Graft-versus-host disease (GVHD) — cutaneous manifestations

Meet the patient

A 42-year-old man is on day 28 after a matched-unrelated-donor stem cell transplant for leukaemia. He arrives with a rising rash that began on his palms and earlobes, now spreading up his trunk, accompanied by cramping diarrhoea and a bilirubin that has crept from 20 to 60. The team that cured his leukaemia now has a new enemy, and it is the donor's immune system.[1][4]

The two questions that decide his next week are the two that decide every GVHD case: is this acute or chronic, and how severe? (timing plus stage set the whole ladder) and will steroids be enough, or do I need the second-line drug ready? Roughly half do not respond to steroids — so the refractory pathway is not the exception, it is the plan.[3]

The graft turns on the host — the Billingham three

GVHD can only happen when three conditions line up at once. Memorise them, because the third is the one examiners probe — it is the difference between GVHD and plain transplant rejection.[1]

1. The graft carries immunocompetent cells — donor T-lymphocytes able to mount a response. (Solid organs come without a T-cell army; stem cell grafts arrive with one.)
2. The host cannot reject them — the recipient is immunocompromised, the conditioning having cleared the native immune system. This is the inverse of solid-organ rejection, where the host attacks the graft.
3. There is histoincompatibility — HLA mismatch, or disparity at minor histocompatibility antigens. This is why even HLA-identical sibling transplants still get GVHD: minor antigens like HA-1 and HY differ.[5]

The clinician's confession: the conditioning regimen does not just prepare space for the graft — it damages host tissues, releasing danger signals (DAMPs, gut-derived LPS) that prime the very immune attack you are about to fight. The harder the conditioning, the worse the GVHD tends to be. The treatment is also the trigger.[5]

The three-phase cascade of acute GVHD is the mechanism every stem must hang on:[5]

[5]

Why skin, gut, and liver? They share rapidly dividing epithelia preferentially injured by conditioning, they express the minor histocompatibility antigens donor T-cells target, and the gut — once breached — leaks LPS that amplifies the whole fire. One broken barrier feeds the next.[1]

The graft-versus-leukaemia (GVL) balance is the catch examiners love: the same donor T-cells that cause GVHD also clear residual leukaemia. Aggressive T-cell depletion stamps out GVHD and lets the leukaemia back — so the aim is control, never total abrogation. Regulatory T-cells (Tregs) hold the leash; conditioning depletes them, and their slow reconstitution tracks with disease control.[1]

Acute GVHD — read the rash where it starts

Acute cutaneous GVHD is a graded maculopapular rash that begins acrally — palms, soles, ears — and the acral onset is the single sign that separates it from a drug eruption. Drug rashes are truncal and symmetrical; GVHD creeps in from the edges. That distinction is worth a viva mark every time.[3][4]

The skin stage is the one every candidate reproduces — it tracks the rash from trivial to lethal:[1]

[1]

The classic trap: a post-transplant patient on ten drugs gets a truncal maculopapular rash and is labelled "drug eruption." If it started on the palms and soles, and the diarrhoea and bilirubin are climbing too, that triad is acute GVHD until proven otherwise — biopsy it, do not just stop an antibiotic.[3][4]

The companion organs define the syndrome: gut involvement is anorexia, nausea, watery or bloody diarrhoea, abdominal pain, and ileus (endoscopy with biopsy confirms and excludes CMV and C. difficile); liver involvement is cholestatic, with rising bilirubin and alkaline phosphatase — distinct from sinusoidal obstruction syndrome, which is painful hepatomegaly and weight gain. Onset is typically 2–6 weeks post-transplant; a rash before day 14 is hyperacute GVHD — fulminant and treatment-refractory, with a grim prognosis.[3]

Chronic GVHD — when the attack turns to fibrosis

Chronic GVHD is a different disease, not a late acute GVHD. The cytokine milieu flips from Th1 cytotoxicity to a Th2/B-cell/TGF-beta fibrotic process — autoimmunity tangled with alloimmunity — and the skin changes from an inflamed rash to lichenoid and sclerodermoid patterns. Roughly 30–50 percent of long-term survivors develop it, and it is the leading cause of late non-relapse mortality.[1][6]

The cutaneous patterns fall into two faces, and the sclerodermoid face is the one that cripples:[1][6]

• Lichenoid — violaceous, polygonal, flat-topped papules indistinguishable from lichen planus on the wrists, forearms, trunk, and oral mucosa (Wickham-striae-like white reticulate patches). May leave hyperpigmentation and atrophy.
• Sclerodermoid — indurated, bound-down, fibrotic plaques like morphea or systemic sclerosis, localised or generalised. Deep fasciitis produces restricted joint mobility and contractures, peau d'orange skin, and a "groove sign" where superficial veins are buried.
• Poikiloderma, nail dystrophy (pterygium, ridging, anonychia), and scarring alopecia complete the cutaneous picture.[6]

