Page 101 - GPD-4-2

Gene & Protein in Disease Significance of MXRA7 in bladder cancer

A C

B

D E

Figure 4. LASSO Cox model identifying key prognostic genes and establishing risk score for BLCA with strong predictive power. (A) Regression coefficient
change curve versus -log (λ) based on LASSO-Cox analysis. (B) Partial likelihood deviance changes in LASSO regression with log (λ). (C) Relationship
between risk score (top), prognostic immune gene expression (bottom), and a scatter plot of prognostic state versus time (middle). (D) Differential
expression profile in gene forest map. (E) ROC curve analysis for the downstream genes is shown in Figure 4D.
Abbreviations: BLCA: Bladder cancer; LASSO: Least absolute shrinkage and selection operator; ROC: Receiver operating characteristic curve.

0.047979*PDLIM4 + 0.013591*UPK2). In addition, ROC variables on time, with results shown in Table S4 in the
curve analysis of the immune prognostic model yielded Supplementary. Seven factors were identified as significant:
area under curve (AUC) values of 0.69, 0.72, and 0.68 for “Age”, “MXRA7”, “MXRA7 expression level”, “Risk score”,
1, 3, and 5 years, respectively, indicating relatively strong “Tumor grade”, “Cancer status”, and “Clinical_N”. Then,
predictive power (Figure 4E). a Cox multifactor analysis was reconstructed for these
significant factors (Table 1). The model rejected the
3.5. Building a nomogram based on the MXRA7 original hypothesis (chi = 103.87, p <0.05), indicating
prognostic model statistical significance in model construction. Notably, the
The Cox multifactor analysis using SPSSAU selected 12 “Age”, “MXRA7”, “Risk score”, “MXRA7 expression level-
factors for screening, including “Age”, “BMI”, “MXRA7”, Low”, “Cancer Status-With tumor”, “Tumor grade-Low
“Risk score”, “Sex”, “MXRA7 expression level”, “Cancer grade”, and “Clinical_N-N2” had a significant positive
status”, “Stage”, “Tumor grade”, “Clinical_T”, “Clinical_N”, effect on survival risk. However, “Clinical_N-N1” and
and “Clinical_M” as independent variables, with the “Clinical_N-NX” did not significantly impact survival.
existence state is represented by the patient status (alive or The nomogram combined these seven significant
dead). The analysis examined the influence of independent variables to predict whether BLCA patients could survive

Volume 4 Issue 2 (2025) 6 doi: 10.36922/gpd.6256