Prognostic significance and immunologic features of the paired-box (PAXs) family: a pan-cancer multi- omics analysis

doi:10.21203/rs.3.rs-3968824/v1

Download PDF

Research Article

Prognostic significance and immunologic features of the paired-box (PAXs) family: a pan-cancer multi- omics analysis

https://doi.org/10.21203/rs.3.rs-3968824/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective

The PAX genes, comprising a family of nine clearly defined paired-box transcription factors, are associated with the onset and progression of certain tumors. Even so, no extensive systematic investigation toward the contribution of PAX genes in pan-cancer has been implemented.

Methods

The development and modulation of the PAX gene family in pan tumor and its correlations with prognosis, tumor mutation burden (TMB), microsatellite instability (MSI), immunological subtypes, immune checkpoint genes, tumour stemness, tumor microenvironment, chemotherapeutics sensitivity, and effectiveness of immunotherapy were explored by bioinformatics analysis, based on multi-omics data from TCGA, GEO, cBioPortal, and TIMER database.

Results

We observed the significant correlations between the regulation of particular PAX family members in pan tumor and the survival prognosis and tumor stage of patients, TMB, MSI, stemness scores, immune cells infiltration, etc. The PAX gene family exhibited some degree of heterogeneity in different cancers in terms of the above mentioned findings. It has also been revealed in the present multiple omics study that the expression for most of the PAX family members, including PAX1/3/5/8/9, is significantly correlated with copy number variation. Moreover, we also found that several PAX family members were clearly associated with expression of immune checkpoint genes, the sensitivity to chemotherapy agents, and anti-PD-L1/PD-1 immunotherapy. Furthermore, the invading immune evaluation in bladder tumors displayed substantial correlations between PAX gene variations in copy number or substitution levels and the extent of multiple immune cell infiltration. In addition, the mRNA and amino acid manifestations of PAX8 in BLCA were validated using real-time PCR (RT-PCR) and the Human Protein Atlas (HPA).

Conclusion

In summary, our findings highlight the importance of PAX family genes in predictions of various tumor types, as evidenced by multiple datasets and identified PAX-associated genes that could be used as targets for therapies. These results suggest that PAX family related genes can be used as potential prognostic markers for cancer. It represents a systematic analysis of the further function of PAX family genes, which can provide new ideas for the prognosis and treatment of various cancers.

the paired-box family

pan cancer

prognosis

tumor microenvironment

immunotherapy

Cancer represents a significant global health challenge, with genetic variations and atypical expressions engaging significant functions in its formation and progression.. As a result, analyzing tumor-associated genes is critical for knowing more about the underlying causes of tumor formation. Understanding the association between significant genes and pan-cancer patterns can aid in cancer prevention and treatment by taking into consideration the several similarities across different types of cancer. Using substantial samples, large-scale approaches, and varied cancer databases may assist in identifying certain genes that regulate malignancy. Several extensive database platforms, including the Gene Expression Overview (GEO) and the Cancer Genome Atlas (TCGA), provide multi-omics information on a variety of tumors, enabling pan-cancer studies across multiple Omics.

The paired box (PAXs) family is mainly divided into four subfamilies based on domain composition and sequence homology: PAX1/9 (PAX1, PAX9), PAX2/5/8 (PAX2, PAX5, PAX8), PAX3/7 (PAX3, PAX7), and PAX4/6 (PAX4, PAX6). Among them, PAX2 and PAX8 have been identified in renal disorders related to embryonic development and abnormal epithelial cell proliferation [1]. And PAX2 has been shown to differ from other genes in the auxiliary diagnosis of endometrial cancer, with marker significance [2]. PAX2, PAX4, and PAX6 have significant functions in the progression and proliferation of the pancreas, providing valuable insights for studying diabetes and its management [3]. The interaction of the PAX1/9 subfamily with carcinoma cells is critical in cancer diagnosis or detection [4]. For example, PAX9 is a transcription element belonging to the PAX family [5], which acts as either a transcription-activating agent or a suppressor. It has been studied in multiple tumor forms, particularly HNSCC, ESCC, lung tumor, and cervical carcinoma of squamous cells. PAX5 was related to the onset and evolution of breast carcinoma [6].

There is a lack of integrated differentiation and correlation analysis in all investigations. In this paper, the activity and the control of PAX expressed genes to pan tumors, including its association for forecasting, immune surveillance, cancer microenvironment, and other aspects, has been studied by bioinformatics analysis and multi omics research. Based on multi-Omics analysis, this review focuses on the distinct regulation of PAX related genes in pan-cancer and the immunological characteristics of the PAX family.

An integrated evaluation of PAX gene members in various tumor types revealed significant variations in PAX gene expression between malignant cells and adjacent non-cancerous tissues, with various family members showing significant correlations with cancer patient prognosis. Furthermore, ongoing research has revealed the roles of PAX genes within the cancer microenvironment, which may improve the efficacy of immune-mediated therapy. Furthermore, we investigated the correlation between PAX gene activation and tumor diagnosis, SNP mutations of PAX familial genes in all cancers and associated with patient outcomes, and the prospective use of PAX family genes in anti-cancer treatments. The research study offers a conceptual and experimental foundation to continue examining the mechanism of the PAX familial genes in tumors that will aid advancements of novel concepts and approaches for treating tumors.

2.1 Collection of Data

The TCGA was a publicly funded program focused on documenting and identifying significant tumor-inducing genetic modifications in large numbers of over 33 distinctive human cancer varieties. This is accomplished through massive DNA sequencing and multifaceted incorporated analysis. The availability of publicly available cancer genomic data sets will lead to improved diagnostic methods, improved standards of care, and ultimately improved cancer prevention [7]. Data on mRNA transcription expression, clinical details, survival statistics, mutation profiles, immunophenotype information, and stemness index evaluations for thirty-three distinct tumors from the TCGA originated at the University of California, Santa Cruz's Xena dataset. According to GSE78220 (n = 28; Palmer show bead sheet resistance to treat melanoma), GSE111636 (n = 11; Pembrolizumab for progressive urothelial cancers) and GSE67501 (n = 11; Nivolumab for renal cell carcinoma) three GEO immunotherapy datasets, quantile normalization and gene annotation were performed by using the microarray framework records GPL11154, GPL17586, and GPL14951.. In addition, 33 kinds of tumors including adrenocortical carcinoma (ACC), bladder (BRCA) (BLCA), breast cancer, cervical cancer (from cesg), bile duct carcinoma (CHOL), colon (COAD), large B cell lymphoma (DLBC), esophageal (ESCA), glioblastoma (GBM), the head and neck cancer (HNSC), renal color cells Carcinoma (KICH), clear cell renal cell carcinoma (KIRC), papillary renal cell carcinoma (KIRP), acute myeloid leukemia (LAML), low grade glioma (LGG), liver cancer (LIHC), Lung adenocarcinoma (LUAD), lung cancer (Lung), lung squamous cell carcinoma (LUSC), mesothelioma (MESO), ovarian cancer (OV), pancreatic cancer (PAAD), pheochromocytoma and paraganglioma (PCPG), prostate cancer (PRAD), rectal cancer (READ), sarcoma (SARC), melanoma (SKCM), gastric cancer (STAD), testicular cancer (TGCT), thyroid cancer (THCA), thymoma (THYM), endometrioid cancer (UCEC), daughter Palace carcinosarcoma (UCS), ocular melanoma (UVM).

Using IMvigor210 mRNA sequencing data set, which includes 348 individuals with muscular invasion bladder carcinoma (MIBC) received therapy with the PD-L1 inhibitor atezolizumab., we tested the potential of PAXs family genes as markers of immunotherapy efficacy. Three common tumor types that received immunotherapy were considered in this study: skin cancer, renal-cell tumors, and bladder tumors. Furthermore, we participated in an ongoing clinical trial, the TRUCE-01 trial (NCT04730219), which focuses on treating Muscle-Invasive Bladder Cancer (MIBC) with an injection of LeiLiZhu single type resistance, low-dose albumin, and paclitaxel. This was a Phase II single-arm study. We collected mRNA transcriptome sequencing data and medical records from 29 participants as part of this trial. The Wilcoxon paired test was used to investigate changes in the function of PAX genes and assess the significance of their distinct expressions.

2.2 Evaluation of PAX family expression in pan cancer tissues

Firstly, the normal data were deleted from all data of 33 TCGA tumors, and the transcript data were used for generating box plots comparing the degree of expression of nine different PAX (PAX1-9) family members. Following that, we performed a statistical assessment of the expression of genes in 18 individuals with cancer, comparing typical and cancerous tissue; this study used over 5 frequent samples. Finally, we examined how the PAX gene family was expressed differently in 18 different cancers, including BLCA, CHOL, KIRP, LUAD, BRCA, ESCA, COAD, PRAD, LUSC, HNSC, GBM, KICH, LIHC, THCA, KIRC, STAD, READ, and UCEC. All the data were processed using the Wilcox test to compare differences between paracancerous healthy tissue and cancer tissue. Following that, the R corrplot package was employed to conduct more analysis of PAX gene associations.

