Abstract
Background/Aim: Seizures represent a major problem for patients with brain metastases. This study evaluated the role of seizures in patients receiving single-fraction radiosurgery (SRS) or multi-fraction stereotactic radiotherapy (FSRT). Patients and Methods: This retrospective study included 195 patients receiving SRS (n=164) or FSRT (n=31) alone for one to three brain metastases. The prevalence of pre-SRS/FSRT seizures and correlations with pre-treatment factors were investigated. These factors plus pre-SRS/FSRT seizures were assessed in regard to survival. Results: Thirty-three patients had pre-SRS/FSRT seizures (prevalence=16.9%). Seizures were significantly correlated with age ≤61 years. Trends were observed for seizures being more frequent in those with NSCLC and those without extra-cranial metastatic spread. On multivariate analysis, significant associations with improved survival were found for Karnofsky performance score ≥80%, breast cancer, and an interval from diagnosis of malignant disease to SRS/FSRT ≥21 months. Conclusion: Younger age, NSCLC and absence of extra-cranial spread appeared to be risk factors for seizures prior to SRS/FSRT. Having seizures prior to SRS/FSRT showed no association with survival.
Cerebral metastases represent a serious palliative situation for patients with malignant diseases (1-6). The majority of these patients have already multiple (more than three) intracerebral lesions at the time of diagnosis of metastatic spread to the brain (1, 2, 6). Many of these patients have very poor prognoses and receive whole-brain radiotherapy alone. The expected survival of patients with limited brain metastasis, i.e. one to three lesions, is generally more favorable, and these patients are often candidates for local therapies, either alone or combined with whole-brain radiotherapy (1, 6). Local therapies include stereotactic radiosurgery (SRS), which is given with a single fraction; fractionated stereotactic radiotherapy (FSRT), which mostly consists of three to seven fractions; and resection of metastases (1, 6-8). Most of these patients are treated with SRS or FSRT. Depending on the risk of intracerebral recurrence, SRS and FSRT are administered with or without whole-brain radiotherapy. According to randomized trials, the addition of whole-brain radiotherapy increased the risk of post-treatment cognitive decline but improved long-term intracerebral control of the disease (9, 10). Intracerebral progression or recurrence can also be associated with cognitive impairment (11-13). Therefore, patients with a higher risk of developing new or progressive brain metastases likely benefit from the addition of whole-brain radiotherapy, whereas those patients with a lower risk appear to be better treated with SRS or FSRT alone (14). The present study focused on patients of the latter group. The quality of life of these patients can be significantly compromised by seizures that most often occur prior to treatment of brain metastases (15). The risk of seizures was reported to be associated with the number of intracerebral lesions (16, 17). Therefore, it is reasonable to investigate the role of seizures separately for patients with single, few and multiple lesions. This study was performed on patients treated with SRS or FSRT alone for a limited number of one to three metastases. It evaluated the prevalence of seizures prior to SRS/FSRT, risk factors for this condition, and potential associations of seizures with survival in a large cohort of patients.
Patients and Methods
This retrospective study included 195 patients treated with single-fraction SRS (n=164) or multi-fraction FSRT (n=31) for one to three brain metastases between 1999 and 2019. It was approved by the local Ethics Committee of the University of Lübeck (reference-number: 20-120A). The median dose of single-fraction SRS was 20 Gy (range=15-25 Gy). FSRT was administered with total doses ranging between 22.5 and 45 Gy (median=33 Gy); the median number of fractions was 3 (range: 5-16), and the median dose per fraction was 8 Gy (range=2.5-12 Gy). In the entire cohort, the prevalence of seizures prior to SRS or FSRT was determined. In addition, correlations between seizures and seven potential prognostic pre-treatment factors (Table I) were investigated. These factors included age at SRS or FSRT (≤61 versus ≥62 years, median=62 years), gender, Karnofsky performance score (≤70% versus ≥80%, median=80%), type of primary tumor [breast cancer versus non-small cell lung cancer (NSCLC), renal cell carcinoma (RCC), melanoma and other types (n<20)], number of brain metastases (one versus two or three), extra-cranial metastatic spread (absence versus presence) and interval from first diagnosis of malignant disease to SRS or FSRT (≤20 versus ≥21 months, median=20 months). Furthermore, these factors plus symptoms and seizures prior to SRS or FSRT (Table I) were assessed for associations with survival.
