OBJECTIVES : To compare the relative efficacy of the ALK-tyrosine kinase inhibitor (TKI) lorlatinib, investigated in the single-arm Phase I/II trial B7461001 as a second-line or later treatment for patients with advanced ALK-positive NSCLC, to chemotherapy, a key comparator within this indication.

METHODS : A clinical systematic literature review (SLR) performed in February 2017 (updated February 2018) identified two relevant studies with PFS data (ALUR and ASCEND-5) investigating chemotherapy treatments (pemetrexed or docetaxel). MAICs were performed (matching on Eastern Cooperative Oncology Group Performance Status [ECOG PS], race, gender, and brain metastases) using the pooled pseudo-PFS data for chemotherapy and the lorlatinib patient cohorts previously treated with at least two ALK-TKIs or one ALK-TKI other than crizotinib. For OS, MAICs were performed using a retrospective study (Ou 2014), although there are substantial limitations with this evidence source. An SLR update in February 2020 identified a retrospective study (Lin 2019) with PFS data for platinum-based chemotherapy. An MAIC matching on age, race, gender and brain metastases (Lin 2019 did not report ECOG PS) compared lorlatinib to platinum-based chemotherapy.

RESULTS : Lorlatinib was consistently associated with a significant decrease in the hazard of progression versus all types of chemotherapy assessed. For the original comparison to docetaxel and pemetrexed, the adjusted hazard ratio (HR) was 0.36 (95% confidence interval [CI]: 0.30-0.44) and the naïve HR was 0.36 (95% CI: 0.27-0.47). For comparison to platinum-based chemotherapy the adjusted HR was 0.40 (95% CI: 0.29-0.55); the naïve HR was 0.37 (95% CI: 0.23-0.57). Lorlatinib provided a significant decrease in hazard of death versus chemotherapy; with HRs ranging from 0.30 (95% CI: 0.18-0.50) to 0.43 (95% CI: 0.27-0.60).

CONCLUSIONS : Lorlatinib consistently demonstrated a significant improvement in the outcomes of both PFS and OS when compared with chemotherapy, although the evidence base is limited.