**3. Results**

Of the 120 patients registered in the trial with material available, 109 had a DNA polymerase β H-score (Polβ) and 74 had both Polβ and *KRAS* mutational status. Figure 1 reports the flowchart of the study.

**Figure 1.** CONSORT diagram showing the flow of participants.

The main demographic characteristics of the population (*n* = 109) and the relationships between characteristics and Polβ are reported in Table 1.

### *3.1. Progression-Free Survival*

The median PFS was, respectively, 5.9 and 7.2 months in the mutated (mut) and wild-type (wt) KRAS groups (adjusted HR mut vs. wt: 1.09, 95% CI: 0.56–2.08, *p* = 0.815).

Polβ, considered a continuous variable, did not have any significant impact on PFS in a multivariable Cox model. HR was 0.99 for each 10-unit increment of the score, with 95% CI 0.97–1.02 and *p* = 0.579. The inclusion of KRAS mutational status in the statistical model did not modify the impact of Polβ on progression or death risk (HR: 0.99, 95% CI: 0.96–1.02, *p* = 0.501). Considering Polβ as a dichotomous variable, median PFS were, respectively, 4 and 6.3 months for negative (neg) and positive (pos) staining. The absence or presence of DNA polymerase β had no impact on the risk of PFS, considering the multivariable models, either including KRAS status or not in the analysis (HR pos vs. neg: 1.10, 95% CI: 0.44–2.70, *p* = 0.847; HR pos vs. neg: 1.08, 95% CI: 0.49–2.38, *p* = 0.857). Detailed results of the multivariable analysis for PFS are reported in Table 2, and the Kaplan–Meier curves for PFS are shown in Figure 2A. The forest plot in Figure 2B graphically shows the effect of KRAS status on the relationship between Polβ and PFS.
