
	Version PDF

Pathological score for the evaluation of allograft rejection in human hand (composite tissue) allotransplantation

European Journal of Dermatology. Volume 15, Number 4, 235-8, July-August 2005, Investigative report

Summary

Author(s) : Jean Kanitakis, Palmina Petruzzo, Denis Jullien, Lionel Badet, Maria Clara Dezza, Alain Claudy, Marco Lanzetta, Nadey Hakim, Earl Owen, Jean-Michel Dubernard , Department of Dermatology/EA 37-32 UCBL, Ed. Herriot Hospital, Lyon, France, Department of Surgery and Transplantation, Ed. Herriot Hospital, Lyon, France, Hand Surgery & Reconstructive Microsurgery Unit, San Gerardo Hospital, Univ. of Milan Bicocca, Milan, Italy, Transplant Unit, St. Mary’s Hospital, London, UK, Microsearch Foundation, Sydney, Australia.

Summary : The experience obtained from the human hand allografts (HHA) performed to date suggests that the skin is a priviledged target of allograft rejection in this setting. The aim of this study was to establish a pathological score for assessing the severity of HHA rejection. The pathological slides of 89 skin biopsies obtained from the allografted limbs of six HHA recipients from day 0 to 5 years post-graft were retrospectively examined. According to the severity of the pathological changes observed, the following grades of rejection are proposed: 0: no rejection, I: mild rejection, II: moderate rejection, III: severe rejection, IV: very severe rejection. This grading system can be used as a basis for monitoring allograft rejection and for assessing the effects of the immunosuppressive treatment aiming at reversing HHA rejection\; it can also be used for monitoring rejection of other skin-containing CTA.

Keywords : composite tissue allografts, human hand allografts, pathology, rejection

Pictures

ARTICLE

Auteur(s) :, Jean Kanitakis^1,*, Palmina Petruzzo², Denis Jullien¹, Lionel Badet², Maria Clara Dezza³, Alain Claudy¹, Marco Lanzetta³, Nadey Hakim⁴, Earl Owen⁵, Jean-Michel Dubernard²

¹Department of Dermatology/EA 37-32 UCBL, Ed. Herriot Hospital, Lyon, France
²Department of Surgery and Transplantation, Ed. Herriot Hospital, Lyon, France
³Hand Surgery & Reconstructive Microsurgery Unit, San Gerardo Hospital, Univ. of Milan Bicocca, Milan, Italy
⁴Transplant Unit, St. Mary’s Hospital, London, UK
⁵Microsearch Foundation, Sydney, Australia

accepté le 17 Février 2005

Advances in immunosuppressive treatments and surgical techniques have made composite tissue allografts (CTA) possible in humans. Aside from allografts of femur, knee, larynx, tendon, muscle, intestine and abdominal wall [1-3], 24 human hand allografts (HHA) have been performed worldwide since 1998, when the first HHA was performed in Lyon [4-7]. HHA can be considered the “gold standard” of CTA since they contain several tissues showing various degrees of antigenicity. The experience we have obtained so far from three HHA performed in our hospital showed that clinicopathologic monitoring of the skin is the most reliable way to detect early allograft rejection, since this tissue proved to be a privileged target of allograft rejection. This rejection manifests primarily with skin lesions that start as faint, hardly-visible pink macules that may progress to red, infiltrated, more or less scaly lichenoid papules. Eventually these may coalesce into diffuse psoriasiform lesions over the allografted limb. In a previous study we reported the clinicopathological changes seen during allograft rejection in the first HHA [8, 9]. These preliminary findings prompted us to reexamine retrospectively the pathological sections of the skin biopsies taken from six HHA performed in Lyon and Milan, in order to establish a pathological score assessing the severity of HHA rejection. Based on our results, we propose a scoring system of five severity grades that can be used to monitor the development of allograft rejection and its regression upon adequate adaptation of the immunosuppressive treatment. Besides HHA, this score is also applicable to other human CTA containing skin; it could be refined in the future to include additional pathological findings that could develop in the long-term.

