Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Responses in the diffusivity and vascular function of the irradiated normal brain are seen up until 18 months following SRS of brain metastases

Responses in the diffusivity and vascular function of the irradiated normal brain are seen up... Background. MRI may provide insights into longitudinal responses in the diffusivity and vascular function of the irradiated normal-appearing brain following stereotactic radiosurgery (SRS) of brain metastases. Methods. Forty patients with brain metastases from non-small cell lung cancer (N = 26) and malignant melanoma (N  =  14) received SRS (15–25 Gy). Longitudinal MRI was performed pre-SRS and at 3, 6, 9, 12, and 18  months post-SRS. Measures of tissue diffusivity and vascularity were assessed by diffusion-weighted and perfusion MRI, respectively. All maps were normalized to white matter receiving less than 1 Gy. Longitudinal responses were as- sessed in normal-appearing brain, excluding tumor and edema, in the LowDose (1–10 Gy) and HighDose (>10 Gy) regions. The Eastern Cooperative Oncology Group (ECOG) performance status was recorded pre-SRS. Results. Following SRS, the diffusivity in the LowDose region increased continuously for 1 year (105.1% ± 6.2%; P  < .001), before reversing toward pre-SRS levels at 18  months. Transient reductions in microvascular cerebral blood volume (P < .05), blood flow (P < .05), and vessel densities (P < .05) were observed in LowDose at 6–9 months post-SRS. Correspondingly, vessel calibers in LowDose transiently increased at 3–9 months (P < .01). The responses in HighDose displayed similar trends as in LowDose, but with larger interpatient variations. Vascular responses followed pre-SRS ECOG status. Conclusions. Our results imply that even low doses of radiation to normal-appearing brain following cerebral SRS induce increased diffusivity and reduced vascular function for up until 18 months. In particular, the vascular responses indicate the reduced ability of the normal-appearing brain tissue to form new capillaries. Assessing the potential long-term neurologic effects of SRS on the normal-appearing brain is warranted. Key Points • SRS increased diffusivity and reduced microvascular function in normal-appearing brain tissue. • Increased diffusivity and reduced microvascular function were observed in low-dose regions. • Microvascular changes were associated with the pre-SRS ECOG status. © The Author(s) 2020. Published by Oxford University Press, the Society for Neuro-Oncology and the European Association of Neuro-Oncology. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS Importance of the Study Stereotactic radiosurgery (SRS) is a well-es- and the impact on neurologic function are tablished treatment option for cancer patients missing. To address this need, we here dem- with a limited number of brain metastases onstrate increased diffusivity and reduced (<5). Although SRS inherently provides rapid vascular function in the normal-appearing dose fall-off to surrounding normal-appearing brain parenchyma in patients with brain me- brain tissue, several preclinical studies sug- tastases treated with SRS even in low-dose gest that damage to the vascular network regions until 18  months follow-up. Our re- plays a key role in radiation-induced effects sults suggest that mapping tissue diffusivity in a normal brain. However, a complete un- and vascular function prior to treatment may derstanding of underlying radiobiological bring us closer to revealing the underlying responses in the surrounding normal brain functional mechanisms of SRS. Stereotactic radiosurgery (SRS) is a well-established first- metastases from non-small cell lung cancer and malignant line treatment option for patients with a limited number melanomas. of brain metastases (<5) and good performance status. Compared to the combined use of SRS and whole-brain radiotherapy (WBRT), or WBRT alone, SRS alone pro- vides similar survival rates, but reduced risk of harm to Material and Methods 2,3 neurocognition and quality of life. Due to continuous im- provements in image-guided identification of targets and Patients and Study Design organs at risk, as well as in radiation delivery techniques, SRS is also becoming increasingly attractive in cases of Forty patients with brain metastases from non-small cell multiple metastases. lung cancer (N  =  26) and malignant melanoma (N  =  14), Brain radionecrosis is the most common complica- participating in an ongoing observational MRI study tion after SRS, reported to occur in 24% of patients and (TREATMENT; clinicaltrials.gov identifier: NCT03458455), causing neurological deficits in 13%. Over the past dec- have been analyzed. The study has been approved by the ades, studies have identified the dosimetric burden on sur - Regional Ethical Committee and written informed consent rounding normal brain to be an important risk factor for the was obtained from all participants. To be eligible for inclu- development of radionecrosis, ie, the risk increases if the sion, patients must receive SRS to at least one untreated volume of brain exposed to doses higher than a threshold brain metastasis with the longest diameter >5 mm on a di- 4,5 dose, typically 10–12 Gy, is above 5–10 cc. However, agnostic MRI exam, henceforth referred to as the pre-SRS a complete understating of the underlying radiobiolog- MRI. Patient, tumor, and treatment characteristics are pro- ical response mechanisms governing the development of vided in Supplementary T able 1. radiation-induced toxicity and followed radionecrosis is Ten patients had previously received WBRT (N = 4), SRS missing. Moreover, the relevance of lower doses from SRS (N  =  3), or both (N  =  3) to other metastases not targeted (<8 Gy) to normal-appearing brain for potential long-term by SRS in this study. Furthermore, 5 patients received con- effects is poorly understood. comitant immunotherapy. Eighteen patients were using Several preclinical studies have recognized damage to corticosteroids (individual dosage) at the pre-SRS MRI, the vascular network to play a key role in radiation-induced while 21 patients did not (information from one patient effects on normal brain tissue. Radiation-induced vascular was not attainable). damage involves early endothelial cell injury and apop- At the time of study data lock, MRI exams were per- tosis, followed by decreased vessel densities, dilation, and formed every 3 months for the first year and at 18 months thickening of the blood vessels. Additionally, microvessel post-SRS. Dropout of patients on follow-up MRIs was thrombosis with vessel occlusion may occur within weeks due to death (N  =  7), deemed clinically unfit to perform to years after irradiation. In patients, increased vessel per- an MRI exam (N = 9), or too short follow-up time (N = 3). meability in normal-appearing brain tissue, measured by The first follow-up MRI at 3  months post-SRS was per - perfusion MRI during fractionated radiotherapy of low- formed within a margin of 1 week, while the remaining grade gliomas, has been shown to correlate with reduced MRI exams were performed within a 2-week margin. neurocognitive function. However, clinical studies of After SRS, 12 patients were treated with additional WBRT vascular responses in irradiated normal-appearing brain (N  =  4) or SRS (N  =  7) or both (N  =  1). MRI exams per- tissue are sparse, and most studies have evaluated acute formed on these patients after the additional brain radi- and early responses to fractionated radiotherapy rather otherapy were excluded from further analysis, leaving than SRS. the following number of patients at each MRI readout: To this end, our study sheds light on responses in the dif- N = 40 (pre-SRS), N = 38 (+3 months), N = 29 (+6 months), fusivity and vascular function of irradiated non-cancerous N  =  22 (+9  months), N  =  18 (+12  months), and N  =  15 normal-appearing brain tissue following SRS of brain (+18 months). Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS Stereotactic Radiosurgery Co-registration of Dose Distribution and Longitudinal MRI Data SRS was delivered using a frameless linear accelerator- based system (Varian TrueBeam vSTx, HD-MLC 120; The planning CT images and corresponding SRS dose multileaf collimator of 2.5  mm). The SRS planning was distribution were exported from the planning system and mainly performed using iPlan RT Dose (v4.5.4, Brainlab co-registered to the pre-SRS high-resolution T1-weighted AG). RayStation (v5.0, Raysearch Laboratories) was used post-contrast MRIs using normalized mutual informa- in one case. Pre-SRS 3D high-resolution T1-weighted post- tion in SPM12 (Statistical Parametric Mapping [SPM] contrast MRIs (distortion-corrected) were used for delin- toolbox version 12, University College London, England). eation of the metastases and organs at risk, while dose For each MRI exam, diffusion-weighted images, high- calculation was performed on co-registered computer to- resolution post-contrast and FLAIR images, with asso- mography (CT) images. The time from the pre-SRS MRI ciated tumor, edema, as well as binary white and gray to the planning CT scan was on average 4  days (range: matter masks, were co-registered to the DSC MRI space 5–12  days). Delineation of the gross tumor volume was by means of normalized mutual information in nordicICE performed by a radiation oncologist. A  2  mm isotropic (NordicNeuroLab AS) or SMP12. Additionally, for the pre- margin, accounting for both subclinical disease and plan- SRS MRIs, the dose distribution was co-registered to the ning uncertainties, was automatically added by the soft- DSC MRI space via the T1-weighted post-contrast MRIs. ware to generate the planning target volume. The SRS The pre-SRS DSC space was used as the reference space dose was set to cover at least 99% of the planning target for co-registration of all longitudinal MRI data. Hence, after volume and ranged from 15 to 25 Gy (single fraction) or final co-registration, the data from all follow-up MRIs and from 20.1 to 27 Gy (3 fractions) (Supplementary Table 1). SRS dose distribution were in the pre-SRS DSC space. The SRS dose and fractionation scheme were determined according to institutional guidelines, which are based on previous treatment history and clinical status of the patient Normal-Appearing Brain together with tumor size, proximity to organs at risk, and White and gray matter was identified on high-resolution normal tissue dose constraints. The mean volume of study T1-weighted post-contrast or FLAIR images by calculation metastases was significantly lower in patients treated with of probability density maps using the segmentation tool in 1 versus 3 fractions (Supplementary Figure 1). SPM12. The binary masks of white and gray matter were All patients received a corticosteroid dose (Medrol) of obtained by applying a probability threshold of >0.85 (1.0 32 mg (single fraction) or 16 mg (3 fractions) immediately being the highest probability). Normal-appearing brain after SRS and on the same night, as well as the following tissue was defined by the white and gray matter masks, day (morning and night). Those treated with 3 fractions excluding areas with pathological contrast enhancement continued to receive 2 doses every day until the night of and edema. The pathological contrast enhancement, not the last fraction. Thereafter, the use of corticosteroids was excluding central necrosis, was delineated on the high- individually managed by the patients treating physicians. resolution T1-weighted post-contrast images by 2 expe- rienced neuro-radiologists. Edema was defined on FLAIR MRI Protocol images in native or DSC MRI space. All delineations were performed