Target trial emulation with real-world data to determine the population-level cost-effectiveness of multi-gene panel sequencing in advanced melanoma.
Bibliographic record
Abstract
e21508 Background: Compared to single-gene BRAF testing to guide targeted treatment for advanced melanoma, multi-gene panels can identify additional gene mutations with known therapeutic or prognostic relevance. No randomized control trials of multi-gene panel sequencing have been completed in advanced melanoma. This study determined the population-level cost-effectiveness of multi-gene panel sequencing compared to single-gene BRAF testing for advanced melanoma. Methods: Our population-based retrospective study emulated a hypothetical pragmatic trial using comprehensive patient-level clinical and health administrative data between September 2016 and December 2018 from British Columbia, Canada. To emulate random treatment assignment, we 1:1 matched multi-gene panel patients to contemporaneous controls using a machine learning approach that maximized balance on 15 covariates. Following matching, we estimated mean three-year survival time and costs (2021 CAD), and calculated incremental net monetary benefit (INMB) for life-years gained (LYG) at $100,000/LYG using inverse probability of censoring weighted linear regression and nonparametric bootstrapping. Besides an intention-to-treat (ITT) effect, we also estimated the per-protocol (PP) effect of initiating treatment within 90 days of receiving test results additionally using inverse probability of treatment weights. We also estimated overall survival using Weibull regression and Kaplan-Meier (KM) survival analysis. Results: We matched 147 patients receiving multi-gene panel sequencing to controls, achieving good balance for all included covariates. ITT mean incremental costs were $19,541 (95%CI: -$18,939, $77,396) and mean incremental LYG were 0.22 (95%CI: -0.06, 0.50). We did not find statistically significant different differences in overall survival using the KM (P = 0.11) and Weibull regression (HR: 0.73 [95%CI: 0.51-1.03]) survival analysis in the ITT analysis. PP incremental costs were $36,367 (95%CI: -$6,653, $120,216) and incremental LYG were 0.56 (95%CI: 0.39, 1.24), with corresponding differences in overall survival using KM (P = 0.02) and Weibull regression (HR: 0.56 [95%CI: 0.36-0.87]) survival analysis. The probability of multi-gene panel sequencing being cost-effective at $100,000/LYG was 54.9% in the ITT analysis and 64.5% in the PP analysis. Conclusions: We found the cost-effectiveness of multi-gene panel sequencing to be evenly poised, with estimates favouring multi-gene panel sequencing with respect to overall survival and cost-effectiveness when accounting for probability of treatment initiation. This real-world evidence generated using randomized trial design principles can support jurisdictions’ deliberations on the reimbursement of precision oncology interventions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.126 | 0.157 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.006 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.011 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".