Detecting Multiple Driver Mutations in a Patient with Essential Thrombocythemia

Background

According to the World Health Organization classification, Philadelphia-negative myeloproliferative neoplasms (MPNs) include polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF) [1]. These disorders have a similar molecular basis, Janus kinase 2 gene mutation (JAK2), which is found in 50–60% of patients with ET or PMF [2]. A somatic mutation in the myeloproliferative leukemia virus oncogene (MPL) is found in 5–10% of patients with ET or PMF [3]. In ET patients with non-mutated JAK2 or MPL, a somatic mutation of CALR, the gene encoding for calreticulin, has been identified [4].

JAK2 is a nonreceptor tyrosine kinase that plays a significant role in transducing signals from several class 1 cytokine receptors that are essential for myelopoiesis. When a ligand binds to its specific receptor (for example, erythropoietin to the erythropoietin receptor), it induces a structural change that leads to the activation of JAK2 signaling. Activated JAK2 induces the phosphorylation of itself and other proteins that serve as binding sites for further downstream signaling proteins. Multiple negative feedback mechanisms attenuate signaling. The substitution of the normal valine residue at position 617 by phenylalanine in exon 14 of JAK2 (JAK2 V617F) activates JAK2 signaling without class I receptor stimulation. JAK2 V617F also escapes the normal negative regulation, further contributing to myeloproliferation.

CALR encodes for the multifunctional protein calreticulin, found in the cytosol, endoplasmic reticulum, and cell membrane. It has 3 distinct domains: N, P, and C-domains. The C-domain is responsible for calcium hemostasis in the endoplasmic reticulum. The KDEL sequence in the C-domain helps retain the protein in the endoplasmic reticulum. Mutations in CALR lead to the loss of most of the C-domain, multiple calcium binding sites, and the KDEL sequence. There are 2 types of mutations in the CALR gene: Type 1 and Type 2. Type 2 mutations are more commonly seen in patients with ET. The exact mechanism by which mutations in CALR cause myeloproliferative tumorigenesis is unknown.

CALR mutations are thought to be mutually exclusive with mutations in JAK2 or MPL [5,6]. We report a rare presentation of co-existing JAK2 and CALR mutations in a patient with ET and delineate the steps taken to identify these mutations.

Case Report

A 55-year-old African American woman with a medical history of chronic migraine and hyperlipidemia presented to the hematology clinic for evaluation of an elevated platelet count noted during a previous hospitalization for renal stones. The patient had no history of thrombotic events or vasomotor symptoms like lightheadedness, headache, or erythromelalgia. She had no family history of any clots or bleeding disorders. Physical examination was unremarkable and did not show splenomegaly. Laboratory workup showed a platelet count of 649 k/uL, a white blood cell count of 9.45 k/uL, hemoglobin of 12.4 g/dL, and a hematocrit of 36.8%. The iron panel showed no evidence of iron deficiency. Initial testing was performed of peripheral blood for JAK2 V617F mutation analysis with a reflex to CALR followed by MPL in case the JAK2 mutation was not identified. Quantitative, allele-specific polymerase chain reaction (PCR) was used. The patient had a positive JAK2 V617F mutation, which was measured at 2% of the total JAK2 DNA. The test did not reflex to look for the CALR and MPL mutations. The patient had a bone marrow biopsy, which showed a hypercellular bone marrow with trilineage hematopoiesis (Figure 1), a myeloid-to-erythroid ratio of 2: 1, megakaryocytes that appeared focally increased, and dysmorphic on Wright-Giemsa staining (Figure 2), and Factor VIII-related antigen staining; no fibrosis was present. Next-generation DNA sequencing was performed on the bone marrow sample. It was positive for a frameshift mutation in the CALR gene located on chromosome 19q13.2, resulting in a 52-base-pair deletion (Chr19(GRCh37): g.13054572_13054623del; NM_004343.3(CALR): c.1099_1150del; p.Leu367Thrfs*46). The variant allele frequency for CALR mutation was 7%. The next-generation sequencing did not show a JAK2 mutation. Because of the inconsistent bone marrow and peripheral blood findings, testing for JAK2 and CALR mutations was repeated in the peripheral blood and it was confirmed they were both positive. The patient was started on aspirin 81 mg for thromboprophylaxis. The patient has been following with the hematology clinic at our institution for the last 2 years. Her platelet count has remained between 600 and 800 K/uL, and she has not had any thrombotic events.

