Radiosurgery combines stereotactic techniques (to achieve precise three-dimensional localization) with highly focused high-energy radiation treatments. This procedure makes it possible to deliver a very large dose of radiation to a small target while minimizing the dose outside the target volume. A previous report surveyed the clinical results and complications associated with radiosurgery for the treatment of metastatic tumor, primary glioma, arteriovenous malformation (AVM), and benign conditions including pituitary tumor, acoustic neuroma, and meningioma.
Over the past several years, however, radiosurgery has advanced rapidly. Physicians now have a better understanding of the use of radiosurgery in the treatment of brain metastases, including whether or not the addition of whole-brain irradiation provides any advantage over stereotactic radiosurgery alone. Large AVMs, previously considered difficult to treat, are now routinely treated with radiosurgery either through staged treatments or as part of a multimodality regimen. Radiosurgery has also evolved as a potential therapy following standard radiation for nasopharyngeal carcinoma. Improved local control and survival rates have been reported for this common cancer.
Advances in both technique and radiosurgery equipment have led to the use of hypofractionated radiosurgery in selected cases (which is believed to reduce patient morbidity) and frameless image-guided radiosurgery (which has resulted in both increased flexibility of treatment, as well as treatment of lesions outside the head). The following discussion focuses on our experience with these radiosurgical advances at Stanford University.
Stereotactic radiosurgery has emerged as a treatment for patients with brain metastasis, either with or without whole-brain radiation therapy (WBRT), with an 85% to 95% control rate.[2-7] Median survival after radiosurgery (6.4 to 10 months)[3,4,6-12] is comparable to survival rates that have been reported with surgical resection followed by conventional fractionated radiotherapy (4 to 13 months),[13,14] with good local control.[3,4,6-8,10-12,15,16] Recent studies have shown that patients treated with radiosurgery for either one or two brain metastases have a prolonged survival similar to that achieved with surgical resection.[2,5,9]
The success of stereotactic radiosurgical treatment of brain metastases has raised the issue of whether WBRT in conjunction with radiosurgery represents a treatment advantage over radiosurgery or WBRT alone. Reporting a series of 105 patients treated from 1991 to 1997, Sneed et al observed that response rates with radiosurgery alone were not statistically different from those achieved with radiosurgery plus WBRT, with respect to freedom from progression (71% vs 79%) and median survival (11.3 vs 11.1 months). In contrast, Fuller et al reported a series of 27 patients with 41 metastatic tumors and noted that radiosurgery plus WBRT achieved statistically superior disease control (P = .0007), compared with radiosurgery alone.
In a review of 97 patients who had been treated for two to four brain metastases, Schoeggl et al concluded that stereotactic radiosurgery provided an equivalent rate of survival, compared to the historic experience of patients treated with WBRT. However, their review did not compare a subset of patients treated with radiosurgery alone to those treated with radiosurgery plus WBRT.
Kondziolka et al reported a randomized trial of WBRT vs radiosurgery plus WBRT, which showed that WBRT combined with radiosurgery in patients with two to four brain metastases significantly improved the control of brain disease. The 1-year rate of local failure was 8% for radiosurgery and WBRT combined, compared to 100% for WBRT alone. The median survival with WBRT alone was 7.5 months, compared to 11 months for WBRT plus radiosurgery. While the median time to local failure and time to any brain failure were statistically significant (P = .0005 and .002, respectively), median survival was not statistically different, perhaps because of the relatively small number of patients (N = 27) in the study.
Stanford Treatment Algorithm
Based on our own experience and that of other institutions, a fairly well-established algorithm for treating patients with multiple brain metastases is followed at Stanford. Whole-brain irradiation is the preferred treatment for patients with progressive systemic disease, poor performance status, or more than four lesions. Patients with multiple lesions who undergo WBRT and have a subsequent reduction in the radiographic number of metastases may be candidates for a stereotactic radiosurgical boost to the residual tumors if there are four or fewer lesions.[3,22]
Patients with stable systemic disease and a good performance status but who have mass effect from one or more tumors are candidates for surgical resection of their symptomatic lesions, with radiosurgery or radiation therapy used for the remaining smaller tumors. Patients with stable systemic disease, a good performance status, and no mass effect from their brain metastases (four or fewer tumors) are optimal candidates for stereotactic radiosurgery. A 91% tumor control rate has been observed in this population.
At our institution, most patients undergoing radiosurgery for multiple brain metastases are treated with WBRT and radiosurgery, whereas patients with a single brain metastasis are often treated with radiosurgery alone. Whole-brain irradiation is reserved for tumor recurrence/progression and new brain tumors associated with a poor performance status or progressive systemic disease. Unlike conventional brain irradiation, radiosurgery can be repeated for metachronous lesions months to years after initial treatment.
Stereotactic radiosurgery is an excellent treatment for small to moderately sized AVMs, particularly those located in surgically inaccessible regions or in patients who are poor candidates for surgery.[23-26] Such obliteration typically takes place over 2 to 3 years. Arteriovenous malformations less than 4 cm in diameter treated with 20 to 25 Gy have a 3-year obliteration rate of 76% to 95%, with low morbidity (2.5% to 4.5% of patients develop permanent neurologic deficits; 2.5% to 4.5%, transient deficits).[23-30]
Although radiosurgical treatment of smaller AVMs is routine, larger AVMs often present a greater challenge. Arteriovenous malformations greater than 4 cm in diameter have only a 33% to 50% rate of obliteration at 3 years after radiosurgery but a 20% to 30% complication rate following treatment at doses of 15 to 20 Gy.[23,25,26] The rate of obliteration increases for larger AVMs with the use of higher doses (25 to 45 Gy), but the risk of radiation-induced complications also increases.
Overall treatment morbidity for large to giant AVMs can be reduced with multimodality therapyie, combinations of embolization, stereotactic radiosurgery, and microsurgery.[31,32] Our previous experience with large and complex AVMs has shown that such multimodality therapy can reduce patient morbidity and mortality.
Embolization obviously reduces the nidus volume requiring resection, but having had stereotactic radiosurgery several years prior to surgical resection also produces a benefit. In some patients, partial AVM thrombosis significantly reduced the volume of residual AVM.
Some authors have recommended staged stereotactic radiosurgery when treating large AVMs. In these cases, multiple radiosurgery treatments are delivered to different portions of the AVM at various time intervals to avoid delivering treatment to a single large target, thereby theoretically reducing the risk of radiation necrosis. This option has been applied in selected patients.[31,34] The disadvantages of such an approach are the second latency period of 1 to 3 years before obliteration occurs, the possibility that a second radiosurgery treatment may still not obliterate the AVM, and the risk of radiation-induced injury, which may be higher with a second radiosurgery treatment.
Nasopharyngeal carcinoma arises in the mucosa or submucosa of the nasopharynx, and frequently spreads to the skull base. Given the radiosensitivity of the tumor, radiotherapy is the primary treatment. However, there is a significant incidence of local failure (26% to 100%) in more advanced cases after treatment with conventional radiotherapy.[35-37] Although higher radiation doses or brachytherapy boosts increase local control,[35,37-40] the possibility of normal tissue injury and/or the inability to effectively treat tumor extension to the skull base limit the usefulness of these techniques. Because of the historically high local failure rate, we developed a protocol to deliver a planned stereotactic radiosurgical boost following conventional radiotherapy in patients with nasopharyngeal carcinoma.
From October 1992 to December 1998, 23 patients at Stanford University Medical Center underwent radiotherapy followed by planned radiosurgery as initial management of a newly diagnosed, biopsy-proven nasopharyngeal carcinoma. The total dose of radiotherapy to the nasopharynx was 66 Gy, divided into daily treatments of 200 cGy. Elective neck irradiation to a dose of 50 Gy was also used; involved lymph nodes received radiation boosts to a total dose of 66 Gy. In addition to the radiotherapy, 15 patients received cisplatin(Drug information on cisplatin)um-based chemotherapy.
Within 4 weeks of completing radiotherapy, patients were treated with radiosurgery. Treatment was administered with one to four isocenters, depending on tumor volume and shape. The prescribed dose of radiation was delivered to the periphery of the original lesion, corresponding to the 80% to 85% isodose contour. The median dose was 12 Gy (range: 7 to 15 Gy). Mean follow-up was 27 months (range: 8 to 72 months), with 12 patients followed for more than 2 years, and 7 patients followed for more than 3 years.
Stanford Study Results
Throughout the course of follow-up, there were no local recurrences among the 23 patients treated with radiosurgery (Figure 1). Cervical lymph node recurrences developed in two patients, and seven patients developed distant metastases including liver (1), lung (2), and bone (4). Of these seven patients, four expired as a result of their metastases.
Simultaneous cisplatin (Platinol)-based chemotherapy and radiotherapy followed by radiosurgery was administered to 15 patients. No significant difference in relapse-free survival was observed, but a trend favored the chemotherapy group: Actuarial relapse-free survival at 3 years was 59% among patients receiving chemotherapy and radiotherapy followed by radiosurgery, compared to 37.5% in the patients who were treated with radiotherapy or radiosurgery alone. No patient developed either acute or late complications following radiosurgery.
Our preliminary experience showed that a high rate of local control can be achieved with stereotactic radiosurgery even in patients with advanced-staged nasopharyngeal carcinoma. We believe that radiosurgery is more effective than brachytherapy in such cases because of the ability of radiosurgery to effectively treat the base of skull. Although intracavitary brachytherapy in the nasopharynx is likely to be effective in treating tumors confined to the mucosa or submucosa, it is unlikely to eradicate a larger mass that extends several centimeters beneath the mucosal surface. Radiosurgery not only provides the same benefits as brachytherapy (ie, relative sparing of normal tissues and the ability to safely boost regions of involvement to high doses), but it can also deliver a high radiation dose to sites that are remote from the nasopharynx.
Although longer follow-up is needed, radiosurgery following radiation has resulted in 100% local control of nasopharyngeal carcinoma in a small group of patients. Based on this experience, patients with advanced nasopharyngeal carcinoma treated at Stanford receive a 66-Gy dose of radiation with concurrent cisplatin chemotherapy, followed by 12-Gy radiosurgery to the primary site, after which three more cycles of cisplatin/fluorouracil chemotherapy are administered. Despite the benefits of this approach, it is obvious that more effective chemotherapy is needed to decrease the incidence of late systemic recurrence in this patient population.