PDZ Binding Kinase/T-LAK Cell-Derived Protein Kinase Plays an Oncogenic Role and Promotes Immune Escape in Human Tumors

Background PDZ binding kinase (PBK)/T-LAK cell-derived protein kinase (TOPK) is an important mitotic kinase that promotes tumor progression in some cancers. However, the pan-cancer analysis of PBK/TOPK and its role in tumor immunity are limited. Methods The oncogenic and immune roles of PBK in various cancers were explored using multiple databases, including Oncomine, Human Protein Atlas, ULCAN, Tumor Immune Estimation Resource 2.0, STRING, and Gene Expression Profiling Interactive Analysis 2, and data collected from The Cancer Genome Atlas and Genotype-Tissue Expression Project. Several bioinformatics tools and methods were used for quantitative analyses and panoramic descriptions, such as the DESeq2 and Tumor Immune Dysfunction and Exclusion (TIDE) algorithm. Results PBK was expressed at higher levels in most solid tumors than in normal tissues in multiple databases. PBK was associated with an advanced tumor stage and grade and a poor prognosis in most cases. PBK was associated with tumor immune cell infiltration in most cases and was especially positively correlated with TAMs, Tregs, MDSCs, and T cell exhaustion in KIRC, LGG, and LIHC. PBK was closely related to TMB, MSI, and immune checkpoint genes in various cancers, and patients with higher expression of PBK in KIRC, LGG, and LIHC had higher TIDE scores and lower immune responses in the predicted results. PBK was closely related to cell cycle regulation and immune-related processes in LIHC and LGG according to GO and KEGG enrichment analyses. Conclusions PBK may play an oncogenic role in most solid tumors and promotes immune escape, especially in KIRC, LGG, and LIHC. This study suggests the potential value of PBK inhibitors combined with immunotherapy.


Introduction
PDZ binding kinase (PBK) is a kinase that binds to the PDZ2 region of HDIG as first identified by an investigator screening a yeast two-hybrid system [1], and the gene was cloned from a HeLa cell cDNA library. T-LAK cell-derived of its expression, PBK/TOPK is rarely expressed in normal tissues, as its expression is detected only in testicular and embryonic tissues but is expressed at high levels in some malignancies; its expression level is negatively correlated with patient prognosis [3][4][5]. Second, regarding the molecular mechanism, PBK/TOPK, a mitotic kinase is involved in regulating multiple important signaling pathways such as p53 [6], MAPK [3], and PI3K/AKT [7], playing an important role in the process of tumorigenesis and tumor progression.
ird, regarding the molecular properties, PBK/TOPK is a serine-threonine protein kinase, and several PBK/TOPK specific inhibitors have been screened, such as HI-TOPK-032, OTS514, and OTS964. In animal models, these inhibitors effectively inhibit the development of multiple malignant tumors with fewer toxic side effects [8][9][10], showing good clinical prospects.
Furthermore, several recent studies have suggested that PBK/TOPK may be associated with tumor-infiltrating immune cells and have a potential target for cancer immunotherapy [5,11,12]. is study is the first to conduct a pancancer analysis of PBK/TOPK using Oncomine, Human Protein Atlas(HPA), ULCAN, the Tumor Immune Estimation Resource 2.0 (TIMER2.0), STRING, Gene Expression Profiling Interactive Analysis 2 (GEPIA2), and the data collected from e Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression Project (GTEx). e results of this study showed the potential correlation and mechanism of PBK/TOPK and tumor-immune interactions.

Analysis of PBK/TOPK Expression. Differences in PBK/
TOPK mRNA expression across cancers in the Oncomine database (https://www.oncomine.org/) were analyzed. e thresholds were set as a fold change of 1.5 and P value of 0.001.
RNAseq data from TCGA and GTEx were downloaded from UCSC XENA (https://xenabrowser.net/datapages/) [13]. All RNAseq data were processed using the Toil process in TPM format. Notably, 33 cancers and 31 types of normal tissues (N � 15776) were included in the gene expression analysis. e difference in PBK/TOPK mRNA expression between tumors and paired adjacent noncancerous tissues from TCGA data was analyzed. Differences in PBK/TOPK mRNA expression between tumors and unpaired normal tissues from TCGA and GTEx data were also analyzed. e RNAseq data in TPM (transcripts per million reads) format were log2 transformed. If not stated otherwise, the analysis of "TCGA" in this study was based on the data downloaded from UCSC Xena.

