BJD
British Journal of Dermatology
C L I N I C A L A N D LA B O R A T O R Y I N V E S T I G A T I O N S
Skin ulcers related to chronic graft-versus-host disease: clinical findings and associated morbidity M. Jachiet,1 A. de Masson,1 R. Peffault de Latour,2 M. Rybojad,1 M. Robin,2 J.-H. Bourhis,3 A. Xhaard,2 N. Dhedin,2 F. Sicre de Fontbrune,2 F. Suarez,4 S. Barete,7 N. Parquet,2 S. Nguyen,8 L. Ades,5 M.-T. Rubio,6 2 and J.-D. Bouaziz1 S. Wittnebel,3 M. Bagot,1 G. Socie 1
Department of Dermatology, AP-HP, Universite Paris VII Sorbonne Paris Cite; and H^opital Saint Louis, Paris, France Department of Haematology, AP-HP, Universite Paris VII Sorbonne Paris Cite; and H^opital Saint Louis, Paris, France 3 Department of Haematology, AP-HP, Institut Gustave Roussy, Villejuif, France 4 Department of Haematology, AP-HP, H^opital Necker, Paris, France 5 Departments of Haematology, AP-HP, H^opital Avicenne, Bobigny, France 6 Department of Haematology, AP-HP, H^opital Saint Antoine, Paris, France Department of 7Dermatology and 8Haematology, AP-HP, H^opital Pitie Salp^etriere, Paris, France 2
Summary Correspondence Jean-David Bouaziz. E-mail: [email protected]
Accepted for publication 6 January 2014
Funding sources No external funding.
Conflicts of interest None declared G.S. and J.-D.B. contributed equally to this study and share last author seniorship. DOI 10.1111/bjd.12828
Background According to the National Institutes of Health classification of chronic graft-versus-host disease (cGVHD), skin ulcers after allogeneic haematopoietic stem-cell transplantation (HSCT) are recorded as having the maximal severity score but published data are scarce. Objectives To describe skin ulcers related to cGVHD with an emphasis on clinical findings, associated morbidity, management and evolution. Patients and methods A multicentre retrospective analysis was carried out of patients with a diagnosis of cGVHD skin ulcers. Results All 25 patients included in the study had sclerotic skin cGVHD and 21 had lichenoid skin lesions associated with the sclerotic skin lesions. Thirteen patients had severe cGVHD without considering the skin, because of the involvement of an extracutaneous organ by cGVHD. The median time from HSCT to the onset of ulcers was 44 months. In addition to scleroderma, initial skin lesions at the site of ulcers were bullous erosive lichen in 21 patients and bullous erosive morphoea in four patients. Fifteen patients had an inaugural oedema. Ulcers were mostly bilateral with a predilection for the lower limbs. They were frequently colonized but few infections occurred. Four patients died during a median follow-up period of 55 months. Conclusions Chronic graft-versus-host disease skin ulcers occur in patients with sclerodermatous skin cGVHD, are associated with severe cGVHD, often start with bullous lichenoid lesions or bullous morphoea and seem to cause more morbidity than mortality, given the low rate of mortality observed in our series of patients.
What’s already known about this topic?
•
Skin ulcers are classified as a specific manifestation of skin chronic graft-versus-host disease (cGVHD) in the National Institutes of Health classification of GVHD but published data on this is scarce.
What does this study add?
• • •
© 2014 British Association of Dermatologists
cGVHD skin ulcers often start with oedema, bullous lichenoid lesions or bullous morphoea and are always associated with sclerodermatous skin cGVHD. cGVHD skin ulcers were not associated with a high mortality. Early recognition of cGVHD skin ulcers should improve their management.
British Journal of Dermatology (2014) 171, pp63–68
63
64 Skin ulcers related to cGVHD, M. Jachiet et al.
Graft-versus-host disease (GVHD) is a major complication associated with high morbidity and mortality after allogeneic haematopoietic stem-cell transplantation (HSCT). Graft-versushost disease is classified into acute GVHD (aGVHD), characterized by a selective epithelial inflammation mainly affecting the skin, gastrointestinal tract and liver, and chronic GVHD (cGVHD), characterized by inflammation and fibrosis frequently affecting the skin, liver, mouth and eyes. Reported incidence rates of cGVHD after HSCT range from 6% to 80% according to recipient age, donor type, HSCT source, graft manipulation and use of post-transplantation donor lymphocyte infusions.1 Most typical cutaneous manifestations of cGVHD include poikiloderma (atrophic and pigmentary changes), lichen planus-like eruption (erythematous/violaceous flat-topped papules or plaques or a silvery or shiny appearance on direct light), deep sclerotic features, morphoealike superficial sclerotic features, or lichen sclerosus-like lesions (discrete to coalescent grey to white moveable papules or plaques with a shiny appearance and leathery consistency). Severe sclerotic features characterized by thickened, tight and fragile skin are often associated with poor wound healing, inadequate lymphatic drainage and skin ulcers.1 According to the National Institutes of Health (NIH) scoring system for cGVHD, each organ affected is scored according to a 4-point scale (0–3), with three reflecting severe impairment. For the skin, ulcerations are scored as 3, which is the maximal severity score.1 Although commonly seen in clinical practice, published data regarding skin ulcers after cGVHD are missing. The frequency of cGVHD skin ulcers is unknown but only a few previous case reports have described this post-HSCT complication. We report the first retrospective multicentre study of patients with cGVHD skin ulcers with an emphasis on clinical findings, associated cutaneous and extracutaneous cGVHD features and their severity.
Patients and methods We conducted a retrospective multicentre study in six French hospitals (bone marrow transplant and dermatology departments). All patients with a well-assessed diagnosis of cGVHD skin ulcers (clinical pictures, n = 15 and/or accurate dermatological description of lesions) between 1995 and 2013 were included in the study. Around 4400 HSCTs have been performed in these centres during the inclusion period (estimated frequency of cGVHD skin ulcers after HSCT = 0.56%). Ulcerations due to causes other than cGVHD were ruled out (in all patients). Doppler analysis was performed when necessary. Patient characteristics, HSCT and post-HSCT data, clinical features, natural history and treatment of skin ulcers were collected retrospectively. aGVHD was diagnosed according to previously described criteria.2 Diagnosis and staging of cGVHD, including skin changes, were assessed according to the NIH consensus criteria.1 In this classification, a clinical scoring system (0–3) is used for the evaluation of cGVHD involvement of individual sites. Superficial skin sclerotic cGVHD is classified as score 2 and deep skin sclerotic cGVHD British Journal of Dermatology (2014) 171, pp63–68
(rippling, firm subcutaneous tissue with overlying movable skin or hidebound sclerosis) is classified as score 3. A global scoring system is used for the assessment of cGVHD gravity: a lung score ≥ 2 or any other site = 3 is classified as severe cGVHD.1 When manifestations of skin cGVHD were not ‘diagnostic’ (clinically sufficient to establish the diagnosis of cGVHD), a skin biopsy was performed to confirm skin cGVHD. Patients who had clinical pictures gave informed consent. The study was conducted according to the Declaration of Helsinki.
