Abstract 965
Aim/Background
Durvalumab (D; MEDI4736), a human IgG1 anti-PD-L1 antibody, and tremelimumab (T), a human IgG2 anti-CTLA-4 antibody, have shown acceptable safety profiles and antitumour activity as monotherapies; preclinical data suggest their combination may have greater antitumour activity vs either agent alone.
Methods
This is a phase 1 open-label dose-escalation/expansion study (NCT02000947) of D + T in patients (pts) with Stage III/IV NSCLC (≤3 prior therapies; immunotherapy naïve). Primary endpoint is safety and tolerability; dose-limiting toxicity (DLT) period is 28 days. Secondary endpoints include RECIST v1.1 response. PD-L1 expression was tested retrospectively.
Results
As of 15 Apr 2015, 102 pts have received D + T in the dose-escalation phase (Table).
| D Q4W + T Q4W | D Q2W + T Q4W | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Cohort | 1a | 2a | 3a | 3b | 4 | 4a | 5 | 5a | 8 | 9 |
| Dose D (mg/kg) + T (mg/kg) | 3 + 1 | 10 + 1 | 15 + 1 | 10 + 3 | 20 + 1 | 15 + 3 | 15 + 10 | 20 + 3 | 10 + 1 | 10 + 3 |
| n | 3 | 3 | 18 | 3 | 18 | 14 | 9 | 6 | 17 | 11 |
| Pts with ≥1 drug-related AE, n | ||||||||||
| All grades | 1 | 3 | 11 | 3 | 9 | 12 | 8 | 5 | 12 | 10 |
| Grade 3/4 | 0 | 2 | 7 | 2 | 4 | 6 | 7 | 5 | 3 | 5 |
| Drug-related discontinuations | 0 | 1 | 2 | 2 | 1 | 4 | 4 | 3 | 0 | 3 |
| Drug-related deaths | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
2 pts in Cohort 5a had a DLT (Grade [G] 3 increased AST/ALT, G4 increased lipase). 73% had ≥1 drug-related AE; most frequent were diarrhoea (27%) and fatigue (23%). 40% had ≥1 G3/4 related AE; most frequent were colitis (9%) and diarrhoea (8%). 20% had drug-related AEs leading to discontinuation; most frequent was colitis (7%). 2 drug-related AEs led to death. AEs were manageable with standard therapy, except G4 myasthenia gravis and G5 polymyositis (n = 1, Cohort 2a) and G5 neuromuscular disorder (n = 1, Cohort 5a). Increasing T doses with fixed D dose were associated with greater frequency of AEs without deeper, earlier or greater number of responses. D + T had dose-proportional PK exposures in line with monotherapy studies. Of 63 efficacy evaluable pts dosed ≥16 weeks before cutoff, 17 (27%) had partial responses (PR; confirmed/unconfirmed); 14 (22%) had stable disease. PRs occurred in 6 of 18 pts (33%) with PD-L1+ tumours, 9 of 33 pts (27%) with PD-L1− tumours.
Conclusions
M + T has a manageable safety profile and evidence of clinical activity in advanced NSCLC, including PD-L1− disease, supporting continued study of the combination. Recruitment is ongoing.
Clinical trial identification
NCT02000947
Disclosure
N. Rizvi: consulting income from MedImmune, Roche, Merck, AstraZeneca and BMS. A. Balmanoukian: Corporate-sponsored research MedImmune, Merck, Genentech, Merck Serono, Pfizer, BMS; other substantive relationships speakers Bureau for Bristol Myers Squibb. S.B. Goldberg: Advisory Board: Clovis, Boehringer Ingelheim Corporate-Sponsored Research: AstraZeneca, Clovis, Genentech, Merck, Immunogen, Kadmon. J. Chaft: Advisory Board: Genentech, Myriad, Biodesix. R.E. Sanborn: no personal financial conflicts of interest, but institution receives research support from Medimmune. M.C. Rebelatto, R. Narwal, P.B. Robbins, Y. Gu, J.J. Karakunnel: employee of MedImmune and owns stock/stock options in AstraZeneca. S. Antonia: Advisory board: Merck Corporate sponsored research; consultant for BMS, MedImmune/AstraZeneca.