The prognostic value of PET/CT metabolic parameters in diffuse large B-cell lymphoma
PET/CT代谢参数对弥漫性大B细胞淋巴瘤预后价值

摘要

目的 比较治疗前（基线）和化疗4疗程（中期）后18F-PET/CT各代谢参数的变化情况联合Deauville五分法对弥漫性大B细胞淋巴瘤预后评估的预测价值。方法 自2020年3月至2024年10月回顾性分析经病理证实为弥漫性大B细胞淋巴瘤的44例患者[其中男25例，女19例，中位年龄为59（17-83）岁]，患者治疗前和化疗4周期后均进行18F-PET/CT检查，收集两次检查的相关代谢参数，分别计算各代谢参数的变化值及其变化百分比，利用ROC曲线寻找最佳临界值，以不同代谢参数的界值进行分组，进行Kaplan-Meier生存分析和COX回归分析。化疗4周期以后，再次行18F-PET/CT检查，以Deauville五分法对每个患者进行评分，分别以1-3分为一组，4-5分为另一组，进行Kaplan-Meier生存分析和COX回归分析。采用卡方检验分析比较数据。结果 （1）44例患者中11例在随访中出现病情进展，进展组的△LLR%为50.00%（5.00%~86.00%），与未进展组84.00%（60.50%~92.50%）（z=-2.36，P＜0.05）相比，未进展组数值明显高于进展组。（2）△LLR%最佳的临界值分别为60.5%。（3）△LLR%≥60.5%与＜60.5%组间，Deauville评分≥4与＜4，2年PFS差异均存在统计学意义（χ²=5.529、7.729，均P＜0.05）。结论 △LLR%以及Deauville五分法对DLBCL的预后均具有预测价值，多因素分析变量中△LLR%为独立影响因素。△LLR%、Deauville五分法联合预测可以对患者预后起到更好的评价效能。

Abstract

Objective To compare the predictive value of changes in metabolic parameters of 18F-PET/CT before treatment (baseline) and after 4 cycles of chemotherapy (interim), combined with the Deauville five-point scale, for prognostic assessment in DLBCL.
Methods From March 2020 to October 2024, a retrospective analysis was conducted on 44 patients with pathologically confirmed diffuse large B-cell lymphoma [including 25 males and 19 females, with a median age of 59 (17-83) years]. All patients underwent 18F-PET/CT scans before treatment and after 4 cycles of chemotherapy. The relevant metabolic parameters from both scans were collected, and the absolute changes and percentage changes in each parameter were calculated. ROC curves were used to determine optimal cutoff values, and patients were grouped based on these thresholds for Kaplan-Meier survival analysis and COX regression analysis. After 4 cycles of chemotherapy, another 18F-PET/CT scan was performed, and each patient was scored using the Deauville five-point scale. Patients were divided into two groups: scores 1-3 in one group and scores 4-5 in another, followed by Kaplan-Meier survival analysis and COX regression analysis. The chi-square test was used for comparative data analysis.
Results (1) Among the 44 patients, 11 experienced disease progression during follow-up. The median △LLR% in the progression group was 50.00% (5.00%~86.00%), significantly lower than that in the non-progression group (84.00% [60.50%~92.50%]) (z=-2.36, P<0.05). (2) The optimal cutoff value for △LLR% was 60.5%. (3) Significant differences in 2-year PFS were observed between the △LLR% ≥60.5% and <60.5% groups, as well as between the Deauville score ≥4 and <4 groups (χ²=5.529, 7.729, both P<0.05).
Conclusion   Both △LLR% and the Deauville five-point scale have predictive value for the prognosis of DLBCL, with △LLR% being an independent influencing factor in multivariate analysis. The combination of △LLR% and the Deauville five-point scale can provide better prognostic evaluation for patients.