in controls (Po0.00001, OR undefined). In contrast to previous studies (Miyagawa et al., 1997; Lombardi et al., 1999; Loiseau et al., 2000), we could not identify any association of DRB1*1401 with PV. However, the DRB1*1454 allele was significantly associated with WE patients (P ¼ 0.00007). Other significantly overexpressed alleles in the WE patient group included DRB1*1404 (P ¼ 0.0009), DQB1*0302 (Po0.00001), and DQB1*0503 (Po0.00001).
As with WE patients, we found significantly higher frequencies of HLA DRB1*0402 (P ¼ 0.0003)* and *1404 (Po 0.00001) and DQB1*0503 (Po0.00001) and *0302 (P ¼ 0.049) alleles in IA PV patients than in the IA control group (Table 1b). Notably, DRB1*1404 was the most significantly overexpressed allele overall in IA patients (42.1% compared with 6.8% in the IA controls). As in WE patients, we could not identify any association with HLA DRB1*1401. However, DRB1*1454 was not observed to a significant degree in IA PV patients. In both ethnic groups, certain DRB1 and DQB1 alleles were in negative association with PV, perhaps suggesting a protective effect. In WE patients, there was a significant decrease in the DQB1*02 allele frequency in PV patients (25.5% in controls, 9.9% in patients, P ¼ 0.0001) and in DRB1*0301 (P ¼ 0.0009) in the WE control group. In IA PV patients, there were also significant reductions in DQB1*02 (P ¼ 0.0001) and DRB1*0301 (P ¼ 0.0003) allele frequencies compared with those in controls. In addition, DRB1*15 also had a lower frequency compared with controls (P ¼ 0.0003). The DRB1 locus is highly polymorphic, with more than 300 alleles. Horn et al. (2007) recently showed that in some samples previously typed as DRB1*1401, a different exon 3 sequence was identified, differing from DRB1*1401 at nucleotide position 51 of exon 3. This results in a change of codon 112 TAC to CAC, resulting in a tyrosine-to-histidine exchange in the transcribed protein. This allele has been named DRB1*1454. Our data clearly show that HLA DRB1*1454 is strongly associated with PV in WE patients but not in IA patients. We only found one patient with HLA DRB1*1401 in the IA patient group and two patients in the WE patient group, despite this allele having been reported numerous times as being associated with PV in a number of populations. It seems likely that all the previously recognized associations of pemphigus with DRB1*1401 are actually with DRB1*1454. Conventional assays by PCR with sequencespecific primers have typed DRB1*14 alleles on the basis of exon 2 and will not have differentiated DRB1*1401 and
DRB1*1454. The use of exon 3 primers now allows for a distinction of these alleles, and we recommend that these be used in future association studies. However, it is unclear whether the amino-acid change associated with DRB1*1454 has an effect on the binding of target antigens in pemphigus; further work is necessary to elucidate this. HLA DRB1*14 and *04 alleles were the most significant alleles associated with PV in both the IA and WE groups. HLA DRB1*0402 was the strongest risk factor in WE patients, whereas HLA DRB1*1404 had the strongest association in IA patients, underscoring the important genetic differences between these groups. Although HLA DQB1* 0503 was significantly associated with PV in both groups, this may well be a reflection of its linkage disequilibrium with DRB1*14 alleles. Similarly, DQB1*0302 is in linkage disequilibrium with DRB1*04. This study represents, to our knowledge, the largest cohort of white European and Indo-Asian patients with PV in whom class II HLA typing has been published to date. We have confirmed previously published associations with the alleles HLA DRB1*0402 and 1404 and DQB1*0302 and 0503, with DRB1*1404 being the strongest risk factor in the IA group and DRB1*0402 being the strongest risk factor in the WE group. We have also shown a significant association with a novel allele, DRB1*1454, in white European individuals with pemphigus vulgaris. Finally, our data indicate that genetic susceptibility differences in ethnic groups are maintained in patients living away from their countries of ethnic origin, underlining the importance of genetic risk factors in this disease. CONFLICT OF INTEREST The authors state no conflict of interest.
ACKNOWLEDGMENTS This work was generously supported by the British Skin Foundation and the Comprehensive Biomedical Research Centre at KCL/GSTT. We thank the PV Network for facilitating the participation of their members and colleagues who provided information on their patients.
