Tomé Lab

From Human Oncology

Therapeutic Radiological Physics, Department of Human Oncology

Current Scientists (in alphabetical order)

• Dr. Nick Hardcastle (PhD, Centre for Medical Radiaton Physics at University of Wollongong, Australia (2009))

Former Scientists and their Locations (in alphabetical order)

• Dr. Hazim Jarradt, Assistant Professor University of Michigan (Ann Arbor, MI)
• Dr. Nigel Orton, Chief Medical Physicist, Cancer Care Northwest (Spokane, WA)

Current Students (in alphabetical order)

• Noah Arvidson, MS
• Ed Bender, MS
• Andrew Ellis, BS
• Dinesh Tewatia, MS (Joint with Dr. Bhudatt Paliwal)
• Dongxu Wang, MS (Joint with Dr. T. Rock Mackie)
• Jialu Yu, BS
• Heming Zhen, BS

Former Doctoral Students and their Current Locations (by year graduated)

• Dr. Susan Richardson, Instructor, Washington University (St. Louis, MO), (Joint with Dr. Bhudatt Paliwal), 2003
• Dr. Alonso Gutierrez, Assistant Professor, University of Texas (San Antonio, TX), (Joint with Dr. T. Rock Mackie), 2007
• Dr. Yusung Kim, Assistant Professor, University of Iowa (Iowa City, IA), 2007. On the web at [1]
• Dr. Eric Ehler, Radiation Oncology Physics Resident, University of Minnesota (Minneapolis-St.Paul, MN), 2009
• Dr. Karl Rasmussen (Joint with Dr. Steven Howard)
• Dr. Emilie Soisson, Staff Medical Physicist, McGill University Health Centre, Montreal, Canada, 2009
• Dr. Leah Schubert, Instructor, University of Nebraska (Omaha, NE), (Joint with Dr. Bhudatt Paliwal), 2009
• Dr. David Westerly, Assistant Professor, University of Colorado (Denver, CO), (Joint with Dr. T. Rock Mackie), 2009

Research Topics and Related Publications

Deformable image registration

There is a lot of interest currently in trying to account for interfraction motion (i.e. changes in the patient anatomy between treatment fractions) and intrafraction motion (i.e. motion during treatment) in the planning process. Examples include 4D planning which is a project listed below, as well as adaptive radiotherapy where a patients changing anatomy is accounted for and changes to the treatment plan are made when necessary. One key technology in these techniques is deformable registration because one is interested in the total dose delivered either over several fractions (adaptive) or during a fraction (4D). The problem is that the patient images during different fractions (adaptive) or different time points (4D) are of course different, and we need to derive deformation maps so that all image sets can be referenced to a single baseline image. We need to do this because we want to know what dose each voxel in the patient recieves in total after several fractions or during an extended period of time, so that we can derive overall dose-volume-histograms. The only way to do this is with knowledge of what voxel goes where, and when. Deformable registration is the technique used to attempt to derive deformation maps between fractions or time points, i.e. to determine what voxel goes where. The huge challenge in this is that there is no gold standard as to how this can be done. There are a dizzying number of different algorithms that have been developed, most of which are based on some sort of simplified physical model of how flexible materials deform. The challenge of developing these algorithms is not the only problem. Perhaps more important is determining whether or not they work. This may sound silly, but there is no clear-cut way to figure out whether or not your deformations are correct, or whether in fact they just look pretty and are completely wrong. It is useful to draw an analogy with dose calculations. A lot of work certainly went in to methods of calculating the dose in a patient, and along the way those who developed the method had a simple way to figure out if their methods worked: They would simply design a phantom of some sort and measure the dose and compare it to the calculation. Dose can be measured at a point very accurately with an ion chamber, and in a plane with somewhat less accuracy using film. Therefore those that develop dose calculation algorithms have a relatively easy way of testing their calculations. Now let's think about deformable registration - how does one test their calculations? For dose calculation purposes one can make a reasonable facimilie of a human by making phantoms of different materials - for example a real skeleton embedded in plastic representing the approximate density of tissue and bone. But how does one make a facimilie of the human body when what's important is not simply the density of the organs but all of their complex mechanical properties? The simple answer is that one does not exist. What's more, even if one did exist there is no easy way to measure a 2-D grid of deformation values like one can do easily with dose and film. Because of these considerations in our lab we are trying to come up with ways to at least attempt to do quality assurance of deformable image registration algorithms. There are certianly some ways of doing this already, like looking at difference images, tracking the boundaries between organs, and tracking easily identifiable points like branches in blood vessels, but what exists now is not complete. We will be adding some pretty pictures and more descriptions of what we are doing soon.

