肝细胞癌
阿替唑单抗
贝伐单抗
胃肠病学
队列
医学
内科学
肿瘤科
肝硬化
单变量分析
实体瘤疗效评价标准
无进展生存期
多元分析
总体生存率
癌症
进行性疾病
无容量
化疗
免疫疗法
作者
Claudia Campani,Jessica Bamba‐Funck,Bertille Campion,Sabrina Sidali,Lorraine Blaise,Nathalie Ganne‐Carrié,Alix Demory,Olivier Sutter,Edouard Larrey,Manon Evain,Haroun Ghannouchi,Mathilde Wagner,Fabio Marra,Angéla Sutton,Manon Allaire,Jean‐Charles Nault
摘要
The combination of atezolizumab and bevacizumab (AtezoBev) is the current first-line treatment for patients with hepatocellular carcinoma (HCC). Our aim was to evaluate the prognostic role of alpha-foetoprotein (AFP) early response and its combination with albumin-bilirubin (ALBI) in these patients.Patients with HCC under AtezoBev with AFP > 20 ng/ml were included in three centres. The optimal threshold of AFP variation after 3 weeks of treatment was identified for overall survival (OS) and radiological response (RR) using RECIST 1.1 and mRECIST and its ability to predict progression-free survival (PFS) and OS was tested using univariate and multivariate analysis in derivation and validation cohorts.Seventy-five patients with AFP values >20 ng/ml were included. Fifty-eight patients were male with a median age of 63.5 years; 73% had cirrhosis and HCC stage was classified as BCLC B (18.7%) or C (81.3%). In the derivation cohort (n = 38), a decline in AFP ≥ 20% at 3 weeks (AFP early response) was associated with RR using mRECIST criteria (OR: 13.09 95% CI: 1.44-19.34 p = .02), PFS (HR: 0.42; 95% CI: 0.19-0.93, p = .03) and OS (HR: 0.35; 95% CI: 0.15-0.83, p = .01). AFP early response was confirmed as predictor of RR (p = .02 for mRECIST) and OS (p = .03) in the validation cohort (n= 37). In the whole cohort, the combination of ALBI and AFP early response was significantly associated with OS (p = .046) and PFS (p = .012) with a poor prognosis in patients belonging to the ALBI2-AFP non-responders class.AFP early response at 3 weeks predicts oncological outcomes in HCC patients treated with AtezoBev and combination with ALBI grade refines prognostic discrimination.
科研通智能强力驱动
Strongly Powered by AbleSci AI