The TT Genotype of the KIAA1524 rs2278911 Polymorphism Is Associated with Poor Prognosis in Multiple Myeloma

Background: The KIAA1524 gene encodes an oncoprotein, CIP2A, which inhibits the phosphorylation of the Akt kinase B, stabilizes the c-Myc protein, and, through that, promotes cancerogenesis. An increase in CIP2A expression has been observed in numerous solid tumors and hematologic malignancies, including multiple myeloma (MM). The aim of our study was to evaluate the clinical impact of the functional single nucleotide polymorphisms (SNP) of the KIAA1524 gene (rs2278911, 686C > T) in MM patients. Methods: The study group consisted of 128 patients with de novo MM. EDTA venous blood samples were collected prior to the treatment. The SNPs were analyzed by Real-Time PCR with the use of specific Taqman probes. Results: Multivariable analysis revealed that variables independently associated with shorter progression-free survival (PFS) included thrombocytopenia, delTP53 and IGH/CCND1 translocation and the TT genotype of the KIAA1524 gene (686C > T) (median PFS: 6 vs. 25 months; HR = 7.18). On the other hand, autologous haematopoietic stem cell transplantation (AHSCT) was related to a lower risk of early disease progression. Moreover, light chain disease, International Staging System (ISS) 3, poor performance status, hypoalbuminemia, IGH/FGFR3 translocation and the TT genotype of the KIAA1524 gene (686C > T) were independent prognostic factors associated with shorter overall survival (OS) (median OS: 8 vs. 45 months; HR = 7.08). Conclusion: The evaluation of the SNP 686C > T of the KIAA1524 gene could be used as a diagnostic tool in MM patients at risk of early disease progression and death.


Introduction
Multiple myeloma (MM) is a malignant neoplasm characterized by the clonal proliferation of atypical plasma cells within the bone marrow, and accounts for around 10% of all hematologic malignancies [1]. The risk factors for MM include age, race, sex, family history, chronic inflammation, immunodeficiency and exposure to radiation and organic solvents [2][3][4][5]. Patients may experience a wide range of unspecific and usually mild symptoms, including fatigue, nausea, vomiting, weight loss and recurring infections [3]. At the time of presentation, 73% of the patients have anemia, almost 80% have osteolytic bone disease and 19% have acute kidney injury [6]. The imbalance between the osteoblasts' and CTD-cyclophosphamide, thalidomide, dexamethsone; ISS-Multiple Myeloma International Staging System; WHO-World Health Organization; V(C)D-bortezomib, (cyclophosphamide), dexamethasone; VTDbortezomib, thalidomide, dexamethasone.

DNA Isolation
The DNA isolation from 200 µL of whole peripheral blood was performed using the column method with a dedicated kit, according to the manufacturer's recommendations (DNA Blood Mini Kit, Qiagen, Toronto, ON, Canada). Subsequently, the spectrophotometric evaluation of the concentration and quality of the obtained DNA was carried out using a NanoDrop Lite Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

Single Nucleotide Polymorphism Analysis
A StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) was used for the SNP genotyping. The real-time PCR was performed according to the manufacturer-established protocol with the use of the Genotyping Master Mix and TaqMan probes specific for the KIAA1524 SNP (rs2278911; 686C > T) (C__11237075_30, Thermo Fisher Scientific, USA). A selected probe was functionally tested by the manufacturer. The qPCR was performed in 10 µL reaction volume on 96-well plates. A rection mix containing 5 µL of Taqman Genotyping Master Mix, 0.5 µL of Taqman SNP assay (20×) and 4.5 µL of 0.2 ng/µL diluted DNA template was added to each well. The thermal cycling protocol was as follows: enzyme activation at 95 • C, 40 cycles of denaturation at 95 • C for 15 s and annealing at 60 • C for 1 min. All sample tests were performed in triplicate. After the amplification, the obtained genotypes were analyzed with StepOne Software v2.3 (Applied Biosystems, Waltham, MA, USA). Additionally, 10% of the samples were randomly selected and re-analyzed in a sequencing device (3500 Genetic Analyzer, Life Technologies). A 100% consistency of the results was obtained.

