Plasma concentrations of dabrafenib and trametinib (PCD/T) monitoring in advanced BRAFV600mut melanoma patients.

9541 Background: Dabrafenib (D) and trametinib (T) are highly ative in BRAFV600mut melanoma pts. One previous study suggested high PCD to be associated with adverse effects (AE) but not with tumor response. We investigated the relationship between PCD/T and tumor control or AE. Methods: PCD/T were evaluated with high-performance liquid chromatography/mass spectrometry in D+T-treated metastatic melanoma patients. We analyzed results at steady state (≥2 d for D, 24 d for T after introduction or dose modification) and far from peak concentrations (8 to 14.5 h for D and 20-28 h for T). We collected data of tumor assessments (RECIST 1.1) prospectively stored in our database and AE (CTCAE 4.0) blinded to PCD/T results. Each AE was associated with the closest sample harvested to the beginning of AE, or in the absence of AE with the highest PCD/T level for each patient. Results: We analyzed 75 D and 58 T assays from 36 pts (19M/17F), treated with D+T for metastatic melanoma (Stage IV: N = 35), mostly in first line (69.4%). Initial D dose was 300 mg/d and 2 mg/d for T, reduced in 10 patients (27.7%) for AE: to 30% of D (N = 8) and 25% of T (N = 8). High interindividual variability of PCD (range: 4-945ng/mL, median 70.0) and of PCT (5-25ng/mL, median 8.6) was observed. No differences between mean PCD/T at the time of evaluations showing progressive disease (PD) compared to those without PD pts (146.6±111.6 and 9.3±2.1) and pts with complete (N = 11), partial (N = 1) or stable response (N = 1) (160.6±127.9, P = 0.81 and 10.6ng/mL±2.6, P = 0.29) were observed. No significant relationship was shown between PCD/T and body mass index (r = 0.22 and -0.31), age (p = 0.19 and 0.26), or between PCD/T and D (p = 0.11) or T (p = 0.17) doses, neither between elevated mean PCD/T and any most common AE. Conclusions: This study shows a high interindividual variability but failed to show a relationship between PCD or T and tumor response nor AE. One limit is we did not explore PC of D active metabolites (hydroxy-and desmethylD). It has been shown that there is an auto-induction of D; T inhibits P-gp, a D substrate, suggesting synergistic pharmacological interactions. Thus, D pharmacokinetic seems to be too complex to be easily monitored.