The myeloproliferative neoplasms, unclassifiable: clinical and pathological considerations
Gianelli U, Cattaneo D, Bossi A, Cortinovis I, Boiocchi L, Lui Y-C, et al. Mod Pathol. Oct 2016. [Epub ahead of print]

Background:

Diagnostic criteria for cases of unclassifiable myeloproliferative neoplasms (MPNs), were defined by the 2001 World Health Organisation (WHO) classification of tumours of hematopoietic and lymphoid tissues. This category includes individuals with some pathological and clinical features of MPNs but that lack some specific criteria, or those who overlap MPN categories. Two subtypes of unclassifiable MPNS were defined: early stage BCR-ABL1-negative MPNs that do not fulfil all criteria for a complete MPN disease, or patients whose bone marrow is examined at the advanced stage of MPN in which previous diagnosis was unknown. However, the unclassifiable category does not include cases which lack molecular data or have suboptimal histological specimens for the correct classification of the disease. The prevalence of unclassifiable MPNs ranges from 10-15% of all BCR-ABL1-negative MPNs and presenting clinical symptoms vary greatly depending on stage of the disease. In the literature there are no studies that address the diagnostic issues relating to WHO unclassifiable MPNs.

Aim:

The study aim was to further establish the morphological, clinical and molecular characteristics of myeloproliferative neoplasm unclassifiable patients, focussing on features which are similar or dissimilar to the established subtypes of BCR-ABL1-negative myeloproliferative neoplasms and reducing the number of patients allocated to this group.

Method:

  • All patients diagnosed as unclassifiable at 2 reference centres (IRCCS Ca’Grandad-Maggiore Policlinico Hospital Foundation (MPHF) or Milan, n=46) and Weill Cornell Medical College (WCMC, n=25)
  • Clinical, laboratory and molecular data and a bone marrow biopsy was collected for each patient
  • Detailed morphological analysis of 16 bone marrow biopsy variables was carried out by 4 experienced haematopathologists
  • The JAK2V617F mutation was detected by PCR and CALR and MPL by sequencing. If relevant, allele burden measured
  • The morphological profile of each patient was compared to a well-defined reference group with clear diagnosis of MPN

 

Results:

Of the 71 MPN unclassifiable patients evaluated, white blood cell count and haemoglobin levels were within the normal range. However, platelet count and lactate hydrogenase were higher than normal values and 44% of patients had splenomegaly. Molecular analysis showed 51 patients with JAK2V617F and of the remaining 20 patients, 2 carried the MPLW515L and 8 had a mutation in exon 9 of the CALR gene. 70% of patients displayed hypercellular bone marrow with normal erythropoiesis (59%) without left-shifting, and increased granulopoiesis (73%) with left-shifting. Loose clusters of megakaryocytes were present in 96% of cases and these cells were often giant forms with hyperlobulated or bolbous nuclei and maturation defects. 59% of individuals were in the advanced fibrotic stage of the disease. Morphological profile analysis found that 67 patients displayed similar morphological profiles and could be clustered into morphological subgroups based on similarities with the defined reference groups (essential thrombocythemia (ET), polycythemia vera (PV), primary myelofibrosis (PMF)). Evaluating the clinical features of these subgroups according to the WHO 2008 criteria established discrepancies between these features (52% of the PMF group, 23% of the ET and 100% of the PV group). In the PMF group, 13 of the 25 patients did not meet at least 3 of the 4 WHO diagnostic criteria for PMF, 6 of the 26 in the ET group did not have a platelet count >450 x109/L, needed to be defined by the criteria, and none of the PV-like patients met the diagnostic criteria to be defined as PV.

