Table 10.
Distribution of patient categorical features over training, validation, and testing sets.
| Train | Validation | Testing | ||||
|---|---|---|---|---|---|---|
| Lymph node | ||||||
| Num | 326 | – | 186 | – | 200 | – |
| NEG | 113 | 34% | 56 | 30% | 67 | 33% |
| POS | 213 | 65% | 130 | 69% | 133 | 66% |
| No data | 4 | – | 4 | – | 2 | – |
| Stage | ||||||
| Num | 240 | – | 141 | - | 155 | – |
| 1 | 50 | 20% | 25 | 17% | 27 | 17% |
| 0 | 1 | 0% | 0 | 0% | 0 | 0% |
| 3 | 27 | 11% | 25 | 17% | 25 | 16% |
| 2 | 161 | 67% | 91 | 64% | 100 | 64% |
| 4 | 1 | 0% | 0 | 0% | 3 | 1% |
| No data | 90 | – | 49 | – | 47 | – |
| Grade | ||||||
| Num | 323 | – | 185 | – | 197 | – |
| 1 | 19 | 5% | 5 | 2% | 9 | 4% |
| 3 | 207 | 64% | 120 | 64% | 127 | 64% |
| 2 | 97 | 30% | 60 | 32% | 61 | 30% |
| No data | 7 | – | 5 | – | 5 | – |
| Subtype | ||||||
| Num | 328 | – | 187 | – | 202 | – |
| Normal | 27 | 8% | 15 | 8% | 10 | 4% |
| Basal | 75 | 22% | 27 | 14% | 32 | 15% |
| Her2 | 52 | 15% | 27 | 14% | 39 | 19% |
| LumB | 67 | 20% | 50 | 26% | 42 | 20% |
| Claudin-low | 37 | 11% | 23 | 12% | 29 | 14% |
| LumA | 70 | 21% | 45 | 24% | 50 | 24% |
| No data | 2 | – | 3 | – | 0 | – |
| Surgery | ||||||
| Num | 326 | – | 186 | – | 200 | – |
| MASTECTOMY | 209 | 64% | 118 | 63% | 130 | 65% |
| BREAST-CONSERVING | 117 | 35% | 68 | 36% | 70 | 35% |
| No data | 4 | – | 4 | – | 2 | – |
| Histology | ||||||
| Num | 330 | – | 190 | – | 202 | – |
| IDC+ILC | 11 | 3% | 14 | 7% | 10 | 4% |
| IDC-MUC | 6 | 1% | 6 | 3% | 5 | 2% |
| ILC | 25 | 7% | 9 | 4% | 15 | 7% |
| OTHER-INVASIVE | 1 | 0% | 1 | 0% | 0 | 0% |
| OTHER | 1 | 0% | 0 | 0% | 0 | 0% |
| IDC-MED | 6 | 1% | 2 | 1% | 3 | 1% |
| INVASIVE-TUMOUR | 3 | 0% | 0 | 0% | 1 | 0% |
| IDC-TUB | 6 | 1% | 3 | 1% | 3 | 1% |
| DCIS | 1 | 0% | 0 | 0% | 0 | 0% |
| IDC | 270 | 81% | 155 | 81% | 165 | 81% |
| No data | 0 | – | 0 | – | 0 | – |
| Menopause | ||||||
| Num | 330 | – | 190 | – | 202 | – |
| Pre | 111 | 33% | 56 | 29% | 60 | 29% |
| Post | 219 | 66% | 134 | 70% | 142 | 70% |
| No data | 0 | – | 0 | – | 0 | – |
| Her2 SNP6 | ||||||
| Num | 330 | – | 190 | – | 202 | – |
| NEUT | 224 | 67% | 131 | 68% | 137 | 67% |
| LOSS | 20 | 6% | 8 | 4% | 8 | 3% |
| GAIN | 86 | 26% | 51 | 26% | 57 | 28% |
| No data | 0 | – | 0 | – | 0 | – |
| Laterality | ||||||
| Num | 306 | – | 181 | – | 189 | – |
| r | 140 | 45% | 92 | 50% | 87 | 46% |
| l | 166 | 54% | 89 | 49% | 102 | 53% |
| No data | 24 | – | 9 | – | 13 | – |
| Cluster | ||||||
| Num | 330 | – | 190 | – | 202 | – |
| 4.5 | 34 | 10% | 19 | 10% | 21 | 10% |
| 10 | 78 | 23% | 36 | 18% | 33 | 16% |
| 1 | 24 | 7% | 17 | 8% | 10 | 4% |
| 3 | 33 | 10% | 22 | 11% | 23 | 11% |
| 2 | 8 | 2% | 12 | 6% | 8 | 3% |
| 5 | 49 | 14% | 27 | 14% | 30 | 14% |
| 4 | 17 | 5% | 5 | 2% | 12 | 5% |
| 7 | 20 | 6% | 11 | 5% | 8 | 3% |
| 6 | 10 | 3% | 7 | 3% | 10 | 4% |
| 9 | 25 | 7% | 16 | 8% | 19 | 9% |
| 8 | 32 | 9% | 18 | 9% | 28 | 13% |
| No data | 0 | – | 0 | – | 0 | – |
| Cohort | ||||||
| Num | 330 | – | 190 | – | 202 | – |
| 1 | 95 | 28% | 58 | 30% | 57 | 28% |
| 3 | 115 | 34% | 61 | 32% | 73 | 36% |
| 2 | 44 | 13% | 29 | 15% | 36 | 17% |
| 5 | 27 | 8% | 19 | 10% | 14 | 6% |
| 4 | 49 | 14% | 23 | 12% | 22 | 10% |
| No data | 0 | – | 0 | – | 0 | – |
| ER IHC | ||||||
| Num | 328 | – | 189 | – | 199 | – |
| Neg | 150 | 45% | 65 | 34% | 72 | 36% |
| Pos | 178 | 54% | 124 | 65% | 127 | 63% |
| No data | 2 | – | 1 | – | 3 | – |
| ER/HER status | ||||||
| Num | 292 | – | 167 | – | 185 | – |
| HER2+ | 53 | 18% | 24 | 14% | 31 | 16% |
| ER−/HER2− | 86 | 29% | 37 | 22% | 43 | 23% |
| ER+/HER2–High-Prolif | 82 | 28% | 65 | 38% | 65 | 35% |
| ER+/HER2–Low-Prolif | 71 | 24% | 41 | 24% | 46 | 24% |
| No data | 38 | – | 23 | – | 17 | – |
| Cellularity | ||||||
| Num | 320 | – | 186 | – | 200 | – |
| High | 180 | 56% | 92 | 49% | 106 | 53% |
| Moderate | 108 | 33% | 70 | 37% | 75 | 37% |
| Low | 32 | 10% | 24 | 12% | 19 | 9% |
| No data | 10 | – | 4 | – | 2 | – |