2.3 Survival Analysis

Individuals were assigned to two groups based on their average activity rates for each PAX gene family: those with significant expression of genes and those exhibiting low expressed genes. We created Kaplan-Meier curves (K-M) to illustrate the differences in patient survival rates. The Univariate Cox regression model was used to determine the predictive value of gene activity on overall survival (OS), disease-specific survival (DSS), and progression-free survival (PFS).

2.4 PAX mutation frequency, patient outcome consequences, and PAX association with CNV or SNV

The cBioPortal for Cancer Genomics is available at http://cbioportal.org, representing an open-access database containing a variety of data types. These include omatic variations, copy number variations (CNVs), expression of mRNA profiles, DNA modification patterns, protein enrichment analyses, and confined anonymous patient data [8]. In a 2020 contains a cohort study on 2683 cases of led/TCGA genome-wide Cancer Analysis union (Pan - Cancer Analysis of Whole Genomes Consortium) a survey of a large amount of data in the queue, And discusses the PAX mutation frequency and the impact on prognosis of patients, and PAX expression level and single nucleotide mutation (SNVs) and the correlation between CNVs.

2.5 Association amongst PAX expression, immunological subgroup, and cancer microenvironment

With the TCGA dataset, study examined the link between PAX transcription and immunological variants in pan-cancer scenarios. We calculated immune and stromal scores for each TCGA tumor sample using the ESTIMATE method. Then, we used Spearman's coefficient of association to explore the relationship between PAX gene expression rates and immune and stromal scores.

2.6 PAX activation linked to cancer mutational burden (TMB), microsatellite instability (MSI), and immunological checkpoint activity

Microsatellite instability (MSI) testing and tumor mutational burden (TMB) are biological markers used to identify patients who may benefit from immune-regulated checkpoint inhibitors [9]. We used an evaluation package to assess the immune microenvironment of cancer patient samples and Spearman correlation tests to investigate the associations among PAX genes and TMB as well as MSI. In addition, we explored the relationship between PAX genes and immunological checkpoint indicators such as PD-L1, PD-1, CTLA-4, and others in 33 TCGA tumor types.

2.7 Tumor Immune Estimation Resource

Tumor Immune Estimation Resource 2.0 (TIMER 2.0), available at http://timer.comp-genomics.org, tool was used to analyze various immune cell infiltrations [10]. It offers insight into the link between the extent of activity for all members of the PAX family and the rate of infiltration of lymphocytes in a variety of malignancies.

2.8 Pan-cancer sensitivity to drugs

Using CellMiner database accessible at https://discover.nci.nih.gov/cellminer/home.do and drug susceptibility in the equivalent information on gene expression samples following FDA standard authentication and certification in clinical laboratory, drug susceptibility data filtering, the Spearman correlation evaluation was employed to compare the associations with PAX family gene regulation and sensitivity to drugs.

2.9 Evaluation of Gene Set Enrichment in Bladder tumor

Employing the Kyoto Encyclopedia of Genes and Genomes (KEGG) gene set (MSigDB), GSEA evaluation of the physiological roles and potential regulatory networks of PAXs in bladder tumors has been identified, and the correlation analysis of PAX family genes was performed. Spearman correlation coefficients were calculated. GSEA is executed using the R package 'cluster Profiler'.

2.10 RNA Extraction and qRT-PCR; human protein atlas (HPA) analysis

We initially isolated the whole RNA with ABP Biosciences, Inc.'s Nuclezol LS RNA Isolation Reagent. Then, through GeneCopoeia's SureScript-First-Strand-cDNA-synthesis kit in Guangzhou, we synthesized cDNA from 1.0 g of total RNA. The key genes were quantified using the GeneCopoeia BlazeTaqTM SYBR Green qPCR Mix 2.0 kit, and the analysis was carried out on a Bio-Rad CFX96 real-time quantitative PCR instrument. The PAX8 primer sequences were as follows: forward, 5’-GCCTTCTCCCTCTGCCTTTA-3’; reverse, 5’- TTATGCAGGCTCCAGTCACA-3’. The 2^−ΔΔCT method was applied to evaluate the findings, and each qPCR reaction was replicated thrice. The Human Protein Mapping Database (HPA), available at https://www.proteinatlas.org, happens to offering comprehensive information about the distribution of human proteins in various tissue and cell types [11]. Use a database of HPA further validation PAX8 protein expression.

2.11 Statistical Analysis

To determine the prevalence of PAX genes in cancer and nearby cells, we used the Wilcoxon rank sum test. The log-rank test served to assess variances among subdivisions, and the Pearson or Spearman correlation coefficient was used to determine associations. We used Kaplan-Meier curve analysis to contrast survivability across the high and low-expression cohorts. In 33 types of cancer, a univariate Cox regression study employed to determine the distinct predictive component of the PAX gene on OS, DFS, DSS, and PFS. SPSS (version 23.0) or the R program (version 4.0.2) were used for all statistical analysis and mapping. Each analysis was executed out via R and the R package, and a finding with a P value less than 0.05 was regarded as significant.

3.1 Identification of PAX family manifestation in pan cancer

Figure 1 illustrates an extensive graphical representation of the current study. Initially, we assessed the gene expression of nine PAX family genes and ranked them from high to low in the 33 types of cancer using the TCGA database, separately (Fig. 2A; Supplementary Figure S1)through the TCGA pan cancer statistics. Among 33 tumor types, PAX8 and PAX9 genes are highly expressed in pan cancer compared to other genes, as shown in Fig. 2A. Afterwards, we plotted the differential heatmap of a single family member between the 18 tumor and matched para-tumor samples with size of more than 5, was given in Fig. 2B, and Supplementary Figure S2. The PAX8 gene has been discovered to be extremely prevalent in tumor cells such as UCEC, CHOL, STAD, BLCA, and low expressed in cancers such as KICH, KIRC, LUSC, and THCA. Expression differences of each PAX family member was observed in LUSC; that is, PAX1-7 and PAX 9 are highly expressed, whereas PAX8 represents low expression in LUSC. In addition, we also analyzed the pearson correlations among PAX family genes. As seen in Fig. 2C, most of these genes were negatively correlated. A substantial favourable association between PAX2 and PAX8 (correlation coefficient = 0.49), while an intriguing adverse relationship exists among PAX2 and PAX9 (correlation coefficient = -0.2).

3.2 KM survival curve analysis and unicox regression study

To examine the consequences of PAX family expression for diagnosis, we studied the association with PAX expression and predicted survival in 33 TCGA cancers. First, via setting the optimal cutoff determined by the “survival” R package, we executed K-M survival study for comparing survival differences of OS, PFS, DFS and DSS between patients with high and low PAXs expression (Supplementary Table S1; Additional Data1). Of note, most PAX family members were significantly positively or negatively associated with OS, PFS, DFS and DSS with large prognosis conformance in many types of cancer.

Afterwards, (Fig. 3A) illustrated the manifestation of PAX family members has been linked to OS time by the univariate cox regression method in many types of cancer The predictive factors were characterized by the risk ratio (HR > 1) and p-value (p < 0.05) in the forest plot. Taking PAX1 as an example, its expression can affect the survival time of GBM (HR = 0.50, 95% CI 0.28-0.89P = 0.02), KIRP (HR = 114.18, 95% CI = 5.68-2292.94, P = 0.002), and THCA (HR = 1.98, 95% CI 1.25–3.12, P = 0.003), and this effect is statistically significant. In addition to investigating survival time, we also conducted HR95% CI analysis on PFS, DSS, and DFS of pan cancer, and visualized the data. Grey blocks with no statistical significance were used to distinguish them (Fig. 3B-D). Similar to Fig. 3A, a HR greater than one indicates a protective variable, whereas an HR less than one indicates a risk factor.

3.3 Genome variation of PAX family in pan-cancer

Based on the mutation database obtained from Pan tumors evaluation of entire genomes (ICGC-TCGA, Nature 2020)containing 2683 samples, we conducted a brief introduction to the mutation situation and survival analysis of members of this family. First, we counted the overall gene mutation of the PAX family in the 2683 samples (Fig. 4A), and examined the gene family type of mutation for all tumor species (Fig. 4B). The findings found that the frequency of mutations of the PAX family genes accounted for 21.8% of pan-cancer cases, of which Amplification accounted for 15.51%. In addition, in order to more accurately show the mutation diversity of each gene in the family, we conducted a statistical analysis of mutation frequency of each gene in the sample (Fig. 4C), and found that the mutation rate of PAX1, PAX4 and PAX5 was higher. We analyzed the overall survival probability of the data, and made a survival curve to express the correlation more intuitively (Fig. 4D). Obviously, the non-mutant cohort showed a greater probability of survival compared to the mutant population.(P = 8.471e-4).

Furthermore, mutation evidence for all PAX family members are provided in Supplementary Figure S3A-I, respectively. Next, we examined how their distinct variants influenced the outcome of cancer compared to non-mutation cohorts (Supplemental Figure S4A-I). Patients with cancer using PAX3/4/8 variants possess a less favorable outlook when compared with non-missense modifications. According to Fig. 5, the research showed mRNA levels of PAX6, PAX8 and PAX9 genes increased with the increase of DNA copy number (P ≤ 0.05). However, there were no significant variations in PAX activation among different genomic variant types, except for PAX5 (Supplementary Figure S5). These results illustrated that copy number variation of some PAXs in pan-cancer are positively associated with their mRNAregulation.