• Publications:
  • The utilization of consistency metrics for error analysis in deformable image registration." Bender ET, Tomé WA. Phys Med Biol. 54(18):5561-77, 2009[2]

Whole brain radiotherapy with hippocampal avoidance

The hippocampus is a deep brain structure (actually a pair of structures, one left and one right) which contains neural stem cells. Damage to the hippocampus during radiotherapy may lead to neurocognitive impairment. It therefore may be desirable to spare a patients hippocampus during whole brain radiotherapy (a typical treatment for patients with multiple brain metastases). The difficulty in this is that it is hard to see the hippocampus on MRI images. We are in the process of developing an automatic method for generating a conformal avoidance region for a patients hippocampus. The figure below shows an example of this automatically generated conformal avoidance region in green, and for comparison the manually segmented hippocampus in red. Note that the green contour was generated automatically, without using the actual hippocampus.

• Publications:
  • "Distribution of Brain Metastases in Relation to the Hippocampus: Implications for Neurocognitive Functional Preservation." Ghia A, Tomé WA, Thomas S, Cannon G, Khuntia D, Kuo JS, Mehta MP. International journal of radiation oncology, biology, physics, (2007) [3]
  • "Whole Brain Radiotherapy With Hippocampal Avoidance and Simultaneously Integrated Brain Metastases Boost: A Planning Study." Alonso Gutiérrez, David Westerly, Wolfgang Tomé, Hazim Jaradat, Thomas Mackie, Søren Bentzen, Depak Khuntia, Minesh Mehta, International Journal of Radiation Oncology Biology Physics 69 pp. 589-597 (2007) [4]

4D-Patient Management and QA of 4D-Radiotherapy

Lung cancer is the leading cause of cancer deaths in the United States for both men and women. Intrafraction motion, which is the motion of tissues during the delivery of therapeutic radiation, is a significant problem in the treatment of lung tumors using radiotherapy primarily due to respiratory motion. Breathing motion affects imaging, treatment planning, and the delivered dose in radiation therapy. Our research on this subject so far has focused on the estimation of the interplay effect, the segmentation of mobile lung tumors, and the development treatment planning methods that allow one to consider temporal changes. The ultimate goal of this research is the development of novel methods to plan, quality assure, and deliver conformal radiation to a mobile target while reducing dose to critical normal anatomy. This includes investigation of the following subjects: tumor motion induced image artifacts in 4D-CT, prediction and mitigation of the interplay effect (i.e. tumor underdose and temporal dose variation) using computer simulation for pretreatment quality assurance, the development of strategies that allow one to reduce or mitigate the interplay effect, and 4D treatment planning strategies.

• Publications:
  • "On the automated definition of mobile target volumes from 4D-CT images for stereotactic body radiotherapy." Zhang T, Orton NP, Tomé WA. "Med Phys." 2005;32(11):3493-502.[5]
  • "On the Dose Delivered to a Moving Target When Employing Different IMRT Delivery Mechanisms", Ehler E, Nelms BE, and Tomé WA, Radiotherapy and Oncology 83, p. 49-56 (2007). [6]
  • "A 4D IMRT/SBRT and respiratory gating QA device for patient-specific intra-fraction motion kernels" Nelms BE, Ehler E, Bragg H, Tomé WA, Journal of Applied Clinical Medical Physics 8, pp. 152-168 (2007). [7]
  • "Lung 4D-IMRT treatment planning: An evaluation of three methods applied to four-dimensional data sets." Ehler ED, Tomé WA. Radiotherapy and Oncology 88(3):319-325, 2008. [8]
  • "A method to automate the segmentation of the GTV and ITV for lung tumors." Ehler ED, Bzdusek K, Tomé WA. Med Dosim. 34(2):145-53, 2009. [9]
  • "Step and shoot IMRT to mobile targets and techniques to mitigate the interplay effect." Ehler ED, Tomé WA. Phys Med Biol. 54(13):4311-24, 2009. [10]

Pulsed reduced dose-rate radiotherapy

Pulsed reduced dose-rate radiotherapy is a radiotherapy technique in which the dose-rate at which the treatment is delivered is substantially reduced. This is done by delivering the entire treatment dose as a series of 0.2 Gy pulses separated by 3 minute time intervals, creating an apparent dose rate of 0.0667 Gy per minute.