Survival Analysis Based on PBK/TOPK Expression.
e patients were stratified based on the expression of the PBK/TOPK mRNA in tumors and adjacent noncancerous tissues using a bipartite method and were divided into two groups: PBK/TOPK high expression and PBK/TOPK low expression groups. e correlation of PBK/TOPK expression with overall survival (OS) was analyzed with the Kaplan-Meier survival analysis.

Analysis of the Correlation of PBK/TOPK Expression with Immune Cell Infiltration.
e immune cell infiltration score data were downloaded from the Timer2.0 database (http:// timer.comp-genomics.org/) [16]. TIMER2.0 is a comprehensive resource for the systematical analysis of immune cell infiltrates and tumor immunological, clinical, and genomic features across diverse cancer types. "PBK/TOPK" was input in the "Gene" of the "Immune Association" module. e association of PBK/TOPK with selected immune cell types was analyzed. Timer2.0 provides six immune estimation algorithms, among which TIMER algorithms is the only one that considers tissue specificity. e associated data was downloaded from the " Table" module of TIMER2.0, and we present the TIMER algorithms-associated data of "T-cell CD8+," "T-cell CD4+," "B cell," "Neutrophil," "Macrophage," and "Myeloid dendritic cell" across cancers as a heatmap with R software v4.0.3 and the "ComplexHeatmap" R package [17]. To further analyze the relationship of PBK with tumor-killing immune cells (CD8+ T cell and NK cell) as well as immune cells that facilitate tumor immune escape (M2 macrophage, Tregs, cancer-associated fibroblast (CAFs), and Myeloid-derived suppressor cells (MDSCs)) in KIRC, LGG, and LIHC, we displayed the images of the relevant data derived from the analysis based on the QUANTISEQ, MCP-COUNTER, and Tumor Immune Dysfunction and Exclusion (TIDE) algorithms in the TIMER2.0. e partial correlation (cor) values and P values were analyzed using Spearman's rank correlation test.  [19]. Spearman's rank correlation coefficients were calculated to analyze the correlations of PBK/TOPK expression with the TMB and the MSI of each tumor sample.

Analysis of the Correlation between PBK/TOPK Expression and the Immune Response.
e association of PBK/ TOPK expression with immune checkpoint gene expression in pan-cancers from TCGA was analyzed.

Analysis of the Correlation between PBK/TOPK Expression and DNA Mismatch Repair Genes and
Methyltransferases.

Gene Enrichment
Analysis. STRING (https://www. string-db.org/) [23] and GEPIA2 (http://gepia2.cancerpku.cn/) were used for the PBK/TOPK-associated enrichment analysis. "PBK/TOPK" was input into the "SEARCH" module, and "Homo sapiens" was subsequently selected. We next set the following parameters: network type ("full network"), meaning of network edges ("evidence"), active interaction source ("experiments"), minimum required interaction score ("medium confidence 0.150"), and max number of interactors to show ("no more than 10 interactors" in the 1st shell). Finally, the proteins binging to PBK/ TOPK based on available experimental data were obtained. e "Similar Gene Detection" module of GEPIA2 was used to obtain the top 100 PBK/TOPK-associated genes in TCGA tumors and normal tissues. en, GEPIA2 "correlation analysis" module was used to analyze the Pearson correlation coefficients of PBK/TOPK with the top 100 selected genes. e results were visualized by constructing a dot plot. en, we used TIMER2.0 to verify the correlation of PBK/TOPK expression with the five highest PBK/TOPK-correlated genes in GEPIA2. BIO-INFORMATICS ＆ EVOLUTIONARY GENOMICS, a web viewer that calculates and draws Venn diagrams (http:// bioinformatics.psb.ugent.be/webtools/Venn/), was used to prepare the Venn diagrams of the PBK/TOPK-binding and related genes.
ese selected genes and the differentially expressed genes between the PBK/TOPK high expression and low expression groups of KIRC, LGG, and LIHC in TCGA were subjected to KEGG and GO pathway enrichment analysis.

