Adult Cutaneous Chronic Graft-Versus-Host Disease: An In-Depth Review

Tissue Injury and Early Inflammation

In early cGVHD, tissue injury and inflammation initiate immune dysregulation and fibrosis (Figure 1). The preparative regimens that include high-dose chemotherapy and irradiation, cause direct cellular injury and disrupt epithelial barriers, leading to the release of damage-associated molecular patterns such as extracellular adenosine triphosphate (ATP). Extracellular ATP acts as a danger signal by binding P2X7 receptors on antigen-presenting cells (APCs), promoting costimulatory molecules (eg, CD80 and CD86) expression and interferon-γ secretion that drives donor T-cell expansion.^5,12 Concurrently, the disruption of the mucosal barrier facilitates microbial translocation, thereby exposing pattern recognition receptors to pathogen-associated molecular patterns and further intensifying the inflammatory cascade. Transcriptomic analyses of cutaneous cGVHD lesions confirm activation of interferon-driven pathways, including STAT1- and TBX21-associated Th1 signaling, as well as upregulation of pattern-recognition receptor and phagocytic pathways, supporting the central role of innate immune activation in early disease.^13,14 This environment activates endothelium and recruits donor immune cells into target tissues. Moreover, neutrophil recruitment, stimulated by chemokine gradients and microbial factors, amplifies tissue damage through the production of reactive oxygen species and proteolytic enzymes, which further deteriorate tissue integrity. ¹⁵ In parallel, early monocyte reprogramming has been demonstrated in preclinical models, where IL-17-dependent cGVHD is associated with transcriptional activation of circulating monocyte subsets enriched for NF-κB and Mitogen-Activated Protein Kinase (MAPK) signaling, indicating that systemic innate immune dysregulation precedes overt fibrosis. ¹⁶ In addition, early inflammatory signals contribute to the loss of regulatory T cell function, compromising peripheral tolerance mechanisms and perpetuating the cycle of tissue injury. ¹⁷ Histopathologic findings in affected tissues demonstrate vascular injury and immune cell infiltration in early disease, features that are thought to contribute to immune dysregulation and, over time, to the fibroproliferative changes characteristic of advanced cGVHD. ¹⁸

OPEN IN VIEWER

Figure 1.

Pathophysiology of chronic cGVHD. The schematic illustrates the multi-step immunopathogenic cascade leading to tissue injury and fibrosis in chronic cutaneous GVHD. Damage to the thymus impairs negative selection of autoreactive donor-derived T cells, resulting in their persistence within the peripheral immune repertoire. Concurrently, epithelial barrier injury in the skin leads to the release of DAMPs such as extracellular ATP and the translocation of PAMPs from the microbiota. DAMPs engage purinergic receptors (notably P2X7) on APCs, while PAMPs further stimulate APC activation, promoting secretion of IFN-γ and other pro-inflammatory cytokines. Activated APCs and the cytokine milieu drive endothelial cell activation, enhancing adhesion molecule expression and vascular permeability, thereby promoting neutrophil recruitment into the dermis. Recruited neutrophils release ROS and proteolytic enzymes, amplifying local tissue damage and sustaining DAMP release. In parallel, donor-derived CD8⁺ T cells acquire a Tc17/Th17-like phenotype, producing IL-17 and other inflammatory mediators that perpetuate inflammation and tissue injury. B-cell activation, stimulated by elevated BAFF and supported by T-cell signaling, results in autoantibody production and further T–B cell cross-talk, which together promote fibroblast activation. These combined immune events culminate in fibroblast and myofibroblast proliferation, excessive extracellular matrix deposition, and cutaneous fibrosis, a hallmark of the chronic phase of the disease. APC, antigen-presenting cell; ATP, adenosine triphosphate; BAFF, B-cell activating factor; DAMP, damage-associated molecular pattern; IFN-γ, interferon-gamma; PAMP, pathogen-associated molecular pattern; ROS, reactive oxygen species; cGVHD, chronic graft-versus-host disease.

