News of the Week

Highlights:

• In 2019, the American Society for Transplantation and Cellular Therapy (ASTCT) consensus panel developed definitions and grading criteria for the established immune effector cell (IEC)- associated toxicities—specifically cytokine release syndrome (CRS), IEC-associated neurotoxicity syndrome (ICANS), and IEC-associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS).
• These updates provide a framework to harmonize safety reporting, improve risk stratification, and compare outcomes across clinical trials, real-world registries, and routine clinical care.

While IEC therapies have transformed the treatment of relapsed and refractory malignancies, they induce unique, complex, and potentially life-threatening adverse events. Due to the emergence of novel and delayed toxicities and the limitations of traditional toxicity grading systems for capturing the unique biology, timing, or severity of IEC toxicities, the ASTCT convened a multidisciplinary panel of 36 international experts in cellular therapy, hematology, oncology, neurology, and pediatrics to update existing guidelines and establish standardized consensus definitions. Standardizing these criteria allows clinicians and researchers to compare safety profiles across products accurately, clarify toxicity mechanisms, and link standardized grades directly to clinical treatment protocols. Through a structured consensus meeting in Washington, D.C., and iterative writing/review stages, the authors established unified definitions and 4-tier severity grading scales across 11 distinct IEC toxicity categories. The key recommendations by category are as follows:

1. Systemic & Hyperinflammatory Toxicities:

• Cytokine Release Syndrome (CRS): Maintained the 2019 Consensus Graded based on fever (≥ 38 °C), with either hypotension (graded by vasopressor requirements), and/or hypoxia (graded by oxygen delivery). Organ toxicities and inflammatory biomarkers (C-reactive protein, ferritin) track alongside CRS but do not alter the CRS grade itself; they should be graded separately using organ-specific criteria (see Table 1 in the main manuscript).
• IEC-HLH-like Syndrome (IEC-HS): Requires a >5-fold elevation in ferritin (from baseline or upper limit of normal) alongside hepatic transaminitis, hypofibrinogenemia (<150 mg/dL), and severe cytopenias. IEC-HLH-like Syndrome consistently develops in patients with antecedent or ongoing/resolving CRS, distinguishing it from acute CRS-associated multiorgan failure (see Tables 3 and 4).

2. Hematological Toxicities (ICAHT)

• Neutropenia (N-ICAHT): Evaluates the depth and cumulative duration of absolute neutrophil count suppression (early: days 0–30; late: > day 30, may occur after early count recovery) to directly stratify infection risk and non-relapse mortality (see Table 2).
• Thrombocytopenia (T-ICAHT): Introduces a platelet grading scale (evaluating depth, duration, and thresholds (≤ 20 k/µL) to track transfusion dependency, bleeding risk, and overall survival.
• Early post-infusion cytopenias (within the first 30 days) should not automatically be classified as dose-limiting toxicities in clinical trials.

3. Neurological Toxicities

• Immune Effector Cell-Associated Neurotoxicity Syndrome (ICANS): In adults and older children aged 9 years or more, the guideline retains the 10-point ICE score (Orientation, Naming, Following commands, Writing, Attention). In younger children and non-verbal baselines, the guideline replaces the labor-intensive Cornell Assesment of Pediatric Delirium (CAPD) tool with the Modified ICE (M-ICE) score, tailored to pediatric cognitive stages. Bilateral cranial nerve VI palsy was reclassified under stand-alone cranial nerve toxicities (see Tables 5, 6, and 7).
• Non-ICANS Neurotoxicities (NINTs):
  1. IEC-Cranial Nerve Palsy (IEC-CNP): Most frequently presents as facial nerve (CN VII) weakness, median 4–5 weeks post-infusion; graded using consolidated Common Terminology Criteria for Adverse Events (CTCAE) scales (see Table 8).
  2. IEC-Polyneuropathies: Ascending motor weakness, gait changes, and radiculoneuritis; graded via CTCAE peripheral motor/sensory scales (see Table 9).
  3. IEC-Parkinsonism: Subacute movement disorder characterized by bradykinesia, gait instability, cognitive slowness, and apathy/indifference. Grading determined by the worst symptom in any domain (motor, cognitive, or personality)(see Table 10). Typically DaTscan-negative and levodopa-refractory.
  4. Tumor Inflammation-Associated Neurotoxicity (TIAN): Transient local inflammation and swelling localized to CNS tumor or CNS lymphoma sites (see Table 11).
  5. IEC-Cerebellar Disorder: Ataxia, Gait disturbance and dysarthria, reported with specific targets (e.g., GPRC5D); graded via CTCAE ataxia/dysarthria scales (see Table 12).

