Introduction

Hemophagocytic lymphohistiocytosis (HLH) is an inflammatory syndrome that can occur de novo or due to the presence of an underlying malignancy. Prior studies have shown the IL2/ferritin ratio is a simple predictive tool in determining HLH associated with lymphoma. Two studies from Japan demonstrated a ratio >2 has 95.6% positive predictive value for an underlying lymphoma diagnosis. We aimed to validate the ratio as a diagnostic tool for lymphoma associated HLH across a heterogenous, Western population as well as investigate for other novel laboratory predictors of LAHS.

Methods

This was a multi-institutional retrospective chart review of patients with an ICD-10 code diagnosis of HLH from 2008-2024. Data obtained included patient demographics, lab values, presumed etiology of HLH and presence of macrophages in the bone marrow. HLH was diagnosed according to the HLH-2004 diagnostic guidelines. The diagnosis of malignant lymphoma was made according to the 2008 WHO classification as data was pulled from 2008-2024. We compared the laboratory findings of the non lymphoma-associated HLH (N-LAHS) group with those of the LAHS group. Patient characteristics were compared using Fisher's exact test, chi-square test, and Mann-Whitney test. Bivariate and multivariate logistic regression models with log transformed ratios were developed.

Results

In total, 126 cases of HLH were identified, with 82 classified as N-LAHS and 44 LAHS. The median age was 55 and 45% were female. Overall, 55.2% were White, 36.8% Black, 2.4% Asian and 5.6 % Hispanic. Black patients were more likely to have N-LAHS (p=0.021). Causes of N-LAHS were primary HLH (1), HLH due to a rheumatologic cause (14), Infectious etiologies (31) and 36 were classified as “other” which included unknown causes for HLH, other hematolymphoid causes for HLH and cellular therapy/immune checkpoint inhibitor associated HLH. Lymphoma subtypes included Aggressive B-cell lymphomas (15), Aggressive T-cell lymphomas (9), Indolent B-cell lymphomas (3) and Unspecified (17).

Patients with LAHS had a median IL2 level of 25,620 (range 1,394 – 266,968) compared to N-LAHS of 8061 (859 – 85840) (p=<0.001). The median ferritin values were 11,704 (695 – 152,384) for LAHS and 15,000 (1,139 – 320,000) for N-LAHS (p=0.147). The IL2/ferritin ratio median for LAHS was 2.2 (0.1 – 36.4) and 0.4 (0.0 – 56.6) for N-LAHS (p=<0.001).

The IL-2/(ALT/AST) ratio for LAHS was 10,255.9 (265.4 – 174,263.9) and 5,277.9 (319.0 – 133,760.8) for N-LAHS (p=<0.001). Bivariate logistic regression models for log transformed sIL2R/ferritin and sIL-2R/(ALT/AST) ratios were developed.

The model revealed that 1-unit increase in the log(sIL2R/ferritin ratio) was associated with a 74% increase in the odds of having lymphoma (OR: 1.74, 95% CI: 1.33–2.29). Therefore, a doubling in the sIL2R/ferritin ratio increased the odds of lymphoma by 47%. When the predictor was categorized as > or < 2, the OR showed that patients with an IL2/ferritin ratio >2 had a nearly 7 times greater odds of having lymphoma compared to those with a ratio <2.

Each 1-unit increase in the log(sIL-2R/(ALT/AST) ratio) was associated with a 90% increase in the odds of having lymphoma (OR: 1.90, 95% CI: 1.38–2.61). Therefore, a doubling in the sIL-2R/(ALT/AST) ratio increased the odds of lymphoma by 56%.

Conclusions

Our study validated the significance of IL2/ferritin ratio as a differentiator for the diagnosis of LAHS and N-LAHS within a Western population with different disease and population demographics. Our group utilized a logistic regression model to predict the outcome of lymphoma based on the value of the Il2/ferritin ratio, rather than the described distribution of the outcome across the >2 / <2 groups in our sample. This was done intentionally to predict the risk in an individual rather than a population. Taken together, these findings suggest that the IL2/ferritin ratio has sufficient diagnostic accuracy in predicting LAHS in a Western population. Validation of the IL2/ferritin ratio suggests integrating it into the diagnostic algorithm for LAHS. Notably, our findings also suggest that the IL2/ (ALT/AST) ratio is potentially a new biomarker that may assist in predicting LAHS.

This content is only available as a PDF.
2025
Sign in via your Institution