An Instrument for Estimating the 6-Month Survival Probability After Whole-brain Irradiation Alone for Cerebral Metastases from Gynecological Cancer

Discussion

Despite considerable research efforts in the field of metastatic gynecological cancer, many of these patients have a poor prognosis, especially in those with multiple brain metastases requiring WBI. Better outcomes may be achieved with the implementation of novel anticancer drugs (7). However, personalized cancer care can contribute to this ambitious goal. Personalized approaches consider many aspects of a patient's specific situation and preferences. One important aspect is the patient's survival prognosis. If it is quite poor, the treatment should not be long or burdensome. If it is favorable, therapy should aim to achieve the best possible long-term disease control with acceptable late toxicities.

These considerations also apply to patients who are assigned to WBI alone for cerebral metastases from gynecological cancer. In the case of a short remaining lifespan, a short course of WBI is preferable. In a large retrospective study performed on 416 patients with cerebral metastases from various tumor types and poor survival prognoses, standard WBI with 10×3 Gy in 2 weeks was as effective as higher dose, longer regimens (15×3 Gy in 3 weeks/20×2 Gy in 4 weeks) with respect to intracerebral control and survival (8). The 6-month intracerebral control rates were 39% after 10×3 Gy and 41% after higher doses, respectively (p=0.61), and the 6-month survival rates 33% and 29%, respectively (p=0.86). It was possible to reduce overall treatment time by 50% from 4 to 2 weeks without loss of efficacy. In another large retrospective study in a similar patient population (n=442), short-course WBI with 5×4 Gy in 1 week was not inferior to 10×3 Gy in 2 weeks (2). Six-month intracerebral control rates were 50% after 5×4 Gy and 37% after 10×3 Gy, respectively (p=0.07), and 6-month survival rates were 24% and 27%, respectively (p=0.29). Thus, the overall treatment time was further reduced to only 1 week. In contrast, patients with a more favorable survival prognosis can benefit from longer-course WBI in terms of improved intracerebral control and survival (3). Moreover, doses per fraction of less than 3 Gy were reported to be associated with less neurocognitive decline than doses of 3 Gy and more per fraction (4, 5).

Since a patient's survival time is an important factor to be considered when selecting the appropriate WBI regimen, it would be helpful if the treating radiation oncologists were able to estimate a specific patient's prognosis, ideally with a simple instrument. Several such instruments are already available for patients with cerebral metastases from specific tumor types such as breast cancer, lung cancer and others (9-13). In the present study, an additional prognostic instrument was developed particularly for patients treated with WBI alone for cerebral metastases from gynecological cancer. Since only patients treated with WBI alone were included, this score is more specific and carries less risk of hidden selection bias due to different treatment approaches than a previous instrument that also included patients receiving local modalities for cerebral metastases from gynecological cancer (14). In the present study, three prognostic groups were designated based on two variables, namely ECOG-PS and metastasis outside the brain: 2-7 points, 9 points and 14 points. The 6-month survival rate in the 2-7 points group was only 10%. Therefore, these patients appear to be good candidates for short-course WBI, for example with 5×4 Gy in 1 week. For the other two groups (9 points and 14 points), the 6-month survival rates were much more favorable, at 53% and 100%, respectively. Patients of both groups appear appropriate for treatment with longer-course WBI. Patients of the group with 9 points may receive 10×3 Gy in 2 weeks, and patients of that with 14 points may receive longer-course WBI with total doses >30 Gy and doses per fraction <3 Gy.

However, when considering these suggestions, the retrospective nature of the study and the risk of hidden selection biases must be considered. In order to further reduce the risk of WBI-related neurocognitive decline, WBI may be performed with hippocampal-sparing for selected patients with a limited number of cerebral lesions that are not close to the hippocampi (15).

In conclusion, a new instrument was designed to predict the 6-month survival probability of patients receiving WBI for cerebral metastases from gynecological cancer. Three predictive groups were created with significantly different survival prognoses, facilitating more personalized treatment in this specific patient group.