A Cochrane review by Hart et al 201112 assessed if resection of single brain metastasis followed by WBRT holds any clinical advantage over WBRT alone. Three RCTs were identified that included a total of 195 patients. Of note, the three RCTs by Mintz et al 1996,42 Patchell et al 199043 and Vetch et al 1993,44 included in the Cochrane review were published before 2001. All studies included populations with mixed primary tumours, including one study (Patchell 1990) with less than ten breast cancer patients. No results were reported by Hart et al for breast cancer patients separately.
Two retrospective studies on surgery among breast cancer patients were identified.
Patient selection
The surgical trials identified in the Hart et al Cochrane review were limited to patients with good performance status, with a single or limited number (1-3) of accessible lesions, inactive or well-controlled primary disease and limited co-morbidities, and patients with raised intracranial pressure or other uncontrolled symptoms. Generally, patients were considered unsuitable for surgery when there were multiple lesions, when the lesion was surgically inaccessible, or patients with active primary disease or significant comorbidities.12
Overall survival
The Hart et al Cochrane review found no significant difference in survival (HR 0.72, 95% CI 0.34 to 1.55, random effects, p = 0.40) although there was heterogeneity between trials
(I2 = 83%).12 There was some indication that surgery and WBRT might reduce the risk of deaths due to neurological cause. The Hart et al Cochrane review reported that those treated with surgery and WBRT were less likely to die from neurological causes although this did not reach statistical significance (RR 0.68, 95% CI 0.43 to 1.09, p=0.11; three trials).12 Mortality at 30 days was similar in both arms of each trial.
Although no statistically significant difference between surgery plus WBRT and WBRT alone was observed, the Patchell 199043 and Vecht 199344 trials were in favour of surgery while the Mintz 199642 trial was in favour of WBRT alone. The Patchell 1990 trial included a majority of patients with non-small cell lung cancer, which is highly radio-resistant and would not be expected to respond well to WBRT. There may have been selection bias in this trial also, as patients were selected for surgery by a single neurosurgeon. The Mintz 1996 trial included patients with a poorer KPS, and a larger proportion of patients had extra-cranial metastases.42
Functionally independent survival
One trial in the Hart et al Cochrane review (Patchell 1990) reported results on functional independent survival.43 The trial found that patients treated by surgery and WBRT maintained their functional independence for longer than those treated by WBRT alone (HR 0.42, 95% CI 0.22 to 0.82, p=0.01).43
Adverse events
The results of each trial identified in the Hart et al Cochrane review found that neither surgery in combination with WBRT or WBRT alone was more likely to cause adverse effects (RR 1.27, 95% CI 0.77 to 2.09, p=0.35).12 It is noted that the reporting of the trials did not allow for clustering of adverse effects within patients. Commonly reported adverse events for patients in the surgical arms of the trials included respiratory problems, haematoma and infections.