Gurajala Ram, Partovi Sasan, DiFilippo Frank P, Li Xin, Coppa Christopher, Shah Shetal N, Karuppasamy Karunakaravel, Obuchowski Nancy, Fayazzadeh Ehsan, McLennan Gordon, Levitin Abraham
Section of Interventional Radiology, Imaging Institute, Cleveland Clinic Main Campus, Cleveland, Ohio, USA.
Department of Nuclear Medicine, Imaging Institute, Cleveland Clinic Main Campus, Cleveland, Ohio, USA.
J Gastrointest Oncol. 2024 Feb 29;15(1):356-367. doi: 10.21037/jgo-23-890. Epub 2024 Feb 26.
Y radioembolization is an established treatment modality for hepatic malignancies. Successful radioembolization requires optimal dose delivery to tumors while minimizing dosages to parenchyma. Post-treatment positron emission tomography (PET)/computed tomography (CT) dosimetry is the established benchmark, whereas PET/magnetic resonance (MR) is an emerging modality. The goal of this study was to assess the intermodality agreement between PET/MR and PET/CT Y dosimetry.
In this single-institution study, 18 patients (20 treatment sessions) with a primary or metastatic hepatic malignancy underwent both PET/MR and PET/CT after Y radioembolization. Patients were randomized to undergo one modality first, followed by the other. The region of interest was delineated using MR images and tumor and liver dosimetry was calculated. Intermodality agreement was assessed using the Bland-Altman method. A generalized linear model was used to assess the effect of baseline variables on intermodality dose differences.
PET/MR underestimated tumor and liver absorbed doses when compared to PET/CT by -3.7% (P=0.042) and -5.8% (P=0.029), respectively. A coverage probability plot demonstrated that 80% and 90% of tumor dose measurements fell within intermodality differences of 11% and 18%, respectively. PET/MR underestimated tumor dose at both low (<1 GBq) and high (>3 GBq) injected activity levels (P<0.001) by -22.3 [standard deviation (SD) =13.5] and -24.3 (SD =18.7), respectively.
Although PET/MR significantly underestimated the absorbed dose when compared to PET/CT, the intermodality agreement was high and the degree of underestimation was better than previously reported. Intermodality differences were more pronounced at low and high injected doses. Additional studies are required to assess the clinical implications of these findings.
钇90放射性栓塞是一种已确立的肝恶性肿瘤治疗方式。成功的放射性栓塞需要向肿瘤输送最佳剂量，同时尽量减少对实质组织的剂量。治疗后正电子发射断层扫描（PET）/计算机断层扫描（CT）剂量测定是既定的基准，而PET/磁共振成像（MR）是一种新兴的方式。本研究的目的是评估PET/MR与PET/CT钇90剂量测定之间的模态间一致性。
在这项单机构研究中，18例患有原发性或转移性肝恶性肿瘤的患者（20个治疗疗程）在钇90放射性栓塞后接受了PET/MR和PET/CT检查。患者被随机分为先接受一种模态检查，然后再接受另一种模态检查。使用磁共振图像勾勒感兴趣区域，并计算肿瘤和肝脏的剂量。使用Bland-Altman方法评估模态间一致性。使用广义线性模型评估基线变量对模态间剂量差异的影响。
与PET/CT相比，PET/MR分别低估肿瘤和肝脏吸收剂量-3.7%（P = 0.042）和-5.8%（P = 0.029）。覆盖概率图显示，80%和90%的肿瘤剂量测量值分别落在模态间差异的11%和18%范围内。在低（<1 GBq）和高（>3 GBq）注射活度水平下，PET/MR均低估肿瘤剂量（P < 0.001），分别低估-22.3 [标准差（SD）= 13.5]和-24.3（SD = 18.7）。
尽管与PET/CT相比，PET/MR显著低估了吸收剂量，但模态间一致性较高，且低估程度优于先前报道。在低注射剂量和高注射剂量下，模态间差异更为明显。需要进一步研究以评估这些发现的临床意义。
