Gamma Knife Radiosurgery For Brain Metastases From Gastrointestinal Cancer





Keywords: brain metastasis, radiosurgery, gamma knife, outcome, brain tumor

Interactive Manuscript

Ask Questions of this Manuscript:

   



Abstract

     Brain metastases (METs) from gastrointestinal cancers are uncommon.
     We evaluated the efficacy of Gamma Knife radiosurgery (GKRS) for brain METs from gastrointestinal cancers.
     The total tumor number was 1153 (median; two per patient). The median KPS was 80% (range 40-100%). The most common primary tumor site was the colon (85 patients) followed by the rectum (79), gastrium (58), esophagus (37), liver (15), pancreas (four), duodenum (three) and bile duct, cholecyst and small intestine (two each). The median volume of the largest tumors was 6.9 (range; 0.02-85.6) cc, and the median peripheral dose was 20 (range; 10-25) Gy. The median period between original cancer diagnosis and brain MET occurrence was 23 (range; 0- 237) months.
     We retrospectively analyzed 287 patients (87 women, 200 men, mean age; 64 range; 19-85 years) who underwent GKRS for gastrointestinal cancer brain METs between July 1998 and September 2009.
     The median survival time (MST) was 4.7 months overall. MSTs according to Recursive Partitioning Analysis (RPA) were 16.9 months in Class 1 (7 patients), 5.2 in Class 2 (243) and 2.0 in Class 3 (37). Single metastasis, lobar tumor locations, KPS better than 80%, non-symptomatic, well-controlled primary tumors, no non-brain metastatic lesions and having prior surgery were significant predictive factors favoring survival. Follow-up imaging studies were available in 182 patients (63%, 635 tumors). The tumor control rates were 93% and 87% at the 6th and 12th post-GKRS months. Neurological survival, qualitative survival, and new lesion-free survival were 93%, 81% and 55% at the 6th post-GKRS month and 89%, 69% and 42% at the 12th post-GKRS month, respectively.
     This was a retrospective study.
       Our present results suggest GKRS to be effective for gastrointestinal cancer brain MET patients of RPA classes 1 and 2.
     Proper patient selection is important to optimize outcomes.


Acknowledgements

Project Roles: