Improved survival outcomes with anakinra over etoposide-based therapies for the management of adults with hemophagocytic lymphohistiocytosis: a retrospective multicenter research network study

Background: Hemophagocytic lymphohistiocytosis (HLH) is a rare life-threatening, hyperinflammatory syndrome for which etoposide-based regimens have historically been the standard of care. Recent reports have described positive outcomes with the utilization of ruxolitinib or anakinra although these studies are often limited to small samples. Objectives: We aimed to compare the efficacy of ruxolitinib, anakinra, and etoposide-based therapies for the management of HLH in adult patients. Design: We performed a population-based, multicenter, retrospective cohort study utilizing the TriNetX Networks database. Methods: Adult patients (⩾18 years) diagnosed with HLH who received first-line treatment with ruxolitinib, anakinra, or etoposide between 2008 and 2023 were analyzed. The primary endpoint was overall survival (OS) at 1 year. A 1:1 propensity-score matching analysis was utilized. Results: Anakinra (p = 0.020) but not ruxolitinib (p = 0.19) resulted in a significantly higher 1-year OS when compared with etoposide-based therapies. Conclusions: Anakinra is effective for the management of adult patients with HLH.


Introduction
Hemophagocytic lymphohistiocytosis (HLH) is a rare life-threatening, hyperinflammatory syndrome that is commonly managed with an etoposide-based regimen such as the HLH-94 protocol. 1Although inciting factors for the development of HLH are widely heterogeneous, with the most common triggers identified as hematologic malignancy, infection, or autoimmune disease, clinical outcomes overall remain poor.Recent reports have described positive outcomes with the utilization of ruxolitinib or anakinra although these publications are limited to case TherapeuTic advances in hematology real-time, real-world electronic health records in an aggregate and de-identified form that aligns with the Health Insurance Portability and Accountability Act, the US Federal Law, and the EU Data Protection Law Regulation 2016/679.Accessible data for research purposes include the International Classification of Diseases, Tenth Revision (ICD-10) diagnosis codes, as well as demographics, laboratory data elements, procedural data, and medication administrations.To protect the identity of patients included in the TriNetX database, all queries resulting in a total patient count of less than 10 are not reported.
We sought to assess the efficacy of ruxolitinib, anakinra, and etoposide-based therapies in the management of adults with HLH.Patients with an ICD-10 diagnosis code for HLH (D76.0) between January 2008 and December 2023 and 18-80 years of age at diagnosis were assessed for inclusion.Patients were excluded if they received a chimeric antigen receptor T-cell therapy or if they received emapalumab or tocilizumab up to 1 year after HLH diagnosis.Patients with a recent history of hematologic malignancy were defined as having a composite ICD-10 code for malignant neoplasms of lymphoid, hematopoietic, and related tissue (C81-C96), within 1 year of HLH diagnosis.The primary outcome of interest was overall survival (OS) at 1 year.To determine whether practice changes over time influenced treatment outcomes, subgroup analyses were also performed over two time periods, January 2008 to December 2015 and January 2016 to December 2023.Analyses that compared ruxolitinib with anakinra or etoposide only included those treated between January 2015 and December 2023 as ruxolitinib utilization was not found prior to 2015 in the TriNetX database.Bivariate regression analysis was performed on baseline characteristics associated with worse OS to determine variables included in the matching analysis.A comparison of 1-year OS based on identified characteristics was also performed between treatment arms.A 1:1 propensity-score matching utilizing logistic regression was performed based on age, male sex, history of hematologic malignancy, management by critical care services, and key baseline laboratory values at initial diagnosis which included: absolute neutrophil count (ANC), serum ferritin, fibrinogen, and triglycerides for the primary outcome of interest.The secondary outcome of interest was the development of secondary hematologic malignancy at 5 years among patients without a prior history of leukemia, lymphoma, myeloma, or other hematologic malignant processes.All statistical analyses were conducted utilizing the TriNetX Analytics platform.

TherapeuTic advances in hematology
Lastly, patients who received an etoposide-based therapy had a significantly higher risk of secondary hematologic malignancy at 5 years compared to those who received anakinra (HR 4.25; 95% CI 1.82-9.95;p < 0.001) (Figure 3).No ruxolitinib-treated patients developed secondary malignancy at 5 years; however, this was based on a limited sample size of 15 patients.

