Clinical Manifestation of Calreticulin Gene Mutations in Essential Thrombocythemia without Janus Kinase 2 and MPL Mutations: A Chinese Cohort Clinical Study.

BACKGROUND: Recently, calreticulin (CALR) gene mutations have been identified in patients with essential thrombocythemia (ET). A high-frequency of ET cases without Janus kinase 2 (JAK2) mutations contain CALR mutations and exhibit clinical characteristics different from those with mutant JAK2. Thus, we investigated the frequency and clinical features of Chinese patients of Han ethnicity with CALR mutations in ET.
METHODS: We recruited 310 Chinese patients of Han ethnicity with ET to analyze states of CALR, JAK2V617F, and MPLW515 mutations by polymerase chain reaction and direct sequencing. We analyzed the relationship between the mutations and clinical features.
RESULTS: CALR, JAK2V617F, and MPLW515 mutations were detected in 30% (n = 92), 48% (n = 149), and 1% (n = 4) of patients with ET, respectively. The mutation types of CALR involved deletion and insertion of base pairs. Most of them were Type 1 (52-bp deletion) and Type 2 (5-bp insertion, TTGTC) mutations, leading to del367fs46 and ins385fs47, respectively. The three mutations were exclusive. Clinically, patients with mutated CALR had a lower hemoglobin level, lower white blood cell (WBC) count, and higher platelet count compared to those with mutated JAK2 (P < 0.05). Furthermore, a significant difference was found in WBCs between wild-type patients (triple negative for JAK2, MPL, and CALR mutations) and patients with JAK2 mutations. Patients with CALR mutations predominantly clustered into low or intermediate groups according to the International Prognostic Score of thrombosis for ET (P < 0.05).
CONCLUSIONS: CALR mutations were frequent in Chinese patients with ET, especially in those without JAK2 or MPL mutations. Compared with JAK2 mutant ET, CALR mutant ET showed a different clinical manifestation and an unfavorable prognosis. Thus, CALR is a potentially valuable diagnostic marker and therapeutic target in ET.

INTRODUCTION

According to the 2008 World Health Organization (WHO) classification of hematopoietic tumors, essential thrombocythemia (ET) belongs to BCR-ABL1 negative myeloproliferative neoplasms (MPNs). The disease is characterized by stem cell-derived clonal myeloproliferation. A diagnosis requires meeting all the following criteria:[1] platelet (PLT) count ≥450 × 109/L; megakaryocyte proliferation with large and mature morphology; not meeting the WHO criteria for chronic myeloid leukemia, polycythemia vera, primary myelofibrosis, myelodysplastic syndromes, or other myeloid neoplasm; demonstration of JAK2V617F or other clonal markers or no evidence of reactive thrombocytosis. Therefore, JAK2V617F and MPL mutations have been characteristic molecular markers in ET since 2005 in addition to clinical features and hemogram and marrow picture, whose frequency is 50–60% and 5–10%, respectively. The above mutations explain the pathogenesis of ET from the Janus kinase/signal transducers and activators of the transcription (JAK/STAT) pathway and provide a theoretical foundation for targeted therapy.[234] However, about one-third of ET cases do not harbor these mutations. Although some other gene aberrations such as mutations in TET2, EZH2, DNMT3A, and ASXL1 have been detected and are associated with different cellular targets and epigenetic regulatory pathways, they are nonspecific and casual.[567] Fortunately, a new mutation in calreticulin (CALR) exon 9 was recently detected in MPNs (especially in ET without JAK2 and MPL mutations) through exome sequencing technology. This novel somatic mutation is associated with clinical features and prognoses of ET and might provide useful leads regarding tumorigenesis.[89] However, few studies have been done on ET patients of Chinese descent. In this study, we analyzed the status of the above mutations and their clinical and laboratory features in a large (n = 310) cohort of Chinese patients with ET.

