Revised response criteria for myelofibrosis: International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) and European LeukemiaNet (ELN) consensus report

Ayalew Tefferi, Francisco Cervantes, Ruben Mesa, Francesco Passamonti, Srdan Verstovsek, Alessandro M. Vannucchi, Jason Gotlib, Brigitte Dupriez, Animesh Pardanani, Claire Harrison, Ronald Hoffman, Heinz Gisslinger, Nicolaus Kröger, Juergen Thiele, Tiziano Barbui and Giovanni Barosi

Key Points

  • Treatment response criteria for MF must capture drug benefit in terms of symptom burden.

  • The current document includes stricter definitions of red cell transfusion need and independence.


The current document is a revision of the International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) criteria for treatment response in myelofibrosis (MF) and represents a collaborative effort by the IWG-MRT and the European LeukemiaNet to objectively assess the value of new drugs in inducing morphologic remission or improvement in MF-associated symptomatic burden (MF-SB). Some of the changes in the current revision include stricter definitions of red cell transfusion dependency and independency and consideration of the Myeloproliferative Neoplasm Symptom Assessment Form as a tool to quantify meaningful changes in disease-related symptoms. Six response categories are listed: complete remission (CR) and partial remission signify treatment effects that are consistent with disease modification, whereas drug-induced improvements in MF-SB were annotated as clinical improvement, anemia response, spleen response, or symptoms response. Additional criteria are provided for progressive disease, stable disease, and relapse. The document also includes recommendations for assessing cytogenetic and molecular remissions, without mandating their inclusion for CR assignment.


Myelofibrosis (MF) is a pathological entity associated with primary MF, postpolycythemia vera MF, and postessential thrombocythemia MF.1 These diseases are characterized by clonal myeloproliferation, ineffective erythropoiesis, bone marrow stromal changes, hepatosplenic extramedullary hematopoiesis, and aberrant cytokine expression.2 At presentation, the disease characteristics of primary MF include palpable splenomegaly in 89%, constitutional symptoms in 27%, moderate-to-severe anemia (hemoglobin < 10 g/dL) in 35%, thrombocytopenia (platelets < 100 × 10[9]/L) in 17%, or marked leukocytosis (leukocytes > 25 × 10[9]/L) in 10%.3 Patients with MF have shortened survival4 and greatly compromised quality of life (QoL).5 Contributing factors for shortened survival include leukemic transformation5 and thrombohemorrhagic complications6 and for the compromised quality of life severe anemia (often requiring red cell transfusions), symptomatic enlargement of the spleen and liver, substantial MF-associated symptoms burden (MF-SB), and cachexia.5 Allogeneic stem cell transplant, using conventional7 or reduced intensity8 conditioning, is currently the only treatment modality in MF with the ability to induce long-term disease-free remission. The value of other treatment options, including drug therapy, splenectomy, and radiotherapy, is mostly palliative with uncertain survival benefit.9 A plethora of new drugs, including thalidomide analogs10 and Janus kinase11-13 or mammalian target of rapamycin14 inhibitors, have been recently developed and evaluated in MF clinical trials. So far, none of these new drugs have displayed selective anti-clonal effect, despite an otherwise remarkable activity in alleviating anemia, splenic discomfort, and constitutional symptoms. In other words, the value of such drugs would be undermined if formal response criteria in MF did not include response categories that capture drug benefit in terms of MF-SB, which impacts health-related QoL. However, there is no good evidence to indicate that responses in anemia, splenomegaly, or symptoms could be used as surrogates for improved survival. Consensus-based definitions of response, in this regard, are designed for the purpose of standardizing response criteria for use in clinical trials and not for use in routine care of patients.

