Stereotatic radiosurgery of 468 brain metastases ≤2 cm: implications for SRS dose and whole brain radiation therapy

Shehata. IJROBP 2004;59:87

For many decades, the standard treatment for brain metastases has been whole brain radiotherapy (WBRT) with various fractionated regimens as determined by the Radiation Therapy Oncology Group (RTOG). In recent years, stereotactic radiosurgery (SRS) and surgery have emerged as effective primary treatments for patients with brain metastases, with improved survival compared with WBRT alone. In select patients with single brain metastasis, the addition of postoperative WBRT has been proved to prolong neurologic survival and prevent recurrence compared with surgery alone. The effect of SRS alone compared with SRS combined with WBRT has not been clearly defined; despite an improvement in local tumor control with combined SRS and WBRT, some retrospective reports failed to demonstrate a survival advantage.

Dose selection for the treatment of brain metastases with SRS has historically been based on the inverse relationship between the SRS dose and the treatment volume, correlating with the incidence of brain necrosis. According to this principle, smaller-sized brain metastases can better tolerate treatment with increasing doses of SRS than can larger metastases. Currently, the RTOG protocol 90-05 has been the only prospective study attempting to identify the maximal tolerable dose of SRS for brain metastases ≤2 cm. We sought to identify the optimal SRS dose and the associated impact of WBRT for brain metastases ≤2 cm as determined by local tumor control rates and complication rates.

The national standard stereostatic radiosurgery (SRS) dose for brain metastases ≤2 cm is 24 Gy as established by the Radiation Therapy Oncology Group study 90–05, in which planned whole brain radiotherapy (WBRT) was not used. On the basis of our institutional experience, the goal of this study was to determine the optimal SRS dose and influence of WBRT on local tumor control among 468 ≤2-cm metastases.  Between October 1992 and May 2001, 468 newly diagnosed or recurrent ≤2-cm brain metastases, among 160 patients, were treated with SRS (dose range, 7–30 Gy; median, 20). A total of 240 metastases received planned WBRT (range, 6.75–50.4 Gy; median, 40.5) vs. 228 metastases that did not. The variables tested by multivariate analysis for their potential effect on tumor control included histologic type, site of metastasis, primary diagnosis, tumor volume, SRS dose, newly diagnosed vs. recurrent metastasis, and planned WBRT vs. no planned WBRT.

Results

Follow-up ranged from 1 to 82 months (median 7). On multivariate analysis, the addition of WBRT was the most significant predictor of local tumor control. Overall, patients who received WBRT had superior local tumor control rates (97% vs. 87% in those who did not receive WBRT; p = 0.0001). Patients receiving WBRT and SRS ≥20 Gy achieved local control rates of 99% compared with 91% control rates when treated with WBRT and SRS <20 Gy (p = 0.0029). Increasing the SRS dose to >20 Gy resulted in no improvement in local tumor control and a higher rate of Grade 3 and 4 neurotoxicity, approaching statistical significance (5.9% vs. 1.9%, p = 0.078).

Discussion  

For many malignancies, the standard of practice is to aggressively treat local tumor recurrences with RT to reduce the morbidity associated with the increased tumor burden. The brain should be no exception, given its extreme sensitivity to mass effect resulting from local tumor recurrence. Since the advent of SRS, many have advocated that better-prognosis brain metastases patients with an anticipated survival of ≥1 year should receive SRS alone without the addition of planned WBRT. This practice was based on the anticipation that adding WBRT would put these long-term survivors at risk of demonstrating the late effects of neurocognitive decline with no benefit to survival. RTOG study 95-08 demonstrated that SRS plus WBRT significantly improved survival for brain metastases compared with WBRT alone among the subset of patients with single brain metastasis, recursive partitioning analysis Class I, age <50 years, and non–small-cell lung cancer. Some studies have demonstrated that the addition of WBRT to SRS compared with SRS alone does not provide any significant benefit to quantitative survival. However, patients with recurrent brain metastases present most commonly with a symptomatic neurologic deficit. The consequences of tumor recurrence may pose a greater threat to a patient's quality of survival and clinical performance than the presumptive neurocognitive late effects of WBRT. The management of brain metastases should emphasize the improvement and preservation of neurologic function, with the maximal therapy provided at the initial presentation of intracranial disease to prevent brain recurrence. WBRT before SRS may provide a theoretical advantage by decreasing the tumor volume to be treated with SRS and subsequently reduce complication rates. In addition, maximal initial therapy consisting of SRS and upfront WBRT was recently demonstrated by Sneed  to be associated with a lower rate of requirement for salvage brain therapy (7%) compared with SRS alone (37%). The increased need for salvage brain treatment and its potential impact on quality of survival needs further evaluation.