METHODS

Subjects and inclusion criteria

Three-hundred and ten patients with ET were recruited from the Department of Hematology at Wuxi People's Hospital and Jiangsu Province Hospital between January 2007 and June 2015. All patients were diagnosed according to the WHO 2008 criteria for ET classification. We also recruited twenty healthy controls. Every patient provided informed consent for sample collection in accordance with the Declaration of Helsinki. The study was approved by the Ethics Committee of both hospitals.

DNA extraction and Janus kinase 2, MPL, and calreticulin mutation analyses

Mononuclear cells were extracted from peripheral blood and/or bone marrow of ET patients before treatment by Ficoll–Hypaque gradient centrifugation and cryopreserved on the day of sample collection. Genomic DNA was extracted from 5 to 10 × 106 mononuclear cells using the DNAzol kit (Invitrogen) (Norway Dynal Company) according to the manufacturer's instructions. Polymerase chain reaction (PCR) and clone sequencing were performed for detection of JAK2V617F and MPL exon 10 mutations as previously described.[10] Mutations in CALR exon 10 were also assessed by PCR followed by direct sequencing. A pair of primers was used to amplify a 306-bp product:forward 5’-ACAACTTCCTCATCACCAACG-3’ and reverse 5’-GGCCTCAGTCCAGCCCTG-3’. The CALR PCR conditions were as follows: 95°C for 5 min; 35 cycles of 95°C for 30 s, 58°C for 30 s, and 72°C for 60 s; and finally, 72°C for 7 min. The mutations were confirmed by at least two repeated analyses in this study.

Clinical data collection

The following clinical data were collected and analyzed: age, sex, peripheral blood count, event of major thrombosis, and International Prognostic Score for ET-Thrombosis system (IPSET-T).

Statistical analyses

Statistical analyses were performed with the statistical software GraphPad Prism 5.0 (GraphPad Software, USA). Discrete variables such as sex, event of major thrombosis, and IPSET-T were analyzed by the Chi-square test between different mutation groups and the wild-type group. Continuous variables including age and peripheral blood count were summarized by median and range and analyzed using the Mann–Whitney test. Overall survival (OS) analysis was considered from the date of diagnosis to the date of death (uncensored) or last contact (censored) and prepared by the Kaplan–Meier method. A value of P < 0.05 was considered statistically significant.

RESULTS

Frequency and pattern of the calreticulin mutation

Among the 310 ET cases, CALR mutations were identified in 92 (30%), JAK2V617F in 149 (48%), and MPLW515 in 4 (1%) cases, whereas 65 (21%) cases were negative for all three mutations (wild-type) [Figure 1a]. All four MPL-mutated cases were the MPLW515 mutation. The three mutations were exclusive in ET. We did not observe the CALR mutation or the other two mutations in healthy controls. In patients without JAK2V617F and MPL (wild-type), the frequency of CALR mutations was much higher at 59% (92/157). All 11 CALR mutation types were detected in our ET cases, including nucleotide deletion or insertion. The most common mutation was Type 1 (c.1092_1143del 52-bp; 50%, 45/92) leading to L367fs*46, followed by Type 2 (c.1154_1155insTTGTC; 35%, 32/92) leading to K385fs*47 [Figures 1b and 2a].

Figure 1

(a) Distribution of Janus kinase 2, MPL, and calreticulin mutations in essential thrombocythemia. (b) Different mutation types in essential thrombocythemia with calreticulin mutation.

Figure 2

(a) Comparison between wild-type and mutant calreticulin gene sequences (deletion and insertion nucleotide in red). (b) Structure and function of calreticulin protein.