Plasma Cells Isolation
The bone marrow mononuclear cells (BMMC) were isolated from 3 mL of bone marrow (diluted 10×) using the density gradient centrifugation method (Biocoll, AG Biochrom, Berlin Germany), then subjected to washing and a viability assessment followed by the cell separation. The cell separation was performed after the incubation with CD138 antibodies and conducted according to the manufacturer's instructions with the use of the magneticactivated cell sorting technique (MACS ® Cell Separation, Bergisch Gladback, Germany). Subsequently, the assessment of viability and quantity of plasma cells was performed.

Gene Expression in the Bone Marrow Plasma Cells
Total RNA was isolated from the plasma cells (the maximum of 10 7 cells were used) with the use of an RNeasy Mini Kit (Qiagen, Toronto, ON, Canada). The expression of the KIAA and MYC genes was measured with the use of a StepOnePlus device and a dedicated High-Capacity cDNA Reverse Transcription Kit, TaqMan™ Fast Advanced Master Mix and TaqMan probes (Thermo Fisher Scientific, USA). For the KIAA1524 (assay ID: Hs00405413) and MYC (assay ID: hs00153408) genes, the ACTB (assay ID: hs99999903) gene was used as the reference gene. The relative expression of the investigated molecular targets was calculated using the 2 −∆Ct and 2 −∆∆Ct formulas.

Cytogenetic Assessment
The bone marrow samples were subjected to cytogenetic analysis. Abnormalities, such as the TP53 gene deletion, IGH gene rearrangements and CKS1B gene amplification, were detected by simultaneous staining of cytoplasmic immunoglobulin with the fluorescence in situ hybridization (cIg-FISH) technique according to the recommendations of Ross et al., with some modifications [27]. The following probes, all from Abbott Molecular (Abbott Park, IL, USA), were used: Vysis TP53/CEP 17 FISH Probe Kit for detection of del(17p13.1), Vysis IGH/FGFR3 Dual Colour, Dual Fusion Translocation Probe for detection of t(4;14)(p16;q32), Vysis IGH/MYC/CEP 8 Tri-colur, Dual Fusion Translocation Probe for detection of t(8;14)(q24;q32), Vysis IGH/CCND1 Dual Colour, Dual Fusion Translocation Probe for detection of t(11;14)(q13;q32), Vysis IGH/MAF Dual Colour, Dual Fusion Translocation Probe for detection of t(14;16)(q32;q23) and Vysis 1q21 CKS1B SpectrumOrange/1p32 CDKN2C SpectrumGreen FISH Probe Kit for the detection of amp(1q32). Fluorescent microscopic analysis was performed by scoring 100 AMCA-positive (anti-mitochondrial antibody-positive) plasma cells to determine the frequency of each aberration. The cut-off values were set at 20% for deletion/amplification probes and 10% for dual fusion probes, according to the recommendations of the European Myeloma Network [15,16].