4. Gastrointestinal, Autoimmune & Off-Target Toxicities

• IEC-Enterocolitis (IEC-EC): Delayed-onset, non-bloody, grade ≥3 diarrhea occurring weeks to months post-infusion (especially with BCMA CAR-T); graded using CTCAE Diarrhea/Enterocolitis criteria regardless of tissue T-cell clonality (see Table 13).
• Local IEC Toxicity Syndrome (LICATS): Organ-specific, subacute inflammation (renal dysfunction, proteinuria, skin rashes, musculoskeletal symptoms) observed uniquely in treatment of autoimmune disease, distinct from systemic CRS.
• On-Target, Off-Tumor Toxicities: Adverse events caused by target expression on normal non-malignant tissues (e.g., B-cell aplasia with CD19, plasma cell aplasia with BCMA, dysgeusia/nail changes, skin rash with GPRC5D); recorded using CTCAE terms with documented target attribution (see Table 14).

Highlights:

• Allogeneic hematopoietic stem cell transplant (allo-HCT) is the primary consolidation strategy in first complete remission (CR1), as well as in CR2 for eligible patients. Autologous HCT (auto-HCT) is conditionally suggested only for unfit patients without bone marrow involvement.
• Myeloablative conditioning (MAC) with total body irradiation (TBI) is recommended for younger/fit patients and reduced-intensity conditioning (RIC) for older/frail patients. In contrast, universal post-transplant intrathecal chemotherapy is recommended for all patients, regardless of prior central nervous system (CNS) involvement.

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare aggressive cancer arising from plasmacytoid dendritic cell precursors. While novel targeted therapies (such as tagraxofusp, pivekimab sunirine, and venetoclax combinations) and conventional chemotherapies achieve high initial response rates, durable long-term disease control is generally not achievable without consolidative allo-HCT. Although allo-HCT is an established practice, no official societal guidelines previously existed to guide clinical decisions for BPDCN transplantation. An expert panel of the American Society for Transplantation and Cellular Therapy evaluated and voted on 17 distinct clinical recommendations using an online platform. Consensus for each recommendation required a majority threshold of >50%. The study applied the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) system to structure recommendations and evaluate evidence. Because no randomized controlled trials or prospective cohort studies exist for BPDCN transplantation, the manuscript’s underlying evidence relies entirely on retrospective cohort studies rated as having low to very low overall certainty. The main recommendations are:

1. Pre-transplant staging and confirmation of remission:

• Comprehensive multiorgan work-up: Before proceeding with transplantation, clinicians must confirm CR across all potential organ sites. The recommended evaluation includes a bone marrow aspirate and biopsy, flow cytometry, PET-CT scan, diagnostic lumbar puncture, and a thorough skin examination. (Strong recommendation, 100% agreement)

2. Transplant indications by disease status:

• First complete remission (CR1):
  • Allogeneic-HCT is strongly recommended as the primary consolidation strategy for all eligible patients. (Strong recommendation, 94% agreement)
  • Autologous HCT is suggested only for patients unfit for an allogeneic transplant, provided there is no bone marrow involvement with BPDCN. (Conditional recommendation, 76% agreement)
• Second complete remission (CR2):
  • Allogeneic HCT is strongly recommended for eligible patients in CR2. (Strong recommendation, 100% agreement)
  • Autologous HCT is conditionally suggested for unfit CR2 patients without bone marrow disease. (Conditional recommendation, 71% agreement)
• Primary induction failure or active relapsed/refractory disease:
  • Neither allo- nor auto-HCT is recommended due to disease resistance and clinical futility. Patients should instead be considered for clinical trials. (Strong recommendation)

3. Conditioning regimen selection (allo-HCT):

• Young and/or fit patients: The panel recommends a MAC regimen, preferably containing TBI, for patients in either CR1 or CR2. (Strong recommendation, 100% agreement)
• Older or frail patients: The panel recommends a RIC regimen for patients in either CR1 or CR2. (Strong recommendation, 100% agreement)