Relation between clinical phenotype and different mutations

We divided all cases into three subgroups according to the different mutation types (MPL-mutated cases were eliminated because of the small sample size): CALR-mutated, JAK2V617F, and wild-type groups. Table 1 shows the clinical and laboratory characteristics of the 306 ET cases. As shown in Table 1, cases with mutated CALR had a lower hemoglobin (Hb) level, lower white blood cell (WBC) count, and higher PLT count, as well as a higher chance of being younger than those with mutated JAK2 (P < 0.05). No significant differences were observed in Hb level, PLT count, or age between the CALR-mutated and wild-type groups; however, a significant difference was found for WBC count. Similar clinical features were observed between the Type 1 or Type 2 groups and the JAK2V617F group, and there was no significant difference in clinical and laboratory features between the Type 1 and Type 2 CALR mutation groups [Table 2].

Table 1

Clinical features of ET patients with CALR and JAK2 mutation

Characteristics	CALR	JAK2V617F	Wild-type	P
				CALR versus JAK2V617F	CALR versus wild-type
Number	92	149	65
Sex (male:female)	44:48	73:76	33:32	0.89	0.74
Age (years)	55 (20–85)	62 (23–95)	45 (18–80)	<0.001	0.08
WBC (×109/L)	8.2 (5.8–12.9)	15.6 (4.5–111.3)	9.5 (3.2–33.7)	<0.001	<0.001
Hb (g/L)	126 (100–167)	149 (76–212)	121.5 (79–221)	0.002	0.36
PLT (×109/L)	996 (465–1941)	601 (460–1602)	629 (567–1881)	<0.001	0.26
Thrombotic event (%)	1	2	4.6	1.0	0.3075
IPSET-T (n)				<0.001	<0.001
Low-risk	58	25	25	–	–
Intermediate risk	30	68	19	–	–
High-risk	4	56	21	–	–

ET: Essential thrombocythemia; IPSET-T: International Prognostic Score of thrombosis in World Health Organization–essential thrombocythem; WBC: White blood cell; PLT: Platelet; Hb: Hemoglobin; JAK2: Janus kinase 2; CALR: Calreticulin; –: Not available.

Table 2

Clinical features of ET patients with Types 1 or 2 of CALR and JAK2 mutation

Characteristics	Type 1	Type 2	JAK2V617F	P
				Type 1 versus Type 2	Type 1 versus JAK2V617F	Type 2 versus JAK2V617F
Number (male/female)	45 (22:23)	32 (17:15)	149 (73:76)	0.81	1	0.7008
Age (years)	52 (20–78)	42 (22–80)	62 (23–95)	0.065	0.020	<0.001
WBC (×109/L)	7.9 (5.8–12.9)	6.8 (6.1–10.9)	15.6 (4.5–111.3)	0.159	<0.001	<0.001
Hb (g/L)	119 (100–167)	122 (102–157)	149 (76–212)	0.671	0.002	0.025
PLT (×109/L)	768 (465–1941)	890 (750–1342)	601 (460–1602)	0.095	0.185	0.005
Thrombotic event (%)	0	0	2	–	1.0	1.0
IPSET-T (n)				0.246	<0.001	<0.001
Low-risk	26	24	25	–	–	–
Intermediate risk	18	7	68	–	–	–
High-risk	1	1	56	–	–	–

ET: Essential thrombocythemia; IPSET-T: International Prognostic Score of thrombosis in World Health Organization–essential thrombocythem; WBC: White blood cell; PLT: Platelet; Hb: Hemoglobin; JAK2: Janus kinase 2; CALR: Calreticulin; –: Not available

Thrombotic event, overall survival, and International Prognostic Score for Essential Thrombocythemia- Thrombosis between different mutations

According to our data [Table 1 and Figure 3], there was no significant difference in thrombotic events and OS among the three groups. According to the IPSET-T stratification, a close relationship of the CALR mutations to the low- or intermediate-risk group was observed (P < 0.05).

Figure 3

Kaplan–Meier estimates of overall survival in 306 patients with essential thrombocythemia stratified by mutation type.

