Background
In SELECT, lenvatinib (LEN) improved progression-free survival in patients (pts) with differentiated thyroid cancer compared with placebo (18.3 vs 3.6 months; hazard ratio 0.21; 99% CI 0.14–0.31; P<0.001). We report baseline characteristics and change in the sum of target lesion diameter for pts from SELECT who did and did not respond to LEN.
Methods
Pts were randomized 2:1 to receive LEN 24 mg/day or placebo. Tumor assessments: independent radiologic review every 8 weeks during the randomization phase and by investigator review every 12 weeks during the extension phase. Responders were defined as pts who demonstrated either a partial or complete response. All other pts were categorized as nonresponders. Data cutoff: 15 November 2013.
Results
Of 261 LEN-treated pts, 169 were responders and 92 were nonresponders. Among responders, 66% were aged ≤65 years and 46% were male; 48% of nonresponders were aged ≤65 years and 51% were male. Responders showed lower tumor burden than nonresponders. For responders, 33% had baseline tumor burden ≤35 mm and 67% had ECOG PS of 0. For nonresponders, 10% had baseline tumor burden ≤35 mm and 34% had ECOG PS of 0. Median duration of treatment was higher for responders than for nonresponders (14.8 months [range: 1.1‒26.8] vs 5.5 months [range: 0.2‒21.5]). Median baseline sum of target lesion diameters was 51.8 mm (range: 15.1‒181.2) for responders and median maximum percent change from baseline was –52% (range: –100% to –30%). For nonresponders, median baseline sum of target lesion diameters was 71.5 mm (range: 15.9‒331.2) and median maximum percent change from baseline was –20% (range: –38% to 66%).
Conclusions
LEN was effective across tumor burdens in SELECT. In this analysis, the percent of responders with lower tumor burden is higher compared with nonresponders. These data suggest earlier use of LEN could be beneficial.