Definition and characteristics of primary myelofibrosis

Primary myelofibrosis (PMF) is a myeloproliferative, BCR-ABL1-negative neoplasm. It is characterized by a dominant proliferation of megakaryocytes and granulocytes in the bone marrow and shows increasing reticulin and/or collagen fibrosis in advanced stages. The incidence of primary myelofibrosis is 0.5-1.5 / 100,000 per year and occurs predominantly at the age of 60-70 years.

Classification of primary myelofibrosis

According to the WHO classification 2017, primary myelofibrosis is classified as myeloproliferative neoplasm (MPN) and is divided into a prefibrotic/early stage and a fibrotic stage.

Primary myelofibrosis - WHO Classification 2017

Myeloproliferative Neoplasm (MPN)

Primary myelofibrosis (PMF):

  • Prefibrotic/early stage
  • Fibrotic stage

Diagnostic criteria for primary myelofibrosis

According to the WHO classification, all 3 major criteria and at least 1 minor criterion of the diagnostic criteria must be met in order to make a diagnosis of primary myelofibrosis (overt fibrotic stage).

Major criteria for primary myelofibrosis

  • Megakaryocyte proliferation and atypia, accompanied by reticulin and/or collagen fibrosis grades 2 or 3 according to WHO
  • WHO criteria for ET, PV, BCR-ABL1-positive CML, MDS or other myeloid neoplasm are not met
  • JAK2, CALR or MPL mutation, or in the absence of these mutations, detection of another clonal marker*, or absence of minor reactive bone marrow reticulin fibrosis**

Minor criteria for primary myelofibrosis

At least 1 of the following criteria are met in 2 consecutive determinations

  • anaemia not attributed to a comorbid condition
  • Leukocytosis ≥ 11 x 109 /l
  • Palpable splenomegaly
  • LDH level above the upper limit of normal (ULN) of the institutional reference range
  • Leukoerythroblastosis

* e.g. ASXL1, EZH2, TET2, IDH1/IDH2, SRSF2, SF3B1
** bone marrow fibrosis due to secondary infection, autoimmune disease or other chronic inflammatory diseases, hairy cell leukemia or other lymphoid neoplasm, metastatic tumor disease or toxic (chronic) myelopathy

Facts

  • Approx.
    60%
    60%

    of patients with PMF have a JAK2 V617F mutation (Onkopedia Guideline PMF)

Diagnostics of primary myelofibrosis

Cytomorphology

The cytomorphological assessment in MPN involves cellularity in the total as well as in the individual hematopoietic lines. It is also important to determine the proportion of blasts. For special questions (in case of suspected refractory anaemia with ring sideroblasts and thrombocytosis; MDS/MPN-RS-T) iron staining is also relevant.

If there is fibrosis of the bone marrow, e.g. in primary myelofibrosis (PMF), the cytomorphological assessability of the preparations is often limited ("punctio sicca").

In all cases, however, the histology is decisive for the diagnosis as well as for the assessment of the degree of fibrosis and the bone marrow architecture, which must always be carried out if MPN is suspected or confirmed.

Chromosome analysis

Chromosomal abnormalities occur in 35-40% of patients with primary myelofibrosis and are predominantly unbalanced events. Thus, gains of material of the long arm of chromosome 1 (+1q), deletions in the long arm of chromosome 20 (del(20q)), trisomy 9 (+9), deletions in the long arm of chromosome 13 (del(13q)), trisomy 8 (+8), abnormality of chromosome 7 (-7, del(7q)) as well as of chromosome 5 (-5, del(5q)), deletions in the short arm of chromosome 12 (del(12p)) and rather rarely an isochromosome of the long arm of chromosome 17 (i(17q)) are observed. None of these abnormalities are specific for primary myelofibrosis, they also occur in other MPN and also in MDS (Swerdlow et al. 2017).

FISH

In the case of punctio sicca, the cytogenetic changes in the blood smear typical of primary myelofibrosis can be detected by FISH analysis.

