Histopathological findings have a key role in diagnosis of primary myelofibrosis (PMF). According to the revised 2016 World Health Organization (WHO) classification of PMF,1 bone marrow (BM)

fibrosis represents a major diagnostic criteria together with abnormal megakaryocyte morphology. The European consensus2has been applied to evaluate the BM fibrosis grade in the revised

2016 WHO classification. According to the European consensus, fibrosis is graded in four levels, from grade 0 to grade 3.Moreover, PMF is further divided into prePMF (MF-0 or MF-1) and overt

PMF (MF-2 or MF-3) according to fibrosis grade.1 Although BM fibrosis is a major criteria for PMF, the fibrosis grade is not incorporated in conventional prognostic scoring systems.

at diagnosis and 95 (28.8%) at referral. The median time between the original diagnosis and the referral was 36 (5–132) months. No patient had received hematopoietic stem cell

transplantation. All patients had a high-quality biopsies collected at diagnosis or referral and gave informed consent compliant with the Declaration of Helsinki. All cases were blind

re-reviewed by two experienced pathologists and reclassified based on the revised 2016 WHO classification. Dynamic International Prognostic Scoring System (DIPSS)6 were calculated as

described. One hundred and ninety-five patients had evaluable cytogenetic results. According to DIPSS-plus,7 karyotypes were classified as the favorable and the unfavorable. _JAK2_, _CALR_

and _MPL_ mutations were tested at diagnosis as described.8 Follow-up data were available for 301 patients, and the median follow-up was 39 (1–255) months. Correlations between sample groups

and clinical and laboratory data were calculated with the χ2 test for qualitative variables with discrete categories and Mann–Whitney _U_-test or Kruskal–Wallis analysis of variance for

continuous variables. Survival distribution was estimated by the Kaplan–Meier method and was compared between subgroups using the log-rank test. The Cox proportional hazards regression model

was used to assess the correlation between variables and survival. Two-tailed _P_-values⩽0.05 were considered significant. In 330 patients, 75 (22.7%) were categorized as DIPPS low-risk

group, 154 (46.7%) intermediate-1-risk group, 93 (28.2%) intermediate-2-risk group and 8 (2.4%) high-risk group. _JAK2_ mutations were detected in 162 subjects (49.1%), _CALR_ mutations in

65 (19.7%), _MPL_ mutations in 8 (2.4%) and triple-negative (no detectable mutation in _JAK2_, _CALR_ or _MPL_) in 95 (28.8%). According to the European consensus, 14 (4.2%) had MF-0, 93

(28.2%) MF-1, 165 (50%) MF-2 and 58 (17.6%) MF-3. Compared with the patients with MF-0 or MF-1, patients with MF-2 or MF-3 were older (_P_=0.014), had more frequent hemoglobin concentrations

<100 g/l (_P_<0.001), less frequent WBC levels >25 × 109/l (_P_=0.028), more frequent platelet levels <100 × 109/l (_P_=0.017), higher DIPSS scores (_P_<0.001) and more

frequent unfavorable karyotype according to DIPSS-plus (_P_=0.017). However, the fibrosis grade was not associated with the size of splenomegaly and driver mutations. There were more

patients with MF-2 or MF-3 who died at last follow-up than patients with MF-0 or MF-1 (31.2% versus 13.1%; _P_<0.001). Supplementary Tables S1 and S2 list baseline clinical and laboratory

variables of the 330 study subjects categorized by BM fibrosis grade. In univariate analysis, patients with higher fibrosis grade had shorter overall survival (OS) (_P_=0.013, Supplementary

Figure S1A). Patients with overt fibrosis (MF-2 or MF-3) had significantly shorter OS compared with subjects with prefibrosis (MF-0 or MF-1) (_P_=0.001, Supplementary Figure S1B). Moreover,