DISCUSSION

ET belongs to MPN and is characterized by thrombocytosis and other features including leukocytosis, splenomegaly, thrombosis, bleeding, microcirculatory symptoms, pruritus, and risk of leukemic or fibrotic transformation. The median survival is 20 years for ET, and the leukemic transformation rate at 20 years is estimated to be <5%.[1112] Thrombosis and bleeding are the major causes of death.[1314] Identification of JAK2V617F and MPL mutations has been applied widely for accurate diagnosis of about 50–60% cases of ET since 2005, whereas CALR mutations seem to play an important role in cases without JAK2V617F and MPL mutations as observed since 2013.[89]

CALR is located on chromosome 19p13.2 and is 5891-bp long with nine exons. The encoded protein has 417 amino acids and consists of three main structural and functional domains [Figure 2b]. As a protein with different functions,[15] CALR exists both in the endoplasmic reticulum (ER) and outside the ER including on the cell surface and in the cytoplasm, nucleus, and extracellular matrix. Within the ER, the main physiological function of the protein includes chaperone activity and modulation of calcium homeostasis; outside the ER, it is associated with the healing of cutaneous wounds, immunogenic response, and regulation of cell adhesion, translation, gene expression, and nuclear export.[161718] Thus, CALR is crucial in cellular activity, and its dysfunction is associated with the disruption of normal cellular physiological and pathological processes, which could lead to oncogenic consequences.[19] Owing to frameshift mutations, proteins encoded by mutated CALR harbor a loss of the C-terminal KDEL domain (Lys-Asp-Glu-Leu), generating a novel peptide sequence terminating with the same 36 amino acids. This alteration transforms the negatively charged glutamic acid-rich C-terminus of wild-type CALR to a positively charged arginine-rich region.[2021] Therefore, the mutant protein may influence Ca2+ binding and normal cell growth. Using retroviral production and transfection of Type-1 CALR-mutated cDNAs into the Ba/F3 cell line, interleukin-3-independent growth, and STAT5 activation have been observed. This suggests that mutant CALR proteins may disrupt cell growth via the STAT pathway.[8]

The total prevalence of CALR mutations in ET is 15–30%, whereas in ET without JAK2 and MPL mutations, CALR mutation frequency ranges from 50% to 70%. The mutations are mutually exclusive of JAK2 and MPL mutations in ET. The mutation type of CALR involves deletion and insertion of a certain number of base pairs, leading to a shift in the reading frame. Most of them are Type 1 (52-bp deletion) and Type 2 (5-bp insertion, TTGTC) mutations, leading to del367fs46 and ins385fs47, respectively. Other complex mutations such as deletion-merging insertion, which might cause a constitutive change, can also be detected.[89] In our study, the total frequency and mutational hotspots were consistent with western countries. However, the frequency of Chinese patients without JAK2 and MPL mutations seemed to be low when compared to that of patients in England (71.4% 80/112; P < 0.05) and complex mutations were not detected.[9] We attribute this to two possible reasons: the different samples or characteristics of the Chinese population.

Clinically, like in previous studies,[22] our CALR- mutated ET patients presented with younger age, lower Hb level, lower WBC count, and higher PLT count compared to patients with the JAK2 mutation. The two major CALR mutant types also displayed the same features compared to patients with the JAK2 mutation. No significant differences were observed in peripheral blood count, age, or sex between the Type 1 and Type 2 mutation groups. This showed that although there were various forms of CALR mutations, their clinical laboratory parameters were similar, which was in contrast to observations in patients with the JAK2 mutation. In addition, patients with CALR mutations had lower a lower risk of a thrombosis, according to our data, although a statistical difference was not observed, whereas some foreign research data have been documented.[23] Therefore, granulocyte/PLT activation and increased WBC counts may be more relevant to the pathogenesis of thrombotic complications than absolute PLT counts.

