A New Score Predicting Survival Prognosis After Whole-Brain Radiotherapy Alone for Brain Metastases in Elderly Patients

Discussion

Most patients with brain metastases receive WBRT-alone. The most commonly used WBRT regimen worldwide is 10×3 Gy over two weeks. Another option would be short-course WBRT with 5×4 Gy in two weeks for patients with a poor estimated survival. According to a retrospective study, 5×4 Gy results in similar survival and intra-cerebral control rates without increased acute toxicity when compared to 10×3 Gy in two weeks (7). On the other hand, patients with a very favorable survival prognosis may benefit from doses beyond 30 Gy such as 20×2 Gy in four weeks in terms of improved survival and intra-cerebral control, which has been suggested by another retrospective study (8). Furthermore, doses per fraction of less than 3 Gy have been reported to result in lower rates of neurocognitive deficits than of 3 Gy and higher (9-10). Late sequelae such as neurocognitive deficits become more important in long-term surviving patients. Patients with a favorable survival prognosis and a very limited number of brain metastases should also be considered for local treatments such as resection, radiosurgery and fractionated stereotactic radiation therapy (11-13). All these considerations demonstrate that it is very important to be able to estimate the individual patient's survival prognosis in order to choose for the best treatment option.

For the clinical routine, physicians need an instrument that allows them to quickly and easily estimate the patient's survival prognosis. Several survival scores already exist (3-5). However, those scores were developed for patients of any age and may, therefore, not be optimal for elderly patients. In the present study, a new survival score was developed specifically for patients of 65 years or older. The current score included three independent predictors of survival, namely gender, KPS, and the number of extracranial organs involved by metastatic disease. The latter prognostic factor has only recently been identified and is for the first time included in a survival score in the present study (1-2). This aspect also makes the current score somewhat unique. Gender has been included in only one previous score, which demonstrates the need for a separate score for elderly patients (14).

Based on scores for each of the three independent prognostic factors, four prognostic groups were created that significantly differed with respect to their 6-month survival rates. Low-scoring, group A patients had a very poor survival prognosis. Only 2% and 4% of the group A patients of the test group and the validation group, respectively, survived for at least six months. These patients obviously did not really benefit from WBRT and may be considered candidates for best supportive care and treatment with corticosteroids instead of irradiation. Patients of group B had survival rates of 17% and 21%, respectively. They should be considered for short-course WBRT such as 5×4 Gy in one week. The patients of group C had an intermediate survival prognosis, with 6-month survival rates of 56% and 50%, respectively. For these patients, 10×3 Gy in two weeks, the most common WBRT regimen, may be appropriate. The patients of group D, who had the most favorable survival prognosis with 6-month survival rates of 90% and 86%, respectively, should be considered for long-course WBRT with doses per fraction of <3 Gy such as 20×2 Gy in four weeks. For group C and D patients, with a very few brain metastases, radiosurgery or neurosurgery, either alone or in addition to WBRT, should be considered (11-13).

The 6-month survival rates of the four prognostic groups of the validation group were quite similar to the 6-month survival rates of the corresponding groups of the test group, which shows that this score is reproducible. One has to be aware that the data this score are based on were retrospective in nature. However, a prospective validation of this score cannot be expected in the near future.

In conclusion, since the 6-month survival rates of the validation and the test groups were quite similar, this score can be considered reproducible. This new score allows for estimation of the survival of elderly patients with brain metastases and, therefore, can assist treating physicians in choosing the best treatment for each of these patients.