Category
Symptoms after
DLI or HSCT
Acute features
Chronic features
Classic acute
< 100 days
Yes
No
Persistent, recurrent and late-onset
> 100 days
Yes
No
Overlap
> 100 days
Yes
Yes
In aGVHD, skin is the most commonly affected organ, followed by the gastrointestinal tract and lastly the liver [10]. Early harbingers of cutaneous disease include pruritus and erythema of the ears, face, palms, and soles [8, 9]. Often, subsequent cutaneous finings include folliculocentric blanching macular erythema, which ultimately progresses to a diffuse and symmetric morbilliform eruption (Fig. 6.1). Skin tenderness is not uncommon and often an ominous sign of severe cutaneous disease. In severe cutaneous aGVHD, the eruption progresses to erythroderma followed by bullae formation with epidermal detachment, mimicking toxic epidermal necrolysis (TEN) (Fig. 6.2). In addition to the skin, the mucous membranes, particularly the conjunctivae and oral mucosa, can become markedly inflamed (Fig. 6.3). It should be noted that, in contrast to the protean presentations of cutaneous chronic GVHD (see Chap. 10), cutaneous aGVHD invariably presents as aforementioned.
Fig. 6.1
Manifestations of stage II acute cutaneous GVHD. A morbilliform and folliculocentric eruption on the trunk (a), palmar involvement (b), folliculocentric involvement of the lower leg (c) and foot (d) (Courtesy of Dr. Ryan Trowbridge)
Fig. 6.2
Manifestations of stage IV acute cutaneous GVHD. Large bullae (a) and diffuse (b) purple-dusky hue, tenderness and sloughing (Courtesy of Dr. Jennifer Tan)
Fig. 6.3
Oral acute GVHD presenting with diffuse ulceration of the ventrolateral tongue, as well as ulceration and crusting of the lips (Reprinted from Ion et al. [11]; with permission)
Differential Diagnosis
During complete evaluation of aGHVD, the alternative diagnoses to consider include: engraftment syndrome, toxic erythema of chemotherapy, viral exanthem, and drug hypersensitivity, the most common alternative diagnosis (Table 6.2) [8, 9, 12]. A diagnosis of aGVHD is supported by an eruption that involves acral sites and is folliculotropic, with concomitant cholestatic liver damage and gastrointestinal manifestations. However, this triad is not specific to the diagnosis of aGVHD.
Diagnosis | Clinical presentation | Onset | Treatment |
|---|---|---|---|
Drug hypersensitivity (Fig. 6.4) | Morbilliform, lack of peri-follicular accentuation | Anytime | Withdrawal of medication |
Viral exanthem | Morbilliform | Anytime | Supportive |
Toxic erythema of chemotherapy (Fig. 6.5) | Variable, often acral | 2–3 weeks after chemotherapy | Supportive |
Engraftment syndrome (Fig. 6.6) | Morbilliform, may begin in acral sites. Concurrent cytokine storm with capillary leak syndrome leading to anasarca and respiratory distress | ~14 days after transplantation | Systemic corticosteroids leads to complete resolution |
Drug Hypersensitivity
A recent retrospective review concluded that drug hypersensitivity could be excluded with certainty in post-HSCT patients who develop a morbilliform eruption if concurrent hyperbilirubinemia and diarrhea are present [12]. The distribution and onset of the cutaneous eruption in drug hypersensitivity often spares acral sites, particularly the palms and soles, and lacks the folliculocentric erythema of aGVHD eruptions (Fig. 6.4) [12].
Fig. 6.4
Drug hypersensitivity reaction to vancomycin presenting as a diffuse morbilliform eruption on the trunk early after HSCT. Note the lack of follicular prominence, a typical feature of acute GVHD
Toxic Erythema of Chemotherapy
This diagnosis is suggested by blanching erythema that is painful and limited to palms and soles (most common) or flexural sites (Fig. 6.5). The pathophysiology is hypothesized to be a result of toxic metabolite accumulation in eccrine-rich sites [13].