3.4 Correlation between clinicopathologic traits and PAX family members

For the tumor stage or the first treatment effect of the patient, the correlation evaluation of each PAX gene activity in pan cancer with clinical traits of interest, such as the tumor stage and the first treatment effect of the patient, were carried out (Fig. 6). Evidently, the activity of PAX3 and PAX9 increases over time with the progression of KIRC or TGCT tumor stages.. Similarly, the expression of PAX5 was increased as tumor stage progression of BLCA, and PAX8 was also increased in LIHC. Conversely, the activation levels of PAX2, PAX5 and PAX8 tended to decrease as the cancer stage of KIRP or TGCT increased.

Moreover, we performed association analysis between first-course therapy results and PAX expression of genes. (Supplementary Figure S6). Notably, after the initial phase of ACC therapy, PAX6 expression was considerably greater within the SD/PD group compared to the CR/PR association, PCPG and UCS, whereas it was significantly downregulated in OV and PRAD. Compared to CR/PR cohort, PAX2, PAX3, or PAX4 expression were up-regulated in ACC, LGG, and PRAD/STAD tumor in SD/PD cohort, respectively; conversely, PAX2 in UCEC, PAX5 in HNSC and LUAD, PAX8 in HNSC, and PAX9 in OV and PRAD, were all decreased in SD/PD cohort.

3.5 Association analyses of the PAXs expression with immunization typing, tumor stemness scores, and tumor microenvironment in pan-cancer

The association of expression between PAX gene in pan-cancer samples and six different subtypes of immune infiltration, including C1 (wound-healing), C2 (IFN-dominated), C3 (inflammatory), C4 (lymphocyte depletion), C5 (immune silence), C6 (TGF-β-dominated) was examined (Fig. 7A). In pan-cancerous tissues, the PAX1, PAX6, and PAX7 manifestation content were found to be highest in subtype C5, while PAX8 was particularly abundant in subtype C3.. Additionally, PAX5 and PAX9 displayed the highest expression in subtype C2. Within each subtype, PAX4 showed minimal activation compared to other PAX family members. Meanwhile, we focused on the link between PAX gene expression and lymphocyte invasion. With the exception of THYM, our studies revealed a distinct significant association between the activity of PAX5 or PAX4 and cytotoxic T-cells (CD8+) infiltrate across all types of cancers. Furthermore, we found a strong beneficial association between PAX6 levels and CD8 + T/NK cell invasion. PAX6 and PAX9 had a substantial adverse association with NK-T cells. Moreover, in nearly all forms of tumors, PAX1 has a significant adverse correlation with the invasion of CD8 + T/NK/NK-T cells as seen in Fig. 7B. Furthermore, study examined the association between PAX activity and immunological and stromal scores in all cancer tissues, results were identified through the R statistical tool (Fig. 7C-D). The results revealed a positively significant correlation between PAX5 regulation and immune/stromal scores in various types of tumors. In contrast, a strong negative association was found between PAX2 activity and immune/stromal scores. Moreover, PAX1 showed an elevated favorable association with immunized scores and an alarmingly low correlation with stromal scores. Following that, tumor stemness scores derived from the expression of mRNA (RNAss) and the methylation of DNA sequence (DNAss) assisted in the analysis of the association between PAX genes and tumor stemness indexes in various tumor types (Fig. 7E-F). It should be noted that the association of PAX genes is complex and the levels of RNAss or DNAss exhibits a certain degree of heterogeneity in different cancer types. For instance, in THYM tumor, there was a significantly positive correlation observed between PAX1/3/9 gene expression and DNAss, while a strong negative correlation was observed with RNAss.

3.6 Correlation analysis of PAXs with immunological checkpoints expression, tumor mutational burden (TMB), and microsatellite instability (MSI)

Following previous reports, TMB MSI, and immune checkpoint-related genes can serve as effective predictors for immune checkpoint inhibitors (ICIs). Therefore, we evaluated the latent correlations of PAX genes with these common ICIs biomarkers, such as immune checkpoint genes, TMB, and MSI in 33 cancer types (Fig. 8; Supplementary Figure S7-9). To begin, we collected data from more than forty prevalent immune checkpoint genes and produced a correlation chart to investigate the link between the activity of all PAX family members and immunological checkpoints such as PD-L1, PD-1, and CTLA-4 etc. (Fig. 8A; Supplementary Figure S7). The results found that the PAX5 gene strongly correlates with PD-L1 immune checkpoints in a variety of cancers., such as BLCA (P < 0.01, Cor = 0.31), BRCA (P < 0.001, Cor = 0.56), TGCT (P < 0.001, Cor = 0.39), THCA (P < 0.001, Cor = 0.23), etc. According to Fig. 8B, the abundance of PAX5 gene was negatively related to various cancers (P < 0.05), such as CHOL, HNSC, KIRC, KIRP, LIHC, TGCT, etc. In most cancer types, the PAX9 gene has been linked with TMB (P < 0.05), such as COAD, LAML, MESO, THYM, etc. Furthermore, across a variety of cancers, the activity level of the PAX5 gene did not correlate well with MSI (Fig. 8C).

3.7 Sensitivity to drugs screening

We explored the association of PAX gene activity and clinical chemotherapy drug efficacy. A scatter plot was drawn in Fig. 9A, sorted by the size of p value (p < 0.002). The PAX3 gene linked with increased sensitivity of ‘Dabrafenib’ (Cor = 0.675, P < 0.001), ‘Vemurafenib’ (Cor = 0.674, P < 0.001), ‘Selumetinib’ (Cor = 0.536, P < 0.001), ‘Hyphenhemycin’ (Cor = 0.470, P < 0.001), ‘Cobimetinib (Isomer 1)’ (Cor = 0.453, P < 0.001), ‘PD-98059’ (Cor = 0.441, P < 0.001), and ‘Bafetinib’ (Cor = 0.436, P < 0.001). PAX8 is positively sensitive to some compounds, such as ‘Erlotinib’ (Cor = 0.514, P < 0.001), ‘Ibrutinib’ (Cor = 0.507, P < 0.001), ‘Afatinib’ (Cor = 0.481, P < 0.001), ‘AZD-9291’ (Cor = 0.456, P < 0.001), ‘Lapatinib’ (Cor = 0.443, P < 0.001), however, it is negatively correlated with ‘Tyrothricin’, ‘Vinblastine’, ‘Paclitaxel’, ‘Okadaic acid’, ‘Actinomycin D’, and ‘Dolastatin 10’ (P < 0.001). Furthermore, we examined the association involving the response to frequently administered chemotherapy agents for bladder tumors and the gene expression of the PAX family (Fig. 9B, p < 0.05). Our findings revealed an inverse association with the manifestation of PAX2 and PAX8 and the IC50 of ‘Paclitaxel’ and ‘Vinblastine’. PAX4 on the other hand showed positive correlations with ‘Paclitaxel’ and ‘Oxaliplatin’.

3.8 The association between the proliferation of PAXs and effectiveness of clinical immunotherapy

Nowadays, immunotherapy becomes a significant component of the treatment of cancer, so understanding the association involving gene activity and immunotherapy outcomes is critical. We examined the relationship of the regulation of PAX family genes and the response to immunotherapy in multiple datasets in the present research. These datasets included three BLCA-related datasets (IMvigor210, GSE111636, and GSE176307), a carcinoma database (GSE78220), and one renal cell carcinoma dataset (GSE67501) (Fig. 10A). In the Imvigor210 and GSE176307 groups, study found a link across PAX3 and PAX6 expression rates and target outcomes of anti-PD-1/PD-L1 treatment. A similar result was observed in the GSE78220 dataset for PAX8 and in the GSE67501 cohort for PAX9. Conversely, a positive association of PAX2 and PAX7 expression and resistance to cancer immunotherapy was noted in the GSE176307.

In addition, we conducted prospective processes underlying PAX family activity and their reaction to immunotherapy (Fig. 10B; Supplementary Figure S10). Our analysis of the Truce-01 sequencing data revealed a significant tendency for decreased PAX3 and PAX6 expressed rates in bladder tumor sufferers who received tislelizumab together with nab-paclitaxel treatment. In summary, these findings indicate that PAX gene regulation could potentially serve as an indicator of reaction to anti-PD-1/PD-L1 immune therapy.

3.9 Correlation analysis of PAX genes with immunological subsets and biological aspects in BLCA

We observed that PAX5 genes had been expressed in the C6 subtype, as seen in Supplementary Figure S11A, was highest than that in the other subtype, and expression of PAX6 genes in the C2 subtype was highest. In addition, we examined the associations of PAX proliferation and clinicopathologic traits in patients with bladder cancer, such as tumor grade, cancer subtype, stage, and T-stage. (Supplementary Figure S11B-E). It is noteworthy that bladder cancer patients with higher-grade, non-papillary, high-stage, and high-T_stage had elevated PAX1/5 expression, and similar results were seen for PAX6 in higher-grade and non-papillary levels. However, high-grade bladder cancer patients had lower PAX8 expression compared with expression in the low-grade.