Molecular genetics

JAK2 V617F mutations are common in primary myelofibrosis

Molecular genetic evidence of a JAK2 V617F mutation is found in about 60% of patients with primary myelofibrosis. The prognostic relevance of this mutation is controversially discussed (Campbell et al. 2006, Guglielmelli et al. 2009). In 10% of all patients with primary myelofibrosis without JAK2 V617F mutation a mutation in codon W515 of the MPL gene was detected. A mutation in the calreticulin gene (CALR) is found in about 70% of patients with myelofibrosis in whom no JAK2 or MPL mutation has been detected (Klampfl et al. 2013, Nangalia et al. 2013). Patients without one of these three mutations ("triple negative") have a higher risk of developing AML than patients with a JAK2, MPL or CALR mutation (Rumi et al. 2014, Tefferi et al. (1) 2014). Mutations in the EZH2 gene are found in 5-7% of patients with PMF and are associated with an unfavourable course of disease according to the data available to date (e.g. Guglielmelli et al. 2011). A mutation of the ASXL1 gene is detected in 20-30% of patients with primary myelofibrosis (Vannucchi et al. 2013).

Table 1: Frequency of different mutations in primary myelofibrosis (Langabeer et al. 2015, Tefferi 2018)

Gene mutation

Frequency (%)

JAK2 V617F

(Exon 14)

55-65

JAK2 Exon 12

rare

MPL

5-10

CALR

25-35

TET2

10-20

IDH1/2

4-5

DNMT3A

5-10

ASXL1

13-30

EZH2

5-10

CBL

5-10

SF3B1

5-10

SRSF2

10-17

U2AF1

5-16

TP53

4

Prognosis

Among the BCR-ABL1-negative myeloproliferative neoplasms, primary myelofibrosis has the most unfavourable course. There is both a risk of progression of the disease into the fibrotic stage and a risk of leukemic transformation. While the median overall survival is about 6 years (Tefferi et al. (4) 2014), the individual clinical courses are very heterogeneous. The prognosis is influenced by a variety of factors, which can be divided into clinical, cytogenetic and molecular genetic factors. Especially for genetic changes, the description of prognostically relevant factors is still a subject of research. With increasing knowledge, various prognostic scoring systems (~PSS) have been successively established or further developed. Table 2 gives an overview of the published risk stratification systems and the prognostic factors considered therein.

Table 2: Prognostic scoring systems for primary myelofibrosis and its risk factors

Score

 

IPSS

 

DIPSS

 

DIPSS Plus

 

GPSS

 

MIPSS

 

GIPSS

 

MIPSS70

 

MIPSS70+

 

MIPSS70+ Version 2.0

International

Prognostic

Scoring

System

 

Dynamic

International

Prognostic

Scoring

System

 

Dynamic

International

Prognostic

Scoring

System

Plus

 

Genetics-Based Prognostic Scoring System

Mutation-Enhanced International Prognostic Scoring System

Genetically Inspired Prognostic Scoring System

Mutation -Enhanced International Prognostic Scoring System

for Transplantation-age Patients

Karyotype

Enhanced MIPSS70

Mutation and Karyotype-Enhanced International Prognostic Scoring System, Version 2.0

Publication

Cervantes et al. 2009

Passamonti et al. 2010

Gangat et al. 2011

Tefferi et al. (5) 2014

Vannucchi et al. 2014

Tefferi et al. (3) 2018

Guglielmelli et al. 2018

Guglielmelli et al. 2018

Tefferi et al. (2) 2018

Considered prognositc factors

 
  • Clinic
 
 
  • Clinic
 
 
  • Clinic
 

 

 
  • Clinic
 

 

 
  • Clinic
 
 
  • Clinic
 
 
  • Clinic
 

 

 

 
  • Karyotype
 
 
  • Karyotype
 

 

 
  • Karyotype
 

 

 
  • Karyotype
 
 
  • Karyotype
 

 

 

 

 
  • Mutations
 
 
  • Mutations
 
 
  • Mutations
 
 
  • Mutations
 
 
  • Mutations
 
 
  • Mutations
 
Clinical prognostic factors in primary myelofibrosis

Various studies have demonstrated an association between reduced overall survival and the following clinical parameters (Cervantes et al. 2009, Passamonti et al. 2010, Caramazza et al. 2011, Gangat et al. 2011, Guglielmelli et al. 2018):

  • Age
  • Anemia
  • Thrombocytopenia
  • Leucocytosis
  • Circulating Bubbles
  • Bone marrow fibrosis
  • Constitutional symptomatology
  • Transfusion dependence

An association with leukemia-free survival was observed for the clinical parameters of platelet count <100 x 109/l and circulating blasts ≥2% (Caramazza et al. 2011, Gangat et al. 2011, Tefferi et al. Leukemia (3) 2018).

