Study of CALR, MPL, and c-kit Gene Mutations in Thai Patients with JAK2 V617F Negative Myeloproliferative Neoplasms

Objective: The aim of this study to determine the prevalence of CALR, MPL and c-kit gene mutations in JAK2 V617F negative-MPN patients. Methods: The retrospective study of CALR, MPL and c-kit mutations were analyzed in 113 samples collected from March 2010 to May 2017 and identified as JAK2 V617F–negative MPN Thai patients. The samples were analysis by gel electrophoresis and direct sequencing. Results: 28.3% of JAK2 V617F–negative MPN patients showed CALR gene mutations. Within the MPN patients with CALR mutation, 46.9% were classified as essential thrombocythemia (ET) and 20.9% were classified as primary myelofibrosis (PMF). Previous studies classified CALR mutations into three types using negatively charged amino acid stretches at the C-terminal domain. Type 1-like mutations were observed in 12 of 49 (24.5%) ET patients and type 2-like mutations were observed in 10 of 49 (20.4%) patients. In addition, 8 of 43 (18.6%) PMF patients showed type 1-like mutations and 1 of 43 (2.3%) showed type 2-like CALR mutation. Interestingly, platelet counts were higher in patients with CALR gene mutation than in patients without CALR gene mutation. MPL mutations (W515K and W515L) were identified in 2 of 109 (1.8%) MPN patients; the MPL mutations were only found in ET patients, which was consistent with previous studies. We did not detect exon 17 c-kit mutation in JAK2-negative MPN patients but detected intronic single nucleotide polymorphisms at c.74,978 and c.75,255 in these samples. Approximately 66% of patients did not have mutations in CALR and MPL genes, in addition to lacking JAK2 gene mutation, and these cases are classified as triple-mutations. Conclusion: Our results showed that 66% of cases were triple-negative mutation MPN because they lacked mutations in JAK2, CALR and MPL genes. The frequencies of CALR and MPL mutation in this study are similar to other CALR and MPL patient data.


Introduction
is detected in about 95% of PV patients (Campbell and Green, 2006). Moreover, mutation in JAK2 exon 12 occurs in approximately 3% of PV patients (Campbell et al., 2005), while 5% of ET and PMF patients negative for JAK2 mutation carry myeloproliferative leukemia virus oncogene (MPL) mutation (Scott et al., 2007). Over 50 MPL mutations have been identified, and the most common mutation occurs at W515 in exon 10 (Bennett and Stroncek, 2006;Pardanani et al., 2006;He et al., 2013). MPL is a thrombopoietin receptor that regulates thrombopoiesis and hematopoietic stem cell maintenance (Ihara K et al.,1999). Mutation in the MPL gene provides an additional mechanistic explanation for JAK-STAT activation in MPN (Pikman, Y. et al.,2006;Greenfield et al., 2021).
Systemic mastocytosis with an associated hematological neoplasm (SM-AMS) is detected in >90% of patients with myeloid neoplasms including myeloproliferative syndrome (SM-MPS) and myeloproliferative neoplasm (SM-MPN) (Pardanani, 2013). Approximately 90% of patients with SM also harbor mutation in KIT, and the most common mutation occurs at D816 (Gleixner et al., 2011).
A study in 2013 using next-generation sequencing revealed a mutation in exon 9 of the CALR gene in many ET or PMF cases negative for JAK2 mutation. Over 36 CALR mutation types have been identified in MPN patients (Klampfl et al., 2013;Nangalia et al., 2013). The majority of CALR mutations are 5 bp insertion and 52 bp deletion. Both types are classified as type I and type II mutations. CALR functions as a chaperone protein and maintains Ca 2+ homeostasis. The interaction between mutant CALR and MPL protein contributes to the induction of cellular transformation (Araki and Komatsu, 2017). Moreover 7.3% of JAK2 V617F mutation patients were also identified CALR mutation while the major of co-mutation patients represented low alleles burden of JAK2 V617F (< 1% allele) (Makarik et al., 2021). In one report, triple-negative JAK2, CALR and MPL mutations were found in ET and PMF patients. Moreover, the triple-negative mutations were found in younger patients compared with cases with those mutations (Shirane et al., 2015). The overall survival and leukemia-free survival in PMF patients are worse than PMF patients with JAK2, CALR or MPL mutation (Tefferi et al., 2014).
In this study, we determined the frequencies of CALR, MPL and c-kit mutations in MPN patients and compared these results with clinical data from MPN patients without JAK2 V617F mutation at our institution.  primers. PCR reactions were as follows: denaturation at 94°C for 4 min, followed by 35 cycles of 94°C for 30 sec, 55°C for 30 sec and 72°C for 30 sec, and extension at 72°C for 5 min. PCR amplification of exon 17 of the c-kit gene was performed using nck17-F (5'TGTGAACATCATTCAAGGCG3') and nck17-R (5'TCCTGCTGTGACCTTCAATG3') primers. PCR reactions were as follows: denaturation at 95°C for 7 min, followed by 30 cycles of 94°C for 30 sec, 64°C for 30 sec and 72°C for 1 min, and extension at 72°C for 5 min. PCR amplification of exon 10 of MPL gene was performed using MPL10-F (5'TAGCCTGGATCTCCTTGGT3') and MPL10-R (5'AGAGGTGACGTGCAGGAAGT3') primers. PCR reactions were as follows: denaturation at 95°C for 2 min, followed by 36 cycles of 94°C for 30 sec, 64°C for 30 sec and 72°C for 30 sec, and extension 72°C for 2 min. The PCR products were examined by 2% agarose gel electrophoresis. Direct sequencing of PCR products was performed with the ABITM 3500xL Genetic Analyzer (Applied Biosystems, Thermo Fisher Scientific ® , MA USA).

