PDF(Pubmed)
Abstract：
Hepatocellular carcinoma is the most common type of liver cancer, accounting for 70% to 85% of individuals with primary liver cancer. Tamoxifen has been evaluated in randomised clinical trials in people with hepatocellular cancer. The reported results have been inconsistent.
To evaluate the benefits and harms of tamoxifen or tamoxifen plus any other anticancer drugs compared with no intervention, placebo, any type of standard care, or alternative treatment in adults with hepatocellular carcinoma, irrespective of sex, administered dose, type of formulation, and duration of treatment.
We searched the Cochrane Hepato-Biliary Group Controlled Trials Register, CENTRAL, MEDLINE, Embase, three other databases, and major trials registries, and handsearched reference lists up to 26 March 2024.
Parallel-group randomised clinical trials including adults (aged 18 years and above) diagnosed with advanced or unresectable hepatocellular carcinoma. Had we found cross-over trials, we would have included only the first trial phase. We did not consider data from quasi-randomised trials for analysis.
Our critical outcomes were all-cause mortality, serious adverse events, and health-related quality of life. Our important outcomes were disease progression, and adverse events considered non-serious.
We assessed risk of bias using the RoB 2 tool.
We used standard Cochrane methods and Review Manager. We meta-analysed the outcome data at the longest follow-up. We presented the results of dichotomous outcomes as risk ratios (RR) and continuous data as mean difference (MD), with 95% confidence intervals (CI) using the random-effects model. We summarised the certainty of evidence using GRADE.
We included 10 trials that randomised 1715 participants with advanced, unresectable, or terminal stage hepatocellular carcinoma. Six were single-centre trials conducted in Hong Kong, Italy, and Spain, while three were conducted as multicentre trials in single countries (France, Italy, and Spain), and one trial was conducted in nine countries in the Asia-Pacific region (Australia, Hong Kong, Indonesia, Malaysia, Myanmar, New Zealand, Singapore, South Korea, and Thailand). The experimental intervention was tamoxifen in all trials. The control interventions were no intervention (three trials), placebo (six trials), and symptomatic treatment (one trial). Co-interventions were best supportive care (three trials) and standard care (one trial). The remaining six trials did not provide this information. The number of participants in the trials ranged from 22 to 496 (median 99), mean age was 63.7 (standard deviation 4.18) years, and mean proportion of men was 74.7% (standard deviation 42%). Follow-up was three months to five years.
Ten trials evaluated oral tamoxifen at five different dosages (ranging from 20 mg per day to 120 mg per day). All trials investigated one or more of our outcomes. We performed meta-analyses when at least two trials assessed similar types of tamoxifen versus similar control interventions. Eight trials evaluated all-cause mortality at varied follow-up points. Tamoxifen versus the control interventions (i.e. no treatment, placebo, and symptomatic treatment) results in little to no difference in mortality between one and five years (RR 0.99, 95% CI 0.92 to 1.06; 8 trials, 1364 participants; low-certainty evidence). In total, 488/682 (71.5%) participants died in the tamoxifen groups versus 487/682 (71.4%) in the control groups. The separate analysis results for one, between two and three, and five years were comparable to the analysis result for all follow-up periods taken together. The evidence is very uncertain about the effect of tamoxifen versus no treatment on serious adverse events at one-year follow-up (RR 0.44, 95% CI 0.19 to 1.06; 1 trial, 36 participants; very low-certainty evidence). A total of 5/20 (25.0%) participants in the tamoxifen group versus 9/16 (56.3%) participants in the control group experienced serious adverse events. One trial measured health-related quality of life at baseline and at nine months\' follow-up, using the Spitzer Quality of Life Index. The evidence is very uncertain about the effect of tamoxifen versus no treatment on health-related quality of life (MD 0.03, 95% CI -0.45 to 0.51; 1 trial, 420 participants; very low-certainty evidence). A second trial found no appreciable difference in global health-related quality of life scores. No further data were provided. Tamoxifen versus control interventions (i.e. no treatment, placebo, or symptomatic treatment) results in little to no difference in disease progression between one and five years\' follow-up (RR 1.02, 95% CI 0.91 to 1.14; 4 trials, 720 participants; low-certainty evidence). A total of 191/358 (53.3%) participants in the tamoxifen group versus 198/362 (54.7%) participants in the control group had progression of hepatocellular carcinoma. Tamoxifen versus control interventions (i.e. no treatment or placebo) may have little to no effect on adverse events considered non-serious during treatment, but the evidence is very uncertain (RR 1.17, 95% CI 0.45 to 3.06; 4 trials, 462 participants; very low-certainty evidence). A total of 10/265 (3.8%) participants in the tamoxifen group versus 6/197 (3.0%) participants in the control group had adverse events considered non-serious. We identified no trials with participants diagnosed with early stages of hepatocellular carcinoma. We identified no ongoing trials.
