Current State of Cancer Imaging—SOTAL
Donald Thrall United States
Donald E. Thrall, DVM, PhD
Dr. Thrall graduated from the Purdue University Veterinary School in 1969 and completed Master of Science and Doctor of Philosophy degrees at Colorado State University in 1971 and 1974, respectively. He held faculty positions at the University of Georgia and the University of Pennsylvania before joining the faculty of North Carolina State University in 1982 where he is currently Professor in the Department of Anatomy, Physiological Sciences and Radiology. Dr. Thrall is board certified by the American College of Veterinary Radiology in both diagnostic radiology and radiation oncology, and has served as President of the College in both capacities.
Dr. Thrall's primary imaging interests are diagnostic radiographic and computed tomographic imaging, particularly of tumors, and experimental and clinical radiation oncology. His current research interest is hyperthermia as an adjunct to cancer therapy. He has authored or co-authored over 180 publications, and is Editor of Veterinary Radiology & Ultrasound the official journal of the American College of Veterinary Radiology, the International Veterinary Radiology Association and the European Association of Veterinary Diagnostic Imaging. He is also the Editor of the Textbook of Veterinary Diagnostic Radiology, published by W.B. Saunders, Co.
Diagnostic radiology’s use in cancer imaging has diminished as newer and more sophisticated modalities have come on line. Today, the main use of diagnostic radiology is for lung metastasis screening and for assessment of bone tumors. In practice, radiologic imaging has fairly limited value for other assessments relating to the cancer patient.
Diagnostic radiology is less sensitive than computed tomography (CT) for evaluating the lung for masses. However, routine screening of the lung using CT has not become established policy in veterinary medicine.
It is absolutely critical that technically excellent radiographs be taken. Poor radiographs are the biggest source of interpretation error in veterinary medicine. Three views of the lung should be routine. Atelectasis of the dependent lung with its increase in opacity, will silhouette with lesions in the dependent lung making them invisible. Thus, failure to make three views will increase one’s interpretive error.
In one study, it was concluded that two views were as accurate as three for detection of lung masses.(1) However, neither the severity nor the number of metastatic lesions present were clearly described. When there are multiple masses or masses are large, even one view will be sufficient. More often, however, there are small or only a few masses and three views are needed for a thorough examination of the lung.
There are multiple pitfalls regarding diagnosis of a lung mass. End-on pulmonary vessels, hetertopic bone, and thoracic wall masses can all be confused with a lung mass. Experience and a systematic review of the radiograph will help in correctly identifying these other conditions. For example, end-on vessels are often adjacent to a bronchus and can be seen to connect with an adjacent vessel. In addition, while heterotopic bone is mineralized and appears very opaque, lung metastases rarely mineralize.
Ultrasonography has revolutionized veterinary medicine. Its real-time features and facilitation of tissue biopsy have been major improvements in clinical practice. It has a very high sensitivity for detecting parenchymal lesions, but its specificity is poor. Thus, incidental and nonspecific findings are commonplace. Clinicians feel compelled to investigate such findings and this has added to the time and expense needed to completely evaluate a patient, especially one being evaluated for cancer staging. Unexpected or nonspecific findings must be interpreted in light of the patient’s problem, and in some instances, it is appropriate to ignore them.(6) However, one must realize that ignoring an abnormality may lead to an incorrect diagnosis.
The “high end” aspects of diagnostic ultrasound include Doppler (to assess blood flow), use of ultrasonographic contrast media, three-dimensional imaging, and detection of tissue harmonics (the tissue’s response to being insonated) to improve resolution. At this time, these features have not contributed measurably to the diagnosis or staging of cancer.(3)
The invention of computed tomography revolutionized imaging in humans. For the first time, anatomy could be assessed in tomographic slices. Additionally, the improved contrast enhancement of CT enables discrimination between tissues that previously were indistinguishable radiographically. An example is detection of acute brain hemorrhage with CT because of the greater attenuation of blood in relation to brain. CT has had less of an impact in veterinary medicine, but it is still a valuable modality and use continues to increase.
Initially, the major use of CT in veterinary medicine was for brain imaging, but this has been almost entirely supplanted by magnetic resonance imaging. In cancer patients, CT is of great use for staging the extent of disease and planning for radiation therapy. Basing the extent of a tumor, or planning a complicated tumor excision without evaluation of the tumor using CT, is very inaccurate. Tumor extension assessed from CT imaging is typically further advanced than expected from radiographic or clinical assessments.(8)
With regard to radiation planning, the use of CT has revolutionized the art. It is now possible to configure treatment beams more precisely. This reduces the radiation dose received by normal tissues and provides for assessment of the radiation dose delivered to the tumor and normal tissue. By trial and error, the radiation oncologist can devise a treatment plan to minimize normal tissue dose and maximize tumor dose. With sophisticated systems, radiation dose can be viewed in multiple planes or even in three-dimensional contours.
