Ferrer Florent, Chauvin Jonathan, DeVictor Bénédicte, Lacarelle Bruno, Deville Jean-Laurent, Ciccolini Joseph
SMARTc Unit, Centre de Recherche en Cancérologie de Marseille, Inserm U1068 Aix Marseille Université, 13385 Marseille, France.
Laboratoire de Pharmacocinétique et Toxicologie, La Timone University Hospital of Marseille, 13385 Marseille, France.
Pharmaceuticals (Basel). 2021 May 24;14(6):494. doi: 10.3390/ph14060494.
Different target exposures with sunitinib have been proposed in metastatic renal cell carcinoma (mRCC) patients, such as trough concentrations or AUCs. However, most of the time, rather than therapeutic drug monitoring (TDM), clinical evidence is preferred to tailor dosing, i.e., by reducing the dose when treatment-related toxicities show, or increasing dosing if no signs of efficacy are observed. Here, we compared such empirical dose adjustment of sunitinib in mRCC patients, with the parallel dosing proposals of a PK/PD model with TDM support. In 31 evaluable patients treated with sunitinib, 53.8% had an empirical change in dosing after treatment started (i.e., 46.2% decrease in dosing, 7.6% increase in dosing). Clinical benefit was observed in 54.1% patients, including 8.3% with complete response. Overall, 58.1% of patients experienced treatment discontinuation eventually, either because of toxicities or progressive disease. When choosing 50-100 ng/mL trough concentrations as a target exposure (i.e., sunitinib + active metabolite N-desethyl sunitinib), 45% patients were adequately exposed. When considering 1200-2150 ng/mL.h as a target AUC (i.e., sunitinib + active metabolite N-desethyl sunitinib), only 26% patients were in the desired therapeutic window. TDM with retrospective PK/PD modeling would have suggested decreasing sunitinib dosing in a much larger number of patients as compared with empirical dose adjustment. Indeed, when using target trough concentrations, the model proposed reducing dosing for 61% patients, and up to 84% patients based upon target AUC. Conversely, the model proposed increasing dosing in 9.7% of patients when using target trough concentrations and in 6.5% patients when using target AUC. Overall, TDM with adaptive dosing would have led to tailoring sunitinib dosing in a larger number of patients (i.e., 53.8% vs. 71-91%, depending on the chosen metrics for target exposure) than a clinical-based decision. Interestingly, sunitinib dosing was empirically reduced in 41% patients who displayed early-onset severe toxicities, whereas model-based recommendations would have immediately proposed to reduce dosing in more than 80% of those patients. This observation suggests that early treatment-related toxicities could have been partly avoided using prospective PK/PD modeling with adaptive dosing. Conversely, the possible impact of model-based adapted dosing on efficacy could not be fully evaluated because no clear relationship was found between baseline exposure levels and sunitinib efficacy measured at 3 months.
对于转移性肾细胞癌（mRCC）患者，已提出舒尼替尼的不同目标暴露量，如谷浓度或曲线下面积（AUC）。然而，大多数情况下，与治疗药物监测（TDM）相比，临床证据更适合用于调整剂量，即当出现治疗相关毒性时降低剂量，或在未观察到疗效迹象时增加剂量。在此，我们将mRCC患者中舒尼替尼的这种经验性剂量调整与有TDM支持的药代动力学/药效学（PK/PD）模型的平行剂量建议进行了比较。在31例接受舒尼替尼治疗的可评估患者中，53.8%在治疗开始后有剂量的经验性改变（即46.2%剂量降低，7.6%剂量增加）。54.1%的患者观察到临床获益，包括8.3%完全缓解。总体而言，最终58.1%的患者因毒性或疾病进展而停药。当选择50 - 100 ng/mL的谷浓度作为目标暴露量（即舒尼替尼+活性代谢物N - 去乙基舒尼替尼）时，45%的患者暴露充分。当将1200 - 2150 ng/mL·h作为目标AUC（即舒尼替尼+活性代谢物N - 去乙基舒尼替尼）时，只有26%的患者处于期望的治疗窗内。与经验性剂量调整相比，采用回顾性PK/PD建模的TDM会建议在更多患者中降低舒尼替尼剂量。实际上，当使用目标谷浓度时，模型建议61%的患者降低剂量，基于目标AUC时高达84%的患者降低剂量。相反，当使用目标谷浓度时，模型建议9.7%的患者增加剂量，使用目标AUC时为6.5%的患者增加剂量。总体而言，与基于临床的决策相比，采用适应性给药的TDM会使更多患者的舒尼替尼剂量得到调整（即53.8%对71 - 91%，取决于所选的目标暴露量指标）。有趣的是，41%出现早期严重毒性的患者经验性降低了舒尼替尼剂量，而基于模型的建议会立即建议在超过80%的此类患者中降低剂量。这一观察结果表明，使用前瞻性PK/PD建模和适应性给药可能部分避免早期治疗相关毒性。相反，基于模型的适应性给药对疗效的可能影响无法完全评估，因为在基线暴露水平与3个月时测量的舒尼替尼疗效之间未发现明确关系。