Risk assessment based on clinical risk factors: IPSS and DIPSS score

In order to be able to make a statement about the course of primary myelofibrosis, the IPSS and DIPSS scoring systems were established on the basis of clinical risk factors. In both scoring systems the risk factors age, B symptoms, haemoglobin value below 10 g/dl, leukocytes above 25 x 109/l and over 1% blasts in peripheral blood are considered (Cervantes et al. 2009, Passamonti et al. 2010). Both scoring systems can be used for the selection of the therapy algorithm according to German guidelines (Onkopedia guideline PMF 2018). While IPSS can only be applied at the initial diagnosis, risk classification according to DIPSS is possible throughout the course of the disease due to a different weighting of the risk factors.

Table 3: Prognosis and risk assessment of primary myelofibrosis after IPSS and DIPSS

Risk factors

Number of risk faktors

Prognosis (Risk)

Median survival risk factors (years)

IPSS

DPSS

IPSS

DPSS

- Age >65 years

- Constitutional symptoms (Fever, weight loss,

night sweat)

- Hb <10g/dl*

- Leukocytes>25 x 109 /l

- Blasts in peripheral blood ≥ 1%

0

0

low

11,2

15,4

1

1-2

intermediate 1

7,9

6,5

2

3-4

intermediate 2

4,0

2,9

≥ 3

≥ 5

high

2,3

1,3

*double weighted at DIPSS

Cytogenetic prognostic factors in primary myelofibrosis

Cytogenetic abnormalities are frequently observed in primary myelofibrosis, although the abnormalities present are heterogeneous and not specific to PMF (see Diagnosis of Primary Myelofibrosis, Chromosomal Analysis). Cytogenetic alterations can influence the prognosis both favourably and unfavourably. Accordingly, stratification based on the prognostic significance of the individual cytogenetic abnormality was included in various scores.

DIPSS Plus Score

For the first time the karyotype was included in the DIPSS Plus score. In addition to the clinical risk factors listed in Table 3, this also integrated the parameters of transfusion need and platelet count below 100 x 109/l as well as cytogenetic abnormalities (Gangat et al. 2011). Cytogenetic abnormalities were divided into two prognostic groups. A complex karyotype (≥ 3 abnormalities) and one or two abnormalities with a trisomy 8, a monosomy 7 or 7q deletion, an abnormalities of chromosome 5, an isochromosome of the long arm of chromosome 17, a 12p deletion, an inversion of chromosome 3 or an 11q23 rearrangement were evaluated as prognostically unfavorable cytogenetic abnormalities.

Refined 3-step cytogenetic risk model

As studies increasingly pointed to cytogenetic heterogeneity in the prognostically unfavourable group, this two-step model was re-examined in a study of 1,002 patients with primary myelofibrosis by Tefferi et al. and the dependence of overall survival on cytogenetic abnormalities was newly validated (Tefferi et al. (1) 2018). This led to a refined 3-step model that favours risk stratification into three prognostic groups: favourable, unfavourable and very unfavourable (= very high risk) abnormalities with regard to overall survival (see Table 4). This 3-step risk model was incorporated into the risk classifications published in 2018 in MIPSS70+ Version 2.0 (Tefferi et al. (2) 2018) and GIPSS (Tefferi et al. (3) 2018).

Table 4: Revised cytogenetic risk stratification according Tefferi et al. (1) 2018

Risk category

Specific abnormalities

Median survival

Favorable

Normal karyotype

Sole 20q-, sole 13q-, sole +9, sole chromosome 1 translocationen/duplikationen, sole sex chromosome-abnormality including -Y

4.4 years

Unfavorable

Sole abnormalities:
Sole +8, 7q-, sole translocation not involving chromosome 1

Two abnormalities: without VHR-abnormality

Single/multiple abnormalities: 5q- abnormalities Complex karyotype without VHR-abnormality, Monosomal karyotype without VHR-abnormality, sole abnormalities not otherwise classified

2.9 years

Very high risk (VHR)

Single/multiple abnormalities: Monosomy 7, inv(3)/3q21, i(17q), 12p-/12p11.2, 11q-/11q23, autosomal trisomies other than +8 or +9 (z.B. +21, +19)

1.2 years

Further Prognostic Assessments of Cytogenetic Abnormalities

Other publications have discussed further prognostic assessments of cytogenetic abnormalities (see Table 5).