4. Post-transplant prophylaxis and maintenance:

• CNS prophylaxis: Post-transplant intrathecal chemotherapy is suggested for all patients, regardless of prior CNS involvement and /or residual CNS disease. (Strong recommendation, 65% agreement)
• Systemic maintenance / targeted therapy:
  • Universal routine maintenance chemotherapy or targeted therapy is not recommended for all post-allo-HCT or post-auto-HCT patients.
  • Systemic maintenance is conditionally suggested only for patients allografted in CR2 or beyond to help mitigate high relapse risk. (Conditional recommendation, 59% agreement)

5. Post-transplant disease surveillance:

• Surveillance schedule: Formal multiorgan disease monitoring (bone marrow evaluation, skin exam, and PET-CT as clinically indicated) is recommended between day +30 and +90, at 6 months, 1 year, 18 months, and at 2 years post-transplantation. (Strong recommendation, 94% agreement)

Highlights:

• Early-phase clinical trials evaluating minor histocompatibility antigens (MiHAs) HA-1 and HA-2–specific T cells with high-affinity, MiHA-specific T-cell receptors (TCR; TCR-T) demonstrate that genetically engineering donor T cells achieves durable antileukemic responses, robust in vivo expansion, and multi-year persistence in patients with high-risk or relapsed hematologic malignancies.
• Because these antigens are restricted exclusively to hematopoietic cells, the treatment selectively eliminates leukemic blasts while sparing non-hematopoietic host tissues and donor blood cells.

Relapse of hematologic malignancies remains the primary cause of failure following allogeneic hematopoietic cell transplantation (allo-HCT). While conventional donor lymphocyte infusions can elicit graft-versus-leukemia (GvL) effects, they carry a substantial risk of graft-versus-host disease (GvHD) and often fail to sustain long-term remission because of poor T-cell persistence. Adoptive TCR-T therapy targeting hematopoietic-restricted MiHAs provides a targeted approach to enhance GvL responses while completely sparing non-hematopoietic host tissues. MiHAs are polymorphic peptides derived predominantly from germline single nucleotide polymorphisms, deletions, or altered antigen processing presented on HLA class I or II complexes. Therapeutically favored MiHAs are expressed exclusively in recipient hematopoiesis. Because normal donor hematopoiesis lacks the recipient’s target variant or restricting HLA allele, donor blood cells are unaffected, avoiding GvHD and post-transplant marrow aplasia. Unlike somatic tumor neoantigens, germline-encoded MiHAs are broadly expressed across myeloid and lymphoid malignancies. Out of more than 170 validated HLA class I-restricted MiHAs, a standardized toolbox targeting just 5 to 10 prioritized MiHAs could cover approximately 60% of all HCT recipients. Current MiHA TCR-T manufacturing incorporates targeted phenotypic selection with depletion of CD45RA⁺ naive T cells to reduce alloreactivity, along with gene-editing strategies to optimize safety and in vivo performance. Transgenes are delivered via lentiviral/retroviral vectors or non-viral CRISPR-Cas9 knock-in platforms. Knockout of endogenous TCR α/β chains via CRISPR-Cas9 eliminates TCR mispairing and alloreactivity. The main clinical trials have evaluated HA-1 (HMHA1/ARHGAP45) and HA-2 (MYO1G), two targets presented by the HLA-A*02:01 allele. An ongoing Phase 1 trial evaluated an HA-1 TCR-T product manufactured from CD45RA⁺-depleted donor memory T cells, engineered with an iC9 safety switch, a CD34 marker, and a CD8 co-receptor. Administered across 15 infusions to 9 patients, the trial demonstrated an exceptional safety profile with no cases of cytokine release syndrome, neurotoxicity (ICANS), or dose-limiting toxicities. Four of the nine patients achieved or sustained complete remission (with the longest remission extending beyond 36 months). Concurrently, BlueSphere Bio is conducting a Phase 1/2a trial infusing its BSB-1001 HA-1 product on day 0 following myeloablative conditioning in high-risk TP53– or MECOM-mutated leukemia and myelodysplastic syndrome patients. The ALLOHA Phase 1 study also evaluated prophylactic infusions of donor-derived TCR-T cells TSC-100 (targeting HA-1) and TSC-101 (targeting HA-2) following reduced-intensity haploidentical HCT. Clinical development concentrated on TSC-101, which markedly reduced relapses to 16% (3 of 19 patients) compared to 44% (7 of 16) in transplant controls, conferring a 50% reduction in relapse risk and improved overall survival. All TSC recipients achieved early complete donor chimerism, and TSC-101 persisted past one year in seven subjects. Clinicians can also intentionally structure donor selection algorithms around MiHA and HLA disparities. For an HLA-A*02:01⁺ / HA-1⁺ recipient with a 10/10 matched donor, clinicians select a donor who is HA-1⁻ (rs1801284 G/G). In haploidentical or 9/10 mismatched HCT, selecting a donor lacking the restricting HLA-A2 allele altogether completely protects donor hematopoiesis from TCR-T attack. When choosing mismatched donors, avoiding the entire HLA-A2 serogroup (e.g., HLA-A*02:06) prevents unexpected off-target cross-reactivity. Infusion protocols require lymphodepleting chemotherapy (fludarabine/cyclophosphamide or clofarabine) to clear cellular space, increase IL-7/IL-15 availability, and debulk leukemia burden to maximize TCR-T expansion.