For the analyses of correlations between seizures prior to SRS or FSRT and the potential prognostic pre-treatment factors, the chi-square test was applied. p-Values of less than 0.05 were considered to indicate a significant correlation, and p-values <0.12 were considered to show a trend. Survival times were calculated from the day of single-fraction SRS or the day of the first fraction of FSRT. For univariate analyses of survival, the Kaplan–Meier method and the log-rank test were applied. In the case of a significant correlation (p<0.05) or a trend (p<0.12), the factors were additionally analyzed in a multivariate Cox regression analysis. When a p-value was found to be less than 0.05 in the Cox regression analysis, the prognostic factor was considered to be independent.
Results
For 33 out of the 195 patients included in this study, seizures prior to SRS or FSRT were documented, corresponding to a prevalence of 16.9%. The occurrence of seizures was significantly correlated with age ≤61 years (prevalence of 24.7%, p=0.008). Trends were observed for seizures being more frequent in those with NSCLC (26.7%, p=0.113) and those without extra-cranial metastatic spread (23.2%, p=0.072). The results of this analysis are shown in Table II.
Moreover, a significant association with improved survival was found in those with a Karnofsky performance score of ≥80% (p<0.001) on univariate analysis (Table III), and trends were observed for improved survival in those with breast cancer (p=0.059), without extra-cranial metastatic spread (p=0.068), and an interval from first diagnosis of malignant disease to SRS or FSRT of ≥21 months (p=0.055). Symptoms (p=0.181) and seizures (p=0.431) prior to SRS or FSRT were not correlated with survival. In the corresponding multivariate Cox regression analysis, a Karnofsky performance score of ≥80% (p<0.001), a favorable type of primary tumor (i.e. breast cancer) (p=0.038) and an interval from first diagnosis of the malignant disease to SRS or FSRT of ≥21 months (p=0.005) proved to be independent predictors of survival (Table IV).
Distribution of the potential prognostic factors investigated in this study.
Discussion
Patients with very few brain metastases often receive SRS or FSRT, which are administered either alone or as a boost following whole-brain radiotherapy (1, 5, 6). Since two randomized trials emphasized that the addition of whole-brain radiotherapy is associated with a significant increase in cognitive deficits, physicians have become more hesitant regarding the combined approach and favor use of SRS/FSRT alone (9, 10). In the first trial, which was stopped after 58 patients, cognitive decline was observed in 24% and 96% of the patients at 4 months following SRS alone and following SRS supplemented by whole-brain radiotherapy, respectively (p<0.001) (9). In the second trial, which included 213 patients, cognitive decline at 3 months was observed in 64% of patients following SRS alone and in 92% of patients following SRS combined with whole-brain radiotherapy, respectively (p<0.001) (10). However, the rates of freedom from new or recurrent brain metastases at 1 year were significantly higher after the combined approach than after SRS alone. Intracerebral control rates were 27% after SRS alone versus 73% after SRS plus whole-brain radiotherapy in the first trial (p<0.001), and 15% versus 50% in the second (p<0.001) (9, 10). One has to be aware that a consequence of omitting whole-brain radiotherapy is worse control of intracerebral metastatic disease with associated symptoms, including seizures (11-13). About 20 years ago, the investigators of a trial of the Radiation Therapy Oncology Group (RTOG) reported that poor cognitive function was caused by intracerebral recurrence but by no other factor they had investigated (11). Similar findings were reported from prospective studies published a few years later (12, 13). Moreover, the risk of cognitive decline due to whole-brain radiotherapy can be significantly reduced with modern approaches including hippocampal-sparing radiotherapy and the drug memantine (18, 19). Memantine was shown to be safe and effective in the treatment of symptomatic Alzheimer's disease (20). In a randomized trial, memantine when added to whole-brain radiotherapy led to a significant protraction of the time to post-treatment cognitive decline (p=0.01) (18). In another prospective study including a historical control group, whole-brain radiotherapy with a hippocampal-sparing technique resulted in a significant reduction of cognitive deficits from 30% to 7% (p<0.001) (19). Both procedures were also successfully used in combination (21). It is important to consider all pros and cons of the addition of whole-brain radiotherapy to SRS or FSRT for each individual patient with one to three brain metastases. A scoring system designed to predict the probability of developing new brain metastases after SRS or FSRT alone may support clinical decision making (14). Three risk groups were identified with 6-month rates of freedom from new brain metastases of 36%, 65% and 80%, respectively. Patients of the third group and many patients of the second group appear to be well treated with SRS or FSRT alone and may not require whole-brain radiotherapy.
