BACKGROUND: DESTINY-Gastric01 (NCT03329690) is a randomized, phase 2 study evaluating trastuzumab deruxtecan (T-DXd) in patients with HER2-positive advanced gastric cancer who progressed after ≥2 regimens. T-DXd significantly improved objective response rate (51% vs. 14%; P < 0.001) and overall survival (median OS; 12.5 vs. 8.4 months; P = 0.01) relative to chemotherapy (irinotecan or paclitaxel), leading to regulatory approval in USA and Japan. This post hoc analysis evaluated the overall effect of treatment differences on the quality of survival after discounting for time spent with toxicities or disease progression by comparing the Q-TWiST for patients who received T-DXd versus those who received chemotherapy.
METHODS: Patients were randomized 2:1 to receive T-DXd or chemotherapy. For each treatment arm, OS, truncated at 10.1 months (the median OS for the entire analysis population, following Q-TWiST convention), was partitioned into three health states: time with grade ≥3 toxicities before disease progression (TOX), time before disease progression without symptoms of disease progression or toxicity (TWiST), and time following disease progression prior to death or censoring (PROG). Mean duration in each state was weighted by a utility score, determined first in a threshold analysis, using a range of hypothetical utility values to generate quality-adjusted states, and then based on observed EQ-5D-5L scores, for that state; the sum of the utility-weighted durations yielded the Q-TWiST value for the time until the end of 10.1 months’ follow-up. In both threshold and observed utility analyses, 95% confidence intervals (CIs) and two-sided P values were calculated using the bootstrap method.
RESULTS: Relative to patients receiving chemotherapy (n = 62), patients receiving T-DXd (n = 125) had significantly longer unweighted durations of TOX (3.0 vs. 1.6 months; P < 0.01) and TWiST (3.1 vs. 2.1 months; P < 0.05) and a shorter unweighted duration of PROG (2.4 vs. 3.7 months; P < 0.01). Using a matrix of 25 hypothetical utility-weight combinations, with a TWiST utility of 1 and TOX and PROG utilities ranging from 0 to 1, Q-TWiST differences between treatment arms ranged from −0.5 to 2.3 months, favoring T-DXd in 22 combinations, of which 15 were statistically significant. Using observed EQ-5D-5L utility values, Q-TWiST was 0.9 months (95% CI, 0.2-1.5) longer for T-DXd than for chemotherapy (6.6 vs. 5.7 months), significantly favoring T-DXd (P < 0.05).
CONCLUSIONS: Over a 10-month follow-up period, treatment with T-DXd was associated with a statistically significant gain in quality-adjusted OS versus chemotherapy among previously treated patients with gastric cancer. An analysis using observed EQ-5D-5L utility scores found a statistically significant advantage in Q-TWiST for T-DXd.