3.10 Correlation studies of PAX genes expression with stemcell index, and immunological features in bladder cancer

The present study conducted an analysis on the correlation across the PAX family genes in BLCA and various factors including the tumor microenvironment, stem cell index, TMB, and MSI (Fig. 11). The results revealed a negative correlation between PAX1/3/5/6 and RNAss or DNAss, while PAX2/4/9 exhibited a positive correlation. Apparently, PAX2/7/9 was negatively correlated with three fractions scores of TME, while PAX5/6/8 were positively associated. By analyzing the related regulation of PAXs gens and TMB or MSI in bladder tumor, we found that PAX2/3/6 expression present positive correlations, PAX5/9 expression show negative correlations, but not significant.

In addition, PAX3/5/6 activation appears to be strongly associated with CD8 + cell proliferation, Macrophages, Neutrophils, and Dendritic cells.; Similarly, PAX1 with Macrophage, PAX8 with Neutrophil, and PAX9 with B cell represented a positive correlation in bladder cancer. Conversely, PAX1 with CD8 + T cells/Neutrophil, and PAX2 with Neutrophil/Dendritic cells in BLCA, displayed a significant negative association (Fig. 12). No clear correlation has been found between PAX4/7 and conventional immune cell types. Although, results revealed the PAX gene expression may potentially influence immune infiltration within the tumor microenvironment, thereby potentially impacting patient survival.

3.11 Gene set enrichment analysis of bladder cancer

To discover the fundamental cause of PAX family genes in bladder cancer, we used the GSEA approach to conduct KEGG pathway enrichment analyses (Fig. 13; Supplementary Table S3). The results suggested that overexpression PAX5 may potentially positively regulate ‘Hematopoietic cell lineage’, ‘Cytokine cytokine receptor interaction’, ‘Chemokine signaling pathway’, ‘T cell receptor signaling pathway’, ‘B cell receptor signaling pathway’, ‘Natural killer cell mediated cytotoxicity’, ‘Toll like receptor signaling pathway’, ‘Antigen processing and presentation’, ‘Neuroactive ligand receptor interaction’ and ‘Calcium signaling pathway’, etc. Similarly, upregulation of PAX6 positively regulates ‘Neuroactive ligand receptor interaction’ ‘Hematopoietic cell lineage’, and ‘Cytokine cytokine receptor interaction’, etc. Upregulation of PAX1 positively regulates ‘Neuroactive ligand receptor interaction’, and ‘Mapk signaling pathway’, etc. Conversely, downregulated PAX9 was primarily enriched for the above-mentioned KEGG pathways, such as ‘Neuroactive ligand receptor interaction’, ‘Antigen processing and presentation’, ‘Hematopoietic cell lineage’, ‘Cytokine cytokine receptor interaction’, and ‘Jak stat signaling pathway’, etc. Next, we performed also functional annotation analysis on the PAX genes using the Hallmark gene sets (Supplementary Figure S12; Supplementary Table S4). As can be seen, high expression of PAX1/5 and low expression of PAX9 all enriched ‘Epithelial Mesenchymal Transition’. Furthermore, high expression of PAX5 and PAX5 enriched ‘Inflammatory response’, ‘Allograft rejection’, ‘Interferon alpha response’, or ‘Interferon gamma response’. Thus, these PAX genes play an important role linking cancers and immunological characteristics.

3.12 Association analysis between mutation and gene copy number of PAX family and the infiltration abundance of immune cells

TIMER database performed for correlation analysis between the copy number variations of each PAX family member and the content of common immune cells (Fig. 14A). Single arm/deep deletion of PAX2/3/5/6/8 decreases CD8+, CD4+, neutrophils, macrophages or dendritic cells, while single arm gain of PAX4 increases CD8 + T cells. Besides, single arm gain/high amplication versus normal of PAX7 decreases CD4 + T cells, neutrophils and dendritic cells, whereas its single arm gain increases B cells. Similarly, single arm gain/high amplification of PAX2/9 decreases B cells, CD4 + T cells or DCs, PAX4/5/6 decreases B cells, and PAX3/8 decreases CD4 + T cells, were also observed.

At the same time, the abundance differences of the major immune effector cells were analyzed between the mutated and non-mutated groups of each PAX family gene (Fig. 14B). Mutation of PAX3 and PAX7 increases the content of CD8 + T cells when compared to wild-type, however PAX4 mutations decrease CD8 + effector memory T cells and activated NK cells.

3.13 PAX8 gene expression in cancer cells and near healthy tissues

According to the HPA database, PAX8 activity was 25%-75% in tumor tissues but fewer than 25% in adjacent tissues, as shown in Fig. 15A-B. Simultaneously, qRT-PCR (P 0.01) confirmed PAX8 mRNA expression (Fig. 15C).

Many tumors have been found to have dysregulation of the PAX family gene [12]. The PAX family genes are evolutionarily quite conserved. They can encode a series of transcription factors and transcriptional regulators related to growth and development regulation, and exist in various organisms. It exerts its effects on cell and tissue formation and differentiation by regulating many intracellular signaling pathways. Although PAX family genes are associated with proliferation, invasion and progression of adverse outcomes in some cancers, there is no integrated analysis of PAX family genes across various tumors [20–22]. The present research systematically investigated PAX family gene expression and prognostic significance among individual malignancies, the association with clinical and immunological characterization, as well as the mechanism of tumor progression based on multi-omics data.

Our research discovered variations in the gene expression of numerous PAX family genes between para-cancerous and healthy tissues in a number of cancer types, indicating that these genes may be significant in cancer development and progression. For example, PAX8 expression of genes varied considerably between tumor tissues with high levels and noncancerous tissues (P ≤ 0.05). We speculate that this gene may promote carcinogenesis when highly expressed, but the specific mechanism remains unclear. Studies have shown that PAX8 to be a co-activator of transcription in renal cancer cells, activating genes that control regulating cell cycle and metabolic process.[23]. In addition, PAX8 can be activated in renal cell carcinoma to regulate the cell cycle and metabolic process of tumor cells. In addition, the clearance of PAX8 may reduce the invasiveness of renal cell carcinoma cells [24]. In addition, long-term Pax4 overexpression has been found to lead to hyperglycemia in pancreatic cancer by affecting β cell dedifferentiation [25]. This indicates that PAX member expression is closely related to the prognosis of some cancers In addition, we investigated the distinctions in the tumor cells and adjacent tissue in pan-cancer, as well as the impact of this gene family on the prognosis of various cancers. The analysis results show an elevated PAX1 expression is a preventive measure for KIRP [26] and THCA, as well as a risk factor for GBM; High expression of PAX8 factor for LGG [27] and susceptible to HNSC. The conclusion shows that high expression of some PAX gene members can affect the prognostic value of individuals with associated tumors, reduce OS time, and have an effect on their survival prediction. At present, our research on specific mechanisms is insufficient and has further exploration significance.

In Fig. 4A-4B, the mutation rate of PAX family genes accounted for 21.8%, and the amplification rate accounted for 15.51%, and it showed that the mutation rate of PAX1, PAX4, and PAX5 was very high, suggesting that the alteration of that gene family contributed to individual survival.. Based on this hypothesis, we analyzed the OS patients probability, and drew the survival curve between the mutant and non-mutation groups to illustrate this relationship more intuitively. The findings showed that the non-mutation group had a considerably greater OS rate than the mutant group (P ≤ 0.05). Moreover, we explored the change of overall survival probability between PAX family members and non-mutation group. The study showed that PAX gene mutation would affect the prognosis of patients and the patient’s quality of life. We investigated the association of copy number changes of PAX gene family members and the prevalence of corresponding genes. The findings revealed that the levels of mRNA expression of PAX1, PAX3, PAX5, PAX7, PAX8, and PAX9 were significantly different from the DNA copy number (P ≤ 0.05). For example, recent studies have shown that PAX8 can regulate tumor-stroma interaction, cell adhesion, proliferation, and migration, and mutations in PAX8 are thought to be associated with the development and progression of 90% of ovarian cancers [28, 29]. PAX9 gene mutation, amplification and deletion can promote the development of lung adenocarcinoma, squamous cell carcinoma, cervical cancer, ovarian cancer and breast cancer [30]. PAX8 is highly expressed in cervical cancer, and PAX8 transcripts are upregulated in cervical cancer samples, suggesting a correlation with cervical cancer development [31], and PAX8 expression is also upregulated in uterine adenocarcinoma, squamous cell carcinoma, endometrioid adenocarcinoma, and adenosquamous carcinoma [32–44]. These results suggest that PAX member gene mutations have regulatory effects on a variety of cancers. We also visualized the mutation type of each PAX gene family member and examined the link between the mutation kind and the activity of each family member.. From Figure S5, we can find that the gene expression of PAX5 gene increased after nucleotide deletion and truncation, and the difference was statistically significant (P ≤ 0.001). Previous studies have shown that PAX5 in B-acute lymphoblastic leukemia (B-ALL) can mediate the development and progression of B-ALL by not expressing or expressing truncated proteins lacking functional domains, resulting in loss of allelic function [45–48]. In addition, study found that down-regulation of PAX2 expression by antisense oligonucleotides is associated with growth inhibition of human renal cancer cells [49]. In other words, the mutation of PAX gene can affect its gene expression, thereby regulating the incidence and tumor proliferation.

