A Bayesian network meta‐analysis of ALK inhibitor treatments in patients with ALK‐positive non‐small cell lung cancer

OBJECTIVE: To date, no direct comparisons have compared the effectiveness of all ALK inhibitors (ALKis) against ALK‐positive non‐small cell lung cancer (NSCLC). The aim of the present study was to investigate the efficacy and safety of ALKis in ALK‐positive NSCLC. METHODS: The effectiveness of ALKis was evaluated by assessing progression‐free survival (PFS), overall survival (OS), overall response rate (ORR), and PFS with baseline brain metastasis (BM). The serious adverse events (SAEs: Grade ≥ 3) and adverse events (AEs) leading to discontinuation were pooled to evaluate safety. We conducted an indirect treatment comparison between all ALKis by using a Bayesian model. RESULTS: Twelve eligible trials including seven treatments were identified. All of the ALKis improved PFS and ORR relative to chemotherapy. Consistent with alectinib, brigatinib, lorlatinib, and ensartinib showed significant differences versus crizotinib and ceritinib. Lorlatinib seemed to prolong PFS compared with alectinib (0.64, 0.37 to 1.07), brigatinib (0.56, 0.3 to 1.05), and ensartinib (0.53, 0.28 to 1.02). No significant difference was found among them in OS except for alectinib versus crizotinib. Moreover, alectinib was significantly more effective than crizotinib (1.54, 1.02 to 2.5) in achieving the best ORR. Subgroup analyses based on BM indicated that PFS was dramatically lengthened by lorlatinib. Compared with other ALKis, alectinib notably reduced the rate of SAEs. There was no striking difference between discontinuation for AEs, except for ceritinib versus crizotinib. The ranking of validity showed that lorlatinib had the longest PFS (98.32%) and PFS with BM (85.84%) and the highest ORR (77.01%). The rank of probabilities showed that alectinib had the potentially best safety in terms of SAEs (97.85%), and ceritinib had less discontinuation (95.45%). CONCLUSION: Alectinib was the first choice for patients with ALK‐positive NSCLC and even for those with BM, whereas lorlatinib was the second choice. Long‐term follow‐up and prospective studies are warranted to compare ALKis and to verify our conclusions directly.