临床试验表明，对于晚期肝细胞癌患者降低甲胎蛋白(AFP)水平可能具有预后价值。然而，在实际世界中，AFP水平的变化与预后的关系尚不清楚。我们选取了2011年1月1日至2021年6月30日期间，首次使用酪氨酸激酶抑制剂治疗的晚期肝细胞癌患者，并分别测定了治疗前和治疗后（首次治疗开始8±2周后）的AFP值。我们将AFP水平的变化定义为与治疗前比降低≥20%或没有/低于20%。以治疗后AFP测量到患者死亡之间的时间为实际过程中的总生存时间和无进展生存时间（rwPFS, the first progression event or death）进行评估。同时，我们通过Cox比例风险模型，根据可能的混杂因素和基线AFP值进行调整，以估计调整风险比（aHR）。还评估了基线AFP和肝细胞癌风险因素对其的影响。共有533名患者参与到本研究中，其中，有AFP减少的患者基线AFP的中位数高于无/AFP减少的患者（N = 166，210 µg / L vs N = 367，150 µg / L）。与无AFP降低相比，AFP减少的患者死亡的风险减少了35%（aHR = 0.65；95% CI，0.52-0.81；中位数分别为10.3和5.9个月）。对于rwPFS，结果也类似（aHR = 0.66；95% CI，0.54-0.81；中位数分别为4.6和2.6个月）。与基线AFP≥400 µg / L或有乙肝病毒、丙肝病毒或酗酒史的患者相比，AFP降低与预后的改善有关。仅基线AFP和rwPFS之间的交互作用具有统计学意义。针对某些病因，治疗后AFP的变化对预后的影响可能比基线AFP更为重要。因此，未来的研究应分析治疗期间AFP的长期变化对预后的影响。在实际世界中，AFP变化对预后影响的价值尚不确定。本研究收集了一大型全国性晚期肝细胞癌患者队列的纵向数据，结果显示相对于基线水平，治疗后AFP水平降低≥20%可显著改善实际世界中的总生存时间和无进展生存时间，此关系在有HCC风险因素（例如乙肝病毒、酗酒等）或基线AFP值较高的患者中可能更为显著。© 2023 Flatiron Health 以及作者。Cancer发表在Wiley Periodicals LLC的期刊上，代表着美国癌症协会（American Cancer Society）。

Clinical trials suggest α-fetoprotein (AFP) reduction may be prognostic among patients with advanced hepatocellular carcinoma. However, the association of AFP reduction with outcomes in real-world settings is unclear.Patients with advanced hepatocellular carcinoma between January 1, 2011, and June 30, 2021, first-line tyrosine kinase inhibitor, and baseline and posttreatment AFP values (closest to 8 ± 2 weeks after first-line initiation) were included. AFP reduction was defined as ≥20% decrease from baseline vs <20% or no decrease. Real-world overall survival and progression-free survival (rwPFS) were defined as time from posttreatment AFP measurement to death, and the first progression event or death, respectively. Adjusted hazard ratios (aHRs) were estimated using Cox proportional hazards models adjusted for potential confounders and baseline AFP. Effect modification by baseline AFP and hepatocellular carcinoma risk factors was assessed.Among 533 patients, median baseline AFP was higher in those with AFP reduction than those without (N = 166, 210 µg/L vs N = 367, 150 µg/L). There was a 35% decrease in hazard of death for patients with reduction vs without (aHR = 0.65; 95% CI, 0.52-0.81; median, 10.3 vs 5.9 months). Results were similar for rwPFS (aHR = 0.66; 95% CI, 0.54-0.81; median, 4.6 vs 2.6 months). AFP reduction was associated with better outcomes among patients with baseline AFP ≥400 µg/L or with history of hepatitis B virus, hepatitis C virus, or alcohol use. Only the interaction between baseline AFP and reduction in association with rwPFS was statistically significant.For certain etiologies, posttreatment AFP change may be more important than baseline AFP for prognosis. Further work should characterize the prognostic implications of longitudinal AFP changes during treatment.The prognostic value of the change in α-fetoprotein (AFP) concentration after treatment initiation is less established, particularly in real-world settings. Longitudinal data from a large nationwide cohort of patients with advanced hepatocellular carcinoma (HCC) treated with first-line tyrosine kinase inhibitor in routine practice revealed that ≥20% reduction in posttreatment AFP levels was associated with better real-world overall survival and progression-free survival after adjusting for baseline AFP levels and other factors. The results also suggested that the associations may be stronger among patients with a history of HCC risk factors (e.g., hepatitis C virus, alcohol) or with higher baseline AFP levels.© 2023 Flatiron Health Inc and The Authors. Cancer published by Wiley Periodicals LLC on behalf of American Cancer Society.