Statistical
Analysis. All statistical analyses of the RNAseq data of TCGA were performed using R software v4.0.3. e Wilcoxon rank-sum test was used to analyze the expression differences between tumor and normal tissues. ANOVA was used for PBK/TOPK mRNA expression comparison between the different pathological stages. e Shapiro-Wilk normality test was used to determine the normal distribution of the data, and Levene's test was used for the homogeneity of variance analysis. One-way ANOVA was used if variances were homogeneous, and Welch's oneway ANOVA was used if variances were not homogeneous. e student's t-test was used for expression comparison of PBK/TOPK between different tumor grade groups. Independent samples t-tests were used if the data had a normal distribution, and Welch's t-tests were used if the data were not normally distributed. e correlation of PBK/TOPK expression and OS was analyzed with the Kaplan-Meier survival analysis.
e "survival" R package was used for statistical analysis. Spearman's rank correlation coefficient was used to analyze the correlations of PBK/TOPK expression and TMB, MSI, immune checkpoint gene expression, DNA mismatch repair gene expression, and methyltransferase gene expression. e "clusterProfiler" was used for GO and KEGG enrichment analysis. And the "org.Hs.eg.db" R package (V.3.10.0) was used for ID transformation. We limit species to "Homo sapiens." e thresholds were set as a fold change of 1.0 and P value of 0.05 in the TCGA datasets. e " ggplot2" and "ggpubr" R packages were used to generate boxplots, violin plots, lollipop charts, and Bubble charts. e "forestplot" R package and "survminer" R package were used to generate forest plots and survival curves. e "ComplexHeatmap" was used to generate heatmaps. A P value <0.05 was considered statistically significant.

PBK/TOPK Is Overexpressed in Various Cancers.
e expression of PBK/TOPK mRNA and protein were explored. e PBK/TOPK mRNA was expressed at higher levels in tumors than in normal tissues in most tumors' datasets from Oncomine, such as bladder cancer, brain and Journal of Oncology 3 CNS cancer, cervical cancer, colorectal cancer, esophageal cancer, gastric cancer, head and neck cancer, liver cancer, lung cancer, lymphoma, ovarian cancer, pancreatic cancer, prostate cancer, and sarcoma ( Figure 1(a)). Meanwhile, lower expression of the PBK/TOPK mRNA was observed in the leukemia datasets. Notably, higher PBK/TOPK mRNA expression was found in 9 datasets, and lower PBK/TOPK mRNA expression was found in 1 dataset of breast cancer. Next, we analyzed PBK/TOPK mRNA expression in various cancers from TCGA. First, we performed paired comparisons of differences in PBK/TOPK mRNA expression levels between cancer and normal tissues from individuals with the same cancer. Only 18 tumors were eligible, and PBK/TOPK mRNA expression showed no significant differences only in PAAD and READ. Higher PBK/TOPK mRNA expression levels were detected in other tumors, such as BLCA, BRCA, CHOL, COAD, ESCA, HNSC, KICH, KIRC, KIRP, LIHC, LUAD, LUSC, PRAD, READ, STAD, THCA, and UCEC, than in tumor tissues compared to normal tissues ( Figure 1(b); for additional details, see Table S1). en, the normal tissues from GTEx were included. We reconfirmed the similar expression in the 18 cancers. Higher expression of the PBK/TOPK mRNA was detected in the following 12 cancers: ACC, CESC, DLBC, GBM, HNSC, LGG, OV, PAAD, PCPG, READ, SKCM, and THYM, while no significance was found in SARC. Because of the lack of matched normal tissues, MESO was not conducted statistical difference analysis (Figure 1(c); for additional details, see Table S1). Similar to the Oncomine results, lower expression of PBK/TOPK was observed in LAML tissues than that in normal tissues. Combining the aforementioned data, we found that the PBK/TOPK mRNA appeared to be expressed at high levels in 30 cancers, except LAML, SARC, and MESO. e expression of the PBK/TOPK protein was obtained from the Human Protein Atlas and CPTAC datasets. Moderate nuclear and cytoplasmic positivity was observed in varying fractions of cells in several cases of ovarian, colorectal, melanoma, breast, stomach, testis, and cervical cancers. e remaining cancer tissues including urothelial carcinoma, prostate cancer, skin basal cell carcinoma, pancreatic adenocarcinoma, Hodgkin lymphoma, glioma, oral squamous carcinoma, hepatocellular carcinoma, and lung squamous carcinoma showed weak staining (Figure 1(d)). Endometrial cancer, renal cancer, carcinoid cancer, and thyroid cancer generally exhibited negative staining. We confirmed the high expression of the PBK/ TOPK protein in COAD and OV in the CPTAC datasets. In addition, we also observed high expression of PBK/TOPK in UCEC and RCC tissues compared to the normal tissues in the CPTAC datasets (Figure 1(e)).

