Department of Dermatology, AP-HP, Ambroise Paré Hospital, Boulogne-Billancourt University of Versailles-Saint-Quentin-en-Yvelines, Research Unit EA 4340 'Biomarkers in Cancerology and in Hemato-oncology', Boulogne-Billancourt.
Department of Toxicology, AP-HP, Raymond Poincaré Hospital, Boulogne-Billancourt University of Versailles-Saint-Quentin-en-Yvelines, Boulogne-Billancourt.
Ann Oncol. 2015 Jul;26(7):1470-5. doi: 10.1093/annonc/mdv189. Epub 2015 Apr 21.
Vemurafenib improves survival in advanced BRAFV600(mut) melanoma patients, but tolerance is often poor and resistance frequently occurs, without predictive factor. Our aim was to investigate for the first time a relationship between plasma vemurafenib concentration (PVC) and efficacy or tolerance.
Plasma samples from unresectable metastatic BRAFV600(mut) melanoma patients treated with vemurafenib monotherapy were prospectively collected at each tumour response evaluation (RECIST 1.1) or when adverse event occurred (CTCAE 4.0). PVC was measured with liquid chromatography-tandem mass spectrometry. Herein, we report on PVC at steady state (≥14 days after vemurafenib introduction or dose modification). Samples collected after first melanoma progression were excluded from the response analysis. All samples were analysed in the tolerance analysis. We kept the closest collected sample from the onset of each adverse effect or the one with the highest PVC in the absence of this adverse effect. Comparisons of means (Student's t-tests and Wilcoxon rank sum tests) and of frequencies (χ(2) tests) were carried out. A logistic regression analysis identified predictors of progression.
We included 105 plasma samples in 23 patients (10M/13F). Initial vemurafenib dose was 960 mg b.i.d., reduced by 25% (8 patients) or 50% (2 patients) for intolerance in 10 patients (44%). PVC displayed high inter-individual variability (13.0-109.8 µg/ml, median 54.0). Mean PVC was lower at time of first progression (38.8 ± 19.7 µg/ml) than mean PVC found when tumour was stable or in partial or complete response (56.4 ± 21.0 µg/ml, P = 0.013, 21 patients). Logistic regression revealed that having a low PVC (P = 0.01) or brain metastasis (P = 0.01) were both significantly and independently associated with tumour progression. High PVC was not statistically significantly associated with the occurrence of adverse effects.
PVC at steady state is highly variable and low PVC was associated with tumour progression, suggesting a new path to melanoma resistance to vemurafenib.
维莫非尼可改善晚期 BRAFV600（mut）黑色素瘤患者的生存，但通常耐受性差，且易发生耐药，目前尚无预测因素。我们的目的是首次探讨血浆维莫非尼浓度（PVC）与疗效或耐受性之间的关系。
前瞻性收集了接受维莫非尼单药治疗的不可切除转移性 BRAFV600（mut）黑色素瘤患者的血浆样本，在每次肿瘤反应评估（RECIST 1.1）或出现不良事件时（CTCAE 4.0）采集。采用液相色谱-串联质谱法测量 PVC。在此，我们报告了稳态时（维莫非尼起始或剂量调整后≥14 天）的 PVC。将首次黑色素瘤进展后采集的样本排除在反应分析之外。所有样本均纳入耐受性分析。我们保留了每个不良事件开始时或在无不良事件时采集的最接近的样本或 PVC 最高的样本。采用学生 t 检验和 Wilcoxon 秩和检验比较均值，采用卡方检验比较频率。通过逻辑回归分析确定进展的预测因子。
我们纳入了 23 例患者的 105 个血浆样本（10 例男性，13 例女性）。初始维莫非尼剂量为 960mg，bid，10 例患者（44%）因不耐受减少 25%（8 例）或 50%（2 例）。个体间 PVC 差异较大（13.0-109.8μg/ml，中位数 54.0）。首次进展时的平均 PVC（38.8±19.7μg/ml）低于肿瘤稳定或部分或完全缓解时的平均 PVC（56.4±21.0μg/ml，P=0.013，21 例）。逻辑回归显示，低 PVC（P=0.01）或脑转移（P=0.01）均与肿瘤进展显著相关且独立相关。高 PVC 与不良事件的发生无统计学显著相关性。
稳态时的 PVC 差异较大，低 PVC 与肿瘤进展相关，提示黑色素瘤对维莫非尼耐药的新途径。
