purpose Fracture of the femur is one of the late complications of adjuvant radiotherapy for patients with soft tissue sarcomas of the thigh, who receive external beam irradiation after limb-sparing surgery. When the target volume approximates the femur, it is often inevitable that a large segment of the femur will receive full prescription dose with conventional radiation techniques. We report the dosimetric feasibility of intensity- modulated radiation therapy (IMRT) techniques to achieve adequate target coverage and bone sparing. Methods and materials Treatment planning was performed using both three-dimensional conformal radiotherapy (3D-CRT) and IMRT techniques for 10 patients with soft tissue sarcoma of the thigh with tumor approaching the femur. None of the patients had bony involvement. For all patients, the gross total volume (GTV) and the femur were contoured. The clinical target volume (CTV) was defined as the GTV with a 1.5-cm margin axially, except at the bone interface where the bone interface was used as CTV if the 1.5-cm axial margin extended beyond the bone interface. In the superior-inferior direction, the CTV margin placed around the GTV varied from 5 to 10 cm. The planning target volume (PTV) was defined as the CTV with 5-mm margin all around. The 3D conformal technique consisted primarily of two to three beams with wedges or partial transmission blocks as compensators. For the IMRT technique, five coplanar beams were used, chosen so as to spare much of the surrounding soft tissue and to clear the other extremity or groin areas. IMRT plans were designed to adequately treat the planning target volume and spare the femur as much as possible. Results Dose distributions and dose–volume histograms were analyzed. PTV coverage was comparable with both IMRT and 3D-CRT plans. Dose distributions were more conformal with IMRT, however, especially for patients with large variations of contours. The volume of the femur receiving at least full prescription (63 Gy) V100 decreased on average by approximately 57%, from 44.7 ± 16.8% with 3D-CRT to 18.6 ± 9.2% with IMRT ( p < 0.01). For 3 patients with a GTV surrounding <50% of the circumference of the femur, the reduction in the V100 to the femur ranged from 61% to 79%. The hot spots in the femur, as measured by D05 (the dose encompassing 5% of volume), reduced on average from 67.2 ± 1.8 Gy with 3D-CRT to 65.0 ± 1.2 Gy with IMRT ( p < 0.01). The mean dose to the femur was on average 38.5 ± 11.5 Gy with IMRT, compared with 40.9 ± 12.7 Gy with 3D-CRT. The volume of the surrounding soft tissues, defined as the ipsilateral limb excluding the PTV and the femur, receiving at least prescription dose (63 Gy) was reduced on average by about 78%, from 997 ± 660 cc with 3D-CRT to 201 ± 144 cc with IMRT ( p < 0.01). The D05 to the surrounding soft tissues was on average 58.7 ± 4.7 Gy with IMRT, compared to 67.8 ± 1.3 Gy with 3D-CRT ( p < 0.01), a reduction of approximately 13%. The mean dose to the surrounding soft tissues was comparable in both plans. The volume of the skin (from surface to 5 mm depth) receiving prescription dose (63 Gy) declined by roughly 45%, from 115 ± 40 cc with 3D-CRT to 61 ± 20 cc with IMRT ( p < 0.01), with IMRT providing full skin dose coverage to scars. The hot spots in the skin decreased from 68.0 ± 1.7 Gy with 3D-CRT to 65.2 ± 1.2 Gy with IMRT ( p < 0.01). The mean dose to the skin lessened from 51.5 ± 4.7 Gy with 3D- CRT to 44.0 ± 4.2 Gy with IMRT ( p < 0.01), a reduction of 14%. Conclusions Intensity-modulated radiation therapy techniques can reduce the dose to the femur without compromising target coverage by achieving concave dose distributions around the interface of the PTV and the femur. At the same time, IMRT can reduce the hot spots significantly in the surrounding soft tissues and skin. Whether such dosimetric improvements can translate into reduction of complications and/or improving local control needs to be investigated.