Monika Saha1,4, Karen Harman2,4, Neil J. Mortimer2, Valentina Binda3, Martin M. Black1, Ellie Kondeatis3, www.jidonline.org
313
P Terheyden et al. Imatinib Therapy of Melanoma
Robert Vaughan3 and Richard W. Groves1 1
Department of Immunodermatology, St John’s Institute of Dermatology, King’s College London, London, UK; 2Department of Dermatology, University Hospitals, Leicester, UK and 3Clinical Transplantation Laboratory, Guy’s Hospital, King’s College London, London, UK E-mail: [email protected] 4 These authors contributed equally to this work.
human leukocyte antigen alleles in pemphigus vulgaris and pemphigus foliaceus Italian patients. J Invest Dermatol 113:107–10
REFERENCES Horn PA, bis-Camps M, Verboom M, Bunce M, Yousaf K, Williams S et al. (2007) The nature of diversity of HLA-DRB1 exon 3. Tissue Antigens 70:335–7 Loiseau P, Lecleach L, Prost C, Lepage V, Busson M, Bastuji-Garin S et al. (2000) HLA class II polymorphism contributes to specify desmoglein derived peptides in pemphigus vulgaris and pemphigus foliaceus. J Autoimmun 15:67–73 Lombardi ML, Mercuro O, Ruocco V, Lo SA, Lombari V, Guerrera V et al. (1999) Common
Miyagawa S, Higashimine I, Iida T, Yamashina Y, Fukumoto T, Shirai T. (1997) HLA-DRB1*04 and DRB1*14 alleles are associated with susceptibility to pemphigus among Japanese. J Invest Dermatol 109:615–8 Tron
F, Gilbert D, Mouquet H, Joly P, Drouot L, Makni S et al. (2005) Genetic factors in pemphigus. J Autoimmun 24:319–28
Response to Imatinib Mesylate Depends on the Presence of the V559A-Mutated KIT Oncogene Journal of Investigative Dermatology (2010) 130, 314–316; doi:10.1038/jid.2009.197; published online 8 October 2009
TO THE EDITOR Identification of markers that predict the response to a targeted therapy is a key goal for the treatment of melanoma. A major advance was the observation that certain subtypes, such as mucous membrane melanoma and acral lentiginous melanoma, are frequently associated with activating mutations in the KIT oncogene (Curtin et al., 2006). Imatinib mesylate, a small-molecule inhibitor of protein tyrosine kinases, has been used with considerable success in the treatment of gastrointestinal stromal tumors, which are induced by abnormal activation and overexpression of the KIT receptor in 90% of cases (Sleifjer et al., 2008). Recent reports have described the usefulness of this molecule in the treatment of rectal (Hodi et al., 2008) and acral lentiginous melanoma (Kim et al., 2008). Here, we report a patient with a rapid response to imatinib in a lung metastasis from an acral lentiginous melanoma. Surprisingly, this was paralleled by progressive disease of locoregional lymph node metastasis. A 64-year-old woman was admitted because of progressive severe dyspnea at rest. A pulmonary mass obstructing the left main bronchus was revealed by computed tomographic scanning (Figure 1a). Grayish pigmentation on
her left big toe revealed an acral lentiginous melanoma. Computed tomographic imaging of the abdomen and pelvis excluded visceral metastases but revealed lymph node metastases in the left ilioinguinal region (Figure 1b). Biopsy specimens were excised from the skin and lymph nodes and by incision via endoluminal access from the pulmonary mass. Probes were processed for routine histology and immunohistochemistry and preserved for molecular biology. Our studies were conducted after obtaining written informed consent and institutional review board approval and were in compliance with the Declaration of Helsinki Principles. Histopathology of the skin pigmentation confirmed the diagnosis and revealed parts of an acral lentiginous melanoma (not shown). Histopathology of both the lung (Figure 1c) and lymph node (Figure 1d) biopsies showed highly proliferative atypical epithelioid cells. Expression analysis by immunohistochemistry using a specific monoclonal mouse antibody against KIT (CD117) revealed that both metastases showed the presence of KIT; however, the lung specimen (Figure 1e) expressed the protein more brightly and uniformly than did the lymph node metastasis (Figure 1f). Furthermore, S100, Melan-A,
Abbreviation: CT, computed tomography
314
Journal of Investigative Dermatology (2010), Volume 130
and HMB-45 were positively stained in both probes (Figures 1g and h). We initiated monotherapy with imatinib mesylate (Glivec, Novartis Pharma Nu¨rnberg, Germany) 400 mg once daily. After 6 weeks, radiological reassessment of the chest showed regression of the pulmonary mass; the largest diameter had decreased from 3.2 to 2.0 cm. Moreover, the atelectasis of the left upper lobe had reversed completely (Figure 1i). The performance status was remarkably improved. However, the lymph node metastasis in the left groin showed marked progression; the largest diameter had increased from 3.0 to 5.8 cm in the inguinal node (Figure 1j) and from 2 to 3 cm in the iliacal node (not shown). In an effort to understand this differential biological behavior of the two tumor manifestations, we investigated the KIT mutation status of the primary tumor, the lymph node, and the lung metastases. To this end, genomic DNA was extracted and exons 11, 13, 17, and 18 of the KIT gene, which harbor the vast majority of mutations identified in human tumors, were analyzed by PCR amplification followed by direct sequencing. Interestingly, in both the primary tumor and the lymph node metastasis, we detected only wild-type sequences, whereas in the lung metastasis, a homo- or hemizygous nucleotide