Discussion

Our understanding of the genetic basis of essential thrombocytosis is expanding as various mutations are identified [7]. Different driver mutations lead to different disease pheno-types and carry prognostic significance.

The JAK2 and CALR mutations are thought to be mutually exclusive in patients with ET. This could be attributed to the low sensitivity of the methods used for detection of multiple mutations in patients in previous studies. Assays that can be used for detection of JAK2 mutation include genomic direct DNA sequencing, RT-PCR-sequencing, PCR- amplification refractory mutation systems (ARMS), PCR-restriction fragment length polymorphism, real-time PCR, DNA-melting curve analysis, allele blocker PCR, and allele-specific quantitative PCR (AS-PCR). [8] Allele-specific amplification combined with real-time PCR is the most widely used assay and has high sensitivity. The allele burden of JAK2 V617F in patients with ET has been found to be lower in comparison to patients with PV and PMF [9,10]. Therefore, a test with high sensitivity should be utilized in detecting JAK2 mutations in ET patients. In our patient, JAK2 mutation was not identified by the next-generation sequencing due to test’s low sensitivity.

In a study comparing the clinical and hematologic features of the patients with JAK2 and CALR mutations, it was seen that the patients with CALR mutations are younger, have a lower hemoglobin level, a lower white blood cell count, a higher platelet count, and a lower erythropoietin level as compared to the patients with JAK2 mutation [5]. Patients with CALR-mutated ET had a lower incidence of thrombotic events than patients with JAK2-mutated ET. There was no significant difference between the 2 groups in the progression to myelofibrosis.

There is limited data on the clinical implications of the co-existence of 2 driver mutations. In a study by Kang et al, 4% of the ET patients had both JAK2 and CALR mutations. They concluded that the JAK2 mutation has a more significant effect on the disease phenotype regardless of the CALR mutation [13]. Patients without mutated JAK2 showed better progression-free survival than those with mutated JAK2, irrespective of CALR mutation, but this finding was not statistically significant. In a French intergroup of myeloproliferative neoplasms study on MPN patients with 2 driver mutations, patients harboring 2 driver mutations had low JAK2 V617F allele burden (median 1% vs 26%, P<.001) [14]. They concluded that in ET, predominant mutation dictates the disease phenotype. Usseglio et al analyzed the association of multiple mutations in patients with MPN. They reported 8 out of 103 ET patients had CALR or MPL mutations in addition to JAK2 mutation [15]. All these patients had a JAK2 allele burden of <4%, which is similar to our patient. The patients had higher platelet counts in comparison with JAK2 and CALR single-mutated patients. They concluded that CALR and MPL mutations should be tested not only in patients with negative JAK2, but also in low allele burden JAK2 mutation.

Clonal hematopoiesis of indeterminate potential (CHIP) is characterized by the presence of a driver mutation at a variant allele frequency (VAF) of at least 2% in the peripheral blood and in the absence of diagnostic criteria for hematological malignancy [16,17]. The JAK2 mutation identified in our patient had a VAF of 2% and could be classified as CHIP. CALR mutation identified by NGS in the bone marrow could be the predominant mutation dictating the patient’s phenotype.

With the recent advances in targeted therapies and the discovery of a new antibody that can specifically target CALR function in MPNs, it becomes important for clinicians to test for all known mutations to ensure patients do not miss out on treatment options [18].

Conclusions

In our patient, initial testing was done to test for the JAK2 mutation, and if the test were negative, only then would the testing for CALR and MPL mutations have been done. This is a widespread practice in many centers for better utilization of resources, since these mutations are considered mutually exclusive. Because of that, the co-existence of the mutations may be underreported. Testing for multiple driver mutations can be considered in patients with low JAK2 V617F allele burden. This will help to determine the true incidence of these mutations and will provide more information on the clinical presentation and outcomes in such patients. Tests with high sensitivity should be utilized for the detection of JAK2 mutations in ET patients. If the JAK2 V617F allele burden is ≤2%, the possibility of clonal hematopoiesis of indeterminate potential should be considered because, then, the co-existing dominant mutation could guide disease phenotype and management.