Skip to main content
. 2018 Feb 15;4(4):483–492. doi: 10.1001/jamaoncol.2017.4996

Table 2. Univariate and Multivariate Analyses of Prognostic Factorsa.

Prognostic Factor No. Survival
Univariate Multivariable
HR (95% CI) P Value HR (95% CI) P Value
Treatment
NACT alone 167 1 [Reference] 1 [Reference]
NACT plus RHT 162 0.73 (0.54-0.98) .04 0.70 (0.52-0.95) .02
Age, y
18-40 88 1 [Reference] ND
41-70 241 0.97 (0.69-1.37) .87 ND
Sex
Men 182 1 [Reference] ND
Women 147 0.85 (0.63-1.16) .31 ND
Grade
G3 176 1 [Reference] 1 [Reference]
G2 153 0.68 (0.50-0.92) .01 0.69 (0.51-0.94) .02
Tumor size, cm
>12.0 130 1 [Reference] 1 [Reference]
5.0-12.0 199 0.63 (0.47-0.86) .003 0.62 (0.46-0.84) .002
Presentation of tumor
Recurrent 37 1 [Reference] ND
Primary 157 0.62 (0.40-0.95) .03 ND
Prior surgery 135 0.42 (0.27-0.67) <.001 ND
Site
Nonextremity 186 1 [Reference] ND
Extremity 143 0.45 (0.33-0.63) <.001 ND

Abbreviations: HR, hazard ratio; NACT neoadjuvant doxorubicin, ifosfamide, and etoposide chemotherapy; ND, no data; RHT, regional hyperthermia.

a

The analyses of subgroups (treatment, age, sex, grade, tumor size) were prespecified and stratified to tumor presentation and site. The univariate HR estimates for the stratification variables (tumor presentation and site) are given; HRs for stratification variables in multivariate analyses cannot be calculated.