Table 2.
Exploratory analysis to investigate the influence of DM status at baseline covariate on PFS (intention-to-treat population).
| Variable (Reference Level) | HR (95% CI) | p Value |
|---|---|---|
| Diabetes at baseline (No) | 1.64 (0.95 to 2.84) | 0.079 |
| Previous therapy at entry (No) | 1.56 (0.88 to 2.77) | 0.130 |
| Progression at baseline (No) | 3.19 (1.34 to 7.61) | 0.009 |
| Treatment (Placebo) | 0.53 (0.31 to 0.89) | 0.017 |
| Interaction between treatment and DM at baseline interaction (No) | 0.78 (0.29 to 2.09) | 0.619 |
CI, confidence interval; DM, diabetes mellitus; HR, hazard ratio; PFS, progression-free survival. Each baseline covariate was included one at a time in a Cox proportional hazards model (treatment and the baseline stratification factors centrally assessed progression at baseline and previous therapy for non-functional-NET at entry are forced into all models). Any factors found to be significant at p < 0.10 (via the Wald Chi-square test statistic) were considered potentially important.