HLH: Time to Transplant; A 10-Year UK Retrospective Cohort Study

Background

Primary haemophagocytic lymphohistiocytosis (HLH) autosomal recessive form, has an estimated incidence of 1:50,000 and an average survival of 2 months if left untreated.In autosomal recessive cases there is often no positive family history, which may lead to a delay in diagnosis. Multiple mutations associated with primary HLH have been described. Treatment for primary HLH should begin as soon as diagnosis is established in order to induce remission and proceed to haematopoietic stem cell transplantation (HSCT), the only curative option. No comprehensive international studies for primary HLH have occurred since the HLH-2004 study, nor have any countries reported population based national outcomes since then.

Aims

The primary aim of this research was to describe the outcomes of all children identified as having a mutation associated with primary HLH via the molecular immunology service in the UK and in particular to elucidate the relationship of time of transplant to OS and event free survival (EFS). Secondary aims include to determine the factors associated with better OS.

Methods

All children (<18-years of age) diagnosed with primary HLH from January 1^st 2010 to December 31^st 2019 through genetic samples processed at either Great Ormond Street Hospital or Manchester Children’s Hospital, UK were identified for 7 children’s hospitals in the UK. Patient details including demographics, diagnostic criteria of HLH at diagnosis, transplant related parameters and information on relapse and death were collected, whereby OS and EFS could be assessed.

Results

90 patients from 7 different centres within the UK had mutations identified in keeping with a new diagnosis of primary HLH. 59% (53/90) of this cohort were male and the median age at diagnosis was 189 days (IQR 1411 days; Q1 53, Q3 1465). The median time to transplant from diagnosis was 117 days (IQR 81 days, Q1 81, Q3 179). Genetic testing was performed on 100% of patients, with 64% of these having a classical mutation for familial HLH, 29% having an immunodeficiency associated HLH, and no confirmed genetic anomaly in 7%. Of the 90 patients, 51% had chemotherapy-based treatment, 25% had chemo- and immunotherapy, 6% immunotherapy-based therapy, 8% other therapies and 10% undeclared. 76 of our cohort received a HSCT, with 5 of these receiving a second HSCT, due to graft failure or relapse.

33 (37%) patients died: 12 (13%) prior to HSCT, 50% of which were due to multi-organ failure. EFS and OS at 1 year were 64% and 65%, falling to 54% and 58% at 5 years respectively.

Subgroup analysis of OS following HSCT was performed on the following groups: no active disease no relapse (CR1); no active disease after at least one relapse (CR2); active disease without relapse (primary refractory; REF1); active disease with relapse (secondary refractory; REF2). OS at 1y and 5y was significantly better in the CR1 group compared to the REF1 group (p<0.0001 and p<0.001, at 1 and 5 years respectively).

There were no statistically significant differences between different age groups in our cohort. However, there was a trend to high mortality within the first 6 months post-HSCT in the 0 to 3-month age group, with an OS of 68% at 6 months, which fell to 58% by 5 years. Conversely the 10 to 17-year-olds and the 1 to 5-year-olds had better OS post-HSCT of 71% and 75% respectively.

Conclusion

This is the largest national retrospective study examining the relationship between primary HLH and timing of HSCT reported since HLH-2004. However, despite the long time that has elapsed, no discernible improvement was found in overall survival for children diagnosed with primary HLH. On closer inspection of the cohort, age had no statistically significant impact on OS. However, importantly, disease status and history of relapse prior to HSCT were significant determinants in OS. This highlights the need to achieve remission rapidly and to move to HSCT in that window. We are now working with other international centres to examine a validation cohort and investigate HSCT outcomes in more detail.