Myeloproliferative neoplasms (MPN) are characterized by relatively specific molecular markers; JAK2 mutations are almost invariable in polycythemia vera whereas approximately 90% of patients with either essential thrombocythemia (ET) or primary myelofibrosis (PMF) display JAK2, Myeloproliferative leukemia virus oncogene (MPL) or calreticulin (CALR) mutations, in a mutually exclusive manner [1, 2]. The recent discovery of CALR mutations was relevant not only for its diagnostic contribution  but also prognostic significance [4-6]. In ET, CALR mutations correlated with younger age, male sex, higher platelet count, lower leukocyte count, lower hemoglobin level, and lower thrombosis risk [5, 6]; in PMF, they correlated with younger age, higher platelet count, lower leukocyte count, higher hemoglobin level, and lower incidence of spliceosome mutations . Survival was favorably affected by CALR mutations in PMF but not in ET [4-6]. Prognosis in CALR-mutated PMF was further modified by the presence (unfavorable) or absence (favorable) of ASXL1 mutations [7, 8] and the number of prognostically detrimental mutations .
CALR mutational frequencies in ET are estimated between 15 and 24% . More than 80% of CALR mutated patients harbor one of only two mutation variants: type 1, a 52-bp deletion (p.L367fs*46) and type 2, a 5-bp TTGTC insertion (p.K385fs*47) . In a recent study of ET patients not included in the current cohort , the frequencies of type 1 and type 2 CALR mutations were 46% and 38%, respectively. In PMF, mutational frequencies of CALR are estimated between 25 and 35% ; in one study, 27% of PMF patients harbored CALR mutations, of which, 80% and 11% were type 1 and type 2, respectively . Comparison of type 1 and type 2 CALR mutations in PMF showed the latter to be associated with higher dynamic international prognostic scoring system (DIPSS)-plus risk distribution, circulating blast percentage, leukocyte count, and inferior survival . In the current study, we looked for phenotypic or prognostic differences between type 1 and type 2 CALR mutations in ET.
The current study was approved by the institutional review boards of Mayo Clinic, Rochester, MN; University of Varese, Varese, Italy; and University of Florence, Florence, Italy. Patients were selected from institutional databases based on availability of archived bone marrow or peripheral blood DNA for mutation screening. Diagnosis of ET was according to the 2001 or 2008 WHO criteria [13, 14]. Previously published methods were used for CALR, JAK2, and MPL mutation analysis . Statistical analyses considered clinical and laboratory parameters obtained at time of diagnosis. Differences in the distribution of continuous variables between categories were analyzed by either Mann–Whitney (for comparison of two groups) or Kruskal–Wallis (comparison of three or more groups) test. Patient groups with nominal variables were compared by chi-square test. Overall and thrombosis-free survivals were calculated from the date of diagnosis to date of death or thrombosis (uncensored), respectively, or last contact (censored) and prepared by the Kaplan–Meier method and compared by the log-rank test. Cox proportional hazard regression model was used for multivariable analysis. P values less than 0.05 were considered significant. The Stat View (SAS Institute, Cary, NC) statistical package was used for all calculations.
Results and Discussion
For the test cohort, a total of 402 ET patients (median age 54 years; 60% females) were recruited from the Mayo Clinic (n = 299) and University of Varese (n = 103); mutational frequencies were 57% for JAK2, 28% for CALR, 3% for MPL, and 12% for “triple-negative.” Table 1 lists the clinical and laboratory features of the study population, stratified by mutational status. Risk distribution according to International prognostic scoring system for essential thrombocythemia (IPSET)  was low in 39% of patients, intermediate in 42%, and high in 19%. The results outlined in Table 1 are in line with previous observations that have shown association of CALR mutations and triple-negative mutational status, compared to JAK2 mutations, with younger age, lower haemoglobin, and lower leukocyte count [5, 6]. In addition, CALR mutations were associated with higher incidence of male sex (48% vs. 37% in triple-negative vs. 30% in JAK2 mutated cases) and higher platelet count. Also as previously shown, the incidence of thrombotic events was significantly lower in CALR-mutated and triple-negative patients, compared to JAK2 mutated cases. There was no difference in overall survival between CALR and JAK2 mutated cases (P = 0.45), but thrombosis-free survival was significantly better in the former (P = 0.04).
Table 1. Clinical and Laboratory Features of 402 Patients with Essential Thrombocytopenia Stratified by the Presence or Absence of CALR, JAK2, and MPL Mutations.
All patients n = 402
JAK2 mutated (n = 227; 57%)
CALR mutated (n = 114; 28%)
MPL mutated (n = 11; 3%)
Triple negative (n = 50; 12%)
IPSET: International Prognostic Scoring System for Essential Thrombocythemia.
