After controlling the clinical and pathological characteristics (age, gender, metastasis, smoke, tumor stage, tuberculosis, tumor size), multivariate (Cox regression) survival analysis was conducted to evaluate the potential effects of the three CCN genes on cancer prognosis.
Due to cohort size, multivariate analysis was not performed in this study.
Considering the small simple size, Multivariate stepwise backward analyses included all variables significant at P ≤ 0.05 in Univariate analysis [ 26].
In our study, the necrosis rate impacted on the outcome but less than the percentage of residual tumour cells weighted by the total tumour size (multivariate analysis).
To assess whether BRAF status was associated with positive lymph nodes independently of size, multivariate analysis was carried out using logistic regression and the corresponding likelihood ratio test.
For the 33 superficial STS with tumour size recorded multivariate analysis revealed size, necrosis and cellularity as independent prognostic variables.
Future research could address this with larger sample sizes and multivariate analyses.
The accuracy and efficiency of the proposed method, and its applicability to various types and sizes of multivariate normal integrals are demonstrated by numerical examples.
Univariate t-tests (unequal variances and equal sample sizes) and multivariate linear regression (coded in the R package LIMMA [20]) were performed.
