Objective Quantification of Osteosclerosis by Area Pixel Count of Bone Marrow Biopsy as a Diagnostic Tool in Myeloproliferative Neoplasms
CJ Teman, AR Wilson, SL Perkins, JT Prchal, M Salama. University of Utah, Salt Lake City, UT; ARUP Laboratories, Salt Lake City, UT
Background: Myeloproliferative neoplasms (MPN) are a heterogeneous disease group with overlapping clinical and histological features, leading to frequent diagnostic difficulties. Fibrosis can occur in late stages of several MPN, including primary myelofibrosis (PM), essential thrombocythemia (ET), chronic myelogenous leukemia (CML) and polycythemia vera (PV). Osteosclerosis frequently accompanies fibrosis, but has not achieved widespread recognition in diagnostic algorithms of MPN, reflecting a lack of objective quantification methods. We propose a method of osteosclerosis quantification in marrow cores, and hypothesize that trabecular volume (TV) can be used as a diagnostic criterion to differentiate MPN.
Design: The study group included 68 patients with established MPN, including PM (n=24), ET (n=21), PV (n=15), and untreated CML (n=8). The control group consisted of 47 negative lymphoma staging marrows in age and gender-matched patients. Biopsies were digitally scanned with the ScanScope XT system (Aperio, Vista, CA). The entire hematopoietic area and each individual bony trabecula were circled manually; and areas of cortical bone, fragmentation, and crush artifact were excluded. Areas were quantitated using the ImageScope software's pixel count algorithm, and trabecular area was divided by total area to calculate the TV.
Results: The control group had an average TV of 15.7 4.7%. The categories of MPN with higher TV included PM (33.8 11.4%, p<0.0001), myelofibrosis secondary to PV (29.7 8.5%, p=0.005), and myelofibrosis secondary to ET (24.8 11.7%, p=0.037). PV without fibrosis (20.0 7.1%), ET without fibrosis (21.6 8.7%), and CML (18.9 5.9%) had greater TV than the control group, but not to a statistically significant degree. Although the PM group had the greatest overall TV, there was no statistically significant difference when compared to groups with myelofibrosis secondary to either ET (p=0.055) or PV (p=0.32). However, PM displayed a striking increase in TV compared to prefibrotic PV (p=0.0007), prefibrotic ET (p=0.0018), and CML (p<0.0001).
Conclusions: These findings suggest that trabecular volume can serve as an objective tool for quantification of osteosclerosis in MPN, and that the degree of osteosclerosis can help differentiate PM from other MPN.
Wednesday, March 11, 2009 1:00 PM
Poster Session VI # 200, Wednesday Afternoon