Results Patients, haematopoietic stem-cell transplant and chronic graft-versus-host disease characteristics Twenty-five patients, 20 males and five females were included in the study. Patient and HSCT characteristics are summarized in Table 1. Some typical examples of cGVHD skin ulcers are depicted in Figures 1 and 2. Data regarding cGVHD characteristics are summarized in Table 2. All 25 patients had skin sclerotic cGVHD including eight of 25 (32%) superficial sclerotic features (score 2) and 17 of 25 (68%) deep sclerotic features (score 3). Twenty-one patients (84%) had lichen planus-like Table 1 Patients and haematopoietic stem-cell transplant (HSCT): characteristics (n = 25) Characteristics Median age at HSCT (range), years Sex (male) Diagnosis Acute myeloblastic leukaemia Acute lymphoblastic leukaemia Chronic myeloid leukaemia Myeloproliferative syndrome Chronic lymphocytic leukaemia Lymphoma Others (multiple myeloma, Waldenstrom) Stem cells source Bone marrow Peripheral blood stem cells Umbilical cord blood Donor type HLA-identical sibling HLA-matched unrelated donor Conditioning regimen Myeloablative conditioning Reduced intensity conditioning Total body irradiation Antithymocyte globulin GVHD prophylaxis Ciclosporin MTX MMF Other (tacrolimus)
47 (14–63) 20 (80%) 2 5 6 1 3 5 3
(8%) (20%) (24%) (4%) (12%) (20%) (12%)
4 (16%) 21 (84%) 0 16 (64%) 9 (36%) 11 14 12 6
(44%) (56%) (48%) (24%)
23 14 10 2
(92%) (56%) (40%) (8%)
GVHD, graft-versus-host disease, HLA, human leucocyte antigen; MMF, mycophenolate mofetil; MTX, methotrexate.
© 2014 British Association of Dermatologists
Skin ulcers related to cGVHD, M. Jachiet et al. 65
Fig 1. Chronic graft-versus-host disease skin ulcers. Typical pictures in three patients. Table 2 Chronic graft-versus-host disease (cGVHD): characteristics (n = 25)
Table 3 Skin chronic graft-versus-host disease ulcers: characteristics Characteristics
Characteristic Prior extracutaneous aGVHD 11 (44%) Prior skin aGVHD 17 (68%) Sites involved with cGVHD and maximal severity score 1/2/3 Mouth 21 (84%) 9/7/5 Eye 14 (56%) 8/4/2 Gastrointestinal tract 8 (32%) 4/2/2 Liver 12 (48%) 0/8/4 Lung 13 (52%) 8/2/3 Joints and fascia 15 (60%) 6/5/4 Genital tract 6 (24%) 4/1/1 Visceral severe cGVHD involvement 13 (52%) Eosinophilia > 500 mm 3 8 (32%) Thrombocytopenia < 100 000 mm 3 10 (40%) Sclerotic skin cGVHD 25 (100%) Superficial (score 2) 8 (32%) Deep (score 3) 17 (68%) Lichenoid cGVHD 21 (84%) aGVHD, acute graft-versus-host disease.
lesions associated with the sclerotic skin lesions. Chronic GVHD also involved the mouth (n = 21, score 1/2/3: n = 9/ 7/5), eyes (n = 14, 8/4/2), gastrointestinal tract (n = 8, 4/ 2/2), liver (n = 12, 0/8/4), lung (n = 13, 8/2/3), joints and fascia (n = 15, 6/5/4) and genital tract (n = 6, 4/1/1). Eosinophilia > 500 mm 3 was noted for eight patients (32%) and thrombocytopenia < 100 000 m 3 for 10 patients (40%). All patients had severe cGVHD because of the skin. Thirteen patients (52%) had severe cGVHD without considering the skin, due to the involvement of an extracutaneous organ by cGVHD. Four patients (16%) died during the follow-up period. All four patients died of severe infection not related to ulcers, in the context of severe refractory cGVHD for two patients and progression of haematological disease for one patient. Skin chronic graft-versus-host disease ulcers Table 3 summarizes data on skin cGVHD ulcer characteristics, treatment and evolution. The median time from HSCT to the onset of ulcers was 44 months (range, 4–111 months). In addition to scleroderma-like GVHD, initial skin lesions at the site of ulcers were bullous erosive lichen in 21 patients (84%) © 2014 British Association of Dermatologists
Skin lesions preceding the ulcers Erosive lichen with sclerosis Bullous morphoea with sclerosis Ulcer localization Lower limbs Upper limbs Trunk Skin infections during study period None Superficial infection Cellulitis Wound healing No Partial Complete Recurrence No Yes Treatment Local care Immunosuppressive treatment change Skin graft Death With ulcers Without ulcers None Median duration of ulcers (range); months Median follow-up after ulcers (range); months Median time from HSCT to the onset of ulcers (range); months
21 (84%) 4 (16%) 21 (84%) 5 (20%) 8 (32%) 13 (52%) 3 (12%) 9 (36%) 4 (16%) 14 (56%) 7 (28%) 15 (60%) 10 (40%) 25 (100%) 16 (64%) 1 (4%) 3 1 21 22 55 44
(12%) (4%) (84%) (1 ≥ 156) (5–264) (4–111)
HSCT, haematopoietic stem cell transplantation.
and bullous erosive morphoea in four patients (16%). Fifteen patients (60%) had an inaugural oedema. Ulcers were localized on lower limbs (anterior tibia or ankle and perimalleolar region) for the majority of the patients (n = 21, 84%), upper limbs for five patients (20%) and trunk for eight patients (32%). Most of the ulcers were bilateral (n = 22, 88%). More than 50% of the ulcers were colonized with bacteria (mostly meticillin-sensitive Staphylococcus aureus and Pseudomonas aeruginosa) but not all colonized wounds were infected. Cutaneous infections occurred in 12 patients (superficial skin infection n = 3, cellulitis n = 9). A skin biopsy of ulcers was performed in four cases. Histological examination revealed nonspecific British Journal of Dermatology (2014) 171, pp63–68