The clinical and laboratory features of type 1 versus type 2 CALR-mutated versus JAK2-mutated patients are listed in Table 2. Compared to JAK2-mutated cases, type 2 CALR-mutated patients were younger (P = 0.001) and displayed lower hemoglobin level (P = 0.001), lower leukocyte count (P = 0.02), and higher platelet count (P < 0.0001). Type 1 CALR-mutated patients also displayed lower hemoglobin level (P = 0.0005), lower leukocyte count (P = 0.01), and higher platelet count (P = 0.008), compared to JAK2-mutated cases. Notably, compared to JAK2 mutations, type 1 (P = 0.005) but not type 2 (P = 0.44) CALR mutations were associated with male sex. The association between young age and type 1 CALR mutations did not reach statistical significance (P = 0.06), as it did for type 2 CALR mutations (P = 0.001). When type 1 and type 2 CALR mutations were directly compared to each other, patients with type 2 variants displayed significantly higher platelet count (P = 0.03; Table 1). The two CALR mutation variants were similar in their hemoglobin level (P = 0.87), leukocyte count (P = 0.95), and IPSET scores (P = 0.48). Overall (Fig. 1A) and thrombosis-free (Fig. 1B) survival was assessed for the Mayo Clinic patients only to ensure adequate follow-up time and showed no difference between the two mutant CALR variants.
Table 2. Clinical and Laboratory Features of 322 Patients with Essential Thrombocythemia with Type 1 (p.L367fs*46) or Type 2 (K385fs*47) Mutant CALR or JAK2V617F
All patients (n = 322)
JAK2 mutated (n = 227; 70%)
CALR Type1 (n = 51; 16%)
CALR Type2 (n = 44; 14%)
P value JAK2 vs. Type 1
P value JAK2 vs. Type 2
P value Type 1 vs. Type 2
IPSET: International Prognostic Scoring Essential Thrombocythemia.
The above-outlined findings were further explored in an independent series from the University of Florence, Italy, which included 625 patients with ET, diagnosed according to the 2008 WHO criteria . Of these 111 (18%) were CALR-mutated, including 58 (52%) type 1 and 34 (31%) type 2. Type 2 (vs. Type 1) CALR-mutated patients were younger (median age 49 vs. 59 years; P = 0.002) and displayed higher platelet count (median 987 vs. 739 × 10(9)/L; P = 0.002). In the Florence series, there was no difference between the two mutant CALR variants in gender distribution, leukocyte count, hemoglobin level, or thrombosis-free survival.
The observations from the current study are in line with recent communication from Vannucchi et al. regarding the preferential accumulation of both wild-type and mutant CALR in megakaryocytes . In the particular study, the authors used a polyclonal antimutant CALR antibody to immunostain bone marrow and showed primarily megakaryocyte staining in CALR-mutated patients but not in other MPN patients who are not mutated for CALR. Immunostaining for wild-type CALR also showed preferential accumulation of the protein in megakaryocytes, both in CALR mutated and unmutated patients and normal controls. Staining of myeloid and erythroid lineage cells was much less pronounced for both antimutant and antiwild type CALR. Preferential overexpression of CALR in megakaryocytes was also demonstrated by gene expression analysis . In relevance to the observations from the current study, gross inspection of the staining patterns of megakaryocytes in CALR-mutated patients suggested more intense staining with type 2 versus type 1 variant (personal communication from A.M. Vannucchi) .
Taken together, the above observations suggest a functional link between CALR and megakaryocytes that is differentially affected by specific CALR mutations. CALR mutations might use Janus kinase-signal transducer and activator of transcription (JAK-STAT)-independent mechanisms to contribute to the MPN phenotype, especially in view of the functional complexity of CALR and its role in multiple cellular processes . Accordingly, CALR might contribute to megakaryocyte maturation and platelet production through an indirect effect on endoplasmic reticulum (ER) homeostasis , and the degree of such an effect might vary depending on specific CALR variants.
The study from the Mayo Clinic was supported by the Mayo Clinic Harvey-Yulman Charitable Foundation for Myelofibrosis Tissue Bank and Clinical Database of Molecular and Biological Abnormalities. The study in Florence was supported by a special grant from Associazione Italiana per la Ricerca sul Cancro-“AIRC 5 per Mille”- to AGIMM, “AIRC-Gruppo Italiano Malattie Mieloproliferative” (#1005); for a description of the AGIMM project and list of investigators, see at www.progettoagimm.it. Supported also by FIRB (RBAP11CZLK) and PRIN (2010NYKNS7) to AMV.
None of the authors have any conflict of interest as it relates to the current manuscript. All authors participated in the writing of the manuscript and gave their approval to the final draft. All authors participated in patient care, information gathering, data analysis, pathology review or karyotype review.