in nordicICE. All MRI exams were performed using a dedicated 20-channel The co-registered dose distribution (Figure  1A) was di- head/neck coil on a 3 T Skyra (Siemens Healthineers) with vided into 3 distinct isodose levels: Reference <1 Gy, the following protocol: 3D T1-weighted images, before LowDose 1–10 Gy, and HighDose >10 Gy (Figure  1B). For and after injection of Gadolinium-based contrast agent patients who received SRS in 3 fractions (N  =  12), the (Dotarem 279.3  mg/mL, 0.2  mL/kg bodyweight; repetition linear-quadratic model with α/β for brain parenchyma = 2 time [TR]/echo time [TE] = 700 ms/12 ms; voxel size = 0.9 × Gy was used to calculate corresponding dose regions: 0.9 × 0.9 mm ; acquisition matrix = 512 × 512), T2-weighted <1.31 Gy (Reference), 1.31–16.2 Gy (LowDose), and >16.2 fluid attenuated inversion recovery (FLAIR) (TR/TE/inver - Gy (HighDose). Changes in tissue diffusivity and vascular sion time [TI] = 5000 ms/387 ms/1800 ms; voxel size = 0.9 × function were assessed in normal-appearing brain tissue 0.9  × 0.9  mm ; acquisition matrix  =  512  × 512; field of in the LowDose and HighDose regions (Figure  1C). The view  =  460.8  × 460.8  mm ), diffusion-weighted imaging volume of normal-appearing brain in the LowDose and (TR/TE  =  5960  ms/71  ms; b-values  =  0 and 1000/1500  s/ HighDose regions at all follow-up MRIs are provided in 2 3 mm ; voxel size = 1.22 × 1.22 × 4.0 mm ; slice gap = 5.0 mm; Supplementary T able 2. acquisition matrix  =  180  × 180; field of view  =  219.6  × 219.6 mm ), and dynamic susceptibility contrast (DSC) MRI with combined gradient-echo and spin-echo acquisitions Quantification of Tissue Diffusivity and Vascular (TR = 1500 ms; TE [gradient-echo] = 13 ms [malignant mel- Function anoma]/15–30  ms [non-small cell lung cancer]; TE [spin- The tissue diffusivity was assessed from ADC maps created echo]  =  104  ms; voxel size  =  2.0–2.2  × 2.0–2.2  × 5.0  mm ; directly on the MRI scanner from the diffusion-weighted slice gap  =  6.5  mm; acquisition matrix  =  120  × 90; field images, using Stejskal–Tanner diffusion approximation. of view  =  240–264  × 180–198  mm ) with a bolus injection Vascular function was assessed by analysis of the per- (3  mL/s) of contrast agent, followed by 30  mL of physio- fusion MRI as previously described. In short, standard logic saline solution. Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS A Original dose distribution B Dose regions C Regions-of-interest Normal brain tissue LowDoseHighDose White matter White matter Gray matter Gray matter 0Gy 25Gy Reference Lowdose Highdose Edema Tumor <1Gy 1–10Gy >10Gy Figure 1 Regions-of-interests used for assessment of normal-appearing brain tissue responses to SRS. (A) A  representative co-registered stereotactic dose distribution for a patient with brain metastasis from non-small cell lung cancer as an overlay of a T1-weighted post-contrast image acquired pre-SRS. The prescribed SRS dose was 25 Gy. (B) The dose distribution was divided into 3 dose regions: Reference: <1 Gy (blue overlay), LowDose: 1–10 Gy (green overlay), and HighDose: >10 Gy (yellow overlay). (C) The final regions-of-interest used for longitudinal response assessments included normal-appearing brain tissue, ie, white and gray matter in LowDose (white matter: light green overlay, gray matter: dark green overlay) and HighDose (white matter: light yellow overlay, gray matter: dark yellow overlay), excluding edema (purple overlay) and tumor (red overlay). voxel-wise DSC MRI kinetic analyses of the spin- and densities, could not be calculated for these patients at the gradient-echo acquisitions were performed in nordicICE, given MRI exams. providing parametric maps of cerebral blood volume (CBV) and cerebral blood flow (CBF). Whereas the spin-echo maps ECOG Performance Status and Eloquent Regions reflect the micro-vasculature, the gradient-echo maps rep- resent the total micro-to-macroscopic vasculature, and Pre-SRS, the Eastern Cooperative Oncology Group the prefixes “Micro” and “Macro” are henceforth used for (ECOG) performance status was recorded for 31 of the vascular metrics obtained from spin-echo and gradient- patients as follows: 0 (N = 13), 1 (N = 16), and 2–3 (N = 2). echo, respectively. The DSC MRI analysis included motion Potential differences in normal-appearing tissue re- correction, automatic detection of the arterial input func- sponses were assessed in patients with pre-SRS ECOG tion with deconvolution by standard single value decom- status 0 versus >0. Due to missing spin-echo data, the position, and contrast agent leakage-correction adapted final number of patients assessed were (ECOG  =  0/>0) for both T1- and T2-shortening effects. Vessel caliber anal- N  =  12/17 (pre-SRS), N  =  11/15 (+3  months), N  =  9/10 ysis was performed in Matlab (v.R2017a, MathWorks Inc.), (+6 months), N = 7/8 (+9 months), N = 6/4 (+12 months), providing estimations of mean vessel calibers and mean and N = 6/4 (+18 months). vessel densities. Binary masks of all the study metastases, as well as any From all MRI exams, normalized parametric maps additional metastases also treated with SRS at the same were calculated by dividing all image voxel values to time, were co-registered to the MNI space for assessment the respective median value of white matter within the of their location related to eloquent regions determined Reference region (Figure  1B), henceforth prefixed n. For from binary masks provided by the SPM12. each normalized parametric map, the mean value, ex- cluding outliers >3 SD away from the mean, of all regions- of-interests were computed if >8 non-zero voxels were Statistical Analysis present. Longitudinal changes were assessed relative to pre-SRS in percent (%), mean ± SD. Mann–Whitney U test or Fisher’s exact test was used to Spin-echo readout data were missing for the following compare groups of continuous and dichotomized data, number of patients: N  =  2 (pre-SRS), N  =  1 (+3  months), respectively. Comparisons between absolute and relative N = 2 (+6 months), N = 1 (+9 months), N = 2 (+12 months), changes (to pre-SRS) in the regions-of-interest were made and N  =  1 (+18  months). Thus, nMicro-CBV and nMicro- using the Wilcoxon signed-rank test. The significance level CBF, as well as nMean vessel calibers and nMean vessel was 5%, including Holm-Bonferroni correction in the case Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS of multiple comparisons. The statistical analyses were per- The HighDose region of patients set to receive 3 frac- formed using Matlab or IBM SPSS Statistics (v25). tions to a minimum of 1 metastasis versus a single fraction showed lower vascular metrics. Specifically, nMicro-CBV was 1.22  ± 0.20 versus 1.44  ± 0.21 (P < .01), nMicro-CBF was 1.23 ± 0.20 versus 1.48 ± 0.28 (P < .05), and Results nMean vessel densities was 0.83 ± 0.19 versus 0.98 ± 0.12 (P < .05). Furthermore, the LowDose region showed lower Diffusivity and Vascular Profiles of the nMacro-CBV in patients having received the previous ra- Pre-irradiated Brain diotherapy to the brain compared with those who had not Compared to the LowDose and HighDose regions, the me- (P < 0.05) (Supplementary Table 5). No differences in vas- tastases and associated edema displayed pathologic dif- cular function in the LowDose and HighDose regions were fusivity and vascular function pre-SRS (Figure  2). In the observed between patients treated with or without cor- tumor, the nADC (P < .001), nMacro-CBV (P < .01), nMacro- ticosteroids (Supplementary Table 6). However, slightly CBF (P < .05), and nMean vessel calibers (P < .001) were higher nADC was observed in the LowDose region in pa- elevated, while nMicro-CBF (P < .01) and nMean vessel tients treated with corticosteroids (1.14 ± 0.05) compared densities (P < .001) were lower. Associated edema showed to untreated patients (1.11 ± 0.05) (P < .05). Moreover, no higher nADC (P < .001), nMicro-CBV, nMicro-CBF, nMacro- differences in diffusivity or vascular function were ob- CBV, and nMacro-CBF (P < .001) and lower nMean vessel served between patients with different primary diagnosis densities (P < .001) compared to normal-appearing brain (Supplementary Table 7), or between patients having re- tissue (Supplementary T able 3). ceived previous immunotherapy or not (Supplementary In the LowDose and HighDose regions, the gray matter Table 8), or between patients with pre-SRS ECOG status was characterized by higher diffusivity and increased 0 versus >0 (Supplementary Table 9). No differences in vascular function compared to white matter, with mean the absolute or relative change in tumor or edema vol- gray-to-withe matter ratios of 1.2 for nADC, and ranging umes were observed depending on the pre-SRS ECOG from 1.1 to 1.8 for the vascular metrics (Supplementary status at any of the MRI examinations. However, pa- Table 4). tients with EOCG status >0 more frequently presented A C nMicro-CBV E nMacro-CBV G nMean vessel T1-weighted post- calibers contrast MRI Edema Tumor BD nADC nMicro-CBF FH nMacro-CBF nMean vessel densities Figure 2 Normalized parametric maps of diffusivity and vascular function pre-SRS. (A) T1-weighted post-contrast image in DSC space with a brain metastasis from non-small cell lung cancer (red overlay) and associated edema (purple overlay), and normalized (n) parametric maps; (B) the apparent diffusion coefficient (nADC), (C) nMicro-vascular cerebral blood volume (CBV), (D) nMicro-vascular cerebral blood flow (CBF), (E) nMacro-CBV, (F) nMacro-CBF, (G) nMean vessel calibers, and (H) nMean vessel densities. Low High Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS with metastases in eloquent regions at pre-SRS (P < .05) (P < .001), returning to 102.5% ± 4.1% at 18  months post- (Supplementary T able 10). SRS. Similarly, an increasing trend was observed for nADC in HighDose, also peaking 1  year post-SRS at 109.6% ± 18.8%, but with larger interpatient variations. The vascular responses in the LowDose region showed Increased Diffusivity and Reduced Vascular transient reduction in nMicro-CBV (Figure 3Bi) and nMicro- Function of Normal Irradiated Brain CBF (Figure 3Bii), combined with stable levels of nMacro- Following SRS, the LowDose region showed increased dif- CBV (Figure  3Diii) and nMacro-CBF (Figure  3Biv), as well fusivity (Figure 3Ai) combined with reduced vascular func- as transient increase in nMean vessel calibers (Figure 3Bv) tion for up until 18 months (Figure  3Bi–vi). The increase in and reduction in nMean vessel densities (Figure  3Bvi). nADC in LowDose peaked 1 year post-SRS at 105.1% ± 6.1% Eighteen months post-SRS, the vascular function in the A Diffusivity C Longitudinal nMicro-CBV Pre-SRS +3 months +6 months nADC N = 40 38 29 22/21 18 15 LowDose T1-weighted post-contrast HighDose 105 * 5.5 LowDose HighDose nMicro-CBV MRI (Months) P < 0.05 Relative to pre-SRS 0.3 # P < 0.01 Relative to pre-SRS * P < 0.001 Relative to pre-SRS B Vascular function i iii v nMicro-CBV nMacro-CBV nMean vessel calibers 115 115 125 N = 38 35 25 20/19 15 14 N = 40 36 27 21/20 17 15 N = 38 35 24 20/19 15 14 110 110 LowDose LowDose HighDose HighDose 105 105 100 100 95 95 LowDose HighDose 90 90 90 MRI (Months) MRI (Months) MRI (Months) ii iv vi nMicro-CBF nMacro-CBF nMean vessel densities 115 115 125 N = 38 35 25 20/19 15 14 N = 40 36 27 21/20 17 15 N = 38 35 25 20/19 15 14 110 110 LowDose LowDose HighDose HighDose 105 105 100 100 95 95 LowDose HighDose 90 90 90 MRI (Months) MRI (Months) MRI (Months) Figure 3 Increased diffusivity and decreased vascular function in normal-appearing brain up until 18  months post-SRS. Mean ± SEM of normalized (n) metrics relative to pre-SRS (%) in normal-appearing brain tissue in LowDose (green) and HighDose (yellow) on a logarithmic scale: (Ai) the apparent diffusion coefficient (nADC), (Bi) nMicro-vascular blood volume (CBV), (Bii) nMicro-vascular blood flow (CBF), (Biii) nMacro- CBV, (Biv) nMacro-CBF, (Bv) nMean vessel calibers, and (Bvi) nMean vessel densities. Transient changes in diffusivity and vascular function are highlighted by shaded green areas. The number of patients (N) included at each time point is shown in gray. At 9 months post-SRS, analysis of the HighDose region was excluded for one patient due to insufficient number of voxels in this region (total number of patients analyzed is shown in yellow). (C) Reduced nMicro-CBV in LowDose at 3 and 6 months post-SRS compared to pre-SRS is shown for a patient with a brain metas- tasis from malignant melanoma (bottom row), with corresponding T1-weighted post-contrast images indicating LowDose (green) and HighDose (yellow) regions (top row). P values from the Wilcoxon Singed-Rank test relative to pre-SRS. Pre-SRS Pre-SRS Pre-SRS Pre-SRS Pre-SRS Pre-SRS Pre-SRS Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS LowDose region had returned to pre-SRS levels. In the CBV and CBF, as well as reduced vessel densities. While no HighDose region, similar trends as in LowDose were ob- apparent changes to SRS in normal-appearing tissue were served for nMicro-CBV and nMicro-CBF, as well as nMean observed on conventional MRI, our findings indicate that vessel calibers. However, nMacro-CBV and nMacro-CBF the micro-vasculature is particularly vulnerable, even to and nMean vessel densities showed a continued increasing low-dose irradiation. Combined with the observed stable trend from 6, 9, and 3 months post-SRS, respectively. levels of total CBV and CBF, as well as increases in vessel No apparent dose-dependency or plateau dose of the calibers, our results further imply that the vasculature of responses in diffusivity or vascularity was observed the irradiated normal brain loses radio-sensitive, yet well- (Supplementary Figure 2). Moreover, the responses in the functioning and highly differentiated small capillaries. LowDose region for nMicro-CBV and nMicro-CBF, as well Instead, the post-radiated tissue is dominated by larger as nMean vessel calibers, were different for patients with vessels with potentially reduced function. pre-SRS ECOG 0 versus >0 (Figure  4). Specifically, pa- The radiation-induced effects on endothelial vascular in- tients with ECOG status >0 showed transient reductions jury in both tumors and normal tissues are known to mod- in nMicro-CBV to 93.6% ± 8.9% (P < .05) at 6 months and ulate angiogenesis and neovascularization. Our findings to 94.1% ± 4.8% at 9  months (P < .01; Figure  4A). The re- in humans are in line with preclinical studies showing that duction at 9  months was significantly lower than for pa- doses as low as 2 Gy may induce significant reductions of tients with pre-SRS ECOG of 0 (P < .01), showing a small small capillaries (diameter ≤10 µm). Moreover, endothelial increase of 101.6% ± 7.3%. Likewise, the transient reduc- apoptosis caused by irradiation has been shown to induce tions in nMicro-CBF at 6–9  months (P < .01) were lower increased vessel dilation and vessel permeability and thus for patients with pre-SRS ECOG >0 versus 0 at 9  months reduced vascular function. Interestingly, in our study we (P < .01) and 1 year post-SRS (P < .05; Figure 4B). Finally, at observed consistent increase in mean vessel densities 6 months post-SRS, nMean vessel calibers was increased from 6 months post-SRS in normal-appearing brain tissue to 109.6% ± 8.8% (P < .01) in the pre-SRS ECOG >0 patients regions having received high doses (>10 Gy) compared compared to stable levels of 100.4% ± 6.2% for the patients to low doses (1–10 Gy). Approximately 90% of endothe- with ECOG status 0 (P < .05; Figure 4C). lial cells receiving conventional fractionated radiotherapy experience mitotic cell death without apoptosis, whereas higher doses also induce apoptotic cell death. Such chronic effects are commonly reflected in upregulated en- dothelial cell senescence in the cerebral vascular system. Discussion The observed abnormal vascular response, and especially In our study of 40 patients with brain metastases from in the HighDose region, could therefore suggest abnormal non-small cell lung cancer and malignant melanoma, we revascularization from capillary rarefaction, increased vas- show by MRI that normal-appearing brain tissue having re- cular permeability, and impaired vascular homeostasis. ceived low doses (1–10 Gy) from SRS displays increased Histopathology of patient specimens of various ce- diffusivity and decreased vascular function for up until rebral neoplasms with radiation-induced injury shows 18 months post-SRS. The decreased vascular function was that damaged tissue, in addition to vascular changes, is particularly expressed by reduced levels of microvascular characterized by macrophage invasion, demyelination, Longitudinal Vascular responses in LowDose by pre-SRS ECOG status nMicro-CBV nMicro-CBF nMean vessel calibers AB C 115 115 130 Pre-SRS ECOG = 0 Pre-SRS ECOG = 0 N = 12, 11, 9, 7, 6, 6 N = 12, 11, 9, 7, 6, 6 110 110 Pre-SRS ECOG > 0 Pre-SRS ECOG > 0 N = 17, 15, 10, 8, 4, 4 N = 17, 15, 10, 8, 4, 4 # 105 105 100 100 < 0.05 < 0.05 < 0.01 < 0.01 95 95 Pre-SRS ECOG = 0 N = 12, 11, 9, 7, 6, 6 90 90 90 Pre-SRS ECOG > 0 N = 17, 15, 10, 8, 4, 4 85 85 MRI (Months) MRI (Months) MRI (Months) P < 0.05 Relative to pre-SRS; # P < 0.01 Relative to pre-SRS; P value between groups Figure 4 Reduced vascular function in normal-appearing brain is more pronounced in a patient with pre-SRS ECOG status >0 compared to 0.  Mean ± SEM of normalized (n) metrics relative to pre-SRS (%) in normal-appearing brain tissue in LowDose on a logarithmic scale for pa- tients with pre-SRS Eastern Cooperation Oncology Group (ECOG) status of 0 (gray) and >0 (brown). (A) Transiently reduced nMicro-vascular blood volume (CBV) and (B) nMicro-vascular blood flow (CBF) in patients with pre-SRS ECOG status >0 compared to increased nMicro-CBV at 9–12 months post-SRS in patients with pre-SRS ECOG status 0. (C) Transient increase in nMean vessel calibers was higher in patients with pre-SRS ECOG status >0 compared to 0. The number of patients (N) with available data in each group is shown in the legend box. P values from Mann–Whitney U test (between groups) and the Wilcoxon Signed-Rank test (relative to pre-SRS). Pre-SRS Pre-SRS Pre-SRS Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS and reactive gliosis. In particular, the presence of de- In line with our findings, reduced CBV and CBF have myelination and reactive gliosis both cause higher tissue been observed in normal-appearing brain tissue up until diffusivity measured by diffusion-weighted MRI and 12  months after SRS to arteriovenous malfunctions 14,15 diffusion tensor imaging (DTI). The increased diffu- and conventional radiotherapy of low- and high-grade 20–24 sivity of normal-appearing tissue observed in our study gliomas. In the latter studies, a dose-dependent re- may be due to microstructural loss caused by such pro- duction of vascular function was observed, with larger cesses. Recent DTI studies have demonstrated increased reductions in dose regions receiving more than 30% of diffusivity of normal-appearing white matter after frac- the prescribed dose (30–66 Gy). Increased vascular re- tionated radiotherapy of primary brain tumors, attrib- sponses in normal-appearing tissue during conventional uted to demyelination, and correlated with neurological radiotherapy of gliomas and early after SRS to brain me- 16,17 18 deficits. Specifically, in one of these studies including tastases have also been reported. While a weak positive 54 adult patients, increased diffusivity in white matter correlation between dose and increased vascular met- beneath the cingulate cortex at 3 and 6  months after ra- rics, with a plateau effect at 10 Gy, was observed after diotherapy was associated with a decline in verbal set- SRS, the strongest dose-dependency was found during shifting ability and cognitive flexibility. In light of these the fractionated radiotherapy of the gliomas and was ac- findings, our results reinforce the potential future clinical tually diminished 6  months post-radiotherapy. In light of value of using advanced MRI to identify dose tolerances to our findings, showing no apparent linear correlation be- subclinical microstructural changes that can be integrated tween responses in diffusivity or vascular function, as- into the treatment planning and subsequently reduce sessed 3–18 post-SRS, the dose-dependency may be more neurocognitive decline. pronounced during and early after irradiation—further The responses in diffusivity and vascular function of indicating a dose-independent ability to restore vascular normal-appearing brain tissue in HighDose (>10 Gy), function. Altogether, the present studies having inves- mainly representing peri-tumoral tissue, showed similar tigated advanced MRI for normal brain tissue response 17–24 trends as in LowDose up until 1 year post-SRS. However, assessment, including ours, demonstrate the poten- larger interpatient variations were observed. This may re- tial for MRI metrics to identify normal brain tissue toler- flect variability in the pre-SRS diffusivity and vascular ances that are essential for optimizing the treatment plan levels of the peri-tumoral region and consequently a dif- accordingly. ferent premise to handle high-dose radiation. Poor vas- Our study observed distinctive longitudinal responses cular function in the peri-tumoral region pre-SRS has been for patients having a pre-SRS ECOG status 0 compared shown to be associated with radiation-induced changes. >0. Though no apparent differences in diffusivity or vas- Moreover, while at 18 months post-SRS the normal brain cular function were detectable by the diffusion or per- tissue diffusivity and vascular function in LowDose ap- fusion MRI pre-SRS, underlying subtle differences may parently returned to pre-SRS levels, all the vascular met- have been present. The vasculature in normal brain tissue rics showed an increasing trend starting at 6–12  months of patients with ECOG status >0 may already have been post-SRS in HighDose. The increase was particularly ap- impaired and becoming reinforced by the radiation dose. parent for the nMacro-CBV and CBF, as well as mean vessel While the association between cerebro-microvascular densities, suggesting a vascular network dominated by dysfunction and early neurologic pathogenesis is in- 26,27 larger vessels and reduced capability of recruiting smaller triguing, it is well recognized that CBF levels and cog- capillaries. This deterioration of the vascular network may nitive impairment are inversely correlated. In our study, be a contributing factor of the underlying mechanisms in- changes in ECOG performance status could not be cor- volved in the development of radionecrosis—which has related to responses in diffusivity or vascular function as an increasing risk of occurring with increasing volume of the ECOG status was not routinely recorded at the clinical brain exposed to high doses, in particular, 10 Gy and 12 follow-ups post-SRS. This is however warranted in future 4 5 Gy. However, as stressed by Milano et  al. the develop- studies. ment of radionecrosis is likely impacted by multiple fac- Our study has limitations. Outside of the study period, tors, including possible regional variations in susceptibility individual treatment management resulted in large var- to radiation injury. From our data, showing more homoge- iations with respect to previous and salvage radiation nous tissue responses in LowDose compared to