Conclusion:

The MPN unclassifiable category is clearly heterogeneous, although subgroups can be defined using clinical or morphological information. Clinical features at presentation were often not useful to suggest a specific BCR-ABL1-negative MPN however morphological features and molecular analysis were used to confirm diagnosis, e.g. 86% of patients carried a ‘driver’ mutation. Morphological analysis and evaluating correlation with classic MPN reference groups allowed the development of three morphological profiles, allowing the identification of several discrepancies between the WHO defined classic disease and the unclassified MPN subgroup. The two most important pathological factors preventing accurate morphological classification of these cases were a morphological profile that does not entirely resemble a classic BCR-ABL1-negative MPN and advanced bone marrow fibrosis. Due to the limitations of morphological analysis, clinical and molecular data was used to re-classify cases based on the identified subgroups but this was only successful in 48% of cases.  This study was the first of its kind to provide a morphological and clinical description of MPN unclassifiable cases, allowing the identification of specific subgroups of patients for more effective patient management and evident clinical consequences.

Prognostic impact of bone marrow fibrosis in primary myelofibrosis. A study of the AGIMM group on 490 patients
Guglielmelli P, Rotunno G, Bacilli A, Rumi E, Rosti V, Dealing F, et al. AJH. Sept 2016;91:918-922

Background:

Primary myelofibrosis (PMF) is a chronic myeloproliferative neoplasm (MPN), in which bone marrow (BM) fibrosis, alongside abnormal megakaryocyte morphology, is considered a major diagnostic criteria based on WHO classification. According to a European consensus, BM fibrosis has to be semi-quantitatively evaluated and can be graded in four levels (grade 0: absent; grade 1: loose network of reticulin with many intersections; grade 2: diffuse and dense increase in reticulin with extensive intersections; grade 3, diffuse and dense increase in reticulin with extensive intersections, coarse bundles of collagen and often associated osteosclerosis). The prognostic impact of fibrosis in PMF is unclear although compared to grade 0 fibrosis, grade 1 and above is associated with a 2-fold greater risk of death. Recent data suggests that survival can be more accurately predicted when fibrosis grading is combined with IPSS score, although the transferability of this data remains unclear.

Aim:

The investigation aim was to analyse the prognostic impact of fibrosis grading by evaluating diagnostic biopsies.

Method:

  • 490 patients were diagnosed with PMF according to the 2008 WHO criteria with fibrosis grade ≥1, at 6 Italian centres of the AIRC-Gruppo Italiano Malattie Mieloproliferative (AGIMM)
  • Clinical, molecular and follow-up data was collected for all patients and individual IPSS score was calculated using parameters collected at diagnosis, coinciding with BM biopsy collection
  • Patterns in clinical and laboratory measures associated to fibrosis grade were analysed by chi-squared test or Wilcoxon rank-sum test
  • Overall survival (OS) and leukaemia free survival (LFS) were calculated

Results:

490 patients were included in the study, 36.7% with grade 1 fibrosis, 40% with grade 2 and 23.3% with grade 3. IPSS scores were calculated and patient median OS was calculated for each group (low-risk 22.8 years (29.4%), intermediate-1 8.2 years (29%), intermediate-2 4.2 years (20.6%) and high risk 2.9 years (21%)). Molecular analysis revealed that 62% of patients had JAK2V617F, 20.6% had CALR mutations and 33.3% were considered high molecular risk (HMR) due to the presence of ASXL1 (23.3%), SRSF2 (9%), EZH2 (7.7%) or IDH1/2 (2.6%) mutations. A higher fibrosis grade was associated with characteristics of advanced disease; such as palpable and larger splenomegaly, thrombocytopenia, leukopenia or anaemia and a higher IPSS score. Interestingly no associations were made between fibrosis grade and phenotypic driver mutations. However, HMR patient frequency increased with fibrosis grade (25.6%, 33.7% and 44.7% respectively) and significant associations were found between fibrosis grade and ASXL1 or EZH2 mutations. Patients with fibrosis grade 2 or above were shown to have reduced survival, thrombocytopenia, large splenomegaly, and mutations including CALR type2 and HMR risk mutations. Once stratified into IPSS categories, further addition of fibrosis grade 2 or above identified patients with significantly shorter survival compared to grade 1. Multivariate analysis of fibrosis grade in the context of individual IPSS variables allocated patients to four risk groups which differed significantly in overall survival.

