Patients (pts) with Optimized Hemophagocytic Lymphohistiocytosis (HLH) Inflammatory (OHI) Index-Confirmed Diagnosis of Malignancy-Associated HLH (mHLH) and Emapalumab Treatment

Introduction: mHLH has historically been diagnosed using the HLH-2004 criteria, but clinical manifestations of the underlying malignancy (esp. hematologic malignancies [HMs]), rather than an independent aberrant inflammatory state, may result in these criteria being met. Moreover, these criteria were developed for children with primary (familial) HLH or derived from pts with underrepresented HMs. The OHI index (soluble CD25 [sCD25]) >3900 U/mL and ferritin >1000 ng/mL) offers alternate diagnostic criteria, although these criteria have not yet been validated (Zoref-Lorenz A, et al. Blood 2022;139:1098–1110). This analysis evaluated biomarkers associated with mHLH to determine if the OHI index can distinguish pts with mHLH from pts with HM only. Emapalumab, a fully human anti-interferon-γ (IFNγ) monoclonal antibody, was also assessed in pts with OHI index-confirmed mHLH.

Methods: A preliminary exploratory analysis pooled data from NCT03259230, an observational, case-control study investigating the use of IFNγ and IFNγ-inducible chemokines as markers of disease activity and prognosis in adult mHLH (N=38), and NCT03985423, an adaptive, interventional, open-label, single-arm, multicenter proof-of-concept study investigating the use of emapalumab in adult pts with secondary HLH (N=7). A principal component analysis (an unsupervised multivariate analysis) was performed on laboratory and inflammatory biomarker observations made at diagnosis (or baseline for pts in NCT03985423). Each biomarker was also tested for association with OHI group by t-tests, corrected for multiple testing by Benjamini-Hochberg False Discovery Rate (FDR) estimation. Biomarkers with a FDR ≤5% were selected. Missing biomarker data at baseline were imputed using first observation carried backwards. NCT03985423 was prematurely terminated for non-safety-related reasons. Investigator and per-protocol (PP) response were assessed. The PP definition of complete response (CR) was resolution of fever, splenomegaly, cytopenias, hyperferritinemia, and any coagulopathy, as well as no sustained worsening of sCD25 levels at Week 4 or end of treatment (EOT), whichever was later. A PP partial response (PR) was defined as a >50% improvement in ≥3 abnormalities assessed for a CR. Pharmacodynamic univariate analyses used linear mixed effect models with changes from baseline as dependent variables and baseline biomarker values, time, response, and time by response interaction as fixed effects, and subject as random effects.

Results: HLH-2004 criteria and the OHI index were retrospectively applied to 45 adult pts enrolled in these studies. Four diagnostic groups were defined: mHLH reported by Investigator (n=27), mHLH confirmed by OHI index (n=13), mHLH confirmed by HLH-2004 criteria (n=22), and malignancy alone (without HLH; n=20). OHI index demonstrated utility in distinguishing between mHLH and malignancy alone. sCD25, ferritin, interleukin (IL)-10, IL-6, and neopterin were the most discriminant markers. Of the 7 pts treated with emapalumab, 6 were evaluable. One pt discontinued treatment after the baseline visit and could not be assessed. One pt with mHLH administered emapalumab had an OHI –ve status at baseline and no improvement with treatment. Among the 5 pts with an OHI +ve status at baseline, 4 (80%) had a best response of investigator-assessed PR during the study, and 1 (20%) proceeded to stem cell transplant. Two (33%) pts had a PP PR. Reductions in chemokine C-X-C motif ligand 9 (CXCL9; a specific marker of IFNγ activity), ferritin and sCD25 levels from baseline were associated with both investigator-assessed and PP response. Three (60%) pts were OHI –ve at EOT. OHI index status evolution from +ve to –ve was closely associated with investigator-assessed response.

Conclusions: Pts with OHI index-confirmed mHLH can be distinguished from pts with malignancy alone by their biomarker signature. This may assist in identifying pts with mHLH for clinical trial purposes, but prospective statistical analysis with a larger sample size, is needed. Investigator assessment of response to emapalumab correlated well with improvements in CXCL9, ferritin, and sCD25 with complete normalization in some cases. Evolution of OHI index corroborated well with investigator-assessed response, suggesting that OHI index status may provide an alternative objective measure for assessing treatment effect in adults with mHLH.