Accuracy of radiological assessment of nodal involvement in lung cancer

<bold>Introduction:</bold> The sensitivity and negative predictive value of Endobronchial Ultrasound (EBUS) in mediastinal nodal staging is around 89% and 91% respectively. It is superior to radiological assessment alone and has an established place in the staging of lung cancer. Our centre relied on MDT expert judgement of radiology pre-2021 due to lack of availability of staging EBUS operators. The decision to proceed to radical treatment was based on a number of factors including but not restricted to tumour size and location, nodal size, PET avidity and patient fitness. <bold>Methodology:</bold> We undertook a retrospective analysis of patient outcomes for radiologically staged N1/2 lung cancer between August 2018 to November 2021. This cohort should ideally have been offered staging EBUS. We excluded patients staged as N0, N3, or M1. We followed their treatment pathway through to surgical staging post-resection. <bold>Results:</bold> Of the 39 cases, 25 underwent surgical resection,15(60%) were thought to have radiological N1 disease and 10(40%) N2 disease based on nodal size (>10mm) and/or PET criteria. Post-surgical pathological staging revealed that 12(48%) were inaccurately staged with 2(8%) being upstaged and 10(40%) patients having a lower nodal status. Fourteen patients received radical chemo-radiotherapy for presumed N1/N2 disease and 12 (85%) in this group did not have evidence of disease progression on follow up imaging at 6 months. Average 1-year survival in this cohort was 88% <bold>Conclusion:</bold> Even an experienced lung cancer MDT could only identify radiological nodal involvement accurately about 50% of the time. This could lead to sub-optimal treatment selection and potential under-treatment. The role of staging EBUS cannot be over-emphasised.