Table 2: Univariate survival analyses

From: IGF-IR: a new prognostic biomarker for human glioblastoma

 Median cancer-specific survival (months)P-value
Age (years)* (n=218) 0.0004
Multifocality (n=207) NS
 No7.9 
 Yes10.6 
Date of surgery* (n=218) NS
Surgical resection (n=203) 0.002
 Partial8.4 
 Complete13.1 
Adjuvant therapy (n=168) 10−5
 Radiotherapy10.6 
 Radiotherapy+temozolomide14.9 
IGFI LI* (n=212) NS
IGFII LI* (n=204) NS
IGF-IR LI* (n=218) 0.046
IGF-IIR LI* (n=213) NS
Binary scores
IGFI (n=212) NS
 Positive8.7 
 Negative10.5 
IGFII (n=204) NS
 Positive8.8 
 Negative9.3 
IGF-IR (n=218) 0.020
 Positive9.0 
 Negative11.6 
IGF-IIR (n=213) NS
 Positive10.3 
 Negative9.2 
  1. Abbreviations: IGF=insulin-like growth factor; LI=labelling index; NS=not significant. Continuous variables were analysed using the univariate Cox regression (see asterisk (*)). The other binary variables were analysed using the log-rank test. For these latter variables, each category is characterised by the median cancer-specific survival time (in months). For IGFI, IGFII, IGF-IR and IGF-IIR, the cases labelled as positive correspond to LI1% and those labelled as negative to LI<1%. The n values indicate the total number of cases taken into account in the univariate analyses (excluding the missing values and certain non-standard clinical categories that are detailed in Table 1).