HighDose, therapy to the brain, prescription dose and fractionations we hypothesize that these changes also play a part in the of the study SRS, use of immunotherapy, systemic treat- development of radionecrosis—further supported by an ment, and corticosteroid treatment. Combined with the occasional manifestation of radionecrosis outside the longitudinal reduction of patients, multivariate analyses high-dose field. were thus not feasible. However, a separate analysis of Although radiation exposure to normal brain tissue is the different treatment groups showed that excluding reduced with SRS compared to WBRT, large volumes of patients previously treated with brain radiotherapy did normal tissue still receive low doses. Our results thus not alter the outcome of the longitudinal response anal- support future studies on effects of low-dose irradiation ysis. In our study and within the power of our sample for development of radionecrosis, neurologic deficits, and size, the patients treated with 3 fractions versus the pa- second malignancies. Such potential long-term effects tients treated with a single fraction did not show any may be especially relevant with increased use of SRS to apparent differences in the longitudinal changes in dif- multiple and/or larger metastases, and with increasing fusivity or vascular function. We aimed to reduce the po- treatment efficacy that potentially can result in longer sur - tential impact of different fractionation regimes on our vival times for these patients. findings by calculating equivalent doses (EQD2) for the Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS dose distributions delivered in 3 fractions. The validity of the linear-quadratic model is less established for higher Acknowledgments doses (>8–10 Gy), but the LowDose region is mainly dom- inated by doses <10 Gy and should therefore be less in- We thank Dr. Knut Håkon Hole at Oslo University Hospital for a fluenced by this restriction. The lower pre-SRS perfusion constructive review of the manuscript. levels in what later became the HighDose region in pa- tients receiving 3 rather than a single SRS fraction may well be due to differences in tumor size. Larger tumors, more commonly treated with fractionated SRS, may in- Conflict of interest statement. Intellectual property right, herently be more prone to reduced vascular function NordicNeuroLab AS, Bergen, Norway (K.E.E). in the peri-tumoral region. Thus, although receiving less than 10 Gy per fraction, which has been shown to be a threshold dose for inducing apoptotic endothelial cells death, the impaired pre-SRS vasculature may re- Authorship Statement. Experimental design: L.B.N, C.S., K.D.J., duce the level of this threshold dose. Furthermore, no Å.H., and K.E.E. Implementation: L.B.N, I.D, E.G., C.S., O.G., differences in pre-SRS diffusivity and vascular function K.D.J., Å.H., and K.E.E. Analysis and/or interpretation of the data: of normal brain tissue or longitudinal responses were L.B.N., I.D., E.G., C.S., A.L., D.O.S., and K.E.E. Writing of the first present between patients having received immuno- manuscript draft: L.B.N., I.D., E.G., C.S., A.L., D.O.S., and K.E.E. therapy or not. However, preliminary analysis of the cur- Editing: all coauthors. Read and approved the final version: all rent study population does imply that concomitant SRS coauthors. Unpublished material: no unpublished material is and immunotherapy render peri-tumoral regions more referenced. prone to changes than SRS alone, indicating dose- dependent synergetic effect immunotherapy to SRS. Finally, corticosteroid treatment has been suggested to decrease vascular functions, but conflicting results are 30,31 reported in the literature. Due to lack of information References about corticosteroid use on the follow-up MRIs, this could not be account for in our data. However, at pre-SRS no differences in any vascular metrics were observed be- 1. Hartgerink  D, van  der  Heijden  B, De  Ruysscher  D, et  al. Stereotactic tween patients using corticosteroids and not. radiosurgery in the management of patients with brain metastases of Our longitudinal MRI study indicates that normal brain non-small cell lung cancer: indications, decision tools and future direc- tissue is sensitive to low doses of irradiation following tions. Front Oncol. 2018;8(9):154. SRS, resulting in loss of, and potentially, reduced ability to 2. Chang  EL, Wefel  JS, Hess  KR, et  al. Neurocognition in patients with form well-functioning capillaries. The long-term implica- brain metastases treated with radiosurgery or radiosurgery plus tions of our findings in terms of neurological function and whole-brain irradiation: a randomised controlled trial. Lancet Oncol. clinical outcomes are thus warranted in future studies. 2009;10(11):1037–1044. 3. Schimmel WCM, Gehring K, Eekers DBP, Hanssens PEJ, Sitskoorn MM. Cognitive effects of stereotactic radiosurgery in adult patients with brain metastases: a systematic review. Adv Radiat Oncol. 2018;3(4):568–581. Supplementary Data 4. Minniti G, Clarke E, Lanzetta G, et al. Stereotactic radiosurgery for brain Supplementary data are available at Neuro-Oncology metastases: analysis of outcome and risk of brain radionecrosis. Radiat Advances online. Oncol. 2011;6(15):48. 5. Milano  MT, Usuki  KY, Walter  KA, Clark  D, Schell  MC. Stereotactic radiosurgery and hypofractionated stereotactic radiotherapy: normal tissue dose constraints of the central nervous system. Cancer Treat Rev. Keywords 2011;37(7):567–578. 6. Sundgren  PC, Cao  Y. Brain irradiation: effects on normal brain brain metastases | diffusion-weighted MRI | normal- parenchyma and radiation injury. Neuroimaging Clin N Am. appearing brain tissue response | perfusion MRI | stereo- 2009;19(4):657–668. tactic radiosurgery 7. Belka  C, Budach  W, Kortmann  RD, Bamberg  M. Radiation induced CNS toxicity—molecular and cellular mechanisms. Br J Cancer. 2001;85(9):1233–1239. 8. Cao Y, Tsien CI, Sundgren PC, et al. Dynamic contrast-enhanced magnetic resonance imaging as a biomarker for prediction of radiation-induced Funding neurocognitive dysfunction. Clin Cancer Res. 2009;15(5):1747–1754. This work was supported by the South-Eastern Norway 9. Brenner DJ. The linear-quadratic model is an appropriate methodology Regional Health Authority (2016102, 2013069); the Norwegian for determining isoeffective doses at large doses per fraction. Semin Cancer Society (6817564); the European Research Council under Radiat Oncol. 2008;18(4):234–239. the European Union’s Horizon 2020 (758657); and the Research 10. Digernes I, Grøvik E, Nilsen LB, et al. Brain metastases with poor vas- Council of Norway (261984). cular function are susceptible to pseudoprogression after stereotactic radiation surgery. Adv Radiat Oncol. 2018;3(4):559–567. Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS 11. Emblem  KE, Farrar  CT, Gerstner  ER, et  al. Vessel caliber—a potential 22. Lee  MC, Cha  S, Chang  SM, Nelson  SJ. Dynamic susceptibility con- MRI biomarker of tumour response in clinical trials. Nat Rev Clin Oncol. trast perfusion imaging of radiation effects in normal-appearing brain 2014;11(10):566–584. tissue: changes in the first-pass and recirculation phases. J Magn Reson 12. Venkatesulu BP, Mahadevan LS, Aliru ML, et al. Radiation-induced en- Imaging. 2005;21(6):683–693. dothelial vascular injury: a review of possible mechanisms. JACC Basic 23. Petr  J, Platzek  I, Seidlitz  A, et  al. Early and late effects of Transl Sci. 2018;3(4):563–572. radiochemotherapy on cerebral blood flow in glioblastoma pa- 13. Sundgren  PC, Fan  X, Weybright  P, et  al. Differentiation of recurrent tients measured with non-invasive perfusion MRI. Radiother Oncol. brain tumor versus radiation injury using diffusion tensor imaging in 2016;118(1):24–28. patients with new contrast-enhancing lesions. Magn Reson Imaging. 24. Price SJ, Jena R, Green HA, et al. Early radiotherapy dose response and 2006;24(9):1131–1142. lack of hypersensitivity effect in normal brain tissue: a sequential dy- 14. Hagen T, Ahlhelm F, Reiche W. Apparent diffusion coefficient in vasogenic namic susceptibility imaging study of cerebral perfusion. Clin Oncol (R edema and reactive astrogliosis. Neuroradiology. 2007;49(11):921–926. Coll Radiol). 2007;19(8):577–587. 15. Natarajan R, Hagman S, Wu X, et al. Diffusion tensor imaging in NAWM 25. Balagamwala  EH, Chao  ST, Suh  JH. Principles of radiobiology of ster- and NADGM in MS and CIS: association with candidate biomarkers in eotactic radiosurgery and clinical applications in the central nervous Sera. Mult Scler Int. 2013;2013:265259. system. Technol Cancer Res Treat. 2012;11(1):3–13. 16. Chapman  CH, Nagesh  V, Sundgren  PC, et  al. Diffusion tensor im- 26. Toth  P, Tarantini  S, Csiszar  A, Ungvari  Z. Functional vascular contribu- aging of normal-appearing white matter as biomarker for radiation- tions to cognitive impairment and dementia: mechanisms and conse- induced late delayed cognitive decline. Int J Radiat Oncol Biol Phys. quences of cerebral autoregulatory dysfunction, endothelial impairment, 2012;82(5):2033–2040. and neurovascular uncoupling in aging. Am J Physiol Heart Circ Physiol. 17. Tringale KR, Nguyen T, Bahrami N, et al. Identifying early diffusion imaging 2017;312(1):H1–H20. biomarkers of regional white matter injury as indicators of executive func- 27. Østergaard  L, Aamand  R, Gutiérrez-Jiménez  E, et  al. The capillary tion decline following brain radiotherapy: a prospective clinical trial in pri- dysfunction hypothesis of Alzheimer’s disease. Neurobiol Aging. mary brain tumor patients. Radiother Oncol. 2019;132(March):27–33. 2013;34(4):1018–1031. 18. Jakubovic R, Sahgal A, Ruschin M, Pejović-Milić A, Milwid R, Aviv RI. 28. Alosco ML, Gunstad J, Jerskey BA, et al. The adverse effects of reduced Non tumor perfusion changes following stereotactic radiosurgery to cerebral perfusion on cognition and brain structure in older adults with brain metastases. Technol Cancer Res Treat. 2015;14(4):497–503. cardiovascular disease. Brain Behav. 2013;3(6):626–636. 19. Taki  S, Higashi  K, Oguchi  M, et  al. Changes in regional cerebral 29. Nilsen  LB, Groevik  E, Digernes  I, et  al. Vascular responses in normal blood flow in irradiated regions and normal brain after stereotactic brain tissue after combined immunotherapy and SRS to brain metas- radiosurgery. Ann Nucl Med. 2002;16(4):273–277. tases. Radiother Oncol. 2019;133(Suppl 1):S551–S552. 20. Fahlström  M, Blomquist  E, Nyholm  T, Larsson  EM. Perfusion magnetic 30. Bastin  ME, Carpenter  TK, Armitage  PA, Sinha  S, Wardlaw  JM, resonance imaging changes in normal appearing brain tissue after radi- Whittle IR. Effects of dexamethasone on cerebral perfusion and water otherapy in glioblastoma patients may confound longitudinal evaluation diffusion in patients with high-grade glioma. AJNR Am J Neuroradiol. of treatment response. Radiol Oncol. 2018;52(2):143–151. 2006;27(2):402–408. 21. Fuss  M, Wenz  F, Scholdei  R, et  al. Radiation-induced regional ce- 31. Koedel U, Pfister HW, Tomasz A. Methylprednisolone attenuates inflam- rebral blood volume (rCBV) changes in normal brain and low-grade mation, increase of brain water content and intracranial pressure, but astrocytomas: quantification and time and dose-dependent occurrence. does not influence cerebral blood flow changes in experimental pneumo- Int J Radiat Oncol Biol Phys. 2000;48(1):53–58. coccal meningitis. Brain Res. 1994;644(1):25–31. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neuro-Oncology Advances Oxford University Press

Responses in the diffusivity and vascular function of the irradiated normal brain are seen up until 18 months following SRS of brain metastases

Loading next page...
