Figure 2. Linear-dependent effect of increasing age on OS. The estimated logarithm HRs (solid line) with 95% CIs (shading) for the association of patients’ age with OS in 2640 patients based on the the dfmacox in a smoothHR – the optimal extended Cox-type additive hazard regression unadjusted model (A) or the model adjusted for Ann Arbor stage, B symptoms, LDH, ECOG PS, and PTI (B). The effect of age on the risk of mortality was modeled using a penalized spline (P-spline) expansion, with patients’ age as a continuous covariate. An age cutoff of 46 years (indicated by the vertical line), defined by AIC analysis, was used as the reference value for calculating the HR. OS, overall survival; HR, hazard ratio; CI, confidence interval; dfmacox, degrees of freedom in multivariate additive Cox models; LDH, lactate dehydrogenase; ECOG, Eastern Cooperative Oncology Group; PS, performance status; PTI, primary tumor invasion; AIC, Akaike’s information criterion.