The present study analyzed the PAX family expression in pan-cancer and the correlation of PAX genes with common clinical features. These clinical characteristics include the patient's disease stage and the response to initial treatment. Previous studies have shown that overexpression of PAX2 can promote tumor proliferation and growth in vivo and in vitro in a mouse ovarian cancer xenograft model, and high PAX2 expression is associated with shorter PFS in 152 patients with stage III-IV serous ovarian cancer [50–52]. In other words, the expression of PAX gene family can affect the progression-free survival of patients, reflecting that the expression of PAX gene can affect the efficacy of patients and the quality of life of patients to a certain extent.

Immune cells, tumor microenvironment and stem cell index are the focus of most scholars. By analyzing the correlation of these hot spots with PAX expression, we found that PAX5 expression correlated with previous immunophenotyping in most cancers. It has been found that PAX5 can activate the chromatin of key genes involved in B cell signal transduction, adhesion, migration and immune function during B cell development, thus affecting the immunophenotype [53]. It can be inferred that the high expression of some PAX family genes can participate in the regulation of immune process, thereby changing the interaction between tumor and immune cells and promoting the occurrence and development of tumors.

TME is composed of tumor cells and non-tumor cell components, and the non-tumor cells mainly include stromal cells and various lymphocytes. They can support the biological behavior of tumors [54], for example, tumor-associated fibroblasts can promote the epithelial-mesenchymal transition of tumor cells, thereby making tumor cells more aggressive [55]. Tumor-associated macrophages can support tumor growth, angiogenesis, invasion and metastasis by secreting a large number of cytokines and growth factors [56]. In addition, the increased content of stromal cells and the infiltration of immune cells in cancer patients will promote the proliferation, invasion and metastasis of tumors, which are positively correlated with TNM stage, T stage and M stage [57]. The matrix score of the tumor microenvironment can quantify the tumor tissue interstitial cells, and the immune score can quantify the immune cells in the tumor tissue, so as to explore the purity of the tumor. We examined the link between PAX family and pan-cancer interstitial and immunological score to investigate the interaction between tumor microenvironment and PAX genes. These members found to be significantly associated with stromal scores in pan carcinoma, as well as immune scores. This suggests that regulation of PAX gene can influence the number of tumor immune cells and stromal cells, and thus the tumor microenvironment.

The expression of PAX family genes was negatively correlated with mRNA expression based dry score (RNAss) in most tumors, and also negatively correlated with DNA methylation based dry score (DNAss) with a few positive correlations. It is speculated that the expression level of PAX gene family may affect the dryness of related tumor cells. Some scholars have shown that in glioblastoma, PAX3 knockdown can inhibit the proliferation and invasion and induce cell apoptosis, while overexpression can promote the proliferation, invasion and inhibit the apoptosis of glioma cell lines [58]. Another study found that PAX9 knockdown can inhibit some biological behaviors of ovarian cancer cells, such as proliferation, invasion and migration [59]. In summary, it can be concluded that the expression of PAX family genes is significantly correlated with the stemness of tumor cells in most cancers. There is little research on its specific mechanism, so it has further exploration significance.

Tumor immunotherapy, which uses lymphocytes to combat cancer, is being shown to be a potent anti-tumor tool and is now more common in clinical trials. However, for many patients with solid cancer, the local immunodeficient cancer environment is difficult to change, and clinical progress is minimal [60]. At present, immune checkpoint inhibitors (ICIs) targeting programmed cell death protein-1 (PD-1) and programmed cell death ligand 1 (PD-L1) have been widely used in clinical practice and are often used to treat advanced cancers [61]. As a result, research into immune checkpoints is required. We extracted over 40 typical immune checkpoint genes and studied the link between the activity of various PAX family members and such immune checkpoint genes (Figure S7). For example, in various cancers, the PAX5 gene is significantly associated with the PD-L1 immune checkpoint, indicating that the expression of the PAX gene is significantly correlated with the expression of many immune checkpoint genes (Fig. 8A), and intervention of the PAX5 gene targeting the relevant immune checkpoint may be an effective therapeutic option.

Tumor mutation burden (TMB), defined as the total number of somatic gene insertions, coding errors, deletion errors, and base substitutions detected per 1 million bases in the tumor genome, is a promising new biomarker [62]. A number of studies have shown that immunotherapy is more effective in patients with high TMB, and TMB also has a certain predictive effect in malignant tumors such as lung cancer [63] and melanoma [64]. Microsatellites are short tandem repeat sequences distributed throughout the genome. When DNA mismatch repair mechanism is damaged, mutations can occur, leading to microsatellite instability (MSI), presenting a highly mutant phenotype of genomic microsatellite [65, 66].

Immune oncology, a novel pillar of modern cancer treatment, is revolutionizing the way we approach cancer therapy. Immune checkpoint inhibitors (ICIs), such as PD-1 and PD-L1, can regulate the signaling pathways of T cell-related immune responses, enabling them to produce effective anti-tumor responses. Although ICI therapy has made a breakthrough progress, it also has side effects, and not all patients respond optimally. Therefore, it is very desirable to find biomarkers that can help identify patients who are likely to benefit from ICI treatment [67]. Currently, biomarkers used in clinical practice include PD-L1 immunohistochemistry, microsatellite instability - high (MSI-H), mismatch repair deficiency (dMMR), and tumor mutational burden (TMB). In order to explore the correlation between the expression of PAX family genes and the therapeutic effect of ICI, we analyzed the correlation among the expression of PAX family members in pancarcinoma, TMB, and MSI. According to the analysis in Fig. 8, we conclude that PAX family genes are significantly correlated with TMB and MSI in some cancers, that is to say, the expression of PAX family genes is also correlated with the therapeutic effect of ICI. We can see that the expression level of PAX5 gene in MSI was negatively correlated with most cancers, such as DLBC, TGCT, etc. (P ≤ 0.05), suggesting that MSI increases with PAX5 gene expression reduction, thus making these tumors highly responsive to immunotherapy and achieving a good ICI treatment response.

In addition, we evaluated the sensitivity of BLCA to commonly used chemotherapeutic agents based on expression of PAX family genes. Studies have shown that PAX9 is highly expressed in drug-resistant ovarian cancer cell lines and tumor tissues of drug-resistant patients, and the sensitivity of drug-resistant cells to cisplatin is also increased through further in vitro experiments by down-regulating PAX9 [68]. In non-small cell lung cancer (NSCLC), studies have shown that PAX6 can regulate the stemness and differentiation of tumor cells, affect cell resistance and proliferation, and has the potential to become an effective prognostic factor of NSCLC [69]. In addition, some scholars have pointed out that PAX6 affects the efficacy of mothiazole in glioma by affecting the mechanism of action of mothiazole in vivo, which is closely related to the drug sensitivity of mothiazole [70]. However, in our study, we observed that PAX3 expression was positively correlated with drug sensitivity of Vemurafenib, Dabrafenib, Selumetinib, and other drugs, that is, patients with higher PAX3 expression had better treatment effects receiving these drugs. While PAX8 gene expression was positively correlated with clinical chemotherapy drug resistance. Patients with elevated PAX8 expression should exercise caution when considering the use of these drugs, as higher PAX8 expression corresponds to poorer treatment outcomes. In essence, PAX family gene expression can affect the chemotherapy drug sensitivity, thus affecting the curative effect and prognosis of patients. The specific mechanisms underlying these effects warrant further exploration.

BLCA is the most common urinary system tumor, the tumor has been our research hot spot. To understand the expression of PAX gene in BLCA, we examined the expression of PAX8 gene in patient tumor tissues and adjacent tissues using qPCR and immunohistochemistry. The results showed that PAX8 gene expression was higher in tumor tissues than in adjacent tissues. This suggests that PAX8 gene is highly expressed in BLCA tumor tissues, which may be related to the occurrence and development of BLCA, and the targeted therapy of PAX8 gene may be a breakthrough point in the treatment of BLCA. In recent years, great progress has been made in the treatment of BLCA, predominantly centering on chemotherapy. Although this therapy has a survival advantage, its efficacy is unsatisfactory due to severe side effects. Therefore, targeted drug therapy for the disease has become necessary, and it can also open up promising new ways to improve the prognosis of patients [71]. At present, targeting the immune checkpoint immune drugs in metastatic bladder cancer also showed very good curative effects. Thus, our endeavor is to identify targeted molecular pathways in BLCA, thereby elevating patient prognosis and survival rates.

