Modern Pathology (2014) 27, 1447–1454 & 2014 USCAP, Inc. All rights reserved 0893-3952/14 $32.00 1447 High concordance in grading reticulin fibrosis and cellularity in patients with myeloproliferative neoplasms Olga Pozdnyakova1, Kaida Wu2, Abhay Patki2,5, Scott J Rodig1, Juergen Thiele3 and Robert P Hasserjian4 1Department of Pathology, Brigham and Women’s Hospital, Boston, MA, USA; 2Sanofi Oncology, Cambridge, MA, USA; 3University of Cologne, Cologne, Germany and 4Department of Pathology, WRN244, Massachusetts General Hospital, Boston, MA, USA The myeloproliferative neoplasms primary myelofibrosis, polycythemia vera, and, rarely, essential thrombo- cythemia are characterized by variable degrees of bone marrow fibrosis, either at presentation or upon progression. The increasing use of emerging therapies that may alter disease biology and morphology demands accurate and reproducible assessment of fibrosis grade. To assess concordance of hematopoietic cellularity and fibrosis grading, three hematopathologists independently evaluated a total of 728 bone marrow biopsies from 261 patients with myeloproliferative neoplasms on three clinical trials using fedratinib (SAR302503), a JAK2 inhibitor, including 249 taken at baseline and 479 on therapy. Concordance between the pathologists was evaluated by Pearson correlation coefficient (cellularity) and unweighted kappa statistic (fibrosis grade). There was high correlation of cellularity assessment (r ¼ 0.92) and fibrosis grading (kappa ¼ 0.83) between the three pathologists. Concordance with World Health Organization (WHO) grade 3 samples was higher compared with grades 0, 1, and 2. Concordance of fibrosis grading in pretreatment samples was superior to that of post-treatment samples (kappa ¼ 0.83 and 0.79, respectively, P ¼ 0.023). Our analysis suggests that the updated 2008 WHO reticulin fibrosis grading system is highly reproducible, even in patients undergoing JAK2 inhibitor therapy. This system is practically applicable to establish baseline fibrosis grade as well as changes in fibrosis in subsequent samples on therapy. Modern Pathology (2014) 27, 1447–1454; doi:10.1038/modpathol.2014.69; published online 25 April 2014 Keywords: anti-JAK2 therapy; bone marrow; collagen deposition; hematopoietic cellularity; inter-rater agreement; reticulin fibrosis Primary myelofibrosis, essential thrombocythemia, Bone marrow fibrosis is measured by reticulin and and polycythemia vera are myeloproliferative trichrome stains of the bone marrow trephine biopsy neoplasms that are associated with mutation in and is graded on a semiquantitative scale.13 The the JAK2 gene and, less often, with mutations in classification proposed by the World Health MPL or CALR genes.1–5 These myeloproliferative Organization (WHO) and updated in 2008 employs neoplasms have highly variable propensities to the European Consensus System to grade bone either present with or to develop progressive bone marrow fibrosis, which has four possible grades (0, marrow fibrosis: this fibrosing process is intrinsic to 1, 2, and 3); distinction between these levels is primary myelofibrosis and affects a significant sub- based on the number, density, and thickness of set of polycythemia vera patients and a very small silver-stained reticulin fibers as assessed by the subset of essential thrombocythemia patients.6–12 pathologist (Table 1).14,15 Previous systems used to grade bone marrow fibrosis included the Manoharan and Bauermeister schemes.13,14 Correspondence: Dr RP Hasserjian, MD, Department of Pathology, Reticulin grade is not included among the factors WRN244, Massachusetts General Hospital, 55 Fruit Street, used in primary myelofibrosis risk stratifica- Boston, MA 02114, USA. tion systems, such as the Dynamic International E-mail: [email protected] 16,17 5Current address: Blueprint Medicines, Cambridge, MA, USA. Prognostic Scoring System and shows variable Received 12 February 2014; revised 12 March 2014; accepted 13 correlation with clinical parameters of disease pro- March 2014; published online 25 April 2014 gression, such as splenomegaly. However, increased www.modernpathology.org Fibrosis/cellularity grading in bone marrow 1448 O Pozdnyakova et al reticulin grade is an adverse prognostic factor in Materials and methods primary myelofibrosis, essential thrombocythemia, and polycythemia vera.18–21 The level of reticulin Patients fibrosis is included among the criteria used to The study included 261 patients with myeloproli- establish a diagnosis of post-polycythemic myelo- ferative neoplasms on three fedratinib (SAR302503) fibrosis and post-thrombocythemic myelofibrosis, trials: one Phase 1/2 trial (38 patients), one Phase 2 which represent clinically progressed stages of trial (26 patients), and one Phase 