HLH Medical Condition: Diagnostic Criteria, HScore, and Treatment Every Heme/Onc Fellow Should Know Cold

Written by Dr. Roupen Odabashian MD, FRCPC
Hematologist-Oncologist | Founder, MeDucation AI | Updated November 2026

Hemophagocytic lymphohistiocytosis (HLH) is a hyperinflammatory syndrome, not a single disease, caused by an immune system that cannot turn itself off. It gets missed because it mimics sepsis, liver failure, or a flare of the underlying malignancy, and because the people best positioned to catch it are often looking for something else entirely^[1]. If you see a patient with fever, cytopenias, and a rising ferritin who is not responding to antibiotics, the things to do are: send a ferritin, triglycerides, fibrinogen, and soluble CD25 (or calculate an HScore with what you already have), start looking for the trigger (infection, malignancy, or a rheumatologic process) in parallel rather than sequentially, and call hematology before you wait for a bone marrow biopsy to confirm hemophagocytosis, because that finding is neither sensitive nor specific and should never be the thing that delays treatment^[1].

Conflict of interest, up front: I'm a practicing hematologist-oncologist and I founded MeDucation AI, which sells a physician-written heme/onc question bank and AI-powered study tools. I have no financial relationship with any drug or diagnostic company mentioned below, and I'll tell you plainly where the adult evidence is thin.

What should a fellow actually do, ranked by urgency?

Here is the ranked approach, from what buys you time to what you need eventually but shouldn't wait on.

What is HLH, and why does it get missed?

HLH is a final common pathway of excessive immune activation, producing fever, hepatosplenomegaly, cytopenias, and a specific laboratory signature of elevated ferritin, triglycerides, soluble CD25, transaminases, LDH, and D-dimers, with low fibrinogen, albumin, and sodium^[1]. It is missed for a structural reason: the patients who develop it are already sick with something else, usually an infection or a cancer, so the fever and cytopenias get attributed to the primary problem instead of recognized as a second, superimposed syndrome^[1]. Hemophagocytosis itself, the bone marrow or lymph node finding that gives the syndrome its name, is neither sensitive nor specific and can appear in septicemia and other unrelated conditions, which is part of why relying on it as a gatekeeper delays diagnosis^[1].

What are the HLH-2004 diagnostic criteria?

Five of these eight must be met, with the explicit comment that if hemophagocytic activity isn't proven at presentation, clinicians should keep looking rather than treat its absence as reassuring^[3]. The important caveat for fellows: there are "currently no generally accepted criteria for the definition of Malignancy-Triggered HLH or HLH During Chemotherapy," and HLH-2004 is used as a substitute despite acknowledged weaknesses in that setting^[1].

What does the HScore add that HLH-2004 doesn't?

The HScore is a weighted probability tool built specifically from adult data, not a pediatric substitute. It was designed by Fardet and colleagues from a study of 312 adult patients, with and without HLH, that had a 45% prevalence of hematologic malignancies in that cohort, using features similar to the HLH-2004 criteria to estimate the likelihood that a patient actually has HLH rather than simply meeting a checklist^[1]. Separately, a group proposed diagnostic criteria for adult lymphoma-associated hemophagocytic syndrome from 142 cases, built from most of the HLH-2004 criteria but in a different configuration that also incorporates LDH and D-dimers; these have not achieved wide acceptance^[1].

The practical takeaway: in an adult oncology patient, lean on an adult-derived probability tool and adult-pattern labs (LDH, D-dimers) rather than forcing a pediatric five-of-eight rule that was never validated for malignancy-associated disease^[1].

What's the difference between primary and secondary HLH?

Primary (familial) HLH is a genetically heterogeneous group including familial HLH types 2 through 5, Griscelli syndrome type II, and the X-linked lymphoproliferative syndromes, and it predominantly presents in childhood, often triggered by infection^[1]. The incidence of the classic autosomal recessive form is estimated at around 1 in 50,000 live-born children, and it is fatal with a median survival under two months if untreated^[3].

Secondary HLH is acquired, and in adults the dominant drivers are infection and malignancy. Roughly 1% of adults with a hematologic malignancy develop HLH, and nearly half of published adult HLH cases overall are malignancy-triggered^[1]. Age is a strong clue: in a large Japanese survey of 799 HLH patients, lymphoma was the trigger in 68% of patients over age 60, compared with 38% aged 30 to 59, 10% aged 15 to 29, and 0% under age 14^[1]. In children, a malignant trigger is far less common, reported at 8% prevalence^[1].

Geography matters too: DLBCL is the predominant trigger in Western countries and Japan, while T-cell neoplasms predominate in China and Korea^[1]. EBV co-infection is common but varies sharply by lymphoma subtype, found in up to 90% of Hodgkin lymphoma-associated HLH versus roughly 33% in peripheral T-cell lymphoma and low rates in DLBCL^[1]. In HIV patients with HLH, more than half have an underlying malignancy, usually lymphoma^[1].

What workup should a fellow actually order?

Order the HLH-2004 panel and an HScore-compatible set in parallel with an aggressive trigger hunt, because treatment decisions can't wait for every result to return^[1][2]. Practically, that means CBC with differential, fibrinogen, triglycerides, ferritin, soluble CD25 if your lab can send it, LDH, D-dimer, liver panel, and a sodium and albumin given their known association with HLH's lab signature^[1]. Trigger-side, that means blood and urine cultures, EBV and CMV PCR given their frequency as primary HLH triggers in children and co-triggers in adult lymphoma-associated disease, and imaging or pathology review aimed at finding an occult lymphoma, since nearly half of adult cases trace back to one^[1]. Bone marrow biopsy remains useful for trigger identification (finding an occult lymphoma or confirming hemophagocytosis as supportive evidence) but a negative marrow should never be used to exclude HLH or delay treatment^[1].