In this study, we initially explored the PAX family genes and suggested that their expression was correlated with the tumor microenvironment and patient prognosis of BLCA. To this end, we divided all samples into high and low expression groups based on the median PAX expression level in BLCA. Subsequently, we performed GSEA to find out which Hallmark and KEGG pathways were enriched in these high and low expression groups for each PAX gene. The data showed that the KEGG_OLFACTORY_TRANSDUCTION pathway was enriched in the high expression group of PAX1 to PAX8 genes and in the low expression group of PAX9 genes. Other pathways in different PAX gene enrichment results are not completely consistent, prompt PAX family genes may participate in the BLCA progress through different mechanisms. In addition, investigating the interplay between PAX family genes and KEGG_OLFACTORY_TRANSDUCTION pathway in BLCA enhances our understanding of targeted molecular pathway therapies for BLCA.

In the TCGA cohort, we analyzed the expression level of PAX in five immune subtypes (C1-C4 and C6). Our investigation revealed notable variations in PAX5 and PAX6 expression levels across different subtypes. In addition, we examined the relationship between PAX expression and tumor grade, tumor subtype (non-papillary or papillary), tumor stage, and tumor size range stage (T stage) in BLCA. Results showed that the different tumor stage, tumor T stage, and tumor subtypes, significantly changed the PAX5 and PAX6 expression levels within bladder tumor tissues. Some studies have shown that Pax-5 can be expressed in Neuroendocrine neoplasm (NENs), and its expression is related to G grade, stage, prognosis, and positively correlated with the malignant degree of tumor [72]. Sun Fei et al. revealed that the higher the expression of PAX8, the higher the pathological classification of glioma, and there was a significant positive correlation between the two, and stressed that PAX8 can promote the occurrence and development of glioma [73]. In the current study, gene expression levels of PAX5 and PAX6 genes significantly correlated with tumor unhealthy grade, cancer stage, and T stage, and the expression levels of PAX5 and PAX6 genes in the non-papillary type of cancer increased compared to those in the papillary type of tumor, showing that PAX5 and PAX6 gene activity relates to the invasion, appearance, and formation of BLCA and may play a role in its development.

We observed that PAX family gene expression was closely related to the infiltration of immune cells in tumors, implying that PAX levels might influence tumor immune system response. To further explore the family genome variation effect on immune infiltration, we have different CNVs BLCA PAX genes or mutations between groups common immune cells abundance variance analysis. It was concluded that PAX5 and PAX6 genes were significantly positively correlated with immune cell content, and PAX3 and PAX7 high mutations could increase CD8 + T cell content. It can be speculated that PAX5 and PAX6 gene high expression, PAX3 and PAX7 high mutation can promote tumor invasion, thereby causing a large number of immune cell infiltration in vivo. Some scholars have shown that PAX5 is expressed in breast cancer and regulates the invasion and progression of breast cancer [74]. It has also been shown that inhibition of Pax6 gene expression can greatly minimize the tendency of retinoblastoma to grow and invade [75]. In conclusion, PAX family gene expression can affect the invasive ability of BLCA tumor cells and regulate its malignant progression process. In addition, copy number variation can affect immune cell content, which may be due to its effect on gene transcription, thereby affecting immune cell infiltration, or it may be that copy number variation directly affects protein modification or expression, thereby affecting immune cell infiltration. Some investigators believe that PAX5 mutations may reduce the quantity of effector T cells that reside in the tumor's microenvironment, resulting in cancer tolerance to the immune system [76]. In summary, it is reasonable to believe that the proliferation of PAX family genes may affect the chemotaxis of immune cells via some mechanism, contribute to the control of immune invasion, and thus affect the immunotherapy effect of patients, but the specific mechanism remains unknown. The primary limitation of our research was that the data utilized came from open databases, and the particular process and method still need to be validated in vivo or in vitro. Furthermore, there is complexity because we analyzed data from multiple omics of the PAX family in pan-cancer.

The PAX gene family has been shown in studies to be abnormally expressed in a range of tumors, and it is significantly related to patient prognosis and existence, as well as tumor invasion and progression. Furthermore, this study reveals the potential mechanism of PAX in the tumor microenvironment, immune checkpoints, drug sensitivity, and other aspects, indicating that PAX can be a novel possible biomarker in many types of cancers, as well as a new target for cancer treatment and that it is critical to develop new comprehensive treatment plans.

Funding

This study received support from various sources, including Tianjin Municipal Health Industry Key Project (grant no. TJWJ2022XK014), the Scientific Research Project of Tianjin Municipal Education Commission (grant no. 2022ZD069), National Key R&D Program of China (grant no. 2023YFC2507003), Tianjin Institute of Urology Talent Funding Program (grant no. MYSRC202310), and Clinical Medicine Research Project of the Second Hospital of Tianjin Medical University (grant no. 2023LC03).

Author contributions

The study design was conceptualized by HH, KW, and XZ. The manuscript was written by YQ, CS, and QX. HY, and XZ were responsible for database screening and data collection. Experiments were conducted by QX and YQ. Bioinformatics analysis was carried out by CS. Manuscript revisions were done by HH, KW, CH, and MT. Critical comments were provided by KW, HH, ZlZ, and LW. All authors read and approved the final version.

Conflicts of Interest

The authors wish to state that they do not have any competing interests.

Acknowledgments

We extend our sincere appreciation to all study participants.

Ethics approval and consent to participate

This study performed in strict adherence to the Declaration of Helsinki and was granted approval by the Ethics Committee of The Second Affiliated Hospital of Tianjin Medical University (approval no. KY2021K003). Written consent was obtained from all participants before study initiation.

Data Availability Statement

The study's original contributions are incorporated within the article/Supplementary Material. If anyone have additional queries, please feel free to reach out to the corresponding authors.