 
/lp/oxford-university-press/responses-in-the-diffusivity-and-vascular-function-of-the-irradiated-GBPr71z2vo

References (31)

Publisher
Oxford University Press
Copyright
© The Author(s) 2020. Published by Oxford University Press, the Society for Neuro-Oncology and the European Association of Neuro-Oncology.
eISSN
2632-2498
DOI
10.1093/noajnl/vdaa028
Publisher site
See Article on Publisher Site

Abstract

Background. MRI may provide insights into longitudinal responses in the diffusivity and vascular function of the irradiated normal-appearing brain following stereotactic radiosurgery (SRS) of brain metastases. Methods. Forty patients with brain metastases from non-small cell lung cancer (N = 26) and malignant melanoma (N  =  14) received SRS (15–25 Gy). Longitudinal MRI was performed pre-SRS and at 3, 6, 9, 12, and 18  months post-SRS. Measures of tissue diffusivity and vascularity were assessed by diffusion-weighted and perfusion MRI, respectively. All maps were normalized to white matter receiving less than 1 Gy. Longitudinal responses were as- sessed in normal-appearing brain, excluding tumor and edema, in the LowDose (1–10 Gy) and HighDose (>10 Gy) regions. The Eastern Cooperative Oncology Group (ECOG) performance status was recorded pre-SRS. Results. Following SRS, the diffusivity in the LowDose region increased continuously for 1 year (105.1% ± 6.2%; P  < .001), before reversing toward pre-SRS levels at 18  months. Transient reductions in microvascular cerebral blood volume (P < .05), blood flow (P < .05), and vessel densities (P < .05) were observed in LowDose at 6–9 months post-SRS. Correspondingly, vessel calibers in LowDose transiently increased at 3–9 months (P < .01). The responses in HighDose displayed similar trends as in LowDose, but with larger interpatient variations. Vascular responses followed pre-SRS ECOG status. Conclusions. Our results imply that even low doses of radiation to normal-appearing brain following cerebral SRS induce increased diffusivity and reduced vascular function for up until 18 months. In particular, the vascular responses indicate the reduced ability of the normal-appearing brain tissue to form new capillaries. Assessing the potential long-term neurologic effects of SRS on the normal-appearing brain is warranted. Key Points • SRS increased diffusivity and reduced microvascular function in normal-appearing brain tissue. • Increased diffusivity and reduced microvascular function were observed in low-dose regions. • Microvascular changes were associated with the pre-SRS ECOG status. © The Author(s) 2020. Published by Oxford University Press, the Society for Neuro-Oncology and the European Association of Neuro-Oncology. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS Importance of the Study Stereotactic radiosurgery (SRS) is a well-es- and the impact on neurologic function are tablished treatment option for cancer patients missing. To address this need, we here dem- with a limited number of brain metastases onstrate increased diffusivity and reduced (<5). Although SRS inherently provides rapid vascular function in the normal-appearing dose fall-off to surrounding normal-appearing brain parenchyma in patients with brain me- brain tissue, several preclinical studies sug- tastases treated with SRS even in low-dose gest that damage to the vascular network regions until 18  months follow-up. Our re- plays a key role in radiation-induced effects sults suggest that mapping tissue diffusivity in a normal brain. However, a complete un- and vascular function prior to treatment may derstanding of underlying radiobiological bring us closer to revealing the underlying responses in the surrounding normal brain functional mechanisms of SRS. Stereotactic radiosurgery (SRS) is a well-established first- metastases from non-small cell lung cancer and malignant line treatment option for patients with a limited number melanomas. of brain metastases (<5) and good performance status. Compared to the combined use of SRS and whole-brain radiotherapy (WBRT), or WBRT alone, SRS alone pro- vides similar survival rates, but reduced risk of harm to Material and Methods 2,3 neurocognition and quality of life. Due to continuous im- provements in image-guided identification of targets and Patients and Study Design organs at risk, as well as in radiation delivery techniques, SRS is also becoming increasingly attractive in cases of Forty patients with brain metastases from non-small cell multiple metastases. lung cancer (N  =  26) and malignant melanoma (N  =  14), Brain radionecrosis is the most common complica- participating in an ongoing observational MRI study tion after SRS, reported to occur in 24% of patients and (TREATMENT; clinicaltrials.gov identifier: NCT03458455), causing neurological deficits in 13%. Over the past dec- have been analyzed. The study has been approved by the ades, studies have identified the dosimetric burden on sur - Regional Ethical Committee and written informed consent rounding normal brain to be an important risk factor for the was obtained from all participants. To be eligible for inclu- development of radionecrosis, ie, the risk increases if the sion, patients must receive SRS to at least one untreated volume of brain exposed to doses higher than a threshold brain metastasis with the longest diameter >5 mm on a di- 4,5 dose, typically 10–12 Gy, is above 5–10 cc. However, agnostic MRI exam, henceforth referred to as the pre-SRS a complete understating of the underlying radiobiolog- MRI. Patient, tumor, and treatment characteristics are pro- ical response mechanisms governing the development of vided in Supplementary T able 1. radiation-induced toxicity and followed radionecrosis is Ten patients had previously received WBRT (N = 4), SRS missing. Moreover, the relevance of lower doses from SRS (N  =  3), or both (N  =  3) to other metastases not targeted (<8 Gy) to normal-appearing brain for potential long-term by SRS in this study. Furthermore, 5 patients received con- effects is poorly understood. comitant immunotherapy. Eighteen patients were using Several preclinical studies have recognized damage to corticosteroids (individual dosage) at the pre-SRS MRI, the vascular network to play a key role in radiation-induced while 21 patients did not (information from one patient effects on normal brain tissue. Radiation-induced vascular was not attainable). damage involves early endothelial cell injury and apop- At the time of study data lock, MRI exams were per- tosis, followed by decreased vessel densities, dilation, and formed every 3 months for the first year and at 18 months thickening of the blood vessels. Additionally, microvessel post-SRS. Dropout of patients on follow-up MRIs was thrombosis with vessel occlusion may occur within weeks due to death (N  =  7), deemed clinically unfit to perform to years after irradiation. In patients, increased vessel per- an MRI exam (N = 9), or too short follow-up time (N = 3). meability in normal-appearing brain tissue, measured by The first follow-up MRI at 3  months post-SRS was per - perfusion MRI during fractionated radiotherapy of low- formed within a margin of 1 week, while the remaining grade gliomas, has been shown to correlate with reduced MRI exams were performed within a 2-week margin. neurocognitive function. However, clinical studies of After SRS, 12 patients were treated with additional WBRT vascular responses in irradiated normal-appearing brain (N  =  4) or SRS (N  =  7) or both (N  =  1). MRI exams per- tissue are sparse, and most studies have evaluated acute formed on these patients after the additional brain radi- and early responses to fractionated radiotherapy rather otherapy were excluded from further analysis, leaving than SRS. the following number of patients at each MRI readout: To this end, our study sheds light on responses in the dif- N = 40 (pre-SRS), N = 38 (+3 months), N = 29 (+6 months), fusivity and vascular function of irradiated non-cancerous N  =  22 (+9  months), N  =  18 (+12  months), and N  =  15 normal-appearing brain tissue following SRS of brain (+18 months). Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS Stereotactic Radiosurgery Co-registration of Dose Distribution and Longitudinal MRI Data SRS was delivered using a frameless linear accelerator- based system (Varian TrueBeam vSTx, HD-MLC 120; The planning CT images and corresponding SRS dose multileaf collimator of 2.5  mm). The SRS planning was distribution were exported from the planning system and mainly performed using iPlan RT Dose (v4.5.4, Brainlab co-registered to the pre-SRS high-resolution T1-weighted AG). RayStation (v5.0, Raysearch Laboratories) was used post-contrast MRIs using normalized mutual informa- in one case. Pre-SRS 3D high-resolution T1-weighted post- tion in SPM12 (Statistical Parametric Mapping [SPM] contrast MRIs (distortion-corrected) were used for delin- toolbox version 12, University College London, England). eation of the metastases and organs at risk, while dose For each MRI exam, diffusion-weighted images, high- calculation was performed on co-registered computer to- resolution post-contrast and FLAIR images, with asso- mography (CT) images. The time from the pre-SRS MRI ciated tumor, edema, as well as binary white and gray to the planning CT scan was on average 4  days (range: matter masks, were co-registered to the DSC MRI space 5–12  days). Delineation of the gross tumor volume was by means of normalized mutual information in nordicICE performed by a radiation oncologist. A  2  mm isotropic (NordicNeuroLab AS) or SMP12. Additionally, for the pre- margin, accounting for both subclinical disease and plan- SRS MRIs, the dose distribution was co-registered to the ning uncertainties, was automatically added by the soft- DSC MRI space via the T1-weighted post-contrast MRIs. ware to generate the planning target volume. The SRS The pre-SRS DSC space was used as the reference space dose was set to cover at least 99% of the planning target for co-registration of all longitudinal MRI data. Hence, after volume and ranged from 15 to 25 Gy (single fraction) or final co-registration, the data from all follow-up MRIs and from 20.1 to 27 Gy (3 fractions) (Supplementary Table 1). SRS dose distribution were in the pre-SRS DSC space. The SRS dose and fractionation scheme were determined according to institutional guidelines, which are based on previous treatment history and clinical status of the patient Normal-Appearing Brain together with tumor size, proximity to organs at risk, and White and gray matter was identified on high-resolution normal tissue dose constraints. The mean volume of study T1-weighted post-contrast or FLAIR images by calculation metastases was significantly lower in patients treated with of probability density maps using the segmentation tool in 1 versus 3 fractions (Supplementary Figure 1). SPM12. The binary masks of white and gray matter were All patients received a corticosteroid dose (Medrol) of obtained by applying a probability threshold of >0.85 (1.0 32 mg (single