66 Skin ulcers related to cGVHD, M. Jachiet et al.
changes, sometimes minimal inflammatory infiltrate in the dermis without vasculitis, slight fibrotic features and dermis oedema. The treatment of cGVHD skin ulcers was always discussed with a haematologist and a dermatologist. All patients had local wound care, debridement and topical treatment, consisting of antiseptics, alginate or hydrocolloid dressing, topical corticosteroids, topical tacrolimus, silver sulfadiazine, silver nitrate or petrolatum. Compression therapy with elastic bandages was used when tolerated by the patient. A change in immunosuppressive therapy was performed in 16 patients (64%). Patients received mostly corticosteroids, extracorporeal photopheresis, sirolimus, mycophenolate mofetil, everolimus or imatinib mesylate. Surgical debridement was performed for one patient. An unsuccessful skin graft was performed for one patient. Antibiotics were given in the case of cellulitis (n = 9). The median follow-up after the onset of skin ulceration was 55 months (range, 5–264 months). The median duration of ulcers was 22 months (range, 1 ≥ 156 months). Complete healing was achieved for seven patients (28%), and partial healing for 14 patients (56%). Ulcers were refractory to healing in four patients. Recurrence was noted in 10 patients (40%). Natural history of skin ulcers in one representative patient A 59-year-old man with a diagnosis of chronic lymphocytic leukaemia received an allogeneic HSCT from his human leucocyte antigen (HLA)-identical sister. Twenty-one days after HSCT, the patient had a grade 3 skin aGVHD, treated with steroids. Nine months after HSCT, he progressively developed extensive skin cGVHD, with oedema and morphoea, stabilized with prednisone and ciclosporin. Five months later, cGVHD evolved into sclerosis unsuccessfully treated with imatinib mesylate. Bullous morphoea (Fig. 2a) evolved into deep welldemarcated erosions, then ulcers on the legs (Fig. 2b) and abdominal skin, with extensive sclerodermatous fasciitis. The patient developed a Pseudomonas aeruginosa infection of the skin ulcers of the abdomen that required hospitalization and intra-
(a)
(b)
venous antibiotherapy. Local wound treatment alternating silver sulfadiazine, alginate and paraffin-based dressing, topical betamethasone, tacrolimus for inflammatory skin lesions and compression therapy, resulted in complete healing of ulcers within 1 year (Fig. 2c). Extracorporeal photopheresis was not efficient but sirolimus treatment induced partial remission of skin sclerosis.
Discussion In this retrospective study we show that cGVHD skin ulcers always occur in patients with sclerodermatous skin cGVHD, are associated with severe cGVHD, and often start with bullous lichenoid lesions or bullous morphoea. Unexpectedly, cGVHD skin ulcers were not associated with a high mortality in our series of patients. Skin is the organ most frequently affected by cGVHD and haematologists and dermatologists will encounter more situations in which early recognition of the cutaneous manifestations of cGVHD can improve patient care. Skin ulcers could represent one of the most severe complications of cutaneous cGVHD. These ulcers are often painful, slow to heal, unresponsive to traditional treatments and can be a nidus for infection; consequently the patient’s quality of life and prognosis may be severely affected. Of note, chronic nonhealing ulcers in the setting of sclerosis may develop into squamous cell carcinoma as we observed in one case in our series of patients. In the majority of cases, ulcers occurred with a background of deep and pronounced sclerosis and some developed in the setting of oedema, followed by sclerosis. Sclerosis usually appears as a late complication of cGVHD.3 Inamoto et al.4 reported that sclerosis was associated with longer time to withdrawal of immunosuppressive treatment but not with risks of overall mortality or nonrelapse mortality. In our study, 12 patients had total body irradiation in their conditioning regimen, known to be strongly associated with development of sclerosis.4,5 The median time of ulcer appearance after HSCT was long (44 months). Although most of the patients had a severe cGVHD, only four patients died during patient follow-up [median time of follow-up after ulcer
(c)
Fig 2. Evolution of the skin area affected by ulcers on the legs of patient 1. Deep fasciitis and bullous morphoea (a), followed by welldemarcated ulcers (b) and complete healing (c). British Journal of Dermatology (2014) 171, pp63–68
© 2014 British Association of Dermatologists
Skin ulcers related to cGVHD, M. Jachiet et al. 67
onset = 55 months (38–77)], while Devillier et al.6 reported a 5-year overall survival of 60% in patients with refractory cGVHD. Although the aetiology of these ulcers is unclear, inflammation and fibrosis seem to play a major role in their formation. Indeed, most cases in our study occurred on sclerotic and inflammatory lesions. Skin ulcers appeared because of bullous lichenoid lesions or bullous morphoea and not because of traumatic causes. Bullous lichenoid lesions and bullous morphoea have already been described in the setting of cGVHD7–9 and cause a detachment between the epidermis and the dermis leading to skin erosion. In bullous morphoea, biopsy findings generally reveal thickened sclerotic deeply eosinophilic collagen bundles that extend to subcutaneous tissue or at the level of deep reticular dermis, subepidermal blisters, variable subepidermal oedema, lymphatic dilatation, and moderate inflammatory infiltration around the blood vessels. The bullae development mechanism in the sclerotic area may involve the deposition of eosinophilic granule major basic protein, lymphatic obstruction and increased hydrostatic pressure.10,11 Histology of bullous lichen planus reveals an intense lichenoid reaction with subepidermal detachment, probably secondary to the basal vacuolization with apoptotic keratinocytes.12 Most of the patients in our series also had unexplained inflammatory oedema of the legs preceding the development of ulcers, with no evidence of cardiac, renal and hepatic causes. Skin oedema may be a specific manifestation of skin cGVHD and be the result of inappropriate endothelial activation that could be another mechanism that explains skin erosions.13 Once erosion is formed, poor healing due to sclerotic tissues and lymphatic failure will ultimately lead to ulcer formation. In some cases, formation of ulcers has been related to drugs. Ghoreschi et al.14 reported 11 cases of ulcers after psoralen plus ultraviolet A bath photochemotherapy, associated with isotretinoin for three patients, for the treatment of sclerotic-type skin cGVHD. Oral etretinate is known to induce skin ulcers, as evidenced by Marcellus et al.15 reporting 10 cases of ulcers in 32 patients treated with etretinate for refractory sclerotic-type cGVHD. Furthermore, they suggested a link between ulcer formation and regression of sclerosis. Thalidomide has also been related to skin ulcers16 as well as the emergence or worsening of sclerosis.17 While it is difficult to absolutely exclude the role of drugs, the chronology of the ulcers in our cases makes this unlikely. To date, no validated treatment of cGVHD skin ulcers has been proposed. In our experience, local therapy plays a major role. Alginate dressing is often required because these wounds are often exudating. Silver sulfadiazine and silver nitrate may also be required to prevent gram-negative bacterial infections. Topical steroids and/or topical tacrolimus may be useful to control local inflammation. Mechanical compression therapy and leg elevation to reverse tissue oedema and improve blood and lymphatic flow may be useful. In our series, only one patient received an unsuccessful skin graft, whereas there are a few reports in the literature regarding treatment with skin transplantation. Some cases report that split thickness skin transplantation © 2014 British Association of Dermatologists
from an HLA-identical donor can be a successful permanent treatment option. Successful allogeneic skin grafting has been reported in more than nine cases.18–26 All but one of the autograft skin transplants darkened.18,20,25,26 In view of these data, the optimal graft source remains unclear, but allogeneic splitthickness skin graft from the HSCT donor seems better than autologous. Crocchiolo et al.23 reported successful transplantation of cultured keratinocytes after glycerolysed skin allograft for multiple ulcerations in the back, groin and flanks. Systemic treatment is a challenge and should ideally allow an improvement of sclerodermatous cGVHD without increasing the risk of infection. Most of our patients were resistant to systemic steroids and calcineurin inhibitors. In our experiment, extracorporeal photopheresis or mTOR inhibitor (everolimus) combined with low-dose systemic steroids may be useful. In conclusion, further longitudinal studies of cohorts of patients with cGVHD should be performed to evaluate the frequency and long-term evolution of cGVHD skin ulcers. General conclusions about the natural evolution and the morbidity of cGVHD skin ulcers in our series of patients should be tempered by the retrospective design of our study. An early and efficient management of skin cGVHD lesions and a better knowledge of the pathophysiology of skin cGVHD ulcers should help decrease their frequency and improve their treatment.