CT and MR can be used for some physiologic assessments. For example, by quantifying the uptake of contrast medium in a tumor following intravenous injection, it is possible to compute a parameter that is related to tumor perfusion. Inturn, perfusion may be related to tumor oxygenation, which has an effect on both the biologic aggressiveness of the tumor as well as its response to irradiation. Perfusion has been quantified in canine nasal tumors and there is considerable heterogeneity in this parameter between patients. Thus, it may be a factor, which when measured prior to therapy, may predict which patients are likely to respond favorably. More work is needed to be done, but assessing perfusion using CT is a reasonable parameter to pursue.(9)
In cancer patients, nuclear imaging has had its widest use in evaluating patients with osteosarcoma. Bone scintigraphy can be used to detect metastasis and in assessing biologic behavior.
Bone metastases from osteosarcoma is not common, but it can be a significant complicating factor if an expensive procedure, such as limb sparing, is being considered. In this instance, it is important to know if metastatatic disease is present prior to the procedure. A search for skeletal metastases can be done using radiography, but this is time consuming and there is a radiation hazard for personnel. Scintigraphy provides an alternative method. Based on available information, approximately 2%–6% of dogs with osteosarcoma will have extrapulmonary metastases that can be detected with either radiography or scintigraphy.(2,5,7) Scintigraphic detection of non-neoplastic bone lesions, such as degenerative joint disease, is definitely problematic. These lesions need to be worked-up and result in increased expenditure of time and money; they are similar in principle to unexpected sonographic findings.
Quantitative scintigraphy of primary bone tumors has been shown to be related to tumor aggressiveness. Tumors characterized by a high level of uptake of the bone-seeking radionuclide are associated with more rapid metastasis. This, pretreatment scintigraphy provides an opportunity to identify tumors likely to be associated with early metastasis.(4)
Magnetic Resonance Imaging
Magnetic resonance (MR) imaging is characterized by exquisite contrast resolution. Additionally, imaging parameters can be varied to maximize signal from a particular tissue type; i.e. water, fat, cartilage. Because of the dependence of local biochemical events on the source of signal generation, information about the chemical makeup of tissue can be attained.
The major use of MR imaging in veterinary medicine is neuroimaging. MR imaging has revolutionized the diagnosis of brain tumors and inflammatory and degenerative brain disorders as well.
Using the spectroscopic capabilities of MR, high-energy phosphate metabolism can be quantified and intracellular pH determined. For example, the intracellular pH of spontaneous canine tumors has been found to be higher than would be expected based on extracellular pH measurements. Thus, the tumor cells have a large ability for buffering. The resultant extra:intra cellular pH gradient will facilitate the intracellular accumulation of chemotherapeutic agents that are weakly acidic. Thus, identification of such tumors or acidification through interventional means may be a future tactic used to improve the response to certain types of chemotherapy.(10)
The next horizon in cancer imaging is biologic/molecular imaging. By binding of appropriate compounds in vivo, or by competitive inhibition with radiotracers, the metabolic characteristics of the tumor may be evaluated using imaging. Imaging hypoxia and lactate are two examples where tumor physiologic parameters can be assessed using imaging methods.
Diagnostic radiology has assumed a diminished role in tumor diagnosis and staging. More sophisticated and expensive modalities are taking its place. Many owners expect these tools to be available. It will be a challenge for the profession to acquire and maintain these tools; whether this activity can be sustained in the event of economic downturn is unknown. Regardless, the implementation of these “high-tech” tools has enhanced learning and improved the response to therapy.
1. Barthez, P., Hornof, W., Theon, A., Craychee, T., and Morgan, J., Receiver operating characteristic curve analysis of the performance of various radiographic protocols when screening dogs for pulmonary metastasis. J Am Vet Med Assoc. 204: 237-240, 1994.
2. Berg, J., Lamb, C., and O'Callaghan, M., Bone scintigraphy in the initial evaluation of dogs with primary bone tumors. J Am Vet Med Assoc. 196: 917-920, 1990.
3. Burns, P., Ultrasound contrast agents in radiological diagnosis. Radiol Med. 87: 71-82, 1994.
4. Forrest, L., Dodge, R., Page, R., et al., Relationship between quantitative tumor scintigraphy and time to metastasis in dogs with osteosarcoma. J Nuc Med. 33: 1542-1547, 1992.
5. Hahn, K., Hurd, C., and Cantwell, H., Single-phase methylene diphosphate bone scintigraphy in the diagnostic evaluation of dogs with osteosarcoma. J Am Vet Med Assoc. 196: 1483-1486, 1990.
6. Pauker, S. and Kopelman, R., Trapped by an incidental finding. Am J Roentgen. 160: 233-236, 1993.
7. Straw, R., Cook, N., LaRue, S., Withrow, S., and Wrigley, R., Radiographic bone surveys (letter). J Am Vet Med Assoc. 195: 1458, 1989.
8. Thrall, D., Robertson, I., McLeod, D., et al., A comparison of computed tomographic findings in 31 dogs with malignant nasal neoplasia. Vet Radiol. 30: 59-66, 1989.
9. vanCamp, S., Fisher, P., and Thrall, D., Dynamic CT measurement of contrast medium washin kinetics in canine nasal tumors. Vet Radiol and Ultrasound. 41: , 2000.
10. Vujaskovic, Z., Poulson, J., Gaskin, A., et al., Temperature dependent changes in physiologic parameters of spontaneous canine soft tissue sarcomas after combined radiotherapy and hyperthermia treatment. Int J Radiat Oncol Biol Phys. 46: 179-185, 2000.
WSAVA Contact Information