Discussion
The results of this large multicenter retrospective cohort study suggest that treatment with anakinra provides greater survival benefit over etoposidebased therapies for the management of adult patients with HLH.Despite long-term survival outcomes in children with HLH who received etoposide-based regimens, durability of response remains poor in adults owing largely to disease relapse and organ dysfunction. 14In adult patients receiving the HLH-94 protocol, 1-year OS rates are dismal ranging between 29% and 35%. 2,15,16n addition, the high cumulative exposure to etoposide from the HLH-94 protocol has been implicated in significantly increasing the risk of secondary hematologic malignancy.
Immune system dysregulation in patients who develop HLH involves numerous cytokines such as IL-1, IL-2, IL-6, IL-10, and interferon-gamma which are signaled through the Janus kinase (JAK) and signal transducers and activators of transcription pathways. 17,18Given the lack of success with etoposide-based therapies in adults, inhibition or interruption of key cytokine pathways associated with the clinical manifestations of HLH has been a clinical area of interest.Anakinra and ruxolitinib, which target IL-1 and the JAK pathway, respectively, have emerged as promising therapeutic modalities. 19,203][4][5] Collectively, treatment response from a pooled analysis of four studies was 78.3% ][4][5] With a median follow-up of 10 months, Bavarez et al. also reported an OS of 88.2% in their subset of adult patients (n = 18) who received anakinra (one patient with prior etoposide treatment was excluded). 3These findings were similar to our analysis in which 1-year OS was 79.3% in the unmatched cohort.With a significantly lower rate of secondary hematologic malignancy and higher short-term survival over etoposide, our findings indicate that anakinra is a highly effective therapy for adults with HLH.Specifically, among patients who develop HLH during pregnancy or require treatment during the peripartum period, anakinra has also been demonstrated to offer a safer therapeutic option. 21th the first reported case of ruxolitinib use for HLH treatment in 2017, 12 published literature investigating ruxolitinib utility is sparse, predominantly seen for relapsed/refractory disease in pediatric patients.][8][9][10][11] With a complete response rate of 71.4% (n = 5/7), ruxolitinib monotherapy appears promising albeit seen with a small sample size.Survival data are also limited to one study with case reports by Acosta et al., 6 Ahmed et al., 7 Slostad et al., 8 and Zandvakili et al. 9 only describing initial response rates.Although 1-year OS was not statistically lower for ruxolitinib-treated patients in our cohort compared to etoposide-based therapies, improvement in survival was evident.In two upfront combination studies using ruxolitinib with etoposide and ruxolitinib, dexamethasone, and etoposide, with doxorubicin for adults with HLH, high response rates were seen suggesting a potential utility for initial therapy. 10,11is study has several limitations to acknowledge.First, although we utilized the propensityscore method to balance patient characteristics, statistically significant differences in several baseline variables were seen in the anakinra versus etoposide cohort.In addition, confounders such as end-organ dysfunction at the time of treatment and underlying HLH triggers were present.Second, similar to other large database studies, patient-level data are not available and factors such as treatment dosages (i.e.low-dose versus high-dose ruxolitinib) and underlying etiologies remain unknown.Although infectious and autoimmune triggers could not be collated due to their broad indications, we were able to balance patients with hematologic malignancies, which have been implicated as a significant source of mortality in HLH.Lastly, validation through diagnostic and prognostic tools such as the HLH-2004 criteria and H-Score could not be determined.

Conclusion
This multicenter, retrospective electronic health cohort study indicates that anakinra is highly effective for the management of adult patients with HLH.Anakinra usage was associated with significantly lower development of secondary hematologic malignancy and higher 1-year OS when compared to etoposide-based therapies.
Further studies are warranted to validate these findings.

Ethics approval and consent to participate
The study was performed using the TriNetX database which uses de-identified aggregate patient data for analysis.No protected health information or personal data are made available to the users of the platform.As a result, institutional review board approval was not required for this study.

Figure 1 .
Figure 1.Overall survival at 1 year between (a) all patients who received RUX, ANAK, or ETOP-based regimens, (b) patients who received RUX and ETOP after the propensity-score match, (c) patients who received ANAK and ETOP after the propensity-score match, and (d) patients who received RUX and ANAK after the propensityscore match.ANAK, anakinra; ETOP, etoposide; HLH, hemophagocytic lymphohistiocytosis; RUX, ruxolitinib.

Figure 2 .
Figure 2. Forest plot of characteristics associated with worse 1-year OS by treatment.ANC, absolute neutrophil count; ICU, intensive care unit; OS, overall survival.

Table 1 .
Baseline demographics and characteristics after propensity-score matching.
ANAK, anakinra; ANC, absolute neutrophil count; ETOP, etoposide-based therapy; ICD-10, International Classification of Diseases, Tenth Revision; IL, interleukin; NR, not reported; RUX, ruxolitinib.a Variables that are significant with a p < 0.05 are bolded.b Reported as mean ± SD unless otherwise stated.c Indicated by composite ICD-10 code for malignant neoplasms of lymphoid, hematopoietic, and related tissue (C81-C96).d Queries resulting in less than 10 findings are not reported by the TriNetX.