For trending disease activity once treatment starts, platelet count tends to fall quickly with disease flares, ferritin rises quickly with active disease but normalizes slowly even after inflammation resolves, and fibrinogen, soluble CD25, and LDH are also used as activity markers^[1].

What does the HLH-2004 treatment protocol actually involve?

Patients without a known family history who achieve complete resolution after eight weeks typically stop therapy, to avoid transplanting a child who may have had secondary rather than familial disease. Patients with confirmed familial disease, genetically verified disease, or severe/persistent/reactivated non-familial disease go on to continuation therapy with the same three drugs, with transplant performed as early as an acceptable donor is available^[3].

In the malignancy-associated and chemotherapy-associated setting, management is explicitly a balancing act: treating the HLH while also treating the underlying cancer and managing other treatment-related toxicity, rather than choosing one over the other^[2]. In practice, adult hematologists frequently incorporate lymphoma-directed chemotherapy regimens themselves, since standard regimens like CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) or its etoposide- or methotrexate-containing variants (CHOEP) double as HLH-directed therapy in lymphoma-triggered disease^[1].

Those numbers tell you two things: outcomes in familial disease improved substantially once the pre-transplant regimen could reliably bridge patients to HSCT^[3], and malignancy-associated HLH in adults still carries a wide and often grim survival range, with T-cell lymphoma-triggered HLH carrying a worse prognosis than B-cell lymphoma-triggered disease, and the presence of HLH in a lymphoma patient independently predicting worse survival and earlier death^[1].

What is the evidence for ruxolitinib, emapalumab, and anakinra in HLH?

Emapalumab is the one drug on this list with a completed phase 2/3 registration trial and FDA approval. It's an anti-interferon-gamma monoclonal antibody, the first approved treatment specifically for primary HLH, and it was studied in a single-arm, open-label trial of 34 patients (median age 1 year) with primary HLH that was refractory, recurrent, or progressive, or in patients intolerant of conventional therapy^[4]. The study achieved its primary endpoint, with 63% of patients demonstrating overall response at the end of the treatment period, including complete response, partial response, or HLH improvement, and 70% of trial participants went on to HSCT^[4]. Adverse events were substantial: infections in 56%, hypertension in 41%, infusion reactions in 27%, and fatal adverse reactions (septic shock, GI hemorrhage) in 6%^[4]. Emapalumab has since also been approved for HLH/macrophage activation syndrome in known or suspected Still's disease, including systemic JIA, with an inadequate response or intolerance to glucocorticoids, or with recurrent MAS, with 54% of patients achieving complete response at week 8 and 82% achieving clinical MAS remission at week 8 in that population^[5].

For ruxolitinib and anakinra, the sources available here do not contain a trial result I can cite, so I'm not going to manufacture one. What I can tell you from the guidance I do have is that cytokine-directed therapy has a conceptual parallel in a related syndrome: interleukin-6 blockade with tocilizumab has been proposed for cytokine release syndrome after T-cell engaging therapies, but because the underlying biology of iatrogenic CRS differs from HLH occurring in the context of malignancy, it is explicitly unclear whether that same approach (or by extension, other cytokine-targeted agents) will be effective for HLH itself^[1]. If you're being tested or asked to counsel a colleague on ruxolitinib or anakinra in HLH, say plainly that the evidence base for emapalumab in primary HLH is far more mature than for these other targeted agents in this specific indication, and check current guideline and product literature directly rather than assuming equivalence.

When do you call transplant?

Call transplant as soon as a patient has confirmed familial disease, a genetically verified diagnosis, or non-familial disease that is severe, persistent, or reactivated, and start the donor search in parallel with continuation therapy rather than after it, since HSCT should happen "as early as possible, when an acceptable donor is available"^[3]. The rationale is that pre-transplant therapy under HLH-94 successfully bridged as many as 80% of patients with verified familial disease to transplant, which is the entire reason the continuation-therapy structure exists^[3]. In malignancy-associated disease, the transplant conversation is different and happens alongside, not instead of, lymphoma- or leukemia-directed therapy, since HLH management in that setting requires balancing treatment of the malignancy against control of the HLH itself^[2].

What should a fellow expect the boards to test on HLH?

• Knowing that hemophagocytosis on marrow biopsy is neither sensitive nor specific and is not required to start treatment^[1].

• Distinguishing primary (genetic, pediatric-predominant) from secondary HLH, and recognizing the age-lymphoma relationship: lymphoma-triggered HLH rises sharply with age^[1].

• Knowing that emapalumab is FDA-approved for refractory/recurrent/progressive primary HLH and separately for HLH/MAS in Still's disease, and being able to state the mechanism (anti-interferon-gamma antibody)^[4][5].

References

1. Consensus recommendations for the diagnosis and management of hemophagocytic lymphohistiocytosis associated with malignancies | Haematologica

2. Diagnosis and Management of Secondary Hemophagocytic Lymphohistiocytosis | NCCN Continuing Education

3. HLH‐2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis

4. FDA Approves Emapalumab for Rare Primary Hemophagocytic Lymphohistiocytosis | HCPLive

5. FDA approves Gamifant® (emapalumab-lzsg) as first-ever treatment for adults and children with Macrophage Activation Syndrome in Still's disease | Sobi