T Viktorová, Babjuk M , J Dusková, et al. Expression of PAX2 and PAX8 genes in conventional type of renal carcinoma and their role in the tumor prognosis[J]. Casopís lékar̆ů c̆eských, 2005, 144 Suppl 2:30-33.
Raffone A , Travaglino A , Saccone G , et al. PAX2 in endometrial carcinogenesis and in differential diagnosis of endometrial hyperplasia: A systematic review and meta-analysis of diagnostic accuracy[J]. Acta obstetricia et gynecologica Scandinavica, 2019, 98(3):287-299.
Sosa-Pineda B . The gene Pax4 is an essential regulator of pancreatic beta-cell development.[J]. Molecules & Cells, 2004, 18(3):289-294.
Xin, Liu, Guang, et al. The role of the Pax1/9 gene in the early development of amphioxus pharyngeal gill slits[J]. Journal of Experimental Zoology Part B Molecular & Developmental Evolution, 2015.
Xiong Z , Ren S , Chen H , et al. PAX9 regulates squamous cell differentiation and carcinogenesis in the oro-oesophageal epithelium[J]. The Journal of Pathology, 2017.
Benzina S, Harquail J, Guerrette R, O'Brien P, Jean S, Crapoulet N, Robichaud GA. Breast Cancer Malignant Processes are Regulated by Pax-5 Through the Disruption of FAK Signaling Pathways. J Cancer. 2016 Oct 22;7(14):2035-2044.
Tomczak K, Czerwińska P, Wiznerowicz M. The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge. Contemp Oncol (Pozn). 2015;19(1A):A68-77.
Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, Sun Y, Jacobsen A, Sinha R, Larsson E, Cerami E, Sander C, Schultz N. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013 Apr 2;6(269):pl1.
Palmeri M, Mehnert J, Silk AW, Jabbour SK, Ganesan S, Popli P, Riedlinger G, Stephenson R, de Meritens AB, Leiser A, Mayer T, Chan N, Spencer K, Girda E, Malhotra J, Chan T, Subbiah V, Groisberg R. Real-world application of tumor mutational burden-high (TMB-high) and microsatellite instability (MSI) confirms their utility as immunotherapy biomarkers. ESMO Open. 2022 Feb;7(1):100336.
Li T, Fan J, Wang B, Traugh N, Chen Q, Liu JS, Li B, Liu XS. TIMER: A Web Server for Comprehensive Analysis of Tumor-Infiltrating Immune Cells. Cancer Res. 2017 Nov 1;77(21):e108-e110.
Chang C. Research on human protein atlas image classification method based on ResNet deep network [J]. Journal of Medical Informatics,2019,40(07):45-49.
Guan Haoran, Li Mengqi, Han Yinglun. Pairing box gene family - an important transcription factor during development [J]. Chinese Journal of Biochemistry and Molecular Biology, 2021,37 (04): 449-457.
Zhang Weiwei, Yan Wenji, Dai Guanghai. PAX gene and tumors [J]. Armed Police Medical Journal, 2020,31 (01): 71-74.
Li Li, Yang Yang, Xue Lei. The regulatory role of Pax gene family in the development of Drosophila [J]. Genetics, 2010,32 (02): 115-121
Geeleher P., Cox N., Huang R.S. pRRophetic: An R package for prediction of clinical chemotherapeutic response from tumor gene expression levels. PLoS ONE. 2014;9:e107468.
Lang D., Powell S.K., Plummer R.S., Young K.P., Ruggeri B.A. PAX genes: Roles in development, pathophysiology, and cancer. Biochem. Pharmacol. 2007;73:1–14.
Eccles M.R., Wallis L.J., Fidler A.E., Spurr N.K., Goodfellow P.J., Reeve A.E. Expression of the PAX2 gene in human fetal kidney and Wilms’ tumor. Cell Growth Differ. 1992;3:279–289.
Kaddour H., Coppola E., Di Nardo A.A., Le Poupon C., Mailly P., Wizenmann A., Volovitch M., Prochiantz A., Pierani A. Extracellular Pax6 Regulates Tangential Cajal-Retzius Cell Migration in the Developing Mouse Neocortex. Cereb. Cortex. 2020;30:465–475.
Harshman L.A., Brophy P.D. PAX2 in human kidney malformations and disease. Pediatr. Nephrol. 2012;27:1265–1275.
Zhu Qingyi. The role and significance of PAX-2 gene in renal tumors and kidney development [J]. Huaihai Pharmaceutical, 2002 (05): 437-439
Stuart ET. PAX mediated transcriptional inhibition leads to loss of p53 function [J]. Foreign Medicine (Physiology, Pathology, and Clinical Division), 1996 (03): 72.
DNA methylation of PAX1 as a biomarker for oral squamous cell carcinoma. Yung-Kai Huang;;Bou-Yu Peng;;Chia-Yo Wu;;Chien-Tien Su;;Hui-Chen Wang;;Hung-Cheng Lai.Clinical Oral Investigations,2014.
Bleu M., Gaulis S., Lopes R., Sprouffske K., Apfel V., Holwerda S., Pregnolato M., Yildiz U., Cordo’ V., Dost A.F.M., et al. PAX8 activates metabolic genes via enhancer elements in Renal Cell Carcinoma. Nat. Commun. 2019;10:3739.
Li L, Li CG, Almomani SN, Hossain SM, Eccles MR. Co-Expression of Multiple PAX Genes in Renal Cell Carcinoma (RCC) and Correlation of High PAX Expression with Favorable Clinical Outcome in RCC Patients. Int J Mol Sci. 2023 Jul 14;24(14):11432.
Hu He K., Lorenzo P.I., Brun T., Jimenez Moreno C.M., Aeberhard D., Ortega J.V., Cornu M., Thorel F., Gjinovci A., Thorens B., et al. In Vivo Conditional Pax4 Overexpression in Mature Islet [1]-Cells Prevents Stress-Induced Hyperglycemia in Mice. Diabetes. 2011;60:1705–1715.
Muratovska A., Zhou C., He S., Goodyer P., Eccles M.R. Paired-Box genes are frequently expressed in cancer and often required for cancer cell survival. Oncogene. 2003;22:7989–7997.
Hung N, Chen YJ, Taha A, Olivecrona M, Boet R, Wiles A, Warr T, Shaw A, Eiholzer R, Baguley BC, Eccles MR, Braithwaite AW, Macfarlane M, Royds JA, Slatter T. Increased paired box transcription factor 8 has a survival function in glioma. BMC Cancer. 2014 Mar 6;14:159.
Chaves-Moreira D, Morin PJ, Drapkin R. Unraveling the Mysteries of PAX8 in Reproductive Tract Cancers. Cancer Res. 2021 Feb 15;81(4):806-810.
Corona RI, Seo JH, Lin X, Hazelett DJ, Reddy J, Fonseca MAS, et al. Non-coding somatic mutations converge on the PAX8 pathway in ovarian cancer. Nat Commun 2020;11:2020.
Chen X, Li Y, Paiboonrungruang C, Li Y, Peters H, Kist R, Xiong Z. PAX9 in Cancer Development. Int J Mol Sci. 2022 May 17;23(10):5589.
Association of genomic variants at PAX8 and PBX2 with cervical cancer risk. Ramachandran D, Wang Y, Schürmann P, et al. Int J Cancer. 2021;149:893–900.
A comprehensive analysis of PAX8 expression in human epithelial tumors. Laury AR, Perets R, Piao H, et al. Am J Surg Pathol. 2011;35:816–826
Expression of PAX8 in normal and neoplastic tissues: a comprehensive immunohistochemical study. Tacha D, Zhou D, Cheng L. Appl Immunohistochem Mol Morphol. 2011;19:293–299.
PAX8 expression in non-neoplastic tissues, primary tumors, and metastatic tumors: a comprehensive immunohistochemical study. Ozcan A, Shen SS, Hamilton C, Anjana K, Coffey D, Krishnan B, Truong LD. Mod Pathol. 2011;24:751–764.
PAX8 expression in uterine adenocarcinomas and mesonephric proliferations. Yemelyanova A, Gown AM, Wu LS, Holmes BJ, Ronnett BM, Vang R. Int J Gynecol Pathol. 2014;33:492–499.
PAX8 and PAX2 expression in endocervical adenocarcinoma in situ and high-grade squamous dysplasia. Shukla A, Thomas D, Roh MH. Int J Gynecol Pathol. 2013;32:116–121.
A case of poorly differentiated adenocarcinoma of the uterine cervix treated with paclitaxel and carboplatin after positive PAX8 immunostaining (article in Japanese) Maenaka T, Iwata N, Egawa-Takata T, et al. https://pubmed.ncbi.nlm.nih.gov/24196083/ Gan To Kagaku Ryoho. 2013;40:1413–1417.
Pax8: a marker for carcinoma of Müllerian origin in serous effusions. Tong GX, Devaraj K, Hamele-Bena D, et al. Diagn Cytopathol. 2011;39:567–574.
NY-BR-1 and PAX8 immunoreactivity in breast, gynecologic tract, and other CK7+ carcinomas: potential use for determining site of origin. Woodard AH, Yu J, Dabbs DJ, et al. Am J Clin Pathol. 2011;136:428–435.
The utility of PAX8 and IMP3 immunohistochemical stains in the differential diagnosis of benign, premalignant, and malignant endocervical glandular lesions. Danialan R, Assaad M, Burghardt J, Newcomb P, Cartun RW, Mandavilli S. Gynecol Oncol. 2013;130:383–388.
Differential patterns of PAX8, p16, and ER immunostains in mesonephric lesions and adenocarcinomas of the cervix. Goyal A, Yang B. Int J Gynecol Pathol. 2014;33:613–619.
Assessment of the utility of PAX8 immunohistochemical stain in diagnosing endocervical glandular lesions. Liang L, Zheng W, Liu J, Liang SX. Arch Pathol Lab Med. 2016;140:148–152.
Immunohistochemistry for the novel markers glypican 3, PAX8, and p40 (ΔNp63) in squamous cell and urothelial carcinoma. Gailey MP, Bellizzi AM. Am J Clin Pathol. 2013;140:872–880.
Comprehensive analysis of PAX8 expression in epithelial malignancies of the uterine cervix. Wong S, Hong W, Hui P, Buza N. Int J Gynecol Pathol. 2017;36:101–106.
Mullighan CG, Goorha S, Radtke I, Miller CB, Coustan-Smith E, Dalton JD, et al.. Genome-wide analysis of genetic alterations in acute lymphoblastic leukaemia. Nature (2007) 446(7137):758–64.
Mullighan CG, Su X, Zhang J, Radtke I, Phillips LA, Miller CB, et al.. Deletion of IKZF1 and prognosis in acute lymphoblastic leukemia. N Engl J Med (2009) 360(5):470–80.
Familiades J, Bousquet M, Lafage-Pochitaloff M, Bene MC, Beldjord K, De Vos J, et al.. PAX5 mutations occur frequently in adult b-cell progenitor acute lymphoblastic leukemia and PAX5 haploinsufficiency is associated with BCR-ABL1 and TCF3-PBX1 fusion genes: a GRAALL study. Leukemia (2009) 23(11):1989–98.
Gu Z, Churchman ML, Roberts KG, Moore I, Zhou X, Nakitandwe J, et al.. PAX5-driven subtypes of b-progenitor acute lymphoblastic leukemia. Nat Genet (2019) 51(2):296–307.
Gnarra JR, Dressler GR. Expression of Pax-2 in human renal cell carcinoma and growth inhibition by antisense oligonucleotides. Cancer Res. 1995 Sep 15;55(18):4092-8.
Zhang HS, Yan XB, Li XB, Fan L, Zhang YF, Wu GH, Li M, Fang J. PAX2 Protein induces expression of cyclin D1 through activating AP-1 protein and promotes proliferation of colon cancer Cells. J Biol Chem. 2012;287:44164–44172.
Zhang LP, Shi XY, Zhao CY, Liu YZ, Cheng P. RNA interference of pax2 inhibits growth of transplanted human endometrial cancer cells in nude mice. Chin J Cancer. 2011;30:400–406.
Hueber PA, Iglesias D, Chu LL, Eccles M, Goodyer P. In vivo validation of PAX2 as a target for renal cancer therapy. Cancer Lett. 2008;265:148–155.
Schebesta A., McManus S., Salvagiotto G., Delogu A., Busslinger G. A., Busslinger M. Transcription factor Pax5 activates the chromatin of key genes involved in B cell signaling, adhesion, migration, and immune function. Immunity. 2007;27(1):49–63.
Quail DF, Joyce JA. Microenvironmental regulation of tumor progression and metastasis[J]. Nat Med,2013,19(11):1423-1437.
Bu L, Baba H, Yoshida Net al. Biological heterogeneity and versatility of cancer-associated fibroblasts in the tumor microenvironment[J]. Oncogene,2019,38(25):4887-4901.
Qian BZ, Pollard JW. Macrophage diversity enhances tumor progression and metastasis[J]. Cell,2010,141(1):39-51.
An W, Pan SW, Duan SJ, et al. Analysis of HLA-DRA expression and its clinical significance in gastric cancer tissues based on GEO matrix score and immunoscore [J]. Journal of China Medical University,2021,50(12):1063-1068.
Xia L, Huang Q, Nie D, Shi J, Gong M, Wu B. PAX3 is overexpressed in human glioblastomas and critically regulates the tumorigenicity of glioma cells. Brain Res. 2013;1521:68–78.
Guo L, Yang Y. Expression of PAX9 in ovarian cancer and its effect on proliferation, migration and invasion of ovarian cancer cells [J]. Chinese Journal of Sex Science, 21,30(12):87-91.
Li X, Liu R, Su X, Pan Y, Han X, Shao C, Shi Y. Harnessing tumor-associated macrophages as aids for cancer immunotherapy. Mol Cancer. 2019 Dec 5;18(1):177.
Kong Y, Ma Y, Zhao X, Pan J, Xu Z, Zhang L. Optimizing the Treatment Schedule of Radiotherapy Combined With Anti-PD-1/PD-L1 Immunotherapy in Metastatic Cancers. Front Oncol. 2021 Mar 30;11:638873.
Gao Z, Song CG, Zhang Y, et al. Prognostic analysis of TMB expression level in advanced NSCLC patients with EGFR mutation [J]. Chin J Lab Diagnostics, 2012,26(08):1135-1139.
Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer[J].Science,2015,348(6230):124.
Goodman AM, Kato S, Bazhenova L, et al. Tumor Mutational Burden as an Independent Predictor of Response to Immunotherapy in Diverse Cancers[J].Mol Cancer Ther,2017,16:2598.
Zheng R, Nie MM. Research progress of microsatellite instability in the treatment of gastric cancer [J]. Chin J Clin Med, 2021,29(05):864-869.
Baretti M, LE D T. DNA mismatch repair in cancer[J]. Pharmacol Ther, 2018, 189:45-62.
Jardim DL, Goodman A, de Melo Gagliato D, Kurzrock R. The Challenges of Tumor Mutational Burden as an Immunotherapy Biomarker. Cancer Cell. 2021 Feb 8;39(2):154-173.
Gu ZW. The role of PAX9 in platinum resistance of epithelial ovarian cancer and its clinical significance [D]. Shanghai jiaotong university, 2020.
Zhang T. The role and mechanism of PAX6 in the metastasis and cisplatin resistance of non-small cell lung cancer [D]. Chengdu Medical College,2019.
Cheng Q. Temozolomide inhibits glioma cell proliferation by regulating miR-223/PAX6 [D]. Central South University,2015.
Zhu Z, Shen Z, Xu C. Targeted therapy for advanced urothelial cancer of the bladder: where do we stand? Anticancer Agents Med Chem. 2012 Nov;12(9):1081-7.
Su T T. Study on the correlation between PAX-5 and FHIT expression and clinical and pathological types of neuroendocrine tumors [D]. Guangxi Medical University,2019.
Sun F, Ni SH, Ma YJ, et al. Expression level of PAX8 protein in brain glioma patients and its clinicopathological characteristics [J]. Medicine and Philosophy (B),2018,39(03):46-50.
Benzina S, Harquail J, Guerrette R, O'Brien P, Jean S, Crapoulet N, Robichaud GA. Breast Cancer Malignant Processes are Regulated by Pax-5 Through the Disruption of FAK Signaling Pathways. J Cancer. 2016 Oct 22;7(14):2035-2044.
Wang H, Wang F, Chu RX et al. The effect of Pax6 gene inhibition on the proliferation and invasion of retinoblastoma cells and its mechanism [J]. Journal of Applied Cancer,2020,35(12):1961-1964.
WANG L. Pax5 gene mutation induces low T cell infiltration in tumor microenvironment through low expression of chemokines [D]. Huazhong university of science and technology, 2022.