Dysregulated B-Cell and T-Cell Immunity

Thymic injury from conditioning and acute GVHD impairs central tolerance, allowing expansion of autoreactive T cells, reducing the expression of tissue-restricted antigens such as Class II-associated Invariant Chain Peptide (CLIP) that are crucial for eliminating autoreactive T cells during negative selection (Figure 1).^19,20 This disruption permits generation of autoreactive T cells, while donor CD8⁺ T cells further damage medullary thymic epithelial cells and impair selection.^21,22 In parallel, there is an expansion of a T helper 17 (Th17)-prone subset characterized by elevated RORγt expression and a mixed Th1/Th17 cytokine profile, including increased production of IL-17 and IFN-γ, which intensifies local inflammation and recruits additional effector cells to target tissues.^23,24 Ribonucleic acid (RNA) sequencing of cutaneous cGVHD has demonstrated that lichenoid disease displays a mixed Th1/Th17 signature, whereas sclerotic disease is predominantly Th1-driven with a unique TSLP-OX40 axis. Strong enrichment of IFN-related gene expression is a shared feature of both cGVHD subtypes, while the TREM1 pathway appears specific to lichenoid disease.^13,14 Moreover, IFN-inducible chemokines such as CXCL9 and CXCL10 are highly expressed in affected skin, supporting recruitment of CXCR3⁺ effector T cells and amplification of local inflammation. ¹³ Reduced thymopoiesis, as evidenced by lower T cell receptor excision circle levels, further impairs the generation of naïve T cells, exacerbating the dominance of pathogenic, autoreactive clones. Deficiencies in regulatory T cell populations, including conventional regulatory T cells, fail to restrain these aberrant responses, thereby perpetuating the cycle of autoimmunity and tissue damage. ²⁵

B-cell dysregulation further fuels chronic inflammation. Activated B cells, acting as potent APCs, enhance the activation of T follicular helper cells and support the formation of ectopic germinal centers, where aberrant B-cell clonal expansion occurs. ²⁶ This process leads to the production of pathogenic autoantibodies that contribute to tissue injury and fibrosis. Transcriptomic data from sclerotic cGVHD skin reveal upregulation of Fc receptor-associated and phagocytic pathways, suggesting that immune complex-mediated activation of macrophages may further amplify tissue inflammation. ¹³ In addition, elevated levels of B-cell-activating factors, such as B-cell activating factor (BAFF), provide survival signals that promote the persistence of autoreactive B cells, thereby reinforcing the chronic inflammatory state. ²⁶ Failure to regulate B-cell responses, combined with deficient regulatory lymphocytes, sustains immune activation, autoantibody production, and end-organ damage central to cGVHD pathogenesis. ²⁵

Fibrosis

Fibrosis in cGVHD results from a persistent, dysregulated wound-healing response in which prolonged immune activation drives extracellular matrix deposition and irreversible tissue remodeling. Initial inflammatory insults, together with sustained release of profibrotic cytokines such as transforming growth factor-β (TGF-β), create a microenvironment that favors fibroblast activation and differentiation into myofibroblasts. ²⁷ Bulk RNA sequencing of sclerotic cGVHD skin demonstrates upregulation of canonical fibrotic pathways, including TGF-β-associated genes (TGFB1, TGFB3), collagen genes (COL3A1, COL11A1), fibronectin (FN1), and wound-healing signatures, confirming active matrix remodeling in affected tissues. ¹³ Moreover, distinct transcriptomic subsets have been identified within sclerotic cGVHD, including a fibroinflammatory subtype characterized by concurrent Th1 and fibrotic signaling and a predominantly fibrotic/TGF-β-driven subtype, suggesting biological heterogeneity in fibrogenic mechanisms. ¹³ Donor-derived macrophages, particularly those activated in a CSF-1-dependent manner, contribute to the establishment of a TGF-β-rich milieu that intensifies collagen deposition and sclerotic changes in target organs like the skin and lungs. ²⁸ Preclinical single-cell analyses have shown that CSF-1 receptor-dependent monocyte/macrophage populations expand during IL-17-driven cGVHD and display transcriptional programs enriched for activation, chemotaxis, and inflammatory signaling, linking systemic monocyte dysregulation to tissue fibrosis. ¹⁶ In addition, IL-17-dependent transcriptional signatures in circulating monocytes can be detected before clinical diagnosis of cGVHD becomes apparent, supporting a role for early systemic immune reprogramming in subsequent fibrotic evolution. ¹⁶