The sicca flag everyone forgets: dry eyes and dry mouth from lymphocytic destruction of the lacrimal and salivary glands is often the earliest and most disabling chronic GVHD symptom, and it is the one patients suffer in silence. Ask about it; sicca plus oral lichenoid lesions plus genital erosions or vaginal stenosis is the hidden morbidity of this disease.[1]

The systemic reach is what makes chronic GVHD a multisystem disease, not a skin disease: bronchiolitis obliterans (the lethal lung complication), myasthenia gravis, autoimmune cytopenias, polymyositis, peripheral neuropathy, and oesophageal webs. And the dark late sequel — squamous cell carcinoma of the skin and oral mucosa, at several-fold increased risk, often multiple and aggressive, demanding lifelong surveillance.[1][6]

Histopathology — the two signs that close the diagnosis

Skin biopsy is not always needed when the classic triad follows transplant, but two signs confirm it when the picture is ambiguous. In acute GVHD: basal vacuolar degeneration with apoptotic keratinocytes (dyskeratotic cells, Civatte bodies) and lymphocytic satellitosis — lymphocytes sitting beside dying keratinocytes. Severe disease shows full-thickness epidermal necrosis with subepidermal blistering, the TEN-like picture.[7]

Chronic GVHD mirrors its two faces histologically: a lichenoid band-like lymphocytic infiltrate obscuring the dermo-epidermal junction, or a sclerodermoid pattern of dermal sclerosis with loss of adnexal structures and thickened collagen bundles — indistinguishable from morphea on a blind slide. Deep biopsies of fasciitis show an eosinophilic-fasciitis-like infiltrate.[7]

Grading — the skin stage is only one column

Overall acute GVHD grade combines skin, liver, and gut, and it is the grade — not the rash alone — that predicts survival. The Glucksberg/IBMTR system is the one to reproduce:[1]

[1]

Grade IV — any organ at stage 4 — is life-threatening; the skin alone rarely kills, but skin stage 4 with gut or liver failure does. High stool volume predicts poor prognosis independently, which is why severe diarrhoea escalates urgency.[3]

Chronic GVHD uses the NIH 2025 consensus severity score across eight organs (skin, mouth, eyes, GI, liver, lungs, joints/fascia, genital tract), classifying disease as mild (at most 2 organs, none over score 1, no lung), moderate (at least 3 organs, or any single organ at score 2, or lung score 1), or severe (any organ at score 3, or lung score 2–3). The NIH moved the field past the 100-day clock — chronic GVHD is defined by features, not just timing.[9]

Acute versus chronic at a glance

The split that earns the marks is the cytokine flip — Th1 cytotoxicity becomes Th2/B-cell fibrosis — and the skin pattern that follows.[1]

[1]

The discriminator line: a rash on the palms within six weeks is acute cytotoxic GVHD; lichenoid papules, bound-down skin, and a dry mouth after three months is chronic fibrotic GVHD. The cytokine milieu flips, and so does the morphology.[1]

Management — steroids first, the refractory pathway ready

Every patient starts with systemic corticosteroids, and roughly half will need more — so plan the second line before you need it. For acute GVHD, methylprednisolone 1–2 mg/kg/day IV (or prednisolone equivalent) is first-line; about 50 percent respond. Steroid-refractory is defined as progression after 5–7 days or no improvement after 14 days — at which point ruxolitinib (JAK1/2), FDA-approved on the REACH-2 trial, is the preferred second agent, with extracorporeal photopheresis, anti-TNF (infliximab), and mesenchymal stromal cells as alternatives.[3]

[1]

For chronic GVHD, first-line is prednisolone 1 mg/kg/day plus a calcineurin inhibitor (tacrolimus or cyclosporine), continued for at least 4–8 weeks before declaring steroid-refractory. Because 50–60 percent of chronic patients become refractory or dependent, three targeted drugs transformed this space — and a fellowship candidate must name each one's class, dose, and trial.[2][8][9]

The three drugs that carry steroid-refractory chronic GVHD:[2]

[10] [11] [12]

Beyond the three targeted drugs: rituximab (anti-CD20, 375 mg/m² weekly for 4 doses) for autoantibody-driven disease (cytopenias, myasthenia, sclerodermoid, oral lichenoid); extracorporeal photopheresis (ECP) — apheresis with 8-methoxypsoralen plus UVA on 2 consecutive days every 1–2 weeks, steroid-sparing and infection-light, excellent for sclerodermoid and refractory acute cutaneous disease; mycophenolate 1–1.5 g twice daily; and low-dose methotrexate for joint and fascia disease.[2][8]

