Radiosurgery and Whole-Brain Radiation Therapy for Brain Metastases: Title and subTitle BreakEither or Both as the Optimal Treatment

Brain metastases are the most feared and debilitating complication of systemic cancers. They affect approximately 170 000 patients a year, with the most common primary tumors being lung cancer, breast cancer, and melanoma.¹ The traditional approach for patients with brain metastases has been whole-brain radiation therapy (WBRT), irrespective of lesion number or primary site.

A shift in treatment occurred for patients with a single brain metastasis after 3 randomized trials were completed.^{2 - 4} In these trials patients were randomized to receive WBRT alone or surgical resection of a single brain metastasis followed by WBRT. In studies by Patchell et al² and Noordijk et al,³ resection of a single brain metastasis prior to WBRT led to an increase in median survival of 25 and 17 weeks, respectively. There was also a significant increase in functionality for the surgical group in both studies. A study by Mintz et al⁴ did not show a survival advantage for the surgery plus WBRT group, but patient selection (poorer performance status and more advanced systemic disease) most likely accounted for this result.

Patchell et al⁵ performed the only randomized trial comparing resection of a single brain metastasis to observation or WBRT. Patients treated with resection of their brain metastasis followed by WBRT had better local (10% vs 46%) and distant (18% vs 70%) brain control (ie, development of metastases in other parts of the brain), with fewer patients dying of neurologic causes; importantly, there was no survival advantage or increase in functional independence with WBRT. Only a small number of patients have a single brain metastasis and are good surgical candidates.

Stereotactic radiosurgery has an advantage over conventional surgery in that it is noninvasive and allows more than 1 lesion to be treated, including those in areas not surgically accessible. One reason to use stereotactic radiosurgery is to increase effectiveness against “radioresistant tumors,” as the dose used is significantly higher than that for WBRT. An additional potential advantage is elimination of the neurotoxicity associated with WBRT.⁶ The main limitation of stereotactic radiosurgery is tumor size; tumor number and location may also limit the use of stereotactic radiosurgery.⁷ Since 50% to 60% of patients have 1 to 4 brain metastases,⁸ they may be candidates for stereotactic radiosurgery but not surgery. Although stereotactic radiosurgery is “functionally” similar to surgery, no randomized trials have been published as to the comparative efficacy of these treatments. Auchter et al⁹ used stereotactic radiosurgery with WBRT to treat patients with a single brain metastasis, and outcomes were similar to those reported by Patchell et al² and Noordijk et al.³ Despite these similar outcomes, prospective trials are needed to define optimal treatment using stereotactic radiosurgery and WBRT.

In this issue of JAMA, Aoyama and colleagues¹⁰ report the first randomized trial of stereotactic radiosurgery with or without WBRT in patients with up to 4 brain metastases; one other trial has been presented only in abstract form,¹¹ and other studies compared WBRT with or without stereotactic radiosurgery.^{12 - 13} The impetus for performing such a trial is to prospectively confirm retrospective data suggesting that stereotactic radiosurgery alone does not decrease survival and patients can have salvage therapy (ie, effective treatment at relapse) with WBRT at progression^{14 - 16} or to prove that WBRT is a necessary adjunct to stereotactic radiosurgery. The rationale for adding WBRT is to treat “micrometastases” not seen on imaging, thereby improving distant and local brain control.^{5 ,11 ,14 - 16}

Aoyama et al¹⁰ provide several important and confirmatory findings. First, there is no survival advantage with the addition of WBRT to stereotactic radiosurgery, confirming what has been previously reported,^{5 ,11 ,14 - 16} so it appears that WBRT can be delayed without compromising patient survival. Pirzkall et al¹⁶ reported a survival advantage for the subset of patients who had no extracranial disease and were treated with stereotactic radiosurgery and WBRT. Second, local control is improved with the addition of WBRT and recurrence rates are lower. Improved control rates have been reported by others,^{5 ,11 ,14 - 16} but brain control did not appear to have an effect on outcome or functional status except in the study by Patchell et al.⁵ Third, patients treated with stereotactic radiosurgery and WBRT require salvage therapy less often, but those patients treated with stereotactic radiosurgery alone could be salvaged at the time of recurrence or progression. And, fourth, the mode of death was similar in both groups, with about 20% to 25% of patients dying of neurologic causes, different than what Patchell et al⁵ reported.

How does this finding compare with treating patients who have WBRT with or without stereotactic radiosurgery? Two published studies have shown that local control is better when stereotactic radiosurgery is added to WBRT; however, survival is unchanged,^{11 - 12} except for a subset of class 1 patients (defined by recursive partitioning analysis [RPA]¹² ), those with a single brain metastasis and those with non–small cell lung cancer.¹²

How should clinicians interpret the findings reported by Aoyama et al¹⁰ and the data available in the literature? Patients who have more than 4 brain metastases should continue to be treated with WBRT. For patients with 4 or fewer brain metastases, the combination of stereotactic radiosurgery and WBRT improves local brain control but does not affect survival. Therefore, either mode is a reasonable first choice; the exception is for patients with a single brain metastasis, non–small cell lung cancer, or RPA class 1 patients,¹² for whom stereotactic radiosurgery should be added to WBRT. Choice of treatment mode should take into account the patient's age, extracranial disease, and performance status. Whether the loss of local brain control with WBRT or stereotactic radiosurgery alone affects quality of life and functional outcome remains an unanswered question but requires further investigation. As both Aoyama et al¹⁰ and Sneed et al^{14 - 15} have shown, patients who are treated with initial stereotactic radiosurgery can have successful salvage therapy with WBRT. Aoyama et al¹⁰ used a stereotactic radiosurgery dose of 30% less when combined with WBRT, whereas Shehata et al¹⁷ found that for patients with brain metastases 2 cm or less, a stereotactic radiosurgery dose of 2 Gy was optimal; this remains another area in which refinements might alter patient outcomes. Additional studies also are needed to determine how to integrate surgery into the treatment paradigm of stereotactic radiosurgery and WBRT, although this approach would have bearing on only a small number of patients.

Aoyama et al¹⁰ have prospectively shown that withholding stereotactic radiosurgery does not affect survival for patients who have 4 or fewer brain metastases; these patients have a higher rate of local brain failure, but apparently withholding stereotactic radiosurgery does not influence how patients die of their disease. Whether overall quality of life is positively or negatively affected is unknown, but for patients who might be cured of their cancer, omitting WBRT could avoid long-term neurotoxic effects.