Hemophagocytic Lymphohistiocytosis: Retrospective Analysis for Prognostic Factors

Background

Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening systemic inflammatory condition. Due to rarity of the cases, it presents difficulties in diagnosis and management. Survival remains poor despite aggressive chemotherapy.

Objective

Patient outcomes varied markedly despite standardize therapy. Reliable prognostic disease markers may help to tailor intensity of therapy and predict long term outcomes. We attempt to look for variables associated with difference in mortality within 30 days of diagnosis.

Methods

We performed a retrospective search on mayo clinic patient database for the patients with the diagnosis of HLH from 2005 to 2015.

HLH-04 criteria were used to select the study population. Patients were divided in two groups based on survival after the diagnosis. We analyzed different clinical and laboratory parameters to detect difference between the patients expired within 30 days of diagnosis and who survived longer than 30 days.

Baseline Characteristics:

Demographics: 40 patients were included in the analysis who met HLH- 04 criteria. Mean age was 49 years, 40% (16/40) were female and 60% (24/40) were male. Underlying HLH etiology was malignancy 37% (15/40), infection 20% (8/40), rheumatological 17% (7/40), idiopathic 20% (8/40). Two patients were peripartum and one with Kikuchi syndrome. EBV DNA PCR were positive in 32% (13/40) of patients. Table 1 show clinical and laboratory characteristics according to HLH-04 criteria.

Treatment: Steroids were used in 92% (37/40), etoposide was used in 55% (22/40), and HLH 04 protocol (Etoposide/dexamethasone/cyclosporine) was used in 40% (16/40) of the patients. IVIG was used in 13% with underlying rheumatological process. Mean follow up was 57 weeks (0.1 to 336 weeks) for the whole group. Total 40% (16/40) died within 30 days of diagnosis.

Results

Risk of 30-days mortality was significantly higher in the patients with ferritin >5000 mcg/L at the time of diagnosis and age > 55 years.

Out of total 40 patients, 54% (12/22) died in ferritin >5000mcg/L group and 22% (4/18) died in ferritin < 5000 mcg/L group within 30 days. (p-0.03) Death rate within 30 days was 65% (11/17) with age > 55 years and 22% (5/23) with age < 55 years at the time of diagnosis of HLH. (p-0.05)

No difference between the groups in terms of gender, EBV positivity, underlying etiology and etoposide use was found in 30 days mortality. Table 2 summarizes the findings.

Discussion

HLH is a complex disorder with significant heterogeneity in terms of underlying etiology and response to treatments.

Diagnosis is based on a set of clinical and investigational parameters. Most commonly used criteria are HLH-04.

Treatment involves rapid immunosuppression with steroids, chemotherapy and calcineurin inhibitors.

Previous retrospective studies have pointed out different risk factors associated with poor survival, which are malignancy, hypoalbuminemia, elevated creatinine and bilirubin.

Ours is a relatively small retrospective analysis, but it shows significant prognostic value to elevated ferritin (>5000 mcg/L) at the time of diagnosis and age > 55 years.

Survival remains poor in high risk patients despite aggressive therapy. Biological agents (IL1 and IL6 blockage) may provide new realm of therapy with tolerable toxicity profile.

Conclusion

Elevated ferritin at the time of diagnosis and older age are associated with significant risk of 30 day mortality in HLH.

These factors can be incorporated in future clinical trials to choose different treatment pathways.