Reclassification of subtypes in Philadelphia chromosome-negative myeloproliferative neoplasm by 2016 WHO diagnostic criteria: focus on the cases classified as myeloproliferative neoplasm, unclassifiable by the 2008 version

The 2016 revised World Health Organization (WHO) Classification of Tumors of Hematopoietic and Lymphoid Tissues was published in September 2017 [1]. Compared to the 2008 WHO guidelines, there were several major improvements derived from clinical pathological, and molecular genetic studies. The discovery of novel molecular findings had an impact on diagnosis and outcome. By lowering the diagnostic hemoglobin (Hb)/hematocrit (Hct) threshold values for covering masked polycythemia vera (PV), and emphasizing the need to discriminate ‘true’ essential thrombocythemia (ET) from prefibrotic primary myelofibrosis (pre-PMF), bone marrow histology now plays a major role in diagnosis. The efforts for standardization of morphological bone marrow (BM) features resulted in an improvement in the differentiation of MPN subtypes, particularly between ET, pre-PMF, and PV [2].

Myeloproliferative neoplasm, unclassifiable (MPN-U) should be applied only to cases that have definite features of a myeloproliferative neoplasm (MPN), but fail to meet the diagnostic criteria for any of the specific MPN entities, or that present with features that overlap between two or more of the MPN categories. It includes early-phase MPN cases in which the characteristic features are not yet fully developed, advanced-stage MPN, which has prominent myelofibrosis, and cases with convincing evidence of an MPN with a coexisting neoplastic or inflammatory disorder. Therefore, MPN-U is a heterogeneous disease entity, and the exact incidence of MPN-U is unknown.

Two studies reported several findings of MPN-U with the same MPN-U patients. One study investigated, in detail, the morphological, clinical, and molecular features of 71 MPN-U patients. The authors morphologically sub-grouped the cases into ET-like, PMF-like, and PV-like. However, the clinical parameters in these three clusters revealed discrepancies with the morphological profile in approximately 55% of the patients [3]. The other study applied the 2016 revised diagnostic criteria to a previously published series of 71 MPN-U cases. Of the 71 cases, seven were reclassified to PV, 20 to ET, and 25 to overt PMF. Accordingly, the rate of discrepancy between clinical and morphologic findings was changed from 55% to 27%. Their approach reduced approximately 30% of the MPN-U, and revealed improvements in the diagnostic capability [4].

We retrospectively investigated the characteristics of 53 MPN patients and diagnosed their MPN subtypes according to both the 2008 and 2016 WHO classifications. In the present study, cytogenetic studies, targeted sequencing for 88 hematopoiesis-related genes (Supplementary Table 1), and survival analyses were performed. Of 53 MPN cases, only MPN-U cases diagnosed by WHO 2008 criteria showed different diagnosis by the WHO 2016 criteria (Figure 1). Of 11 MPN-U^{WHO 2008}, six were diagnosed as other MPN subtypes by WHO 2016 criteria, specifically, four patients (36.4%) to pre-PMF^{WHO 2016}, and two patients (18.2%) to overt PMF^{WHO 2016}. Five patients (45.5%) remained as MPN-U^{WHO 2008 & 2016}. The clinical parameters, BM morphology, treatments, and prognosis, and diagnostic evidences are summarized in Table 1. The WHO 2008 criteria required at least two minor criteria to diagnose prefibrotic/overt PMF; leukoerythroblastosis, increase in serum lactate dehydrogenase level, anemia, and splenomegaly. Some MPN cases that have PMF-like morphological features but do not satisfy the minor criteria were diagnosed as MPN-U^{WHO 2008}. However, since WHO 2016 criteria required at least one minor criterion to diagnose pre-fibrotic/overt PMF, more than half of MPN-U^{WHO 2008} were reclassified into PMF^{WHO 2016}.

Figure 1. Transitions of the diagnosis in patients with ET, PMF, and MPN-U between WHO 2008 and 2016 Classifications. ET: essential thrombocythemia; MPN-U: myeloproliferative neoplasm, unclassifiable; overt PMF: primary myelofibrosis, overt fibrotic stage; pre-PMF: primary myelofibrosis, prefibrotic/early stage; MF: myelofibrosis; PV: polycythemia vera.