Statistical Analysis
MedCalc software (v.15.8) was used for statistical analysis of the acquired data. A chi-square test was performed to assess the KIAA1524 genotypes' distribution depending on the selected demographic, as well as clinical and molecular variables. Due to the abnormal distribution of the studied continuous data in the D'Agostino-Pearson test, either the nonparametric Mann-Whitney U Test (for comparing two independent groups) or the ANOVA Kruskal-Wallis test (for comparing more than two independent groups) was performed in order to compare the KIAA1524 s expression levels depending on the distribution of SNP located in this gene. The Spearman's rank correlation coefficient test was applied to assess the correlation between the expression of the KIAA1524 gene and the MYC gene.
The influence of selected demographic, clinical and molecular variables on the risk of no response to chemotherapy was estimated by the odds ratio (OR) and corresponding 95% confidence intervals (CI). In terms of the survival evaluation, the log-rank test was applied for the univariable analysis, whereas Cox regression models were used in the multivariable analysis. The influence of selected demographic, clinical and molecular variables on the risk of death was described by the hazard ratio (HR) and corresponding 95% CI. The Kaplan-Meier estimation method was utilized to generate the survival curves, whereas forest plot graphs were used to illustrate the results of the multivariable analysis. In cases in which both the composite (e.g., disease stage according to ISS) and its constituent variables (e.g., albumin or B2M levels) were statistically significant according to the univariable analysis, only the first one of them was included in the multivariate models. The backward elimination method was used for the selection of variables to be included in the results adjustment. Finally, in the multivariable PFS analysis, all results were adjusted by the following variables: gender, AHSCT, PLT, del17/TP53, t(11;14) IGH/CCND1 and KIAA gene genotype. On the other hand, in the multivariable OS analysis, all results were adjusted by diagnosis, performance status, anemia before treatment, t(4;14) IGH/FGFR3 and KIAA gene genotype. The progression-free survival (PFS) was defined as the period between the start of the treatment and the occurrence of disease progression (complete data), or of the last documented follow-up (censored data). The overall survival (OS) was calculated from the start of treatment until the patient's death (complete data) or the last documented follow-up (censored).
The data from the MMRF CoMMpass project were retrieved from "https://portal. gdc.cancer.gov/ (accessed on 21 March 2023)" through the R language wrapper library GenomicDataCommons (installed by BiocManager). Based on the obtained data, Pearson correlations between genes of interest were computed, and further selected results were illustrated with the use of the corrplot library (with heat maps presenting correlation matrix).
In all analyses, results with a p-value < 0.05 was considered statistically significant. Additionally, p-values in the range of 0.05-0.06 were considered to show a trend toward statistical significance.

The KIAA1524 Genotypes' Distribution in Relation to the Demographic, Clinical and Molecular Factors
The only significant difference observed between the patients with various KIAA1524 genotypes was the prevalence of the t(4;14) IGH/FGFR3, whose absence was more common in patients with the CC genotype than in the carriers of the T allele (80% vs. 20%, p = 0.0095). The detailed data is presented in the Supplementary Table S1.

Response to Treatment
A significantly higher risk of non-response after the 2nd cycle of chemotherapy was observed in patients with poor performance status (PS: 2-4: OR = 3.95), those who were treated with a CTD regimen (OR = 5.35), those with elevated CRP (OR = 3.32) and those with the TT genotype of the KIAA1524 gene (OR = 8.33). On the other hand, the risk of non-response after the 2nd cycle of chemotherapy was significantly lower in patients treated with the VTD regimen (OR = 0.05) and those with the CC genotype of the KIAA1524 gene (OR = 0.23). In turn, a significantly higher risk of non-response after the 4th cycle of chemotherapy was observed in patients with elevated CRP (OR = 6.70), those with the presence of t(11.14)/CCND1 translocation (a trend toward significance; OR = 6.42) and those who were carriers of the TT genotype of the KIAA1524 gene (OR = 20.20). On the other hand, a significantly higher risk of non-response after the 6th cycle of therapy was observed in patients with thrombocytopenia (OR = 9.22), those with the presence of 17p/TP53 deletion (OR = 12.46) and those with the presence of t(11.14)/CCND1 translocation (OR = 55.20). Moreover, a significantly higher risk of non-response after the 8th cycle of therapy was observed in patients with the presence of t(11.14)/CCND1 translocation (OR = 16.00). Detailed data on the influence of the selected demographic, clinical and molecular variables on chemotherapy response are presented in Supplementary Table S2.

Univariable Analysis
According to the univariable analysis, the factors significantly correlated with an increased risk of PFS reduction were: male sex (HR = 1.65), stage 3 disease according to the ISS classification (HR = 1.74), higher stage of chronic kidney disease (>G2: HR = 1.85), poor performance status (PS: 2-4; HR = 1.60), anemia before treatment (HR = 2.40) low platelet count (HR = 2.08), low albumin level (HR = 2.29), high LDH level (HR = 1.95), high β2-microglobulin level (B2M) (HR = 2.58), high level of creatinine (HR = 1.64), the presence of del17p/TP53 (HR = 1.84) and the presence of t(11;14) IGH/CCND1 (HR = 2.49). Additionally, a higher risk of PFS reduction was observed in patients with the TT variant of the KIAA1524 gene (rs2278911) as compared to patients with other variants (median PFS: 6 vs. 25 months; HR = 4.67) ( Figure 1A). A decreased risk of PFS reduction was observed in patients treated with VTD as compared to those treated with CTD or V(C)D schemes (HR = 0.56), those in whom AHSCT was applied (HR = 0.37) and those with the CC genotype (compared to carriers of the T allele, HR = 0.40).

