Definition of lymph node areas for radiotherapy of prostate cancer: A critical literature review by the French Genito-Urinary Group and the French Association of Urology (GETUG-AFU)

•Recommendations for pelvic LN contouring rely on dated studies, using discordant definitions.•In 10-70% of cases, sites of drainage are not included in proposed pelvic LN CTVs.•Assessment of efficacy and safety of LN radiotherapy is still debated.•A new consensus for delineation of LN regions is needed.

PurposeRecommendations for pelvic lymph node (LN) contouring rely on relatively dated studies that defined the Clinical Target Volume (CTV) of interest proposed for radiotherapy. The aim of this article was to review these recommendations with a critical analysis of published data on prostate cancer drainage.MethodsWe performed a review of data on LN drainage in prostate cancer, based on anatomy texts and studies on lymphography, pelvic LN dissections, sentinel LN techniques, magnetic resonance imaging, computed tomography and functional imaging. We also present the GETUG experts' opinion, based on a survey on nodal CTV definition.ResultsFor lymphatic drainage of prostate cancers, pelvic LN areas classically considered are: distal common iliac, external iliac, internal iliac and obturator regions. Recently published data allow a mapping of sites at risk of pathological LN invasion. In 10-70% of cases, these sites are not included in the pelvic LN CTVs defined in consensuses. In accordance with other cooperative groups, the GETUG experts' survey showed that proximal common iliac, para-aortic, para-rectal and pre-sacral regions could include sites at risk of invasion in extended LN CTV, but were not considered in CTV contouring common practice. New recommendations are needed for nodal CTV in radiotherapy of prostate cancer.ConclusionsThe assessment of the efficacy and safety of LN radiotherapy is still the subject of several randomised studies. Whether or not meaningful results are obtained depends directly on the quality and homogeneity of the data analysed. A new consensus for delineation of LN regions appears necessary.