Gamma Knife Surgery of Colorectal Brain Metastases: A High Prescription Dose of 25 Gy May Improve Growth Control

Objective

There are few reports on the effect of gamma knife surgery (GKS) for brain metastases from colorectal cancer. The purpose of this study was to identify prognostic factors for local control, complications, and survival in our series of patients treated with GKS.

Methods

Eighty patients (36 males, 44 females) with 140 metastases who received GKS between 1996 and 2008 were retrospectively reviewed. The mean tumor volume was 6.13 (0.01–35.5) cm³; the prescription dose was 21.1 (10–25.1) Gy and the maximum dose 42.7 (17.2–66.7) Gy; and the tumor cover was 95.0% (72%–100%).

Results

Growth control was achieved in 93 of 121 tumors (76.9%) and 42 of 68 (61.8%) patients, while treatment failure was seen in 28 of 121 tumors (23.1%). Local control was better if a high prescription dose of 25 Gy was used, 88.4% vs. 71.4% (P = 0.017), or if tumor volume was <5 cm³ (86.4%), compared with 69.9% for 5–20 cm³ and 51.9% for >20 cm³ (P = 0.002). The hazard ratio for local failure with lower prescription doses was 2.8 (P = 0.026) in the unadjusted, and 8.5 (P = 0.055) in the adjusted multivariate analysis (tumor volumes >5 cm³). The median survival was 6 months (range 0–75) after GKS. Age <70 years (P < 0.001) and high RPA class (P = 0.032) were associated with longer survival. Fifteen patients (22.1%) had persistent edema on follow-up MRI, possibly because of radiation damage to the tumor. Radiation-induced edema was asymptomatic in 93.8%. We found neither a decrease in the incidence of new metastases nor improved survival when whole-brain radiation therapy was given prior to GKS.

Conclusions

GKS provides reasonable local tumor control. Local control rate is highest if the margin dose is 25 Gy and the tumor volume <5 cm³. Radiation edema was common but rarely symptomatic. Survival is longest for young, well-functioning patients.

Introduction

Even though colorectal cancer is the third commonest cancer type in women and fourth in men worldwide (4, 20), brain metastasis is rare, with an incidence of 2.3% (10, 17), and represents only 1%–9% of all brain metastases. Brain metastases are associated with a poor prognosis. Stereotactic radiosurgery used as the only modality without whole brain radiation has become an established treatment for brain metastases. This is because of its minimal invasiveness and high intratumoral radiation dose combined with a sharp dose fall outside the lesion. Cerebral metastases are attractive candidates for gamma knife surgery (GKS) as the lesions are usually small, approximately spherical and well defined from surrounding normal brain tissue. Generally, GKS is associated with a high rate of local tumor growth control, and a low rate of treatment-related complications (26). However, the histopathologic type of metastatic brain tumor may influence the effect of GKS (11). Surgical results after resection of colorectal brain metastases show a survival rate of 7.5–8.3 months with a postoperative mortality of 3%–4% (6, 29), whereas a median survival of 3.2 months is seen after whole-brain radiation therapy (WBRT) (2). There are few reports in the literature on the outcome after stereotactic radiosurgery of brain metastases from gastrointestinal cancer (4, 10) and only two previous reports exclusively deal with colorectal cancer (15, 21). Schoeggl et al. and Matsunaga et al. report very high local tumor control rates (96% and 91.2%, respectively) (4, 15, 21). In our experience, colorectal brain metastases are difficult to control, with a tendency to persistent peritumoral edema on follow-up imaging.

We have therefore reviewed our series of patients with brain metastases from colorectal cancer to identify the local control rate, complication rate, and survival after GKS and looked for treatment- and patient-related factors that could influence the efficiency of GKS with special emphasis on the prescription dose.