Analysis of PBK/TOPK mRNA Expression Based on the Pathological Stage and Tumor Grade.
We explored PBK/ TOPK mRNA expression in various tumors with different pathological stages in TCGA. We found that in the downloaded TCGA data, only 23 cancer datasets contained tumor staging information, while 12 cancer datasets contained histologic grade information. e expression of PBK/TOPK increased from pathologic stages I to IV in various cancers including ACC, KICH, KIRC, KIRP, and LUAD. e expression of PBK/TOPK increased in LIHC from stages I to II but was a lower expression in stage IV. Although elevated PBK expression in stages II and III compared to stage I was found in BRCA, no statistical difference was found in the comparison between the other staging groups. However, PBK/TOPK mRNA expression was negatively correlated with the pathological stage of COAD (Figure 2(a)). And no significance was found in 15 cancers including BLCA, CESC, CHOL, DLBC, ESCA, HNSC, LUSC, MESO, OV, READ, SKCM, TGCT, THCA, UCS, and UVM (Figures S1A-S10). In addition, we explored the correlation between the expression of the PBK/TOPK mRNA and the tumor grade. Higher expression of PBK/ TOPK mRNA was found in high-grade (G3&G4) tumors than in low-grade (G1&G2) tumors of BLCA, CESE, CHOL, HNSC, KIRC, LGG, LIHC, PAAD, UCEC, and UCS ( Figure 2(b)). And no significance was found in ESCA and OV (Figures S1P-S1Q).

PBK/TOPK Is Correlated with Tumor Immune Cell
Infiltration. Immune infiltration cells in the tumor microenvironment have been shown to play a key role in the occurrence and progression of cancer. Immune infiltration cells affect the effect of clinical treatments for cancer. A comprehensive analysis of immune infiltration cells will elucidate the mechanism of cancer immune escape, thus providing an opportunity to develop new therapeutic strategies [24]. Tumor purity was a major confounding factor in this analysis since most immune cell types were negatively correlated with tumor purity. As TIMER is the only algorithm that considers tissue specificity, we displayed the results of the correlation between PBK/TOPK and immune infiltration cells based on the TIMER algorithm of TIMER2.0. According to the data analyzed from the TIMER2.0, we redrew the heatmap (Figure 4(a)) to help readers more intuitively understand the correlation between PBK/TOPK and immune infiltration cells in pan-cancer datasets (Figure 4(a); for additional details, see Table S2). We

Total Unique Analyses
Cell color is determined by the best gene rank percentile for the analyses within the cell.
NOTE: An analysis may be counted in more than one cancer type.

