Prognostic Stratification of Baseline Total Metabolic Tumor Volume on PET/CT Combined with MYC/Bcl-2 Dual Expression in Patients with Primary Gastrointestinal Diffuse Large B-Cell Lymphoma

Purpose: The aim of the present study was to explore whether pretreatment total metabolic tumor volume (TMTV) combined with MYC/BCL-2 dual expression (DE) would improve the prognostic stratification of patients who were newly diagnosed with primary gastrointestinal diffuse large B-cell lymphoma (PGI-DLBCL). Materials: and methods Eighty-three patients between March 2011 and November 2019, diagnosed with PGI-DLBCL prior to treatment, were included in this retrospective study. Baseline TMTV on PET/CT scans were calculated automatically using the boundaries of voxels presenting a SUVmax≥2.5. Expression of MYC/BCL-2 were detected at protein levels via immunohistochemistry（IHC）. The distributions of Progression-free survival (PFS) and overall survival (OS) rates were estimated using the Kaplan-Meier method and differences were compared using a log-rank test followed by multivariate analysis using the Cox proportional hazards model. Results: TMTV and DE were significantly associated with a worse PFS and OS. Multivariate analysis revealed that TMTV (HR=6.090, P<0.001) and DE (HR=2.761, P=0.021) were each independent predictors of PFS, whereas TMTV (HR=9.512, P<0.001) was the only independent predictor of OS. A monogram comprised of TMTV and DE expression identified four groups with very different outcomes: (PFS:χ 2 =32.178, P<0.001; OS:χ 2 =23.091, P<0.001): low-risk group (low TMTV and non-DE, 46 patients); low-intermediate risk group (low TMTV and DE, 16 the patients); high-intermediate risk group (high TMTV and non-DE, 12 patients); and high-risk group (high TMTV and DE patients, 9 patients). Conclusions: TMTV and DE independently predicted survival outcomes in PGI-DLBCL patients. Furthermore, our findings suggest that combination of TMTV and DE may further improve the ability of clinicians to stratify patients in terms of differential prognoses.