ALEX final OS analysis: Alectinib superior to crizotinib in 1L ALK+ NSCLC

Final overall survival (OS) data analysis from the phase III ALEX study presented at ESMO 2025 indicates alectinib is superior to crizotinib in patients with previously untreated ALK-positive (ALK+) non-small-cell lung cancer (NSCLC).

Crizotinib was the first ALK tyrosine kinase inhibitor (TKI) to be approved for advanced ALK+ NSCLC after demonstrating superior progression-free survival (PFS) and objective response rate (ORR) over platinum-based chemotherapy. [Transl Lung Cancer Res 2023;12:369-378; N Engl J Med 2014;371:2167-2177] However, eventual resistance to and frequent early central nervous system (CNS) progression with crizotinib prompted the development of more potent and ALK-selective TKIs, such as alectinib.

ALEX was a phase III global, randomized trial evaluating alectinib (n=152) vs crizotinib (n=151) in patients with previously untreated advanced ALK+ NSCLC. In the last data analysis before ESMO 2025, which was conducted after a median survival follow-up of 48.2 months, the OS data were not mature, but a clinically meaningful trend favouring alectinib was observed. [Ann Oncol 2020;31:1056-1064]

At the updated data cut-off, after a median follow-up of 53.5 months for alectinib and 23.3 months for crizotinib, the median OS was 81.1 vs 54.2 months (hazard ratio [HR], 0.78; 95 percent confidence interval [CI], 0.56–1.08). The respective 7-year OS rates were 48.6 and 38.2 percent. [Ann Oncol 2025;doi:10.1016/j.annonc.2025.09.018] “The high OS rate for alectinib [and to a lesser extent crizotinib] at the 7-year mark suggests that there may be a subgroup of patients with more favourable prognostic markers; further studies are needed to determine what these prognostic markers may be,” remarked the researchers.

Improved median OS with alectinib vs crizotinib was observed across multiple prespecified subgroups, notably including a greater than 30-month improvement (63.4 vs 30.9 months) in patients with CNS metastases at baseline, a group which is typically associated with poorer prognosis. In patients without CNS metastases at baseline, the median OS was 94.0 vs 69.8 months (HR, 0.87; 95 percent CI, 0.58–1.32) in the alectinib vs crizotinib arm.

The median duration of treatment was 28.1 months with alectinib and 10.8 months with crizotinib. Almost all patients experienced any-grade adverse events (AEs) in the alectinib arm (96.7 percent) and in the crizotinib arm (98.0 percent). The most common AEs in the alectinib arm were constipation (40.1 percent), anaemia (28.9 percent), and increased blood bilirubin level (24.3 percent). In the crizotinib arm, the most common AEs were nausea (49.7 percent), diarrhoea (46.4 percent), and vomiting (41.1 percent). Rates of grade 3–5 AEs were similar across alectinib (57.9 percent) and crizotinib (57.6 percent) arms. AEs leading to dose reduction, dose interruption, or treatment withdrawal were reported in 23.0, 32.2, and 17.8 percent of patients treated with alectinib and in 19.9, 28.5, and 14.6 percent, of patients in the crizotinib arm, respectively. The most common AEs leading to dose reduction or discontinuation were increased blood bilirubin in the alectinib arm and increased alanine aminotransferase level in the crizotinib arm.

“Even with a longer duration of treatment and follow-up of patients treated with alectinib compared with crizotinib, the safety profile of alectinib remained tolerable and consistent,” commented the researchers.