Normal Tumor
(c) Figure 1: Continued.  observed a significant correlation between PBK/TOPK expression and TILs in most cancers. We focused on the 10 cancers in which PBK/TOPK affects prognosis. Finally, we focused on 3 tumors, KIRC, LGG, and LIHC, in which PBK/ TOPK expression has a similar correlation with TILs. As shown in Figure 4

Correlations of PBK mRNA Expression with the Infiltrating Immune Cells in KRIC,
LGG, and LIHC Based on TIMER2.0. We analyzed the correlations of PBK/TOPK expression with the infiltrating immune cells that exert tumor-killing effect (CD8+, NK cell) and promote tumor immune escape (M2 macrophage, Tregs, CAFs, MDSCs) based on TIMER2.0 to further explore the potential role of PBK in cancer immunity. As shown in Figure 5, similar to the results based on the TIMER algorithm, the relationships between PBK expression level and CD8+ T cell in 3 tumors Histologic grade      Tables S3 and S4). Immune checkpoints are inhibitory pathways in the immune system that are regulated by ligand and receptor interactions. ey are important for maintaining autoimmune tolerance and regulating the duration and amplitude of physiological immune responses to avoid the destruction and damage caused by immune cells to normal tissues. However, they may be exploited to participate in immune  escape from cancer. e current immunotherapy drugs used in the clinic are mainly antibodies against these targets. PBK/ TOPK expression showed different correlations with different immune checkpoint genes in various cancers (see Figure 6(c); for additional details, see Table S5). In KIRC, LGG, and LIHC, the expression of PBK/TOPK was positively correlated with the selected immune checkpoint genes. Furthermore, the potential ICI response was predicted with the TIDE algorithm. TIDE uses a set of gene expression markers to estimate 2 distinct mechanisms of tumor immune evasion, including dysfunction of tumor infiltration cytotoxic T lymphocytes (CTLs) and exclusion of CTLs by immunosuppressive factors. Patients with higher TIDE scores have a higher chance of antitumor immune escape, thus exhibiting a lower response rate to ICI treatment. Figure 6(d) shows the numbers of immune responses and the distribution of TIDE scores of samples in the PBK/TOPK high expression and low expression groups of patients with KIRC, LGG, and LIHC in the predicted results. In all the 3 cancers, the PBK/TOPK overexpression group had a lower immune response rate and higher TIDE score in the predicted results (for additional details, see Table S6).

Correlations of PBK/TOPK Expression with DNA Mismatch Repair Genes and Methyltransferase Expression in
Pan-Cancer. We further analyzed the correlation between PBK/TOPK expression and the expression of DNA mismatch repair genes and methyltransferase in pan-cancer. As shown in Figure 7(a), PBK/TOPK expression was positively correlated with the expression of DNA mismatch genes, including MLH1, MSH2, MSH6, and PMS2, in the majority of cancers except CHOL and UCS (see Figure 7(a); for additional details, see Table S7), suggesting that PBK/TOPK may promote cancer progression by upregulating DNA mismatch repair-related genes. In addition, DNA methylation may play an important role in tumor progression, and PBK/TOPK expression was positively correlated with methyltransferase expression, including DNMT1, DNMT3A, and DNMT3B, in most cancers except CHOL (see Figure 7(b); for additional details, see Table S7). is result suggested that PBK/TOPK potentially plays an important role in promoting tumorigenesis and progression by regulating gene methylation.
We further performed GO and KEGG pathway enrichment analyses of PBK/TOPK-related differentially genes in KIRC, LGG, and LIHC. Similar to the results of the pancancer analysis, PBK/TOPK-related differentially expressed genes were enriched in cell cycle regulation. In addition, PBK/TOPK was involved in immune regulation processes such as "Antigen Binding," "Immunoglobulin Receptor Binding," and "Immunoglobulin Complex" in both LGG and LIHC. Moreover, PBK/TOPK-related differential genes of LIHC were also enriched in immune processes such as "humoral immune response mediated by circulation" and "complement Activation classic pathway" in LIHC (see in Figures 8(g) and 8(i), for additional details, see Tables S13 and S12). In KRIC, the PBK/TOPK-related differentially genes were enriched not only in cell cycle-related pathways but also in metabolism-related pathways such as "Alcoholism," "Collecting Duct Acid secretion," "Bicarbonate Transmembrane Transporter Activity," and other processes such as "Serine-Type Endopeptidase Activity" which may be associated with the complement system (see in Figure 8(f ); for detailed data, see Table S11).