Concurrently, dysregulated B-cell immunity plays a pivotal role in sustaining fibrosis through disrupted B-cell homeostasis and autoantibody production. Excess levels of BAFF rescue autoreactive B cells from peripheral deletion, enabling these cells to persist and become activated, which in turn leads to the production of pathogenic antibodies that form immune complexes and stimulate the release of additional fibrogenic cytokines. ²⁹ Altered B-cell subsets, marked by a reduction in naïve B cells and a proportional increase in activated or plasmablast-like cells, further contribute to chronic inflammatory signals that drive collagen synthesis.^30,31 This breakdown in normal B-cell tolerance, compounded by the impact of autoantibody formation, reinforces a self-sustaining cycle of inflammation and fibrosis that ultimately can culminate in the scleroderma-like features observed in cGVHD. ²⁵

The skin is the largest and most immunologically active barrier organ, containing abundant resident dendritic cells, macrophages, and mast cells, that facilitate alloimmune recognition and effector cell recruitment. Transcriptomic analyses confirm enrichment of Th1 signaling, macrophage activation pathways, neuroinflammatory signatures, and TGF-β-driven fibrotic programs within sclerotic skin lesion.^13,14 Donor-derived T cells and macrophages, particularly those activated in a CSF-1-dependent manner, infiltrate the skin and drive local production of profibrotic cytokines such as TGF-β and IL-17, leading to fibroblast activation and extracellular matrix deposition.^28,32-35 Keratinocytes, the main targets of donor T cells, can amplify fibrosis by producing TGF-β1 in response to IFN-γ and apoptotic signals, further linking inflammatory and fibrotic pathways. ³⁵ Mast cells also contribute to fibrosis and effector cell recruitment in the dermis. ³³

Although photoexposure and other local insults can precipitate or localize cGVHD lesions, these are not the primary reason for the skin’s overall predominance as a target organ. Instead, the skin’s barrier function, immune cell composition, and regenerative dynamics make it especially vulnerable to the dysregulated immune responses and maladaptive repair that characterize cGVHD.^5,25,36 Table 1 summarizes currently recognized risk factors for cGVHD highlighting their molecular significance.

OPEN IN VIEWER

Table 1.

Risk Factors for Cutaneous cGVHD with Proposed Mechanisms.

Risk factor	Proposed mechanism
Older recipient age 37	Thymic involution impairs central tolerance, increasing autoreactive T cell output and fibrotic potential.5,38,39
Prior acute GVHD 37	Establishes chronic immune activation and tissue damage, priming for dysregulated repair and cGVHD. 40
Peripheral blood stem cells 37	Contain higher T cell counts than marrow, increasing risk of alloimmune activation and fibrosis. 41
Total body irradiation (>450 cgy) 42	Damages epithelial and endothelial barriers, releases DAMPs, and promotes profibrotic cytokine signaling (eg, TGF-β).5,43-45
Major ABO mismatch 42	May reduce alloreactive responses due to altered antigen presentation, potentially reducing fibrotic manifestations. 42
Varicella-zoster virus reactivation46-48	Triggers local immune activation and tissue damage in dermatomes, leading to lesion flare or new-onset cGVHD.5,49,50
Photoexposure 51	UV light increases antigen presentation and skin inflammation, promoting overlap GVHD features. 5

Abbreviations: cGHVD, chronic graft-versus-host disease; DAMP, damage-associated molecular patterns; TGF-β, transforming growth factor-beta.

