Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The prevalence of chronic ocular graft-versus-host disease (oGVHD) may be rising due to the increased number of cases of allogeneic hematopoietic stem cell transplantations (HSCT) performed to treat malignancies.
oGVHD is an immune-mediated disease of the ocular surface and the lacrimal gland and can occur in the absence of systemic GVHD.
Its clinical spectrum ranges from mild dry eye to vision-threatening inflammation and debilitating cicatricial scarring, but unlike other systemic ocular inflammatory diseases, organ-specific treatment rather than maximizing systemic immunosuppression is important in the management of oGVHD.
Early diagnosis and management of oGVHD is important in optimizing visual function and quality of life in HSCT patients and should involve comanagement by an ocular surface specialist and the HSCT team.
Hematopoietic stem cell transplantation (HSCT) is one of the most important advancements in the treatment of hematologic/immune disorders, metabolic disorders, and malignancies. Types of transplantation depend on the source of the donor cells: autologous (self), syngeneic (twin), and allogeneic (another individual). Classically referred to as bone marrow transplantation (BMT), in the modern era of HSCT, donor cells are harvested from the bone marrow, peripheral blood, and placental cord blood.
Graft-versus-host disease (GVHD) occurs in the setting of allogeneic HSCT where the cells from human leukocyte antigen (HLA)-matched donors (related or unrelated) can cause autoimmune-like disease for the host. Unlike graft rejection in a solid organ transplantation, where the host immune system responds to the transplanted organ, in GVHD the donor cells (graft) mount an immune response against the recipient (host patient), precipitating immunologic attack typically to the skin, gastrointestinal (GI) system, liver, mouth, lungs, and eyes. , In essence, donor-derived T-cells recognize host antigens, called minor histocompatibility antigens, as foreign and create an attack against them. Minor histocompatibility antigens are not included in routine HLA typing. , The number of affected organs and the severity of GVHD can vary widely, and the management of GVHD remains one of the most difficult medical challenges after HSCT.
The incidence of GVHD varies greatly, ranging from 10% to 90%, and multiple factors influence its prevalence, such as the degree of histocompatibility, the source of donor tissue, age, prophylaxis, and underlying disease among others. , GVHD has been traditionally classified as “acute” or “chronic,” purely by an arbitrary division of events occurring either within the first 100 days posttransplantation or after this mark, respectively. The distinction of acute and chronic GVHD has now been updated to reflect the disease characteristics rather than the time-point at which GVHD occurs posttransplantation. ,
For acute GVHD, the current grading system includes “classic acute GVHD” with skin, GI, and liver involvement within 100 days posttransplantation, and “persistent, recurrent, or new onset/late acute GVHD” for GVHD without the diagnostic signs of chronic GVHD occurring after 100 days posttransplantation. Similarly, reclassification of chronic GVHD criteria includes “classic chronic GVHD” for GVHD without acute features and “overlap syndrome” for chronic GVHD with acute features. ,
The National Institutes of Health (NIH) 2005 Chronic GVHD Consensus was initially established with its main objectives to: (1) streamline GVHD diagnostic criteria with emphasis on distinction between acute and chronic, (2) establish criteria for severity score of affected organs, and (3) identify categories of the disease and treatment indications, with the aim of providing a structure for future prospective clinical trials in chronic GVHD. , Of these, four are relevant to chronic ocular GVHD (oGVHD) and describe the diagnostic criteria, histology, and management, all of which were written for a target audience of HSCT specialists. , A second NIH Consensus Conference took place in 2014 to update the original criteria, , with similar application for ocular considerations, which are discussed under the diagnosis of chronic oGVHD section below.
