The syndrome at the center of the Norwegian case series, hemophagocytic lymphohistiocytosis, is best understood as a runaway positive feedback loop in the immune system rather than a discrete disease entity. In its normal state, the immune response to infection or cellular stress is self-limiting: activated macrophages and cytotoxic T cells eliminate infected or malignant targets, then recede as regulatory signals reassert control. In HLH, this braking mechanism fails. Cytotoxic cells — whether defective through inherited mutations in primary HLH or overwhelmed by persistent antigenic stimulation in secondary HLH — fail to clear their targets, and the sustained activation of macrophages and T cells floods the circulation with inflammatory cytokines such as interferon-gamma, interleukin-6, and tumor necrosis factor. The name of the syndrome describes its histological signature: hemophagocytosis, the engulfment of red cells and other blood elements by activated macrophages in the bone marrow, spleen, and lymphoid tissues. Clinically, the result is persistent fever, cytopenias, hepatosplenomegaly, coagulopathy, and, in severe cases, progressive multiorgan failure. Untreated, the mortality of the syndrome approaches one hundred percent, and even with modern therapy, survival in high-risk adult populations remains poor.

Ferritin, the iron-storage protein that served as the screening tool in the Norwegian study, deserves particular attention because of its unusual role in this syndrome. Hyperferritinemia is not specific to HLH — it rises in sepsis, liver failure, adult-onset Still disease, and many other inflammatory states — but the magnitude of elevation is informative. Values above 10,000 micrograms per liter carry a meaningful probability of HLH, and the patients in the Oslo series far exceeded this threshold, with a median of 107,500 micrograms per liter and a maximum of 326,000. Ferritin in HLH is not merely a passive marker of inflammation; it is thought to reflect macrophage activation directly, since activated macrophages secrete ferritin into the plasma. The investigators chose an arbitrary cutoff of 20,000 micrograms per liter precisely because it is high enough to be highly suggestive of hyperinflammatory syndromes in a transplant population, where moderate elevations are common and nonspecific. Their screening strategy — extracting every transplant recipient with EBV reactivation and ferritin above this level — illustrates both the promise and the limits of biomarker-driven case finding in a condition that lacks a gold-standard diagnostic test.

The diagnostic instrument most widely used for secondary HLH, the H-score, was developed in French cohorts of immunocompetent adults and assigns points for fever, organomegaly, cytopenias, ferritin levels, triglycerides, fibrinogen, and evidence of hemophagocytosis on bone marrow examination. In the Oslo series, the eight patients in whom the score could be calculated retrospectively all exceeded 200 points, corresponding to a probability of HLH above ninety percent — a striking validation that these were genuine cases rather than diagnostic overreach. Yet the authors are careful to note the caveat that the score was never validated in immunocompromised hosts. Bone marrow hemophagocytosis, one scored variable, is notoriously insensitive and often absent early in the disease course, and three of the eight scored patients lacked marrow sampling entirely. In a transplant recipient, fever and cytopenias are ambiguous findings that could equally signal graft failure, drug toxicity, sepsis, or graft-versus-host disease, and the treating physician’s judgment — which served as the diagnostic standard in this series — remains the operative reality in most centers. The absence of validated diagnostic tools for this population is not a bureaucratic gap but a genuine obstacle to early intervention, since every day of diagnostic uncertainty in a syndrome that killed its victims within weeks of diagnosis is a day of untreated cytokine storm.

The relationship between EBV reactivation and HLH in this setting is mechanistically coherent. EBV infects B cells and establishes lifelong latency in memory lymphocytes, kept in check by cytotoxic T cells that recognize viral antigens. When T-cell depletion removes this surveillance, infected B cells can proliferate unchecked, producing the spectrum of post-transplantation lymphoproliferative disease, from polymorphic lesions to monomorphic lymphomas. In the Oslo series, all ten patients had histologically confirmed PTLD, and nine of the ten had monomorphic disease — the more aggressive end of that spectrum. The massive antigenic load presented by uncontrolled EBV-infected B cells, combined with the cytokine milieu of the post-transplant period, provides exactly the kind of persistent, unresolvable stimulation that drives secondary HLH. The very high viral loads observed — with median whole-blood EBV DNA exceeding one million international units per milliliter, the ceiling of the assay — underscore the degree of uncontrolled viral replication these patients experienced. The temporal pattern is also telling: reactivation occurred at a median of fifty-two days post-transplant, a window consistent with the waning of pharmacologic immunosuppression from the conditioning period and the slow reconstitution of donor T-cell immunity, which typically takes months after ATG-conditioned transplantation.