Conclusion:

These findings indicate that higher BM fibrosis grade is associated with unique clinical characteristics, such as more frequent cytopenias, circulating blasts and larger spleen. These patients were significantly more likely to have a higher IPSS score and be enriched in prognostically negative mutations. Fibrosis grade represented an independent prognostically adverse variable for survival which gave extra information in the context of lower IPSS scores. This study also provided proof of concept to the inclusion of fibrosis grade in statistical analysis of individual IPSS variables allows resolution of different patient groups with significantly different survival. This investigation both substantiates and expands on the claim that fibrosis grade can have a prognostic impact on PMF but before this can be formally included in decision-making, the methods used to assess fibrosis grade must be reappraised for reproducibility.

 

JAK2V617F and calreticulin mutations in recurrent venous thromboembolism: results from the EDITH prospective cohort
Ianotto JC, Chauveau A, Mottier D, Ugo V, Berthon C, Lippert E, et al. Ann Hematol. Oct 2016; [Epub ahead of print]

Background:

One year following venous thromboembolism (VTE) diagnosis (e.g. unprovoked deep vein thrombosis (DVT) or pulmonary embolism (PE)) 4-10% of patients develop cancer. Screening for JAK2V617F and calreticulin (CALR) mutations is an attractive, non-invasive approach to diagnosing myeloproliferative neoplasms (MPNs) in these patients. However, this approach is limited and previous studies suggest systematic screening for MPN clonal mutations after one unexplained VTE may not be recommended. Alternatively, patients with recurrent VTE events have a higher incidence of cancer and therefore screening for MPN mutations in these patients may be more beneficial but has been under-evaluated.

Aim:

This study investigated the incidence rate of JAK2V617F and CALR mutations in a large cohort of patients being treated for recurrent unprovoked VTE.

Method:

  • All patients hospitalised for a first or recurrent unprovoked VTE event were included in an ongoing prospective single centre observational cohort
  • Unprovoked VTE events were included if there was no evidence of post-surgical or traumatic context, no cancer and no recent pregnancy
  • DNA samples were collected and retrospectively tested for JAK2V617F and CALR mutations

Results:

787 patients had an unprovoked VTE event between May 2000 and December 2013, half of which (372) were tested for the presence of JAK2V617F and CALR mutations. 138 patients had an isolated DVT and 234 suffered a PE. Of the 372 patients tested, the JAK2V617F mutation was present in 10 patients and the CALR was absent from all. 6 of the 353 (1.7%) patients who had a VTE recurrence whilst off anticoagulation therapy were JAK2V617F positive, and 4 of the 19 (21%) patients whose recurrence happened whilst on therapeutic anticoagulation. The median age of the 10 patients carrying the JAK2V617F mutation, 8 of which were female, was 75 years and the JAK2V617F allele burden varied between 2-32%. 7 of these patients had unusual hemogram results, 4 with isolated hyperleukocytosis, 2 with isolated thrombocytosis, and one had both hyperleukocytosis and thrombocytosis. Within 6 months of VTE recurrence, MPN was diagnosed in 3 of the JAK2V617F positive patients, of which none had normal a hemogram at VTE recurrence. The other 7 JAK2V617F patients did not develop an overt MPN in the 45 months in which they were followed-up.

Conclusion:

The prevalence of the JAK2V617F mutation was high in patients with recurrent VTE on therapeutic anticoagulation (21%) but significantly lower in those off anticoagulation therapy (1.7%), and surprisingly none carried the CALR mutation. Compared to previous reports, the prevalence of the JAK2V617F mutation in this cohort in patients with recurrent VTE not on coagulation (0-0.5%) was significantly higher, however the prevalence rate was consistent with a first episode of unprovoked VTE or that of the general population. Finding the JAK2V617F mutation in 21% of patients with recurrent VTE on anticoagulation therapy suggests it is reasonable to systematically screen when VTE recurs on full anticoagulation. The absence of the CALR mutation in this cohort cannot completely rule out that CALR screening for patients with recurrent VTE may be useful, given the low number of patients included in the study. This data also suggests that a complete blood count can be as accurate as systematic screening for the JAK2V617F mutation to identify patients with an MPN, as all patients with a defined MPN exhibited an abnormal blood count at the time of VTE recurrence and the remaining patients with the JAK2V617F mutation did not develop an MPN during follow up. Overall screening for JAK2V617F mutations in patients with unprovoked, recurrent VTE events occurring on, but not off, therapeutic anticoagulation could be advantageous. However, screening in patients with thrombocytosis, leukocytosis and/or polycythemia on follow-up hemograms may still be worthwhile.