References 1 Filipovich AH, Weisdorf D, Pavletic S et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant 2005; 11:945–56. 2 Leisenring WM, Martin PJ, Petersdorf EW et al. An acute graft-versus-host disease activity index to predict survival after hematopoietic cell transplantation with myeloablative conditioning regimens. Blood 2006; 108:749–55. 3 Saurat JH. Cutaneous manifestations of graft-versus-host disease. Int J Dermatol 1981; 20:249–56. 4 Inamoto Y, Storer BE, Petersdorf EW et al. Incidence, risk factors, and outcomes of sclerosis in patients with chronic graft-versushost disease. Blood 2013; 121:5098–103. 5 Martires KJ, Baird K, Steinberg SM et al. Sclerotic-type chronic GVHD of the skin: clinical risk factors, laboratory markers, and burden of disease. Blood 2011; 118:4250–7. 6 Devillier R, Castagna L, Gonzague L et al. TLI in refractory chronic GVHD. Bone Marrow Transplant 2013; 48:854–8. 7 Hymes SR, Farmer ER, Burns WH et al. Bullous sclerodermalike changes in chronic graft-vs-host disease. Arch Dermatol 1985; 121:1189–92. 8 Moreno JC, Valverde F, Martinez F et al. Bullous scleroderma-like changes in chronic graft-versus-host disease. J Eur Acad Dermatol Venereol 2003; 17:200–3. 9 Del Pozo J, Garcıa-Silva J, Yebra-Pimentel MT. [Chronic graft-versus-host disease presenting as bullous lesions]. Actas Dermosifiliogr 2008; 99:803–7 (in Spanish). 10 Daoud MS, Su WP, Leiferman KM, Perniciaro C. Bullous morphea: clinical, pathologic, and immunopathologic evaluation of thirteen cases. J Am Acad Dermatol 1994; 30:937–43. 11 Yasar S, Mumcuoglu CT, Serdar ZA, Gunes P. A case of lichen sclerosus et atrophicus accompanying bullous morphea. Ann Dermatol 2011; 23(Suppl. 3):S354–9. British Journal of Dermatology (2014) 171, pp63–68
68 Skin ulcers related to cGVHD, M. Jachiet et al. 12 Wagner G, Rose C, Sachse MM. Clinical variants of lichen planus. J Dtsch Dermatol Ges 2013; 11:309–19. 13 Tardieu M, Rybojad M, Peffault de Latour R et al. Localized edema with sclerodermatous evolution: a possible form of skin chronic graft-versus-host disease associated with endothelial activation. Blood 2013; 122:463–5. 14 Ghoreschi K, Thomas P, Penovici M et al. PUVA-bath photochemotherapy and isotretinoin in sclerodermatous graft-versus-host disease. Eur J Dermatol 2008; 18:667–70. 15 Marcellus DC, Altomonte VL, Farmer ER et al. Etretinate therapy for refractory sclerodermatous chronic graft-versus-host disease. Blood 1999; 93:66–70. 16 Schlossberg H, Klumpp T, Sabol P et al. Severe cutaneous ulceration following treatment with thalidomide for GVHD. Bone Marrow Transplant 2001; 27:229–30. 17 White JML, Creamer D, Du Vivier AWP et al. Sclerodermatous graft-versus-host disease: clinical spectrum and therapeutic challenges. Br J Dermatol 2007; 156:1032–8. 18 Knobler HY, Sagher U, Peled IJ et al. Tolerance to donor-type skin in the recipient of a bone marrow allograft. Treatment of skin ulcers in chronic graft-versus-host disease with skin grafts from the bone marrow donor. Transplantation 1985; 40:223–5. 19 Ammer J, Prantl L, Holler B et al. Successful treatment of a refractory skin ulcer in chronic cutaneous GvHD after allogeneic HSCT
British Journal of Dermatology (2014) 171, pp63–68
20
21
22
23
24
25
26
with split-thickness skin allografting from the stem cell donor. Bone Marrow Transplant 2012; 47:1368–9. Kondo S, Tamura K, Makino S, Takamatsu Y. Allogeneic skin grafting for extensive skin chronic graft-versus-host disease. Int J Hematol 1994; 60:151–5. Kamei Y, Torii S, Hasegawa M et al. Simultaneous allografting and autografting of skin for ulcers in a bone marrow transplant patient. Transplantation 1993; 56:1530–1. Baumeister RG, Riel KA, Kolb HJ. [Allogeneic skin transplantation in chronic graft-versus-host disease following bone marrow transplantation]. Chirurg 1990; 61:438–40 (in German). Crocchiolo R, Nicolini F-E, Sobh M et al. Treatment of a severe extensive cutaneous chronic GVHD after allo-HSCT using glycerolyzed skin allografts and cultured epidermis from the same donor. Bone Marrow Transplant 2011; 46:1153–5. Berg JO, Vindeløv L, Schmidt G, Drzewiecki KT. Allogeneic splitskin grafting in stem cell transplanted patients. J Plast Reconstr Aesthet Surg 2008; 61:1512–15. Storb R, Thomas ED, Buckner CD et al. Allogeneic marrow grafting for treatment of aplastic anemia: a follow-up on long-term survivors. Blood 1976; 48:485–90. Caffee HH, Miller TA. Skin rejection in graft-versus-host disease. Plast Reconstr Surg 1978; 61:753–5.