Univariable Analysis
The univariable analysis identified the following factors as significantly related to higher risk of OS death: light chain disease (HR = 2.37), stage 3 disease according to the ISS classification (stage 3; HR = 2.03), higher stage of chronic kidney disease (>G2: HR = 1.78), poor performance status (PS: 2-4; HR = 2.01), the presence of anemia before treatment (HR = 2.43), low albumin levels (HR = 2.28), high B2M levels (HR = 2.73), high creatinine levels (HR = 1.74), the presence of del17p/TP53 (HR = 1.84) and the presence of t(11;14) IGH/CCND1 (HR = 2.62). Moreover, we observed a higher risk of death in patients with the TT variant of the KIAA1524 gene (rs2278911) as compared to those with other genotypes (median OS: 8 months vs. 45 months; HR = 4.39) ( Figure 1B). A significantly lower risk of OS reduction was observed in the subjects in whom AHSCT was applied (HR = 0.53) and those with the CC variant of the KIAA1524 gene (rs2278911) as compared to carriers of the T allele (HR = 0.41).
A graphical representation of the results of the multivariable analysis of PFS and OS is shown in Figure 2. Detailed data on the influence of the selected demographic, clinical and molecular variables on PFS and OS are presented in Supplementary Table S3. classification (HR = 2.11), performance status (a trend toward significance; HR = 2.19), low albumin levels (HR = 4.08), the presence of t (4;14), IGH/FGFR3 (HR = 3.22) and the presence of the TT genotype of the KIAA1524 gene (rs2278911) (HR = 6.89) [ Figure 2B].
A graphical representation of the results of the multivariable analysis of PFS and OS is shown in Figure 2. Detailed data on the influence of the selected demographic, clinical and molecular variables on PFS and OS are presented in Supplementary Table S3.

Relationship between the Expression of the KIAA1524 Gene, Its SNP (rs2278811) and the Expression of the MYC Gene
We noted that the expression of the KIAA1524 gene was significantly lower in the carriers of the CC genotype of this gene compared to the carriers of the CT and TT genotypes (in both cases p < 0.05) ( Figure 3A). Moreover, we found an intermediate, positive correlation between expression of the KIAA1524 gene and the MYC gene (rho = 0.515; p = 0.0345) ( Figure 3B).

Relationship between the Expression of the KIAA1524 Gene, Its SNP (rs2278811) and the Expression of the MYC Gene
We noted that the expression of the KIAA1524 gene was significantly lower in the carriers of the CC genotype of this gene compared to the carriers of the CT and TT genotypes (in both cases p < 0.05) ( Figure 3A). Moreover, we found an intermediate, positive correlation between expression of the KIAA1524 gene and the MYC gene (rho = 0.515; p = 0.0345) ( Figure 3B). and overall survival (B).

Relationship between the Expression of the KIAA1524 Gene, Its SNP (rs2278811) and the Expression of the MYC Gene
We noted that the expression of the KIAA1524 gene was significantly lower in the carriers of the CC genotype of this gene compared to the carriers of the CT and TT genotypes (in both cases p < 0.05) ( Figure 3A). Moreover, we found an intermediate, positive correlation between expression of the KIAA1524 gene and the MYC gene (rho = 0.515; p = 0.0345) ( Figure 3B

Relationship between the Expression of the KIAA1524 Gene and the Expression of Other Genes Based on Data from the CoMMpass Project
Based on data from the CoMMpass project, we assessed the correlation between the expression of the KIAA1524 gene (known in the above database as CIP2A) and other genes known to be involved in the biology and pathomechanism of MM (Figure 4). A statistically significant and positive correlation between the expression of the CIP2A gene and the TP53, IL6ST, MCL1, BCL-2, AKT1 and E2F1 genes was found. In addition, a positive correlation between the expression of CIP2A and the MYC gene, which was demonstrated in our study, was further validated in the data retrieved from this project. Moreover, we found that the studied gene was strongly correlated with genes involved, among others, in the regulation of the cell cycle and differentiation, DNA replication, DNA damage detection and repair, mRNA processing and transport, spindle dynamics, chromosome processing and segregation ( Figure 5).