Correlations between potential prognostic pre-treatment factors and seizures prior to single-fraction (SRS) or fractionated (FSRT) stereotactic radiotherapy.
Summary of the univariate analyses of potential prognostic pre-treatment factors including symptoms and seizures with respect to survival.
Multivariate analysis (Cox regression analysis) of survival.
The present study focused on those patients treated with SRS or FSRT alone. It was performed to contribute to the understanding of the role of seizures prior to SRS or FSRT in this group. The prevalence of pre-SRS/FSRT seizures was 16.9%, which was within the range of 12-35% reported from previous studies not particularly focusing on SRS/FSRT (15, 17, 22, 23). Furthermore, the current study aimed to identify risk factors for occurrence of seizures prior to SRS or FSRT. A significant association with seizures was shown for younger age (≤61 years). A similar finding was reported from a previous retrospective study including 286 patients with brain metastases who underwent neurosurgical resection (24). In addition, trends were found in the present study for associations between pre-SRS/FSRT seizures and NSCLC or absence of extra-cranial metastatic spread. Associations between the occurrence of seizures in patients with brain metastases and the type of primary tumor were previously described. In the study of Oberndorfer et al. and the review article of Ruda et al., patients with lung cancer had the second highest prevalence of seizures (15, 25). Moreover, in the study of Wolpert et al., the highest prevalence of pre-operative seizures was also found in patients with lung cancer (17). In contrast, a positive correlation between absence of extra-cranial metastatic spread and pre-SRS/FSRT seizures was not reported previously.
In this study, an association between pre-SRS/FSRT seizures and survival was not observed. This agrees with the literature, since a correlation was only described for patients with glioma but not for those with brain metastases (26). In contrast to pre-SRS/FSRT seizures, the performance status, the type of primary tumor and the interval from first diagnosis of the malignant disease to SRS or FSRT were independently associated with survival. Extra-cranial metastatic spread was significant on univariate analysis. These factors were previously identified in patients receiving SRS or FSRT for very few brain metastases (27-29). This showed consistency of our data with previous studies regarding survival. However, the retrospective nature of the data analyzed in this study needs to be considered during their interpretation.
In conclusion, the prevalence of seizures prior to SRS or FSRT in the present study was within the range of the reported data. Younger age, primary NSCLC and absence of extra-cranial spread appeared to be risk factors for the occurrence of seizures prior to SRS/FSRT. In contrast to other significant prognostic factors, the occurrence of pre-SRS/FSRT seizures was not associated with survival.
Acknowledgements
As part of the project NorDigHealth project, the study achieved funding from the European Regional Development Fund through the Interreg Deutschland-Danmark program (087-1.1-18).
Footnotes
Authors' Contributions
All Authors participated in the design of the study. J.W. and D.R. provided the data. The data were analyzed by D.R. and S.E.S. and interpreted by all Authors. D.R. and S.E.S. drafted the article, which was reviewed and finally approved by all Authors.
This article is freely accessible online.
Conflicts of Interest
The Authors state that there are no conflicts of interest related to this study.
- Received April 25, 2020.
- Revision received May 13, 2020.
- Accepted May 15, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.