Nonsclerotic Manifestations

Nonsclerotic cutaneous cGVHD is a highly heterogeneous entity that encompasses a broad range of clinical morphologies, which is why the clinical community now prefers the term “nonsclerotic” over the prior/legacy nomenclature of “lichenoid GVHD.” ⁵ One of the most common presentations is the lichen planus-like variant, in which patients develop well-demarcated, pink to purple/violaceous papules and plaques with lacy skin changes that closely resemble idiopathic lichen planus, yet tend to be more extensive and involve lips/oral mucosa and atypical sites including face, neck, palms, and soles. ⁶⁴ In individuals with darker skin tones/SOC, the conventional erythema is often masked by baseline pigmentation and may instead present as a grayish hue, making subtle inflammatory changes more challenging to discern. ¹¹

Beyond lichen planus-like lesions, nonsclerotic cGVHD may also manifest as poikiloderma, psoriasis-like eruptions, eczema-like lesions, or keratosis pilaris-like changes, all of which frequently begin with symptoms of dry, scaly, and intensely pruritic skin.^66,67 While primary care physicians may consider using antihistamines for the management of itch, their use should be avoided in cGVHD patients as they can exacerbate existing sicca symptoms in this population. As the acute inflammatory phase subsides, residual dyspigmentation and ichthyosis often emerge as post inflammatory changes; although these features reflect prior damage rather than active disease, they can be cosmetically disfiguring and significantly impact quality of life. ⁶⁸

Adnexal involvement is common, with hair changes ranging from nonscarring alopecia—characterized by scalp scaling, pruritus, and diffuse thinning—to scarring (cicatricial) variants, including lichen planopilaris-like alopecia that may result in permanent follicular loss.^69,70 Nail abnormalities range from mild longitudinal ridging and splitting to severe dystrophy/pterygium that may interfere with daily function (Table 2).^69,70 Moreover, nonsclerotic cutaneous cGVHD can mimic various autoimmune conditions such as vitiligo/leukoderma, alopecia areata, cutaneous lupus erythematosus, and even dermatomyositis, blurring the lines between distinct cGVHD pathology and other autoimmune phenomena.^71-73

Interestingly, additional studies have reported that adnexal involvement, including vitiligo and alopecia areata, may correlate with more severe disease course, suggesting that nonsclerotic manifestations may herald a broader dysregulation of the immune system post-transplant.^74,75 These presentations illustrate the complexity of nonsclerotic cGVHD. ⁷²

Sclerotic Manifestations

Sclerotic manifestations of cGVHD can involve both the superficial and deeper layers of the skin and may arise de novo or evolve as a progression from non-sclerotic lesions.^{8,25,59,65,76} Superficial sclerotic changes are typically characterized by lichen sclerosus-like plaques that typically present as well-demarcated, ivory, and shiny plaques/patches with cigarette paper like atrophy and/or follicular plugging sometimes appearing in a setting of poikiloderma (Table 2).^59,76 These changes are readily appreciated on visual inspection and may also be detected on palpation, with reduced skin pliability or elasticity compared to adjacent uninvolved skin. ⁹ In contrast, deeper fibrosis presents as morphea-like or eosinophilic fasciitis (EF)-like lesions that may extend into the reticular dermis, subcutaneous tissue (eg, panniculitis), and fascia (eg, eosinophilic fasciitis).^59,76,77 While in idiopathic forms, morphea and EF generally start as round or oval plaques that gradually become indurated and firm, in cGVHD plaques are often ill defined and more widespread.^76,78 Early morphea-like and/or EF-like cGVHD is challenging to diagnose as early lesions present with cellulitis-like edema and/or violaceous erythema (depending on the depth of involvement).^76,79 Once fibrosis sets, the skin becomes tight, difficult to pinch and often hyperpigmented in the center with or without lilac halo or peripheral erythema. When skin overlying joints is affected, the risk of joint motion restriction and functional limitation is high necessitating prompt and aggressive therapies. (Supplemental Appendix 1)

