The HLH-Risk-Calculator is a machine learning-based tool to predict course & mortality of secondary hemophagocytic lymphohistiocytosis - Report - MDSpire

The HLH-Risk-Calculator is a machine learning-based tool to predict course & mortality of secondary hemophagocytic lymphohistiocytosis

  • By

  • Michael Ruzicka

  • Hans Christian Stubbe

  • Josia Fauser

  • Manuel Trebo

  • Thomas Wimmer

  • Lena Horvath

  • Hans-Joachim Stemmler

  • Stefanie Susanne Stecher

  • Hendrik Schulze-Koops

  • Fabian Hauck

  • Michael Medinger

  • Claire Seydoux

  • Michael Starck

  • Clemens-Martin Wendtner

  • Peter Bojko

  • Marcus Hentrich

  • Katharina Elisabeth Nickel

  • Katharina Goetze

  • Florian Bassermann

  • Sabine Janina Ehrlich

  • Marion Subklewe

  • Andreas Pircher

  • Dominik Wolf

  • Michael von Bergwelt-Baildon

  • Karsten Spiekermann

  • June 30, 2026

  • 0 min

Share

Clinical Report: A Machine Learning Tool for Predicting Disease Progression

Overview

The HLH-Risk-Calculator predicts initial disease severity and mortality in adult patients with secondary hemophagocytic lymphohistiocytosis (sHLH). Key predictors of adverse outcomes include low platelet counts and high levels of soluble interleukin-2 receptor.

Background

Secondary hemophagocytic lymphohistiocytosis is a rare but life-threatening syndrome characterized by hyperinflammation, primarily affecting adults. Accurate and timely diagnosis is crucial due to high mortality rates, which range from 26% to 75% according to various studies. Current diagnostic frameworks lack robust tools for early identification of high-risk patients and prognosis estimation.

Data Highlights

The HLH-Risk-Calculator was developed using data from 167 patients diagnosed with sHLH, incorporating laboratory and demographic features to predict outcomes.

Key Findings

  • The HLH-Risk-Calculator utilizes six Random Forest models to predict disease severity and mortality.
  • Low platelet counts and high soluble interleukin-2 receptor levels are strong predictors of adverse outcomes.
  • External validation of the HLH-Risk-Calculator is necessary before clinical implementation.

Clinical Implications

Further validation of the HLH-Risk-Calculator is necessary before it can be used in clinical settings.

Conclusion

The development of the HLH-Risk-Calculator requires further validation for clinical use.

Related Resources & Content

  1. Frontiers in Pediatrics, 2026 -- Integration of Boruta algorithm and latent class analysis for risk factors of 30-day mortality in pediatric hemophagocytic lymphohistiocytosis based on peripheral blood indicators
  2. the asco post, 2026 -- Machine-Learning Model for HCC Risk Prediction May Outperform Current Methods
  3. the asco post, 2025 -- New Model Predicts Risk of Progression in Early-Stage Classical Hodgkin Lymphoma
  4. Frontiers in Oncology, 2026 -- Machine learning prediction of 30-day all-cause mortality risk factors in HCC rupture
  5. Multicenter validation of secondary hemophagocytic lymphohistiocytosis diagnostic criteria - PMC
  6. A Risk Model Based on sCD25 for Early Mortality in Adult Patients with Hemophagocytic Lymphohistiocytosis - PMC
  7. Multicenter validation of secondary hemophagocytic lymphohistiocytosis diagnostic criteria - PMC
  8. A Risk Model Based on sCD25 for Early Mortality in Adult Patients with Hemophagocytic Lymphohistiocytosis - PMC
  9. Assessing the effectiveness of etoposide treatment in adult haemophagocytic lymphohistiocytosis: a systematic review and meta-analysis - PMC

Original Source(s)

The HLH-Risk-Calculator is a machine learning-based tool to predict course & mortality of secondary hemophagocytic lymphohistiocytosis

  • by Michael Ruzicka, Hans Christian Stubbe, Josia Fauser, Manuel Trebo, Thomas Wimmer, Lena Horvath, Hans-Joachim Stemmler, Stefanie Susanne Stecher, Hendrik Schulze-Koops, Fabian Hauck, Michael Medinger, Claire Seydoux, Michael Starck, Clemens-Martin Wendtner, Peter Bojko, Marcus Hentrich, Katharina Elisabeth Nickel, Katharina Goetze, Florian Bassermann, Sabine Janina Ehrlich, Marion Subklewe, Andreas Pircher, Dominik Wolf, Michael von Bergwelt-Baildon, Karsten Spiekermann

  • June 30, 2026

Related Content

Immunological and clinical characteristics of severe thrombocytopenia in neonates with Kasabach–Merritt phenomenon

Factors associated with pulmonary embolism in children with refractory Mycoplasma pneumoniae pneumonia and elevated D-dimer

The value of speckle tracking echocardiography in early identification of subclinical cardiac involvement in patients with multiple myeloma: a pilot retrospective study