• publications:
  • "Pulsed reduced dose-rate radiotherapy: case report: a novel re-treatment strategy in the management of recurrent glioblastoma multiforme.", George M. Cannon, Wolfgang A. Tomé, H. Ian Robins and Steven P. Howard, Journal of Neuro-Oncology 83, 307-311 (2007) [11]
  • "On the possible increase in local tumour control probability for gliomas exhibiting low dose hyper-radiosensitivity using a pulsed schedule", W A Tomé, and S P Howard, The British Journal of Radiology 80, 32-37 (2007) [12]
  • "Pulsed reduced dose-rate radiotherapy: a novel locoregional retreatment strategy for breast cancer recurrence in the previously irradiated chest wall, axilla, or supraclavicular region.", GM Richards, WA Tomé, HI Robins, JA Stewaer, JS Welsh, PA Mahler, SP Howard, "Breast Cancer Res Treat." "2008 Apr 4. [Epub ahead of print] PMID: 18389365 [13]

Theragnostics and Risk Adaptive Radiotherapy

Risk Adaptive Radiotherapy is a biological optimization strategy that is based on the possible risk characteristics for local recurrence in tumor sub-volumes rather than individual tumor voxels and treatment plans are optimized using biological objective functions that are region specific, rather than voxel specific. Risk Adaptive optimization can be employed in the generation of theragnostic treatment plans.

• Publications:
  • "Risk-adaptive optimization: selective boosting of high-risk tumor subvolumes" Yusung Kim and Wolfgang A Tomé, International Journal of Radiation Oncology, Biology, Physics 66, pp. 1528–1542 (2006) [14]
  • "Radiobiological and treatment planning study of a simultaneously integrated boost for canine nasal tumors using helical tomotherapy." Gutíerrez AN, Deveau M, Forrest LJ, Tomé WA, Mackie TR. "Vet Radiol Ultrasound." 2007;48(6):594-602.[15]
  • "Optimization of radiotherapy using biological parameters." Kim Y. and Tomé WA, "Cancer Treat Res." 2008;139:257-78. Review [16]
  • "Is it beneficial to selectively boost high-risk tumor subvolumes? A comparison of selectively boosting high-risk tumor subvolumes versus homogeneous dose escalation of the entire tumor based on equivalent EUD plans." Kim Y. and Tomé WA, "Acta Oncol." 2008;47(5):906-16. [17]
  • "On the impact of functional imaging accuracy on selective boosting IMRT." Yusung Kim and Wolfgang A Tomé, "Phys Med." 2008 Jan 16. [Epub ahead of print] PMID: 18206411 [18]

Dosimetry in Regions of Electronic Disequilibrium

• Publications
  • "Dosimetric verification of helical tomotherapy for total scalp irradiation." Hardcastle N, Soisson E, Metcalfe P, Rosenfeld AB, Tome WA. Med Phys. 35(11):5061-8, 2008.[19]
  • "Endo-rectal balloon cavity dosimetry in a phantom: performance under IMRT and helical tomotherapy beams." Hardcastle N, Metcalfe PE, Rosenfeld AB, Tomé WA. Radiother Oncol. 92(1):48-56, 2009.[20]

Image guided and bioeffect parameter guided Stereotactic Radiotherapy

High precision intracranial radiotherapy:

