PurposeAdjuvant internal mammary lymph node (IMN) radiation is often delivered with 2-dimensional techniques that use anatomic landmarks and predetermined depths for field placement and dose specification. In contrast, 3-dimensional planning uses the internal mammary vessels (IMVs) to localize the IMNs for planning. Our purpose was to determine if localization of the involved IMN (i-IMN) by 18F-labeled fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG PET-CT) offers opportunities to improve treatment. Methods and materialsBreast cancer patients (n = 80) who had i-IMNs (n = 112) on PET-CT for initial staging (n = 40) or recurrence (n = 40) were studied. Size, intercostal space (IC), and distance from skin, sternum, and IMVs were recorded. Effects on 2- and 3-dimensional planning were evaluated. ResultsMost i-IMNs (94.6%) were in the first to third ICs. Few were in the fourth (4.5%) or fifth (0.9%) IC. Mean i-IMN depth was 3.4 cm (range, 1.1-7.3 cm). Prescriptive depths of 4, 5, and 6 cm would result in undertreatment of 25%, 10.7%, and 5.3% of IMNs, respectively. Most IMNs (86.6%) were lateral or adjacent to the sternal edge. Only 13.4% of IMNs were posterior to the sternum. Use of the ipsilateral or contralateral sternal edge for field placement increases the risk of geographic miss or excess normal tissue exposure. Most i-IMNs were adjacent to (83%) or ≤0.5 cm (14%) from the IMV edge. Three (3%) were >0.5 cm beyond the IMV edge. The clinical target volume (CTV) defined by the first to third ICs encompassed 78% of i-IMNs. IMN-CTV coverage of i-IMNs increased with inclusion of the fourth IC (82%), 0.5 cm medial and lateral margin expansion (93%), or both (96.5%). ConclusionTwo-dimensional treatment techniques risk geographic miss of IMNs and exposure of excess normal tissue to radiation. An IMN-CTV defined by the IMVs from the first to third ICs with 0.5-cm medial and lateral margin expansion encompasses almost all i-IMNs identified on PET-CT imaging. Inclusion of the fourth IC offers modest coverage improvement, and its inclusion should be weighed against potential increase in cardiac exposure. SummaryThe use of 2-dimensional treatment techniques for adjuvant internal mammary lymph node (IMN) radiation may cause geographic miss of tumor and expose normal tissue to radiation injury. Conformal 3-dimensional planning improves coverage and reduces risk of normal tissue damage by using the internal mammary vessel to define an IMN clinical target volume (CTV). Contouring the IMN-CTV from the first to third intercostal spaces with a 0.5-cm expansion medially and laterally encompasses most IMN. Positron emission tomography-computed tomography may have a role in radiation planning by identifying involved-IMN for dose escalation.