The Cost-Effectiveness of Lorlatinib Versus Chemotherapy as a Second- or Third-Line Treatment in Anaplastic Lymphoma Kinase (ALK)-Positive Non-small-cell Lung Cancer in Sweden

Abstract

Background Lorlatinib is a third-generation anaplastic lymphoma kinase (ALK)/c-ros oncogene 1 (ROS1) tyrosine kinase inhibitor (TKI) with efficacy in patients with ALK-rearranged non-small-cell lung cancer (NSCLC) previously treated with a second-generation ALK inhibitor or with first-and second-generation ALK inhibitors. We examined the cost-effectiveness of second-or third-line+ (2L+ or 3L+) lorlatinib in Sweden, versus chemotherapy. Methods A partitioned survival model with three health states (progression free, progressed, or death) was used. Lorlatinib relative efficacy versus chemotherapy was derived using unanchored matching adjusted indirect treatment comparisons from a phase 2 clinical trial. Utility data were derived from the same trial and published studies. Costs (year 2019) were obtained from Swedish national data. Costs and benefits were discounted at 3% per annum using a societal perspective (base case). Model robustness was evaluated with deterministic and probabilistic sensitivity analyses. Results For 2L+, the average discounted total quality-adjusted life year (QALY) gain was 1.29 years. Total incremental costs were Swedish krona (SEK) 731,791, resulting in an incremental cost-effectiveness ratio (ICER) of SEK 566,278 per QALY gained. Non-discounted survival gain amounted to 1.94 years. For 3L+, the average discounted total QALY gain was 1.25 years. Total incremental costs were SEK 754,801, resulting in an ICER of SEK 603,934 per QALY gained. Nondiscounted survival gain was 1.88 years. Sensitivity analyses were consistent. Conclusions ICERs ranged from SEK 421,000 to SEK 384,066 less than the boundary for a cost-effective treatment for a high-severity disease in Sweden (SEK 988,000), suggesting 2L+ or 3L+ lorlatinib is a cost-effective treatment for ALKpositive NSCLC versus chemotherapy.

Cost-effectiveness Analysis of Lorlatinib in Patients Previously Treated with Anaplastic Lymphoma Kinase Inhibitors for Non-small Cell Lung Cancer in Greece

“…A published partitioned survival model 15 in Microsoft Excel was locally adapted to reflect the natural progression of patients with advanced, ALK-positive NSCLC previously treated with ≥1 second-generation ALK TKI. A partitioned survival model is a common modeling structure in oncology and an approach consistent with previous submissions to the National Institute for Health and Care Excellence (NICE) and published literature.…”

“…A partitioned survival model is a common modeling structure in oncology and an approach consistent with previous submissions to the National Institute for Health and Care Excellence (NICE) and published literature. 15,16 Therefore, a partitioned survival methodology was the most suitable for this decision problem. Hence, the model evaluated the cost-effectiveness of lorlatinib compared with a combination of pemetrexed and P-ChT, such as carboplatin or cisplatin, from a public payer perspective over a lifetime horizon.…”

“…Therefore, to compare treatments incrementally, accounting for heterogeneity between each of the trials, a matching-adjusted indirect comparison (MAIC) was performed. 15 MAIC and indirect comparisons have been widely used and accepted in health technology submissions to NICE, including the submissions in the ALK-positive NSCLC population. 15 To assess the feasibility of indirect comparisons, we found that it was most appropriate to match the identified comparator studies to a subset of the lorlatinib cohorts that were most similar, particularly in terms of the line of therapy.…”

“…15 MAIC and indirect comparisons have been widely used and accepted in health technology submissions to NICE, including the submissions in the ALK-positive NSCLC population. 15 To assess the feasibility of indirect comparisons, we found that it was most appropriate to match the identified comparator studies to a subset of the lorlatinib cohorts that were most similar, particularly in terms of the line of therapy. Hence, for chemotherapy, the hazard ratio (HR) of PFS was estimated using the EXP2-3A lorlatinib cohorts, which align with the populations in the ALUR and ASCEND-5 studies.…”

“…7,20 Further, these estimates have been used in the recent publication of the model. 15 Extrapolation of OS and PFS for lorlatinib was performed to estimate likely outcomes beyond the observed duration of the clinical trial. Extrapolation was undertaken by fitting standard parametric survival curves, each being fit to patient-level data from the lorlatinib trial.…”