High precision intracranial radiotherapy (HPIR) has evolved from Fractionated Stereotactic Radiotherapy (FSRT) and just as FSRT is a hybrid between conventionally fractionated radiotherapy and stereotactic radiosurgery that incorporates conventional fractionation schemes with stereotactic localization and targeting techniques to spare critical neurological structures. In the early FSRT experience treatments were given using a multiple-arc set delivered through circular collimators which for highly irregularly shaped lesions lead to the inclusion of a large amount of normal brain and yielded inferior conformality. However, more generally one can think of such a system as a Patient Localization System with submillimetric accuracy. Using this point of view I have developed HPIR employing a High Precision Patient Localization System (HPPLS). In this technique a 3D conformal radiotherapy treatment plan is generated following three primary considerations; beam directions are chosen to (a) provide each treatment beam with a unique entrance and exit pathway, (b) avoid critical structures, and (c) minimize the beam’s-eye-view projection of the target volume. In clinical practice HPIR works as follows: At the time of CT-simulation a custom made bite plate for localization, and a custom made head cushion and head mask for immobilization are manufactured. The patient is scanned in the treatment position with the bite plate linked to his or her maxillary dentition, to which an array of passive infrared markers has been attached. Hence, the localization system is encoded into the patient image set. The image set is then imported into a three-dimensional (3D) radiotherapy planning system. The tumor and critical structures are delineated and the treatment isocenter is established. A treatment plan is generated using five different couch positions, and the gantry angles are chosen such that each beam has a unique entrance and exit pathway, avoids the critical structures, and has a minimal beam’s eye view projection of the target. The treatment plan is exported to the HPPLS. The optical tracking system that is part of the HPPLS emits light in the infrared range, and detects the reflected infrared light from passive markers. The passive markers are identified in the patient image set and a reference is established. Using this reference, the expected patient position in the treatment room is calculated using the isocenter coordinates determined during treatment planning. On the first day the patient is localized in the treatment room using the HPPLS to within 0.3mm translation error and 0.3 degree rotation. The computer reports the patient's current position based on the location of the passive markers relative to the established reference position. Using this readout from the computer, the patient is setup to the desired position in between for each treatment fraction. The use of the HPPLS in conjunction with noncoplanar radiotherapy treatment planning using fixed fields allows one to generate of highly conformal treatment plans that exhibit a high degree of dose homogeneity and a steep dose gradient. Since the HPPLS improves the accuracy of patient localization relative to the linac isocenter and allows real-time monitoring of patient position, one can choose treatment field margins that only account for beam penumbra and image resolution without adding margin to account for larger and poorly defined setup uncertainty. Therefore, using this approach allows one to enhance the normal tissue sparing, the degree of conformality and the homogeneity characteristics that one could achieve using only 3D conformal radiotherapy.

• Publications:
  • Tomé WA, Meeks SL, Li Z, Buatti JM, Bova F, Friedman W: A High Precision System for conformal intracranial radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 47(4): 1137—1143, 2000.[21]
  • Tomé WA, Meeks SL, McNutt TR, Buatti JM, Bova FJ, Friedman WA, Mehta M: Optically Guided Intensity Modulated Radiotherapy. Radiother Oncol 61(1): 33—44, 2001.[22]
  • Tomé WA, Mehta MP, Meeks SL, Buatti JM: Fractionated Stereotactic Radiotherapy: Present and Future. Technol. in Cancer Res. and Treat. 1(3): 153—172, 2002.[23]
  • Meeks SL, Tomé WA, Willoughby TR, Kupelian PA, Wagner TH, Buatti JM. Optically Guided Patient Positioning Techniques. Sem. Rad. Oncolo. 15 (3): 192-201, 2005.[24]
  • Wagner TH, Meeks SL, Bova FJ, Friedmann WA, Willoughby TR, Kupelian PA, Tomé WA. Optical tracking technology in stereotactic radiation therapy. Med. Dosim. 32:111-20, 2007.[25]

Image and bioeffect parameter guided Stereotactic Body Radiotherapy:

Image and bioeffect parameter guided Stereotactic Body Radiotherapy uses image guidance for target localization for the treatment of T1/T2 N0M0 non-small cell lung cancer. Even though this new and very exciting field is still in its infancy, it however offers tremendous possibilities for curative and palliative patient care. In stereotactic body radiotherapy the patient receives the entire treatment in a time span of two weeks in 3 to 5 high dose fractions ranging form 12 to 15Gy a fraction. Therefore, it is absolutely crucial that the patient position during treatment can be monitored and that target motion due to breathing and involuntary patient motion is minimized. We have developed and implemented a image guided system for stereotactic body radiotherapy using Tomotherapy that allows for reproducible patient setup, guarantees minimal patient motion during treatment, as well as allows one to restrict the internal motion of the target during treatment. Moreover, we have developed a bioeffect parameter model to select depending on target size an appropriate fraction schema that allows one to maximize the expected tumor control while keeping the major expected toxicity of this treatment, namely the incidence of radiation pneumonities of grade 2 or higher, below 20%.

