Three-dimensional medical imaging has revolutionized radiological diagnosis and treatment planning, with reconstructions providing lifelike views of anatomic structures. In the field of oncology, 3D reconstructions obtained from CT scans are being used to facilitate treatment plans for prostate cancer, lung tumors, brain tumors, head and neck tumors, gynecological tumors, and pediatric tumors. More than a decade ago, 3D CT data helped drive the important development of 3D conformal radiation therapy (3D-CRT), which made it possible to improve tumor localization and minimize doses to surrounding healthy tissue. More recently, 3D-CRT has evolved into the advanced delivery of intensity modulated radiation therapy (IMRT). IMRT more precisely shapes high-dose radiation to the cancerous volume, sculpting it around adjacent normal tissues. This makes it possible to deliver even higher doses to tumors without increasing side effects. The development of advanced therapeutic methods is made possible by the availability of 3D CT data.
According to John Wong, PhD, director of clinical physics in the Department of Radiation Oncology, William Beaumont Hospital, Royal Oak, Mich, the majority of, if not all, US clinics use 3D volumes for treatment planning, while some 50% of them have a dedicated CT scanner in the department.
“Imaging and radiation treatment planning have always gone hand in hand, and enhancements in imaging are currently having a major impact in treatment planning,” says Michael Zelefsky, MD, chief of brachytherapy service at Memorial Sloan-Kettering Cancer Center, New York. “For instance, IMRT has allowed us to use dose levels that far exceeded what we ever thought we could deliver to the prostate gland. We have observed that higher doses translated into higher tumor control rates for all subsets of prostate risk groups.”
Growing usage of volumetric data in concert with the latest treatment technology has compelled radiation oncologists to sharpen their focus on problem areas. They have also been searching for systems that can provide real-time 3D image information that allows the most accurate treatment delivery. “We have gone to full volumes, so now the question is How good is this static snapshot image?'” Wong says. “Three-dimensional imaging gave us a much better appreciation of where the tumor is, but we realize that this snapshot is not continually accurate. We need a better understanding of how the patient and organs are moving.”
If the motion is predictable, he says, a series of scans are taken early in treatment to understand how to set up the margins for that particular patient. If the motion changes too much from day to day, however, then scanning may be required before each treatment. Conversely, some clinics use radiomarkers implanted in the body because they do not have the resources to do such numerous CT scans.
“But we need to be careful that the CT scan taken on the day before treatment does not lure us into thinking that we can treat the tumor with very, very small margin,” he says. “We need to be concerned that organs can move during the 15- to 20-minute treatment session.
“With prostate cancer, we worry about the day-to-day treatment, like how the prostate might not move as much during a single session as compared to the lung,” Wong says. “But depending on what the rectal contents are at the time of treatment, the prostate can move a few millimeters also, and we need to get the treatment over quickly to make sure that the CT is good.
“With lung cancer, we have to deal with cyclical motion of the tumor during treatment,” Wong continues. “There are two ways of working with that: you can gate the patient, only turning on the machine at certain points during the cycle, or you can use the deep breath hold and a computerized assisted breath hold mechanism. In that case, we implement a breath hold to immobilize the patient, then turn the beam on.
“We can achieve a significant reduction in our treatment margin,” Wong says. “But there is a limit to it. There is a time dependency in terms of how well the reduced margin will hold.”
The Dynamics of Work Flow
While the ideal approach for image-guided treatment may sometimes involve taking more images of the patient during the course of treatment, that remains too labor-intensive for many facilities. Image guidance work flow therefore must be changed in many locations in the clinic to take full advantage of the real-time images provided by 3D software. One element that streamlines this process is the fact that images already are sent to treatment planning stations in a seamless manner.
“Generally, the CT scan images are immediately conveyed to the treatment planning computer,” Zelefsky says.
“In the 3D world, the physician and technologist no longer need to spend as much time simulating the treatment,” Wong adds. “All the information goes straight from the computer to the machine.”
What follows is a lot of coordination between the physicians, the physicists, the therapists, and the computer planners. At Beaumont, the radiation oncology department has combated the work-flow issue by consolidating staff into clinical and technical divisions. The clinical side involves physicians and nurses, and the technical side includes the radiation therapists, the dosimetrists, and the physicists.
“We are careful to make sure this whole chain understands how the patient flows from planning into treatment, because the therapists who set the patient up on the couch for treatment are no longer involved in the simulation,” Wong says. “We improve the quality of treatment with more information, and we speed up treatment because all the changes are performed via network.”
There are several newer technologies designed to expedite treatment with the added information supplied through 3D re-creations. Tomotherapy is a slice-based delivery system that uses an accelerator on a helical CT scanner gantry. A second innovation, which Wong has been involved with along with David A. Jaffray, PhD, the head of radiation physics at the University of Toronto, uses a kV cone-beam CT imaging system based on a large-area, flat-panel detector adapted to a medical linear accelerator.
“We are putting a CT scanner on an accelerator or vice versa to facilitate image guidance,” Wong says. “We then fuse our images with biological images, and whatever new markers we get we’ll incorporate into treatment. What we want to do is eventually evolve radiation therapy into short-course treatment or something almost like radiosurgery.”
Economics and Outlook
With technology and work-flow advances must also come an evolution in economics, which is not currently the case for 3D imaging used to facilitate treatment planning.
“We are still at the beginning of this era of image guidance in terms of reimbursement,” Wong says. “We don’t get reimbursement for image-guided CRT, though IMRT does give a reimbursement that is two to three times higher than 3D-CRT.”
In the future, the reimbursement issue will have to take into account the use of other modalities in conjunction with treatment planning, such as PET for cases of lung cancer. “Information from MR spectroscopy and PET scanning already is being incorporated into treatment planning systems to ensure that regions within the target that contain greater concentration of tumor cells receive an intense dose of radiation. Such approaches in the future may allow the radiation oncologist to treat even less volume of normal tissue, limit the radiation to be delivered to areas where functional imaging tells us the tumor is located, and reduce the side effects of therapy,” Zelefsky says.
Overall, image guidance represents a change in the philosophy of treating cancer. “We are truly entering an era of 3D image-guided therapy,” Wong asserts.