Based on the low- and very low-certainty evidence, the effects of tamoxifen on all-cause mortality, disease progression, serious adverse events, health-related quality of life, and adverse events considered non-serious in adults with advanced, unresectable, or terminal stage hepatocellular carcinoma when compared with no intervention, placebo, or symptomatic treatment could not be established. Our findings are mostly based on trials at high risk of bias with insufficient power (fewer than 100 participants), and a lack of trial data on clinically important outcomes. Therefore, firm conclusions cannot be drawn. Trials comparing tamoxifen administered with any other anticancer drug versus standard care, usual care, or alternative treatment as control interventions were lacking. Evidence on the benefits and harms of tamoxifen in participants at the early stages of hepatocellular carcinoma was also lacking.
This Cochrane review had no dedicated funding.
Protocol available via DOI: 10.1002/14651858.CD014869.
To evaluate the benefits and harms of tamoxifen or tamoxifen plus any other anticancer drugs compared with no intervention, placebo, any type of standard care, or alternative treatment in adults with hepatocellular carcinoma, irrespective of sex, administered dose, type of formulation, and duration of treatment.
We searched the Cochrane Hepato-Biliary Group Controlled Trials Register, CENTRAL, MEDLINE, Embase, three other databases, and major trials registries, and handsearched reference lists up to 26 March 2024.
Parallel-group randomised clinical trials including adults (aged 18 years and above) diagnosed with advanced or unresectable hepatocellular carcinoma. Had we found cross-over trials, we would have included only the first trial phase. We did not consider data from quasi-randomised trials for analysis.
Our critical outcomes were all-cause mortality, serious adverse events, and health-related quality of life. Our important outcomes were disease progression, and adverse events considered non-serious.
We assessed risk of bias using the RoB 2 tool.
We used standard Cochrane methods and Review Manager. We meta-analysed the outcome data at the longest follow-up. We presented the results of dichotomous outcomes as risk ratios (RR) and continuous data as mean difference (MD), with 95% confidence intervals (CI) using the random-effects model. We summarised the certainty of evidence using GRADE.
We included 10 trials that randomised 1715 participants with advanced, unresectable, or terminal stage hepatocellular carcinoma. Six were single-centre trials conducted in Hong Kong, Italy, and Spain, while three were conducted as multicentre trials in single countries (France, Italy, and Spain), and one trial was conducted in nine countries in the Asia-Pacific region (Australia, Hong Kong, Indonesia, Malaysia, Myanmar, New Zealand, Singapore, South Korea, and Thailand). The experimental intervention was tamoxifen in all trials. The control interventions were no intervention (three trials), placebo (six trials), and symptomatic treatment (one trial). Co-interventions were best supportive care (three trials) and standard care (one trial). The remaining six trials did not provide this information. The number of participants in the trials ranged from 22 to 496 (median 99), mean age was 63.7 (standard deviation 4.18) years, and mean proportion of men was 74.7% (standard deviation 42%). Follow-up was three months to five years.
