Editorial: Radiosurgery for colorectal brain metastases

Abstract

In this issue of the Journal of Neurosurgery, Matsunaga and colleagues1 present the results of a retrospective review of their large experience of 152 patients with colorectal brain metastases managed over a 16-year period. The average time from diagnosis of the primary tumor to diagnosis of brain involvement was 27 months, and most patients had multiple metastases. The authors found a median survival of 6 months, and 91% of tumors were controlled over that time frame. In patients for whom they could determine the cause of death, a neurological demise occurred in 13 patients, and problems related to extracranial cancer were much more frequent, leading to death in 112 patients. The most common brain metastases are related to lung, breast, renal cancer, and metastatic melanoma. Tumors from the gastrointestinal tract, thyroid, prostate gland, ovary or uterus, or metastatic sarcomas are much less frequent. Nevertheless, in certain parts of the world, gastrointestinal tract cancers are more common, and this report from Japan provides insight into management based on a large clinical series. For example, in our own institutional series of over 3200 patients with brain metastases, only 93 had cancers of gastrointestinal origin. In a series of such patients with 246 tumors, we found a similar median survival of 7 months. With some cancer types, synchronous presentation of the primary cancer and brain metastasis is relatively common. This is frequently the case in non–small cell lung cancer, where a neurologic event is documented, a brain scan performed, and then a chest radiograph or CT scan shows evidence of a primary tumor in the lung. This is also relatively frequent in melanoma because of its propensity to involve the brain. On the other hand, brain metastases related to colorectal cancer or other primary gastrointestinal tumors usually occur in the late stages of the disease, often in association with other metastases to the lung or liver. For this reason, expectations are more guarded. In our own experience among different histological subtypes, we found that gastrointestinal tract tumors were associated with a shorter median survival than even melanoma, a tumor often considered to have the shortest survival expectations amongst the common cancers. Colorectal cancers are sometimes considered more radiation resistant than lung or breast cancer, despite the fact that most are adenocarcinomas, which in the lung constitute a more favorable subtype. The authors raise the question of whether brain imaging should be performed on a screening basis in patients with colorectal cancer. Typically, patients with colorectal cancer are not well screened compared to those with lung or breast cancer due to the relatively low rate of brain tumors. This could result in a delay in tumor detection. While they acknowledge that routine surveillance is controversial, they argue that those patients with adverse prognostic factors, particularly those with hepatic metastases (who have the highest incidence of brain metastases) should be screened before neurological symptoms develop. I agree with this concept. I also agree with their recommendation for close follow-up after radiosurgery so that any new tumors can be detected while small and asymptomatic. For all of us who manage brain metastases, the most common form of intracranial cancer, therapeutic advances have reduced some of the nihilism associated with this disease. If we can turn cancer into a more chronic disease managed with early detection, minimally invasive therapies, and serial imaging so that tumors are detected while small and asymptomatic, we will be able to prolong survival, maintain function and quality of life, and allow patients to participate in new research protocols that hopefully may improve the management of extracranial disease.