Selective mediastinal node irradiation in non-small cell lung cancer in the IMRT/VMAT era: How to use E(B)US-NA information in addition to PET-CT for delineation?

Background: FDG-PET-CT-based selective lymph node (LN) irradiation is standard using 3D-conformal techniques for locally advanced NSCLC. With newer techniques (intensity-modulated/volumetric-arc therapy (IMRT/VMAT)), the dose to non-involved adjacent LN decreases, which raises the question whether FDG-PET-CT-delineation is still safe. We therefore evaluated the impact of adding linear endosonography with needle aspiration (E(B)US-NA) to FDG-PET-CT in selective nodal irradiation. Methods: Based on literature data on sensitivity and specificity of E(B)US-NA in FDG-PET-CT-staged NSCLC, false negative (FN) rates for different constellations of CT, PET and E(B)US-NA were calculated. The algorithm was tested on consecutive patients with N2/N3 disease referred for radiotherapy in Leuven and Maastricht. Results: An algorithm determining when to include LN in the GTV is proposed, based on data from 5 meta-analyses. Adding E(B)US-NA to FDG-PET-CT decreases the FN-rate, but for PET-positive and E(B) US-negative LN, FN rates are still 14-16%. In Leuven 520 LN were analyzed, in Maastricht 364 LN; with E(B)US-NA a geographical miss was. avoided in 2 (2/40 = 5%) and 1 (1/28 = 4%) patients, respectively. Conclusions: E(B)US-NA in addition to FDG-PET-CT for mediastinal staging decreases the risk of a geographical miss with 4-5%. The impact of this small decrease on survival is unknown. The proposed algorithm may guide the radiation oncologist when to include LN in the nodal GTV.