Purpose: To evaluate treatment planning and delivery for patients treated during our initial year of experience with RapidArc™ radiation therapy. Methods and Materials: RapidArc was used to treat 52 patients at The University of Alabama at Birmingham between May 2008 and April 2009. A single ionization chamber phantom with film and a two-dimensional ionization chamber array were used for quality assurance measurements. Of the 52 patients, 44 had a static gantry dynamic multileaf collimated (SG-DMLC) IMRT treatment plan, seven of which had quality assurance (QA) measurements. Results: The mean difference between ionization chamber measurement and calculation was 1.2% ± 0.9% (1 standard deviation). For film, the mean fraction of pixels with γ > 1 (3%/3 mm criterion) was 4.6% and for the two-dimensional chamber array was 1.4%. For the seven corresponding SG-DMLC plans, the results were similar. Differences in important dosimetric indicators were typically within 1% relative to SG-DMLC. The volume of nontarget tissue that received >20 Gy was less for RapidArc compared with SG-DMLC, whereas the volume that received more than 10 Gy was larger. The mean difference between the measured and planned leaf positions and the monitor units obtained from machine log files was 0.0 ± 0.5 mm and 0.4 ± 0.3 MU, respectively. Mean delivery times were 1.5 ± 0.2 and 3.3 ± 0.4 min for one- and two-arc plans, respectively. On average, SG-DMLC delivery took 4.4 min longer. Conclusions: RapidArc plans have quality comparable to our standard SG-DMLC IMRT technique, and are delivered with similar accuracy in shorter time. © 2010 Elsevier Inc. All rights reserved.