Interim FDG-PET in lymphoma, a questionable practice in hematology

Predictors of metabolic response in propensity-matched lymphoma patients on interim 18F-fluorodeoxyglucose positron-emission tomography/computed tomography using standardized imaging and reporting protocol: Do we really have one?

The purpose of this prospective study was to determine metabolic response predictor(s) in propensity-matched patients having lymphomas who had baseline and interim ¹⁸fluorodeoxyglucose (FDG) positron-emission tomography/computed tomography (PET/CT) using strict standardized imaging and reporting protocols. This prospective study was conducted at PET/CT section of a JCI-accredited healthcare facility from April 2017 to February 2018. Patients with baseline and interim ¹⁸FDG PET/CT scans using standardized protocol were selected. Interim scans were performed not earlier than 2^nd or later than 4^th chemotherapy. During the study period, 97 of 112 consecutive patients with lymphomas (Hodgkin-HL: 32/97 and Non-Hodgkin-NHL: 65/97) were included in the study. Mean age of cohort was 45 ± 19 years (71% male and 29% female) having a mean body mass index (BMI) of 25.57 ± 5.54 Kg/m² having Stage I (21%), Stage II (18%), Stage III (16%), and Stage IV (45%) disease. Bulky disease was found in 14% and ¹⁸FDG-avid marrow deposits in 33%. Standardized PET/CT imaging protocol as per EANM guidelines was strictly adopted for baseline and interim studies. %Δ changes in fasting blood sugar, ¹⁸FDG dose, uptake time, and liver SUV mean were 3.96%, 2.83%, 2.49%, and 12.15%, respectively. Based on Deauville's scoring, cohort was divided into responders having Score 1-3 (49/97) and nonresponders having Score 4-5 (48/97). The demographic analysis found no significant difference between responders and nonresponders for age, gender, BMI, staging, bulky disease or marrow involvement, and study protocol. No significant coefficient or odd ratios were found on multivariate analysis for age, gender, maximum standardized uptake value (SUV_max), size, BMI, NHL, and advance disease (Stage III and IV) in both groups (χ²: 5.12; receiver operating characteristic [95% confidence interval]: 0.616 [0.51-0.713]; P =0.528). Among responders, baseline SUV_max and tumor size had a direct correlation with a metabolic response on iPET, more pronounced in NHL than HL groups (SUV_max: 13.4 vs. 19.5 and size: 52 vs. 87 mm; P < 0.0001). We conclude that no significant predictor was found for response in propensity-matched patients with lymphomas (both HL and NHL) who had baseline and interim PET/CTs acquired with a standardized protocol. However, NHL responders were found to have higher baseline median SUV_max and larger lesion size as compared to HL responders. Although, these data are not in concordance with published findings but need to be validated with larger studies using standardized imaging and reporting protocols in propensity-matched patients with lymphomas.