Background

Hemophagocytic lymphohistiocytosis(HLH) is a rare frequently fatal group of immunodeficiencies that mostly present with cytopenias and infections. It is characterized by an immune dysregulation that results in excessive cytokine expression, phagocytosis of blood cells by activated macrophages, and end-organ damage. There are familial forms (“primary HLH”), typically presenting in childhood, and sporadic forms (“secondary HLH”), often presenting in adults. While primary HLH is often limited to mutations affecting perforin dependent cytotoxicity, limited data are available on secondary HLH which is associated with conditions, such as malignancy and infection, without identifiable genetic abnormalities.

We reviewed diagnoses of HLH in adult treated at Barnes-Jewish Hospital in St. Louis since 1998. We hypothesized that serum ferritin – which is elevated in most HLH cases – can help in differentiating HLH etiology and might predict outcome.

Methods

We identified 56 adult patients with HLH seen during inpatient and outpatient visits using HLH ICD-9 billing code 288.4. Clinical and laboratory/hematopathology records were reviewed; 12 cases were excluded from the analysis – 9 for missing data and 3 for insufficient diagnosis of HLH. We followed the 2004 Histiocyte Society Criteria for HLH. All results were presented as median plus range, mean +/- standard deviation, or percentages as indicated. We compared clinical and laboratory parameters between HLH groups by using the Student T-test, Mann-Whitney, or Chi square test, where appropriate. Predictive/prognostic factors were evaluated by univariate analysis. Results were considered significant at p <0.05. All statistical analyses were carried out using SPSS software, version 17 (SPSS Inc., Chicago, IL).

Results

Among the included 44 cases, the median age was 45 years (range, 20-80); 21 patients were male. Fever and splenomegaly were seen in 77.3% and 52.3% of patients, respectively. The most common laboratory abnormalities during hospitalization were hyperferritinemia (95.3%; median 17,414; range 225-684,000), pancytopenia/bicytopenia (72.7%), elevated triglycerides (52.5%; median 267.5; range 71-857), and low plasma fibrinogen (30.8%). Pathology revealed the characteristics hemophagocytosis of HLH in most cases. The recognized causes of HLH were: 29.5% malignancies (B-cell lymphoma [n=3], T-cell lymphoma [n=6], acute myeloid leukemia [n=1], myeloproliferative/myelodysplastic syndromes [n=2], metastatic breast cancer [n=1]), 25% viral (HCV [n=5], HIV [n=2], HBV [n=1], EBV [n=1], CMV [n=1], influenza [n=1]), 11.4% rheumatologic (systemic rheumatoid arthritis [n=3], Still's disease [n=1], lupus-like syndrome [n=1]), 4.5% bacterial sepsis, 4.5% post solid organ transplantation (liver [n=1], lung [n=1]), 2.27% fungal infection, 2.27% histoplasma infection, 20.4% unknown. Corticosteroids were frequently used in treatment; few patients received cyclosporine, etoposide, intravenous immunoglobulins, or stem cell transplantation. Thirteen patients died, 9 (69%) of septic shock.

The cohort was then dichotomized based on disease etiology into malignancy- and infection-associated HLH; this resulted in 13 and 15 cases in each category, respectively. Patients with malignancy-associated HLH had higher serum ferritin levels than those with infection-associated HLH but the different was not significant (84,592.8 +/- 185,009.4 vs. 23,262 +/- 19,090.5 ng/mL) (p = 0.23). Triglycerides were higher in the infection-associated HLH group compared to the other group with a difference approaching significance (358.6 +/- 201.7 vs. 233.8 vs. 124.6; p = 0.076). Overall, patient with higher ferritin levels were more likely to have a poor prognosis. At a cutoff ferritin level of 60,000 ng/mL, 62.5% of patients with higher levels died compared to 22.8% who died while having lower ferritin levels (p = 0.042).

Conclusion

HLH in the adults is associated with a wide spectrum of underlying diseases. It has a high mortality rate, particularly in patients with markedly elevated ferritin levels. Further studies are needed to develop prognostic criteria for outcomes in adult HLH. Analysis of additional cases is ongoing and will be presented at the meeting.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution