John Adler, MD, has functioned as neurosurgical leader of the team involved in developing the CyberKnife frameless stereotactic radiosurgery systemunquestionably one of the most innovative and promising methods for delivering high-dose precision radiation to any target in the body. He and his colleague, Steven Chang, md, are in a unique position to assess the evolution of stereotactic radiosurgery over the last decade. This includes not only a summary of the current status of the conventional indications for stereotactic radiosurgery, but also a concise report on their unique and fundamental work in the treatment of nasopharyngeal carcinoma.
Evolution of the Gamma Knife
Stereotactic radiosurgery was the brainchild of visionary Swedish neurosurgeon Lars Leksell. Almost half a century ago, by combining a neurosurgical localization device (a "stereotactic head frame") with multiple cross-firing beams of x-rays from cobalt-60 sources, he was able to deliver very precise high-dose radiation (eg, 80 to 160 Gy in a single session) to image-defined targets. His initial device, the Gamma Knife, was introduced into clinical practice in Stockholm in 1968. It was used to create small focal lesions (4 to 8 mm) deep in the brain for the treatment of neurologic disorders such as Parkinson’s disease, chronic pain, and behavioral disorders. Stereo skull x-rays provided guidance, with arteriography or pneumoencephalography used to define internal soft-tissue targets.
The armamentarium was later expanded with the addition of precisely aimed heavy particle beams (protons, for the most part, but also helium, carbon, or neon nuclei, etc) that required expensive cyclotrons for delivery. The number of therapeutic targets also expanded rapidly to include a host of benign and malignant tumors and vascular lesions in the brain.
The sea change that enabled widespread use of stereotactic radiation was the newfound capability of modifying linear accelerators into devices capable of converging x-ray beams delivered from many directions into a controlled distribution aimed at an image-defined target (using computed tomography [CT] or magnetic resonance imaging [MRI] for localization). Multiple variations on the stereotactic radiosurgery theme originally conceived by Winston and Lutz, Columbo, and Betti have evolved over the past 15 years.
The wunderkind of these techniques and devices is the CyberKnife, born of the union of a compact industrial linear accelerator with a large but precise industrial robot, and guided via real-time monitoring of stereotactic x-rays. This ingenious system allows the delivery of the fairly complex summation of multiple nonisocentric x-ray beams to achieve a three-dimensional dose distribution that is highly conformal to irregular target shapes and surrounding normal tissue anatomy. Combining the large robot with the small linear accelerator allows for a great increase in system flexibility, enabling treatment not only at the end of the cylinder (ie, the head, which is the only realistic target for other stereotactic radiosurgery systems), but also throughout the entire volume of the cylinder (ie, the body).
Neurosurgeons have grown to appreciate the power of large single doses (eg, 12 to 25 Gy) in reliably treating various tumors and vascular lesions of the brain. The evolution of the CyberKnife will confer significant power to physicians treating other targets throughout the body.
Barriers to Use
The two barriers that have limited the utility of stereotactic radiosurgery are large target volumes (generally > 25 to 35 cm3, due to the escalating volume receiving higher doses) and the proximity of very radiosensitive normal structures (ie, the optic nerves and chiasm or the cochlear nerve). Neurosurgeons have often "punted" in these situations, resorting to the use of fractionation of stereotactically delivered radiation to reestablish a safety margin for surrounding normal tissue.
Drs. Adler and Chang provide a practical summary of "fractionated stereotactic radiosurgery," although I personally regard "fractionated radiosurgery" as an oxymoron; ie, I consider all radiosurgery as single dose, and "stereotactic radiotherapy" to be a more appropriate term for fractionated treatment. The authors discuss how these techniques are especially useful for targets around the optic apparatus (using hypofractionation, ie, a few moderately sized fractions) or in the treatment of acoustic neuromas (or vestibular schwannomas).
Many targets, especially metastatic tumors, respond so well to radiosurgical doses that one would never consider wasting the patient’s time and effort with a less effective and more time-consuming fractionated course. Some targetsfor example, arteriovenous malformationsdo not respond to conventional fractionation schemes, thereby necessitating hypofractionation to achieve therapeutic effect.
Drs. Adler and Chang entitle the final section of their article "Image-Guided Stereotactic Radiosurgery"a relatively bland and uninformative heading that cloaks the underlying revelation: The CyberKnife can track the target in real time! I recall discussions I had with Dr. Adler a decade ago, deriding his vision of the complete obliteration of skull fixation during treatment. Indeed, the CyberKnife in a milder "fixed" form might have been available years ago were it not for the problems that arose with his more radical vision. Of course, the true value of this tool concerns the ability to take such powerful treatment outside of the cranium, and into targets throughout the body.
Stereotactic radiosurgery is one of the most revolutionary advances in neurosurgery in the last half century. The techniques continue to evolve, allowing safe and effective treatment of many conditions today that were simply not treatable 15 years ago. The article by Drs. Chang and Adler provides a state-of-the-art summary of the field, as well as a glimpse of the huge impact this technique will have on the treatment of cancers located elsewhere in the body.