MRI, FDG-PET/CT and Image-Guidance for Re-Irradiation of Locoregionally Recurrent or Second Primary Head-and-Neck Squamous Cell Carcinoma Patients – Results of a Multicenter Cohort Study

Abstract

To investigate patterns of care and prognostic benefits of MRI, FDG-PET/CT and image-guidance in re-irradiation of locoregionally recurrent or second primary head-and-neck squamous cell carcinomas (r/s HNSCCs) within a multicenter cohort study. Patients receiving re-irradiation for r/s HNSCC between 2009 and 2020 at 16 tertiary cancer centers in Germany were retrospectively analyzed in terms of MRI and FDG-PET/CT usage for treatment planning and regarding image-guidance frequency during re-irradiation. Patterns of use of these modalities over time were analyzed by binary logistic regression analysis, and the association between the usage of these modalities and best locoregional treatment response was analyzed with chi-square tests. Cumulative incidence analyses of locoregional failures with death as competing event were performed. In the total cohort of 297 patients, 226 (76%) were male, median age was 62 years (IQR, 56-70), and median ECOG was 1 (IQR, 1-2). There were 260 locoregionally recurrent HNSCCs, and 37 second primary HNSCCs; 44 patients (15%) had distant metastases at the time of re-irradiation. MRI and FDG-PET/CT was used for re-irradiation planning in 117 (39%) and 71 patients (24%), respectively. In median, image guidance (IGRT) was performed twice weekly (IQR, 1-5), usually with cone beam CTs or megavolt-CTs, and 85 patients (29%) received daily IGRT during re-irradiation. Usage of MRI (OR = 0.967; 95% CI, 0.892-1.048; p = .416), FDG-PET/CT (OR = 1.053; 95% CI, 0.960-1.156; p = .274), or daily IGRT (OR = 1.057; 95% CI, 0.968-1.115; p = .218) did not increase in frequency over time within the analyzed time span but was significantly dependent on the treatment center (χ2(15), P<.001 for all modalities). Daily IGRT was associated with a higher rate of at least stable disease after re-irradiation as assessed by RECIST criteria (χ2(1) = 4.011, p<.05). There was a trend towards better RECIST-assessed treatment response for MRI (χ2(1) = 3.223, p = .073) and FDG-PET/CT (χ2(1) = 2.792, p = .095) as part of the re-irradiation planning process. Incidence of locoregional failures was not dependent on MRI (SHR = 0.94; 95% CI, 0.67-1.33; p = 0.741, Fine-Gray), FDG-PET/CT (SHR = 0.88; 95% CI, 0.59-1.33; p = 0.552) or daily IGRT (SHR = 0.76; 95% CI, 0.51-1.14, p = 0.182), There was a trend towards lower acute grade 3/4-toxicities in patients receiving daily IGRT (χ2(1) = 3.354, p = 0.067). Our data suggest that daily IGRT may increase disease control and should be regularly applied for re-irradiation of r/s HNSCCs. MRI and FDG-PET/CT usage were not associated with the incidence of locoregional failures after re-irradiation. However, prospective trials with multiparametric MRI and/or FDG-PET/CT for optimal re-irradiation planning are warranted.