Each year more than 100,000 cancer patients in the United States develop brain metastases . Of these, the majority will have either multiple lesions or uncontrolled primary or metastatic disease [2-4]. The treatment and care of patients with brain metastases remains a serious health-care problem. The article by Berk is a comprehensive review of completed or ongoing clinical trials worldwide evaluating the role of radiotherapy in the treatment of patients with brain metastases.
The major radiotherapy treatment themes that have been or are currently being evaluated are:
- The effect of various fractionation schemes on survival,
- The impact of dose escalation using hyperfractionated regimens,
- The effect of biochemical modification of radiotherapy,
- The use of radiosurgery as a replacement for surgical resection or as a boost to whole-brain radiotherapy, and
- The results of surgery alone vs surgery and whole-brain radiotherapy for patients with single metastatic lesions.
When designing trials for patients with brain metastases, clear end points should be established. Unfortunately, most studies have used survival as the sole end point. In the majority of patients, survival is affected predominantly by the extent, location, and response to therapy of non-central nervous system (CNS) sites of disease rather than by the success or failure of radiotherapy or surgery to control brain metastases. Whereas survival remains the most important end point, local control, neurologic quality of life, and costs are other end points of significance. For the uncommon long-term survivors, the neurocognitive and neuroendocrine effects of various radiotherapy treatment schemes might also be considered in study design.
Dr. Berk has extensively reviewed the current status of radiotherapy treatment regimens and results for patients with brain metastases. We will review and comment upon two specific issues related to this topic: (1) the appropriate treatment for patients with a single brain metastasis and (2) the results of biochemical modification of radiotherapy.
Most Appropriate Treatment for a Single Brain Metastasis
The most appropriate treatment for patients with a single brain metastasis has been a source of controversy for nearly two decades. Surgery has been used for patients with a single brain metastasis for many years. The advantage of surgical resection in immediately relieving neurologic signs and symptoms secondary to tumor mass effect is undeniable.
Patchell et al  were the first to address this issue in a well-designed, prospective, randomized study. The results demonstrated that patients with or without active systemic cancer and a single brain metastasis treated with surgical resection combined with whole-brain radiotherapy (3,600 cGy in 12 fractions) lived longer, had fewer local recurrences in the CNS, and had a better quality of life compared to patients treated with the same dose of whole-brain radiotherapy alone.
Most investigators would have anticipated such an advantage with the addition of complete surgical resection to whole-brain radiotherapy. That there was also an advantage with regard to survival and quality of life in the surgical arm demonstrated that some patients suffered neurologic effects and died from progression of their single brain metastasis. Some have interpreted the data from Patchell et al as an indication to recommend surgery for all patients, while others have argued that, by using higher doses of radiotherapy, results similar to those achieved with surgery and radiotherapy could be obtained.
Noordijk et al  have recently reported ther results of their similarly designed study. One of the important issues from this study deals with the selection of appropriate treatment for these patients. The major end point of the study was overall survival, with an advantage being demonstrated for those patients undergoing surgical resection and whole-brain radiotherapy as compared with whole-brain radiotherapy alone (median survival, 10 vs 6 months; P = 0.04). In a subset analysis (accepting the potential pitfalls of this statistical method), older patients (more than 60 years) and patients with active systemic disease outside the CNS did not appear to benefit from the addition of surgery.
It is not surprising that older patients had decreased survival rates compared with their younger counterparts, as it is well known that older age is one of the most important adverse prognostic factors for patients with a wide variety of intracranial histologies, both primary and metastatic. In patients with active systemic disease outside the CNS, the competing risk of dying from non-CNS causes overwhelms any potential advantage that could be achieved with combined-modality therapy for brain metastases.
The results of these two prospective randomized studies indicate a survival advantage for patients with a single brain metastasis treated with surgery and radiotherapy compared to radiotherapy alone. It is clear that more aggressive treatment than standard whole-brain radiation therapy is warranted in younger patients who present with a single brain lesion without evidence of other systemic disease. Some investigators believe that radiosurgery can serve as a surgical alternative [6,7].