Materials and methods

The slides of skin biopsies obtained from six HHA recipients grafted in Lyon (France) and Milan (Italy) from 1998 to 2003 and followed histologically until January 2005 were retrospectively reexamined. All patients were men with a mean age at transplantation 33.8 years (range 22-48). The duration of follow-up ranged from 16 to 60 months. The patients received a maintenance immunosuppressive treatment similar to that given to renal graft patients, consisting in an association of FK506 (blood levels 5-10 ng/mL), prednisolone (10-20 mg/d) and mycophenolate mofetil (1.5-2 g/d). During the induction phase, polyclonal antilymphocyte antidodies and/or monoclonal anti-CD25 antibody were occasionally associated. More details regarding surgical technique and immunosuppressive treatment can be found elsewhere [4, 6-10]. Punch or scalpel skin biopsies were obtained from the allografted forearms (and occasionally the hands) at various time points post-graft (from day 0 to 5 years) during systemic surveillance from clinically normal skin, or during episodes of graft rejection, manifesting clinically with lesions ranging from faint erythematous macules to diffuse erythematous-scaly plaques. The skin specimens were formalin-fixed and paraffin-embedded, and on several occasions also fresh-frozen. 5-μm-thick sections were stained with hematoxylin-eosin or labeled immunohistochemically with antibodies detecting various cell components of normal and inflamed skin [11](table 1)( Table 1 ). A total of 89 skin biopsies were examined, with multiple section levels examined for each biopsy specimen. The following features were particularly looked for in the epidermis: thickness, presence of spongiosis, exocytosis, keratinocyte necrosis/apoptosis, basal cell vacuolization. The following changes were searched for in the dermis and hypodermis: inflammatory cell infiltrate; density, composition and localization of the infiltrate (perivascular, periadnexal, perineural or interstitial); presence of hemorrhage; vascular changes (dilatation, necrosis, thrombosis); adnexal (hair follicle, sweat gland) alterations (necrosis, spongiosis); dermal edema, fibrosis and/or sclerosis.
Table 1 Main antibodies used in the study

Antibody (clone)	Source	Antigen recognized
Rabbit polyclonal	Dako	CD3 (T cells)
MT310	Dako	CD4 (T helper cells)
C8/144B	Dako	CD8 (T suppr/cytotoxic cells)
UCHL1	Dako	CD45 RO (T memory cells)
L26	Dako	CD20 (B cells)
LN3	Novocastra Lab.	HLA DR (class II)
AA1	Dako	Tryptase (mast cells)
HLA A23/24	One lambda	HLA A23/24

Results

The pathological changes observed in the skin biopsies varied greatly from none to very severe. Based on the careful reexamination of the 89 biopsies, the following grades were defined according to the severity of the changes observed:

– Grade 0: no rejection. The skin does not show obvious histological changes, or may occasionally contain a small number of lymphocytes around dermal blood vessels; however the density of this infiltrate is not sufficient to suspect graft rejection ( (figure 1A) ). This grade corresponded to biopsies taken from clinically normal-looking skin ( (figure 1B) ). i.e. when no rejection was clinically suspected, or after regression of clinical lesions of (mild) graft rejection following an increase of the immunosuppressive treatment.
– Grade I: mild rejection. This is characterized by a mild dermal lymphocytic infiltrate of recipient’s origin, forming small perivascular cuffs in the upper and occasionally also the mid dermis. The epidermis is as a rule unaffected ( (figure 2A) ). This grade corresponded to biopsies taken from normal-looking skin or occasionally from asymptomatic faint macular, non-infiltrated lesions ( (figure 2B) ) that clinically raised concern about graft rejection.
– Grade II: moderate rejection. This is characterized by a moderately dense dermal infiltrate, forming perivascular cuffs and diffusing interstitially between collagen bundles. The infiltrate is predominantly lymphocytic, with occasional histiocytic or epithelioid cells insinuating between collagen bundles. The epidermis is either unaffected or shows a mild degree of epidermal exocytosis and/or spongiosis, but does not contain necrotic keratinocytes ( (figure 3A) ). This grade was seen in biopsies taken from erythematous, slightly infiltrated maculopapular skin lesions suggestive of graft rejection ( (figure 3B) ).
– Grade III: severe rejection. This is characterized by both epidermal and dermal changes. The most consistent finding is a dense, mainly CD3+ lymphocytic infiltrate made of CD4+ and CD8+ cells forming nodules around capillaries of the upper dermis, larger blood vessels of the mid and lower dermis, and eccrine sweat glands. The epidermis may show spongiosis or exocytosis, and contains a variable number of scattered apoptotic keratinocytes in the stratum spinosum ( (figure 4A) ). The epidermis may develop lichenoid changes reminiscent of chronic (lichenoid) GVHD, i.e. orthokeratotic hyperkeratosis, hypergranulosis, acanthosis, papillomatosis, basal cell layer vacuolization and infiltration with lymphocytes (interface dermatitis). In these cases, the dermal lymphocytic infiltrate forms a band in the papillary dermis ( (figure 4B) ). Similar changes are seen on epidermal adnexae (hair follicles, eccrine glands). This grade was observed in biopsies from erythematous, infiltrated, scaly papules that were either isolated or confluent into plaques ( (figure 4C) ). These lesions were suggestive of graft rejection and were observed in the first HHA recipient, resulting from the progression of maculopapular lesions because of non-compliance with the immunosuppressive treatment.
– Grade IV: very severe rejection. This is characterized by an epidermis of variable thickness, comprising highly hyperplastic areas but also zones of epidermal thinning and necrosis, resulting from the confluence of necrotic keratinocytes ( (figure 5A) ). In areas where the epidermis is not (completely) necrotic, it shows intraepidermal lymphocytic exocytosis and basal cell vacuolization with focal invasion by lymphocytes. The dermis contains an inflammatory infiltrate forming large aggregates around blood vessels ( (figure 5B) ) and sweat glands ( (figure 5C) ), and smaller ones around tactile corpuscles and nerves; this is polymorphous, made mainly of activated (HLA-class II+) CD45RO+ memory T-cells, but containing also abundant eosinophils, occasional CD20+ B-cells, plasma cells, tryptase+ mast cells and histiocytic cells. The dermal eccrine secretory ducts show basal cell vacuolization, infiltration by lymphocytes, and often contain necrotic/apoptotic keratinocytes; they may also display malpighian metaplasia. The infiltrate extends focally to the hypodermis as perivascular nodules ( (figure 6A) ). This grade was observed in the amputation specimen of the first HHA recipient obtained during the 28^th month post-graft, that showed macroscopically, along with changes observed in previous grades, superficial erosive and necrotic areas ( (figure 6B) ).

