An atlas to aid delineation of para-aortic lymph node region in cervical cancer: Design and validation of contouring guidelines

Introduction: Previous studies have investigated the anatomical distribution of para-aortic lymph nodes (PAN) in patients with cervical cancer. However, an atlas for accurate clinical target volume (CTV) delineation has yet to be defined. The purpose of this study was to design and verify a computerized tomography (CT) atlas to provide guidance for contouring the PAN CTV in patients with cervical cancer.

Materials and methods: This prospective study included 21 cervical cancer patients (design cohort) with 39 pathological PAN identified on (18)F-FDG PET-CT. PAN [left lateral para-aortic (LLPA), aorto-caval (AC), right para-caval (RPC) nodes] were delineated on CT simulation scans. Measurements were taken from the volumetric centre of the nodes to the edge of aorta and inferior vena-cava (IVC). Initially the aorta and IVC were expanded by the mean distance to the lymph node centre to create a CTV. Expansion margins were then increased asymmetrically until the CTV resulted in a clinically acceptable number of PAN included. The CTV was validated on a further 10 patients (validation cohort) with 29 PAN. A detailed contouring guide and accompanying visual atlas for elective PAN CTV delineation was created based on the validated margins.

Results: For the design cohort (n = 21 patients, 39 PAN), the mean distance from the centre of the node to the aorta was 8 mm (range 4-17) for both LLPA (range 4-17) and AC (range 4-15) regions. Mean distance from the IVC to the centre of the nodes was 5 mm (range 4-6) in the RPC region and 6 mm (range 3-15) in the AC region. No PAN was superior to the T12-L1 interspace or the left renal vein or inferior to the L5-S1 interspace. For validation cohort (n = 10 patients, 29 PAN), mean distance from centre of the node to the aorta was 9 mm (range 5-15) in the LLPA region, 7 mm (range 6.5-14) in the AC region. Mean distance from the ICV to the centre of the nodes was 3 mm (range 2.5-4) in the RPC region and 5 mm (range 3-10) in the AC region. A CTV expansion from the aorta of 10 mm circumferentially and 15 mm laterally, and from the IVC of 8 mm anteromedially and 6 mm posterolaterally resulted in coverage of 97% (38/39) of PAN in the design cohort. On prospective validation, the described CTV included 97% (28/29) of PAN in the validation cohort.

Conclusion: We propose the following PAN CTV; expansion from aorta of 10 mm circumferentially except 15 mm laterally, expansion from the IVC of 8 mm anteromedial and 6 mm posterolaterally. The suggested CTV includes 97% (28/29) PAN in a validated patient cohort. A detailed guide and accompanying visual atlas is provided to aid delineation of the PAN CTV in patients with cervical cancer.