Study design

The current work is the result of a collaborative project by the International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) and European LeukemiaNet (ELN). The document was developed through extensive discussions that took place during the IWG-MRT annual meetings of 2011 and 2012 in Florence, Italy, as well as the ELN MPN subcommittee conferences at the 2011 and 2012 American Society of Hematology annual meetings in San Diego, CA, and Atlanta, GA, respectively. These meetings were led by an expert panel that included authors of the 2006 IWG-MRT response criteria for MF,15 as well as myeloproliferative neoplasm subcommittee members for ELN. Post- and pre-meeting input from study participants were sought through electronic communications and adjudicated through consensus (see the supplemental Data Set link at the top of the online article for details of the decision process).16

The basic principles behind the current revision were to include response categories that suggest disease modification, as well as those that provide objective quantification of drug activity in improving anemia, splenomegaly, and symptoms. In this regard, the expert panel acknowledged the need for strict definitions of red cell transfusion dependency and independency, confirmation of spleen response by imaging studies, and the Myeloproliferative Neoplasm Symptom Assessment Form (MPN-SAF) tool to measure meaningful changes in MF symptoms.17 The revised criteria also provide recommendations for assessing cytogenetic and molecular responses.

Results and discussion

Table 1 outlines 9 separate categories for the revised IWG-MRT and ELN response criteria for treatment in MF. Table 1 also includes recommendations for cytogenetic and molecular responses, but these responses are not required for assignment as complete (CR) or partial remission (PR). Definitions of red cell transfusion dependency and independency and other items are added as footnotes to Table 1.

Table 1

Revised IWG-MRT and ELN response criteria for MF

The definitions of CR and partial (PR) remissions are somewhat akin to those employed in acute myeloid leukemia18 and myelodysplastic syndromes,19 and are meant to highlight drug effects that suggest disease modification or substantial anti-clonal activity. Bone marrow morphologic remission is a requisite for CR and its definition in the current document was intentionally toned down to minimize subjective differences in assessing megakaryocyte morphology. Morphologic remission in the peripheral blood, but not necessarily in the bone marrow, is required for PR assignment. Patients meeting criteria for CR, but who have inadequate blood count recovery are also included in the PR response category to capture disease-modifying activity confounded by drug-related cytopenia.18 In other words, some drugs might induce prolonged myelosuppressive effect that prevents normal recovery of blood counts despite morphologically normal-appearing marrow.

The IWG-MRT and ELN response categories other than CR and PR were developed in recognition of the profound impact of MF-SB to health-related QoL. The primary contributors of decreased health-related QoL in MF are anemia, marked splenomegaly, and constitutional symptoms. Accordingly, the current revised document includes response definitions for each one of these specific disease features and an additional composite response category, labeled as clinical improvement (CI), and defined as a response in anemia, splenomegaly, or MF-SB that is not associated with progressive splenomegaly (Table 1) or increase in severity of anemia, thrombocytopenia, or neutropenia (Table 1 footnotes). Accordingly, an anemia response that might be associated with progressive splenomegaly (as has been seen with pomalidomide therapy)20 or spleen response associated with drug-induced anemia (as has been seen with some Janus kinase inhibitors),11,21 would still be included in an individual response category, although not counted as CI. Similarly, for a symptom response to count as CI, it requires the absence of progressive splenomegaly and treatment-associated anemia.

Recent experience with clinical trials in MF has highlighted the need to establish strict definitions for red cell transfusion dependency and treatment-induced transfusion independency22; these are now outlined as footnotes in Table 1. We fully recognize the fact that our consensus-based definitions in this regard are imprecise and do not adequately address the confounding effects of age and race on blood volume and cultural differences in indications for blood transfusion. Some patients with transfusion needs may not meet the strict criteria for transfusion dependency at the time of study enrollment; the expert panel recommends the use of the pre-transfusion hemoglobin level as baseline in such cases. Another confounding element in phase 1 and 2 studies is the possibility that improvement in anemia might be the result of discontinuation of myelosuppressive therapy (eg, hydroxyurea), in preparation for enrollment into clinical trials. This is why phase 3 studies are important in validating observations from phase 2 studies, especially in terms of anemia response. In other words, anemia response in a phase 2 study, without placebo control, should be viewed with caution, but standardizing the criteria should help compare results between phase 2 studies.