Organ-specific therapy — systemic alone never controls chronic GVHD

The recurring junior error is to escalate only the systemic drug and leave the sicca, the mouth ulcers, and the lungs untreated. Organ-specific topical, intralesional, and inhaled therapy runs in parallel with the systemic regimen from the start — the NIH 2025 consensus and HSCT curriculum both insist on it.[2][9]

[1]

Bronchiolitis obliterans deserves its own warning. It is the leading non-cutaneous cause of chronic GVHD death — progressive dyspnoea with an obstructive, irreversibly declining FEV1 — and by the time it is symptomatic the lung is already scarred. The FAM regimen slows but rarely reverses it; recognise and treat early, not late.[6]

[1]

Prophylaxis — the regimen that sets the risk

Prophylaxis is decided before the transplant and is the single biggest determinant of whether GVHD happens at all. The standard is a calcineurin inhibitor (tacrolimus or cyclosporine) plus short-course methotrexate. Post-transplant cyclophosphamide (PTCy) is now favoured for haploidentical transplants — it depletes alloreactive T-cells while sparing regulatory T-cells. Ex vivo T-cell depletion (CD34+ selection or alemtuzumab) cuts GVHD but raises relapse and infection risk, an honest trade-off.[1]

Differential diagnosis — the post-transplant rash trap

Three mimics account for most mislabels; the discriminator is distribution plus the accompanying organs.[1]

• Drug eruption — morbilliform, truncal and symmetrical, from antibiotics or conditioning drugs; temporal relationship to a new drug. Acral onset favours GVHD.
• Viral exanthem — CMV, adenovirus, HHV-6 reactivation post-transplant; often with the virus detectable in blood.
• Morphea or systemic sclerosis — indistinguishable from sclerodermoid chronic GVHD histologically; the transplant history settles it.[1]

The classic trap: calling Stage IV acute GVHD "TEN from a drug" in a recent transplant. The timing, the acral onset, the rising bilirubin and diarrhoea, and satellitosis on biopsy separate true GVHD from a drug reaction — and the management overlaps (steroids, supportive care) but the prognosis and escalation pathway differ.[4]

How GVHD patients come to harm — the preventable list

• Stage III–IV acute GVHD read as "a bad drug rash", delaying steroids in a TEN-like dermatological emergency.[3]
• A chronic sinus or non-healing oral ulcer left unbiopsied in chronic GVHD — squamous cell carcinoma, often metastatic at diagnosis.[6]
• Sicca syndrome never asked about — corneal scarring and a destroyed mouth from untreated ocular and oral disease.[1]
• Fasciitis and joint contractures left without physiotherapy and belumosudil — permanent immobility that was preventable.[6]
• Progressive dyspnoea attributed to "deconditioning" when it is bronchiolitis obliterans — the FAM regimen started too late.[6]
• A non-irradiated blood product given to an immunocompromised host — transfusion-associated GVHD, fatal in over 90 percent.[1]
• Steroid-refractory acute GVHD kept on escalating steroids instead of escalating to ruxolitinib (REACH-2).[13]

Outcomes and prognosis

Severity at grade and steroid responsiveness set survival. Grade I acute GVHD barely dents it; Grade II keeps roughly 70–80 percent five-year survival; Grade III–IV carries 60–80 percent non-relapse mortality. Steroid-refractory acute GVHD historically meant under 20 percent one-year survival, modestly rescued by ruxolitinib (REACH-2).[3][13]

By NIH 2025 severity, chronic GVHD five-year overall survival runs over 90 percent for mild, 60–80 percent for moderate, and under 50 percent for severe disease. Bronchiolitis obliterans and steroid-refractory status are the principal killers; thrombocytopenia at chronic GVHD onset is an independent poor prognostic marker.[1][9]

Special situations — the ones outside the standard timeline

Hyperacute GVHD (onset before day 14) is fulminant and treatment-refractory with a grim prognosis; consider anti-thymocyte globulin. Engraftment syndrome — fever, rash, weight gain, pulmonary oedema at neutrophil recovery (day 7–14) — mimics acute GVHD but responds to brief corticosteroids. Overlap syndrome carries simultaneous acute and chronic features per NIH criteria.[1]

Transfusion-associated GVHD (TA-GVHD) is the one to fear outside transplant: an immunocompetent host transfused with cellular blood from an HLA-homozygous donor (often a relative) develops donor T-cell attack on marrow and skin — mortality over 90 percent. The prevention is absolute: irradiated blood products for every immunocompromised patient and for all directed family donations. Maternal-fetal GVHD occurs in SCID/Omenn syndrome from engrafted maternal T-cells.[1]

The mantra, and the memory device

[1]

The mantra: steroids first, the refractory pathway ready — ruxolitinib for acute, ruxolitinib-belumosudil-ibrutinib for chronic — and never leave the sicca, the mouth, or the lungs untreated.[1][3]