Histopathologically, sclerotic cGVHD is distinguished by a homogenized, edematous collagen band with a loss of elastic fibers and only minimal inflammatory infiltrate, features that distinguish it from other fibrosing disorders such as systemic sclerosis. ⁷⁶ Clinical series have also described variants where sclerotic changes are associated with features resembling hypertrophic lupus erythematosus and eosinophilic fasciitis, further expanding the recognized spectrum of sclerotic GVHD. ⁸⁰ Although sclerotic cGVHD more commonly affects the trunk and extremities, involvement of the facial and buccal fascia can occur and may result in reduced oral aperture and perioral tightness. In contrast, genital tract involvement is common yet often underrecognized, necessitating thorough examination and prompt referral to gynecology or urology.^9,80,81 These sclerotic alterations lead to significant functional impairments, including joint contractures and chest wall rigidity that can result in pulmonary restriction, as well as chronic ulcers, angiomatosis, calcinosis, and an elevated risk for cutaneous squamous cell carcinoma (cSCC), which in this setting is best described as Marjolin Ulcer^82,83 and has distinct pathgenesis and high-risk/poor prognosis, when compared to conventional ultraviolet-driven cSCC. Early recognition and management may mitigate fibrosis progression.^1,8,76,80,84

Considerations for SOC Patients

In patients with richly pigmented skin, conventional findings such as erythema may be subtle or absent, and cGVHD may manifest as dyspigmentation (ash-color, hypo- or hyperpigmentation), poikiloderma, or lichen planus-like violaceous papules that may be less apparent than in lighter skin (Table 2). Sclerotic changes may present as areas of induration, loss of skin markings, or shiny, bound-down plaques, but the degree of visible sclerosis may be underestimated in the SOC.^25,85,86

Histopathologic features of cutaneous cGVHD appear broadly similar across skin phototypes, with epidermal/interface changes (eg, basal keratinocyte apoptosis, vacuolar change, and lymphocytic satellitosis) commonly observed in epidermal cGVHD, and sclerosis affecting the dermis and sometimes subcutis/fascia in sclerotic disease.^59,86 In a retrospective cohort of Korean patients, lichenoid patterns predominated clinically and histologically, with sclerodermoid changes and acute/chronic overlap was also reported.^59,86 Data specifically evaluating histopathologic differences by SOC remain limited; however, pigment incontinence and dermal melanophages may be more conspicuous in lichenoid/interface patterns and may contribute to clinically apparent dyspigmentation.

Grading System for Cutaneous cGVHD

The grading of cutaneous cGVHD relies primarily on the NIH Consensus Criteria, which were first introduced in 2005 and subsequently updated in 2014. ⁹⁶ This system remains the most widely used and standardized approach for assessing the severity of cGVHD across multiple organ systems. The NIH criteria employ a four-point grading scale (0-3), where 0 indicates no involvement, 1 represents mild disease, 2 indicates moderate severity, and 3 corresponds to severe disease with significant functional impairment. ⁹⁷ For cutaneous cGVHD, key features incorporated into the grading system include BSA involvement, the presence of erythema or violaceous skin changes, sclerosis, panniculitis, fascial restriction, ulceration, and the extent of genital, oral, and ocular involvement. ⁹⁸ Skin findings vary in SOC patients, where erythema is less prominent. These parameters enable clinicians to categorize global disease severity as mild, moderate, or severe, facilitating consistent clinical assessments and therapeutic decision-making. ⁹⁹

Patients with mild disease typically exhibit limited skin involvement (<18% BSA, following “The Rule of Nines” assessment) without significant functional disability. Moderate disease involves more extensive skin manifestations (18%-50% BSA) and may affect multiple organs, though without critical impairment. Severe disease is characterized by widespread skin involvement (>50% BSA), deep sclerotic features, erosions/ulcers and/or functional limitations such as restricted joint mobility due to fascial sclerosis. ⁹⁶

In addition to the NIH clinical grading criteria, adjunct PRO measures such as the Lee Symptom Scale (LSS) and the SF-36 have been explored to complement the assessment of cGVHD severity and better capture symptom burden and quality of life impact.^100-102 The LSS, a validated PRO measure, is frequently used to quantify the symptomatic burden of cGVHD across different organ systems (Supplemental Appendix 2). ¹⁰³ Additionally, the Short Form-36 (SF-36), a general health-related quality of life instrument, has been widely applied in cGVHD research and clinical trials (Supplemental Appendix 3). ¹⁰⁴ In some cases, the Modified Rodnan Skin Score (mRSS), originally developed for systemic sclerosis, has been used to assess skin thickening and fibrosis in cGVHD patients (Supplemental Appendix 1). ⁹⁸ However, these methods have not outperformed the NIH scoring system in predicting clinical outcomes. ¹⁰⁵

