A Systematic Review and Meta-Analysis of the Prognostic Value of Tumor Response Measurements During Radiotherapy for Head and Neck cancers.

AIMS: This review evaluates the literature on the prognostic value of in-treatment tumor response measurements during radiotherapy for head and neck cancer, assessing the most commonly used imaging modalities through meta-analysis. MATERIALS AND METHODS: A systematic literature search was conducted following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines. Studies correlating in-treatment tumor response during radiotherapy with clinical endpoints such as loco-regional control (LRC) or overall survival (OS) were included. Meta-analyses were performed for comparable hazard ratios (HR) using the random-effects model. RESULTS: 98 studies were included. 18F-fluorodeoxy-d-glucose positron emission tomography (FDG-PET) (32 studies), magnetic resonance imaging (MRI) (31 studies), and computed tomography (CT) (11 studies) were the most frequently used modalities. Meta-analysis showed that a low tumor volume reduction rate on CT was strongly associated with poorer LRC (HR: 4.6). High SUVmax on FDG-PET during treatment was a significant prognostic marker for both poorer OS (HR: 1.97) and LRC (HR: 1.72). In contrast, changes in apparent diffusion coefficient (ADC) on MRI showed no significant prognostic value for LRC (HR: 0.97). Only 19.4% of studies could be included in the meta-analyses due to missing data. Attempts to retrieve additional information from authors were largely unsuccessful. CONCLUSION: Tumor volume reduction rate on CT and SUVmax on FDG-PET are promising prognostic markers for treatment response during radiotherapy in head and neck cancer. ADC measurements from MRI show limited prognostic value.