Phase II Study of Bevacizumab in Combination with 1st-Line Chemotherapy or 2nd-Line Erlotinib in Non-Squamous Nsclc (Ns-Nsclc) Patients with Asymptomatic Untreated Brain Metastases (ML21823)

ABSTRACT Background Brain metastases (BM) occur in up to 56% of pts with advanced cancer and are no longer a contraindication to bevacizumab treatment based on safety data. The non-comparative phase II BRAIN trial (NCT00800202) is the first study to assess the efficacy/safety of bevacizumab in combination with systemic treatments in ns-NSCLC pts with untreated BM. Methods Eligible pts (stage IV ns-NSCLC; PS 0–1; untreated, asymptomatic BM; ineligible for surgery/radiosurgery) received: arm A (n = 67),1st-line bevacizumab (15mg/kg q3w until progression/unacceptable toxicity) plus carboplatin (AUC 6 q3w, ≤6 cycles) and paclitaxel (200mg/m2 q3w, ≤6 cycles) (B + CP); or arm B (n = 24), bevacizumab (as above) plus erlotinib (150mg/day until progression/unacceptable toxicity) (B + E) in 2nd line. Primary endpoint: 6-month progression-free survival (PFS) by treatment arm. Secondary endpoints: response rate (RR; RECIST v1.1), overall survival (OS) and safety. The trial could be halted if the incidence of brain haemorrhage (ICH) was >3 in B + CP or >2 in B + E. All sites of disease were assessed every 6 weeks including mandatory MRI for BM assessment. Results Pt baseline characteristics and outcomes are shown (Table). Efficacy: the observed 6-month PFS was 56.5% (B + CP) and 58.0% (B + E). Safety: ICH frequency comparable to previously published data in a similar pt population (B + CP, n = 1 grade 1; B + E, n = 0). Grade 3–5 adverse events of special interest (AESI) were seen in 19.4% (B + CP) and 20.8% (B + E) of pts. No erlotinib-related grade 3–5 AESIs were seen. There were 3 AEs leading to death (1 epilepsy, B + CP; 1 hypertensive encephalopathy, 1 ischaemic stroke, B + E). Conclusion Bevacizumab with 1st-line chemotherapy or 2nd-line erlotinib demonstrated efficacy and acceptable safety in pts with ns-NSCLC with asymptomatic untreated brain metastases when compared to historical controls. Baseline characteristics and clinical outcomes for patients in the BRAIN study. B + CP B + E n = 67 n = 24 Baseline characteristics Gender, n (%) Male 46 (68.7) 11 (45.8) Female 21 (31.3) 13 (54.2) Median age, years (range) 61.0 (40–79) 54.0 (34–70) ECOG performance status, n (%) 0 37 (55.2) 13 (54.2) 1 30 (44.8) 11 (45.8) WHO histology, n (%) Adenocarcinoma 59 (88.1) 23 (95.8) Large-cell carcinoma 8 (11.9) 1 (4.2) Recurrence, n (%) No 61 (91.0) 13 (54.2 Yes 6 (9.0) 11 (45.8) Clinical outcomes (95% CI) 6-month PFS, % 56.5 (43.8–67.4) 58.0 (36.0–74.8) Median PFS, months 6.7 (5.7–7.1) 6.3 (2.5–8.4) Median OS, months 15.1 (11.8–NR) 13.6 (7.5–26.3) 12-month OS, % 62.8 (49.7–73.4) 50.7 (28.7–69.0) 18-month OS % 41.1 (27.4–54.3) 40.5 (20.2–60.0) Overall RR, % 62.7 (50.0, 74.2) 12.5 (2.7, 32.4) RR, % Intracranial metastases 61.2 (48.5–72.9) 20.8 (7.1–42.2) Extracranial metastases 64.2 (51.5–75.5) 12.5 (2.7–32.4) NR = not reached Disclosure B. Besse: Received grants from Roche. H. Senellart: Attended advisory boards. F. Barlesi: Attended adivsory boards for Roche. Received research funding from Roche. E. Dansin: Attended advisory boards for Roche, Lilly and Boehringer-Ingelheim. M. Perol: Attended advisory boards for Roche, Pfizer, Lilly and Boehringer-Ingelheim. D. Moro-Sibilot: Attended adivsory boards for Roche, Pfizer and Lilly. Received reserarch funding from Roche, Pfizer, Lilly, Astra Zeneca and BIF. Substantive Relationships with Roche, Pfizer, Lilly, Astra Zeneca and BIF. J. Soria: Attended Roche Advisory board. All other authors have declared no conflicts of interest.