Relationship between the Expression of the KIAA1524 Gene and the Expression of Other Genes Based on Data from the CoMMpass Project
Based on data from the CoMMpass project, we assessed the correlation between the expression of the KIAA1524 gene (known in the above database as CIP2A) and other genes known to be involved in the biology and pathomechanism of MM (Figure 4). A statistically significant and positive correlation between the expression of the CIP2A gene and the TP53, IL6ST, MCL1, BCL-2, AKT1 and E2F1 genes was found. In addition, a positive correlation between the expression of CIP2A and the MYC gene, which was demonstrated in our study, was further validated in the data retrieved from this project. Moreover, we found that the studied gene was strongly correlated with genes involved, among others, in the regulation of the cell cycle and differentiation, DNA replication, DNA damage detection and repair, mRNA processing and transport, spindle dynamics, chromosome processing and segregation ( Figure 5).

Discussion
MYC affects a number of metabolic pathways and plays a crucial key role in the development of MM. It plays a role in cycle regulation, proliferation, differentiation and metabolism of the cells, and its dysregulation has been shown to promote the transition from monoclonal gammopathy of undetermined significance (MGUS) to MM. Its activity can be regulated by its mutations, as well as by changes in pathways involved in its regulation (e.g., it may be controlled by mTORC1 and RAS on a post-transcriptional level) [28][29][30].
The two most important phosphorylation sites on MYC, responsible for its stabilization and degradation are Serine62 (S62), which stabilizes MYC and participates in the target gene selection when phosphorylated, and Threonine 58 (T58), which runs the MYC degradation when phosphorylated [30,31]. S62 is phosphorylated by the extracellular-regulated kinases (ERKs) 1 and 2, while the glycogen synthase kinase (GSK-3ꞵ) is responsible for the T58′s phosphorylation, which can only be carried out after the S62′s phosphorylation [30]. Before MYC can be delivered to and degraded by the ubiquitin-proteasome system, it must be dephosphorylated in position S62 by PP2A [32,33]. CIP2A inhibits the PP2A, and, through that, stabilizes MYC [30,32]. Moreover, PP2A's inhibition increases ERK's activation and promotes cell proliferation and survival, as well as the expression of invasive proteases [30].
Since CIP2A also influences the RAS pathway, it may indirectly regulate the MYC pathway through its elements [34][35][36] The Ras proto-oncogene has been shown to enhance the ERK's activity, restrict the GSK-3ꞵ's activity and affect the PI3K pathway-mediated T58′s phosphorylation, making it a MYC stabilizer [30]. In addition, the mutations in MYC ubiquitin ligase (FBW7) may prevent the MYC's degradation, and the ubiquitin-specific protease USP28 has been shown to contribute to the MYC stabilization in neoplastic cells.

