Table 2. Univariate analysis of potential predictors of ECS.
| Clinicopathologic parameters | N−* | pN+ ECS−** | pN+ ECS+*** | P-valuea |
|---|---|---|---|---|
| Gender | ||||
| Male | 45 | 14 | 27 | 0.023b |
| Female | 29 | 7 | 5 | |
| Age (years) | ||||
| ⩽60 | 35 | 10 | 15 | NS |
| >60 | 39 | 11 | 17 | |
| Alcohol habit | ||||
| (−) | 38 | 8 | 10 | 0.005b |
| (+) | 33 | 13 | 21 | |
| Smoking habit | ||||
| (−) | 39 | 12 | 11 | NS |
| (+) | 32 | 9 | 20 | |
| Tumor site | ||||
| Tongue | 45 | 13 | 19 | NS |
| Lower gingiva | 18 | 4 | 5 | |
| Floor of mouth | 3 | 2 | 5 | |
| Upper gingiva | 2 | 1 | 1 | |
| Buccal mucosa | 6 | 1 | 2 | |
| Growth pattern | ||||
| Superficial | 18 | 3 | 6 | NS |
| Exophytic | 28 | 8 | 7 | |
| Endophytic | 28 | 10 | 19 | |
| Pathological T stage | ||||
| 1–2 | 65 | 14 | 23 | 0.042c |
| 3–4 | 9 | 7 | 9 | |
| Cellular differentiation | ||||
| Well/moderate | 64 | 19 | 24 | NS |
| Poor | 10 | 2 | 8 | |
| Mode of invasion | ||||
| 1–3 | 53 | 13 | 14 | 0.009b |
| 4C–D | 21 | 8 | 18 | |
| Number of positive nodes d | ||||
| 1 | — | 13 | 5 | 0.001e |
| ⩾2 | — | 8 | 27 | |
| Size of positive nodesf (mm) | ||||
| Minor axis (3–20) | ||||
| ⩽10 | — | 13 | 16 | NS |
| >10 | — | 8 | 16 | |
| Major axis (5–27) | ||||
| ⩽14 | — | 14 | 14 | NS |
| >14 | — | 7 | 18 | |
| Gene status in primary tumor | ||||
| CCND1 numerical aberration | ||||
| (−) | 56 | 16 | 12 | 0.0001b |
| (+) | 18 | 5 | 20 | 0.011e |
| EGFR numerical aberration | ||||
| (−) | 51 | 18 | 14 | 0.018b |
| (+) | 23 | 3 | 18 | 0.004e |
Abbreviations: ECS=extracapsular spread; EGFR=epidermal growth factor receptor gene; NS=not significant.
The P-value was determined using the two-tailed Fisher exact test.
* vs ***.
* vs **.
Determined according to histopathological diagnosis.
** vs ***.
Detected by computed tomography.