查看2014.Doxorubicin.HSA的源代码

=== 统计分析 (Statistical analysis) ===

使用分类变量（例如品种、性别、转移的存在和分期程序）的 Fischer 精确检验和连续变量（例如体重、年龄）的不配对 t 检验进行组比较（DOX 治疗组与未治疗组）。

PFS 和 MST 使用 Kaplan-Meier 乘积极限法计算。在DOX治疗组中，检查与狗的PFS和OS预后因素相关的变量包括体重，诊断时是否存在转移，诊断时贫血，诊断时血小板减少症，诊断时心电图异常，诊断时填塞，肿瘤位置，肿瘤大小（平均值和中位数），进行心包穿刺术的次数，DOX剂量（30mg m-2 vs 1 mg kg-1）， DOX 时间表（每 2 周一次与每 3 周一次）、同时使用节律化疗、使用抢救疗法（仅适用于 OS）和对治疗的反应。使用两个数据集的对数秩 （Mantel-Cox） 检验和趋势的对数秩检验（当输入两个以上数据集时）比较生成的生存曲线，并报告双尾 P 值。P 值为 ≤0.05 被认为是显著的。所有统计分析均使用商业软件包（Prism v5.0，GraphPad Software，LaJolla，CA，USA）进行。

Group comparisons (DOX-treated versus untreated groups) were made using Fischer’s exact test for categorical variables (e.g. breed, sex, presence of metastasis and staging procedures) and the unpaired t-test for continuous variables (e.g. weight, age).

The PFS and MST were calculated using the Kaplan–Meier product-limit method. Variables examined for association as prognostic factors for PFS and OS for dogs in the DOX-treated group included body weight, presence of metastasis at diagnosis, anaemia at diagnosis, thrombocytopenia at diagnosis, ECG abnormalities at diagnosis, tamponade at diagnosis, tumour location, tumour size (mean and median), number of pericardiocenteses performed, DOX dose (30 mg m−2 versus 1 mg kg−1), DOX schedule (every 2 weeks versus every 3 weeks), concurrent use of metronomic chemotherapy, use of rescue therapy (for OS only) and response to treatment. Generated survival curves were compared using the log-rank (Mantel–Cox) test for two data sets and the log-rank test for trend (when more than two datasets were entered) with two-tailed P values reported. A P value of ≤0.05 was considered significant. All statistical analyses were performed using a commercial software package (Prism v5.0, GraphPad Software, LaJolla, CA, USA).

<br>