Additional Data 1 is not available with this version.

The OS, PFS, DFS and DSS Kaplan‐Meier curves were plotted across 33 cancer types according to the PAXs expression level.

No competing interests reported.

FigureS1.tif
Supplementary Figure S1. (A-I) Expression boxplot of each PAX genes across 33 TCGA cancer tissues, with ranking them from high to low.
FigureS2.tif
Supplementary Figure S2. (A-I) The PAX regulation of genes in cancer versus corresponding normal samples in 18 cancer types with normal sample sizes greater than 50.
FigureS3.tif
Supplementary Figure S3. (A-I) Alternations frequency of each PAX family member in pan-cancer and multiple types of tumors had been examined from the ICGC/TCGA PanCancer Atlas (2683 samples).
FigureS4.tif
Supplementary Figure S4. (A-I) Impact of genetic variations of each PAX family member on overall survival outcomes were explored based on the ICGC/TCGA PanCancer Atlas (2683 samples). of tumor patients.
FigureS5.tif
Supplementary Figure S5. (A-I) The associations of pan-cancer PAX mRNA expression with SNV.
FigureS6.tif
Supplementary Figure S6. (A-I) Correlation analysis of the expression of PAX family (each member) in pancarcinoma with response status after initial treatment of patients.
FigureS7.tif
Supplementary Figure S7. (A-I) TCGA database and Spearman correlation tests performed to compare the gene activity of each PAX member to the activation of 47 common immune regulatory genes in 33 kinds of cancer.
FigureS8.tif
Supplementary Figure S8. The association of each PAX member (A-I) and TMB across 33 tumor types.
FigureS9.tif
Supplementary Figure S9. The association between the expression of each PAX member (A-I) and MSI from 33 tumor types.
FigureS10.tif
Supplementary Figure S10. (A-G) Utilizing the TRUCE-01 facts, alterations in PAX1, PAX2, PAX4, PAX5, PAX7, PAX8, and PAX9 activity preceding and following immune therapy have been contrasted in 15 cases using a pair-wise Wilcox test.
FigureS11.tif
Supplementary Figure S11. (A)Correlation analysis of each PAX gene in BLCA tumor and previous immunophenotyping. (B-E) Correlation analyses of each PAX gene in BLCA and tumor grade, tumor subtypes, stage, and T_stage.
FigureS12.tif
Supplementary Figure S12. GSEA analysis for Hallmark gene sets identified the expression of higher and lower groups of PAX1 (A), PAX3 (B), PAX4 (C), PAX5 (D), PAX6 (E), PAX8 (F) and PAX9 (G).
SupplementaryTableS1S2S3S4.xlsx
SupplementaryTable S1. K-M analysis findings for OS (A), DFS (B), DSS (C), PFS (D)were aggregated across 33 cancer types according to the expression levels of PAXs. SupplementaryTable S2. Drug sensitivity analysis of PAX family genes based on the CellMiner database. Supplementary Table S3. The GSEA was performed for each PAX family member using the KEGG gene sets based on TCGA_BLCA datasets. Supplementary Table S4. The GSEA was performed for each PAX family member using the Hallmark gene sets based on TCGA_BLCA datasets.

Download PDF

Version 1

posted

You are reading this latest preprint version

Prognostic significance and immunologic features of the paired-box (PAXs) family: a pan-cancer multi- omics analysis

Status:

Version 1

Abstract

Objective

Methods

Results

Conclusion

Figures

Introduction

Materials and Methods

2.1 Collection of Data

2.2 Evaluation of PAX family expression in pan cancer tissues

2.3 Survival Analysis

2.4 PAX mutation frequency, patient outcome consequences, and PAX association with CNV or SNV

2.5 Association amongst PAX expression, immunological subgroup, and cancer microenvironment

2.7 Tumor Immune Estimation Resource

2.8 Pan-cancer sensitivity to drugs

2.9 Evaluation of Gene Set Enrichment in Bladder tumor

2.10 RNA Extraction and qRT-PCR; human protein atlas (HPA) analysis

2.11 Statistical Analysis

Results

3.1 Identification of PAX family manifestation in pan cancer

3.2 KM survival curve analysis and unicox regression study

3.3 Genome variation of PAX family in pan-cancer

3.4 Correlation between clinicopathologic traits and PAX family members

3.5 Association analyses of the PAXs expression with immunization typing, tumor stemness scores, and tumor microenvironment in pan-cancer

3.6 Correlation analysis of PAXs with immunological checkpoints expression, tumor mutational burden (TMB), and microsatellite instability (MSI)

3.7 Sensitivity to drugs screening

3.8 The association between the proliferation of PAXs and effectiveness of clinical immunotherapy

3.9 Correlation analysis of PAX genes with immunological subsets and biological aspects in BLCA

3.10 Correlation studies of PAX genes expression with stemcell index, and immunological features in bladder cancer

3.11 Gene set enrichment analysis of bladder cancer

3.12 Association analysis between mutation and gene copy number of PAX family and the infiltration abundance of immune cells

3.13 PAX8 gene expression in cancer cells and near healthy tissues

Discussion

Conclusion

Declarations

References

Additional Data

Additional Declarations

Supplementary Files

Status:

Version 1