In conclusion, MiHA TCR-T cell therapy is a safe, effective, and genetically precise adjunct to allo-HCT that can improve GvL responses and improve relapse-free survival without increasing the risk of GvHD.

Highlights:

• In FLT3-ITD and NPM1–mutated acute myeloid leukemia (AML), pre-transplant measurable residual disease (MRD) detectability strongly correlates with relapse risk and can guide decisions on conditioning intensity or post-transplant targeted maintenance.
• In IDH1/2–mutated AML, commercially validated molecular MRD assays are lacking, and persistent pre-transplant mutations have suboptimal risk-stratification value.
• In TP53–mutated AML, pre-transplant clearance of the mutation has not been shown to predict improved overall survival (OS), so that a negative test result may be falsely reassuring.

A recent manuscript published in Blood addressed the clinical challenge of navigating the current complexities, assay limitations, and subtype-specific realities of MRD testing to deliver actionable, precision-based care for patients undergoing allogeneic hematopoietic cell transplantation (allo-HCT) in AML. Test modalities—such as multiparameter flow cytometry (MFC), quantitative PCR (qPCR), and next-generation sequencing (NGS)—vary significantly in sensitivity, limit of detection, and assay performance. Furthermore, patients entering transplant in morphologic complete remission (CR) do not share equal risk of relapse and death, even within equivalent baseline genomic risk groups. Clinicians face real-world dilemmas in translating MRD test results into actionable care. For example, delaying transplant to attempt “MRD erasure” is unproven and risks disease progression or toxicity. The critical points of each section are:

FLT3-ITD-Mutated AML

Approximately 50% of adult patients in morphologic (CR) harbor detectable FLT3-ITD MRD before transplant. Pre-transplant detection of persistent FLT3-ITD is the single variable most strongly associated with post-transplant relapse and OS in a dose-dependent manner. However, detectable FLT3-ITD MRD is not a contraindication to allo-HCT. In this context, post-transplant maintenance with the FLT3 inhibitor gilteritinib is strongly supported for patients with detectable pre- or early post-transplant MRD, demonstrating a statistically significant relapse-free survival benefit in this subgroup (hazard ratio 0.515). Patients who are undetectable for FLT3-ITD MRD before transplant derive little to no overall benefit from post-HCT gilteritinib maintenance. While bone marrow aspirate provides maximum sensitivity, peripheral blood testing offers high positive predictive value and a convenient non-invasive alternative for serial monitoring. Ultrasensitive FLT3-ITD testing is routinely performed starting at 3 months post-transplant, repeated every 3 months during year 1, and every 6 months during year 2. Maintenance can be discontinued if MRD remains undetectable at 2 years. Persistent mutations in genes such as DNMT3A, TET2, or ASXL1 during remission typically represent benign clonal hematopoiesis of indeterminate potential (CHIP) rather than active leukemia. Conversely, persistent FLT3-ITD reflects true residual leukemic burden.

NPM1-Mutated AML

NPM1-mutated AML (in the absence of co-occurring adverse lesions like FLT3-ITD) is generally considered favorable risk, with a substantial proportion of patients cured by chemotherapy alone. Molecular MRD testing after initial induction/consolidation chemotherapy is a critical tool for stratifying patients in first CR. Patients with persistent NPM1 transcripts/mutations after chemotherapy face a high risk of relapse, reversing the baseline favorable risk and identifying those who strictly require upfront consolidative allo-HCT. Molecular assays—specifically qPCR and ultrasensitive DNA-based NGS-MRD, limit of detection approx. 0.005% VAF)—are the standard of care because they are more sensitive than MFC. For fit patients aged <65 years with pre-transplant NPM1-MRD positivity, myeloablative conditioning is strongly favored over reduced-intensity conditioning. Guidelines recommend serial molecular monitoring (qPCR or ultrasensitive NGS-MRD) every 3 months for bone marrow (or every 4–6 weeks for blood) during years 1–2 post-transplant. Finally, clinical trials evaluating menin inhibitors as post-HCT maintenance or pre-emptive therapy for persistent NPM1 MRD are currently ongoing.

