AAPM Task Group 119 has produced quantitative confidence limits as baseline expectation values for IMRT commissioning. A set of test cases was developed to assess the overall accuracy of planning and delivery of IMRT treatments. Each test uses contours of targets and avoidance structures drawn within rectangular phantoms. These tests were planned, delivered, measured, and analyzed by nine facilities using a variety of IMRT planning and delivery systems. Each facility had passed the Radiological Physics Center credentialing tests for IMRT. The agreement between the planned and measured doses was determined using ion chamber dosimetry in high and low dose regions, film dosimetry on coronal planes in the phantom with all fields delivered, and planar dosimetry for each field measured perpendicular to the central axis. The planar dose distributions were assessed using gamma criteria of 3%/3 mm. The mean values and standard deviations were used to develop confidence limits for the test results using the concept confidence limit = /mean/ + 1.96sigma. Other facilities can use the test protocol and results as a basis for comparison to this group. Locally derived confidence limits that substantially exceed these baseline values may indicate the need for improved IMRT commissioning.

Through a systematic analysis of multiparametric MR imaging features, we are able to build models with improved predictive value over conventional imaging metrics. The results are encouraging, suggesting the wealth of imaging radiomics should be further explored to help tailoring the treatment into the era of personalized medicine. Clin Cancer Res; 22(21); 5256-64. ©2016 AACR.

Background Our previous analysis of papillary thyroid carcinomas (PTC) from the Ukrainian-American cohort exposed to 131I from the Chernobyl accident found RET/PTC rearrangements and other driver mutations in 60% of tumors. Methods In this study, we analyzed the remaining, mutation-negative tumors using RNA-Seq and RT-PCR to identify novel chromosomal rearrangements and characterize their relationship with radiation dose. Results The ETV6-NTRK3 rearrangement was identified by RNA-Seq in a tumor from a patient who received a high 131I dose. Overall, it was detected in 9/62 (14.5%) of post-Chernobyl and in 3/151 (2%) of sporadic PTCs (p=0.019). The most common fusion type was between exon 4 of ETV6 and exon 14 of NTRK3. The ETV6-NTRK3 prevalence in post-Chernobyl PTCs was associated with increasing 131I dose, albeit at borderline significance (p=0.126). The group of rearrangement-positive PTCs (ETV6-NTRK3, RET/PTC, PAX8-PPARγ) was associated with significantly higher dose response compared to the group of PTCs with point mutations (BRAF, RAS) (p<0.001). In vitro exposure of human thyroid cells to 1 Gy of 131I and γ-radiation resulted in the formation of ETV6-NTRK3 with a rate of 7.9 × 10−6 and 3.0 ×10−6 cells, respectively. Conclusions We report here the occurrence of ETV6-NTRK3 rearrangements in thyroid cancer and show that this rearrangement is significantly more common in tumors associated with exposure to 131I and has a borderline significant dose response. Moreover, ETV6-NTRK3 can be directly induced in thyroid cells by ionizing radiation in vitro and therefore may represent a novel mechanism of radiation-induced carcinogenesis.

No abstract

Purpose To improve image quality and computed tomography (CT) number accuracy of daily cone beam CT (CBCT) through a deep learning methodology with generative adversarial network. Methods One hundred and fifty paired pelvic CT and CBCT scans were used for model training and validation. An unsupervised deep learning method, 2.5D pixel‐to‐pixel generative adversarial network (GAN) model with feature mapping was proposed. A total of 12 000 slice pairs of CT and CBCT were used for model training, while ten‐fold cross validation was applied to verify model robustness. Paired CT–CBCT scans from an additional 15 pelvic patients and 10 head‐and‐neck (HN) patients with CBCT images collected at a different machine were used for independent testing purpose. Besides the proposed method above, other network architectures were also tested as: 2D vs 2.5D; GAN model with vs without feature mapping; GAN model with vs without additional perceptual loss; and previously reported models as U‐net and cycleGAN with or without identity loss. Image quality of deep‐learning generated synthetic CT (sCT) images was quantitatively compared against the reference CT (rCT) image using mean absolute error (MAE) of Hounsfield units (HU) and peak signal‐to‐noise ratio (PSNR). The dosimetric calculation accuracy was further evaluated with both photon and proton beams. Results The deep‐learning generated sCTs showed improved image quality with reduced artifact distortion and improved soft tissue contrast. The proposed algorithm of 2.5 Pix2pix GAN with feature matching (FM) was shown to be the best model among all tested methods producing the highest PSNR and the lowest MAE to rCT. The dose distribution demonstrated a high accuracy in the scope of photon‐based planning, yet more work is needed for proton‐based treatment. Once the model was trained, it took 11–12 ms to process one slice, and could generate a 3D volume of dCBCT (80 slices) in less than a second using a NVIDIA GeForce GTX Titan X GPU (12 GB, Maxwell architecture). Conclusion The proposed deep learning algorithm is promising to improve CBCT image quality in an efficient way, thus has a potential to support online CBCT‐based adaptive radiotherapy.