© 2014 British Association of Dermatologists
British Journal of Dermatology
C L I N I C A L A N D LA B O R A T O R Y I N V E S T I G A T I O N S
Skin ulcers related to chronic graft-versus-host disease: clinical findings and associated morbidity M. Jachiet,1 A. de Masson,1 R. Peffault de Latour,2 M. Rybojad,1 M. Robin,2 J.-H. Bourhis,3 A. Xhaard,2 N. Dhedin,2 F. Sicre de Fontbrune,2 F. Suarez,4 S. Barete,7 N. Parquet,2 S. Nguyen,8 L. Ades,5 M.-T. Rubio,6 2 and J.-D. Bouaziz1 S. Wittnebel,3 M. Bagot,1 G. Socie 1
Department of Dermatology, AP-HP, Universite Paris VII Sorbonne Paris Cite; and H^opital Saint Louis, Paris, France Department of Haematology, AP-HP, Universite Paris VII Sorbonne Paris Cite; and H^opital Saint Louis, Paris, France 3 Department of Haematology, AP-HP, Institut Gustave Roussy, Villejuif, France 4 Department of Haematology, AP-HP, H^opital Necker, Paris, France 5 Departments of Haematology, AP-HP, H^opital Avicenne, Bobigny, France 6 Department of Haematology, AP-HP, H^opital Saint Antoine, Paris, France Department of 7Dermatology and 8Haematology, AP-HP, H^opital Pitie Salp^etriere, Paris, France 2
Summary Correspondence Jean-David Bouaziz. E-mail: [email protected]
Accepted for publication 6 January 2014
Funding sources No external funding.
Conflicts of interest None declared G.S. and J.-D.B. contributed equally to this study and share last author seniorship. DOI 10.1111/bjd.12828
Background According to the National Institutes of Health classification of chronic graft-versus-host disease (cGVHD), skin ulcers after allogeneic haematopoietic stem-cell transplantation (HSCT) are recorded as having the maximal severity score but published data are scarce. Objectives To describe skin ulcers related to cGVHD with an emphasis on clinical findings, associated morbidity, management and evolution. Patients and methods A multicentre retrospective analysis was carried out of patients with a diagnosis of cGVHD skin ulcers. Results All 25 patients included in the study had sclerotic skin cGVHD and 21 had lichenoid skin lesions associated with the sclerotic skin lesions. Thirteen patients had severe cGVHD without considering the skin, because of the involvement of an extracutaneous organ by cGVHD. The median time from HSCT to the onset of ulcers was 44 months. In addition to scleroderma, initial skin lesions at the site of ulcers were bullous erosive lichen in 21 patients and bullous erosive morphoea in four patients. Fifteen patients had an inaugural oedema. Ulcers were mostly bilateral with a predilection for the lower limbs. They were frequently colonized but few infections occurred. Four patients died during a median follow-up period of 55 months. Conclusions Chronic graft-versus-host disease skin ulcers occur in patients with sclerodermatous skin cGVHD, are associated with severe cGVHD, often start with bullous lichenoid lesions or bullous morphoea and seem to cause more morbidity than mortality, given the low rate of mortality observed in our series of patients.
What’s already known about this topic?
•
Skin ulcers are classified as a specific manifestation of skin chronic graft-versus-host disease (cGVHD) in the National Institutes of Health classification of GVHD but published data on this is scarce.
What does this study add?
• • •
© 2014 British Association of Dermatologists
cGVHD skin ulcers often start with oedema, bullous lichenoid lesions or bullous morphoea and are always associated with sclerodermatous skin cGVHD. cGVHD skin ulcers were not associated with a high mortality. Early recognition of cGVHD skin ulcers should improve their management.
British Journal of Dermatology (2014) 171, pp63–68
63
64 Skin ulcers related to cGVHD, M. Jachiet et al.
Graft-versus-host disease (GVHD) is a major complication associated with high morbidity and mortality after allogeneic haematopoietic stem-cell transplantation (HSCT). Graft-versushost disease is classified into acute GVHD (aGVHD), characterized by a selective epithelial inflammation mainly affecting the skin, gastrointestinal tract and liver, and chronic GVHD (cGVHD), characterized by inflammation and fibrosis frequently affecting the skin, liver, mouth and eyes. Reported incidence rates of cGVHD after HSCT range from 6% to 80% according to recipient age, donor type, HSCT source, graft manipulation and use of post-transplantation donor lymphocyte infusions.1 Most typical cutaneous manifestations of cGVHD include poikiloderma (atrophic and pigmentary changes), lichen planus-like eruption (erythematous/violaceous flat-topped papules or plaques or a silvery or shiny appearance on direct light), deep sclerotic features, morphoealike superficial sclerotic features, or lichen sclerosus-like lesions (discrete to coalescent grey to white moveable papules or plaques with a shiny appearance and leathery consistency). Severe sclerotic features characterized by thickened, tight and fragile skin are often associated with poor wound healing, inadequate lymphatic drainage and skin ulcers.1 According to the National Institutes of Health (NIH) scoring system for cGVHD, each organ affected is scored according to a 4-point scale (0–3), with three reflecting severe impairment. For the skin, ulcerations are scored as 3, which is the maximal severity score.1 Although commonly seen in clinical practice, published data regarding skin ulcers after cGVHD are missing. The frequency of cGVHD skin ulcers is unknown but only a few previous case reports have described this post-HSCT complication. We report the first retrospective multicentre study of patients with cGVHD skin ulcers with an emphasis on clinical findings, associated cutaneous and extracutaneous cGVHD features and their severity.
Patients and methods We conducted a retrospective multicentre study in six French hospitals (bone marrow transplant and dermatology departments). All patients with a well-assessed diagnosis of cGVHD skin ulcers (clinical pictures, n = 15 and/or accurate dermatological description of lesions) between 1995 and 2013 were included in the study. Around 4400 HSCTs have been performed in these centres during the inclusion period (estimated frequency of cGVHD skin ulcers after HSCT = 0.56%). Ulcerations due to causes other than cGVHD were ruled out (in all patients). Doppler analysis was performed when necessary. Patient characteristics, HSCT and post-HSCT data, clinical features, natural history and treatment of skin ulcers were collected retrospectively. aGVHD was diagnosed according to previously described criteria.2 Diagnosis and staging of cGVHD, including skin changes, were assessed according to the NIH consensus criteria.1 In this classification, a clinical scoring system (0–3) is used for the evaluation of cGVHD involvement of individual sites. Superficial skin sclerotic cGVHD is classified as score 2 and deep skin sclerotic cGVHD British Journal of Dermatology (2014) 171, pp63–68
(rippling, firm subcutaneous tissue with overlying movable skin or hidebound sclerosis) is classified as score 3. A global scoring system is used for the assessment of cGVHD gravity: a lung score ≥ 2 or any other site = 3 is classified as severe cGVHD.1 When manifestations of skin cGVHD were not ‘diagnostic’ (clinically sufficient to establish the diagnosis of cGVHD), a skin biopsy was performed to confirm skin cGVHD. Patients who had clinical pictures gave informed consent. The study was conducted according to the Declaration of Helsinki.
