Reticulin-Free Quantitation of Bone Marrow Fibrosis in MPNs: Utility and Applications

Area of Science:

• Hematology
• Computational Pathology
• Medical Diagnostics

Background:

• Accurate assessment of marrow fibrosis is crucial for diagnosing and managing myeloproliferative neoplasms (MPNs).
• Current methods relying on reticulin staining face technical challenges and inter-institutional variability.
• Automated quantitation of fibrosis could significantly improve diagnostic consistency and efficiency.

Purpose of the Study:

• To develop and validate a machine learning model for quantitative fibrosis assessment.
• To utilize routinely processed H&E-stained bone marrow trephine sections for fibrosis analysis.
• To compare the performance of the H&E-based model with existing reticulin-based methods.

Main Methods:

• Development of a machine learning algorithm for quantitative fibrosis analysis.
• Application of the model to H&E-stained bone marrow trephine tissue sections.
• Comparative analysis against the Continuous Indexing of Fibrosis (CIF) reticulin-stained model.

Main Results:

• The H&E-based machine learning model demonstrated comparable performance to the established reticulin-based CIF model.
• H&E staining offers advantages in tissue retention and staining characteristics.
• Quantitative fibrosis assessment directly from H&E slides is feasible and reliable.

Conclusions:

• Quantitative marrow fibrosis assessment using H&E-stained sections is a viable and effective approach.
• This method has the potential to enhance routine clinical practice and support clinical trials.
• The H&E-derived quantitative fibrosis analysis can contribute to advancements in spatial multi-omic studies.