In terms of ET prognosis, the following three aspects are currently evaluated: OS, acute myeloid leukemia (AML) evolution, and thrombosis-free survival (TFS). The earlier OS models considered three risk factors – age (≥60 years), WBC count (>15 × 109/L or 11 × 109/L), and anemia – and categorized patients into low-risk (no risk factor), medium-risk (1 risk factor), and high-risk groups (≥2 risk factors). AML evolution categorizes patients into these groups based on PLT >1000 × 109/L and anemia: low-risk (no risk factor), medium-risk (1 risk factor), and high-risk groups (2 risk factors).[242526] In 2012, Passamonti et al. proposed the IPSET system based on analysis of 867 cases of WHO-diagnosed ET. This model includes three parameters: age (2 points for ≥60 years) while blood cell count (1 point for ≥11 × 109/L) and previous thrombotic events (1 point). Patients were then categorized into low-risk (sum = 0; median survival not reached), medium-risk (sum = 1–2; median survival 24.5 years), and high-risk (sum = 3–4; median survival 13.8 years) groups based on overall scores to evaluate different OS of patients in different risk stratifications. This model is both suitable for the OS evaluation of ET corresponding to the WHO diagnostic standard and those that do not meet the WHO diagnostic standard but have suitable clinical presentations, as well as TFS prognosis evaluation.[27]

TFS prognosis evaluation in ET patients previously considered two factors—age (≥60 years) and previous thrombotic events—to categorize patients into low-risk (no risk factor) and high-risk groups (more than 1 risk factor).[28] Along with evidence that JAK2 mutations and cardiovascular risk factors are closely linked to thrombotic events,[2930] a new thrombosis prognosis model, IPSET-T was established. This model included JAK2 mutations (2 points), cardiovascular risk factors (1 point for tobacco use, hypertension, or diabetes mellitus), age (1 point for ≥60 years), and previous thrombotic events (2 points) and categorized patients into low-, medium-, and high-risk groups. Based on this model, almost 50% of low-risk patients were re-grouped into the new medium-risk (47%) or high-risk groups (5%), while nearly two-thirds of the previously high-risk patients were now in the low- or medium-risk groups (one-third each). This categorization resulted in three groups of patients with distinctly different TFSs and presented the different thrombosis event risks of patients who have different clinical and biological characteristics.[31] According to our short follow-up period (median follow-up time was 102 months), no difference was observed between different mutation groups on OS. However, current reports about OS are not consistent. Klampfl et al. show that patients with CALR mutations had better OS than those with JAK2 mutations, whereas no difference was observed according to Nangalia et al. and Rotunno et al.[8920] Recently, the long-term follow-up study of 266 ET cases by Tefferi et al. showed no difference in the OS of patients with CALR and JAK2 mutations even when analyzed in different age groups. In addition, the two groups of patients showed no significant difference in leukemia prognosis and myelofibrosis.[32] In comparison, Palandri et al. found from long-term follow-up of 217 ET patients who were <40 years old that those with CALR mutations had better OS and TFS than those with JAK2 mutations.[33] Thus, the effect of CALR mutations on the OS of ET patients requires further clarification with larger sample sizes and longer follow-up time in multi-race clinical studies. Although OS of our data had no statistical difference, the patients were divided into low-, medium-, and high-risk groups based on IPSET-T, and the different mutation distributions were compared in these groups. CALR mutations mainly distributed in the medium- and low-risk groups while JAK2 mutations showed a focused distribution in the medium- and high-risk groups. This could be indirect evidence of the effect of CALR mutations on prognosis. However, a recent study found no independent effect of CALR mutations on prognosis in a multivariate analysis of TFS and had no substantial guiding significance to the IPSET-T model.[34] Further clarification from additional studies with larger sample sizes is needed.

In summary, CALR mutations are novel somatic mutations in ET. This finding will assist with understanding ET that does not involve JAK2 and MPL mutations. Compared to ET with JAK2 mutations, ET with CALR mutations in Chinese patients showed a different clinical manifestation and different prognosis, which is similar to results from western countries. CALR might be a valuable diagnostic maker and therapeutic target in ET[35] leading to the development of innovative diagnostic criteria and therapies for MPN patients.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.