Table 5: Prognostic assessment of cytogenetic changes

 

Favorable

Intermediate

Unfavorable

Very high risk

Tam et al. 2009

+9 (also with additional abnormality), 13q-, 20q-

NK

Abnormalities including  chromosome 5 or 7, complex

Abnormality of

Chromosom 17

Hussein et al. 2010

+9, 13q-, 20q-

NK

Other abnormalities

+8, complex

Caramazza et al. 2011

NK and abnormalities, which are not unfavorable

 

Complex, +8, -7/7q-, i(17q), -5/5q-, 12p-, inv(3), 11q23-

rearrangements

 

Brecqueville et al. 2014

Assoziation between 12p-, 17q-, 20q- and transformation into AML

Tefferi et al. (5) 2014

NK, +9, 13q-

20q-, 1q+, -Y

Complex-not-monosomal,

5q-, +8, sole or two other abnormalities

monosomal KT, inv(3), i(17q), -7/7q-, 11q or 12p abnormality

KT: karyotype, NK: normal karyotype, complex ≥ 3 abnormalities

Molecular genetic prognostic factors in primary myelofibrosis

Influence of driver mutations

The driver mutations influence the overall survival in different ways. With a median overall survival of 15.9 years, patients with a CALR mutation have the best prognosis (Tefferi et al. (4) 2014), as long as there is no concomitant ASXL1 mutation (Tefferi et al. (2) 2014). Patients with MPL mutation (9.9 years) and JAK2 mutation (5.9 years) follow. With a median overall survival of 2.3 years, triple negative cases have the worst prognosis (Tefferi et al. (4) 2014).

For patients with a CALR mutation the subtype is of prognostic relevance. A distinction is mainly made between type 1 and type 2 mutations. Both involve exon 9 of the CALR gene. With ~69-80%, the type 1 mutation represents the majority of the CALR mutations, it leads to a deletion in exon 9. Type 2 mutations are detected in 11-21% of the cases, this mutation results in an insertion (Klampfl et al. 2013, Tefferi et al. (3) 2014). The apparent prognostically favorable course for CALR-mutated PMF is due to the high number of CALR type 1 mutations and possibly limited to these. PMF patients with CALR type 2 mutations show a similar clinical course as patients with JAK2 mutation (Tefferi et al. (3) 2014).

"High molecular risk” mutations affect overall and leukemia-free survival

Cooperative non-driver mutations can greatly affect the prognosis in primary myelofibrosis. For this purpose, the category of HMR mutations ("high molecular risk") was introduced (Guglielmelli et al. 2014), which is now considered in various scoring systems. The established HMR mutations include ASXL1, EZH2, SRSF2, IDH1 and IDH2. Independent of IPSS and DIPSS Plus, they are associated with a shorter survival and higher risk of transformation into acute leukemia (Vannucchi et al. 2013). The occurrence of two or more mutations of these genes is 20-051505

less favourable compared to one or no mutation (Guglielmelli et al. 2014). Mutations of the splicing factor U2AF1 affecting the amino acid glutamine at position 157 (Q157) have also been identified as a risk factor. They are independently associated with reduced survival in DIPSS, but the transformation rate was not affected (Tefferi et al. (4) 2018). Based on this study, U2AF1 Q157 mutations are now also counted as HMR category (Tefferi et al. (2) 2018, Tefferi et al. (3) 2018).

A further study indicates that NRAS/KRAS mutations also strongly impair the prognosis. Although they occur rarely in primary myelofibrosis (about 6% of patients), they are associated with reduced overall and leukemia-free survival (Santos et al. 2020).

Epigenetic and splicing factors with potential association to rapid progression

In a cohort of 77 PMF patients with disease progression an association to the presence of mutations of certain splicing factors (SRSF2, U2AF1, SF3B1) or epigenetic factors (IDH2, EZH2) could be observed. For 27 patients with stable clinical course, which were used as control group, no mutation in any of the mentioned genes could be detected in any case. For the 19 patients with mutation in SRSF2, U2AF1, SF3B1, IDH2 or EZH2 a rapid progression was observed, in the median after 2 years. In comparison, the median progression-free survival for the 58 patients with a different genetic background was 7.25 years (Bartels et al. 2020).

