Autores: Herbst, R. S.; Garon, E. B. ; Kim, D. W.; Cho, B. C.;
Pérez Gracia, José Luis; Han, J. Y.; Arvis, C. D.; Majem, M. ; Forster, M.; Monnet, I.; Novello, S.; Szalai, Z. ; Gubens, M. A.; Su, W. C.; Ceresoli, G. L.; Samkari, A.; Jensen, E.; Lubiniecki, G. M.; Baas, P.
Resumen:
Background: In the global, open-label, phase 2/3 study KEYNOTE-010, pembro 10¿mg/kg or 2¿mg/kg Q3W improved OS vs docetaxel in pts with previously treated advanced NSCLC with PD-L1 TPS ¿50% and ¿1% (coprimary analyses) at median follow-up of 13.1¿mo. We present long-term results overall, in pts who completed 35 cycles (~2 y) of pembro, and in pts who received a second course of pembro.
Methods: Pts aged >18 y with previously treated advanced NSCLC with PD-L1 TPS ¿1% were randomized 1:1:1 to pembro 10¿mg/kg or 2¿mg/kg Q3W, or docetaxel 75¿mg/m2 Q3W. Pts received pembro for 35 cycles, until disease progression/intolerable toxicity. Response was assessed every 9 wk (RECIST 1.1 by independent central review), and survival every 2¿mo posttreatment. There was no difference between pembro doses in the primary analysis, thus doses were pooled in this analysis.
Results: As of March 16, 2018, median (range) follow-up was 42.6 (35.2¿53.2) mo overall (N¿=¿1033). Pembro improved OS vs docetaxel in pts with PD-L1 TPS ¿50% (HR, 0.53; 95% CI, 0.42¿0.66; P¿<¿0.00001) and TPS ¿1% (HR, 0.69; 95% CI, 0.60¿0.80; P¿<¿0.00001). In pts with PD-L1 TPS ¿50%, median (95% CI) OS was 16.9 (12.3¿21.4) mo with pembro vs 8.2 (6.4¿9.8) mo with docetaxel; 36-mo OS rates were 35% vs 13%, respectively. Similar to the primary analysis, 16% of pembro pts and 36% of docetaxel pts had grade 3¿5 treatment-related AEs. 79 of 690 pembro pts received 35 treatment cycles (~2 y). 36-mo OS rate among these 79 pts was 99% and 75 (95%) had PR/CR as best response; 72 pts (91%) remained alive. 48 pts (64%) had an ongoing response; median duration of response was not reached (range, 4¿46+ mo). 25 of 79 pts (32%) had PD (investigator review) after stopping 35 cycles of pembro. 14 pts received second course pembro, 5 of whom completed 17 cycles; 6 (43%) had PR, 5 (36%) had SD, and 11 (79%) remained alive.
Conclusions: At 43-mo follow-up, pembro continued to prolong OS vs docetaxel in pts with previously treated, PD-L1¿expressing advanced NSCLC, with manageable long-term safety. Most pts who completed 35 cycles (~2 y) of pembro had durable response. The majority of pts with PD by investigator review who received second course pembro had either PR or SD and remained alive.