Fig. 6.5
Toxic erythema of chemotherapy due to cytarabine presenting as confluent erythema and edema of the palms
Engraftment Syndrome
Engraftment syndrome (ES) may be favored in patients who received an autologous transplant and develop symptoms reminiscent of capillary leak syndrome with cytokine storm [14, 15]. ES has also been described in allogeneic HSCT [16]. Clinical findings include non-infectious fever (persistently negative culture data and no response to empiric antibiotics), hypoxemia, diarrhea, weight gain, anasarca, and a morbilliform cutaneous eruption (Fig. 6.6). Timing of onset is, by definition, in the peri-engrafment period. ES is remarkably responsive to systemic corticosteroids, and in a recent retrospective review, 41 of 42 patients with ES showed rapid symptoms resolution leading to hospital discharge in a median of 10 days [15]. The same review found excellent correlation between markedly elevated C-reactive protein (CRP) and ES diagnosis (Table 6.2) [15].
Fig. 6.6
Engraftment syndrome. Diffuse morbilliform eruption mimicking acute GVHD early after allogeneic HSCT
Many studies have investigated the utility of a diagnostic skin biopsy for aGVHD. The histological findings are often non-specific and do not reliably improve diagnostic distinction between aGVHD and drug hypersensitivity, particularly in early GVHD [17–19]. The results of a decision analysis showed that a skin biopsy should be performed only in cases in which the pretest probability for aGVHD is exceptionally low [20, 21].
Grading and Prognostication
The most commonly utilized aGVHD grading system is the modified Seattle Glucksberg system (Tables 6.3 and 6.4), which is an evolution from the original 1974 Glucksberg study [5, 7]. The key variables that determine staging of aGVHD include: amount of body surface area of skin involvement, extent of elevation in serum bilirubin for hepatic involvement, and stool output volume for gastrointestinal involvement (Table 6.3). The integration of stages determines the grade, which is closely correlated with mortality (Table 6.4) [22].
Stage | Skin (% BSA) | Liver (bilirubin) | Gastrointestinal | Non-relapse mortality at 100 days [21] |
|---|---|---|---|---|
1 | Maculopapular rash < 25 % | 2.0–2.9 mg/dL | Diarrhea 0.5–1 L/day or nausea/emesis with positive gut biopsy | 27 % |
2 | Maculopapular rash 25–50 % | 3.0–5.9 mg/dL | Diarrhea 1–1.5 L/day | 43 % |
3 | Maculopapular rash >50 % | 6.0–14.9 mg/dL | Diarrhea > 1.5 L/day | 68 % |
4 | Generalized erythema (i.e. erythroderma) with desquamation or bullae | >14.9 mg/dL | Severe abdominal pain with or without ileus | 92 % |
Grade | Skin stage | Liver stage | Gastrointestinal stage |
|---|---|---|---|
I | 1–2 | 0 | 0 |
II | 3 or | 1 or | 1 |
III | 0–3 and | 2–3 or | 2–4 |
IV | 4 or | 4 or | 0–4 with Grade 4 skin or liver |
A recently developed and validated Ann Arbor aGHVD severity score utilizes biomarkers alone and showed excellent concordance with response to therapy at 28 days as well as non-relapse mortality at 6 months [23]. The three biomarker (TNFR1, ST2, and REG3a) plasma concentrations are combined to create a single score (Table 6.5), which subsequently stratifies the patient as Ann Arbor score 1 (likelihood of complete or partial response to treatment, 81 % at day 28) to score 3 (likelihood of complete or partial response to treatment, 46 % at day 28) [23].
Table 6.5
Ann Arbor Risk Score for aGVHD based on plasma concentrations of three biomarkers: TNFR1, REG3α, and ST2
Ann Arbor Score | CR/PR at 28 days | Non-relapse mortality at 12 months, % (95 % CI) |
|---|---|---|
1 | 81 % | 8 % (3–16) |
2 | 68 % | 27 % (20–24) |
3 | 46 % | 46 % (33–58) |
Interestingly, the Ann Arbor severity score more accurately predicted non-relapse mortality when compared with the Seattle Glucksberg grading system [23]. Although the Ann Arbor score is not a practical tool at the bedside, it will be essential for patient stratification and surrogate endpoint outcome analysis in future interventional studies.