IDH1/2-Mutated and TP53-Mutated AML

Unlike FLT3-ITD or NPM1, no well-validated, standardized commercial molecular MRD tests are currently tailored specifically for IDH1 or IDH2 mutations. Detecting persistent IDH1 mutations in pre-transplant remission is common but does not significantly correlate with post-transplant relapse or OS. However, while pre-transplant persistence is largely uninformative, rising or re-emerging IDH1/2 mutation levels post-transplant strongly signal an increased risk of relapse and should prompt consideration of starting an IDH inhibitor. TP53-mutated AML carries a dismal prognosis and is rarely cured by chemotherapy alone, making allo-HCT the best available option for durable remission despite high relapse rates. Unlike other subtypes, pre-transplant clearance of TP53 mutations has not been shown to predict improved OS. Clinicians should not interpret a negative pre-HCT MRD result as indicating low relapse risk. Finally, because pre-transplant MRD status does not reliably stratify risk, consider proactive post-transplant strategies (e.g., hypomethylating agent maintenance, early immunosuppression taper, donor lymphocyte infusions, or clinical trial enrollment) regardless of pre-HCT MRD results.

Highlights:

• Post-transplant failure in myelofibrosis does not occur along a single pathway. It occurs through two largely uncorrelated dimensions: the disease-intrinsic axis, driven by genomic lesions and blood parameters, which specifically predicts relapse risk, and the host axis, driven by patient resilience and transplant logistics, which specifically predicts procedural non-relapse mortality (NRM).
• Recombining both dimensions into the iMTSS framework yields higher overall survival (OS) discrimination (concordance index [C-index] 0.640) than any established disease- or transplant-specific scoring tool.

Allogeneic stem cell transplantation (allo-HCT) remains the only curative therapy for myelofibrosis, but it carries significant early toxicity and procedural risk. NRM, driven by patient-related factors (such as performance status or portal vein thrombosis) and host-graft tolerance, is particularly prominent and accounts for a large proportion of early transplant failures compared with other myeloid malignancies. Traditional prognostic frameworks (such as DIPSS, MIPSS70+, MTSS, or CIBMTR scores) cannot tell clinicians why an individual patient is at risk or which targeted strategy (e.g., disease-directed pre/post-transplant therapy versus donor optimization and toxicity reduction) would best mitigate that risk. To evaluate this problem, the authors established a two-axis prognostic framework (iMTSS) that independently quantifies biology-driven relapse risk and patient-driven procedural risk to guide individualized clinical counseling.

Using a multicentric and international cohort of 1550 adult patients with primary or secondary myelofibrosis who underwent their first allo-HCT between 2000 and 2023, the authors evaluated OS and 2-year outcomes for relapse and NRM. The dataset was split 60/40: a development (training) set of 930 patients to construct the score and an independent validation set of 620 patients to test model discrimination and calibration. The combined iMTSS continuous score (integrating the disease axis and host axis) achieved a C-index of 0.640 (95% confidence interval: 0.616–0.662) for 5-year OS, outperforming the existing prognostic systems. In the independent validation set, iMTSS separated patients into three 5-year OS tiers: low (72%), intermediate (58%), and high (39%) (p<0.01). With respect to the disease axis (C-index 0.69), relapse was driven by aggressive lesions and cell counts: TP53 allelic state (subdistribution hazard ratio [sHR]: 1.33 per level), CALR/MPL genotype (sHR 0.54), RAS–pathway composite (sHR 1.62), and peripheral blasts (sHR 1.20 per standard deviation [SD] increase). The host axis-driven NRM (C-index 0.63), with the platelet count (sHR 0.72 per SD increase) as a surrogate for marrow fibrosis, portal hypertension, and delayed engraftment. Other related factors included portal vein thrombosis (sHR 2.48) and donor type mismatch (mismatched unrelated donor [MMUD], sHR 2.27, and haploidentical donor [sHR 1.96]), which carried heavy prognostic weight relative to a matched related donor. By mapping predicted 2-year relapse against predicted 2-year NRM, the study categorizes patients into four actionable management quadrants:

1. No dominant axis (66% of patients): Standard transplant protocols and monitoring.
2. Toxicity-driven / high NRM risk (15%): Elevated procedural toxicity risk with low relapse risk. Clinical levers: Donor optimization (avoiding MMUD/haploidentical donors), portal hypertension management, and conditioning de-intensification.
3. Disease-driven/ high relapse risk (13%): High relapse risk with low procedural toxicity risk. Clinical levers: Pre-transplant debulking/JAK inhibition, conditioning intensification, and post-transplant measurable residual disease-directed maintenance therapy.
4. High risk on both axes (5%): High relapse and high NRM. Clinical levers: Exploration of non-transplant therapies, novel clinical trials, or explicit shared decision-making regarding transplant candidacy.

An interactive calculator is openly available online at imtss-calculator.com.

Highlights

• Post-transplant cyclophosphamide (PTCy) at the standard dose of 50 mg/kg/d on days +3 and +4 after allogeneic hematopoietic cell transplantation (HCT) has been associated with cardiac, renal and pulmonary toxicity, delayed engraftment and graft failure.
• Prior reduced-dose studies were small and limited to haploidentical or human leucocyte antigen-matched donors. The authors therefore assessed graft-versus-host disease (GVHD), engraftment, and survival after uniform dose reduction to 40 mg/kg/d on days +3 and +4 (PTCy 40-40).

Retrospective cohort from 2 US academic centers A total of 115 patients were analyzed (median age, 65 years). The most common diagnoses were acute myeloid leukemia and myelodysplastic syndrome (66.9%). Most patients received reduced-intensity conditioning (RIC; 80%) and grafts from matched unrelated donors (63%). GVHD prophylaxis consisted of PTCy 40 mg/kg/day IV on days +3 and +4, followed by tacrolimus starting on day +5 and mycophenolate mofetil through day +35. Median follow-up was 9.0 months (range, 4.9–28.8).

The primary objectives were to assess cumulative incidence of acute GVHD, chronic GVHD, and relapse. Cumulative incidence of grades 2 to 4 acute GVHD by day +180 was 13% (95% confidence interval [CI] 8-20),Only one case of grade III acute GVHD occurred, with no grade IV cases. The 12-month cumulative incidence of moderate-to-severe chronic GVHD was 13% (95% CI 7-22). At 1 year,  relapse was 18% (95% CI 10-27) progression-free survival was 77% (95% CI 66–85) and overall survival was 84% (95% CI, 74–91).

By day +30, 18.3% of patients had experienced a clinically significant organ toxicity event, including acute kidney injury (7.8%), arrhythmia (7.0%), acute decompensated heart failure (1.7%), acute respiratory distress syndrome (1.7%), and myocarditis and pericarditis (0.8% each). No cases of hemorrhagic cystitis or attributable severe hepatotoxicity were observed. By day +100, bacterial infections had occurred in 30.4% of patients, clinically significant viral infections in 10.4%, and invasive fungal infections in 7.0%.

The authors conclude that PTCy at 40 mg/kg/day on days +3 and +4 is a safe and effective GVHD prophylaxis strategy after either RIC- or myeloablative-HCT, regardless of donor type.

Highlights

• Cytokine release syndrome (CRS) remains a common complication of haploidentical hematopoietic cell transplantation (haplo-HCT): it affects up to 95% of patients on post-transplant cyclophosphamide (PTCy)-based regimens. Early immunosuppression (EIS) has been proposed because alloreactive T-cell expansion occurs mainly within the first 72 hours post-transplant.
• The study’s objective was to evaluate the clinical impact of EIS in the incidence and severity of CRS in haplo-HCT recipients.

This systematic review and meta-analysis included trials and observational studies of adults undergoing haplo-HCT with EIS. EIS was defined as starting a calcineurin inhibitor (CNI) and/or mycophenolate mofetil (MMF) before day +5. Ten studies were included (n=1577), of which seven were single-arm meta-analyses. Acute leukemia was the most frequent indication, and most patients were in complete remission. EIS arms started CNI and MMF between day −1 and day +1, versus day +5 in standard arms. Among 294 EIS-treated patients, pooled CRS incidence was 37.8% (95% CI 27.3–49.7; I² 56.9%), and events were predominantly mild. Three retrospective comparative studies (265 patients) were pooled; in these, EIS was associated with lower odds of CRS (odds ratio 0.18; 95% CI 0.03–0.98; p=0.049; I² 42.1%). Grade II–IV acute GVHD ranged from 17% to 31% at early timepoints and up to 29% at 1 year. One study reported more grade III–IV acute GVHD with EIS (30% vs. 7.5%). Moderate-to-severe chronic GVHD ranged from 5.9% to 18%, with follow-up of 1–4 years.