Discussion

We have previously reported the pathological changes observed in the skin during allograft rejection in the first HHA [8, 9], and propose here, on the basis of the study of five additional HHA recipients, a scoring system for assessing the severity of allograft rejection. Monitoring the skin is justified considering previous observations in animals showing that in CTA the skin is the main target of graft rejection [12, 13], and the pathological study of the amputation specimen of the first HHA confirming that this is also the case in human CTA [9]. The fact that the skin can be easily observed macroscopically and biopsied, allowing the early clinical detection of graft rejection, is also an obvious advantage over the study of underlying tissues such as muscles, nerves or bones.

Most studies concerning limb allografts have been performed in animal models, and scoring systems assessing the histological severity of graft rejection have been proposed in the setting of rat [14, 15] or swine forelimb allotransplantation [16]. However, animal skin shows considerable histologic differences from human skin (concerning the quantitative and qualitative distribution of epidermal adnexae, thickness and pigmentation of the epidermis, and thickness of the dermis to name but a few), so that the pathological changes observed during allograft rejection in animals are not expected to be identical with those seen in human skin. In humans, a scoring system for assessing skin rejection in the setting of intestine and abdominal wall has been proposed [17]. This system includes five grades, and is similar to the one proposed here. Whether the difference in the grafted organ (intestine vs limb) has an influence on the nature of the rejection observed in the allografted skin (regarding intensity, delay post-graft, etc.) will remain speculative until more relevant experience is gained. More recently, based on the observation of two patients, a scoring system of five grades was also proposed for the rejection of HHA, which is similar to our own [18]. However, our study was carried out on a larger number of patients, which allowed us to observe a wider range of pathological alterations, including namely a case of very severe rejection observed in the first HHA recipient, after he had discontinued his immunosuppressive treatment. This resulted in severe clinicopathological alterations, including epidermal necrosis and sloughing, i.e. changes that had not been hitherto considered in previously reported scoring systems [17, 18]. This grade (IV) is equivalent to the grade IVb in the score of Schneeberger et al. [18], considered to occur only in animals.

