A New Predictive Tool for Optimization of the Treatment of Brain Metastases from Colorectal Cancer After Stereotactic Radiosurgery

Discussion

Many patients with metastatic colorectal cancer have a poor survival prognosis. This applies particularly to patients with brain metastases. Therefore, much experimental, translational and clinical research has been and is being still performed for metastatic colorectal cancer (14-17). New combinations of anticancer drugs have been introduced (2-4). However, most systemic agents are not able to pass the blood–brain barrier. Therefore, radiotherapy is still the most frequently used treatment modality for brain metastases. In order to optimize the radiotherapy of cerebral metastases from colorectal cancer, personalized treatment strategies tailoring the therapy to individual situations are important. Such strategies need to take into account the patient's survival prognosis. For example, long-term intracerebral control and radiation-related late toxicities are less important in patients with a very limited survival prognosis (18). In general, patients with up to three cerebral metastases live longer than patients with more than three lesions and, therefore, often receive stereotactic radiosurgery with whole-brain irradiation rather than whole-brain irradiation alone (10-12). However, a certain proportion of patients with 1-3 brain metastases have a poor survival prognosis, mostly due to the presence of visceral metastases (8, 9, 18). Since stereotactic radiosurgery may be associated with discomfort for patients, it may be questioned whether patients with a poor survival prognosis are really appropriate candidates for stereotactic radiosurgery. These patients may be better treated with short-course whole-brain irradiation alone or even best supportive care alone (18).

In order to choose the best treatment for an individual patient, a predictive tool for estimation of the remaining survival time would be of great benefit. In the current study, we created such a predictive tool for patients with no more than three brain metastases from colorectal cancer. In order to avoid selection bias caused by different types of treatment, only patients treated with stereotactic radiosurgery were included. According to the results of the study, the Karnofsky performance score and the interval from first diagnosis of colorectal cancer until radiosurgery were significantly positively associated with survival. In addition, the presence of extracranial metastases showed a strong trend for an association with poorer survival. These results mostly agree with the findings of previous studies including patients with brain metastases from colorectal cancer. In the study of Heisterkamp et al. in patients receiving whole-brain irradiation alone, survival was significantly associated with the Karnofsky performance score and extracranial metastases (8). In the studies of Meyners et al. (9) and Dziggel et al. (18) of patients receiving whole-brain irradiation for brain metastases from less radiosensitive tumors including colorectal cancer, the Karnofsky performance score and extracranial metastases had a significant impact on survival. The prognostic significance of the interval between the first diagnosis of colorectal cancer and radiotherapy has not yet been described for patients with brain metastases from colorectal cancer but for patients with cerebral metastases from other primary tumors (19, 20).

Based on the three factors Karnofsky performance score, interval from first diagnosis of colorectal cancer until radiosurgery, and presence of extracranial metastases, a predictive tool was designed to estimate the survival time of patients treated with stereotactic radiosurgery for brain metastases from colorectal cancer. Two prognostic groups were identified, one for patients with 0 or 1 point and another for those with 2 or 3 points. Only 20% of the patients with 0 or 1 point survived six months, and no patient of this group was alive at one year following stereotactic radiosurgery. Therefore, such patients may be candidates for a short course of whole-brain irradiation or for best supportive care (18). Out of the patients who achieved scores of 2 or 3 points, 78% survived at least six months and almost half of the patients were alive at one year following stereotactic radiosurgery. Since 78% of the patients with 2 or 3 points died from extracranial metastatic disease and not from an intracerebral recurrence, intensification of systemic treatment is important for this group, whereas whole-brain irradiation in addition to stereotactic radiosurgery appears dispensable.

In conclusion, the present study provides a new predictive tool that allows for estimation of the survival prognosis of patients with up to three brain metastases from colorectal cancer treated with stereotactic radiosurgery. Patients with 0 or 1 point have a very poor survival and appear better-treated with short-course whole-brain irradiation or best supportive care following stereotactic radiosurgery. Patients with 2 or 3 points have significantly more favorable prognoses and should receive stereotactic radiosurgery plus intensive systemic treatment.