Signal Intensity in Dentate Nucleus Does Not Always Correlate with Accumulative Gadolinium doses after Whole-Brain Radiotherapy

Synopsis Whole-brain radiotherapy (WBRT) is often performed asynchronously with head contrast-enhanced MRI examination (CEME) in patients with brain metastasis (BM). Data regarding T1 signal intensity (SI) changes in the dentate nucleus (DN) in BM patients after WBRT and repeated exposure to gadolinium (Gd)-based contrast agents (GBCAs) are lacking. In this study, we explored the correlation between T1 SI in the DN and cumulative doses of linear GBCAs in BM patients treated with WBRT.
Introduction Gadolinium (Gd)-based contrast agents (GBCAs) are commonly used in various clinical MRI examinations. Some previous studies have demonstrated increased signal intensity (SI) in the dentate nucleus (DN) and globus pallidus (GP) on unenhanced T1-weighted image (UT1WI) after repeated exposure to GBCAs (1-4). Histological examination has confirmed that the increased T1 SI in the deep brain nuclei was caused by Gd deposition (5-7). Since free Gd (Gd3+) is a heavy metal with toxicity, the Gd deposition in these neurological functional nuclei has raised concerns about potential neurological complications (6, 8-10). Whole-brain radiotherapy (WBRT) is a type of irradiation treatment delivered to the entire brain over multiple visits. Due to this wide range of irradiation, DNs are always included in the irradiation field of WBRT. Additionally, head contrast-enhanced MRI examinations (CEMEs) are usually performed to assess whether there is BM when a patient is diagnosed as cancer, especially lung cancer or mammary cancer. As a result, patients with BM often receive multiple doses of GBCAs. Irradiation can cause vascular damage to increase its permeability, which may cause an accelerated leakage of Gd (11-14). The accelerated leakage of Gd may further lead to an additional increase in T1 SI in the DN after repeated exposure to GBCAs. Up to now, the extent of this impact in patients with BM has not been investigated.

Methods
Patients: The electronic medical records (EMR), PACS, and radiology information systems (RIS) were searched from January 2010 to December 2018 to identify consecutive patients with BM who had undergone WBRT and head CEMEs. Imaging Protocol: All head CEMEs were performed with one of six MRI systems at our institution. All MR examinations included an axial unenhanced T1WI sequence, T2 FLAIR, and contrast-enhanced sequence (CES).
Data Analysis: The radiation doses (RDs) to the left and right DNs were calculated. The SIs were measured using the Uniweb Viewer software (EBM Technologies, China). All measurement screenshots were saved and were all reviewed. Oval regions of interest (ROIs) were drawn as large as possible within the DN, and the same size was used for CWM near the DN.
Statistical Analysis: All analyses were performed with software SPSS and GraphPad Prism. Analysis included the Shapiro-Wilk test, paired t test, Wilcoxon test, and Spearman correlation coefficient (r).
Results WBRT plan maps for patients with BM were shown in Figure 1. WBRT could cause Gd enhancement promotion within DNs as shown in Figure 2. Comparisons of DN/CWML ratio and R between WBRT patients and control patients were shown in Figure 3. Correlations of DN/CWML ratio or R with total number of IDs or number of IDs after WBRT in Group 1 and Group 2 were shown in Figure 4 and Figure 5, respectively. Demographics and patient characteristics by group were shown in Table 1. Correlations between DN/CWML ratio or RFL dependent variables in WBRT group and control group were shown in Table 2 and Table 3, respectively.

Discussion and Conclusion The progressive increase T1 SI in the DN was found to associate with the cumulative IDs of GBCAs by many groups (4, 9, 10, 15). Our results showed that T1 SI in the DN from control patients associated with the total number of IDs, consistent with findings in other studies (2, 4, 6, 10). However, T1 SI in the DN from irradiated patients did not always correlate with cumulative doses of GBCAs, as shown in the results from Group 2. In practical terms, when a patient is diagnosed as lung cancer or mammary cancer, head CEMEs are usually performed to observe whether there is BM. Meanwhile, WBRT will start for the patient when the BM is found. Therefore, this uncertainty of irradiation time indicates that more total number of IDs does not mean more number of IDs after WBRT. Because the increase rate of T1 SI improved markedly after irradiation, the T1 SI was more likely to be determined by the number of IDs after WBRT than the total number of IDs. This is different from previous studies that have showed a continued significant relationship between T1 SI in the DN and the total number of IDs in the irradiated patients with nasopharyngeal malignancy or primary brain tumors (14, 16, 17). In these patients, irradiation and head CEME are often started closely, and thus the number of IDs after irradiation nearly equals to the total number of IDs. In order to reduce the potential neurological complications, WBRT can be improved to confine the irradiation field to the cerebrum if the brain metastases occur only in the cerebrum. Besides, more precise radiotherapy method may be used to prevent the irradiation exposure to the DNs.