Preferences of patients with myeloproliferative neoplasms for accepting anxiety or depression treatment
Daniel C. McFarland, Megan Johnson Shen, Heather Polizzi, John Mascarenhas, Marina Kremyanskaya, Jimmie Holland and Ronald Hoffman. Psychosomatics, http://dx.doi.org/10.1016/j.psym.2016.08.006

Background:

The physical symptom burden for Philadelphia chromosome (BCR-ABL) negative myeloproliferative neoplasms (MPNs) patients is considerable. Generally, the physical symptom burden of an MPN correlates to an increased prevalence of depression, anxiety and distress. In the United States (US), the National Comprehensive Cancer Network (NCCN) and the American Society of Clinical Oncology have both published clinical practice guidelines for the management of anxiety and depression. However, these are only seen as recommendations and are not consistently included as a part of patient care, leaving depression in cancer patients inadequately treated. Addressing this issue will require adequate assessment of the depression and take into account a patient’s treatment preferences. Currently the preferences of MPN patients in the US for the treatment of depression are unknown.

Aim:

This study aim was to determine MPN patient perspective and willingness to accept antidepressants including if they prefer to accept treatment from their haematologist of a psychiatrist.

Method:

  • Individuals were recruited to the study over 4 months if they had a confirmed tissue MPN diagnosis, but excluded if they had another cancer diagnosis
  • Participants were asked to carry out an anonymous survey, including the Distress Thermometer and Problem List (DT&PL), Hospital Anxiety and depression Scale (HADS), Risky Families Questionnaire (RFQ) and to provide demographic information
  • Patients were questioned about:
    1. If they became depressed during treatment would they consider taking antidepressants?
    2. Would they consider letting their haematologist treat this depression?
    3. Would they prefer to be treated by a psychiatrist/mental health professional?

Results:

117 participants were recruited to the study (69 women, 45 men). 63% of patients were willing to accept an antidepressant, and these individuals were more likely to be willing to accept it from their haematologist and did not have a preference of receiving treatment from a mental health professional. 58.1% of patients were willing to accept an antidepressant from a haematologist although a minority (39%) preferred to be treated by a mental health provider for anxiety or depression. There was no significant relationship between a participant’s willingness to accept an antidepressant from their haematologist and their preference for a mental health professional. Patients already receiving antidepressants were more likely to willingly accept them and from their haematologist and had no preference for being treated by a mental health professional. Patients who scored highly for chronic stress were more likely to prefer to receive antidepressants from their haematologist. Anxiety and treatment preferences were not affected by MPN type, time since MPN diagnosis or any other demographic factor analysed, however it should be noted that a 26.7% of patients would not accept antidepressants from either healthcare professional.

Conclusion:

These results show that the majority of MPN patients included in this US based study would be willing to accept antidepressants and would be willing to have them prescribed to them by their haematologist, including those who were already taking antidepressants and those who experience chronic stress. This suggests there is a need to explore how haematology/oncology teams can deliver pharmacological psychological therapy and if they are willing to do so. Administering timely, preference specific treatment for depression and anxiety to MPN patients may relieve the physical symptom burden of these patients and therefore should be both standardized and prioritized in clinical care. This study provides further understanding to the exploration of personalised patient care and suggests psycho-oncologists could better serve the population if they are embedded in haematology clinics. Patient preferences are essential when considering the simultaneous provision of psychosocial care for MPN patients and oncology teams have the opportunity to be directly involved in the psychosocial care of their patients by early referral or co-management with mental health professionals. Potential psychopharmacological educational interventions should be explored.

back to top