Ten trials evaluated oral tamoxifen at five different dosages (ranging from 20 mg per day to 120 mg per day). All trials investigated one or more of our outcomes. We performed meta-analyses when at least two trials assessed similar types of tamoxifen versus similar control interventions. Eight trials evaluated all-cause mortality at varied follow-up points. Tamoxifen versus the control interventions (i.e. no treatment, placebo, and symptomatic treatment) results in little to no difference in mortality between one and five years (RR 0.99, 95% CI 0.92 to 1.06; 8 trials, 1364 participants; low-certainty evidence). In total, 488/682 (71.5%) participants died in the tamoxifen groups versus 487/682 (71.4%) in the control groups. The separate analysis results for one, between two and three, and five years were comparable to the analysis result for all follow-up periods taken together. The evidence is very uncertain about the effect of tamoxifen versus no treatment on serious adverse events at one-year follow-up (RR 0.44, 95% CI 0.19 to 1.06; 1 trial, 36 participants; very low-certainty evidence). A total of 5/20 (25.0%) participants in the tamoxifen group versus 9/16 (56.3%) participants in the control group experienced serious adverse events. One trial measured health-related quality of life at baseline and at nine months\' follow-up, using the Spitzer Quality of Life Index. The evidence is very uncertain about the effect of tamoxifen versus no treatment on health-related quality of life (MD 0.03, 95% CI -0.45 to 0.51; 1 trial, 420 participants; very low-certainty evidence). A second trial found no appreciable difference in global health-related quality of life scores. No further data were provided. Tamoxifen versus control interventions (i.e. no treatment, placebo, or symptomatic treatment) results in little to no difference in disease progression between one and five years\' follow-up (RR 1.02, 95% CI 0.91 to 1.14; 4 trials, 720 participants; low-certainty evidence). A total of 191/358 (53.3%) participants in the tamoxifen group versus 198/362 (54.7%) participants in the control group had progression of hepatocellular carcinoma. Tamoxifen versus control interventions (i.e. no treatment or placebo) may have little to no effect on adverse events considered non-serious during treatment, but the evidence is very uncertain (RR 1.17, 95% CI 0.45 to 3.06; 4 trials, 462 participants; very low-certainty evidence). A total of 10/265 (3.8%) participants in the tamoxifen group versus 6/197 (3.0%) participants in the control group had adverse events considered non-serious. We identified no trials with participants diagnosed with early stages of hepatocellular carcinoma. We identified no ongoing trials.