• Publications:
  • Folwer JF, Tomé WA, Fenwick JD, Mehta MP. A challenge to conventional radiation oncology. Int. J. Radiat. Oncol. Biol. Phys. 60(4): 1241—56, 2004. [26]
  • Zhang T, Orton NP, Tomé WA. On the automated definition of 4D target volumes for Stereotactic Body Radiotherapy from 4D-CT. Med Phys. 32(11): 3493—3502, 2005. [27]
  • Tomé WA, Fenwick JD, Mehta MP. How can tumor effect and normal tissue effect be balanced in stereotactic body radiotherapy. Radiosurgery (6): 87—97, 2006.[28]
  • Hodge W, Tomé WA, Jaradat HA, Orton NP, Khuntia D, Traynor A, Weigel T, Mehta M. Feasibility report of image guided stereotactic body radiotherapy (IG-SBRT) with tomotherapy for early stage medically inoperable lung cancer using extreme hypofractionation. Acta Oncologica 45: 890-96, 2006.[29]
  • Arvidson NB, Mehta MP, Tomé WA. Dose coverage beyond the GTV for various Stereotactic Body Radiotherapy (SBRT) planning techniques reporting similar control rates for stage I non-small-cell lung cancer. Int J. Radiat. Oncol. Biol. Phys. 72(5):1597-603, 2008 [30].

Development of a hybridized Optically and Image Guided Stereotactic Radiosurgery System for Tomotherapy:

Intracranial Conformal Avoidance Radiotherapy

Intracranial Conformal Avoidance Radiotherapy (ICAR) has evolved from the FSRT and just as FSRT is a hybrid between conventionally fractionated radiotherapy and stereotactic radiosurgery that incorporates conventional fractionation schemes with stereotactic localization and targeting techniques to spare critical neurological structures and allows one to conformally avoid eloquent brain that harbors areas critical to neurocognitve function identified form fMRI.

• Grants: CA R01-109656

Image Guided and Dose Guided Radiotherapy

• Publications:
  • Tomé WA, Jaradat HA, Nelson IA, Ritter MA, Mehta MP. Helical Tomotherapy: Image Guidance and Adaptive Dose Guidance. Front Radiat Ther Oncl. 40:162-78, 2007. Review[31]

Treatement planning for GBM using physiologic MRI data:

Our overall goal in this project is to learn how to design better radiotherapy treatement plans for patients with GBM, whose survival rate is currently less than 3% after 5 years. We are currently investigating the utility of various physiologic MRI data for use in the treatement planning process.

• Most of the patients who receive treatement for GBM experience a recurrence of the disease within 2-4 cm of the treatement margin.
• Some patients have a recurrence within the original treatement volume. Therefore, subvolume boosting may be benificial, if high risk areas can be identified
  • "Selective Boosting of Tumor Subvolumes", Wolfgang A. Tome, Jack F. Fowler, International Journal of Radiation Oncology, Biology, Physics 48 pp. 593-599 (2000) [32]
  • On the incorporation of multi-modality image registration into the radiotherapy treatment planning process." Hazim A Jaradat, Wolfgang A Tomé, Todd R McNutt, M. Elisabeth Meyerand, "Technol Cancer Res Treat." 2003 2(1):1-12. [33]
• The MRI modalities (aside from T1/T2 weighted imaging) that we will use in this study include:
  • Hypoxia mapping using blood oxygen level dependent (BOLD) MRI with carbogen breathing
    • The lack of oxygen in hypoxic cells leads to a resistance to radiation
    • The use of carbogen breathing will help in distinguishing between tumors and necrosis or dried blood
    • We will attempt to identify regions of chronic hypoxia and use this information in the treatement plans
  • Chemical shift imaging (CSI) and the choline:N-acetylaspartate index (CNI)
    • Several metabolites can be detected, including choline, creatine, N-acetylaspartate, and lactate
    • The metabolic fingerprint of tumors is different than that of normal brain tissue and this can be used in the treatement planning process
  • Contrast enhanced perfusion imaging
    • Used to measure blood flow and volume
    • Hypervascular regions show up as areas of high signal intensity
    • Cerebral blood volume can be used to identify highly proliferative tumors
  • Diffusion imaging and the apparent diffusion coefficient (ADC)
    • Regions with a low ADC indicate a high cellularity and may indicate a region of rapid tumor growth

Image guided Intensity Modulated Proton Therapy

Radiotherapy using targeted radionuclides

Useful Links

• The American Association of Physicists in Medicine [34]

• The American Board of Radiology - Radiologic Physicists [35]