Results Patients, haematopoietic stem-cell transplant and chronic graft-versus-host disease characteristics Twenty-five patients, 20 males and five females were included in the study. Patient and HSCT characteristics are summarized in Table 1. Some typical examples of cGVHD skin ulcers are depicted in Figures 1 and 2. Data regarding cGVHD characteristics are summarized in Table 2. All 25 patients had skin sclerotic cGVHD including eight of 25 (32%) superficial sclerotic features (score 2) and 17 of 25 (68%) deep sclerotic features (score 3). Twenty-one patients (84%) had lichen planus-like Table 1 Patients and haematopoietic stem-cell transplant (HSCT): characteristics (n = 25) Characteristics Median age at HSCT (range), years Sex (male) Diagnosis Acute myeloblastic leukaemia Acute lymphoblastic leukaemia Chronic myeloid leukaemia Myeloproliferative syndrome Chronic lymphocytic leukaemia Lymphoma Others (multiple myeloma, Waldenstrom) Stem cells source Bone marrow Peripheral blood stem cells Umbilical cord blood Donor type HLA-identical sibling HLA-matched unrelated donor Conditioning regimen Myeloablative conditioning Reduced intensity conditioning Total body irradiation Antithymocyte globulin GVHD prophylaxis Ciclosporin MTX MMF Other (tacrolimus)
47 (14–63) 20 (80%) 2 5 6 1 3 5 3
(8%) (20%) (24%) (4%) (12%) (20%) (12%)
4 (16%) 21 (84%) 0 16 (64%) 9 (36%) 11 14 12 6
(44%) (56%) (48%) (24%)
23 14 10 2
(92%) (56%) (40%) (8%)
GVHD, graft-versus-host disease, HLA, human leucocyte antigen; MMF, mycophenolate mofetil; MTX, methotrexate.
© 2014 British Association of Dermatologists
Skin ulcers related to cGVHD, M. Jachiet et al. 65
Fig 1. Chronic graft-versus-host disease skin ulcers. Typical pictures in three patients. Table 2 Chronic graft-versus-host disease (cGVHD): characteristics (n = 25)
Table 3 Skin chronic graft-versus-host disease ulcers: characteristics Characteristics
Characteristic Prior extracutaneous aGVHD 11 (44%) Prior skin aGVHD 17 (68%) Sites involved with cGVHD and maximal severity score 1/2/3 Mouth 21 (84%) 9/7/5 Eye 14 (56%) 8/4/2 Gastrointestinal tract 8 (32%) 4/2/2 Liver 12 (48%) 0/8/4 Lung 13 (52%) 8/2/3 Joints and fascia 15 (60%) 6/5/4 Genital tract 6 (24%) 4/1/1 Visceral severe cGVHD involvement 13 (52%) Eosinophilia > 500 mm 3 8 (32%) Thrombocytopenia < 100 000 mm 3 10 (40%) Sclerotic skin cGVHD 25 (100%) Superficial (score 2) 8 (32%) Deep (score 3) 17 (68%) Lichenoid cGVHD 21 (84%) aGVHD, acute graft-versus-host disease.
lesions associated with the sclerotic skin lesions. Chronic GVHD also involved the mouth (n = 21, score 1/2/3: n = 9/ 7/5), eyes (n = 14, 8/4/2), gastrointestinal tract (n = 8, 4/ 2/2), liver (n = 12, 0/8/4), lung (n = 13, 8/2/3), joints and fascia (n = 15, 6/5/4) and genital tract (n = 6, 4/1/1). Eosinophilia > 500 mm 3 was noted for eight patients (32%) and thrombocytopenia < 100 000 m 3 for 10 patients (40%). All patients had severe cGVHD because of the skin. Thirteen patients (52%) had severe cGVHD without considering the skin, due to the involvement of an extracutaneous organ by cGVHD. Four patients (16%) died during the follow-up period. All four patients died of severe infection not related to ulcers, in the context of severe refractory cGVHD for two patients and progression of haematological disease for one patient. Skin chronic graft-versus-host disease ulcers Table 3 summarizes data on skin cGVHD ulcer characteristics, treatment and evolution. The median time from HSCT to the onset of ulcers was 44 months (range, 4–111 months). In addition to scleroderma-like GVHD, initial skin lesions at the site of ulcers were bullous erosive lichen in 21 patients (84%) © 2014 British Association of Dermatologists
Skin lesions preceding the ulcers Erosive lichen with sclerosis Bullous morphoea with sclerosis Ulcer localization Lower limbs Upper limbs Trunk Skin infections during study period None Superficial infection Cellulitis Wound healing No Partial Complete Recurrence No Yes Treatment Local care Immunosuppressive treatment change Skin graft Death With ulcers Without ulcers None Median duration of ulcers (range); months Median follow-up after ulcers (range); months Median time from HSCT to the onset of ulcers (range); months
21 (84%) 4 (16%) 21 (84%) 5 (20%) 8 (32%) 13 (52%) 3 (12%) 9 (36%) 4 (16%) 14 (56%) 7 (28%) 15 (60%) 10 (40%) 25 (100%) 16 (64%) 1 (4%) 3 1 21 22 55 44
(12%) (4%) (84%) (1 ≥ 156) (5–264) (4–111)
HSCT, haematopoietic stem cell transplantation.
and bullous erosive morphoea in four patients (16%). Fifteen patients (60%) had an inaugural oedema. Ulcers were localized on lower limbs (anterior tibia or ankle and perimalleolar region) for the majority of the patients (n = 21, 84%), upper limbs for five patients (20%) and trunk for eight patients (32%). Most of the ulcers were bilateral (n = 22, 88%). More than 50% of the ulcers were colonized with bacteria (mostly meticillin-sensitive Staphylococcus aureus and Pseudomonas aeruginosa) but not all colonized wounds were infected. Cutaneous infections occurred in 12 patients (superficial skin infection n = 3, cellulitis n = 9). A skin biopsy of ulcers was performed in four cases. Histological examination revealed nonspecific British Journal of Dermatology (2014) 171, pp63–68
