Local recurrence and survival following stereotactic radiosurgery for brain metastases from small cell lung cancer.

PURPOSE

Stereotactic radiosurgery (SRS) represents a treatment option for patients with brain metastases from small cell lung cancer (SCLC) following prior cranial radiation. Inferior local control has been described. We reviewed our failure patterns following SRS treatment to evaluate this concern.

METHODS AND MATERIALS

Individuals with SCLC who received SRS for brain metastases from 2004 to 2011 were identified. Central nervous system (CNS) disease was detected and followed by gadolinium-enhanced, high-resolution magnetic resonance (MR) imaging. SRS dose was prescribed to the tumor periphery. Local recurrence was defined by increasing lesion size or enhancement, MR-spectroscopy, and perfusion changes consistent with recurrent disease or pathologic confirmation. Any new enhancing lesion not identified on the SRS planning scan was considered a regional failure. Overall survival (OS) and CNS control were evaluated using the Kaplan-Meier method. Factors predicted to influence outcome were tested by univariate log-rank analysis and Cox regression.

RESULTS

Fifteen males and 25 females (median age of 61 years [range, 36-79]) of which 39 received prior brain irradiation were identified. In all, 132 lesions (3.3 per patient) between 0.4 and 4.7 cm received a median dose of 16 Gy (12-22 Gy). Thirteen metastases (10%) ultimately recurred locally with 6- and 12-month control rates of 81% and 69%, respectively. Only 1 of 110 metastases <2 cm recurred. Local failure was more likely for size >2 cm (P < .001) and dose <16 Gy (P < .001). The median OS was 6.5 months, and the time to regional CNS recurrence was 5.2 months. For patients with single brain metastases, both OS (P = .037) and regional CNS recurrence (P = .003) were improved. CNS control (P = .001), and survival (P = .057), were also longer for patients with controlled systemic disease.

CONCLUSIONS

Local control following SRS for SCLC metastases is achievable for lesions <2 cm. For metastases >2 cm, local failure is more common than expected. Patients with controlled systemic disease and limited CNS involvement would benefit most from aggressive treatment.