Table 1 Patient characteristics in the training and validation sets
From: Effect of training-sample size and classification difficulty on the accuracy of genomic predictors
| Â | Training set (n = 130) | Validation set (n = 100) | Pvalue |
|---|---|---|---|
Median age | 51 years (28-79 years) | 50 years (26-73 years) | Â |
Race | Â | Â | 0.804 |
   Caucasian | 85 (65%) | 68 (68%) |  |
   African American | 13 (10%) | 12 (12%) |  |
   Asian | 9 (7%) | 7 (7%) |  |
   Hispanic | 21 (16%) | 13 (13%) |  |
   Mixed | 2 (2%) | 0 |  |
Cancer histology | Â | Â | 0.047 |
   Invasive ductal (IDC) | 119 (92%) | 85 (85%) |  |
   Mixed ductal/lobular (IDC/ILC) | 8 (6%) | 8 (8%) |  |
   Invasive lobular (ILC) | 1 (0.7%) | 7 (7%) |  |
   Others | 2 (1.3%) | 0 |  |
Tumor size | Â | Â | 0.643 |
   T0 | 1 (1%) | 2 (2%) |  |
   T1 | 12 (9%) | 8 (8%) |  |
   T2 | 70 (54%) | 62 (62%) |  |
   T3 | 21 (16%) | 13 (13%) |  |
   T4 | 26 (20%) | 15 (15%) |  |
Lymph node stage | Â | Â | 0.935 |
   N0 | 39 (30%) | 27 (27%) |  |
   N1 | 60 (46%) | 47 (47%) |  |
   N2 | 14 (11%) | 13 (13%) |  |
   N3 | 17 (13%) | 13 (13%) |  |
Nuclear grade (BMN) | Â | Â | 0.005 |
   1 | 2 (2%) | 11 (11%) |  |
   2 | 52 (40%) | 42 (42%) |  |
   3 | 76 (58%) | 47 (47%) |  |
Estrogen receptor | Â | Â | 0.813 |
   Estrogen receptor positive | 80 (62%) | 60 (60%) |  |
   Estrogen receptor negative | 50 (38%) | 40 (40%) |  |
HER-2 | Â | Â | < 0.001 |
   HER-2 positive | 33 (25%) | 7 (7%) |  |
   HER-2 negative | 96 (74%) | 93 (93%) |  |
Neoadjuvant therapy | Â | Â | 0.005 |
   Weekly T × 12 + FAC × 4 | 112 (86%) | 98 (98%) |  |
   3-Weekly T × 4 + FAC × 4 | 18 (14%) | 2 (2%) |  |
Pathologic complete response (pCR) | 33 (25%) | 15 (15%) | 0.055 |