Discussion
MYC affects a number of metabolic pathways and plays a crucial key role in the development of MM. It plays a role in cycle regulation, proliferation, differentiation and metabolism of the cells, and its dysregulation has been shown to promote the transition from monoclonal gammopathy of undetermined significance (MGUS) to MM. Its activity can be regulated by its mutations, as well as by changes in pathways involved in its regulation (e.g., it may be controlled by mTORC1 and RAS on a post-transcriptional level) [28][29][30].
The two most important phosphorylation sites on MYC, responsible for its stabilization and degradation are Serine62 (S62), which stabilizes MYC and participates in the target gene selection when phosphorylated, and Threonine 58 (T58), which runs the MYC degradation when phosphorylated [30,31]. S62 is phosphorylated by the extracellular-regulated kinases (ERKs) 1 and 2, while the glycogen synthase kinase (GSK-3β) is responsible for the T58 s phosphorylation, which can only be carried out after the S62 s phosphorylation [30]. Before MYC can be delivered to and degraded by the ubiquitin-proteasome system, it must be dephosphorylated in position S62 by PP2A [32,33]. CIP2A inhibits the PP2A, and, through that, stabilizes MYC [30,32]. Moreover, PP2A's inhibition increases ERK's activation and promotes cell proliferation and survival, as well as the expression of invasive proteases [30].
Since CIP2A also influences the RAS pathway, it may indirectly regulate the MYC pathway through its elements [34][35][36] The Ras proto-oncogene has been shown to enhance the ERK's activity, restrict the GSK-3β's activity and affect the PI3K pathway-mediated T58 s phosphorylation, making it a MYC stabilizer [30]. In addition, the mutations in MYC ubiquitin ligase (FBW7) may prevent the MYC's degradation, and the ubiquitin-specific protease USP28 has been shown to contribute to the MYC stabilization in neoplastic cells. RalA is a Ras effector, and its phosphorylation is also an important element of PP2Amediated tumor suppression [30]. Altogether, MYC's activity is regulated not only by CIP2A, but also by a number of other factors. The complexity of these regulatory systems would justify the intermediate, positive correlation we observed between the expression of the KIAA1524 gene and the MYC gene (rho = 0.515; p = 0.0345) ( Figure 2B).
Nevertheless, CIP2A is not the only factor capable of changing the PP2A's activity. GFI-1 has been shown (growth factor independence-1) to inhibit the SGPP1 (sphingosine-1-phosphate phosphatase 1) gene transcription, leading to increased S1P (sphingosine-1phosphate) synthesis. This resulted in a S1P/Ceramides ratio imbalance, triggering the PP2A inactivation and, as a consequence, c-Myc stabilization in the MM cells [33]. Such interactions must be considered when judging the prognostic roles of KIAA1524 and CIP2A.
In the case of the 686 SNP of the KIAA1524 (rs2278911) gene, when cystein becomes replaced by tymin, the 229th amino acid coded by the gene changes from arginine (Arg) to glutamine (Gln) [37]. Such an alteration has been shown to lower the charge and change the surface properties of the CIP2A's N-terminal domain, affecting the protein-to-protein interactions [34]. In addition, Preuss et al. found that certain modifications within the B subunit affected PP2A's activity towards its targets [38]. A research paper by Wang et al. [35] revealed that, with its N-terminal (amino acids 1-560), CIP2A could bind directly to the B56α and B56γ subunits of the PP2A protein-both of which are tumor suppressor units-and that the region necessary for this was located between the amino acids 159 and 245 and positively charged. Moreover, the researchers found that changing the 230th (asparagine-neutrally charged) nucleotide to a negatively charged glutamic acid led to decreased binding of CIP2A to the B56α, and concluded that the B56 subunit was most likely to bind to a positively charged surface of the CIP2A [35].
In our study, the presence of the TT genotype of the KIAA1524 gene (rs2278911) had an independent prognostic value, with the afflicted patients being over seven times more likely to experience PFS and OS reduction compared to the carriers of other variants. Since the 686C > T alteration resulted in an exchange of a positively charged amino acid for a neutral one, it seems that in order to explain the TT genotype's correlation with worse prognosis, we must search beyond the currently known direct CIP2A-PP2A interactions. In a paper published in 2014, three years before Wang's publication, Puustinen et al. wrote that the mechanism in which CIP2A modulated the mTORC1-related PP2A activity was most likely that of an allosteric inhibitor, and stipulated that it was more likely to affect the substrate availability rather than the enzyme's activity [39]. In 2021, Laine et al. discovered that poor clinical outcomes in basal-like breast cancer (BLBC) patients was related to high CIP2A expression and not to the alterations in its gene's sequence. They also established that CIP2A expression was upregulated by the DNA-PF/CHK1 activity, TP53 inactivation and EGFR pathway activation [40]. As we have also observed a significantly lower expression of CIP2A among patients with the CC genotype, we would stipulate that that the consequences of indirect CIP2A-mediated PP2A modulation may outweigh the effects of the direct PP2A-binding, and that the most likely cause for the reported outcome-SNP correlation is, in fact, altered CIP2A expression. Our results would also align with a finding by Li et al., who speculated that Arg229Gln altered the CIP2A expression [37].