Recent advancements in technology-assisted and imaging-based methods have sought to improve the objectivity and reproducibility of cGVHD grading. Artificial/augmented intelligence (AI)-based algorithms have demonstrated promising accuracy in analyzing skin photographs to quantify erythema and sclerosis, offering an automated and standardized approach to disease assessment. ¹⁰⁶ Additionally, 3D photography and hyperspectral imaging have been introduced to enhance BSA mapping and segmentation of affected skin, allowing for more precise measurements.^107,108 These imaging technologies have been particularly useful in distinguishing active versus “burned out” sclerotic cGVHD. ¹⁰⁹ Some studies have also explored the utility of skin punch biopsies to measure changes in fibrosis depth over time, though these remain an adjunct rather than a primary grading tool. ⁹⁹

Furthermore, automated computer-aided image segmentation has shown promise in improving the detection of erythema and sclerosis in cGVHD patients. AI models trained on patient images have demonstrated the ability to identify subtle disease activity that may be overlooked during routine clinical evaluations. ¹⁰⁵ One study suggested that erythema scoring is a potential prognostic marker, with higher erythema levels correlating with poorer survival outcomes in cGVHD. ¹⁰⁶ Unfortunately, though, erythema can be difficult to assess in SOC patients. While these technological innovations are promising, further validation in large-scale clinical studies is required before they can be fully integrated into routine practice. ¹⁰⁷

Despite these advances, challenges remain in achieving widespread standardization of non-invasive imaging techniques and AI-driven diagnostic tools. While the NIH criteria continue to serve as the cornerstone of cGVHD assessment, incorporating novel PRO measures, AI-powered image analysis, and advanced imaging modalities may enhance diagnostic accuracy, enable early detection of disease progression, and facilitate more precise therapeutic interventions. ¹⁰⁹

Skin-Directed Therapies

Systemic Therapies

Systemic corticosteroids remain the cornerstone of first-line therapy for NIH-defined moderate to severe cGVHD, including its cutaneous manifestations. They are typically initiated at a prednisone-equivalent dose of 0.5 to 1 mg/kg/day, often combined with a CNI such as cyclosporine or tacrolimus or a JAK inhibitor (as detailed later). ¹⁷³ This approach is consistent with the recently published Cell Therapy Transplant Canada Consensus-Based Guideline, which provides the first nationwide recommendations for diagnosis and treatment of cGVHD and emphasizes standardized systemic management across Canadian centers. ⁸⁷

Despite broad anti-inflammatory effects, corticosteroids provide only partial and often temporary control. In a large single-center study by Inamoto et al, the 12-month failure-free survival (FFS) following initial systemic therapy, which included corticosteroids, was only 54%. Nearly half of patients either required second-line/another systemic treatment, experienced non-relapse mortality, or had recurrent malignancy during initial therapy. ¹⁷³ Notably, high initial doses did not correlate with improved outcomes.

Real-world data further emphasize the limitations of corticosteroids. In a claim-based analysis by Bachier et al, corticosteroids were the most frequently prescribed therapy in first-, second-, and third-line settings. Yet over 70% of patients progressed beyond first-line treatment, and 47% required a third-line, illustrating the frequent need for additional agents and the lack of durable control with systemic corticosteroids alone. ⁴ Additionally, the chronic use of steroids contributes significantly to morbidity, particularly in patients with skin involvement, where steroid-induced myopathy, skin thinning, and delayed wound healing compound disease-related impairment. They are associated with a high burden of toxicity and suboptimal long-term efficacy. The combined inflammatory and fibrotic pathophysiology likely explains limited steroid responsiveness in fibrotic skin disease. The development of fibrosis in cGVHD involves macrophage infiltration, Th17 signaling, and fibroblast activation, processes that are not easily addressed by steroids alone. ¹⁷⁴

Recent therapeutic approaches for cGVHD have transitioned from broad immunosuppressive regimens relying heavily on prolonged, high-dose corticosteroids to more targeted strategies directed at specific immune pathways implicated in disease mechanisms (Table 3, Figure 2). These emerging therapies selectively target donor-derived alloreactive T lymphocytes, auto- and alloreactive B lymphocytes, as well as CD4⁺ FoxP3⁺ regulatory T cells. ¹⁷⁵ Significant progress in understanding cGVHD pathogenesis, alongside improved diagnostic guidelines, refined disease severity grading, clearer response criteria, and standardized clinical trial endpoints, has culminated in the FDA’s approval of multiple agents for patients requiring second-line therapy or beyond.

