Telford Jennifer J, Saltzman John R, Kuntz Karen M, Syngal Sapna
Division of Gastroenterology, Brigham and Women's Hospital, Boston, MA, USA.
J Natl Cancer Inst. 2004 Feb 4;96(3):191-201. doi: 10.1093/jnci/djh026.
Although radical resection and postoperative chemoradiation have been the standard therapy for patients with rectal cancer, preoperative staging by local imaging and chemoradiation are widely used. We used a decision analysis to compare the two strategies for rectal cancer management.
We developed a decision model to compare survival outcomes after postoperative chemoradiation versus preoperative staging and chemoradiation in patients aged 70 years with resectable rectal cancer. In the postoperative chemoradiation strategy, patients undergo radical resection and receive postoperative chemoradiation. In the preoperative staging and chemoradiation strategy, patients with locally advanced cancer receive preoperative chemoradiation and radical resection, whereas those with amenable localized tumors undergo local excision. The cohorts of patients were entered into a Markov model incorporating age-adjusted and disease-specific mortality. Outcomes were evaluated by modeling 5-year disease-specific survival for preoperative chemoradiation as less than, equal to, or greater than that of postoperative chemoradiation. Base-case probabilities were derived from published data; the Surveillance, Epidemiology, and End Results (SEER) Program database; and U.S. Life Tables. One-way and two-way sensitivity analyses were performed. The outcome measures were life expectancy and quality-adjusted life expectancy.
Life expectancy and quality-adjusted life expectancy were 9.72 and 8.72 years, respectively, in the postoperative chemoradiation strategy. In the preoperative staging and chemoradiation strategy, life expectancy was 9.36, 9.72, and 10.09 years and quality-adjusted life expectancy was 8.71, 9.04, and 9.37 years when 5-year disease-specific survival was less than, equal to, or greater than that of postoperative chemoradiation, respectively. The decision model was sensitive to differences in the long-term toxicity of pre- and postoperative chemoradiation. When the 5-year disease-specific survival for patients after pre- or postoperative chemoradiation was equal, the decision model was sensitive to surgical mortality and to the probability of residual lymph node disease after local excision.
If efficacy and toxicity after preoperative chemoradiation are equal to or better than that after postoperative chemoradiation in patients with locally advanced rectal cancer, then preoperative staging to select patients appropriate for preoperative chemoradiation is beneficial.
尽管根治性切除和术后放化疗一直是直肠癌患者的标准治疗方法，但通过局部影像学进行术前分期及放化疗也被广泛应用。我们采用决策分析来比较这两种直肠癌治疗策略。
我们建立了一个决策模型，以比较70岁可切除直肠癌患者术后放化疗与术前分期及放化疗后的生存结局。在术后放化疗策略中，患者接受根治性切除并接受术后放化疗。在术前分期及放化疗策略中，局部晚期癌症患者接受术前放化疗和根治性切除，而那些适合局部切除的局限性肿瘤患者则接受局部切除。将患者队列纳入一个包含年龄调整和疾病特异性死亡率的马尔可夫模型。通过模拟术前放化疗的5年疾病特异性生存率低于、等于或高于术后放化疗的情况来评估结局。基础概率来自已发表的数据、监测、流行病学和最终结果（SEER）计划数据库以及美国生命表。进行了单向和双向敏感性分析。结局指标为预期寿命和质量调整预期寿命。
术后放化疗策略的预期寿命和质量调整预期寿命分别为9.72年和8.72年。在术前分期及放化疗策略中，当5年疾病特异性生存率低于、等于或高于术后放化疗时，预期寿命分别为9.36年、9.72年和10.09年，质量调整预期寿命分别为8.71年、9.04年和9.37年。决策模型对术前和术后放化疗的长期毒性差异敏感。当术前或术后放化疗患者的5年疾病特异性生存率相等时，决策模型对手术死亡率和局部切除后残留淋巴结疾病的概率敏感。
对于局部晚期直肠癌患者，如果术前放化疗后的疗效和毒性等于或优于术后放化疗，那么进行术前分期以选择适合术前放化疗的患者是有益的。