66 Skin ulcers related to cGVHD, M. Jachiet et al.
changes, sometimes minimal inflammatory infiltrate in the dermis without vasculitis, slight fibrotic features and dermis oedema. The treatment of cGVHD skin ulcers was always discussed with a haematologist and a dermatologist. All patients had local wound care, debridement and topical treatment, consisting of antiseptics, alginate or hydrocolloid dressing, topical corticosteroids, topical tacrolimus, silver sulfadiazine, silver nitrate or petrolatum. Compression therapy with elastic bandages was used when tolerated by the patient. A change in immunosuppressive therapy was performed in 16 patients (64%). Patients received mostly corticosteroids, extracorporeal photopheresis, sirolimus, mycophenolate mofetil, everolimus or imatinib mesylate. Surgical debridement was performed for one patient. An unsuccessful skin graft was performed for one patient. Antibiotics were given in the case of cellulitis (n = 9). The median follow-up after the onset of skin ulceration was 55 months (range, 5–264 months). The median duration of ulcers was 22 months (range, 1 ≥ 156 months). Complete healing was achieved for seven patients (28%), and partial healing for 14 patients (56%). Ulcers were refractory to healing in four patients. Recurrence was noted in 10 patients (40%). Natural history of skin ulcers in one representative patient A 59-year-old man with a diagnosis of chronic lymphocytic leukaemia received an allogeneic HSCT from his human leucocyte antigen (HLA)-identical sister. Twenty-one days after HSCT, the patient had a grade 3 skin aGVHD, treated with steroids. Nine months after HSCT, he progressively developed extensive skin cGVHD, with oedema and morphoea, stabilized with prednisone and ciclosporin. Five months later, cGVHD evolved into sclerosis unsuccessfully treated with imatinib mesylate. Bullous morphoea (Fig. 2a) evolved into deep welldemarcated erosions, then ulcers on the legs (Fig. 2b) and abdominal skin, with extensive sclerodermatous fasciitis. The patient developed a Pseudomonas aeruginosa infection of the skin ulcers of the abdomen that required hospitalization and intra-
(a)
(b)
venous antibiotherapy. Local wound treatment alternating silver sulfadiazine, alginate and paraffin-based dressing, topical betamethasone, tacrolimus for inflammatory skin lesions and compression therapy, resulted in complete healing of ulcers within 1 year (Fig. 2c). Extracorporeal photopheresis was not efficient but sirolimus treatment induced partial remission of skin sclerosis.
Discussion In this retrospective study we show that cGVHD skin ulcers always occur in patients with sclerodermatous skin cGVHD, are associated with severe cGVHD, and often start with bullous lichenoid lesions or bullous morphoea. Unexpectedly, cGVHD skin ulcers were not associated with a high mortality in our series of patients. Skin is the organ most frequently affected by cGVHD and haematologists and dermatologists will encounter more situations in which early recognition of the cutaneous manifestations of cGVHD can improve patient care. Skin ulcers could represent one of the most severe complications of cutaneous cGVHD. These ulcers are often painful, slow to heal, unresponsive to traditional treatments and can be a nidus for infection; consequently the patient’s quality of life and prognosis may be severely affected. Of note, chronic nonhealing ulcers in the setting of sclerosis may develop into squamous cell carcinoma as we observed in one case in our series of patients. In the majority of cases, ulcers occurred with a background of deep and pronounced sclerosis and some developed in the setting of oedema, followed by sclerosis. Sclerosis usually appears as a late complication of cGVHD.3 Inamoto et al.4 reported that sclerosis was associated with longer time to withdrawal of immunosuppressive treatment but not with risks of overall mortality or nonrelapse mortality. In our study, 12 patients had total body irradiation in their conditioning regimen, known to be strongly associated with development of sclerosis.4,5 The median time of ulcer appearance after HSCT was long (44 months). Although most of the patients had a severe cGVHD, only four patients died during patient follow-up [median time of follow-up after ulcer
(c)
Fig 2. Evolution of the skin area affected by ulcers on the legs of patient 1. Deep fasciitis and bullous morphoea (a), followed by welldemarcated ulcers (b) and complete healing (c). British Journal of Dermatology (2014) 171, pp63–68
© 2014 British Association of Dermatologists
Skin ulcers related to cGVHD, M. Jachiet et al. 67
onset = 55 months (38–77)], while Devillier et al.6 reported a 5-year overall survival of 60% in patients with refractory cGVHD. Although the aetiology of these ulcers is unclear, inflammation and fibrosis seem to play a major role in their formation. Indeed, most cases in our study occurred on sclerotic and inflammatory lesions. Skin ulcers appeared because of bullous lichenoid lesions or bullous morphoea and not because of traumatic causes. Bullous lichenoid lesions and bullous morphoea have already been described in the setting of cGVHD7–9 and cause a detachment between the epidermis and the dermis leading to skin erosion. In bullous morphoea, biopsy findings generally reveal thickened sclerotic deeply eosinophilic collagen bundles that extend to subcutaneous tissue or at the level of deep reticular dermis, subepidermal blisters, variable subepidermal oedema, lymphatic dilatation, and moderate inflammatory infiltration around the blood vessels. The bullae development mechanism in the sclerotic area may involve the deposition of eosinophilic granule major basic protein, lymphatic obstruction and increased hydrostatic pressure.10,11 Histology of bullous lichen planus reveals an intense lichenoid reaction with subepidermal detachment, probably secondary to the basal vacuolization with apoptotic keratinocytes.12 Most of the patients in our series also had unexplained inflammatory oedema of the legs preceding the development of ulcers, with no evidence of cardiac, renal and hepatic causes. Skin oedema may be a specific manifestation of skin cGVHD and be the result of inappropriate endothelial activation that could be another mechanism that explains skin erosions.13 Once erosion is formed, poor healing due to sclerotic tissues and lymphatic failure will ultimately lead to ulcer formation. In some cases, formation of ulcers has been related to drugs. Ghoreschi et al.14 reported 11 cases of ulcers after psoralen plus ultraviolet A bath photochemotherapy, associated with isotretinoin for three patients, for the treatment of sclerotic-type skin cGVHD. Oral etretinate is known to induce skin ulcers, as evidenced by Marcellus et al.15 reporting 10 cases of ulcers in 32 patients treated with etretinate for refractory sclerotic-type cGVHD. Furthermore, they suggested a link between ulcer formation and regression of sclerosis. Thalidomide has also been related to skin ulcers16 as well as the emergence or worsening of sclerosis.17 While it is difficult to absolutely exclude the role of drugs, the chronology of the ulcers in our cases makes this unlikely. To date, no validated treatment of cGVHD skin ulcers has been proposed. In our experience, local therapy plays a major role. Alginate dressing is often required because these wounds are often exudating. Silver sulfadiazine and silver nitrate may also be required to prevent gram-negative bacterial infections. Topical steroids and/or topical tacrolimus may be useful to control local inflammation. Mechanical compression therapy and leg elevation to reverse tissue oedema and improve blood and lymphatic flow may be useful. In our series, only one patient received an unsuccessful skin graft, whereas there are a few reports in the literature regarding treatment with skin transplantation. Some cases report that split thickness skin transplantation © 2014 British Association of Dermatologists
from an HLA-identical donor can be a successful permanent treatment option. Successful allogeneic skin grafting has been reported in more than nine cases.18–26 All but one of the autograft skin transplants darkened.18,20,25,26 In view of these data, the optimal graft source remains unclear, but allogeneic splitthickness skin graft from the HSCT donor seems better than autologous. Crocchiolo et al.23 reported successful transplantation of cultured keratinocytes after glycerolysed skin allograft for multiple ulcerations in the back, groin and flanks. Systemic treatment is a challenge and should ideally allow an improvement of sclerodermatous cGVHD without increasing the risk of infection. Most of our patients were resistant to systemic steroids and calcineurin inhibitors. In our experiment, extracorporeal photopheresis or mTOR inhibitor (everolimus) combined with low-dose systemic steroids may be useful. In conclusion, further longitudinal studies of cohorts of patients with cGVHD should be performed to evaluate the frequency and long-term evolution of cGVHD skin ulcers. General conclusions about the natural evolution and the morbidity of cGVHD skin ulcers in our series of patients should be tempered by the retrospective design of our study. An early and efficient management of skin cGVHD lesions and a better knowledge of the pathophysiology of skin cGVHD ulcers should help decrease their frequency and improve their treatment.
