Doherty M K, Leung Y, Su J, Naik H, Patel D, Eng L, Kong Q Q, Mohsin F, Brown M C, Espin-Garcia O, Vennettilli A, Renouf D J, Faluyi O O, Knox J J, MacKay H, Wong R, Howell D, Mittmann N, Darling G E, Cella D, Xu W, Liu G
Division of Medical Oncology and Hematology, Princess Margaret Cancer Centre, Toronto, Ontario, Canada.
Division of Medical Oncology and Hematology, Sunnybrook Odette Cancer Centre, Toronto, Ontario, Canada.
Dis Esophagus. 2018 Dec 1;31(12). doi: 10.1093/dote/doy058.
Esophageal cancer and its treatment can cause serious morbidity/toxicity. These effects on health-related quality of life (HRQOL) can be measured using disease-specific scales such as FACT-E, generic scales such as EQ-5D-3L, or through symptoms. In a two-year cross-sectional study, we compared HRQOL across esophageal cancer patients treated in an ambulatory clinic and across multiple disease states, among patients with all stages of esophageal cancer. Consenting patients completed FACT-E, EQ-5D, a visual analog scale, and patient reported (PR)-ECOG. Symptom complexes were constructed from FACT-E domains. Responses were categorized by disease state: pre-, during, and post-treatment, surveillance, progression, and palliative chemotherapy. Spearman correlation and multivariable linear regression characterized these associations. In total, 199 patients completed 317 questionnaires. Mean FACT-E and subscale scores dropped from baseline through treatment and recovered during post-treatment surveillance (P < 0.001); EQ-5D health utility scores (HUS) displayed a similar pattern but with smaller differences (P = 0.07), and with evidence of ceiling effect. Among patients with stage II/III esophageal cancer, mean EQ-5D HUS varied across disease states (P < 0.001), along with FACT-E and subscales (P < 0.001). Among patients with advanced disease, there was no significant difference between baseline and on-treatment total scores, but improved esophageal cancer-specific scales were noted (P = 0.003). Strong correlation was observed between EQ-5D and FACT-E (R = 0.73), along with physical and functional subscales. In addition, the association between FACT-E and EQ-5D HUS was maintained in a multivariable model (P < 0.001). We interpret these results to suggest that in a real-world clinic setting, FACT-E, EQ-5D HUS, and symptoms were strongly correlated. Most HRQOL and symptom parameters suggested that patients had worse HRQOL and symptoms during curative therapy, but recovered well afterwards. In contrast, palliative chemotherapy had a neutral to positive impact on HRQOL/symptoms when compared to their baseline pre-treatment state.
食管癌及其治疗可导致严重的发病/毒性反应。这些对健康相关生活质量(HRQOL)的影响可使用特定疾病量表(如FACT-E)、通用量表(如EQ-5D-3L)或通过症状进行测量。在一项为期两年的横断面研究中,我们比较了在门诊接受治疗的食管癌患者以及处于多种疾病状态的食管癌各阶段患者的HRQOL。同意参与的患者完成了FACT-E、EQ-5D、视觉模拟量表以及患者报告的(PR)-ECOG。症状复合体由FACT-E领域构建。根据疾病状态对回答进行分类:治疗前、治疗期间、治疗后、监测、进展期以及姑息化疗。采用Spearman相关性分析和多变量线性回归来描述这些关联。总共199名患者完成了317份问卷。FACT-E总分及各子量表得分从基线到治疗期间下降,并在治疗后监测期间恢复(P < 0.001);EQ-5D健康效用得分(HUS)呈现类似模式,但差异较小(P = 0.07),且有天花板效应的证据。在II/III期食管癌患者中,EQ-5D HUS平均分在不同疾病状态下有所不同(P < 0.001),FACT-E及其子量表也是如此(P < 0.001)。在晚期疾病患者中,基线总分与治疗期间总分无显著差异,但食管癌特异性量表有所改善(P = 0.003)。观察到EQ-5D与FACT-E之间存在强相关性(R = 0.73),与身体和功能子量表也存在强相关性。此外,在多变量模型中,FACT-E与EQ-5D HUS之间的关联依然存在(P < 0.001)。我们对这些结果的解读表明,在现实世界的临床环境中,FACT-E、EQ-5D HUS和症状之间存在强相关性。大多数HRQOL和症状参数表明,患者在根治性治疗期间的HRQOL和症状较差,但之后恢复良好。相比之下,与治疗前基线状态相比,姑息化疗对HRQOL/症状有中性至积极的影响。