*3.4. Multivariate Analysis of the Role of TP53 Mutation: Combined Cohorts of Patients*

To obtain a more precise estimate of the effect of *TP53* exon 8 mutation on PFS and OS, and to determine its potential independent role considering other information, a pooled analysis considering either data from our previously analyzed cohort and the one described in the present study, was performed.

The final multivariate model for PFS included both *EGFR* exon 19 deletion as well as TP53 mutation. As soon as the effect of exon 19 deletion on the hazard of disease progression or death is not constant over time, that is, the proportional hazards assumption underlying the Cox model was violated, to obtain a better model fit, this variable was entered into the model with a time-dependent coefficient. Table 3 shows that the effect of exon 19 mutation changes over time, showing a strong protective effect over the first six months that vanishes afterward.


**Table 3.** Multivariate Cox analysis of progression-free survival (PFS) (*n* = 272).

Adjusting for presence of *EGFR* exon 19 deletion, *TP53* mutations affecting exon 8 demonstrated to be the unique independent negative prognostic factor for PFS (HR 1.81, 95% CI: 1.13–2.92, Table 3). With regard to OS, only deletion in *EGFR* Exon 19 resulted associated to OS, probably due to data from our previous cohort (HR 0.52 (95% CI: 0.26–1.03) for the first 6 months of follow-up, HR 0.44 (95% CI: 0.22–0.90), for successive 6 months, and HR 1.08 (95% CI: 0.72–1.61) after 12 months).