Based on the low- and very low-certainty evidence, the effects of tamoxifen on all-cause mortality, disease progression, serious adverse events, health-related quality of life, and adverse events considered non-serious in adults with advanced, unresectable, or terminal stage hepatocellular carcinoma when compared with no intervention, placebo, or symptomatic treatment could not be established. Our findings are mostly based on trials at high risk of bias with insufficient power (fewer than 100 participants), and a lack of trial data on clinically important outcomes. Therefore, firm conclusions cannot be drawn. Trials comparing tamoxifen administered with any other anticancer drug versus standard care, usual care, or alternative treatment as control interventions were lacking. Evidence on the benefits and harms of tamoxifen in participants at the early stages of hepatocellular carcinoma was also lacking.
This Cochrane review had no dedicated funding.
Protocol available via DOI: 10.1002/14651858.CD014869.
摘要：
背景：肝细胞癌是最常见的肝癌类型，占原发性肝癌个体的70%～85%。他莫昔芬已在肝细胞癌患者的随机临床试验中进行了评估。报告的结果不一致。
目的：评估他莫昔芬或他莫昔芬联合任何其他抗癌药物与无干预相比的益处和危害，安慰剂,任何类型的标准护理，或成人肝细胞癌的替代治疗，不论性别,给药剂量,配方类型,和治疗的持续时间。
方法：我们搜索了Cochrane肝胆组对照试验注册，中部,MEDLINE,Embase,另外三个数据库,和主要的审判登记处，和截至2024年3月26日的人工搜索参考清单。
方法：平行组随机临床试验，包括诊断为晚期或不可切除肝细胞癌的成年人（18岁及以上）。如果我们找到了交叉试验，我们只包括第一阶段的试验.我们没有考虑来自准随机试验的数据进行分析。
结果：我们的关键结果是全因死亡率，严重不良事件,和健康相关的生活质量。我们的重要结果是疾病进展，和认为非严重的不良事件。
■我们使用RoB2工具评估了偏差风险。
方法：我们使用了标准的Cochrane方法和ReviewManager。我们在最长的随访中对结果数据进行了荟萃分析。我们将二分结果的结果呈现为风险比(RR)，将连续数据呈现为平均差(MD)，使用随机效应模型的95%置信区间(CI)。我们使用GRADE总结了证据的确定性。
方法：我们纳入了10项试验，随机抽取了1715名晚期参与者，不可切除,或晚期肝细胞癌。六项是在香港进行的单中心试验,意大利,西班牙,而三项是在单一国家作为多中心试验进行的(法国，意大利,和西班牙），一项试验在亚太地区的九个国家进行(澳大利亚，香港,印度尼西亚,马来西亚,缅甸,新西兰,新加坡,韩国,和泰国）。在所有试验中，实验干预均为他莫昔芬。对照干预是无干预(三项试验)，安慰剂（6项试验），和对症治疗（一项试验）。共同干预措施是最佳支持治疗（三项试验）和标准治疗（一项试验）。其余六项试验没有提供这一信息。试验参与者的数量从22到496不等（中位数99），平均年龄为63.7(标准差4.18)岁，男性的平均比例为74.7%(标准差42%)。随访时间为3个月至5年。
结果：10项试验评估了5种不同剂量的他莫昔芬口服（每天20mg至每天120mg）。所有试验都调查了我们的一个或多个结果。当至少有两项试验评估了相似类型的他莫昔芬与相似的对照干预措施时，我们进行了荟萃分析。8项试验评估了不同随访点的全因死亡率。他莫昔芬与对照干预措施(即不治疗，安慰剂,和对症治疗）导致1至5年死亡率几乎没有差异（RR0.99，95％CI0.92至1.06；8项试验，1364名参与者；低确定性证据)。总的来说,他莫昔芬组488/682（71.5%）参与者死亡，对照组487/682（71.4%）。其中一个的单独分析结果，在两个和三个之间，5年与所有随访期的分析结果相媲美.在一年的随访中，关于他莫昔芬与未治疗对严重不良事件的影响的证据非常不确定（RR0.44，95％CI0.19至1.06；1项试验，36名参与者；非常低的确定性证据)。他莫昔芬组共有5/20（25.0％）的参与者与对照组的9/16（56.3％）的参与者发生严重不良事件。一项试验在基线和9个月随访时测量了与健康相关的生活质量，使用斯皮策生活质量指数。关于他莫昔芬与不治疗对健康相关生活质量的影响的证据非常不确定（MD0.03，95％CI-0.45至0.51；1项试验，420名参与者；非常低的确定性证据)。第二项试验发现，全球健康相关生活质量评分没有明显差异。没有提供进一步的数据。他莫昔芬与对照干预措施(即不治疗，安慰剂,或对症治疗）在1年和5年的随访中，疾病进展几乎没有差异（RR1.02，95％CI0.91至1.14；4项试验，720名参与者；低确定性证据)。他莫昔芬组的191/358（53.3％）参与者与对照组的198/362（54.7％）参与者的肝细胞癌进展。他莫昔芬与对照干预措施（即无治疗或安慰剂）可能对治疗期间认为非严重的不良事件几乎没有影响，但证据非常不确定(RR1.17，95%CI0.45至3.06；4项试验，462名参与者；非常低的确定性证据)。他莫昔芬组共有10/265（3.8％）的参与者与对照组的6/197（3.0％）的参与者发生了非严重不良事件。我们没有发现被诊断为早期肝细胞癌的参与者的试验。我们没有发现正在进行的试验。
