Phase I study of gefitinib (G) + durvalumab (D) for locally advanced/metastatic non-small cell lung cancer (NSCLC) harbouring epidermal growth factor receptor (EGFR) sensitising mutations

Background: G and D have both shown efficacy in patients (pts) with NSCLC; G + D may improve durability of response. Methods: This Phase 1 dose escalation (Part A) and expansion (Part B) study (NCT0208811) assessed G 250 mg once daily + D 3 mg/kg (Part A) or 10 mg/kg (Parts A + B) every 2 weeks in pts with locally advanced/metastatic NSCLC. Part A pts were all comers who had failed to respond/relapsed following standard treatment (Tx). Part B pts had sensitising EGFR mutations and were tyrosine kinase inhibitor naïve: Arms 1 + 1a received G + D; Arm 2 received G (4 weeks) before G + D. Primary objective: safety/tolerability. Secondary objectives: pharmacokinetics (PK), pharmacodynamics, immunogenicity (anti-drug antibodies [ADAs]) and efficacy. Exploratory objective: evaluation of biomarkers (e.g. tumour programmed cell death ligand-1 [PD-L1]) and relationship with efficacy. Results: There were no dose limiting toxicities in Part A (n = 16) and D 10 mg/kg was used in Part B. In Part B (n = 40) all pts had possible Tx related adverse events (TRAEs; Table): diarrhoea (68%) and elevated alanine aminotransferase (ALT; 58%) were the most common TRAEs; elevated ALT (20%) and aspartate aminotransferase (15%) were the most common TRAEs leading to discontinuation. PK were as expected, inhibition of soluble PD-L1 was observed in all pts and no Tx emergent ADAs were observed. In Arms 1 + 1a, most patients achieved objective response (63.3%; 95% confidence intervals [CI]: 43.9, 80.1), median duration of response was 9.2 months (95% CI: 3.7, 14.0) and median progression-free survival (mPFS) was 10.1 months (95% CI: 5.5, 15.2; Table). PD-L1 expression ≥20% was associated with numerical improvements in mPFS (Table).Table84O Summary of safety, exposure, efficacy and exploratory analysesSafetyArm 1 (n = 10)Arm 1a (n = 20)Arm 2 (n = 10)Any grade AE, n (%)10 (100.0)20 (100.0)10 (100.0)Grade ≥3 AE, n (%)5 (50.0)15 (75.0)8 (80.0)Any grade TRAE, n (%)10 (100.0)20 (100.0)10 (100.0)Grade ≥3 TRAE, n (%)4 (40.0)11 (55.0)7 (70.0)Any grade TRAE leading to treatment discontinuation, n (%)010 (50.0)6 (60.0)Grade ≥3 TRAEaOne cycle equalled 28 days. leading to treatment discontinuation, n (%)09 (45.0)5 (50.0)ExposureArm 1 (n = 10)Arm 1a (n = 20)Arm 2 (n = 10)Median cyclesaOne cycle equalled 28 days. of D, n10.53.06.0Median total G treatment duration, months (min, max)12.0 (5, 13)5.7 (1, 12)7.1 (1, 13)EfficacyArm 1 (n = 10)Arm 1a (n = 20)Arm 2 (n = 10)Achieved OR, % (95% CI)80.0 (44.4, 97.5)55.0 (31.5, 76.9)70.0 (34.8, 93.3)Median DoR, months (95% CI)8.8 (3.0, 14.8)7.4 (3.7, 20.7)12.6 (5.5, 20.4)Median PFSbPatients with progression events: Arm 1, n = 9; Arm 1a, n = 16; Arm 2, n = 8; PD-L1 positive, n = 8; PD-L1 negative, n = 22., months (95% CI)10.5 (4.6, 17.0)9.3 (4.6, 15.2)12.0 (2.7, 15.6)PD-L1 statuscDetermined using fresh and archival baseline tumour biopsies (archival biopsies permitted in Arm 1a only) and SP263 Roche Tissue Diagnostics protocol: PD-L1 positivity was defined as tumour cell PD-L1 expression ≥20%; negative was defined as PD-L1 expression <20%. Results from Arms 1, 1a, and 2 were combined for this analysis.Positive (n = 12)Negative (n = 24)Median PFSbPatients with progression events: Arm 1, n = 9; Arm 1a, n = 16; Arm 2, n = 8; PD-L1 positive, n = 8; PD-L1 negative, n = 22., months (95% CI)15.9 (2.8, 21.3)9.1 (5.5, 11.9)HR (95% CI)0.461dHR was not statistically significant and should be interpreted with caution due to low patient numbers. AE, adverse event; CI, confidence interval; D, durvalumab; DoR, duration of response; G, gefitinib; HR, hazard ratio; max, maximum; min, minimum; n, number of patients; OR, objective response; PD-L1, programmed cell death ligand-1; PFS, progression-free survival; TRAE, treatment-related adverse event. (0.187, 1.029)a One cycle equalled 28 days.b Patients with progression events: Arm 1, n = 9; Arm 1a, n = 16; Arm 2, n = 8; PD-L1 positive, n = 8; PD-L1 negative, n = 22.c Determined using fresh and archival baseline tumour biopsies (archival biopsies permitted in Arm 1a only) and SP263 Roche Tissue Diagnostics protocol: PD-L1 positivity was defined as tumour cell PD-L1 expression ≥20%; negative was defined as PD-L1 expression <20%. Results from Arms 1, 1a, and 2 were combined for this analysis.d HR was not statistically significant and should be interpreted with caution due to low patient numbers. AE, adverse event; CI, confidence interval; D, durvalumab; DoR, duration of response; G, gefitinib; HR, hazard ratio; max, maximum; min, minimum; n, number of patients; OR, objective response; PD-L1, programmed cell death ligand-1; PFS, progression-free survival; TRAE, treatment-related adverse event. Open table in a new tab Conclusions: G + D had a high discontinuation rate due to liver related TRAEs and there was no additional benefit vs historical data for G alone. However, tumours expressing PD-L1 had favourable PFS and could be investigated further. Clinical trial identification: NCT02088112; March 14, 2014. Editorial acknowledgement: Medical writing support, under the direction of the authors, was provided by Lauren McNally, MSci, of CMC CONNECT, a division of McCann Health Medical Communications Ltd, Glasgow, UK, with funding from AstraZeneca PLC, in accordance with Good Publication Practice (GPP3) guidelines. Legal entity responsible for the study: MedImmune LLC (a wholly owned subsidiary of AstraZeneca PLC). Funding: AstraZeneca PLC. Disclosure: T. Yeh, W. Tang, M. Tang, H.K. Angell, M.P. Roudier, M. Marotti: Employee: AstraZeneca. R. Taylor: Employee, contractor: AstraZeneca. D.L. Gibbons: Advisory boards/research funding: AstraZeneca. All other authors have declared no conflicts of interest.