The authors conclude that EIS shows potential utility reducing the incidence of CRS, but clinical trials are needed to answer this question.

Highlights

• The skin is the organ most often involved in chronic graft versus host disease (cGVHD). Topical corticosteroids remain the mainstay despite their adverse effects, no other skin-directed therapies are approved by the FDA.
• This study aimed to evaluate the clinical activity and safety of ruxolitinib 1.5% cream (used for atopic dermatitis and non-segmental vitiligo) in patients with epidermal or superficially sclerotic cutaneous cGVHD.

This single-centre, randomised, double-blind, intra-patient-controlled phase 2 trial randomised 24 patients (median age=47.5 years). Cutaneous cGVHD was epidermal in 21 (88%). Ruxolitinib 1.5% cream was applied twice daily for 28 days to one randomly assigned body side, with placebo vehicle on the contralateral side. The primary endpoint was the absolute day-28 between-side difference in affected lesional body surface area (BSA), recorded as percentage of total BSA on each assigned side and analysed in evaluable patients. Mean affected lesional BSA decreased from 14.3% to 6.2% with ruxolitinib versus 14.5% to 10.4% with vehicle (difference −4.2 percentage points; 95% confidence interval [CI] −6.8 to −1.5; p=0.0030). After mixed-effects adjustment for baseline between-side difference, the difference was −4.0 (95% CI −5.5 to −2.4; p<0.0010). Physician’s Global Assessment response on the ruxolitinib side occurred in 17/23 patients (74%) and five (22%) had complete response. Grade ≥3 treatment-emergent adverse events were sinusitis, lung infection and central nervous system leukaemia progression (one [4%] each), all deemed unrelated or unlikely related.

The authors conclude that ruxolitinib 1.5% cream improved active lesional BSA over 28 days compared with placebo vehicle in selected patients with epidermal or superficially sclerotic cutaneous cGVHD, with little short-term toxicity. They propose it as an adjuvant to systemic therapy or as sole therapy when BSA involvement is not extensive.

Highlights

• BEAM is the most widely used conditioning for autologous transplantation in relapsed HL (Hodgkin lymphoma). However, carmustine-related toxicity and limited availability of carmustine and melphalan have led centers to adopt alternative regimens, and prior comparisons with BEAM came mostly from retrospective series in mixed lymphoma populations, with divergent results.
• The study aimed to compare outcomes of adults with HL autografted in EBMT centers after BEAM versus FEAM, TEAM, EAM and BeEAM.

This study included 13207 adults with HL who underwent autologous hematopoietic stem cell transplantation (HSCT) in 2010–2022. At HSCT, 57% were in complete remission (CR), 29% in partial remission and 14% refractory. Conditioning was BEAM (carmustine, etoposide, cytarabine, melphalan; n=10167 [77%]) or an alternative that replaced carmustine with fotemustine (FEAM; n=1702 [13%]), bendamustine (BeEAM; n=729 [6%]) or thiotepa (TEAM; n=365), or omission of these (EAM; n=244 [2%]). The primary endpoint, 100-day non-relapse mortality (NRM), was 1.3% overall, with no significant difference among regimens (p=0.076). Median follow-up was 2.3 years. Non-CR status was associated with worse overall survival (OS, hazard ratio [HR] 2.01; 95% confidence interval [CI] 1.79–2.26; p<0.001). In center-stratified multivariable Cox models, BeEAM versus BEAM was associated with worse outcomes: OS  (HR 1.61; 95% CI 1.17–2.20; p=0.003), progression-free survival (HR 1.48; 95% CI 1.21–1.82; p<0.001), relapse incidence (HR 1.43; 95% CI 1.15–1.78; p<0.001) and NRM (HR 1.85; 95% CI 1.03–3.34; p=0.04). EAM regimen was also associated with decreased OS (HR 1.55; p=0.019). Propensity score matching confirmed the findings. The authors conclude that BEAM is an effective and safe conditioning regimen for HSCT in HL. Among the alternatives, bendamustine-based conditioning was associated with worse outcomes.

