Page 11 - TD-1-2

Tumor Discovery BAK1 as a novel prognostic biomarker

significantly downregulated in two additional cancer types BAK1, and genes with differences in expression between
(colon adenocarcinoma [COAD] and kidney chromophobe the high and low expression groups were identified
[KICH]; all P < 0.001). We then looked at how BAK1 differed (Figure 3A). The differential genes were found to be
between HCC and non-HCC tissues. Figures 1C and D involved in neuroactive ligand-receptor interaction,
reveal that BAK1 was expressed at a greater level in HCC cytokine-cytokine receptor interaction, adenosine
tissues and that this level differs considerably from that of 3’,5’-cyclic monophosphate (cAMP) signaling pathway,
non-HCC tissues. We subsequently compared the prognosis and hematopoietic cell lineage by KEGG enrichment
of HCC patients with high and low BAK1 expression; analysis (Figure 3B). We then ran a GSEA, and the results
the results revealed that the overall survival rate of HCC in Figure 3C showed that the four functions HUMORAL
patients with high BAK1 expression was considerably lower IMMUNE RESPONSE MEDIATED BY CIRCULATING
than that of patients with low BAK1 expression (Figure 1E). IMMUNC, IMMUNOGLOBULIN COMPLEX,
According to univariate and multivariate independent IMMUNOGLOBULIN COMPLEX CIRCULATING,
prognostic analyses (Figures 1F and G), BAK1 is associated and IMMUNOGLOBULIN RECEPTOR BINDING were
with prognosis and can be a prognostic factor independent active in the high BAK1 expression group, and STEROID
of other factors. In univariate Cox regression analysis, HYDROXYLASE ACTIVITY is active in the high and low
the hazard ratio (HR) and 95% confidence interval (CI) expression group of BAK1. The five pathways FATTY ACID
were 1.421 and 1.152 – 1.753, respectively (P = 0.001); in METABOLISM, GLYCINE SERINE AND THREONINE
multivariate Cox regression analysis, the HR was 1.262, and METABOLISM, PEROXISOME, PRIMARY BILE ACID
the 95% CI was 1.013 – 1.572 (P = 0.038). BIOSYNTHESIS, and RETINOL METABOLISM were
all active in the low BAK1 expression group, as shown in
3.2. Relationship between BAK1 expression and Figure 3D. We constructed a nomogram using risk classes
clinical features and clinical data to predict the 1-, 3-, and 5-year survival
A clinical correlation study was performed to determine if in LIHC patients, as shown in Figure 3E. Assuming a
BAK1 expression differed between clinical groups. BAK1 patient’s composite score is 394, the 1-year survival rate
exhibited significant differences in gender, tumor grade, is 0.941, the 3-year survival rate is 0.882, and the 5-year
tumor stage, and T stage. Figure 2A shows that BAK1 survival rate is 0.839. Correlation plots revealed that the
expression does not differ significantly with age. Figure 2B observed and expected rates of survival in LIHC patients
demonstrates that BAK1 expression is significantly higher in at 1, 3, and 5 years were in perfect agreement (Figure 3F).
female HCC patients than in males. Figure 2C depicts BAK1
expression levels in various tumor stages, with substantial 3.4. Expression of BAK1 and immunity and drug
statistical differences between Stage 1 and Stage 2, as well sensitivity
as Stage 1 and Stage 3. Figure 2D shows a statistically Figure 4A shows a differential study of immune cells, which
significant difference in BAK1 expression among T1, T2, revealed a statistically significant difference in dendritic
and T3, but no such difference was seen in the other T stages. cell activation between high and low BAK1 expression
The clinical correlation heatmap in Figure 2E shows that groups, suggesting that the activation is favorably regulated
the three clinical parameters tumor grade, tumor stage, and by BAK1. Figure 4B, we then examined the relationship
T stage have substantial statistical differences between high between BAK1 and immunological checkpoint-related
and low BAK1 expression groups. Coexpression analysis genes. BAK1 positively regulates CD276, CD86, CD80,
revealed that BAK1 is positively regulated by SMARCD1, TNFRSF8, TNFSF15, LGALS9, TNFRSF18, PDCD1,
MFSD10, RCC2, CDK16, PKM, and MACROH2A1 but VTCN1, and HAVCR2, while IDO2 and BAK1 have a
negatively regulated by GLYATL1, ALDH2, CDO1, DCXR, skewed relationship. Unfortunately, no significant statistical
and SLC27A5 (Figure 2F). The diagnostic value of BAK1 difference was observed when receiving CTLA-4 and PD-1
in LIHC was assessed by drawing a receiver operating treatment regardless of whether BAK1 expression was high
characteristic (ROC) curve (Figure 2G). We discovered that or low (Figure 4C and D). Figure 4E and F illustrate the
the area under the ROC curve was 0.694, 0.582, and 0.611 immunohistochemistry (IHC) status of BAK1 in normal
at 1, 3, and 5 years, indicating that this gene may be a good and cancerous liver tissue, respectively. Representative
prospective LIHC diagnostic marker. IHC photos reveal that BAK1 protein is more abundant
in tumors than in non-tumor tissues. We then identified
3.3. Analysis of differential genes and construction drugs that showed substantial changes in their sensitivity in
of nomogram
patients with high and low BAK1 expression. Ninety drugs
All samples were classified into high and low expression were discovered. Figure 4G to Figure 4N, Fluorouracil,
groups according to the expression of the target gene bosutinib, bleomycin, cyclopamine, and other drugs were

Volume 1 Issue 2 (2022) 5 https://doi.org/10.36922/td.v1i2.221