fraction) or 16 mg (3 fractions) immediately being the highest probability). Normal-appearing brain after SRS and on the same night, as well as the following tissue was defined by the white and gray matter masks, day (morning and night). Those treated with 3 fractions excluding areas with pathological contrast enhancement continued to receive 2 doses every day until the night of and edema. The pathological contrast enhancement, not the last fraction. Thereafter, the use of corticosteroids was excluding central necrosis, was delineated on the high- individually managed by the patients treating physicians. resolution T1-weighted post-contrast images by 2 expe- rienced neuro-radiologists. Edema was defined on FLAIR MRI Protocol images in native or DSC MRI space. All delineations were performed in nordicICE. All MRI exams were performed using a dedicated 20-channel The co-registered dose distribution (Figure  1A) was di- head/neck coil on a 3 T Skyra (Siemens Healthineers) with vided into 3 distinct isodose levels: Reference <1 Gy, the following protocol: 3D T1-weighted images, before LowDose 1–10 Gy, and HighDose >10 Gy (Figure  1B). For and after injection of Gadolinium-based contrast agent patients who received SRS in 3 fractions (N  =  12), the (Dotarem 279.3  mg/mL, 0.2  mL/kg bodyweight; repetition linear-quadratic model with α/β for brain parenchyma = 2 time [TR]/echo time [TE] = 700 ms/12 ms; voxel size = 0.9 × Gy was used to calculate corresponding dose regions: 0.9 × 0.9 mm ; acquisition matrix = 512 × 512), T2-weighted <1.31 Gy (Reference), 1.31–16.2 Gy (LowDose), and >16.2 fluid attenuated inversion recovery (FLAIR) (TR/TE/inver - Gy (HighDose). Changes in tissue diffusivity and vascular sion time [TI] = 5000 ms/387 ms/1800 ms; voxel size = 0.9 × function were assessed in normal-appearing brain tissue 0.9  × 0.9  mm ; acquisition matrix  =  512  × 512; field of in the LowDose and HighDose regions (Figure  1C). The view  =  460.8  × 460.8  mm ), diffusion-weighted imaging volume of normal-appearing brain in the LowDose and (TR/TE  =  5960  ms/71  ms; b-values  =  0 and 1000/1500  s/ HighDose regions at all follow-up MRIs are provided in 2 3 mm ; voxel size = 1.22 × 1.22 × 4.0 mm ; slice gap = 5.0 mm; Supplementary T able 2. acquisition matrix  =  180  × 180; field of view  =  219.6  × 219.6 mm ), and dynamic susceptibility contrast (DSC) MRI with combined gradient-echo and spin-echo acquisitions Quantification of Tissue Diffusivity and Vascular (TR = 1500 ms; TE [gradient-echo] = 13 ms [malignant mel- Function anoma]/15–30  ms [non-small cell lung cancer]; TE [spin- The tissue diffusivity was assessed from ADC maps created echo]  =  104  ms; voxel size  =  2.0–2.2  × 2.0–2.2  × 5.0  mm ; directly on the MRI scanner from the diffusion-weighted slice gap  =  6.5  mm; acquisition matrix  =  120  × 90; field images, using Stejskal–Tanner diffusion approximation. of view  =  240–264  × 180–198  mm ) with a bolus injection Vascular function was assessed by analysis of the per- (3  mL/s) of contrast agent, followed by 30  mL of physio- fusion MRI as previously described. In short, standard logic saline solution. Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS A Original dose distribution B Dose regions C Regions-of-interest Normal brain tissue LowDoseHighDose White matter White matter Gray matter Gray matter 0Gy 25Gy Reference Lowdose Highdose Edema Tumor <1Gy 1–10Gy >10Gy Figure 1 Regions-of-interests used for assessment of normal-appearing brain tissue responses to SRS. (A) A  representative co-registered stereotactic dose distribution for a patient with brain metastasis from non-small cell lung cancer as an overlay of a T1-weighted post-contrast image acquired pre-SRS. The prescribed SRS dose was 25 Gy. (B) The dose distribution was divided into 3 dose regions: Reference: <1 Gy (blue overlay), LowDose: 1–10 Gy (green overlay), and HighDose: >10 Gy (yellow overlay). (C) The final regions-of-interest used for longitudinal response assessments included normal-appearing brain tissue, ie, white and gray matter in LowDose (white matter: light green overlay, gray matter: dark green overlay) and HighDose (white matter: light yellow overlay, gray matter: dark yellow overlay), excluding edema (purple overlay) and tumor (red overlay). voxel-wise DSC MRI kinetic analyses of the spin- and densities, could not be calculated for these patients at the gradient-echo acquisitions were performed in nordicICE, given MRI exams. providing parametric maps of cerebral blood volume (CBV) and cerebral blood flow (CBF). Whereas the spin-echo maps ECOG Performance Status and Eloquent Regions reflect the micro-vasculature, the gradient-echo maps rep- resent the total micro-to-macroscopic vasculature, and Pre-SRS, the Eastern Cooperative Oncology Group the prefixes “Micro” and “Macro” are henceforth used for (ECOG) performance status was recorded for 31 of the vascular metrics obtained from spin-echo and gradient- patients as follows: 0 (N = 13), 1 (N = 16), and 2–3 (N = 2). echo, respectively. The DSC MRI analysis included motion Potential differences in normal-appearing tissue re- correction, automatic detection of the arterial input func- sponses were assessed in patients with pre-SRS ECOG tion with deconvolution by standard single value decom- status 0 versus >0. Due to missing spin-echo data, the position, and contrast agent leakage-correction adapted final number of patients assessed were (ECOG  =  0/>0) for both T1- and T2-shortening effects. Vessel caliber anal- N  =  12/17 (pre-SRS), N  =  11/15 (+3  months), N  =  9/10 ysis was performed in Matlab (v.R2017a, MathWorks Inc.), (+6 months), N = 7/8 (+9 months), N = 6/4 (+12 months), providing estimations of mean vessel calibers and mean and N = 6/4 (+18 months). vessel densities. Binary masks of all the study metastases, as well as any From all MRI exams, normalized parametric maps additional metastases also treated with SRS at the same were calculated by dividing all image voxel values to time, were co-registered to the MNI space for assessment the respective median value of white matter within the of their location related to eloquent regions determined Reference region (Figure  1B), henceforth prefixed n. For from binary masks provided by the SPM12. each normalized parametric map, the mean value, ex- cluding outliers >3 SD away from the mean, of all regions- of-interests were computed if >8 non-zero voxels were Statistical Analysis present. Longitudinal changes were assessed relative to pre-SRS in percent (%), mean ± SD. Mann–Whitney U test or Fisher’s exact test was used to Spin-echo readout data were missing for the following compare groups of continuous and dichotomized data, number of patients: N  =  2 (pre-SRS), N  =  1 (+3  months), respectively. Comparisons between absolute and relative N = 2 (+6 months), N = 1 (+9 months), N = 2 (+12 months), changes (to pre-SRS) in the regions-of-interest were made and N  =  1 (+18  months). Thus, nMicro-CBV and nMicro- using the Wilcoxon signed-rank test. The significance level CBF, as well as nMean vessel calibers and nMean vessel was 5%, including Holm-Bonferroni correction in the case Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS of multiple comparisons. The statistical analyses were per- The HighDose region of patients set to receive 3 frac- formed using Matlab or IBM SPSS Statistics (v25). tions to a minimum of 1 metastasis versus a single fraction showed lower vascular metrics. Specifically, nMicro-CBV was 1.22  ± 0.20 versus 1.44  ± 0.21 (P < .01), nMicro-CBF was 1.23 ± 0.20 versus 1.48 ± 0.28 (P < .05), and Results nMean vessel densities was 0.83 ± 0.19 versus 0.98 ± 0.12 (P < .05). Furthermore, the LowDose region showed lower Diffusivity and Vascular Profiles of the nMacro-CBV in patients having received the previous ra- Pre-irradiated Brain diotherapy to the brain compared with those who had not Compared to the LowDose and HighDose regions, the me- (P < 0.05) (Supplementary Table 5). No differences in vas- tastases and associated edema displayed pathologic dif- cular function in the LowDose and HighDose regions were fusivity and vascular function pre-SRS (Figure  2). In the observed between patients treated with or without cor- tumor, the nADC (P < .001), nMacro-CBV (P < .01), nMacro- ticosteroids (Supplementary Table 6). However, slightly CBF (P < .05), and nMean vessel calibers (P < .001) were higher nADC was observed in the LowDose region in pa- elevated, while nMicro-CBF (P < .01) and nMean vessel tients treated with corticosteroids (1.14 ± 0.05) compared densities (P < .001) were lower. Associated edema showed to untreated patients (1.11 ± 0.05) (P < .05). Moreover, no higher nADC (P < .001), nMicro-CBV, nMicro-CBF, nMacro- differences in diffusivity or vascular function were ob- CBV, and nMacro-CBF (P < .001) and lower nMean vessel served between patients with different primary diagnosis densities (P < .001) compared to normal-appearing brain (Supplementary Table 7), or between patients having re- tissue (Supplementary T able 3). ceived previous immunotherapy or not (Supplementary In the LowDose and HighDose regions, the gray matter Table 8), or between patients with pre-SRS ECOG status was characterized by higher diffusivity and increased 0 versus >0 (Supplementary Table 9). No differences in vascular function compared to white matter, with mean the absolute or relative change in tumor or edema vol- gray-to-withe matter ratios of 1.2 for nADC, and ranging umes were observed depending on the pre-SRS ECOG from 1.1 to 1.8 for the vascular metrics (Supplementary status at any of the MRI examinations. However, pa- Table 4). tients with EOCG status >0 more frequently presented A C nMicro-CBV E nMacro-CBV G nMean vessel T1-weighted post- calibers contrast MRI Edema Tumor BD nADC nMicro-CBF FH nMacro-CBF nMean vessel densities Figure 2 Normalized parametric maps of diffusivity and vascular function pre-SRS. (A) T1-weighted post-contrast image in DSC space with a brain metastasis from non-small cell lung cancer (red overlay) and associated edema (purple overlay), and normalized (n) parametric maps; (B) the apparent diffusion coefficient (nADC), (C) nMicro-vascular cerebral blood volume (CBV), (D) nMicro-vascular cerebral blood flow (CBF), (E) nMacro-CBV, (F) nMacro-CBF, (G) nMean vessel calibers, and (H) nMean vessel densities. Low High Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS with metastases in eloquent regions at pre-SRS (P < .05) (P < .001), returning to 102.5% ± 4.1% at 18  months post- (Supplementary T able 10). SRS. Similarly, an increasing trend was observed for nADC in HighDose, also peaking 1  year post-SRS at 109.6% ± 18.8%, but with larger interpatient variations. The vascular responses in the LowDose region showed Increased Diffusivity and Reduced Vascular transient reduction in nMicro-CBV (Figure 3Bi) and nMicro- Function of Normal Irradiated Brain CBF (Figure 3Bii), combined with stable levels of nMacro- Following SRS, the LowDose region showed increased dif- CBV (Figure  3Diii) and nMacro-CBF (Figure  3Biv), as well fusivity (Figure 3Ai) combined with reduced vascular func- as transient increase in nMean vessel calibers (Figure 3Bv) tion for up until 18 months (Figure  3Bi–vi). The increase in and reduction in nMean vessel densities (Figure  3Bvi). nADC in LowDose peaked 1 year post-SRS at 105.1% ± 6.1% Eighteen months post-SRS, the vascular function in the A Diffusivity C Longitudinal nMicro-CBV Pre-SRS +3 months +6 months nADC N = 40 38 29 22/21 18 15 LowDose T1-weighted post-contrast HighDose 105 * 5.5 LowDose HighDose nMicro-CBV MRI (Months) P < 0.05 Relative to pre-SRS 0.3 # P < 0.01 Relative to pre-SRS * P < 0.001 Relative to pre-SRS B Vascular function i iii v nMicro-CBV nMacro-CBV nMean vessel calibers 115 115 125 N = 38 35 25 20/19 15 14 N = 40 36 27 21/20 17 15 N = 38 35 24 20/19 15 14 110 110 LowDose LowDose HighDose HighDose 105 105 100 100 95 95 LowDose HighDose 90 90 90 MRI (Months) MRI (Months) MRI (Months) ii iv vi nMicro-CBF nMacro-CBF nMean vessel densities 115 115 125 N = 38 35 25 20/19 15 14 N = 40 36 27 21/20 17 15 N = 38 35 25 20/19 15 14 110 110 LowDose LowDose HighDose HighDose 105 105 100 100 95 95 LowDose HighDose 90 90 90 MRI (Months) MRI (Months) MRI (Months) Figure 3 Increased diffusivity and decreased vascular function in normal-appearing brain up until 18  months post-SRS. Mean ± SEM of normalized (n) metrics relative to pre-SRS (%) in normal-appearing brain tissue in LowDose (green) and HighDose (yellow) on a logarithmic scale: (Ai) the apparent diffusion coefficient (nADC), (Bi) nMicro-vascular blood volume (CBV), (Bii) nMicro-vascular blood flow (CBF), (Biii) nMacro- CBV, (Biv) nMacro-CBF, (Bv) nMean vessel calibers, and (Bvi) nMean vessel densities. Transient changes in diffusivity and vascular function are highlighted by shaded green areas. The number of patients (N) included at each time point is shown in gray. At 9 months post-SRS, analysis of the HighDose region was excluded for one patient due to insufficient number of voxels in this region (total number of patients analyzed is shown in yellow). (C) Reduced nMicro-CBV in LowDose at 3 and 6 months post-SRS compared to pre-SRS is shown for a patient with a brain metas- tasis from malignant melanoma (bottom row), with corresponding T1-weighted post-contrast images indicating LowDose (green) and HighDose (yellow) regions (top row). P values from the Wilcoxon Singed-Rank test relative to pre-SRS. Pre-SRS Pre-SRS Pre-SRS Pre-SRS Pre-SRS Pre-SRS Pre-SRS Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS LowDose region had returned to pre-SRS levels. In the CBV and CBF, as well as reduced vessel densities. While no HighDose region, similar trends as in LowDose were ob- apparent changes to SRS in normal-appearing tissue were served for nMicro-CBV and nMicro-CBF, as well as nMean observed on conventional MRI, our findings indicate that vessel calibers. However, nMacro-CBV and nMacro-CBF the micro-vasculature is particularly vulnerable, even to and nMean vessel densities showed a continued increasing low-dose irradiation. Combined with the observed stable trend from 6, 9, and 3 months post-SRS, respectively. levels of total CBV and CBF, as well as increases in vessel No apparent dose-dependency or plateau dose of the calibers, our results further imply that the vasculature of responses in diffusivity or vascularity was observed the irradiated normal brain loses radio-sensitive, yet well- (Supplementary Figure 2). Moreover, the responses in the functioning and highly differentiated small capillaries. LowDose region for nMicro-CBV and nMicro-CBF, as well Instead, the post-radiated tissue is dominated by larger as nMean vessel calibers, were different for patients with vessels with potentially reduced function. pre-SRS ECOG 0 versus >0 (Figure  4). Specifically, pa- The radiation-induced effects on endothelial vascular in- tients with ECOG status >0 showed transient reductions jury in both tumors and normal tissues are known to mod- in nMicro-CBV to 93.6% ± 8.9% (P < .05) at 6 months and ulate angiogenesis and neovascularization. Our findings to 94.1% ± 4.8% at 9  months (P < .01; Figure  4A). The re- in humans are in line with preclinical studies showing that duction at 9  months was significantly lower than for pa- doses as low as 2 Gy may induce significant reductions of tients with pre-SRS ECOG of 0 (P < .01), showing a small small capillaries (diameter ≤10 µm). Moreover, endothelial increase of 101.6% ± 7.3%. Likewise, the transient reduc- apoptosis caused by irradiation has been shown to induce tions in nMicro-CBF at 6–9  months (P < .01) were lower increased vessel dilation and vessel permeability and thus for patients with pre-SRS ECOG >0 versus 0 at 9  months reduced vascular function. Interestingly, in our study we (P < .01) and 1 year post-SRS (P < .05; Figure 4B). Finally, at observed consistent increase in mean vessel densities 6 months post-SRS, nMean vessel calibers was increased from 6 months post-SRS in normal-appearing brain tissue to 109.6% ± 8.8% (P < .01) in the pre-SRS ECOG >0 patients regions having received high doses (>10 Gy) compared compared to stable levels of 100.4% ± 6.2% for the patients to low doses (1–10 Gy). Approximately 90% of endothe- with ECOG status 0 (P < .05; Figure 4C). lial cells receiving conventional fractionated radiotherapy experience mitotic cell death without apoptosis, whereas higher doses also induce apoptotic cell death. Such chronic effects are commonly reflected in upregulated en- dothelial cell senescence in the cerebral vascular system. Discussion The observed abnormal vascular response, and especially In our study of 40 patients with brain metastases from in the HighDose region, could therefore suggest abnormal non-small cell lung cancer and malignant melanoma, we revascularization from capillary rarefaction, increased vas- show by MRI that normal-appearing brain tissue having re- cular permeability, and impaired vascular homeostasis. ceived low doses (1–10 Gy) from SRS displays increased Histopathology of patient specimens of various ce- diffusivity and decreased vascular function for up until rebral neoplasms with radiation-induced injury shows 18 months post-SRS. The decreased vascular function was that damaged tissue, in addition to vascular changes, is particularly expressed by reduced levels of microvascular characterized by macrophage invasion, demyelination, Longitudinal Vascular responses in LowDose by pre-SRS ECOG status nMicro-CBV nMicro-CBF nMean vessel calibers AB C 115 115 130 Pre-SRS ECOG = 0 Pre-SRS ECOG = 0 N = 12, 11, 9, 7, 6, 6 N = 12, 11, 9, 7, 6, 6 110 110 Pre-SRS ECOG > 0 Pre-SRS ECOG > 0 N = 17, 15, 10, 8, 4, 4 N = 17, 15, 10, 8, 4, 4 # 105 105 100 100 < 0.05 < 0.05 < 0.01 < 0.01 95 95 Pre-SRS ECOG = 0 N = 12, 11, 9, 7, 6, 6 90 90 90 Pre-SRS ECOG > 0 N = 17, 15, 10, 8, 4, 4 85 85 MRI (Months) MRI (Months) MRI (Months) P < 0.05 Relative to pre-SRS; # P < 0.01 Relative to pre-SRS; P value between groups Figure 4 Reduced vascular function in normal-appearing brain is more pronounced in a patient with pre-SRS ECOG status >0 compared to 0.  Mean ± SEM of normalized (n) metrics relative to pre-SRS (%) in normal-appearing brain tissue in LowDose on a logarithmic scale for pa- tients with pre-SRS Eastern Cooperation Oncology Group (ECOG) status of 0 (gray) and >0 (brown). (A) Transiently reduced nMicro-vascular blood volume (CBV) and (B) nMicro-vascular blood flow (CBF) in patients with pre-SRS ECOG status >0 compared to increased nMicro-CBV at 9–12 months post-SRS in patients with pre-SRS ECOG status 0. (C) Transient increase in nMean vessel calibers was higher in patients with pre-SRS ECOG status >0 compared to 0. The number of patients (N) with available data in each group is shown in the legend box. P values from Mann–Whitney U test (between groups) and the Wilcoxon Signed-Rank test (relative to pre-SRS). Pre-SRS Pre-SRS Pre-SRS Relative to pre-SRS (%) Relative to pre-SRS (%) Relative to pre-SRS (%) Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS and reactive gliosis. In particular, the presence of de- In line with our findings, reduced CBV and CBF have myelination and reactive gliosis both cause higher tissue been observed in normal-appearing brain tissue up until diffusivity measured by diffusion-weighted MRI and 12  months after SRS to arteriovenous malfunctions 14,15 diffusion tensor imaging (DTI). The increased diffu- and conventional radiotherapy of low- and high-grade 20–24 sivity of normal-appearing tissue observed in our study gliomas. In the latter studies, a dose-dependent re- may be due to microstructural loss caused by such pro- duction of vascular function was observed, with larger cesses. Recent DTI studies have demonstrated increased reductions in dose regions receiving more than 30% of diffusivity of normal-appearing white matter after frac- the prescribed dose (30–66 Gy). Increased vascular re- tionated radiotherapy of primary brain tumors, attrib- sponses in normal-appearing tissue during conventional uted to demyelination, and correlated with neurological radiotherapy of gliomas and early after SRS to brain me- 16,17 18 deficits. Specifically, in one of these studies including tastases have also been reported. While a weak positive 54 adult patients, increased diffusivity in white matter correlation between dose and increased vascular met- beneath the cingulate cortex at 3 and 6  months after ra- rics, with a plateau effect at 10 Gy, was observed after diotherapy was associated with a decline in verbal set- SRS, the strongest dose-dependency was found during shifting ability and cognitive flexibility. In light of these the fractionated radiotherapy of the gliomas and was ac- findings, our results reinforce the potential future clinical tually diminished 6  months post-radiotherapy. In light of value of using advanced MRI to identify dose tolerances to our findings, showing no apparent linear correlation be- subclinical microstructural changes that can be integrated tween responses in diffusivity or vascular function, as- into the treatment planning and subsequently reduce sessed 3–18 post-SRS, the dose-dependency may be more neurocognitive decline. pronounced during and early after irradiation—further The responses in diffusivity and vascular function of indicating a dose-independent ability to restore vascular normal-appearing brain tissue in HighDose (>10 Gy), function. Altogether, the present studies having inves- mainly representing peri-tumoral tissue, showed similar tigated advanced MRI for normal brain tissue response 17–24 trends as in LowDose up until 1 year post-SRS. However, assessment, including ours, demonstrate the poten- larger interpatient variations were observed. This may re- tial for MRI metrics to identify normal brain tissue toler- flect variability in the pre-SRS diffusivity and vascular ances that are essential for optimizing the treatment plan levels of the peri-tumoral region and consequently a dif- accordingly. ferent premise to handle high-dose