Clinical Management of Acute Graft-Versus-Host Disease: An Evidence-Based Review from the ASTCT Committee on Practice Guidelines

Diagnosis (expert consensus; not formally graded)

• Acute graft-versus-host disease (GVHD) is a clinical diagnosis, and biopsy is not required to establish it. However, alternative or coexisting diagnoses should always be considered.
• Endoscopy with biopsy is recommended for suspected lower gastrointestinal (GI) GVHD, particularly when symptoms are moderate or severe (e.g., ≥3 diarrheal stools/day above baseline, severe abdominal pain with or without ileus, or grossly bloody stool). A nondiagnostic biopsy does not rule out GVHD.
• Upper endoscopy is generally recommended to evaluate persistent anorexia, nausea or vomiting, or unexplained GI symptoms or weight loss.
• Skin biopsy may be useful in selected cases, although histologic findings often do not reliably distinguish acute GVHD from drug reactions or infection.
• Liver biopsy is recommended when the liver is the only potential site of involvement, since isolated hepatic GVHD occurs in <2% of cases. It is generally avoided when skin or GI involvement is also present.
• Biopsies should be interpreted by a pathologist with expertise in GVHD. Treatment should not be delayed while awaiting results when clinical suspicion is high.
• Classic acute GVHD occurs before Day 100 post-transplantation, and late acute GVHD occurs at or after Day 100. In both cases, diagnostic or distinctive features of chronic GVHD must be absent.

Grading and risk stratification

• The MAGIC (Mount Sinai Acute GVHD International Consortium) should be adopted as the consensus staging and grading scheme for acute GVHD. (Grade C; Highest Level of Evidence [LoE] 2+)
• The MAGIC Algorithm Probability, based on biomarkers ST2 and REG3α, is a validated risk-stratification tool for newly diagnosed acute GVHD, primarily for estimating the risk of non-relapse mortality. (Grade B; Highest LoE 2++)
• The Minnesota risk score, the Manhattan risk system and the MAGIC Composite Score are validated risk-stratification tools for newly diagnosed acute GVHD. They should not be used to diagnose acute GVHD. (Grade B; Highest LoE 2++)
• Risk-stratification tools are not yet validated to guide treatment decisions in newly diagnosed acute GVHD. (Grade B; Highest LoE 1+)

First-line treatment

• Systemic corticosteroids remain the standard first-line therapy for acute GVHD. An initial dose of methylprednisolone 2 mg/kg/day (or equivalent) is appropriate for newly diagnosed grade III–IV disease; a randomized trial showed no benefit from higher doses. (Grade B; Highest LoE 1+)
• Lower initial doses (methylprednisolone 0.5–1 mg/kg/day or equivalent) can be appropriate for stage 1 skin and/or upper GI involvement, as they likely preserve clinical benefit while reducing toxicity. (Grade B; Highest LoE 1+) The panel notes this approach could likely extend to isolated stage 2–3 skin involvement, but not to higher-stage lower GI or liver disease.
• Systemic agents should not routinely be used as alternatives to corticosteroids, nor added to them, for initial treatment. (Grade B; Highest LoE 1+)

Steroid-refractory acute GVHD

The current consensus definition of steroid-refractory acute GVHD should continue to be used as an eligibility criterion in clinical trials. (Grade D; Highest LoE 4) It comprises: (1) progression after 3 days of methylprednisolone 2 mg/kg/day (or equivalent); (2) no improvement after 7 days at that dose; (3) progression to a new organ during treatment with methylprednisolone 1 mg/kg/day (or equivalent) for skin and upper GI GVHD; or (4) recurrence during or after corticosteroid taper.

Second-line treatment

• Ruxolitinib is the consensus second-line systemic therapy for acute GVHD, based on the REACH1 and REACH2 trials. (Grade A; Highest LoE 1+)
• In pediatric patients, remestemcel-L-rknd can be considered an alternative to ruxolitinib for second-line therapy. (Grade B; Highest LoE 2++)

Third-line therapy and beyond

No specific agent is endorsed as the preferred third-line therapy. Options include extracorporeal photopheresis, mesenchymal stromal cells, alpha-1 antitrypsin, vedolizumab and fecal microbiota transplantation. (Grade B; Highest LoE 1+)