Skip to main content

Table 5 Univariate and multivariate Cox regression model associated with the SLC32A1/MSR1 gene signature

From: A machine learning analysis of a “normal-like” IDH-WT diffuse glioma transcriptomic subgroup associated with prolonged survival reveals novel immune and neurotransmitter-related actionable targets

Characteristics Univariate Cox regression Multivariate Cox regression
Beta HRa (95% CIb for HR) p value Beta HRa (95% CIb for HR) p value
Age 0.034 1.035 (1.021–1.048) 6.75e−07 0.02 1.020 (1.004–1.037) 1.42e−02
Gender
 Female Reference      
 Male 0.325 1.384 (0.991–1.932) 5.65e−02 0.073 1.076 (0.756–1.531) 6.83e−01
Grade
 G2 Reference      
 G3 0.883 2.419 (0.850–6.881) 9.77e−02 0.547 1.728 (0.576–5.187) 3.29e−01
 G4 1.75 5.752 (2.113–15.660) 6.18e−04 1.186 3.273 (1.045–10.250) 4.18e−02
EGFR
 Amplification Reference      
 WT − 0.888 0.412 (0.223–0.761) 4.64e−03 − 0.337 0.714 (0.328–1.555) 3.96e−01
CDKN2A
 CDKN2A deletion Reference      
 CDKN2A WT − 0.702 0.496 (0.332–0.741) 6.13e−04 − 1.691 0.184 (0.023–1.472) 1.11e−01
CDKN2B
 CDKN2B deletion Reference      
 CDKN2B WT − 0.653 0.520 (0.350–0.774) 1.25e−03 1.276 3.584 (0.465–27.617) 2.21e−01
`Chr 7 gain/chr 10 loss
 `Chr 7 gain/chr 10 loss combined CNA Reference      
 `Chr 7 gain/chr 10 loss no combined CNA − 0.381 0.683 (0.482–0.968) 3.20e−02 0.022 1.023 (0.682–1.533) 9.14e−01
SLC32A1/MSR1 signature
 SLC32A1_low–MSR1_low − 0.172 0.842 (0.385–1.842) 6.67e−01 0.304 1.355 (0.568–3.233) 4.94e−01
 SLC32A1_medium–MSR1_low − 0.91 0.403 (0.193–0.841) 1.55e−02 − 0.801 0.449 (0.203–0.991) 4.74e−02
 SLC32A1_high–MSR1_low − 0.766 0.465 (0.283–0.762) 2.41e−03 0 1.000 (0.554–1.808) 9.99e−01
 SLC32A1_low–MSR1_high 0.136 1.145 (0.728–1.800) 5.57e−01 − 0.109 0.897 (0.552–1.456) 6.60e−01
 SLC32A1_medium–MSR1_high − 0.149 0.862 (0.566–1.313) 4.89e−01 − 0.084 0.920 (0.584–1.448) 7.18e−01
  1. aHazard ratio
  2. bConfidence interval