Table 1. Peripheral blood and bone marrow findings, cytogenetic and molecular studies, diagnostic evidences^{WHO 2008 & 2016}, treatment, and prognosis of 11 previously diagnosed with MPN-U^WHO2008.

UPN	Age/ Sex	WBC-Hb-PLT (10^9/L-g/ dL-10^9/L)	Hypercellular	Megakaryocytic proliferation	Morphology of megakaryocytes	Reticulin/ Collagen grade	Cytogenetic studies (G-banding and/or FISH)	Gene mutations in the 88-gene panel	WHO version	Diagnosis	Diagnostic criteria^a		Treatment and Prognosis and/or Follow-up
											Satisfied for pre-PMF/overt PMF/post-PV MF	Missing for ET/pre-PMF/overt PMF/post-PV MF
1	70/M	10.26-10.4-187	No	No	Focal increase in number of megakaryocytes with clustering but on average 2.6/x400 HPF, irregularly folded nuclei, aberrant nuclear/cytoplasmic ratio	3/3	normal	ASXL1, CALR, CBL	2008	MPNU	[Overt PMF] Grade ≥2 bone marrow fibrosis, anemia, leukoerythroblastosis, and splenomegaly as minor criteria, CALR mutation	[Overt PMF] Megakaryocytic proliferation	HU with supportive care/expired No follow-up BM examination
									2016	MPNU
2	56/M	10.97-9.3-71	No	No	Small size, dense clustering with small number of megakaryocytes, and aberrant nuclear/cytoplasmic ratio, 0-1/x400 HPF	3/3	46,XY,inv(12) (q15q24.1)[20]	ASXL1, CALR, EZH2, ZRSR2	2008	MPNU			After alloPBSCT, expired due to GVHD. No follow-up BM examination before alloPBSCT.
									2016	MPNU
3*^b	54/M	14.90-13.0-349	Yes	Yes	Pleomorphic including multinucleation, hypolobulated nuclei, small size	3/2	normal	ASXL1, JAK2, LRP1B, SF3B1	2008	MPNU	[Post-PV MF] Previous diagnosis of PV, grade ≥2 bone marrow fibrosis, leukoerythroblastosis as an additional criterion	[Post-PV MF] At least one additional criteria except leukoerythroblastosis	HU After 56 months, the patient presented anemia (Hb 8.3 g/dL) and leukoerythroblastosis, which met the post-PV MF criteria WHO 2008 & 2016. Transformed to AML, expired after alloPBSCT.
									2016	MPNU
4	19/F	9.29-12.8-508	Yes	Yes	Aberrant nuclear/cytoplasmic ratio and irregularly folded nuclei	0-1/0	normal	None	2008	MPNU	[Pre-PMF] Megakaryocytic proliferation and atypia, hypercellularity,	[Pre-PMF] Minor criteria	Anagrelide/alive No follow-up BM examination Normal range of hemoglobin and WBC counts and mild thrombocytosis continued.
									2016	MPNU
5	62/M	7.47-13.8-838	No	Yes	Anisocytosis, loose clustering, and irregularly folded and hyperchromatic nuclei	1-2/0	normal	CALR	2008	MPNU	[Pre-PMF] Megakaryocytic proliferation with atypia, CALR mutation	[Pre-PMF] Hypercellularity, minor criteria (the patient presented borderline splenomegaly) [ET] Increased numbers of enlarged, mature megakaryocytes with hyperlobulated nuclei	Anagrelide, HU, and ASA/alive No follow-up BM examination Normal range of hemoglobin and WBC counts and moderate thrombocytosis continued.
									2016	MPNU
6	84/F	23.56-14.5-1202	Yes	Yes	Small to large megakaryocytes with bulbous nuclei and clustering	2-3/0	46,XX,t(7;8)(q32;p12) [19]/46,XX[1]	POLG, ZMYM3	2008	MPNU	[Overt PMF] Megakaryocytic proliferation and atypia, grade ≥2 bone marrow fibrosis, absence of reactive myelofibrosis, leukoerythroblastosis as a minor criterion	[Overt PMF] At least one minor criteria except leukoerythroblastosis	HU No follow-up BM examination 53 months later, the patient presented CBC 80.47-5.2-84, 10% immature cells, and left-shifted neutrophils in peripheral blood, which suggested acute transformation of PMF. After that, there is no follow up visit.
