Factors Influencing 1-Year Survival in Haemphagocytic Lymphohistiocytosis: A National Cohort Study in England

Background: Haemophagocytic lymphohistiocytosis (HLH) is a rapidly progressive, life-threatening clinical syndrome of excessive immune activation. Multiple predisposing factors influence the likelihood of any individual developing HLH at any time, encompassing genetic, auto-immune disease, immunosuppression and malignancy. In partnership with the National Congenital Anomalies and Rare Disease Registration Service (NCARDRS) we have undertaken a nationwide study in England of all cases of HLH diagnosed between 2003 and 2018, and modeled interactions of demographics and comorbidities with survival.

Methods: Using national linked electronic health data from hospital admissions and death certification we identified cases of HLH that were diagnosed in England between 1^st January 2003 and 31^st December 2018 using a previously validated approach (Bishton et al Brit J Haem 2021). Patients of all ages admitted to hospital with an admission coded with HLH ICD-10 codes D76.1 and D76.2, or who died with a death code of D76.1, D76.2 or D76.3 provided in the latter situation there was confirmatory free text on the death certificate indicating HLH, were included. We calculated 1-year survival using Kaplan-Meier methods by calendar year, age group, sex and comorbidity (haematological malignancy, inflammatory rheumatological or bowel diseases (IBD). We modelled 1-year survival and the interactions between age and associated comorbidity using Cox regression.

Results: There were 1628 people with HLH included in the analysis. 706 (68.6%) patients were aged ≥15 years, 920 (56.5%) were male and 855 (52.5%) had at least one associated comorbidities recorded. The most common comorbidities were lymphoma (n=270 [16.6%]), leukaemia (n=125 [7.7%]), systemic juvenile idiopathic arthritis (SJIA) (n=78 [4.8%]), IBD (n=70 [4.3%]), vasculitis (n=64 [3.9%]) and systemic lupus erythematosus (SLE) (n=47 [2.9%]). Overall, crude 1-year survival was 50% (95% CI 48-53%) with no improvement observed over the study period. Survival varied significantly with age (0-4, 61%; 5-14 76%; 15-54 61%; >55 24% p<0.01)(figure 1), sex (males, 46%, females, 55% p<0.01) and associated comorbidity (rheumatological/IBD, 69%, haematological malignancy 28%, other malignancy, 37% p <0.01). Those aged 0-14 and 15-54 years had a 3-fold increased risk of death at 1-year among HLH associated with haematological or non-haematological malignancy compared to rheumatological disease/IBD. Predicted 1-year survival decreased markedly with age in those with rheumatological disease/IBD (age 0-14, 84%; 15-54, 73%; >55, 27%) meaning survival for those aged >55 years was poor regardless of the underlying disease process. Notably there was no difference in outcomes by lymphoma subtype (B-cell, T-cell or Hodgkin lymphoma) in contrast to the auto-immune sub-groups, where outcomes for adult onset Stills disease and SJIA were excellent (77% and 89% respectively) whilst survival in rheumatoid arthritis (35%) and vasculitis (39%) were similar to non-haematological malignancy driven HLH.

Conclusion: We provide the first estimates of survival for HLH by the associated trigger factors (vasculitis, rheumatoid arthritis, IBD, malignancy), on a national scale, and make the critical observation that outcomes have remained static for the study period. Furthermore outcomes for haematological malignancies are dismal regardless of disease subtype, whilst significant variation in outcomes for different auto-immune disease subtypes are seen indicating that the severe impact of this syndrome is not limited to those with underlying malignancy.

Lanyon:Vifor Pharma: Research Funding. Bishton:Celltrion: Honoraria; Roche Pharmaceuticals: Consultancy, Honoraria; Gilead: Honoraria; Incyte: Consultancy; Lilly: Consultancy.