A number of drugs [41][42][43][44][45] have shown the ability to influence KIAA1524, CIP2A or their affected pathways. Bedewy and Elmaghraby observed that MM patients had higher CIP2A expression, which decreased during bortezomib treatment. Furthermore, higher CIP2A levels were correlated with a worse response to bortezomib, as well as with shorter PFS [45]. The research on KIAA1524 and CIP2A as potential treatment targets is ongoing, but in order for it to progress, many obstacles must be overcome.
One of the biggest issues we are currently facing is a very convoluted net of CIP2A-PP2A interactions, which may be affected by a number of factors and has yet to be fully explored [34,[46][47][48], as well as the partially contradicting data on the matter. A high CIP2A concentration has been found to serve as a negative prognostic factor in a number of human cancers [35], and CIP2A knockdown has been shown to inhibit the activity of MYC, E2F1 and Akt [30]; restrict tumor growth in oral cancer mice models [47] and result in decreased proliferation and increased apoptosis of human MM cells [48]. However, such results could not be achieved in all tumor types [46]. Additionally, although there seems to be a general consensus that CIP2A-mediated PP2A inhibition supports MM development, we did come across some research which pointed towards an oncogenic role of PP2A. In 2001 Kang and Chou found that PP2A regulated the ROI (reactive oxygen intermediates) levels within MM cells, which promoted the expression of IL-6 and BCL-2, thus supporting the MM's development [49]. Three years later, Mitsuthashi et al. demonstrated that PP2A prevented the degradation of the glycoprotein 130 (also known as IL-6 signal transducer). The researchers observed that bortezomib-mediated gp130 degradation provoked the MM cells' apoptosis by decreasing their sensitivity to the IL-6 [50]. In 2021, Slomp et al. found that PP2A reversed the dephosphorylated and stabilized MCL1 (induced myeloid leukemia cell differentiation protein) in MM [45].
In the context of multiple myeloma, the expression of the HNRNPA2B1, USP1, RRM1, SPAG5, PCNA and TOP2A genes has been studied. It has been proven that HNRNPA2B1 increases the stability of mRNA transcripts for interleukin enhancer-binding factor 3 (ILF3), and its overexpression induces cell proliferation through the AKT3 kinase. Increased HN-RNPA2B1 gene expression is associated with an unfavorable prognosis [57]. The USP1 protein leads to regulated DNA repair through deubiquitination and homologous recombination pathways, and inhibits cell differentiation by stabilizing tumor-promoting inhibitors of DNA-binding proteins. Inhibition of USP1 has been shown to reduce myeloma cell viability, inhibit the growth of multiple myeloma cells and overcome bortezomib resistance [58]. Higher expression of the RRM1 gene is associated with shorter overall survival. A blockade of the RRM1 gene results in the upregulation of DNA damage response genes and p53-regulated genes, thereby inhibiting the growth and promoting apoptosis of myeloma cells and cells of the bone marrow microenvironment [59]. SPAG5 gene expression correlates with myeloma cell malignancy, and SPAG5 overexpression is associated with poorer clinical outcomes. In vitro, knockdown of the SPAG5 gene leads to anti-MM effects, cell growth arrest and apoptosis via the PI3K/AKT/mTOR pathway [67]. The PCNA gene encodes a multifunctional protein which is essential for DNA replication and repair and is overexpressed in myeloma cells. PCNA inhibition induces apoptosis and sensitizes cells to melphalan, a common anti-myeloma drug [75]. TOP2A has been demonstrated to be another potential target for anti-myeloma therapy. It has been observed that overexpression of TOP2A topoisomerase in myeloma cells is associated with resistance to proteasome inhibitors [77].

Conclusions
The varied clinical course of MM, differences in the efficacy of specific therapeutic strategies, and different lengths of time needed for chemoresistance development imply a need to identify risk stratification factors that would enable the personalization of therapy and the improvement of treatment outcomes. For this purpose, we can use prognostic markers, which assess the risk of an unfavorable course of a disease and its recurrence, as well as predictive factors, which determine the probability of obtaining remission with a specific treatment method. Although our results are preliminary, they indicate that the assessment of the SNP 686C > T of the KIAA1524 gene may be a useful biomarker for determining the prognosis of MM patients.