GVHD remains a significant limiting factor for successful HSCT. , Prospective studies using the 2005 NIH Consensus criteria further demonstrated that the chronic GVHD score of skin, lung, and gastrointestinal tract each predicts the risk of transplant-related mortality (TRM). GVHD is associated with high morbidity and mortality post-HSCT even though GVHD is also associated with a lower relapse rate, presumably due to graft-versus-tumor effect. Success of allo-HSCT depends on the balance of harboring the benefit of GVHD in the form of graft-versus-tumor effect (graft-versus-leukemia/lymphoma effect [GVL]) while managing the morbidity from GVHD. , A large study by Boyiadzis et al. showed that the antileukemia effect of chronic GVHD may only be clinically beneficial in chronic myeloid leukemia patients and not in those with acute myeloid leukemia and acute lymphocytic leukemia. The field of GVL as it relates to GVHD is evolving, but the management of GVHD remains a critical challenge in the management of HSCT patients. ,
The pathogenesis of GVHD is a highly complex process, largely involving a cascade of interactions between donor-derived CD4+ and CD8+ T-cells and recipient host antigens. Even with HLA-identical grafts, moderate-to-severe acute GVHD can occur in approximately 40%–70% of all HSCT patients. , For acute GVHD pathophysiology, animal studies have elucidated three phases in its development: (I) recipient-conditioning tissue damage; (II) donor T-cell activation; and (III) target cell apoptosis. , The preparative regimen is a risk factor in the development and the severity of GVHD. During the first phase, intensive preparative conditioning, which includes high-dose chemotherapy and possibly radiation, causes tissue injury, releasing inflammatory cytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-1, and IL-7, which leads to activation of host antigen presenting cells (APC). Although both host-derived and donor-derived APCs have been shown to induce GVHD, host-derived APCs play a crucial role in the development of acute GVHD. , The preparative regimen also causes intestinal wall damage, through which bacterial cell wall components and lipopolysaccharide (LPS) can leak and stimulate mononuclear cells. In phase II, host APCs activate donor T-cells, which leads to the production of helper T-cell type I (Th1) cytokines (IL-2, IL-6, and IFN-γ). IFN-γ can further stimulate mononuclear cells to produce proinflammatory cytokines, such as IL-1 and TNF-α. In phase III, the Th1 cells promote the proliferation and differentiation of cytotoxic T lymphocytes (CTLs) and stimulate natural killer (NK) cells. These cells, together with mononuclear cells, induce apoptosis of the target organs. , Apoptosis is the hallmark histopathologic finding in various organs affected by GVHD. Current therapeutic and prophylactic approaches to GVHD focus on preventing the activation and proliferation of T-cells, using glucocorticoids, cyclosporine, tacrolimus, posttransplant cyclophosphamide, and monoclonal antibodies, , as well as promoting regulatory T-cell function through treatment of extracorporeal photopheresis. In 2017, ibrutinib—an inhibitor of Bruton tyrosine kinase in B-cells and IL-2-inducible T-cell kinase in T-cells—became the first US Food and Drug Administration (FDA)-approved drug for the treatment of steroid-refractory chronic GVHD.
For chronic GVHD, similar to acute GVHD, the vast literature on its topic underscores the complexity of its immune pathology, with the involvement of donor T-cells and B-cells, as well as other cells. Two target organs from the acute phase are thought to be among important areas in the development of chronic GVHD: (1) thymic epithelial cells (TECs) damaged by alloreactive T-cells, decreasing regulatory T-cells (Tregs) and increasing self-reactive T-cells, and (2) bone marrow microenvironment disturbance resulting in B-cell dysregulation.
