PurposeWith proper beam setup and optimization constraints in the treatment planning system, volumetric modulated arc therapy (VMAT) can improve target dose coverage and conformity while reducing doses to adjacent structures for whole breast radiation therapy. However, the low‐dose bath effect on critical structures, especially the heart and the ipsilateral lung, remains a concern. In this study, we present a VMAT technique with the jaw offset VMAT (JO‐VMAT) to reduce the leakage and scatter doses to critical structures for whole breast radiation therapy.Materials and MethodsThe data of 10 left breast cancer patients were retrospectively used for this study. CT images were acquired on a CT scanner (GE, Discovery) with the deep‐inspiration breath hold (DIBH) technique. The planning target volumes (PTVs) and the normal structures (the lungs, the heart, and the contralateral breast) were contoured on the DIBH scan. A 3D field‐in‐field plan (3D‐FiF), a tangential VMAT (tVMAT) plan, and a JO‐VMAT plan were created with the Eclipse treatment planning system. An arc treatment field with the x‐jaw closed across the central axis creates a donut‐shaped high‐dose distribution and a cylinder‐shaped low‐dose volume along the central axis of gantry rotation. Applying this setup with proper multi‐leaf collimator (MLC) modulation, the optimized plan potentially can provide sufficient target coverage and reduce unnecessary irradiation to critical structures. The JO‐VMAT plans involve 5–6 tangential arcs (3 clockwise arcs and 2–3 counterclockwise arcs) with jaw offsets. The plans were optimized with objective functions specified to achieve PTV dose coverage and homogeneity; For organs at risk (OARs), objective functions were specified individually for each patient to accomplish the best achievable treatment plan. For tVMAT plans, optimization constraints were kept the same except that the jaw offset was removed from the initial beam setup. The dose volume histogram (DVH) parameters were generated for dosimetric evaluation of PTV and OARs.ResultsThe D95% to the PTV was greater than the prescription dose of 42.56 Gy for all the plans. With both VMAT techniques, the PTV conformity index (CI) was statistically improved from 0.62 (3D‐FiF) to 0.83 for tVMAT and 0.84 for JO‐VMAT plans. The difference in the homogeneity index (HI) was not significant. The Dmax to the heart was reduced from 12.15 Gy for 3D‐FiF to 8.26 Gy for tVMAT and 7.20 Gy for JO‐VMAT plans. However, a low‐dose bath effect was observed with tVMAT plans to all the critical structures including the lungs, the heart, and the contralateral breast. With JO‐VMAT, the V5Gy and V2Gy of the heart were reduced by 32.7% and 15.4% compared to 3D‐FiF plans. Significantly, the ipsilateral lung showed a reduction in mean dose (4.65–3.44 Gy) and low dose parameters (23.4% reduction for V5Gy and 10.7% reduction for V2Gy) for JO‐VMAT plans compared to the 3D‐FiF plans. The V2Gy dose to the contralateral lung and breast was minimal with JO‐VMAT techniques.ConclusionA JO‐VMAT technique was evaluated in this study and compared with 3D‐FiF and tVMAT techniques. Our results showed that the JO‐VMAT technique can achieve clinically comparable coverage and homogeneity and significantly improve dose conformity within PTV. Additionally, JO‐VMAT eliminated the low‐dose bath effect at all OARs evaluation metrics including the ipsilateral/contralateral lung, the heart, and the contralateral breast compared to 3D‐FiF and tVMAT. This technique is feasible for the whole breast radiation therapy of left breast cancers.