The results of a large radiosurgery experience for brain metastases has recently been published . The actuarial 1- and 2-year rates of CNS local control (defined as lack of radiographic progression) were 85% and 65%, respectively. The ability of radiosurgery to control these lesions without significant complications suggests that this therapy may be a reasonable alternative to surgery. In a multivariate analysis, the two factors associated with worse survival were active systemic disease (relative risk, 4.428; P = 0.0001) and age more than 60 years (relative risk, 1.618; P = 0.002). The experience with radiosurgery is remarkably similar to that described in the studies of Noordijk et al  and Patchell et al .
A randomized study comparing radiosurgery to surgery in the treatment of single brain metastases in terms of survival, local control, and quality of life was begun at our institution 5 years ago. However, accrual to this study was poor due to physician and/or patient preference for either surgery or radiosurgery in individual circumstances, and the study was closed early in 1995. A proposed Radiation Therapy Oncology Group (RTOG) study will randomize patients with one to three brain metastases to whole-brain radiotherapy alone vs whole-brain radiotherapy and a radiosurgery boost. Since this study will be open to patients with multiple lesions, there is optimism that it will actively accrue patients.
Radiotherapy After Resection of a Single Brain Metastasis
There is growing interest in determining the need for postoperative radiotherapy in patients with a completely resected single brain metastasis. The rationale for recommending whole-brain radiotherapy in these patients is twofold: (1) to reduce the incidence of local recurrence within the surgical bed and (2) to decrease the incidence of new sites of CNS metastases, thus potentially improving survival and quality of life. However, because of the potential neurocognitive effects of postoperative whole-brain radiotherapy, investigators have questioned its routine use.
We are aware of two ongoing clinical trials in which patients are being randomized to surgery alone vs surgery followed by whole-brain radiotherapy. It is possible that patients with "radioresistant" histologies, such as melanoma and renal-cell carcinoma, may not benefit from prophylactic whole-brain radiotherapy after a complete resection of a single lesion. Until such studies are completed and the results published, we believe that it is reasonable to withhold postoperative whole-brain radiotherapy in young, compliant patients with no evidence of systemic disease and a long interval from diagnosis to the development of a single brain metastasis.
In summary, there is compelling evidence to suggest that aggressive local therapy (surgery and radiotherapy) for patients with a single brain metastasis produces superior survival and quality of life compared to treatment with radiotherapy alone. However, the current data suggest that aggressive therapy should be restricted to the minority of patients for whom brain metastases represent the life-threatening site of their disease. For asymptomatic or mildly symptomatic patients with a small lesion, radiosurgery appears to be an excellent alternative to surgery. While radiosurgery is a noninvasive procedure, the same selection criteria should be considered as for patients undergoing surgical resection. Surgery or radiosurgery cannot be justified as an adjunct to whole-brain radiotherapy in patients with progressive systemic disease. The value of whole-brain radiotherapy for all patients with a resected single lesion remains to be determined in a randomized prospective trial.
Biochemical Modification of Radiotherapy
The results of whole-brain radiotherapy alone for patients with multiple lesions are disappointing. There is continuing interest in investigating methods of chemical modification of radiation effect (radiation sensitizers). The use of misonidazole and radiotherapy for these patients did not improve survival.
When early results indicated that brain metastases showed high bromodeoxyuridine (BrdUrd) labeling , RTOG investigators began a trial of whole-brain radiotherapy alone (37.5 Gy in 15 fractions) vs whole-brain radiotherapy and BrdUrd (RTOG 89-05) . The halogenated pyrimidine, BrdUrd, has been of interest for a number of years in the treatment of CNS tumors because of its uptake into malignant lesions and low uptake in normal brain cells. In RTOG 89-05, BrdUdr was administered in a dose of 0.8 g/m²/d for 4 days of each of the 3 weeks of radiotherapy.
Between October 1989 and March 1993, 72 patients were accrued into this study. While the toxicity of BrdUrd was tolerable, the median survival in the radiotherapy-alone group was 6.12 months vs 4.3 months in the radiotherapy-BrdUrd group (P = 0.9040). Although there was a higher complete response rate in the patients receiving BrdUrd (27%, vs 12% in the radiotherapy-only group), this did not reach statistical significance. Despite the fact that the total number of patients entered into this study was insufficient to show any benefit of BrdUdr administration, the search for ways to enhance radiation effect continues enthusiastically, particularly for patients with multiple lesions, for whom current therapies are less than optimal.