Donor T-cells play a major role in the pathology of chronic GVHD, as evidenced by prophylactic in vivo T-cell depletion with alemtuzumab or antithymocyte globulin, lowering the incidence of chronic GVHD. The paradigm of acute GVHD being a T-helper cell (Th1) process and chronic being Th2 is being revisited owing to the fact that human data showing a mixed Th1/Th17 process have been found in chronic skin GVHD , The balance between Tregs and conventional T-cells (Tconv) is critical in achieving tolerance between donor-derived immunity and recipient, and a relative deficiency in Tregs has been associated with chronic GVHD. ,
While the cross talk between T- and B-cells in chronic GVHD remains elusive, the role of B-cells in chronic GVHD with presence of circulating antibodies to recipient cells has been long recognized. ,
Two types of autoantibodies associated with chronic GVHD are: (1) antibodies specific to alloantigen derived from gender disparity (male patient with female donor, developing anti-Y chromosome antibody), , and (2) antibodies against nonpolymorphic antigen such as platelet-derived growth factor receptor (PDGFR), suggesting a role in fibrosis development. Whether these autoantibodies are directly responsible for the pathogenesis of chronic GVHD or simply reflect dysregulation of B-cells remains unclear. B-cells may also play a role in chronic GVHD independent of autoantibodies by producing cytokines and chemokines as antigen presentation and as regulatory cells. The role of B-cells in the pathophysiology of GVHD continues to be a focus of active investigation. ,
Most of the available murine models for chronic GVHD simulate only a few pathologic manifestations of chronic GVHD (sclerotic chronic GVHD, fibrosis of skin and liver, anti-DANA antibodies), and some do not involve clinically relevant T-cell conditioning regimens. A few murine models do mirror a clinically relevant course, transitioning from acute to chronic scleroderma-like features and serum antibodies, or involving aggressive disease of multiorgans with fibrosis in lung and liver, associated with CD4 T-cells and B-cell infiltration. Murine models of oGVHD have also been described, involving the lacrimal gland , cornea, and limbus. Multiorgan fibrosis in chronic GVHD involving the liver, lung, and lacrimal glands has been demonstrated in a murine model, in which attenuation of the fibrosis process has been shown with a systemic antifibrotic agent (in the form of angiotensin II type 1 receptor antagonist). Another murine model has been described in which clinical phenotypes of corneal and ocular adnexal chronic GVHD in the context of systemic GVHD have been elucidated. Closely mimicking the findings of donor-derived cells playing an important role in human studies, , the importance of donor-derived cells in the oGVHD process has been recapitulated in the current model.
Potential therapeutics of chronic GVHD interventions may be in the area of targeting effector T-cells, Tregs, B-cells, or interrupting the fibrotic process for steroid refractory chronic GVHD. With enhanced understanding of the pathophysiology, additional potential preventive and therapeutic considerations for chronic GVHD will be possible. , , ,
oGVHD affects 40%–60% of patients receiving allo-HSCT and may affect various parts of the eye and eyelid, in both acute and chronic settings (onset varying from weeks to years). Ocular complications such as infections, cataracts, and dry eye syndrome occur in 60%–90% of bone marrow transplant patients, given their overall immunocompromised status and cancer treatments (i.e., chemotherapy, radiation, corticosteroid). oGVHD is an umbrella term to describe a postallogeneic transplantation condition that most commonly affects the ocular surface and lacrimal gland through inflammation, fibroblastic proliferation, and consequent fibrosis from donor-derived immune cells attacking the periductal areas of the lacrimal and meibomian glands as well as their secretory apparatus. , Onset of oGVHD can widely vary from weeks to years and may be delayed in HLA-matched HSCT and related donor HSCT patients, compared to HLA-mismatched and unrelated donor HSCT.
Ocular involvement in the setting of acute GVHD (e.g., conjunctivitis, epithelial sloughing, and pseudomembranes), though rarely observed, is a poor prognostic factor for mortality. , , The relatively low observation may be due to delay in ocular evaluation in patients who are systemically unstable. Risk factors for oGVHD include presence of skin and mouth involvement (or GVHD in multiple organs), , male hosts with female donors, repeated allo-HSCT, and preexisting diabetes. In patients with concomitant liver GVHD causing vitamin A deficiency, oGVHD involvement may be more severe. While devastating ocular sequelae (e.g., perforation, endophthalmitis) are rare, , treated oGVHD can maintain a stable clinical course without permanent vision loss, although it can significantly impair quality of life.
oGVHD typically involves the ocular surface, the cornea, and the conjunctiva, with histologic findings similar to those seen in cutaneous GVHD. Jabs et al. described a clinical staging for conjunctival oGVHD: hyperemia (even if subtle, Fig. 67.1 ) alone for stage I; hyperemia with chemosis and/or serosanguineous exudates for stage II; pseudomembranous conjunctivitis for stage III; and corneal epithelial sloughing for stage IV. Even if the clinical findings are subtle, the diagnosis of oGVHD should be considered when it occurs in the setting of systemic GVHD and no suspicion for infectious etiology. Typical symptoms include photophobia, redness, foreign body sensation, and other symptoms related to decreased aqueous tear function. Conjunctival swabs for viral culture should be considered to rule out other causes of mild hyperemia. For stage II, conjunctival hyperemia is associated with a spectrum of moderate chemosis to exuberant serosanguineous exudates ( Fig. 67.2A ). As systemic fluid retention can occur with corticosteroid treatment, it can be challenging to determine whether a chemotic conjunctiva is the result of a systemic fluid imbalance or oGVHD. Assessing the patient’s overall fluid status along with other causes of increased fluid retention such as hypoalbuminemia and hyponatremia can aid in the diagnosis of oGVHD. The presence of systemic GVHD, history of a recent medication change, reduction in aqueous tear function, and conjunctival biopsies may also help differentiate. Despite its aggressively inflamed appearance with significant hyperemia (stage II, see Fig. 67.2B ), symptoms may vary from severe pain to only mild photophobia.