Discussion
Although previous studies [1,26,27] have reported higher expression of the PBK/TOPK mRNA and protein in tumor tissues than in adjacent noncancerous tissues from different tumors, this study was the first pan-cancer analysis of PBK/ TOPK and enabled us to observe the expression across all cancers. We only detected lower PBK/TOPK mRNA expression in LAML tissues than in normal tissues, which has not been reported in previous studies, although PBK/TOPK may mediate cell proliferation and viability in promyelocyte cell lines [28]. In addition, PBK/TOPK is reported to be upregulated and phosphorylated in HTLV-1-transformed T-cell lines and ATLL-derived T-cell lines [8]. Based on the complex classification of leukemia, the expression and role of PBK in each type of leukemia must be further studied. Images of immunohistochemical staining for PBK in various cancers obtained from the HPA database revealed the low to moderate expression of the PBK protein in many tumors. In previous studies, PBK was expressed at high levels in both tumors in which PBK was positively stained in the HPA database and in the negatively stained tumors in the HPA database [29,30], such as CCA, ESCC, and GBM. Notably, the expression of the PBK protein in melanoma has not been reported; however, moderate nuclear and cytoplasmic positivity was observed in the melanoma IHC images we obtained from the HPA database but requires further confirmation in the future. As a mitotic kinase, PBK plays an important role in the process of tumorigenesis and tumor progression. Not surprisingly, PBK expression was positively correlated with a more advanced pathological stage or differentiation grade of tumors, with the exception of COAD. e correlation of PBK/ TOPK protein expression with the tumor stage in COAD remains unclear. No correlation was reported in the study by Su [31], while Zlobec et al. reported that diffuse TOPK was linked to an advanced pT stage only in patients with hereditary colorectal cancer but not in patients with sporadic colorectal cancer [32], which should be explored in further studies. Furthermore, we confirmed that PBK was associated with a poor prognosis in many solid tumor patients although high expression of PBK seems to be a good prognostic factor for THYM. It is noteworthy that PBK/TOPK expression was significantly negatively correlated with TMB in THYM which is also in contrast to other cancers. Integrated proteomics result of PBK based on ProteomicsDB, MaxQB, and MOPED obtained from the GENECARD website (https://www. genecards.org/cgi-bin/carddisp.pl?gene�PBK) suggests that PBK is likely expressed in immune cells. Because the thymus is an immune organ, the role of PBK in THYM may be different from other cancers and it deserves further exploration. PBK/TOPK expression was positively correlated with the expression of DNA mismatch genes and methyltransferase expression, in the majority of cancers except CHOL and UCS.
is difference may be due to different genetic backgrounds of different cancers. Current studies of PBK in CHOL and UCS are limited, and further studies are needed. Overall, we propose that PBK represents a potential therapeutic target in cancer therapy.
Cancer immunity cycle can be divided into three stages: elimination, equilibrium, and escape. Cancer cells can escape the monitoring, recognition, and attack of the immune system and continue to proliferate by modifying their own surface antigens, recruiting inhibitory immune cells and molecules, and modifying the tumor microenvironment [33]. In this study, PBK expression was found closely related to immune infiltration cells in various cancers, especially in KIRC, LGG, and LIHC. PBK expression was positively correlated not only with immune infiltration cells including M2 macrophage, Tregs, CAFs, and MDSCs but also with T-cell exhaustion, which are involved in tumor immune escape [24,34]. Similarly, Wang et al. suggested PBK/MSL1/ CD276 signaling axis, which may play an important role in immune evasion of nasopharyngeal carcinoma and may be targeted for cancer immunotherapy [12]. Huang et al. identified PBK as one of the 14 hub genes correlated with immune cell infiltration in LIHC [11]. ey found that the expression of these hub genes was significantly positively correlated with Tregs, TFH cells, and M0 macrophages, consistent with our results, but negatively correlated with monocytes, which is different from our results.
is difference may be due to the weight of PBK in the hub genes. Moreover, PBK expression was negatively correlated with NK cells in LGG and LIHC. However, we did not find a consistent result of the correlation between PBK and CD8+ T cell in KIRC and LGG based on the TIMER and QUANTISEQ algorithm of TIMER2.0. On the one hand, the correlations were too weak to reach statistical differences; on the other hand, it also suggests that we should perform different bioinformatics algorithms to improve the dependability of analyzed results. is result suggested a very interesting phenomenon in which PBK, a mitotic kinase,  may promote the proliferation of immune infiltration cells to potentially enhance immune escape. Moreover, although, in LIHC, PBK showed a positive correlation with CD8+ T cells, it still cannot be excluded that these cells may contain bystander CD8+ T cells that are not tumor killing [35], which remains to be verified by further experiments. We observed significant positive correlations between PBK expression and the TMB in KIRC and LGG, as well as MSI in LIHC. PBK expression was positively correlated with the expression of immune checkpoint genes in all 3 tumors. Given the complexity of effectors of immune efficacy, the efficacy of immune checkpoint inhibitor treatment cannot be effectively predicted by a single biomarker. We used the TIDE algorithm to evaluate the association of the immune response with PBK expression in the 3 cancers. TIDE is a bioinformatic algorithm to model 2 distinct mechanisms of tumor immune escape. PBK overexpression groups presented a higher TIDE score, suggesting more T-cell dysfunction and elimination characteristics in the 3 tumors, which again suggested the important role of PBK in tumor immune escape.
In addition to the important role of PBK in cell cycle regulation, we confirmed that PBK was closely related to immune regulation especially in LIHC and LGG via the enrichment analysis. In fact, it has been reported that potent kinase inhibitors of CDK4/6 with the goal of slowing tumor growth have been observed as a secondary unanticipated effect: exhibited a change in the TME, and increased sensitivity to ICI relative to controls in certain tumors including breast cancer, pancreatic cancer, and melanoma in preclinical studies [33]. Moreover, enhanced cell cycle progression signatures were significantly enriched in immunologically "cold" tumors compared with "hot" tumors [33,36]. Due to the oncogenic role of PBK to promote the cell cycle in most tumors and its possible role in promoting immune escape, especially in KIRC, LGG, and LIHC, this study provides broad research space and clinical application prospects for PBK inhibitors combined with immunotherapy.
is study is a bioinformatics and computational biology analysis based on multiple databases and TCGA data that provides a new perspective on immune combination therapy but requires extensive validation in the future.

