An optimized hemophagocytic lymphohistiocytosis (HLH) inflammatory (OHI) index may be able to accurately identify malignancy-associated HLH and risk of mortality in patients with hematologic malignancies, according to research findings published in Blood. The OHI index assesses soluble interleukin-2 receptor (sCD25) and ferritin levels, which researchers noted should be considered early in various hematologic malignancies.
“The OHI index can be utilized to more accurately and quickly establish the diagnosis of HLH and high mortality risk in patients with hematologic malignancies,” said corresponding study author Michael Jordan, MD, of Cincinnati Children’s Hospital Medical Center in Ohio.
Dr. Jordan explained that the early recognition of HLH is a prerequisite to patients receiving effective treatment, which is critical given the poor natural history of the syndrome. “While the optimal treatment for HLH remains to be defined, we believe that improving the diagnostic process [with the OHI] would enable the development and testing of better therapeutic strategies,” he said.
The study authors wrote that HLH is frequently defined based on enrollment criteria from the HLH-2004 study, which were developed to diagnose pediatric familial HLH. Despite the HLH-2004 diagnostic criteria’s usefulness for this condition, there is a lack of data on how well the criteria can accurately diagnose HLH in a broader group of hematologic malignancies.
The multicenter, retrospective study by Dr. Jordan and colleagues included 225 adult patients with hematologic malignancies from Israel, the U.S., and Japan. In all patients, investigators measured sCD25 because of either suspected HLH or as part of routine surveillance.
A total of 112 patients who fulfilled up to five out of eight diagnostic criteria from the HLH-2004 pediatric study were classified as having HLH, while 113 patients who met fewer than five of the criteria were classified as having hematologic malignancies. To identify useful diagnostic and prognostic parameters, the researchers used classification and regression tree (CART) and receiver operating curve analysis.
In the overall cohort, documented hematologic malignancies were lymphoid malignancies such as B-cell lymphoma (40%), T-cell/natural killer cell lymphoma (26%), Hodgkin lymphoma (8%), and chronic lymphocytic leukemia (4%), as well as myeloid malignancies such as acute myeloid leukemia (8%), myelodysplastic syndromes (8%), and myeloproliferative neoplasm (6%). Patients with hematologic malignancies and HLH were significantly younger than patients with only hematologic malignancies (median age = 62.5 vs. 66 years, respectively; p=0.0002).
The researchers found that there was “extensive overlap” of HLH-2004 parameters between patients with only hematologic malignancies and those with HLH. Inflammatory markers sCD25 and ferritin were the only parameters that differed significantly between groups and rose above diagnostic thresholds.
A CART analysis with all parameters at initial presentation was used to identify important markers for HLH-2004 diagnosis. The analysis used initial sCD25 >3,900 U/mL and ferritin >2,636 ng/mL to identify most patients with hematologic malignancies and HLH. The developed OHI index, comprising a combination of sCD25 >3,900 U/mL and ferritin >1,000 ng/mL, improved diagnostic prediction of HLH-2004 with a sensitivity of 84% and a specificity of 81%. Additionally, the OHI index improved prediction of prognosis at presentation compared with the HLH-2004 or an “optimized” version of the HLH-2004.
The investigators explained that the OHI index was also better for identifying patients at high risk of poor outcomes. In an analysis that compared the OHI index with the HLH-2004 without sCD25, the latter tool missed a larger high-risk population than the former, emphasizing the importance of sCD25 assessment.
In addition to the primary findings, the investigators noted that many of the defining features for diagnosing HLH overlap with typical laboratory abnormalities seen in patients with uncomplicated hematologic malignancies.
“Perhaps due to this overlap, the inclusion of parameters other than sCD25 and ferritin actually dilutes the definition of ‘HLH’ and impedes recognition of a very high-risk inflammatory syndrome that we currently recognize due to its similarity to familial HLH,” Dr. Jordan said.
A limitation of the study was its retrospective nature. Additionally, the researchers noted that “the risks/benefits of intensifying or altering therapy based on the OHI [index] is still unknown and warrants further study.”
To address these limitations, Dr. Jordan and colleagues noted that they have started a prospective study to validate both the diagnostic and prognostic power of the OHI index in patients with hematologic malignancies. l
The authors report no relevant conflicts of interest.
Reference
Zoref-Lorenz A, Murakami J, Hofstetter L, et al. An improved index for diagnosis and mortality prediction in malignancy associated hemophagocytic lymphohistiocytosis [published online ahead of print, 2021 Nov 15). Blood. doi: 10.1182/blood.2021012764.
