Abstract 3882: Exposure-toxicity analysis of unbound regorafenib and its active metabolites by dose escalation strategy with low starting dose in patients with colorectal cancer

Background: As regorafenib (REG) and its active metabolites are extensively bound to serum proteins, their unbound exposure are considered to be more relevant to pharmacological and toxicological responses than total (unbound plus bound) exposure. We thus investigated the relationships between toxicity and serum unbound concentrations of REG and active metabolites M-2 and M-5 in patients with colorectal cancer.Patients and Methods: REG was administered orally once daily for the first 21 days of each 28-day cycle. The dose was started at 120 mg/day, and escalated to standard dose of 160 mg/day on day 15 of the first cycle in patients who experienced neither hand-foot skin reaction nor grade ≥2 REG-related adverse events (AEs) without dose reduction or interruption until day 14. Serum concentrations of REG, M-2 and M-5 were evaluated on days 8, 15, and 22 of the first cycle in patients without interruption of REG treatment until each sampling point. Unbound fraction was obtained by equilibrium dialysis, and concentrations were determined by the UPLC-MS/MS method. AEs were graded according to CTCAE ver. 4.0.Results: Among all 68 enrolled patients, 57 patients on day 8, 42 patients on day 15, and 23 patients on day 22 were assessable. On day 8, the median total concentrations of REG, M-2 and M-5 in serum were 6801 nM (range, 2487-17621), 2596 nM (321-12107), and 834 nM (49-12315), respectively. Unbound fraction of REG, M-2 and M-5 varied from 0.019-0.441 %, 0.000-0.477 % and 0.041-2.381 %, respectively, showing no association with levels of serum albumin (28-50 mg/mL) which is a major binding protein. Serum concentrations of total REG or sum of total REG, M-2 and M-5 (total SUM) on day 8 were not related with maximum grade (grade ≤2 vs. grade ≥3) of AEs (excluding laboratory abnormalities) in the first cycle. On the other hand, higher serum concentrations of unbound REG on day 8 tended to associate with severity of maximum grade of AEs in the first cycle (P=0.0711), and concentrations of sum of unbound REG, M-2 and M-5 (unbound SUM) on day 8 were significantly correlated with maximum grade of AEs in the first cycle (P=0.0345). In addition, concentrations of total REG and unbound REG on day 8 in six patients whose dose were escalated to 160 mg/day on day 15 were significantly lower than those in 50 patients whose dose were not escalated (total, 3978 vs. 7005 nM in median, P=0.0270; unbound, 2834 vs. 7244 pM in median, P=0.0455). There were also significant association between concentrations of unbound REG or unbound SUM on day 15 and severity of REG-related AEs (grade 1 vs. grade ≥2) on day 15 (P=0.0495, P=0.0126, respectively).Conclusion: Unbound exposure was well correlated with toxicity in patients treated with REG. Exposure of sum of REG, M-2 and M-5 was associated with toxicity more than exposure of REG alone. Serum albumin levels didn't affect unbound fraction of REG, M-2 and M-5.Citation Format: Hironaga Satake, Takeshi Suzuki, Chiyo K. Imamura, Yasutaka Sukawa, Toshiki Masuishi, Yosuke Kumekawa, Shinsuke Funakoshi, Masahito Kotaka, Yoshiki Horie, Sadayuki Kawai, Hiroyuki Okuda, Tetsuji Terazawa, Chihiro Kondoh, Ken Kato, Kenichi Yoshimura, Hideki Ishikawa, Yasuo Hamamoto, Narikazu Boku, Hiromasa Takaishi, Takanori Kanai. Exposure-toxicity analysis of unbound regorafenib and its active metabolites by dose escalation strategy with low starting dose in patients with colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3882.