In POSEIDON (NCT03164616), 1L T plus D and 4 cycles of platinum-based CT significantly improved PFS and OS vs CT in patients (pts) with EGFR/ALK wild-type mNSCLC leading to the approval of T+D+CT by the FDA; objective response rate (confirmed; 38.8% [95% CI, 33.6–44.3] and 41.5% [95% CI, 36.1–47.0] vs 24.4% [95% CI, 19.9–29.4]) and duration of response were also improved with both T+D+CT and D+CT vs CT. However, the relationship between number of CT cycles and patterns of response in pts with mNSCLC has not been fully established. Here we report outcomes in POSEIDON after 2 vs 4 CT cycles.
Pts (n=1013) were randomised (1:1:1) to 1L T+D+CT, D+CT or CT. Exploratory analyses of objective response (intent-to-treat [ITT] population and subgroups with mutations [m] in STK11, KEAP1 or KRAS) and safety (safety population) were conducted after 2 vs 4 CT cycles (week 6 vs 12 scans, respectively).
78%, 82% and 74% of pts who received T+D+CT, D+CT or CT completed 4 CT cycles. 560 pts had stable disease (SD) after cycle (C) 2. Of these, 22.9% had partial response (PR) after C4 (Table). A similar trend was observed in pts with STK11m, KEAP1m or KRASm mNSCLC; in these pts, improvement appeared greatest with T+D+CT, although 95% CIs were wide and overlapping. Of 252 pts in the ITT with CR/PR after C2, 89.7% remained in response after C4 (Table). Reductions in median target lesion size occurred between C2 and C4 in all arms (data will be presented). The frequency of grade 3/4 adverse events (AEs) and serious AEs originating in C1–2 and C3–4 was similar (data will be presented).
Table 1
| T+D+CT | D+CT | CT | Total | |
| ITT | ||||
| SD@C2 and PR@C4 n/n % (95% CI) | 47/180 26.1 (19.9–33.2) | 42/179 23.5 (17.5–30.4) | 39/201 19.4 (14.2–25.6) | 128/560 22.9 (19.4–26.6) |
| CR/PR@C2 and CR/PR@C4 n/n % (95% CI) | 87/96 90.6 (82.9–95.6) | 87/94 92.6 (85.3–97.0) | 52/62 83.9 (72.3–92.0) | 226/252 89.7 (85.2–93.1) |
| STK11m (NSQ) | ||||
| SD@C2 and PR@C4 n/n % (95% CI) | 5/17 29.4 (10.3–56.0) | 2/13 15.4 (1.9–45.4) | 2/16 12.5 (1.6–38.3) | 9/46 19.6 (9.4–33.9) |
| KEAP1m (any histology) | ||||
| SD@C2 and PR@C4 n/n % (95% CI) | 4/11 36.4 (10.9–69.2) | 4/14 28.6 (8.4–58.1) | 0/5 0.0 (0.0–52.2) | 8/30 26.7 (12.3–45.9) |
| KRASm (NSQ) | ||||
| SD@C2 and PR@C4 n/n % (95% CI) | 13/31 41.9 (24.5–60.9) | 9/33 27.3 (13.3–45.5) | 2/26 7.7 (0.9–25.1) | 24/90 26.7 (17.9–37.0) |
NSQ, non-squamous.
These exploratory data support the use of 4 CT cycles, when given with T (limited course) and D (until progression), to optimise response and tumour shrinkage in pts with mNSCLC, including some harder to treat subgroups. Two further CT cycles did not meaningfully add to toxicity or compromise the ability to administer planned CT.
NCT03164616
Medical writing support for the development of the abstract, under the direction of the authors, was provided by James Holland of Ashfield MedComms (Manchester, UK) an Inizio company, and funded by AstraZeneca.