Etiologic heterogeneity in secondary hemophagocytic lymphohistiocytosis: A retrospective real-world study from a quaternary academic center

Abstract

Abstract Hemophagocytic lymphohistiocytosis (HLH) is a rare, life-threatening hyperinflammatory syndrome marked by excessive immune activation and a cytokine storm that can rapidly progress to multiorgan failure. Secondary HLH (sHLH) is most commonly triggered by underlying conditions such as malignancies, infections, or autoimmune disorders. Although early recognition and timely initiation of therapy are essential to improving outcomes, real-world data on sHLH remain limited. In this retrospective study spanning over a decade, we comprehensively characterized the clinical, laboratory, and etiologic features of sHLH at a quaternary academic center. We further examined survival outcomes between different etiologies and assessed the influence of underlying cardiometabolic comorbidities on overall survival (OS), which has not been systematically explored. We conducted a retrospective cohort study of patients diagnosed with HLH between January 2012 and March 2025. HLH diagnoses were confirmed based on fulfilment of at least 5 out of 8 clinical or laboratory criteria outlined in the HLH-2004 diagnostic guidelines. Patients with primary (familial) HLH or those with a historical diagnosis of HLH before 2012 with insufficient clinical data were excluded from the analysis. OS was calculated from the date of symptom onset and estimated at 6 and 12 months (mo) using the Kaplan-Meier method and compared across etiologic subgroups using the log-rank test. Among 142 patients diagnosed with HLH, 113 fulfilled eligibility criteria for secondary HLH and comprised the final study cohort. The median age at diagnosis was 58 years (IQR: 33.5–66), and 61.9% were male. The underlying etiologies of HLH were predominantly malignancy (50%), with DLBCL being the most common subtype (23%). Other causes included infections (31%), autoimmune diseases (17%), and unknown origins (3%). All patients underwent bone marrow biopsy, and hemophagocytosis was documented in 46.0% of patients. Regarding cardiometabolic comorbidities, 38% had hypertension (HTN), 21% had diabetes mellitus (DM), and 40% had a BMI >30 kg/m². Low HDL (<50 mg/dL) was observed in 60%, and hypertriglyceridemia (>150 mg/dL) in 72%. Metabolic syndrome, as defined by NCEP ATP III criteria, was present in 44% of the cohort. Laboratory evaluation showed a median H-Score of 210 (IQR: 169–238, n=93). Inflammatory and hematologic markers included median values of ferritin (11,518 ng/mL), fibrinogen (244 mg/dL), hemoglobin (8.5 g/dL), platelet count (58.5 ×10⁹/L), and ANC (2.77 ×10⁹/L). Median triglycerides (TAGs) and soluble IL-2 receptor (sCD25) were 278 mg/dL and 11,972 pg/mL, respectively. At the time of last follow-up, 57% of patients were deceased. The median follow-up among survivors was 42 mo (range: 0.4–132.3). The median OS for the entire cohort was 6 mo, with 6 mo and 12 mo OS rates of 48% (95% CI: 40–59%) and 46% (95% CI: 38–57%), respectively. Patients who had malignancy-associated HLH demonstrated significantly worse outcomes (median OS: 1.6 mo; 6 mo OS: 31%; 12 mo OS: 27%, p<.0001) compared to those with autoimmune or infection-associated HLH (both groups: 12 mo OS: 66% and 65%, respectively). TAG levels ≥150 mg/dL were associated with significantly inferior OS compared to TAG <150 mg/dL (median OS: 3 vs. 132 mo, p=.023). While not statistically significant, there was a trend toward worse OS in patients with metabolic syndrome (median OS: 2.5 vs. 8 mo, p=.195) and those with DM (median OS: 1 vs. 8 mo, p=.1636). No significant survival differences were observed based on sex, evidence of hemophagocytosis on bone marrow, BMI, HDL levels, or HTN status. In our large, single-center cohort of sHLH patients, malignancy-associated HLH conferred the poorest survival outcomes as compared to infection and autoimmune-associated HLH. Importantly, elevated TAG levels were independently associated with inferior OS, and trends suggest that metabolic syndrome and diabetes may further exacerbate outcomes. These results highlight the need to consider cardiometabolic risk factors as potential modifiers of HLH prognosis and support the incorporation of metabolic profiling into future risk stratification strategies.