3 trial (197 polycythemia vera and essential thrombocythemia, patients). The diagnoses were primary myelofibrosis 22 respectively. Moreover, bone marrow fibrosis is in 120 patients (46%), post-polycythemia myelofi- a dynamic process and has been shown to resolve brosis in 54 patients (21%), and post-thrombocythe- over time following eradication of the neoplastic mic myelofibrosis in 19 patients (7%). In 68 patients clone by allogeneic stem cell transplantation in (26%), the myeloproliferative neoplasm subtype 23–25 primary myelofibrosis. Regression of bone before study entry was not available. Before enroll- marrow fibrosis is also associated with successful ment on the studies, 70% of patients had been treatment of other diseases associated with bone treated with hydroxyurea and 5% of patients had marrow fibrosis, such as autoimmune myelofibrosis received other cytotoxic agents. A total of 728 bone 26–28 and chronic myelogenous leukemia. marrow biopsies were evaluated, with 249 biopsies The JAK1/JAK2 inhibitor ruxolitinib is a novel taken at baseline (pre-fedratinib) and 479 biopsies agent that has been shown to ameliorate clinical taken on therapy. symptoms and splenomegaly and prolong survival in patients with primary myelofibrosis, post-throm- bocythemic myelofibrosis and post-polycythemic Bone Marrow Examination myelofibrosis.29–35 Although prior therapies used to treat primary myelofibrosis (hydroxyurea Bone marrow biopsies were examined for hemato- and interferon) have shown little or no effect on poietic cellularity and fibrosis grade based on the bone marrow fibrosis grade,36,37 treatment with hematoxylin and eosin, silver impregnation reticu- ruxolitinib has been associated with reduction lin, and trichrome stains. For the Phase 3 and Phase and even complete resolution of bone marrow 2 studies, hematoxylin and eosin, reticulin (Gordon fibrosis.38,39 The selective JAK2 inhibitor fedratinib and Sweet method), and trichrome (Masson has also shown efficacy in treating patients with Wiegert Hematoxylin and Biberich) stains with primary myelofibrosis, post-thrombocythemic were performed at the Covance Central Laboratory myelofibrosis, and post-polycythemic myelofibrosis (Indianapolis, IN, USA) according to the manufac- and similarly shows evidence of reduced reticulin turer’s protocols. For the Phase 1/2 study, hema- fibrosis in treated patients.40 This accumulating toxylin and eosin, reticulin, and trichrome stains evidence suggests that reticulin grade may repre- were performed locally at different US institutions sent an additional parameter to measure treat- where the biopsies had originated. Trichrome stains ment response, in conjunction with existing were available for 659 biopsies (90.5% of total clinical and biological parameters such as spleno- biopsies evaluated). Hematopoietic cellularity inclu- megaly, symptoms, peripheral counts, and JAK2 ded only cells of the myeloid, erythroid, mega- allele burden.30,31,34,35,41 Indeed, a reticulin fibrosis karyocytic, and lymphoid lineages; stromal cells grade of r1 (as well as normal age-adjusted normo- were explicitly excluded from the cellularity assess- cellularity for primary myelofibrosis and poly- ment. Hematopoietic cellularity was scored in 10% cythemia vera) is included among required criteria increments from 0 to 100%. Reticulin fibrosis was for a complete response to therapy in primary graded according to the 2008 updated WHO classi- myelofibrosis, essential thrombocythemia, and fication guidelines, from grades 0 to 3 (Table 1). polycythemia vera patients according to the As recommended, reticulin fibers were graded only recently published European LeukemiaNet guide- in areas of hematopoiesis. On the basis of agreement lines.42,43 However, there has been some controversy between the three hematopathologists before slide as to the reproducibility of reticulin grading and its review, biopsies showing heterogeneous morpho- applicability to routine diagnosis setting outside of a logy with variability in reticulin fibers were graded clinical trial setting.44–48 In order to address this according to the predominant reticulin grade issue, we examined the concordance of reticulin (450% of hematopoietic area). For example, if a grading in a large number of patients with primary 60% of the biopsy demonstrated grade 1 and 40% of myelofibrosis, post-thrombocythemic myelofibrosis, the biopsy showed grade 2, the final assigned grade and post-polycythemic myelofibrosis, in the context was grade 1. of clinical fedratinib trials. Our goal was to The three hematopathologists (OP, JT, and RH) determine the reproducibility of the WHO-adopted examined the same slide under a multi-headed reticulin grading system in patients before and on microscope and independently