OPEN IN VIEWER

Figure 2.

Integrated cellular and molecular mechanisms underlying cGVHD pathogenesis and therapeutic targets. The figure illustrates the complex interplay between donor-derived immune cells and tissue-resident effectors that drive the pathogenesis of cGVHD. Thymic injury impairs central tolerance and permits the emergence of autoreactive donor T cells that become activated upon encountering host APCs. Activated APCs promote differentiation of Th and cytotoxic (Tc) subsets through cytokine and TCR signaling cascades involving JAK1/2–STAT1/3 and ITK pathways. Ruxolitinib, a JAK1/2 inhibitor, attenuates this signaling, thereby reducing effector T-cell activation. Activated donor T cells polarize toward Th1/Tc1 (T-bet, STAT1) and Th17/Tc17 (RORγt, STAT3) lineages, producing IFN-γ and IL-17, which perpetuate tissue inflammation. ROCK2 signaling contributes to Th17/Tc17 differentiation and suppression of Tregs through inhibition of FOXP3/STAT5; blockade by belumosudil restores immune balance. Tfh cells within germinal centers secrete IL-21, driving B-cell maturation and differentiation into plasma cells that produce auto- and alloantibodies. Elevated BAFF supports survival and activation of memory B cells through BCR signaling mediated by BTK and SYK kinases, which are inhibited by ibrutinib. Circulating monocytes differentiate into alternatively activated tissue macrophages, which interact with immune complexes through Fc receptors (FcR) and secrete IL-17 and TGF-β, promoting fibroblast activation and tissue fibrosis, the pathological hallmark of cGVHD. The CSF-1/CSF-1R axis sustains macrophage survival and differentiation; axatilimab, a monoclonal antibody targeting CSF-1R, interrupts this profibrotic signaling loop. Together, these interconnected immune and stromal pathways link defective thymic tolerance, sustained T- and B-cell activation, and macrophage-driven fibrogenesis in the chronic phase of GVHD, highlighting multiple therapeutic targets currently under clinical investigation. APC, antigen-presenting cell; BAFF, B-cell activating factor; BCR, B-cell receptor; BTK, Bruton’s tyrosine kinase; CSF-1(R), colony-stimulating factor 1 (receptor); IFN-γ, interferon-gamma; IL, interleukin; ITK, IL-2-inducible T-cell kinase; JAK, Janus kinase; ROCK2, Rho-associated coiled-coil kinase 2; STAT, signal transducer and activator of transcription; TCR, T-cell receptor; Tfh, T follicular helper cell; Th, T helper cell; Treg, regulatory T cell; TGF-β, transforming growth factor-beta, cGVHD, chronic graft-versus-host disease.

Axatilimab

Axatilimab is a humanized IgG4 monoclonal antibody that selectively targets the colony-stimulating factor 1 receptor (CSF-1R), a key regulator of macrophage survival, proliferation, and differentiation. Given the central role of profibrotic macrophages in cGVHD pathogenesis, particularly in driving fibrosis and sustained inflammation, axatilimab has emerged as a therapeutic strategy for patients with advanced or refractory cGVHD.

