Abstract 1260P
Background
Patients (pts) with BRAF V600-mutated NSCLC are a rare population for whom targeted treatments are sparse and evidence on their efficacy is limited to uncontrolled trials. The combination of dabrafenib + trametinib (D+T), inhibitors of BRAF V600 and MEK, respectively, was approved for BRAF V600-mutated mNSCLC based on a single-arm phase II study showing clinically meaningful efficacy and a manageable safety profile. This retrospective study compared 1L D+T vs other standard of care (SoC) in BRAF V600-mutated mNSCLC patients in real-world (RW) practice to provide comparative effectiveness data.
Methods
Pts with BRAF V600-mutated mNSCLC receiving 1L therapy from the nationwide electronic health record-derived de-identified Flatiron Health NSCLC database (January 2011–June 2020) were included. RW progression-free survival (rwPFS) and RW overall survival (rwOS) were compared between 1L D+T vs immune checkpoint inhibitor (ICI; pembrolizumab) + platinum doublet chemotherapy (PDC), ICI alone, or PDC alone. Stabilized inverse probability of treatment weighting using the propensity score was used to balance baseline covariates for each comparison and estimate the average treatment effect.
Results
After weighting, baseline characteristics were balanced across cohorts. Median rwOS was numerically longer and rwPFS was similar with D+T vs ICI+PDC in unweighted and weighted analyses (Table). Median rwOS and rwPFS were numerically longer with D+T vs ICI in unweighted and weighted analyses. Median rwOS was numerically longer and rwPFS was similar with D+T vs PDC in unweighted analyses, and both were statistically significantly longer with D+T vs PDC in weighted analyses. Table: 1260P
Kaplan-Meier rwOS and rwPFS in weighted analyses
| D+T | ICI+PDC | D+T | ICI | D+T | PDC | |
| N pre-weighting | 48 | 31 | 44 | 31 | 48 | 30 |
| N post- weighting a | 47.2 | 27.7 | 44.4 | 30.3 | 48.3 | 30.3 |
| Median rwOS, months (95% CI) | 29.3 (16.4, NR) | 17.7 (10.5, NR) | 29.3 (16.2, NR) | 10.9 (5.6, 36.9) | 34.7 (16.4, NR) | 9.7 (0.9, 32.0) |
| P = 0.73 | P = 0.24 | P < 0.01 | ||||
| HR (95% CI) | 0.83 (0.32, 2.15) P = 0.71 | 0.6 (0.3, 1.19) P = 0.15 | 0.36 (0.18, 0.72) P < 0.01 | |||
| Median rwPFS, months (95% CI) | 9.6 (6.5, 15.2) | 10.5 (3.7, NR) | 10.5 (6.5, 15.2) | 5.9 (3.4, 7.6) | 13.7 (7.0, NR) | 4.9 (0.9, 15.2) |
| P = 0.51 | P = 0.09 | P = 0.05 | ||||
| HR (95% CI) | 1.35 (0.63, 2.92) P = 0.44 | 0.57 (0.32, 0.99) P = 0.05 | 0.52 (0.3, 0.91) P = 0.02 |
aSum of weighted patients. CI, confidence interval; HR, hazard ratio; NR, not reached.
Conclusions
In this RW comparison 1L D+T was associated with a significant improvement in rwPFS and rwOS vs 1L PDC, and numerical improvement in OS vs 1L ICI alone or combined with PDC, supporting the 1L use of D+T for BRAF V600 -mutated mNSCLC.
Clinical trial identification
Editorial acknowledgement
Isabella Kaufmann, PhD, of Novartis Pharmaceuticals UK Ltd., UK, provided medical editorial assistance.
Legal entity responsible for the study
Novartis Pharmaceuticals.
Funding
Novartis Pharmaceuticals.
Disclosure
B. Melosky: Financial Interests, Personal, Advisory Board: Novartis; Financial Interests, Personal, Advisory Board: Merck; Financial Interests, Personal, Advisory Board: AstraZeneca; Financial Interests, Personal, Advisory Board: Bristol Myers Squibb; Financial Interests, Personal, Advisory Board: Lilly; Financial Interests, Personal, Advisory Board: Pfizer. S.M. Knoll: Financial Interests, Personal, Full or part-time Employment: Novartis. I.S. Souef: Financial Interests, Personal, Full or part-time Employment: Novartis. W. Wu: Financial Interests, Institutional, Other, Research support: Genesis Research. W.C. Rhodes: Financial Interests, Institutional, Other, Contract for research: Genesis Research. J. Martinalbo: Financial Interests, Personal, Full or part-time Employment: Novartis. F. Ye: Financial Interests, Personal, Full or part-time Employment: Novartis.