Establishment of a Machine Learning Model for the Risk Assessment of Perineural Invasion in Head and Neck Squamous Cell Carcinoma.

These ML B2/PNI− tumors had a significantly worse disease-specific survival (DSS, HR = 2.288, p = 3.8 × 10 −2 ) and showed a clear trend towards unfavorable overall survival (OS, HR = 1.711, p = 9.1 × 10 −2 ) and progression-free interval (PFI, HR = 1.669, p = 1.1 × 10 −1 ) compared to ML A/PNI− tumors ( Figure 3 D).

also in 7,418 other papers

PNI+ tumors were strongly enriched in subcluster B2 ( Supplemental Table S4 ), and differences in gene set variation analysis scores for up- (ANOVA p < 2.2 × 10 −16 ) or downregulated (ANOVA p < 2.2 × 10 −16 ) genes of the PNI-related 44-gene set were highly significant between subcluster B2 and cluster A or subcluster B1 ( Figure 1 B).

also in 36,814 other papers