DIPSS variables (_P_<0.0001), no palpable splenomegaly (_P_=0.004), thrombocytopenia (_P_<0.001) and CALR-type-2 or triple-negative mutation (_P_<0.001) were associated with

reduced OS. In the lower-risk DIPSS group (low- and intermediate-1-risk group), MF-2 or MF-3 identified patients with shorter OS compared with MF-0 or MF-1 (_P_=0.014, Supplementary Figure

S1C) while for patients in the higher-risk group (intermediate-2- and high-risk group), fibrosis grade had no impact on OS (Supplementary Figure S1D). In multivariable Cox proportional

hazard regression analysis (Table 1), MF-2 or MF-3 remained significant for OS (hazard ratio (HR): 2.51, 95% confidence interval (CI): 1.37–4.59; _P_=0.003) together with DIPSS variables

(HR: 2.40, 95% CI: 1.64–3.51; _P_<0.001), no palpable splenomegaly (HR: 1,72, 95% CI: 1.03–2.86; _P_=0.036), thrombocytopenia (HR: 2.65, 95% CI: 1.62–4.34; _P_<0.001) and CALR-type-2

or triple-negative mutation (HR: 1.82, 95% CI: 1.10–3.02; _P_=0.02). Based on these data, we developed a new prognostic model using the HRs defined in the Cox regression. We assigned each

factor a weight: (1) 2 for DIPSS high risk; (2) 1 for DIPSS intermediate-2 risk and platelets <100 × 109/l; (3) 0.5 for no splenomegaly and CALR-type-2 or triple-negative mutation.

Patients were categorized into four risk cohorts: (1) low (0–1); (2) intermediate-1 (1.5 and 2); intermediate-2 (2.5 and 3); and high (⩾3.5). One hundred and thirty-one subjects (39.7%) were

categorized into low-risk cohort, 100 (30.3%) intermediate-1-risk cohort, 68 (20.6%) intermediate-2-risk cohort and 31 (9.4%) high-risk cohort. The median OS for the four risk categories

was not reached, 240, 72 and 18 months, respectively, and the difference was highly significant (_P_<0.001; Figure 1). Compared with estimated HRs for survival in the low-risk cohort, HRs

were 2.94 (95% CI, 1.45–5.99; _P_=0.003) for the intermediate-1-risk cohort, 6.18 (95% CI, 3.05–12.52; _P_<0.001) for the intermediate-2-risk cohort and 22.70 (95% CI, 10.81–47.67;

_P_<0.001) for the high-risk cohort. Histopathological findings are major criteria for PMF together with polycythemia vera and essential thrombocythemia (ET) according to the revised 2016

WHO classification.1 Hematopoietic cellularity, granulocytic, erythrocytic and megakaryocytic proliferation, abnormal arrangement, location and morphology of megakaryocyte and reticulin

and/or collagen fibrosis are key points concerning the distinction between polycythemia vera, ET, prePMF and overt PMF.9, 10 Reproducibility and clinical usefulness of the WHO classification

to differential diagnosis for Ph− myeloproliferative dysplasia persisted to be a controversial issue in recent years. Although some studies offered some criticisms of WHO morphological

classification,11, 12 a number of clinico-pathological studies by independent working groups demonstrated that definite diagnosis could be made by strictly regarding histopathological

features according to the WHO criteria.13, 14 Compared with ET and overt PMF, prePMF has unique clinical and laboratory features and outcome.15 Therefore, discriminating prePMF from ET and

overt PMF is necessary and accurate evaluation of BM fibrosis grade is a key issue to diagnosis and prognostic evaluation for prePMF. Our study indicated that higher BM fibrosis grade was

associated with some poor prognostic characteristics, including older age, anemia, thrombocytopenia, unfavorable karyotype and a higher DIPSS risk category, but fibrosis grade was not

associated with driver mutations. Multivariable analyses confirmed that fibrosis grade was independent of DIPSS score for PMF patients, especially in the lower-risk group.Findings from this

study agreed with previous studies.3, 4, 5 This study indicated that there were obvious differences in clinical characteristics and prognosis between prePMF (MF-0 or MF-1) and overt PMF

fibrosis (MF-2 or MF-3) as currently defined by WHO. Therefore, adding fibrosis grade into the traditional prognostic scoring system is necessary to accurate evaluation of prognosis. In

conclusion, we confirmed the independent prognostic impact of fibrosis grade in PMF and the important clinical meaning of the revised 2016 WHO classification for PMF. The main limitation of

this study is the lack of validation in an independent cohort of the proposed score; ideally this score system should be validated in another data set in the future. REFERENCES * Arber DA,

Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM _et al_. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. _Blood_ 2016;

127: 2391–2405. Article  CAS  Google Scholar  * Thiele J, Kvasnicka HM, Facchetti F, Franco V, van der Walt J, Orazi A . European consensus on grading bone marrow fibrosis and assessment of

cellularity. _Haematologica_ 2005; 90: 1128–1132. PubMed  Google Scholar  * Guglielmelli P, Rotunno G, Pacilli A, Rumi E, Rosti V, Delaini F _et al_. Prognostic impact of bone marrow

fibrosis in primary myelofibrosis. A study of the AGIMM group on 490 patients. _Am J Hematol_ 2016; 91: 918–922. Article  CAS  Google Scholar  * Elliott MA, Tefferi A . The prognostic impact

of bone marrow fibrosis in primary myelofibrosis. _Am J Hematol_ 2016; 91: E454. Article  Google Scholar  * Guglielmelli P, Vannucchi AM,, AGIMM Investigators. The prognostic impact of bone

marrow fibrosis in primary myelofibrosis. _Am J Hematol_ 2016; 91: E454–E455. Article  Google Scholar  * Passamonti F, Cervantes F, Vannucchi AM, Morra E, Rumi E, Pereira A _et al_. A

dynamic prognostic model to predict survival in primary myelofibrosis: a study by the IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment). _Blood_

2010; 115: 1703–1708. Article  CAS  Google Scholar  * Gangat N, Caramazza D, Vaidya R, George G, Begna K, Schwager S _et al_. DIPSS-Plus: a refined Dynamic International PrognosticScoring

System (DIPSS) for primary myelofibrosis that incorporates prognostic information from karyotype, platelet count and transfusion status. _J Clin Oncol_ 2011; 29: 392–397. Article  Google

Scholar  * Li B, Xu J, Wang J, Gale RP, Xu Z, Cui Y _et al_. Calreticulin mutations in Chinese with primary myelofibrosis. _Haematologica_ 2014; 99: 1697–1700. Article  CAS  Google Scholar 

* Barbui T, Thiele J, Gisslinger H, Finazzi G, Vannucchi AM, Tefferi A . The 2016 revision of WHO classification of myeloproliferative neoplasms: clinical and molecular advances. _Blood Rev_

2016; 30: 453–459. Article  CAS  Google Scholar  * Barbui T, Thiele J, Vannucchi AM, Tefferi A . Problems and pitfalls regarding WHO-defined diagnosis of early/prefibrotic primary

myelofibrosis versus essential thrombocythemia. _Leukemia_ 2013; 27: 1953–1958. Article  CAS  Google Scholar  * Wilkins BS, Erber WN, Bareford D, Buck G, Wheatley K, East CL _et al_.

Bonemarrow pathology in essential thrombocythemia: interobserver reliability and utility for identifying disease subtypes. _Blood_ 2008; 111: 60–70. Article  CAS  Google Scholar  * Buhr T,

Hebeda K, Kaloutsi V, Porwit A, Van der Walt J, Kreipe H . European Bone Marrow Working Group trial on reproducibility of WHO criteria to discriminate essential thrombocythemia from

prefibrotic primary myelofibrosis. _Haematologica_ 2012; 97: 360–365. Article  Google Scholar  * Thiele J, Kvasnicka HM, Müllauer L, Buxhofer-Ausch V, Gisslinger B, Gisslinge H . Essential

thrombocythemia versus early primary myelofibrosis: a multicenter study to validate the WHO classification. _Blood_ 2011; 117: 5710–5718. Article  CAS  Google Scholar  * Pozdnyakova O, Wu K,

Patki A, Rodig SJ, Thiele J, Hasserjian RP . High concordance in grading reticulin fibrosis and cellularity in patients with myeloproliferative neoplasms. _Mod Pathol_ 2014; 27: 1447–1454.