The therapeutic dilemma the authors describe is one of the most difficult in transplant medicine. The standard approach to PTLD after allogeneic transplantation is twofold: reduction of immunosuppression to allow recovered T cells to attack the EBV-infected B cells, and rituximab, a monoclonal antibody against CD20, to deplete the B-cell compartment directly. But in a patient simultaneously developing HLH, reduction of immunosuppression risks pouring fuel on the hyperinflammatory fire, while intensifying anti-inflammatory therapy risks further impairing the T-cell recovery needed for viral control. The Oslo patients received rituximab and corticosteroids in all cases, with additional agents — chemotherapy, intravenous immunoglobulin, the interleukin-1 receptor antagonist anakinra, the interleukin-6 receptor antibody tocilizumab, and the JAK inhibitor ruxolitinib — deployed in various combinations. That none survived despite this arsenal reflects both the aggressiveness of the syndrome and the absence of an established treatment algorithm for EBV-HLH in this population. Anakinra and ruxolitinib, which target specific cytokine pathways implicated in HLH pathogenesis, represent a newer generation of rational therapy, and small series in other HLH contexts have suggested benefit, but controlled data in transplant-associated EBV-HLH do not exist. EBV-specific cytotoxic T lymphocyte therapy, which could in principle restore viral control without broad immunosuppression, is an attractive theoretical solution but remains logistically complex and unavailable at most centers, including the authors’ own institution.

The preventive strategies the authors propose deserve scrutiny in light of broader trends in transplantation. Post-transplantation cyclophosphamide, which selectively depletes proliferating alloreactive T cells while sparing regulatory and memory populations, has gained substantial ground as graft-versus-host disease prophylaxis and is associated with lower rates of viral reactivation than T-cell-depleting serotherapy, precisely because it preserves more antiviral immune competence. Its adoption as an alternative to ATG-based prophylaxis in appropriate settings could plausibly reduce the incidence of the syndrome described here. Prophylactic rituximab, meanwhile, acts by a different mechanism: by depleting B cells before EBV reactivation occurs, it removes the target cell population in which the virus replicates. The retrospective study the authors cite, in which a single peritransplant dose reduced EBV reactivation and PTLD compared with a preemptive strategy, is suggestive but not definitive, and the finding that long-term survival was similar between groups tempers enthusiasm — preventing PTLD episodes does not necessarily translate into preventing death when competing risks such as relapse and graft-versus-host disease remain. Both strategies, in any case, address the trigger rather than the hyperinflammatory response itself, and it remains unknown whether any intervention can reliably interrupt the transition from EBV-driven PTLD to fulminant HLH once it has begun.

The incidence figure of 1.1 percent among ATG-exposed patients should be interpreted with appropriate caution. The authors themselves note that underdiagnosis is very likely, since milder cases may resolve when treatment of the EBV trigger proves sufficient, and the retrospective design, the reliance on an arbitrary ferritin cutoff, and the exclusion of four probable cases with incomplete data all suggest the true burden may be higher. The denominator — 930 ATG-containing transplants over nearly sixteen years — is large enough to establish that this is not an isolated curiosity, but small enough that the confidence interval around any incidence estimate is wide. What the series establishes unambiguously is the phenotype: early onset within three months of transplant, exclusive occurrence in ATG-treated recipients of EBV-seropositive grafts, universal progression to histologically confirmed PTLD, and uniform fatality within weeks of diagnosis. For clinicians managing ATG-conditioned transplant recipients, the practical implication is concrete: EBV reactivation in this population should prompt ferritin measurement alongside viral load monitoring, and a markedly elevated ferritin in a febrile, cytopenic patient should trigger a full HLH work-up without waiting for the classical diagnostic constellation to complete itself. The syndrome described here is rare, but its rarity is precisely what makes it dangerous — few clinicians encounter it often enough to recognize it early, and by the time the diagnosis is secure, the therapeutic window has usually closed.

Subject of Research: Hematology

Article Title: Epstein-barr virus-associated hemophagocytic lymphohistiocytosis after antithymocyte globulin-conditioned allogeneic hematopoietic stem cell transplantation: a single-center case series

Article References: Abrahamsen, I. W., Myhre, A. E., Vo, C. D., Gedde-Dahl, T., & Tvedt, T. H. (2026). Epstein-barr virus-associated hemophagocytic lymphohistiocytosis after antithymocyte globulin-conditioned allogeneic hematopoietic stem cell transplantation: a single-center case series. Annals of Hematology, 105(11), Article 461. https://doi.org/10.1007/s00277-026-07307-3

Image Credits: AI Generated

DOI: 10.1007/s00277-026-07307-3

Keywords: Epstein-Barr virus, hemophagocytic lymphohistiocytosis, stem cell transplantation, ATG, PTLD, graft-versus-host disease, rituximab, ferritin, Epstein-barr, virus-associated, hemophagocytic, lymphohistiocytosis