Statistical Analysis
Associations between mutations and demographic data were analyzed by Wilcoxon rank-sum (Mann-Whitney) test or T-test.

Results
Blood and bone marrow samples from 113 JAK2 V617F-negative MPN patients were examined for CALR exon 9, MPL exon 10 and c-kit exon 17 mutations ( Table  1). The overall patient group included 49 ET patients, 43 PMF patients and 21 PV patients.
In the overall group of JAK2 V617F-negative MPN patients, 32 cases (28.32%) harbored a CALR gene mutation. Among the 43 ET patients, 23 ET patients had CALR gene mutations; 12 mutations were deletions and 11 mutations were insertions. Among the PMF patients, we found 9 patients with CALR gene mutation, which included 8 patients with deletions and 1 patient with an insertion. No PV patients showed mutation in CALR.
We evaluated the predicted amino acid sequences from the identified CALR gene mutations. For 32 PMF patients that have CALR mutation, there are 8 CALR mutation types were found in this study. Four mutation types were categorized as type 1-like mutation group, 3 mutation types were types 2-like mutation group from and 1 mutation type were other type group mutations ( Figure  1 and 3). We also found the extra-band show as lane 3 in agarose gel of Figure 1, then we analyzed the sample using DNA sequencing. We found that the extra-band contain both 52 bp deletion mutant allele and wild-type allele of CALR gene as shown the details in Figure 2.
We found significantly differences in platelet count (Plt) between ET patients with wild-type CALR and ET patients with CALR mutation (808.5x10 9 /L vs. 1395x10 9 /L, p<0.001) ( Table 2). Hemoglobin (Hb), hematocrit (Hct) and white blood cell count (WBC) were similar in the two groups.
We identified two cases (1.83%) with MPL gene mutations in the overall 109 MPN patients. Both cases with MPL gene mutations were ET patients. The two MPL mutations were W515K and W515L mutations (Figure 4).
DNA sequencing for c-kit exon 17 mutation was   performed for 40 ET, 38 PMF and 10 PV patients. We did not detect c-kit exon 17 mutation in any patient. However, we found single nucleotide polymorphisms (SNPs) at intron positions between exon 16 and exon 17 as well as between exon 17 and exon 18: c.74,978 and c.75,255 ( Figure 5). For the SNP at c.74,978, the heterozygous G/A allele was found in 8 patients and the homozygous G/G allele was found in 80 patients. For the SNP at c.75,255, the heterozygous T/C allele was found in 40 patients while the homozygous T/T allele was found in 3 patients and the homozygous C/C allele was found in 45 patients. There were no significant differences in Hb, Hct, Plt and WBC between patients of each SNP type (Table 2).