Article  CAS  Google Scholar  * Barbui T, Thiele J, Carobbio A, Passamonti F, Rumi E, Randi ML _et al_. Disease characteristics and clinical outcome in young adults with essential

thrombocythemia versus early/prefibrotic primary myelofibrosis. _Blood_ 2012; 120: 569–571. Article  CAS  Google Scholar  Download references ACKNOWLEDGEMENTS Supported in part by National

Natural Science Funds (Nos. 81530008, 81370611, 81270585, 81470297), Program for Peking Union Scholars and Innovative Research Team, PUMC Youth Fund and Fundamental Research Funds for the

Central Universities (No. 3332016089) and Science and technology project of Tianjin (No. 15ZXLCSY00010). AUTHOR INFORMATION Author notes * B Li and P Zhang: These authors contributed equally

to this work. AUTHORS AND AFFILIATIONS * MDS and MPN Centre, Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College,

Tianjin, China B Li, G Feng, Z Xu, T Qin, Y Zhang, Z Sha, D Dong, H Zhang, L Fang, L Pan, N Hu, S Qu & Z Xiao * State Key Laboratory of Experimental Hematology, Institute of Hematology

and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China B Li, Z Xu, Y Zhang & Z Xiao * Department of Pathology, Institute of

Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China P Zhang & W Cai * Divisions of Experimental Hematology and

Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA G Huang Authors * B Li View author publications You can also search for this author inPubMed Google Scholar

* P Zhang View author publications You can also search for this author inPubMed Google Scholar * G Feng View author publications You can also search for this author inPubMed Google Scholar

* Z Xu View author publications You can also search for this author inPubMed Google Scholar * T Qin View author publications You can also search for this author inPubMed Google Scholar * Y

Zhang View author publications You can also search for this author inPubMed Google Scholar * Z Sha View author publications You can also search for this author inPubMed Google Scholar * D

Dong View author publications You can also search for this author inPubMed Google Scholar * H Zhang View author publications You can also search for this author inPubMed Google Scholar * L

Fang View author publications You can also search for this author inPubMed Google Scholar * L Pan View author publications You can also search for this author inPubMed Google Scholar * N Hu

View author publications You can also search for this author inPubMed Google Scholar * S Qu View author publications You can also search for this author inPubMed Google Scholar * W Cai View

author publications You can also search for this author inPubMed Google Scholar * G Huang View author publications You can also search for this author inPubMed Google Scholar * Z Xiao View

author publications You can also search for this author inPubMed Google Scholar CORRESPONDING AUTHOR Correspondence to Z Xiao. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no

conflict of interest. ADDITIONAL INFORMATION Supplementary Information accompanies this paper on Blood Cancer Journal website SUPPLEMENTARY INFORMATION SUPPLEMENTARY TABLES AND FIGURES (DOC

204 KB) RIGHTS AND PERMISSIONS This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in

the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain

permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ Reprints and permissions ABOUT THIS ARTICLE

CITE THIS ARTICLE Li, B., Zhang, P., Feng, G. _et al._ Bone marrow fibrosis grade is an independent risk factor for overall survival in patients with primary myelofibrosis. _Blood Cancer

Journal_ 6, e505 (2016). https://doi.org/10.1038/bcj.2016.116 Download citation * Published: 09 December 2016 * Issue Date: December 2016 * DOI: https://doi.org/10.1038/bcj.2016.116 SHARE

THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to

clipboard Provided by the Springer Nature SharedIt content-sharing initiative