DeNardo G L, Lamborn K R, Goldstein D S, Kroger L A, DeNardo S J
Department of Internal Medicine, University of California Davis Medical Center, Sacramento, USA.
Cancer. 1997 Dec 15;80(12 Suppl):2706-11. doi: 10.1002/(sici)1097-0142(19971215)80:12+<2706::aid-cncr49>3.3.co;2-m.
Because most patients with advanced non-Hodgkin's lymphoma (NHL) and chronic lymphocytic leukemia (CLL) respond to radioimmunotherapy (RIT), the study was designed to evaluate the relationship between response and survival in patients treated with radiolabeled Lym-1.
Fifty-seven patients with NHL (52) or CLL (5) were treated with radiolabeled Lym-1 antibody between 1985 and 1994. Influence of response to treatment on survival was examined directly and was also adjusted for other factors previously found to be significant predictors of survival. A multivariate model that was based on baseline Karnofsky performance status (KPS) and serum lactic dehydrogenase (LDH) predicted response and survival and was used to define risk groups. Proportional-hazards, Kaplan-Meier, and Landmark models were used to evaluate parameters for their ability to predict outcome.
By using a proportional-hazards model with response as a time-dependent variable, overall response (P < 0.001) and complete response (P = 0.006) were predictive of increased survival in univariate analyses. Overall response continued to be significant (P = 0.02) in multivariate analyses, even when risk groups that also predicted survival were included. Median survival of responders was 84 weeks, whereas that of nonresponders was 22 weeks when the Landmark method, based on response status at 16 weeks from start of therapy, was used to generate survival curves.
Response to treatment with radiolabeled Lym-1 was associated with increased survival, even when adjusted for baseline clinical parameters that also predicted for survival. The results provide the first unbiased evidence for survival effects of RIT in patients.
由于大多数晚期非霍奇金淋巴瘤（NHL）和慢性淋巴细胞白血病（CLL）患者对放射免疫疗法（RIT）有反应，本研究旨在评估接受放射性标记的Lym-1治疗的患者的反应与生存之间的关系。
1985年至1994年间，57例NHL患者（52例）或CLL患者（5例）接受了放射性标记的Lym-1抗体治疗。直接检查了治疗反应对生存的影响，并针对先前发现的其他生存显著预测因素进行了调整。基于基线卡诺夫斯基表现状态（KPS）和血清乳酸脱氢酶（LDH）的多变量模型预测了反应和生存情况，并用于定义风险组。使用比例风险模型、Kaplan-Meier模型和地标模型评估参数预测结果的能力。
在单变量分析中，使用以反应作为时间依赖性变量的比例风险模型，总体反应（P < 0.001）和完全反应（P = 0.006）可预测生存期延长。在多变量分析中，即使纳入了也能预测生存的风险组，总体反应仍然具有显著性（P = 0.02）。当地标法基于治疗开始后16周的反应状态用于生成生存曲线时，反应者的中位生存期为84周，而非反应者为22周。
即使针对也能预测生存的基线临床参数进行调整后，对放射性标记的Lym-1治疗的反应仍与生存期延长相关。这些结果为RIT对患者的生存影响提供了首个无偏倚证据。