结论：基于低确定性和极低确定性的证据，他莫昔芬对全因死亡率的影响，疾病进展,严重不良事件,与健康相关的生活质量，和不良事件被认为是非严重的成人晚期，不可切除,或终末期肝细胞癌与无干预相比，安慰剂,或无法建立对症治疗。我们的发现主要是基于偏倚风险较高但力量不足的试验（少于100名参与者），缺乏临床重要结局的试验数据。因此,无法得出确切的结论。将他莫昔芬与任何其他抗癌药物与标准治疗进行比较的试验，日常护理,或缺乏对照干预措施的替代治疗。在肝细胞癌早期阶段的参与者中，他莫昔芬的益处和危害的证据也缺乏。
背景：本次Cochrane审查没有专项资金。
背景：可通过DOI获得的协议：10.1002/14651858。CD014869。
目的：评估他莫昔芬或他莫昔芬联合任何其他抗癌药物与无干预相比的益处和危害，安慰剂,任何类型的标准护理，或成人肝细胞癌的替代治疗，不论性别,给药剂量,配方类型,和治疗的持续时间。
方法：我们搜索了Cochrane肝胆组对照试验注册，中部,MEDLINE,Embase,另外三个数据库,和主要的审判登记处，和截至2024年3月26日的人工搜索参考清单。
方法：平行组随机临床试验，包括诊断为晚期或不可切除肝细胞癌的成年人（18岁及以上）。如果我们找到了交叉试验，我们只包括第一阶段的试验.我们没有考虑来自准随机试验的数据进行分析。
结果：我们的关键结果是全因死亡率，严重不良事件,和健康相关的生活质量。我们的重要结果是疾病进展，和认为非严重的不良事件。
■我们使用RoB2工具评估了偏差风险。
方法：我们使用了标准的Cochrane方法和ReviewManager。我们在最长的随访中对结果数据进行了荟萃分析。我们将二分结果的结果呈现为风险比(RR)，将连续数据呈现为平均差(MD)，使用随机效应模型的95%置信区间(CI)。我们使用GRADE总结了证据的确定性。
方法：我们纳入了10项试验，随机抽取了1715名晚期参与者，不可切除,或晚期肝细胞癌。六项是在香港进行的单中心试验,意大利,西班牙,而三项是在单一国家作为多中心试验进行的(法国，意大利,和西班牙），一项试验在亚太地区的九个国家进行(澳大利亚，香港,印度尼西亚,马来西亚,缅甸,新西兰,新加坡,韩国,和泰国）。在所有试验中，实验干预均为他莫昔芬。对照干预是无干预(三项试验)，安慰剂（6项试验），和对症治疗（一项试验）。共同干预措施是最佳支持治疗（三项试验）和标准治疗（一项试验）。其余六项试验没有提供这一信息。试验参与者的数量从22到496不等（中位数99），平均年龄为63.7(标准差4.18)岁，男性的平均比例为74.7%(标准差42%)。随访时间为3个月至5年。
结果：10项试验评估了5种不同剂量的他莫昔芬口服（每天20mg至每天120mg）。所有试验都调查了我们的一个或多个结果。当至少有两项试验评估了相似类型的他莫昔芬与相似的对照干预措施时，我们进行了荟萃分析。8项试验评估了不同随访点的全因死亡率。他莫昔芬与对照干预措施(即不治疗，安慰剂,和对症治疗）导致1至5年死亡率几乎没有差异（RR0.99，95％CI0.92至1.06；8项试验，1364名参与者；低确定性证据)。总的来说,他莫昔芬组488/682（71.5%）参与者死亡，对照组487/682（71.4%）。其中一个的单独分析结果，在两个和三个之间，5年与所有随访期的分析结果相媲美.在一年的随访中，关于他莫昔芬与未治疗对严重不良事件的影响的证据非常不确定（RR0.44，95％CI0.19至1.06；1项试验，36名参与者；非常低的确定性证据)。他莫昔芬组共有5/20（25.0％）的参与者与对照组的9/16（56.3％）的参与者发生严重不良事件。一项试验在基线和9个月随访时测量了与健康相关的生活质量，使用斯皮策生活质量指数。关于他莫昔芬与不治疗对健康相关生活质量的影响的证据非常不确定（MD0.03，95％CI-0.45至0.51；1项试验，420名参与者；非常低的确定性证据)。第二项试验发现，全球健康相关生活质量评分没有明显差异。没有提供进一步的数据。他莫昔芬与对照干预措施(即不治疗，安慰剂,或对症治疗）在1年和5年的随访中，疾病进展几乎没有差异（RR1.02，95％CI0.91至1.14；4项试验，720名参与者；低确定性证据)。他莫昔芬组的191/358（53.3％）参与者与对照组的198/362（54.7％）参与者的肝细胞癌进展。他莫昔芬与对照干预措施（即无治疗或安慰剂）可能对治疗期间认为非严重的不良事件几乎没有影响，但证据非常不确定(RR1.17，95%CI0.45至3.06；4项试验，462名参与者；非常低的确定性证据)。他莫昔芬组共有10/265（3.8％）的参与者与对照组的6/197（3.0％）的参与者发生了非严重不良事件。我们没有发现被诊断为早期肝细胞癌的参与者的试验。我们没有发现正在进行的试验。
结论：基于低确定性和极低确定性的证据，他莫昔芬对全因死亡率的影响，疾病进展,严重不良事件,与健康相关的生活质量，和不良事件被认为是非严重的成人晚期，不可切除,或终末期肝细胞癌与无干预相比，安慰剂,或无法建立对症治疗。我们的发现主要是基于偏倚风险较高但力量不足的试验（少于100名参与者），缺乏临床重要结局的试验数据。因此,无法得出确切的结论。将他莫昔芬与任何其他抗癌药物与标准治疗进行比较的试验，日常护理,或缺乏对照干预措施的替代治疗。在肝细胞癌早期阶段的参与者中，他莫昔芬的益处和危害的证据也缺乏。
背景：本次Cochrane审查没有专项资金。
背景：可通过DOI获得的协议：10.1002/14651858。CD014869。