Occurring in 12%–17% of acute GVHD patients and 11% of chronic GVHD patients in the same study by Jabs et al., stage III is characterized by hyperemia accompanied by conjunctival epithelial sloughing leading to pseudomembranous changes ( Fig. 67.3 ). The pseudomembranes of oGVHD can quickly scar and become “membranous” with associated bleeding upon removal. , , In the most severe cases, the corneal epithelium is also involved, resulting in up to one-third with corneal epithelial sloughing ( Fig. 67.4 ). Stage IV oGVHD resembles severe skin GVHD, with generalized erythema progressing to bullous formation and desquamation, and GI GVHD, with crypt dropouts, epithelial necrosis, and sloughing.
Conjunctival biopsy may aid in the diagnosis and management of oGVHD when obvious features are absent (low-grade GVHD) or when GVHD is confined to the ocular structures. , , , Histologically, the features of conjunctival GVHD include lymphocyte exocytosis, satellitosis, vacuolization of the basal epithelium, dyskeratosis with loss of epithelium, and epithelial cell necrosis with apoptotic bodies in the epithelium. , , , , Nonspecific features can also be observed, including epithelial attenuation, depolarization of the epithelial layer, and goblet cell depletion ( Fig. 67.5 ); these findings alone are insufficient for the diagnosis of oGVHD. T-cell analysis performed on one acute conjunctival GVHD biopsy specimen showed an abundance of CD4+ helper T-lymphocytes, whereas a conjunctival specimen from a patient with chronic GVHD showed predominantly suppressor/cytotoxic cells, the significance of which remains unclear. The histology of pseudomembranes/membranes ( Fig. 67.6 ) consists of inflammatory cells, fibrinous material, and cellular debris, and one case report identified donor-derived T-cells. Corneal histologic findings in murine GVHD models include epithelial atrophy, stromal edema, neovascularization, and inflammatory infiltrates (comprised primarily of donor T-cells); however, the endothelium did not appear involved. ,
Conjunctival oGVHD may serve as a marker for the severity of acute systemic GVHD and may reflect the course of systemic GVHD. In those affected by acute systemic GVHD and conjunctival involvement, the oGVHD occurred approximately two weeks after the onset of the acute systemic GVHD. These patients had worse survival than those with the same grade of acute GVHD without conjunctival involvement, suggesting that the presence of conjunctival disease may be a poor prognostic factor.
Ocular manifestations of chronic GVHD mimic those of other immune-mediated inflammatory diseases of the ocular surface. Dry eye symptoms of irritation, foreign body sensation, light sensitivity, burning, tearing, and redness are most commonly reported. Impression cytology has demonstrated atypical epithelial cells, goblet cell loss, altered epithelial cell morphology, squamous metaplasia, and predominant inflammation in chronic GVHD patients (91% in ocular and 68% in nonocular GVHD patients in one study). , There may be ongoing subclinical immune damage to the ocular surface in post-HSCT patients who have not yet developed oGVHD clinical features.
Other clinical manifestations due to cicatricial conjunctival changes result in conjunctival scarring, symblepharon formation, and altered eyelid anatomy (ectropion, entropion, punctal stenosis, meibomian gland atrophy). , , Histologically, there is infiltration of the conjunctiva by cytotoxic T-cells. Chronic oGVHD can occur in the absence of systemic GVHD, which suggests that the ocular surface can serve as an isolated target months or years after bone marrow or stem cell transplantation. ,
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here