Lung Sparing for SBRT with beam's-eye-view images and real time tumor tracking

描述（由申请人提供）：在美国，死于肺癌的人数比死于前列腺癌、乳腺癌、结肠癌和直肠癌的人数总和还要多。在2009年美国估计有187，000人被诊断患有非小细胞肺癌（NSCLC），其中超过37，000人表现为局部或早期疾病。随着早期检测方法的改进和更广泛的传播，这一比例预计会增加。立体定向放射治疗（SBRT）已被证明为早期肺癌患者提供了极好的局部控制（85%-95%）。然而，最近的一项II期研究发现，相当一部分患者出现3级或更高的毒性，特别是那些肿瘤位于中心的患者。研究表明，SBRT边缘的减少可以显著降低正常组织并发症的概率。随着SBRT越来越多地成为早期局部非小细胞肺癌的治疗选择，减少有害的副作用变得越来越重要。我们正在提出一种新的故障安全技术，以减少毒性，同时保持对这一不断增长的患者群体的局部控制。我们的假设是，在SBRT过程中直接跟踪肺肿瘤，使用束眼视图（BEV）成像加上动态多叶准直器（DMLC），将导致临床上显着的正常组织保留。目前的提案是第一个采用先进的多模板无标记BEV跟踪算法来导出用于DMLC递送的实时肿瘤位置的提案。多模板无标记物创新的临床受益包括1）自动选择相关标志，2）在变形、旋转和部分遮蔽期间连续跟踪，3）患者无需额外的成像剂量，4）整个肿瘤的直接成像，以及5）无需侵入性基准校准和相关不良反应。这代表了对先前技术的实质性改进。在临床条件下，将在拟人体模系统中通过实验量化治疗优势。最终的结果将是一个集成的实时目标跟踪和动态输送系统的放射治疗以及量化的预期临床效益。与DMLC跟踪的临床整合是一个具有挑战性的工程问题，对人类健康有着大规模的可衡量的积极影响。该项目很好地体现了NIBIB支持物理科学和工程研究与开发以改善人类健康的理想。 公共卫生相关性：本项目的目标是量化使用治疗辐射束跟踪早期肺肿瘤的潜在临床受益，以避免治疗期间过度照射健康组织。我们提出的方法是最安全的，侵入性最小的。如果发现成功，该技术可以应用于其他解剖部位，为大量癌症患者提供更好的结果。

DESCRIPTION (provided by applicant): More people in the U.S. die from lung cancer than from prostate, breast, colon and rectum cancers combined. Of the estimated 187,000 people in the U.S. who were diagnosed with non-small cell lung cancer (NSCLC) in 2009, over 37,000 of them presented with localized or early stage disease. This fraction is expected to grow as methods of early detection improve and become more widely disseminated. Stereotactic body radiation therapy (SBRT) has been shown to provide excellent local control (85%-95%) for early-stage lung cancer patients. However, a recent phase II study found Grade 3 or greater toxicities in a significant fraction of the patients, particularly those with centrally located tumors. Studies have shown that decreases in SBRT margins can significantly reduce the probability of normal tissue complications. As SBRT is increasingly becoming the therapy of choice for early-stage, localized non-small cell lung cancer, reducing the harmful side effects becomes increasingly important. We are proposing a novel failsafe technology to reduce toxicity while retaining local control for this growing population of patients. Our hypothesis is that tracking lung tumors directly during SBRT, using beam's-eye-view (BEV) imaging coupled with a dynamic multileaf collimator (DMLC), will lead to clinically significant normal tissue sparing. The current proposal is the first to employ an advanced multi- template marker-less BEV tracking algorithm to derive the real-time tumor location for DMLC delivery. Clinical benefits of the multi-template marker-less innovation include 1) automatic selection of relevant landmarks, 2) continuous tracking during deformations, rotations and partial obscurations, 3) no additional imaging dose to the patient, 4) direct imaging of the entire tumor, and 5) no need for invasive fiducial implantations and the associated adverse effects. This represents a substantial improvement over previous techniques. The therapeutic advantage will be quantified experimentally in an anthropomorphic phantom system under clinical conditions. The end result will be an integrated real-time target tracking and dynamic delivery system for radiation therapy as well as quantification of the anticipated clinical benefits. Clinical integration with the DMLC tracking is a challenging engineering problem with a measurable positive impact on human health on a large scale. This project represents well the ideals of the NIBIB to support research and development in the physical sciences and engineering for the improvement of human health. PUBLIC HEALTH RELEVANCE: The goal of this project is to quantify the potential clinical benefits of tracking early stage lung tumors with the therapeutic radiation beam in order to avoid over irradiation of healthy tissues during treatment. Our proposed method is the safest, least intrusive to have been proposed. If found to be successful, the technique could be applied to other anatomical sites, providing better outcomes for a very large number of cancer patients.