Background: The CRAFITY score, integrating baseline C-reactive protein (CRP) and alpha-fetoprotein (AFP), has been validated as a prognostic biomarker in hepatocellular carcinoma (HCC) treated with immunotherapy, but many patients present with non-elevated AFP, limiting its accuracy. This study evaluated a composite model incorporating the CRAFITY score with AFP/PIVKA-II kinetic changes. Methods: We retrospectively enrolled 69 patients with unresectable HCC (BCLC stage B/C) receiving immunotherapy between September 2021 and June 2023. Baseline CRP, AFP, and PIVKA-II, as well as 4-week changes, were recorded. The CRAFITY-100 RULE combined CRAFITY (0–2) with AFP/PIVKA-II kinetics (0–3), yielding three risk levels (I–III). Clinical outcomes included objective response (OR) and overall survival (OS). Results: Of the cohort, 10 (14.5%), 29 (42%), and 30 (43.5%) patients had CRAFITY scores 0, 1, and 2, respectively, but this score did not clearly stratify OS (median 24, 12, and 15 months;p= 0.267). In contrast, the CRAFITY-100 RULE classified 5 (7.3%), 35 (50.7%), and 29 (42%) patients into levels I–III, respectively, with significantly different survival (median OS 24, 15, and 7 months;p= 0.048). OR rates were lowest at level III (17%). Time-dependent ROC analysis confirmed superior discrimination of CRAFITY-100 RULE over CRAFITY scores at 6 months (AUROC 0.673 vs. 0.604) and 12 months (0.732 vs. 0.656). Conclusions: The CRAFITY-100 RULE provided clearer stratification and higher discrimination. This simple model integrating baseline and dynamic biomarkers may assist clinical decision-making in unresectable HCC treated with immunotherapy.
背景：整合基线C反应蛋白（CRP）与甲胎蛋白（AFP）的CRAFITY评分已被证实可作为肝细胞癌（HCC）免疫治疗预后的生物标志物，但许多患者AFP水平未升高，限制了其预测准确性。本研究评估了结合CRAFITY评分与AFP/异常凝血酶原（PIVKA-II）动态变化的复合模型。方法：回顾性纳入2021年9月至2023年6月期间接受免疫治疗的69例不可切除HCC患者（BCLC分期B/C期）。记录基线CRP、AFP、PIVKA-II水平及治疗4周后的动态变化。CRAFITY-100规则将CRAFITY评分（0-2分）与AFP/PIVKA-II动态评分（0-3分）相结合，形成三个风险等级（I-III级）。临床结局指标包括客观缓解率（OR）和总生存期（OS）。结果：队列中CRAFITY评分0、1、2分患者分别为10例（14.5%）、29例（42%）和30例（43.5%），但该评分对OS的分层作用不显著（中位OS分别为24、12、15个月；p=0.267）。相比之下，CRAFITY-100规则将5例（7.3%）、35例（50.7%）和29例（42%）患者分别归入I-III级，各级生存期差异显著（中位OS分别为24、15、7个月；p=0.048）。III级患者的OR率最低（17%）。时间依赖性ROC分析证实，CRAFITY-100规则在6个月（AUROC 0.673 vs. 0.604）和12个月（0.732 vs. 0.656）的区分能力均优于CRAFITY评分。结论：CRAFITY-100规则提供了更清晰的风险分层和更高的区分效能。这一整合基线及动态生物标志物的简易模型，或有助于指导不可切除HCC免疫治疗的临床决策。
CRAFITY and AFP/PIVKA-II Kinetics Predict Prognosis in Hepatocellular Carcinoma on Immunotherapy
肿瘤(癌症)患者之家