Integration of molecular genetic factors in prognostic risk models

In order to take account of the prognostic influence of mutations, these were increasingly integrated into risk stratification models. Table 6 provides an overview.

Table 6: Overview of risk scores that take molecular genetic factors into account. For CALR driver mutations a distinction is made between type 1 and type 2 mutations. CALR type 1 mutations have the most favourable prognosis among driver mutations and therefore do not represent a risk factor. In contrast, CALR type 2 mutations appear to be prognostically equivalent to JAK2 mutations (Tefferi et al. (3) 2014).

 

MIPSS

GPSS

GIPSS

MIPSS70

&

MIPSS70+

MIPSS70+  Version 2.0

Publication

Vannucchi et al. 2014

Tefferi et al. (5) 2014

Tefferi et al. (3) 2018

Guglielmelli et al. 2018

Tefferi et al. (2) 2018

Risc factors taken into account with regard to Driver mutations / mutation status 
  • JAK2 or MPL
  • triple negative
 
 
  • JAK2
  • MPL
  • CALR type 2 mutation
  • triple negative
 
 
  • any driver-mutation or - mutation-status, except CALR type 1 mutation
 
 
  • any driver mutation or -mutation status, except CALR type 1 mutation
 
 
  • any driver mutation or -mutation status, except CALR type 1 mutation

 

Risk factors considered in relation to HMR mutations 
  • ASXL1
  • SRSF2
 
 
  • ASXL1
  • SRSF2

 

 
  • ASXL1
  • SRSF2
  • U2AF1 Q157
 
 
  • 1 HMR-mutation*
  • ≥2 HMR-mutations*

 

*ASXL1, SRSF2, EZH2, IDH1/2

 
  • 1 HMR-mutation*
  • ≥2 HMR-mutations*

 

*ASXL1, SRSF2, EZH2, IDH1/2,

U2AF1 Q157

Calculation for primary myelofibrosis (PMF)

The calculation of the clinical scoring systems IPSS or DIPSS is recommended for the selection of the therapy algorithm according to the DGHO guideline (Onkopedia guideline PMF 2018). Here you can access the forecast calculation of the DIPSS score.

Consideration of cytogenetic and molecular genetic factors would lead to re-classification into a higher risk category for an estimated 25% of patients assigned to the low or intermediate group according to IPSS/DIPSS classification (Onkopedia Guideline PMF 2018). The determination of genetic risk factors can therefore be useful for weighing up for or against a stem cell transplantation (Kröger et al. 2015, Tefferi 2018).

Here you can access the calculation of the DIPSS plus score, which takes a two-step cytogenetic risk model into account. A refined three-level cytogenetic risk model as well as molecular genetic risk factors are integrated in the scoring systems of GIPSS and MIPSS70+ Version 2.0.

Recommendation

In addition to the collection of clinical and laboratory chemical parameters, histological and cytomorphological examination of the bone marrow and blood, cytogenetic analysis and molecular genetic examinations (JAK2 V617F mutation, if negative, CALR, if negative MPL) are recommended. According to the WHO 2017 classification, additional molecular genetic analyses should be added for primary myelofibrosis (ASXL1, EZH2, TET2, IDH1/IDH2, SRSF2, SF3B1).

References

Bartels S et al. Mutations associated with age-related clonal hematopoiesis in PMF patients with rapid progression to myelofibrosis. Leukemia 2020;34:1364–1372.

Brecqueville M et al. Array comparative genomic hybridization and sequencing of 23 genes in 80 patients with myelofibrosis at chronic or acute phase. Haematologica 2014;99(1):37-45.

Campbell PJ et al. Mutation of JAK2 in the myeloproliferative disorders: timing, clonality studies, cytogenetic associations, and role in leukemic transformation. Blood 2006;108(10):3548-3555.

Caramazza D et al. Refined cytogenetic-risk categorization for overall and leukemia-free survival in primary myelofibrosis: a single center study of 433 patients. Leukemia 2011;25(1):82-88.

Cervantes F et al. New prognostic scoring system for primary myelofibrosis based on a study of the International Working Group for Myelofibrosis Research and Treatment. Blood 2009;113: 2895-2901.

Gangat N et al. DIPSS plus: a refined Dynamic International Prognostic Scoring System for primary myelofibrosis that incorporates prognostic information from karyotype, platelet count, and transfusion status. J Clin Oncol 2011;29(4):392-397.