									2016	overt PMF	[Overt PMF] Megakaryocytic proliferation and atypia, grade ≥2 bone marrow fibrosis, absence of reactive myelofibrosis, leukoerythroblastosis and leukocytosis as minor criteria	[Overt PMF] None
7	56/F	5.33-11.9-665	No	Yes	Anisocytosis, clustering, and folded nuclei	2-3/1	normal	None	2008	MPNU	[Overt PMF] Megakaryocytic proliferation and atypia, grade ≥2 bone marrow fibrosis, absence of reactive myelofibrosis, borderline anemia as a minor criterion	[Overt PMF] At least one minor criterion except borderline anemia	Anagrelide/alive the next 54 months after the diagnosis No follow-up BM examination In follow-up CBC parameters, only hemoglobin concentration slightly decreased without leukoerythroblastosis and LDH test was not performed.
									2016	overt PMF	[Overt PMF] Megakaryocytic proliferation and atypia, grade ≥2 bone marrow fibrosis, absence of reactive myelofibrosis, borderline anemia as a minor criterion	[Overt PMF] None
8	73/F	11.35-13.5-791	Yes	No	Anisocytosis, aberrant nuclear/cytoplasmic ratio, and irregularly folded and hyperchromatic nuclei	1-2/0	normal	DNMT3A, JAK2, TET2	2008	MPNU	[Pre-PMF] Megakaryocytic proliferation and atypia, JAK2 mutation	[Pre-PMF] At least 2 minor criteria	HU and ASA/alive the next 34 months No follow-up BM examination In follow-up CBC parameters, there is no abnormalities and LDH test was not performed.
									2016	pre-PMF	[Pre-PMF] Megakaryocytic proliferation and atypia, JAK2 mutation, leukocytosis as a minor criterion	[Pre-PMF] None
9^b	68/M	12.75-15.2-986	Yes	Yes	Anisocytosis, clustering, hypolobulated and multinucleated megakaryocytes, and bulbous nuclei	1-2/0	normal	CALR, SF3B1	2008	MPNU	[Pre-PMF] Megakaryocytic proliferation and atypia, CALR mutation	[Pre-PMF] At least one minor criteria except splenomegaly	Anagrelide, HU, and ASA/alive No follow-up BM examination In follow-up CBC parameters, only left-shifted neutrophils were observed without leukoerythroblastosis or anemia. LDH test was not performed.
									2016	pre-PMF	[Pre-PMF] Megakaryocytic proliferation and atypia, CALR mutation, leukocytosis and splenomegaly as a minor criteria	[Pre-PMF] None
10^b	72/M	19.80-15.0-1068	Yes	Yes	Clustering and folded nuclei	1-2/0	normal	JAK2, SF3B1	2008	MPNU	[Pre-PMF] Megakaryocytic proliferation and atypia, JAK2 mutation	[Pre-PMF] At least 2 minor criteria	Anagrelide/alive the next 45 months No follow-up BM examination In follow-up CBC parameters, only leukocytosis and occasionally left-shifted neutrophils were observed without leukoerythroblastosis or anemia. LDH test was not performed.
									2016	pre-PMF	[Pre-PMF] Megakaryocytic proliferation and atypia, JAK2 mutation, leukocytosis as a minor criterion	[Pre-PMF] None
11	48/F	12.00-14.2-1165	Yes	Yes	Loose clustering, hyperchromatic and irregular folded nuclei	1-2/NT	normal	CALR	2008	MPNU	[Pre-PMF] Megakaryocytic proliferation and atypia, CALR mutation	[Pre-PMF] At least 2 minor criteria	HU/alive No follow-up BM examination In follow-up CBC parameters, there is no abnormalities and LDH test was not performed.
									2016	pre-PMF	[Pre-PMF] Megakaryocytic proliferation and atypia, CALR mutation, leukocytosis as a minor criterion	[Pre-PMF] None