The AGAVE-201 trial, a pivotal phase II multinational randomized study, evaluated axatilimab at 3 dosing regimens (0.3 mg/kg every 2 weeks, 1 mg/kg every 2 weeks, and 3 mg/kg every 4 weeks) in 241 patients with recurrent or refractory cGVHD leading to FDA approval. The primary endpoint, ORR within the first 6 cycles, was met in all dosing groups: 74% in the 0.3 mg/kg group, 67% in the 1 mg/kg group, and 50% in the 3 mg/kg group. Additionally, a clinically significant reduction in symptom burden (>5-point drop in the modified LSS) was reported in 60%, 69%, and 41% of patients, respectively. In patients with sclerotic skin manifestations, which constituted over 90% of the cohort, reduction in sclerotic skin surface area was observed in 44% of those receiving 0.3 mg/kg every 2 weeks, 34% of those on 1 mg/kg every 2 weeks, and 60% of those on 3 mg/kg every 4 weeks. Correspondingly, improvements in skin sclerosis were reported in 66%, 56%, and 60% of patients across these dose levels, respectively. These objective measures were mirrored by patient-reported symptom relief: 73%, 77%, and 68% of participants in the 0.3 mg/kg, 1 mg/kg, and 3 mg/kg arms, respectively, experienced a clinically meaningful reduction in skin-thickening symptoms on the modified LSS. ¹²³

The median time to response was under 2 months and 91% of cutaneous improvements occurring within 6 months. Many patients experienced rapid corticosteroid tapering or discontinuation. Twelve-month overall survival was high across all groups, exceeding 80%, with a 12-month FFS ranging from 44% to 59%. Axatilimab-related adverse events were generally dose-dependent and consistent with CSF-1R inhibition but rarely resulted in treatment discontinuation, especially at the 0.3 mg/kg dose. ¹²³

Building on these findings, the phase I/II open-label study by Kitko et al evaluated axatilimab in 40 patients with active cGVHD after failure of at least 2 prior lines of systemic therapy. The study identified 3 mg/kg every 4 weeks as the optimal biologic dose, with a recommended phase II dose (RP2D) of 1 mg/kg every 2 weeks based on favorable safety, pharmacodynamic activity, and response outcomes. Among evaluable patients, the ORR was 67%, with responses across all organs, including skin and lung involvement. In the phase II cohort, the ORR at cycle 7 day 1 was 50%, but when considering best responses within the first 6 cycles, the ORR increased to 82%. At baseline, 35 of the 40 patients (87.5%) enrolled in the Kitko et al study had active cutaneous involvement (77.5% with a skin features score ≥2), reflecting the heavy fibrotic burden in this cohort. Although only 5 of these 35 patients (14%) achieved a formal partial or complete skin response by NIH 2014 criteria during treatment, the majority experienced meaningful improvements in fibrotic manifestations: among the 31 patients with sclerotic skin disease, 26 (84%) showed investigator-rated decreases in skin tightening and reported symptom relief on the LSS. Responses were rapid, with a median time to first documented skin improvement of 5 weeks (range 4-20), and durable, as many patients maintained reduced induration and greater range of motion through 20 weeks of follow-up. These findings underscore axatilimab’s capacity to ameliorate both clinical and patient-reported measures of cutaneous sclerosis in refractory cGVHD, even when formal response rates appear modest. Axatilimab also demonstrated a favorable safety profile with most adverse events being mild and on-target (eg, liver enzyme elevations, fatigue, periorbital edema) and no cytomegalovirus reactivations or invasive fungal infections were observed. ¹⁹⁰ Taken together, these organ-specific analyses underscore axatilimab’s capacity to both diminish fibrotic skin involvement and alleviate patient-reported symptom burden in refractory cGVHD.

Prospects in Systemic Therapies

Multiple clinical trials are under way to evaluate novel therapies in cGVHD (Table 5).

OPEN IN VIEWER

Table 5.

Ongoing Clinical Trials in Cutaneous cGVHD.

Trial name/ID	Intervention	Phase	Target population	Endpoints	Status
NCT06638879	Tacrolimus 0.1% oral ointment (toothpaste)	1	1 to 40 y old individuals oral cGVHD	Utilizing the Mucosal Ulceration Index, the patient will be evaluated for the presence of mucosal ulcerations.	Active
NCT05675930	Photobiomodulation therapy (Thor LX2.3) vs sham device	2	Patients ≥ 4 y with oral chronic GVHD post-allo-HCT	Treatment response rate at day 28, defined as the proportion of oral cGVHD patients achieving complete or partial response	Active

Abbreviations: cGVHD, chronic graft-versus-host disease; HCT, hematopoietic cell transplantation.