The grading system we propose could be used as a basis for assessing rejection in other skin-containing CTA, since it can also be expected that in other types of CTA the skin will behave as the most antigenic tissue. This system could be further refined in the future by studying additional patients, and (more importantly) by observing them in the long term, since additional pathologic changes (such as dermal sclerosis or fibrosis) could develop after several years, as happens with skin lesions of GVHD. Another point that needs clarification is the functional role of the lymphoid infiltrate present in the skin during episodes of graft rejection. Indeed, this infiltrate consists of lymphocytes of recipient’s origin, as shown by the expression of the recipient-specific HLA antigens [8]. These cells express mainly the CD3+CD4+ or CD3+CD8+ phenotype. Although for technical reasons it has not been possible to demonstrate the presence of T regulatory cells on tissue sections, results obtained recently in vitro after isolation of skin-based cells from these biopsies show that a subset of them (in contrast with circulating lymphoid cells) expresses a phenotype of immunoregulatory cells (i.e. CD4/CD25/FoxP3) that induce tolerance rather than rejection [19, 20]. If this is confirmed, the functional properties of the lymphocytic infiltrate (more than its density) will obviously have to be considered in the assessment of the severity of rejection.

References

1 Hettiaratchy S, Randolph M, Petit F, et al. Composite tissue allotransplantation-a new era in plastic surgery? Br Ass Plast Surg 2004; 57: 381-91.

2 Tai C, Goldenberg M, Shuster K, et al. Composite Tissue Allotransplantation (CTA). J Invest Surg 2003; 16: 193-201.

3 Levi D, Tzakis A, Kato T, et al. Transplantation of the abdominal wall. Lancet 2003; 361: 2173-6.

4 Lanzetta M, Petruzzo P, Vitale G, et al. Human Hand Transplantation : what have we learned? Transplant Proc 2004; 36: 664-8.

5 Dubernard JM, Owen E, Lefrançois N, et al. First human hand transplantation. Case report. Transpl Int 2000; 13(suppl 1): 521-4.

6 Dubernard JM, Owen E, Herzberg G, et al. Human hand allograft: report on the first six months. Lancet 1999; 2: 1315-20.

7 www.handregistry.com (accessed May 2005).

8 Kanitakis J, Jullien D, Nicolas JF, et al. Sequential histological and immunohistochemical study of the skin of the first human hand allograft. Transplantation 2000; 69: 1380-5.

9 Kanitakis J, Jullien D, Petruzzo P, et al. Clinicopathologic features of graft rejection in the first human hand allograft. Transplantation 2003; 76: 688-93.

10 Petruzzo P, Revillard JP, Kanitakis J, et al. First human double hand transplantation: efficacy of a conventional immunosuppressive protocol. Clin Transplant 2003; 17: 455-60.

11 Kanitakis J. Anatomy, histology and immunohistochemistry of normal human skin. Eur J Dermatol 2002; 12: 390-401.

12 Lee WP, Yaremchuk MJ, Pan YC, et al. Relative antigenicity of components of a vascularized limb allograft. Plast Reconstr Surg 1991; 87: 401-11.

13 Lanzetta M, Ayrout C, Gal A, et al. Experimental limb transplantation, part II: excellent return of function and indefinite survival after withdrawal of immunosuppression. Transplant Proc 2004; 36: 675-9.

14 Fritz W, Swartz W, Rose S, et al. Limb allografts in rats immunosuppressed with cyclosporine. Ann Surg 1984; 199: 211-5.

15 Van den Helder T, Benhaim P, Anthony J, et al. Efficacy of RS-61443 in reversing acute rejection in a rat model of hindlimb allotransplantation. Transplantation 1994; 57: 427-33.

16 Zdichavsky M, Jones J, Ustuner E, et al. Scoring of skin rejection in a swine composite tissue allograft model. J Surg Res 1999; 85: 1-8.

17 Bejarano P, Levi D, Nassiri M, et al. The pathology of full-thickness cadaver skin transplant for large abdominal defects : a proposed grading system for skin allograft acute rejection. Am J Surg Pathol 2004; 28: 670-5.

18 Schneeberger S, Kreczy A, Brandacher G, et al. Steroid- and ATG-resistant rejection after double forearm transplantation responds to Campath-1H. Am J Transplant 2004; 4: 1372-4.

19 El Aljaafari A, Badet L, Farré A, et al. Isolation of skin T lymphocytes from hand transplant, and evaluation of their allo-response against donor APCs. Transplantation 2004; 78(suppl): 50; (abstract).

20 Dubois B, Chapat L, Goubier A, Kaiserlian D. CD4+CD25+ T cells as key regulators of immune responses. Eur J Dermatol 2003; 13(2): 111-6.