Only a single implanted marker is needed for tracking lung cancers for IGRT

Abstract

Arguing against the Proposition is Sonja Dieterich, Ph.D. After completing her Ph.D. in Nuclear Physics at Rutgers University in 2002, Dr. Dieterich received training in Medical Physics at Georgetown University Hospital, Washington DC, from 2002–2003. In 2003 she accepted a faculty position at Georgetown, where she became Chief of the CyberKnife program in 2006. In 2007 she moved to Stanford University Hospital, Stanford, CA, as Clinical Associate Professor and Chief of Radiosurgery Physics. Dr. Dieterich is certified in Therapeutic Radiologic Physics by the ABR and is Chair of the AAPM Task Group 135 (QA for Robotic Radiosurgery). Her current interests are the development of QA/QM programs for new technologies, motion management, and SRS dosimetry. Advances in imaging technologies have allowed dose escalation in treating localized soft-tissue tumors that move and deform internally due to respiration and the movement of their adjacent organs. With advanced 4D-CT, both the integrated target volume that accounts for tumor excursion and the transient tumor volume at a particular breathing phase can be obtained. I assert that the transient tumor volume with limited margin should be the preferred approach for ablative types of treatment, that is, stereotactic radiosurgery (SRS) or stereotactic body radiotherapy (SBRT), to minimize normal tissue toxicity. If such a target volume with limited margin is to be used for treatment with high doses, image-guidance based on skeletal structure (commonly used in conventional RT and intracranial SRS) is not adequate for treatment delivery. A tumor volume-specific localization technique with high precision is essential. A widely used technique is to localize soft-tissue tumors indirectly through fiducials, that is, implanted tumor surrogates such as radio-opaque markers or radiofrequency transponders.1 It has been a general trend that, when target localization is focused on the tumor volume itself, the six degrees of freedom of the tumor volume are used for treatment guidance. Here we encounter two difficulties. First, for soft-tissue tumors for which internal motion is of concern, the tumor location and orientation vary and are no longer consistent with the global body orientation. Second, due to the often noncoherent tumor deformation, the accuracy of the rotational parameters computed from the implanted fiducial markers is greatly affected. By following all tumor-specific localization parameters (translational and rotational), dosimetric error will be unavoidably introduced due to the changes in geometric parameters such as SSD and effective computation depths. I believe that, until a fully adaptive 4D planning and 4D delivery system is at hand, both geometric and dosimetric errors should be minimized by using global patient/skeletal rotational parameters and tumor-based translational setup parameters. This reduces the need for multiple fiducial implantations and could save patients from complications due to fiducial placement.2 In many cases, incorporated with global alignment, a single fiducial marker centrally located in the tumor could result in good accuracy for tumors with size suitable for SRS and SBRT (our upper limit tumor size is usually about 4 cm in average dimension for definitive SRS or SBRT to assure high dose gradient outside the PTV). Single fiducials should be used with careful management to avoid fiducial migration. The fiducial marker (preferably with antimigration features) should be placed inside the tumor. Sufficient time should elapse between the implant and the planning CT acquisition to allow the fiducial to “scar” into position. 4D-CT sets fused to the fiducial marker can be used to construct the tumor volume to account for tumor deformation and rotation. Maximum tumor/fiducial excursion can also be mapped by the planning 4D-CT to assure that during treatment the fiducial's position remains within an acceptable range. Post-treatment CT has demonstrated that there h