Conclusion
is study analysis revealed that PBK plays not only the oncogenic role in most solid tumors but also the role in promoting immune escape especially in KIRC, LGG, and LIHC. is study provides a potential for PBK inhibitors combined with immunotherapy.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.

Authors' Contributions
Tingting Feng and Yan Zhang contributed equally to this work. Table S1: expression comparison of PBK/TOPK expression in tumor and normal tissues across cancers from TCGA and GTEx. Table S2: the correlation of PBK/TOPK expression with immune cells in TIMER2.0. Table S3: the correlation of PBK/TOPK expression with TMB across cancers from TCGA. Table S4: the correlation of PBK/ TOPK expression with MSI across cancers from TCGA. Table S5: the correlation of PBK/TOPK expression with the expression of immune checkpoints genes across cancers from TCGA. Table S6: analysis of the correlation between PBK/TOPK expression and the immune response based on TIDE in KRIC, LGG, and LIHC. Table S7: the correlations of PBK/TOPK with DNA mismatch genes and methyltransferases. Table S8: similar genes of PBK/TOPK from GEPIA2. Table S9: the correlation of PBK/TOPK with top 5 similar genes from GEPIA2. Table S10: the result of Venn. Table S11: the GO and KEGG enrichment analysis of PBK/ TOPK-related differentially genes in KIRC. Table S12: the GO and KEGG enrichment analysis of PBK/TOPK-related differentially genes in LGG. Table S13: the GO and KEGG enrichment analysis of PBK/TOPK-related differentially genes in LIHC. Table S14: Gene_outcome of PBK in the TIMER2.0 database. Figure S1: PBK mRNA expression based on the pathological stage and tumor grade of other cancers in TCGA. Click the link to download the supplements: (https://pan.baidu.com/s/1GFqYHhkAK0Y_34 zLnH049g) (password 1234). (Supplementary Materials)