Guglielmelli P et al. Identification of patients with poorer survival in primary myelofibrosis based on the burden of JAK2V617F mutated allele. Blood 2009;114(8):1477-1483.

Guglielmelli P et al. EZH2 mutational status predicts poor survival in myelofibrosis. Blood 2011;118(19):5227-5234.

Guglielmelli P et al. The number of prognostically detrimental mutations and prognosis in primary myelofibrosis: an international study of 797 patients. Leukemia 2014;28(9):1804-1810.

Guglielmelli P et al. MIPSS70: Mutation-Enhanced International Prognostic Score System for Transplantation-Age Patients With Primary Myelofibrosis. J Clin Oncol 2018;36(4):310-318.

Hussein K et al. International Prognostic Scoring System-independent cytogenetic risk categorization in primary myelofibrosis. Blood 2010;115(3):496-499.

Klampfl T et al. Somatic mutations of calreticulin in myeloproliferative neoplasms. NEJM 2013;369(25):2379-2390.

Kröger NM et al. Indication and management of allogeneic stem cell transplantation in primary myelofibrosis: a consensus process by an EBMT/ELN international working group. Leukemia 2015;29:2126–2133.

Langabeer SE et al. Molecular diagnostics of myeloproliferative neoplasms. Eur J Haematol 2015;95(4):270-279.

Nangalia J et al. Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. NEJM 2013;369(25):2391-2405.

Onkopedia Leitlinie „Primäre Myelofibrose (PMF)”; DGHO 2018.

Passamonti F 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.

Rumi E et al. Clinical effect of driver mutations of JAK2, CALR, or MPL in primary myelofibrosis. Blood 2014;124:1062-1069.

Swerdlow SH et al. WHO classification of tumours of haematopoetic and lymphoid tissue. International Agency of Research on Cancer 2017; 4. überarbeitete Version.

Santos FPS et al. Prognostic impact of RAS-pathway mutations in patients with myelofibrosis. Leukemia 2020;34:799–810.

Tam CS et al. The role of cytogenetic abnormalities as a prognostic marker in primary myelofibrosis: applicability at the time of diagnosis and later during disease course. Blood 2009;113(18):4171-4178.

Tefferi A et al. (1) CALR vs JAK2 vs MPL-mutated or triple-negative myelofibrosis: clinical, cytogenetic and molecular comparisons. Leukemia 2014;28(7):1472-1477.

Tefferi A et al. (2) CALR and ASXL1 mutations-based molecular prognostication in primary myelofibrosis: an international study of 570 patients. Leukemia 2014;28(7):1494-1500.

Tefferi A et al. (3) Type 1 vs type 2 calreticulin mutations in primary myelofibrosis: differences in phenotype and prognostic impact. Leukemia 2014;28(7):1568–1570.

Tefferi A et al. (4) Long-term survival and blast transformation in molecularly annotated essential thrombocythemia, polycythemia vera, and myelofibrosis. Blood 2014;124(16):2507-2513.

Tefferi A et al. (5) Integration of mutations and karyotype towards a genetics-based prognostic scoring system (GPSS) for primary myelofibrosis. Blood 2014;124:406.

Tefferi A et al. (1) Revised cytogenetic risk stratification in primary myelofibrosis: analysis based on 1002 informative patients. Leukemia 2018;32(5):1189-1199.

Tefferi A et al. (2) MIPSS70+ Version 2.0: Mutation and Karyotype-Enhanced International Prognostic Scoring System for Primary Myelofibrosis. J Clin Oncol 2018;36(17):1769-1770.

Tefferi A et al. (3) GIPSS: genetically inspired prognostic scoring system for primary myelofibrosis. Leukemia 2018;32(7):1631–1642.

Tefferi A et al. (4) U2AF1 Mutation Types in Primary Myelofibrosis: Phenotypic and Prognostic Distinctions. Leukemia 2018;32(10):2274-2278.

Tefferi A. Primary myelofibrosis: 2019 update on diagnosis, risk-stratification and management. Am J Hematol 2018;93(12):1551-1560.

Vannucchi AM et al. Mutations and prognosis in primary myelofibrosis. Leukemia 2013;27(9):1861-1869.

Vannucchi AM et al. Mutation-Enhanced International Prognostic Scoring System (MIPSS) for Primary Myelofibrosis: An AGIMM & IWG-MRT Project. Blood 2014;124:405.