Discussion
There are limit data of JAK2 V617F mutation in Thai patients. Our previous study showed that JAK2 V617F mutation of 186 Thai MPN patients were 49% for PV, 54% for ET and 43% for PMF patients (Bunyoo et al., 2018). The other study showed that JAK2 V617F was detected in 41 from 58 patients (71%) of ET patients (Limsuwanachot et al., 2017)   PV patients in this study showed CALR mutation. The incidence of CALR mutation in our overall MPN patient group was similar to those reported in other Caucasian and Asian populations (Rumi et al., 2014;Kim et al., 2015;Shirane et al., 2015;Vu et al., 2019) (Table 3). Pietra et al, (2016). classified the CALR mutations into three types based on the three negatively charged amino acid stretches at the C-terminal domain. Type 1-like CALR mutations group result in the loss of two negatively charged amino acid stretches, while the type 2-like mutations group maintain all three negatively charged amino acid stretches. And the other type group mutations resulted in the loss of one negatively charged amino acid stretch. Type 2-like CALR mutations group are mainly associated with an ET phenotype, a low risk of thrombosis and indolent clinical course, while type 1-like mutations group are mainly associated with a myelofibrosis phenotype and a high risk of progression from ET to myelofibrosis. The PML patients (without JAK2 V617F mutation) with type 1-like CALR mutation    were good prognosis than patients with type 2-like CALR mutation. Whereas the prognosis in PML patients (without JAK2 V617F mutation) with type 2-like CALR mutation were not different with PML patients contained only JAK2 V617F mutation (Guglielmelli et al., 2015). MPN patients with CALR mutation treated with Ruxolitinib, a JAK2 inhibitor, showed decreased spleen size and increased overall survival compared with patients treated with the best available therapy (Guglielmelli et al., 2016). Furthermore, Fedratinib, a JAK inhibitor, reduced both spleen size and Plt count in myelofibrosis patients with CALR mutation (Passamonti et al., 2014). In our study, the Plt count of ET patients with CALR mutation was higher than ET patients with wild-type CALR, and these findings were similar to previous reports in other populations (Klampfl et al., 2013;Nangalia et al., 2013;Shirane et al., 2015;Fu et al., 2014) (Table 3).
The c-kit gene proto-oncogene encodes a tyrosine kinase that is expressed in mast cells and hematopoietic stem cells (Yarden et al., 1987). The c-kit mutations are found in CML and CMML patients at a prevalence rate of approximately 1%-4%. Almost c-kit mutation in CML and CMML associated with systemic mastocytosis case (Valent et al., 2010). Over 90% of systemic mastocytosis patients carry the c-kit D816V mutation (Nagata et al., 1995;Valent et al., 2017). In this study, we did not find this c-kit mutation at exon 17 in the overall MPN patients. However, we detected SNPs at introns between exon 16 and 17 as well exon 17 and 18. We did not observe any relationship between the intronic variations and hematological parameters including Hb, Hct, Plt and WBC. Because of the limited data of mastocytosis in our study, the result of no c-kit mutation in sample may from MPN patients have no mastocytosis. These data are similar to the recent report showing that c-kit mutations were only found in systemic indolent mastocytosis cases, while no c-kit mutations were observed in aggressive mastocytosis and myelodysplastic syndrome patients (Fritsche-Polanz et al., 2001).
MPN patients are identified MPL mutations approximately 1% to 8% of patients. Almost MPL mutation found in ET patients whereas this mutation rarely found in PMF patients. The majority of MPL mutations are W515L/K (Kim et al., 2015;Shirane et al., 2015;Shams et al., 2018). These findings are consistent with our results showing MPL gene mutations in two ET patients (W515K and W515L). Because of the small numbers of MPL mutant cases, we were unable to compare hematological parameters between wild-type and mutant. In addition, there are previous studies reported CALR mutation patients showed higher Plt count than patients who carry MPL mutation (Kim et al., 2015;Shirane et al., 2015;Shams et al., 2018;Tefferi et al., 2014). In conclusion, 70% of the overall MPN patients, which were negative for JAK2 gene mutation, did not have CALR and MPL gene mutations; this patient group can be classified as triple-negative (TN) mutation MPN cases. Next-generation sequencing has been used to identify other gene mutations in triple-negative MPN patients. Whole-exome sequencing in TN MPN cases identified other mutations, including MPL T119I, S204F, E230G, P453R, R537W, S505N, R537W, and Y591D and germ-line mutations of MPL and JAK2 including MPL S204P, V285E and R321W and JAK2 F556V and V625F (Milosevic et al., 2016). Interestingly, the JAK2 V617F mutation was detected in a TN ET patient by nextgeneration sequencing (Alimam et al., 2021), suggesting that highly sensitive methods may be necessary for the diagnosis of low-level mutations in MPN patients. A JAK1/2-inhibitor, Ruxolitinib (RUX), was used for MPN patients with JAK2 V617F or CALR mutation. But the overexpress of MPL leads to RUX-resistance in CALR mutation cell line (Yasuda et al., 2022). The vaccine called "CALRLong36" contained CALR exon 9 mutant peptides, was tested in CALR mutation MPN patients. The result showed that the vaccine could activate the immune respond in PV patients but it could not activate in PMF patients (Handlos et al., 2021). According to previous data shown by various studies that CALR and MPL mutations were related to the response of treatment and prognosis of the patients. Although, there are low prevalence of CALR exon 9, MPL exon 10 mutations, the analysis of these mutation is still necessary for diagnosis and treatment in Thai patients.