The IWG-MRT and ELN expert panel also recognized the highly subjective nature of spleen and liver size assessment by physical examination, and recommended objective confirmation by magnetic resonance imaging (MRI) or computed tomography. The spleen volume reduction thresholds for response in this regard were set at 35% based on recent studies that compared physical examination and MRI assessment of spleen size in patients with MF.21 In some cases, imaging studies might reveal significant volume reduction that is not captured by physical examination; in such cases, a ≥ 35% reduction in spleen or liver volume overrides the measurements by physical examination for the purposes of response assignment.

Response in MF-SB is assessed by The MPN-SAF total symptom score (TSS)17; the TSS is assessed by the patients themselves and includes fatigue, concentration, early satiety, inactivity, night sweats, itching, bone pain, abdominal discomfort, weight loss, and fevers. Each of the 10 symptoms is scored from 0 (absent/as good as it can be) to 10 (worst imaginable/as bad as it can be). The MPN-SAF TSS is a summation of all the individual scores (0-100 scale). A response in MF-SB requires a ≥50% reduction in the MPN-SAF TSS. An assessment of health-related QoL may also be done in the context of a clinical trial, but given its multifactorial nature, the expert panel agreed that assessing therapy response should focus on MF-SB.

Laboratory investigations on the biology and genetics of MF are likely to identify new drug targets and clarify their pathogenetic contribution of Janus kinase–signal transducer and activator of transcription. This is important considering the failure of currently available drugs in securing selective suppression of clonal myeloproliferation. The availability of more effective and selective anti-neoplastic drugs in MF will mandate the formal incorporation of cytogenetic and molecular information in future revisions of the current response criteria. In the meantime, strictly defined measurements of palliative value are necessary to justify the therapeutic use of new drugs and allow comparison of their efficacy. In the end, we would like to emphasize the fact that consensus statements do not necessarily provide either accurate or validated surrogates of clinical benefit or survival, and that their value as standardized tools of comparison for clinical trials should be approached with caution.


Contribution: A.T., G.B., and T.B. organized and designed the research; A.T., F.C., R.M., F.P., S.V., A.M.V., J.G., B.D., A.P., C.H., R.H., H.G., N.K., J.T., T.B., and G.B. participated in consensus development; and Novartis, Sanofi, Shire to C.H.A.T. wrote the paper.

Conflict-of-interest disclosure: A.T., T.B., A.P., B.D., R.H., F.P., S.V., H.G., J.T., and N.K. declared no conflict of interest; A.M.V. is on the advisory board of Novartis; G.B. and F.C. is on the advisory board of Novartis and Sanofi; R.M. is a consultant for Novartis; J.G. is on the advisory board of Incyte, Novartis, Gilead, and Sanofi; C.H. is on the advisory board of Novartis, Celgene, Sanofi, Gilead, SBio, YM Bioscience, and CTI, and has been a speaker for Shire, Sanofi, and Novartis.

Correspondence: Ayalew Tefferi, Division of Hematology, Department of Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905; e-mail: tefferi.ayalew{at}


The authors acknowledge the key contribution of all IWG-MRT and ELN members who participated in the IWG-MRT annual meetings of 2011 and 2012 in Florence, Italy, as well as the ELN MPN subcommittee conferences at the 2011 and 2012 American Society of Hematology annual meetings in San Diego, CA, and Atlanta, GA, respectively; Prof Rudiger Hehlmann for facilitating the current collaboration and his continued encouragement and support; and Moni and Mary Terner, and the Terner Foundation for providing financial support for the Florence IWG-MRT meetings.

This work was supported by a research grant from Celgene, Incyte, Sanofi, Genentech, NS Pharma, Lilly, and Gilead to R.M.; a research grant from Incyte, Novartis, Gilead, Sanofi to J.G.; and a research grant and honoraria from Incyte, Gilead, and Sanofi and travel support from Incyte, Novartis, and Sanofi to J.G.; and research grants from Novartis, Sanofi, and Shire, and educational materials from Peerview, Medscape, and Articulate Science to C.H.


  • The online version of this article contains a data supplement.

  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

  • Submitted March 4, 2013.
  • Accepted June 27, 2013.


View Abstract