AUTHORS
Victoria Beatrice Anthes, MD, Mahmood Aminilari, PhD2, Katrina Hueniken, MPH MSc3, Ankita Kanoja, MD4, Ryan Karimi, BSc5, Anca Prica, MD6, Abi Vijenthira, MD, MS7, Badr Id Said, MD, MSc8, Danielle Rodin, MD9, David Hodgson, MD, MPH10
AFFILIATIONS
1 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Ontario, Canada, University of Toronto, Toronto, Ontario, Canada
2 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
3 Department of Biostatistics, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
4 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
5 University of Toronto, Toronto, Ontario, Canada
6 Division of Hematology, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
7 Division of Hematology, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
8 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
9 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
10 Radiation Medicine Program, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
ABSTRACT
Introduction: Significant practice variation exists in the use of consolidative radiotherapy (RT) following upfront chemoimmunotherapy for patients with diffuse large B-cell lymphoma (DLBCL), particularly when considering metabolic response informed by end-of-chemotherapy (EOC) FDG-PET. The primary objective of this study was to evaluate the association between consolidative RT and patient outcome according to strata of EOC PET response, defined as complete metabolic response (CMR; Deauville score 1-3) and PET-positive residual disease (Deauville score 4-5). Secondary objectives included evaluating the association between consolidative RT and survival by stage and in the FDG-PET era regardless of PET response.
Methods: We performed a PRISMA-compliant systematic review and meta-analysis (PROSPERO: CRD420251077343). We searched EMBASE, MEDLINE, Web of Science, and the Cochrane Central Register of Controlled Trials (CENTRAL) from 2010 to March 2026 for studies of adults with DLBCL treated with upfront chemoimmunotherapy (R-CHOP or similar) comparing consolidative RT versus observation following EOC PET response assessment. Two pairs of reviewers independently assessed risk of bias and extracted data. Co-primary outcomes were progression-free survival (PFS) and overall survival (OS), with primary analyses performed in patients achieving CMR and PET-positive residual disease separately. We pooled hazard ratios (HRs) using a random-effects model when possible and used the GRADE approach to appraise the uncertainty of the evidence. A Knapp-Hartung adjustment was applied to the CMR OS and Deauville 4-5 PFS analyses given the small number of contributing studies.
Results: Of 7258 studies identified, 23 (1 randomized controlled trial [RCT] n=319, and 22 observational studies, n=6585) comprising 6904 patients were included in the final analysis (4909 CMR, 944 Deauville 4-5, 1051 not retrievable). The mean of study-level averages for patient age was 60 years, 48% female, 44% advanced-stage, 34% with IPI≥3, and 32% with bulky disease (variably defined). R-CHOP was administered in an average of 90% of patients, ranging from 3-8 cycles. Consolidative RT included IFRT and ISRT, with doses ranging from 15 to 60 Gy. Most observational studies were at moderate risk of bias from residual confounding.
Among patients achieving CMR, low-certainty evidence indicates that consolidative RT was associated with improved PFS (pooled HR 0.47, 95% CI 0.24-0.92; prediction interval -2.44 to 0.94; I2 = 49.7%) and OS (pooled HR 0.44, 95% CI 0.22-0.90; prediction interval -1.72 to 0.07). Among patients with PET-positive residual disease, low-certainty evidence indicates that consolidative RT was associated with markedly improved PFS (pooled HR 0.26, 95% CI 0.13-0.50; prediction interval -2.18 to -0.56); OS was not analyzed in this subgroup due to limited data. The association between consolidative RT and superior PFS was significantly stronger among patients with PET-positive residual disease than those with CMR (interaction p = 0.024), suggesting potential effect modification by EOC PET response. In the one RCT of non-bulky limited-stage DLBCL, consolidative RT following R-CHOP did not significantly improve EFS or OS, although the outcome of CMR vs PET-positive EOC patients was not reported separately.
In exploratory analyses of patients with limited-stage (I-II) DLBCL (mixed PET responses), consolidative RT was associated with significant improvements in PFS (HR 0.48, 95% CI 0.31-0.75; prediction interval -1.71 to 0.23; I2 = 25.5%) and OS (HR 0.38, 95% CI 0.20-0.71; prediction interval -2.65 to 0.70; I2 = 44.0%). Among advanced stage (III-IV) DLBCL, a similar improvement in PFS (HR 0.32, 95% CI 0.12–0.83; prediction interval -4.42 to 2.11; I2 = 45.3%) and trend toward superior OS that did not reach statistical significance (HR 0.50, 95% CI 0.24–1.05; prediction interval -2.33 to 0.94) were observed.
Conclusion: This meta-analysis supports consideration of PET-guided consolidative RT, particularly among patients with residual PET-avid EOC findings. However, the evidence is almost entirely derived from observational non-randomized studies, and remains susceptible to selection bias, confounding, and publication bias. This emphasizes the necessity of conducting randomized controlled trials to establish causality.
DISCLOSURES
Victoria Beatrice Anthes, MD: No relevant conflicts of interest to declare.