Treatment of aGVHD
The incidence of aGVHD after allogeneic HSCT is between 10 and 80 %, and ongoing research in HLA matching, conditioning and prophylactic regimens holds promise to decrease that rate [4, 8, 9, 24]. In newly diagnosed aGVHD, early recognition, intervention, and close monitoring are the cornerstones of effective management. The most recent 2012 guidelines from Haemato-oncology Task Force of the British Committee for Standards in Haematology (BCSH) and the British Society for Blood and Marrow Transplantation (BSBMT) and the American Society of Blood and Marrow Transplantation (ASBMT) emphasize that therapeutic choice should be in accordance with proper staging and grading, in conjunction with the rate of disease progression [8, 24]. Treatment strategies discussed here are primarily based on cutaneous aGVHD staging, although it is imperative to utilize the grade of aGVHD as defined earlier.
Brief Note on the Prevention of aGVHD
A noteworthy area of research revolves around the use of mammalian target of rapamycin (mTOR) inhibitors instead of calcineurin inhibitors after allogeneic HSCT for GVHD prophylaxis. In 2002, Hoffmann et al. [25] showed that transfer of donor CD4+ CD25+ regulatory T Cells (Tregs) after allogeneic HSCT can prevent severe aGVHD while maintaining graft-versus-leukemia effect in mice. More recent work by Satake et al. [26] showed (in mice) that the mTOR inhibitor sirolimus in combination with low dose IL-2 showed Treg expansion and decreased risk of aGVHD. Peccatori et al. [27] showed that use of sirolimus prophylaxis led to a much lower than expected incidence of GVHD in patients receiving high GVHD risk transplants (peripheral blood stem cell grafts from partially HLA-matched family donors). Prophylactic use of mTOR inhibitors in combination of Tregs modulation is a very active area of research with multiple ongoing trials (NCT02528877, NCT01795573, NCT02118311, NCT00105001, NCT01251575, NCT01903473, NCT00406393, NCT00993343, NCT00602693).
Treatment of Grade I aGVHD
General consensus guidelines from the BCSH, BSBMT, and ASBMT recommend management of grade I aGVHD (<50 % body surface area involvement) with topical therapies in addition to optimization of prophylactic immunosuppression [8, 9, 28]. The most common approach to grade I aGVHD includes topical corticosteroids and/or topical calcineurin inhibitors (CNIs). However, for cases in which 25–50 % of body surface area is involved, progression must be monitored closely and early initiation of first-line systemic therapy should be considered. Additionally, patients who have a clinical grade of I, although an Ann Arbor risk score of 3, are at a significantly increased risk for aGVHD to involve more organs and should be considered for systemic therapy [22]. It is foreseeable that the Ann Arbor score will play a larger role in early grade aGVHD management, particularly for cases with clinical and biomarker discordance.
High-potency topical corticosteroids (class 1 or 2), the cornerstone for therapy on the extremities and trunk, can be considered for short periods of time on the face, major body folds, and genital region with close monitoring. Topical CNIs, such as tacrolimus or pimecrolimus, provide benefits equivalent to class 4 or class 5 topical corticosteroids and they should be employed in tandem, especially to the face and genital region (Table 6.6) [29].
In addition to topical corticosteroids and CNIs, clinicians should stress skin hygiene and sun-safe practices. Excellent skin hygiene (i.e. use of emollients at least once per day) will optimize skin barrier function and minimize transepidermal water loss; recommendations include: minimize the duration of showers, minimize the use of soaps or other skin cleansers, favor skin care products that are unscented, and use thick emollients immediately after showers plus at least one other time per day. Regular use of sun protection factor creams and appropriate clothing, in addition to avoidance of high-intensity sun exposure will minimize the potential for additional keratinocyte damage.
Table 6.6
Therapeutic options in aGVHD and key studies or reviews