Ward-round test — three stems, thirty seconds each

Stem 1 — the day-28 transplant rash (answer)

A man on day 28 after a matched-unrelated-donor transplant has a rash that began on his palms and earlobes, now on his trunk, with diarrhoea and a bilirubin of 60. What is it, and what is the first step? Model: This is acute GVHD, Stage II–III — the acral onset plus diarrhoea and cholestasis after transplant is the classic triad; acral onset distinguishes it from a drug eruption. Confirm with skin biopsy (basal vacuolar change, apoptotic keratinocytes, lymphocytic satellitosis) if there is doubt, exclude CMV and C. difficile in the stool, and start methylprednisolone 1–2 mg/kg/day IV. Grade overall severity across skin, liver, and gut. Have ruxolitinib (REACH-2) ready, because about half are steroid-refractory.[3][4]

Stem 2 — the changing oral ulcer (answer)

A patient with five years of chronic lichenoid GVHD has an oral leukoplakic plaque that has ulcerated and will not heal over two months. Next step? Model: Urgent biopsy to exclude squamous cell carcinoma — chronic GVHD skin and mucosa carry a several-fold increased SCC risk, often multiple and aggressive with early metastasis. Do not keep dressing it or attributing it to a flare. Stage with imaging, involve head-and-neck surgery and oncology, and tighten photoprotection and surveillance. Annual full-skin and oral examination is mandatory for every chronic GVHD survivor.[6]

Stem 3 — sclerodermoid chronic GVHD failing steroids (answer)

A patient with sclerodermoid chronic GVHD and progressive joint contractures is still worsening on prednisolone 1 mg/kg/day plus tacrolimus after eight weeks. What now, and why this drug? Model: Declare steroid-refractory chronic GVHD and escalate. For sclerodermoid disease with fasciitis where fibrosis dominates, belumosudil 200 mg once daily is the logical first targeted choice — ROCK2 inhibition rebalances Th17/Treg and downregulates TGF-beta-driven fibrosis (ROCKstar, ~74 percent ORR, steroid taper in two-thirds). Ruxolitinib 10 mg BD and ECP are alternatives; add urgent physiotherapy to prevent permanent contracture. Never leave the fasciitis untreated while you escalate only the systemic drug.[2][12]

References

1. [1]Baumrin E, Loren AW, Falk SJ, et al. Chronic graft-versus-host disease. Part I: Epidemiology, pathogenesis, and clinical manifestations J Am Acad Dermatol, 2024.PMID 36572065
2. [2]Baumrin E, Loren AW, Falk SJ, et al. Chronic graft-versus-host disease. Part II: Disease activity grading and therapeutic management J Am Acad Dermatol, 2024.PMID 36572064
3. [3]Patel DA, Crain M, Pusic I, et al. Acute Graft-versus-Host Disease: An Update on New Treatment Options Drugs, 2023.PMID 37247105
4. [4]Hong J, Fraebel J, Yang Y, et al. Understanding and treatment of cutaneous graft-versus-host-disease Bone Marrow Transplant, 2023.PMID 37730800
5. [5]Zeiser R, Blazar BR. Pathophysiology of Chronic Graft-versus-Host Disease and Therapeutic Targets N Engl J Med, 2017.PMID 29281578
6. [6]Shi CR, Ferreira AL, Kaur M, et al. Cutaneous Chronic Graft-Versus-Host Disease: Clinical Manifestations, Diagnosis, Management, and Supportive Care Transplant Cell Ther, 2024.PMID 39370234
7. [7]Aractingi S, Chosidow O Cutaneous graft-versus-host disease Arch Dermatol, 1998.PMID 9606330
8. [8]Malard F, Mohty M Updates in chronic graft-versus-host disease management Am J Hematol, 2023.PMID 37483142
9. [9]Lee SJ, Williams KM, Sarantopoulos S, et al. NIH Chronic Graft-Versus-Host Disease Consensus Conference 2025 Update Transplant Cell Ther, 2025.PMID 40409691
10. [10]Zeiser R, Polverelli N, Ram R, et al. Ruxolitinib for Glucocorticoid-Refractory Chronic Graft-versus-Host Disease N Engl J Med, 2021.PMID 34260836
11. [11]Miklos D, Cutler CS, Arora M, et al. Ibrutinib for chronic graft-versus-host disease after failure of prior therapy Blood, 2017.PMID 28924018
12. [12]Cutler C, Lee SJ, Arai S, et al. Belumosudil for chronic graft-versus-host disease after 2 or more prior lines of therapy: the ROCKstar Study Blood, 2021.PMID 34265047
13. [13]Zeiser R, von Bubnoff N, Butler J, et al. Ruxolitinib for Glucocorticoid-Refractory Acute Graft-versus-Host Disease N Engl J Med, 2020.PMID 32320566