Table 1.
Patient demographics, tumor properties and outcome of patients included for model building.
Institution | TUM training | UW validation | p-Valuesa | |
---|---|---|---|---|
Total patients | 122 p | 103 p | ||
Location | Extremity or trunk | 115/122 p | 99/103 p | 1 |
(94%) | (96%) | |||
Abdomen/retroperitoneal | 7/122 p | 4/103 p | ||
(6%) | (4%) | |||
Age | m 57 (sd: 16.7) | m 52.7 (sd:15.5) | .2 | |
Gender | Female | 57/122 p | 47/103 p | 1 |
(47%) | (46%) | |||
Male | 65/122 p | 56/103 p | ||
(53%) | (54%) | |||
T-stageb | 1 | 18/122 p | 17/103 p | 1 |
(15%) | (17%) | |||
2 | 104/122 p | 86/103 p | ||
(85%) | (83%) | |||
a | 11/122 p | 1/103 p | 1 | |
(9%) | (1%) | |||
b | 111/122 p | 102/103 p | ||
(91%) | (99%) | |||
M-stageb | 0 | 117/122 p | 102/103 p | 1 |
(96%) | (99%) | |||
1 | 5/122 p | 1/103 p | ||
(4%) | (1%) | |||
N-stageb | 0 | 120/122 p | 103/103 p (100%) | 1 |
(98.0%) | ||||
1 | 2/122 p | 0/103 p | ||
(2%) | (0%) | |||
Gradingc | 1 | 48/122 p | 20/103 p (19.5%) | .05 |
(4.8%) | ||||
2 | 28/122 p (32.1%) | 31/103 p (32.2%) | ||
3 | 46/122 p (46.4%) | 52/103 p (48.3%) | ||
Tumor volume | 197.6 cc | 164.5 cc | ||
(sd: 391.4) | (sd: 428.8) | |||
AJCC-stage | IA | 8/122 p | 4/103 p | p < .001 |
(7%) | (4%) | |||
IB | 40/122 p | 16/103 p | ||
(33%) | (16%) | |||
IIA | 8/122 p | 13/103 p | ||
(7%) | (13%) | |||
IIB | 5/122 p | 23/103 p | ||
(4%) | (22%) | |||
III | 57/122 p | 47/103 p | ||
(47%) | (46%) | |||
IV | 4/122 p | 0/103 p | ||
(3%) | (0%) | |||
Median OS | 35.7 mo | 44.7 mo | .2 |
Abbreviations: cc: cubic centimeters, m: median, mo: months, p: patients, TUM: Technical University of Munich, UW: University of Washington, sd: standard deviation.
Categorial variables: Fisher’ exact test (2 cohorts)) and continuous/Rank variables: Wilcoxon rank sum test (2 cohorts), log-rank test for survival, with bonferroni correction for multiple testing.
Following AJCC staging system version 7 [36].
According to French Federation of Cancer Centers Sarcoma Group (FFCCS) [5].