Small Cell Lung Cancer

These data are described in (Ying et al., 1995) and originally collected in (Maksymuik et al., 1994). The patients were randomly assigned to one of two treatments referred to as Arm A and Arm B. Arm A patients received cisplatin (P) followed by etoposide (E), while Arm B patients received (E) followed by (P). There were a total of 62 patients in Arm A with 15 right censored survival times, while Arm B consisted of 59 patients with 8 right censored survival times.<br><br>The variables recorded in the dataset(s) are:<br>- age: age of the patient upon entry (in years)<br>- time: survival time of the patient (in days)<br>- cens: indicator for survival time - 0 if observed, 1 if right censored<br><br>We analyze these data using a Dirichlet Process Mixture Model with a Gamma kernel distribution (Poynor and Kottas, 2017). The primary emphasis of our work is to obtain inference for the mean residual life function. Our code is shared through github (see link in references). <br><br>Journal References:<br><i><br></i><i>Jett, R, J D Earle, J Q Su, F A Diegert, J A Mailliard, C G Kardinal, J E Krook, M H Veeder, and M Wiesenfeld. Sequencing and schedule effects of cisplatin plus etoposide in small-cell lung cancer: results of a North Central Cancer Treatment Group randomized clinical trial</i>. A W Maksymiuk, J Journal of Clinical Oncology 1994 12:1, 70-76<br><br>Poynor V. and Kottas A. (2017). <i>Nonparametric Bayesian Inference for Mean Residual Life Functions in Survival Analysis</i>. Submitted to Biostatistics. arXiv:1411.7481 [stat.ME]. <br><br><i>Ying, Z., S. H. Jung and L. J. Wei. Survival Analysis with Median Regression Models. </i>Journal of the American Statistical Association, Vol. 90, No. 429 (Mar., 1995), pp.178-184<br> <br><br>

本数据集的相关分析描述见于（Ying等，1995），其原始数据采集自（Maksymuik等，1994）的研究。所有患者被随机分配至A组或B组两种治疗方案中。A组患者接受顺铂（cisplatin，P）序贯依托泊苷（etoposide，E）治疗，而B组患者则接受依托泊苷（E）序贯顺铂（P）治疗。其中A组共纳入62例患者，包含15例右删失生存时间（right censored survival times）数据；B组共纳入59例患者，包含8例右删失生存时间数据。

数据集所记录的变量如下：
- 年龄（age）：患者入组时的年龄（单位：年）
- 生存时间（time）：患者的生存时长（单位：天）
- 删失指示符（cens）：生存时间的指示变量——0代表生存时间已被观测到，1代表为右删失生存时间

本研究采用带有Gamma核分布（Gamma kernel distribution）的狄利克雷过程混合模型（Dirichlet Process Mixture Model）对上述数据进行分析（Poynor与Kottas，2017）。本研究的核心重点为对平均剩余寿命函数（mean residual life function）开展统计推断。本研究的代码已通过GitHub（github）开源共享（详见参考文献中的链接）。

### 期刊参考文献
1. Jett R, Earle JD, Su JQ, Diegert FA, Mailliard JA, Kardinal CG, Krook JE, Veeder MH, Wiesenfeld M, Maksymiuk AW. 顺铂联合依托泊苷治疗小细胞肺癌的给药顺序与方案时序效应：北美中部癌症治疗组随机临床试验结果[J]. 《临床肿瘤学杂志》（Journal of Clinical Oncology）, 1994, 12(1): 70-76.
2. Poynor V, Kottas A. 生存分析中平均剩余寿命函数的非参数贝叶斯推断[J/待投稿]. 投稿至《Biostatistics》期刊. arXiv预印本编号：1411.7481 [stat.ME].
3. Ying Z, Jung SH, Wei LJ. 基于中位数回归模型的生存分析[J]. 《美国统计协会会刊》（Journal of the American Statistical Association）, 1995年3月, 第90卷第429期, 第178-184页.