Current data have shown a statistically significant improvement in local control of brain metastases by the addition of WBRT to SRS. The results of this study further emphasized a statistically significant improvement in local control achieved by the addition of WBRT to SRS. On multivariate analysis, we found the addition of WBRT to be the most significant predictor of local tumor control. Overall, patients who received WBRT had superior local tumor control rates (97% vs. 87% for those who did not receive WBRT. At 1 year, patients who received SRS plus WBRT had a local control rate of 96% compared with 77% for SRS alone. The implications of improved local control can be demonstrated in two recent studies that revealed a statistically significant association between intracranial tumor progression and neurocongnitive decline as measured by Folstein Mini-Mental Status examination scores.

In addition to an improvement in local control, Sneed  found statistically significant improvement in the interval of brain freedom from progression by the addition of WBRT to SRS compared with SRS alone (15.9 months vs. 8.3 months, p = 0.008). Chidel  demonstrated statistically significant improvement in the intracranial disease-free survival rate at 2 years with SRS plus WBRT compared with SRS alone (60% vs. 34%, p = 0.027). Furthermore, the combination of upfront WBRT and SRS may be a more cost-effective method of managing brain metastases than SRS alone, given the significant cost associated with repeated SRS or additional craniotomy for new metastases. Cost effectiveness, as relates to this subject, is also in need of further evaluation.

Currently, limited data exist in the literature on identifying the optimal SRS dose for the treatment of brain metastases. Kjellberg  described the inverse relationship between the SRS dose and treatment volume with regard to SRS dose selection on the basis of the incidence of radionecrosis. The SRS dose has traditionally been influenced by location, physician preference, total dose exposure (with or without WBRT), and the RTOG study 90-05 findings. The RTOG study 90-05 was a dose-escalation study attempting to identify the maximal tolerable dose of a single fraction of SRS in patients with recurrent, previously irradiated, brain metastases or previously irradiated primary brain tumors. That study concluded that 24-Gy SRS was the maximal tolerable dose for tumors ≤2 cm on the basis of an acute toxicity rate of <30% as the primary end point (RTOG Grade 3 or greater toxicity).

In this series, we attempted to identify the optimal SRS dose with planned WBRT that provided the maximal level of tumor control and the lowest level of complications (RTOG Grade 3 or greater). According to these criteria, we found 20 Gy to be the optimal SRS dose for brain metastases ≤2 cm. SRS doses ≥20 Gy resulted in significantly improved local control compared with <20 Gy. Additional stratification of the SRS dose  demonstrated superior local control rates with SRS doses of 20 Gy and >20 Gy compared with <20 Gy. However, a higher level of complications was seen with SRS doses >20 Gy compared with ≤20 Gy. A recent prospective trial of 79 patients by Majhail  found SRS doses >20 Gy on multivariate analysis to be significantly associated with a higher incidence of early complications (p = 0.0083). On the basis of the results presented by Majhail, as well as data from this series, we found 20 Gy to be the optimal SRS dose.

Conclusion

First, optimal control of brain metastasis ≤2 cm was seen with 20-Gy SRS combined with planned WBRT. Second, SRS doses >20 Gy resulted in no obvious improvement in local control and appeared to be associated with a greater rate of complications.

1