M, male; F, female; NA, not assessed; NT, not tested; PMF, primary myelofibrosis; pre-PMF, prefibrotic primary myelofibrosis; PV, polycythemia vera; post-PV MF, post-PV myelofibrosis; MPNU, myeloproliferative neoplasm, unclassifiable; alloPBSCT, allogenic peripheral blood stem cell transplantation; ASA, aspirin (acetylsalicylic acid); GVHD, graft-versus-host disease; HU, hydroxyurea; UPN, unique patient number.

*Has a previous history of polycythemia vera.

^aIn 2016 WHO classification, the satisfied minor criteria in patients were confirmed in two consecutive determinations.

^bLess than 5% ring sideroblasts were observed.

Of the remaining five MPN-U^{WHO 2008 & 2016} cases, two (UPN 1 and 2) showed marked myelofibrosis and had no megakaryocytic proliferation: 2.6/HPF and 0–1/HPF, respectively (X400 high-power field). These cases were considered as MPN-U possibly presented in advanced-stage MPN-U^{WHO 2008 & 2016}. UPN 3 was previously diagnosed with PV, and the patient’s clinical and histological findings met the criteria of overt PMF^{WHO 2016}, but did not meet the diagnostic criteria of post-PV myelofibrosis^{WHO 2008 & 2016}. After 56 months, the patient presented anemia (Hb 8.3 g/dL) and leukoerythroblastosis, and at that time the case met the post-PV MF criteria^{WHO 2008 & 2016}. The other two patients (UPN 4 and 5) expressed mild fibrosis and had no PMF minor criteria, and were sub-classified to early stages of MPN-U^{WHO 2008 & 2016}.

SF3B1 mutation is one of diagnostic criteria in myelodysplastic syndrome with ring sideroblasts (MDS-RS) and myelodysplastic/myeloproliferative neoplasm with ring sideroblasts and thrombocytosis (MDS/MPN-RS-T). In our 11 MPN cases, three patients (UPN 3, 9, and 10) harbored SF3B1 mutations. For differential diagnosis from MDS/MPN-RS-T, prussian blue staining of BM aspirates was performed and ring sideroblasts were less than 5% among total BM nucleated cells in all cases. Additionally, no three cases id not present persistent thrombosis. Therefore, the three patients were not compatible with MDS/MPN-RS-T.

We compared survival curves of MPN between the diagnosis of WHO 2008 and 2016 classifications. In the 2008 version, the overall survival and progression-free survival were as follows: ET, undefined for both (all ET cases were alive and did not show progression); pre-PMF, 83 months for both; overt PMF, 60 months and 49 months, respectively; MPN-U, 91 months for both. In the 2016 version, the overall survival and progression-free survival were as follows: ET, undefined for both (all ET cases were alive and did not show progression); pre-PMF, undefined for both (less than 50% of pre-PMF patients died within the observed period, and none showed progression rather than death); overt-PMF, 58 months and 52 months, respectively; MPN-U, 72 months and 91 months, respectively. Both the previous and revised versions did not show a significant difference in survival among ET, pre-PMF, and overt PMF. However, compared to the previous version, the revised one presented a trend of better-stratified prognosis of MPN subtypes (p-value: 0.2308 versus 0.0974 in overall survival rate, 0.1787 versus 0.0703 in progression-free survival rate among ET, pre-PMF, and overt PMF) (Supplementary Figure 1). These survival analyses were similar with previous studies, in which pre-PMF exhibit poorer prognosis compared to ET [5–8] and better prognosis compared to overt PMF [9,10].

The proportion of unclassifiable cases may be considered as a true yardstick for accuracy to discriminate MPN subtypes. Reported incidence of MPN-U varies significantly among different studies, with a range up to >20%. However, most studies show an incidence of 10–15% or less. When the 2016 WHO criteria were applied, Iurlo et al. observed that the incidence of MPN-U was reduced to <5% [2,4].

The 2016 WHO classification and diagnostic criteria defines pre-PMF as a distinct clinicopathologic entity [1], and emphasizes the need to discriminate ET and pre-PMF because of distinction in prognosis and therapeutic relevance between the two diseases [2]. By utilizing criteria of pre-PMF by WHO 2016 classification, in the present study, some unclassifiable cases were re-diagnosed as pre-PMF. According to WHO 2016 criteria, a subset of MPN-U^{WHO 2008} was shifted to pre-PMF, enabling to predict patient outcomes better, and to assist in the development of optimal therapeutic plans.

According to the WHO 2016 criteria, MPN-U was markedly reduced, and our results support that WHO 2016 criteria successfully differentiated MPN subtypes, leading to better prognostication. We expect that the revised diagnostic criteria will help clinicians to make effective decisions. Moreover, we described MPN-U, a heterogeneous group that can be sub-grouped by BM morphology.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (NRF-2017R1A2A1A17069780).

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.