References 1 Filipovich AH, Weisdorf D, Pavletic S et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant 2005; 11:945–56. 2 Leisenring WM, Martin PJ, Petersdorf EW et al. An acute graft-versus-host disease activity index to predict survival after hematopoietic cell transplantation with myeloablative conditioning regimens. Blood 2006; 108:749–55. 3 Saurat JH. Cutaneous manifestations of graft-versus-host disease. Int J Dermatol 1981; 20:249–56. 4 Inamoto Y, Storer BE, Petersdorf EW et al. Incidence, risk factors, and outcomes of sclerosis in patients with chronic graft-versushost disease. Blood 2013; 121:5098–103. 5 Martires KJ, Baird K, Steinberg SM et al. Sclerotic-type chronic GVHD of the skin: clinical risk factors, laboratory markers, and burden of disease. Blood 2011; 118:4250–7. 6 Devillier R, Castagna L, Gonzague L et al. TLI in refractory chronic GVHD. Bone Marrow Transplant 2013; 48:854–8. 7 Hymes SR, Farmer ER, Burns WH et al. Bullous sclerodermalike changes in chronic graft-vs-host disease. Arch Dermatol 1985; 121:1189–92. 8 Moreno JC, Valverde F, Martinez F et al. Bullous scleroderma-like changes in chronic graft-versus-host disease. J Eur Acad Dermatol Venereol 2003; 17:200–3. 9 Del Pozo J, Garcıa-Silva J, Yebra-Pimentel MT. [Chronic graft-versus-host disease presenting as bullous lesions]. Actas Dermosifiliogr 2008; 99:803–7 (in Spanish). 10 Daoud MS, Su WP, Leiferman KM, Perniciaro C. Bullous morphea: clinical, pathologic, and immunopathologic evaluation of thirteen cases. J Am Acad Dermatol 1994; 30:937–43. 11 Yasar S, Mumcuoglu CT, Serdar ZA, Gunes P. A case of lichen sclerosus et atrophicus accompanying bullous morphea. Ann Dermatol 2011; 23(Suppl. 3):S354–9. British Journal of Dermatology (2014) 171, pp63–68
68 Skin ulcers related to cGVHD, M. Jachiet et al. 12 Wagner G, Rose C, Sachse MM. Clinical variants of lichen planus. J Dtsch Dermatol Ges 2013; 11:309–19. 13 Tardieu M, Rybojad M, Peffault de Latour R et al. Localized edema with sclerodermatous evolution: a possible form of skin chronic graft-versus-host disease associated with endothelial activation. Blood 2013; 122:463–5. 14 Ghoreschi K, Thomas P, Penovici M et al. PUVA-bath photochemotherapy and isotretinoin in sclerodermatous graft-versus-host disease. Eur J Dermatol 2008; 18:667–70. 15 Marcellus DC, Altomonte VL, Farmer ER et al. Etretinate therapy for refractory sclerodermatous chronic graft-versus-host disease. Blood 1999; 93:66–70. 16 Schlossberg H, Klumpp T, Sabol P et al. Severe cutaneous ulceration following treatment with thalidomide for GVHD. Bone Marrow Transplant 2001; 27:229–30. 17 White JML, Creamer D, Du Vivier AWP et al. Sclerodermatous graft-versus-host disease: clinical spectrum and therapeutic challenges. Br J Dermatol 2007; 156:1032–8. 18 Knobler HY, Sagher U, Peled IJ et al. Tolerance to donor-type skin in the recipient of a bone marrow allograft. Treatment of skin ulcers in chronic graft-versus-host disease with skin grafts from the bone marrow donor. Transplantation 1985; 40:223–5. 19 Ammer J, Prantl L, Holler B et al. Successful treatment of a refractory skin ulcer in chronic cutaneous GvHD after allogeneic HSCT
British Journal of Dermatology (2014) 171, pp63–68
20
21
22
23
24
25
26
with split-thickness skin allografting from the stem cell donor. Bone Marrow Transplant 2012; 47:1368–9. Kondo S, Tamura K, Makino S, Takamatsu Y. Allogeneic skin grafting for extensive skin chronic graft-versus-host disease. Int J Hematol 1994; 60:151–5. Kamei Y, Torii S, Hasegawa M et al. Simultaneous allografting and autografting of skin for ulcers in a bone marrow transplant patient. Transplantation 1993; 56:1530–1. Baumeister RG, Riel KA, Kolb HJ. [Allogeneic skin transplantation in chronic graft-versus-host disease following bone marrow transplantation]. Chirurg 1990; 61:438–40 (in German). Crocchiolo R, Nicolini F-E, Sobh M et al. Treatment of a severe extensive cutaneous chronic GVHD after allo-HSCT using glycerolyzed skin allografts and cultured epidermis from the same donor. Bone Marrow Transplant 2011; 46:1153–5. Berg JO, Vindeløv L, Schmidt G, Drzewiecki KT. Allogeneic splitskin grafting in stem cell transplanted patients. J Plast Reconstr Aesthet Surg 2008; 61:1512–15. Storb R, Thomas ED, Buckner CD et al. Allogeneic marrow grafting for treatment of aplastic anemia: a follow-up on long-term survivors. Blood 1976; 48:485–90. Caffee HH, Miller TA. Skin rejection in graft-versus-host disease. Plast Reconstr Surg 1978; 61:753–5.
© 2014 British Association of Dermatologists
