3.1. Survival analysis related to NXNL1
As shown in Fig. 1, increased expression of NXNL1 is significantly correlated with good prognosis in BCs (Fig. 1A, P = 0.082). The expression of NXNL1 in ER and PR positive is significantly lower than normal tissue and vice versa, the expression of NXNL1 in normal tissue is higher than negative in ER and PR status (Fig. 1C, 1E, P < 0.001). In addition, NXNL1 expression in Breast Cancer tumor samples is obviously lower than in normal (Fig. 1C, P < 0.01). We download form TCGA show the relationship with the expression of NXNL1 and clinicopathological features of BC patients. Univariate analysis using Cox regression revealed that some factors in table 1, including Tumor stage (HR = 1.482, P-value =0.046), lymph node stage (HR = 2.239, P-value < 0.001) along with pathologic stage (HR = 2.21, P= 0.003) are significantly associated with overall survival. multivariate analysis showed in Table 1 and Fig 2, the up-regulated NXNL1 expression, N status is independent prognostic factors of favorable prognosis.
3.2. NXNL1 expression in Pan-Cancer and Breast Cancer
The underlying mechanism of NXNL1 expression in Pan-cancer requires further study, but we used box plots to visually analyze NXNL1 expression in tumor tissues and normal tissues in pan-cancers (fig 3). It shows that the expression of NXNL1 higher in tumor tissues than normal tissues in BRCA, CHOL, KIRC, KIRP, OV, PAAD, STAD, THCA, THYM. Lower expression of NXNL1 in tumor tissues in CESC, COAD, ESCA, GBM, KICH, LGG, BC, LUSC, PRAD, READ, SARC, SKCM, TGCT, UCEC and UCS.
BC cases with eligible clinical information were analyzed by R-3.6.3. As shown in Table 2, univariate analysis using logistic regression with NXNL1 expression as a categorical dependent variable revealed that expression of NXNL1 correlated significantly with the T stage add (T2&T3&T4 vs. T1, p=0.0046), Pathologic stage add (Stage II &Stage III &Stage IV vs. Stage I, p=0.003), and N stage (Positive vs. Negative, p < 0.001).
3.3. NXNL1 expression with tumor-infiltrating immune cells
Immune system affects cancer development and progression has been one of the most challenging questions over the past two decades [8]. Therefore, we tried to find whether NXNL1 expression relates to immune infiltration in pan-cancer. Just show as fig 3A, because of lacked the necessary data, UVM and MESO were excluded. The expression of NXNL1 is higher in normal tissues than tumor tissue in ACC, BRCA, CESC, COAD, ESCA, GBM, LGG, LUAD, PCPG, READ, TGCT and UCEC. On the contrary, the expression of NXNL1 in tumor tissue is higher than that in normal tissue in HNSC, KICH, KIRP, LAML, LIHC, OV, SARC, SKCM and THYM. No difference inexpression only in BLCA, DLBC, KIRC, PAAD and UCS. Among 1083 BC samples, samples with the top 1/2 and the next 1/2 NXNL1 expression were included into high expression group and low expression group, respectively. To infer the proportion of 23 immune cells in NXNL1 high expression group and low expression group. we used computational resource (CIBERSORT) established to explore the gene expression profile of downloaded samples. Finally, 555 samples of high expression group and 554 samples of low expression group inclusion analysis of screening criterion. As shown in Fig. 4A, T cells, aDC, B cells, Cytotoxic cells, DC, Epsinphils, iDC, Macrophages, Mast cells, Eosinophils, T helper cells, Neutrophils, NK CD56bright cells, NK CD56dim cells, NK cells, pDC, Tem, Th1 cells, Th17 cells, Th2 cells and Treg are main immune cells affected by NXNL1 expression. NXNL1 expression were positively correlated with Th17 cells, iDC, pDC, Eosinophils, Mast cells, NK CD56 bright cells and NK cells. In contrast, the proportion of CD8 T cells (p = 0.808), TFH cells(p = 0.063) and Tcm(p = 0.613) are apparently lower. In addition, lollipop chart (Fig. 4B) revealed that the proportions of different TIICs subpopulations were weakly to moderately correlated.
“Correlation” module of GEPIA helped us to reanalyze the link between NXNL1 expression and gene markers of different types of tumors infiltrating immune cells, including T cells, aDC, B cells, Cytotoxic cells, DC, Epsinphils, Marophils, Mast cells, Neutrophils, NK CD56 bright cells, NK CD56 dim cells, NK cells, pDC, Tem, TFH, Tgd, Th1 cells, Th17 cells, Th2 cells and Treg (Table 3). Results confirmed that NXNL1 expression is correlated with over most of the marker sets of different immune cells in BC. The gene markers effected by NXNL1 expression include T cells, aDC, B cells, Cytotoxic cells, DC, Epsinphils, Marophils, Mast cells, Neutrophils, NK CD56 bright cells, NK CD56 dim cells, NK cells, pDC, Tem, Tgd, Th1 cells, Th17 cells, Th2 cells and Treg. Correlations were evaluated using Spearman correlation coefficient. Correlation results between NXNL1 and markers of B cells, Neutrophils, Mast cells and T cells were similar to CIBERSORT. Thus, these findings suggest that NXNL1 may play an important role in regulating the abundance of B cells, Neutrophils, Mast cells and NK cells.