radiation. Poor vas- Our study observed distinctive longitudinal responses cular function in the peri-tumoral region pre-SRS has been for patients having a pre-SRS ECOG status 0 compared shown to be associated with radiation-induced changes. >0. Though no apparent differences in diffusivity or vas- Moreover, while at 18 months post-SRS the normal brain cular function were detectable by the diffusion or per- tissue diffusivity and vascular function in LowDose ap- fusion MRI pre-SRS, underlying subtle differences may parently returned to pre-SRS levels, all the vascular met- have been present. The vasculature in normal brain tissue rics showed an increasing trend starting at 6–12  months of patients with ECOG status >0 may already have been post-SRS in HighDose. The increase was particularly ap- impaired and becoming reinforced by the radiation dose. parent for the nMacro-CBV and CBF, as well as mean vessel While the association between cerebro-microvascular densities, suggesting a vascular network dominated by dysfunction and early neurologic pathogenesis is in- 26,27 larger vessels and reduced capability of recruiting smaller triguing, it is well recognized that CBF levels and cog- capillaries. This deterioration of the vascular network may nitive impairment are inversely correlated. In our study, be a contributing factor of the underlying mechanisms in- changes in ECOG performance status could not be cor- volved in the development of radionecrosis—which has related to responses in diffusivity or vascular function as an increasing risk of occurring with increasing volume of the ECOG status was not routinely recorded at the clinical brain exposed to high doses, in particular, 10 Gy and 12 follow-ups post-SRS. This is however warranted in future 4 5 Gy. However, as stressed by Milano et  al. the develop- studies. ment of radionecrosis is likely impacted by multiple fac- Our study has limitations. Outside of the study period, tors, including possible regional variations in susceptibility individual treatment management resulted in large var- to radiation injury. From our data, showing more homoge- iations with respect to previous and salvage radiation nous tissue responses in LowDose compared to HighDose, therapy to the brain, prescription dose and fractionations we hypothesize that these changes also play a part in the of the study SRS, use of immunotherapy, systemic treat- development of radionecrosis—further supported by an ment, and corticosteroid treatment. Combined with the occasional manifestation of radionecrosis outside the longitudinal reduction of patients, multivariate analyses high-dose field. were thus not feasible. However, a separate analysis of Although radiation exposure to normal brain tissue is the different treatment groups showed that excluding reduced with SRS compared to WBRT, large volumes of patients previously treated with brain radiotherapy did normal tissue still receive low doses. Our results thus not alter the outcome of the longitudinal response anal- support future studies on effects of low-dose irradiation ysis. In our study and within the power of our sample for development of radionecrosis, neurologic deficits, and size, the patients treated with 3 fractions versus the pa- second malignancies. Such potential long-term effects tients treated with a single fraction did not show any may be especially relevant with increased use of SRS to apparent differences in the longitudinal changes in dif- multiple and/or larger metastases, and with increasing fusivity or vascular function. We aimed to reduce the po- treatment efficacy that potentially can result in longer sur - tential impact of different fractionation regimes on our vival times for these patients. findings by calculating equivalent doses (EQD2) for the Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Neuro-Oncology Advances Nilsen et al. Normal brain tissue responses to cerebral SRS dose distributions delivered in 3 fractions. The validity of the linear-quadratic model is less established for higher Acknowledgments doses (>8–10 Gy), but the LowDose region is mainly dom- inated by doses <10 Gy and should therefore be less in- We thank Dr. Knut Håkon Hole at Oslo University Hospital for a fluenced by this restriction. The lower pre-SRS perfusion constructive review of the manuscript. levels in what later became the HighDose region in pa- tients receiving 3 rather than a single SRS fraction may well be due to differences in tumor size. Larger tumors, more commonly treated with fractionated SRS, may in- Conflict of interest statement. Intellectual property right, herently be more prone to reduced vascular function NordicNeuroLab AS, Bergen, Norway (K.E.E). in the peri-tumoral region. Thus, although receiving less than 10 Gy per fraction, which has been shown to be a threshold dose for inducing apoptotic endothelial cells death, the impaired pre-SRS vasculature may re- Authorship Statement. Experimental design: L.B.N, C.S., K.D.J., duce the level of this threshold dose. Furthermore, no Å.H., and K.E.E. Implementation: L.B.N, I.D, E.G., C.S., O.G., differences in pre-SRS diffusivity and vascular function K.D.J., Å.H., and K.E.E. Analysis and/or interpretation of the data: of normal brain tissue or longitudinal responses were L.B.N., I.D., E.G., C.S., A.L., D.O.S., and K.E.E. Writing of the first present between patients having received immuno- manuscript draft: L.B.N., I.D., E.G., C.S., A.L., D.O.S., and K.E.E. therapy or not. However, preliminary analysis of the cur- Editing: all coauthors. Read and approved the final version: all rent study population does imply that concomitant SRS coauthors. Unpublished material: no unpublished material is and immunotherapy render peri-tumoral regions more referenced. prone to changes than SRS alone, indicating dose- dependent synergetic effect immunotherapy to SRS. Finally, corticosteroid treatment has been suggested to decrease vascular functions, but conflicting results are 30,31 reported in the literature. Due to lack of information References about corticosteroid use on the follow-up MRIs, this could not be account for in our data. However, at pre-SRS no differences in any vascular metrics were observed be- 1. Hartgerink  D, van  der  Heijden  B, De  Ruysscher  D, et  al. Stereotactic tween patients using corticosteroids and not. radiosurgery in the management of patients with brain metastases of Our longitudinal MRI study indicates that normal brain non-small cell lung cancer: indications, decision tools and future direc- tissue is sensitive to low doses of irradiation following tions. Front Oncol. 2018;8(9):154. SRS, resulting in loss of, and potentially, reduced ability to 2. Chang  EL, Wefel  JS, Hess  KR, et  al. Neurocognition in patients with form well-functioning capillaries. The long-term implica- brain metastases treated with radiosurgery or radiosurgery plus tions of our findings in terms of neurological function and whole-brain irradiation: a randomised controlled trial. Lancet Oncol. clinical outcomes are thus warranted in future studies. 2009;10(11):1037–1044. 3. Schimmel WCM, Gehring K, Eekers DBP, Hanssens PEJ, Sitskoorn MM. Cognitive effects of stereotactic radiosurgery in adult patients with brain metastases: a systematic review. Adv Radiat Oncol. 2018;3(4):568–581. Supplementary Data 4. Minniti G, Clarke E, Lanzetta G, et al. Stereotactic radiosurgery for brain Supplementary data are available at Neuro-Oncology metastases: analysis of outcome and risk of brain radionecrosis. Radiat Advances online. Oncol. 2011;6(15):48. 5. Milano  MT, Usuki  KY, Walter  KA, Clark  D, Schell  MC. Stereotactic radiosurgery and hypofractionated stereotactic radiotherapy: normal tissue dose constraints of the central nervous system. Cancer Treat Rev. Keywords 2011;37(7):567–578. 6. Sundgren  PC, Cao  Y. Brain irradiation: effects on normal brain brain metastases | diffusion-weighted MRI | normal- parenchyma and radiation injury. Neuroimaging Clin N Am. appearing brain tissue response | perfusion MRI | stereo- 2009;19(4):657–668. tactic radiosurgery 7. Belka  C, Budach  W, Kortmann  RD, Bamberg  M. Radiation induced CNS toxicity—molecular and cellular mechanisms. Br J Cancer. 2001;85(9):1233–1239. 8. Cao Y, Tsien CI, Sundgren PC, et al. Dynamic contrast-enhanced magnetic resonance imaging as a biomarker for prediction of radiation-induced Funding neurocognitive dysfunction. Clin Cancer Res. 2009;15(5):1747–1754. This work was supported by the South-Eastern Norway 9. Brenner DJ. The linear-quadratic model is an appropriate methodology Regional Health Authority (2016102, 2013069); the Norwegian for determining isoeffective doses at large doses per fraction. Semin Cancer Society (6817564); the European Research Council under Radiat Oncol. 2008;18(4):234–239. the European Union’s Horizon 2020 (758657); and the Research 10. Digernes I, Grøvik E, Nilsen LB, et al. Brain metastases with poor vas- Council of Norway (261984). cular function are susceptible to pseudoprogression after stereotactic radiation surgery. Adv Radiat Oncol. 2018;3(4):559–567. Downloaded from https://academic.oup.com/noa/article/2/1/vdaa028/5764161 by guest on 26 August 2020 Nilsen et al. Normal brain tissue responses to cerebral SRS 11. Emblem  KE, Farrar  CT, Gerstner  ER, et  al. Vessel caliber—a potential 22. Lee  MC, Cha  S, Chang  SM, Nelson  SJ. Dynamic susceptibility con- MRI biomarker of tumour response in clinical trials. Nat Rev Clin Oncol. trast perfusion imaging of radiation effects in normal-appearing brain 2014;11(10):566–584. tissue: changes in the first-pass and recirculation phases. J Magn Reson 12. Venkatesulu BP, Mahadevan LS, Aliru ML, et al. Radiation-induced en- Imaging. 2005;21(6):683–693. dothelial vascular injury: a review of possible mechanisms. JACC Basic 23. Petr  J, Platzek  I, Seidlitz  A, et  al. Early and late effects of Transl Sci. 2018;3(4):563–572. radiochemotherapy on cerebral blood flow in glioblastoma pa- 13. Sundgren  PC, Fan  X, Weybright  P, et  al. Differentiation of recurrent tients measured with non-invasive perfusion MRI. Radiother Oncol. brain tumor versus radiation injury using diffusion tensor imaging in 2016;118(1):24–28. patients with new contrast-enhancing lesions. Magn Reson Imaging. 24. Price SJ, Jena R, Green HA, et al. Early radiotherapy dose response and 2006;24(9):1131–1142. lack of hypersensitivity effect in normal brain tissue: a sequential dy- 14. Hagen T, Ahlhelm F, Reiche W. Apparent diffusion coefficient in vasogenic namic susceptibility imaging study of cerebral perfusion. Clin Oncol (R edema and reactive astrogliosis. Neuroradiology. 2007;49(11):921–926. Coll Radiol). 2007;19(8):577–587. 15. Natarajan R, Hagman S, Wu X, et al. Diffusion tensor imaging in NAWM 25. Balagamwala  EH, Chao  ST, Suh  JH. Principles of radiobiology of ster- and NADGM in MS and CIS: association with candidate biomarkers in eotactic radiosurgery and clinical applications in the central nervous Sera. Mult Scler Int. 2013;2013:265259. system. Technol Cancer Res Treat. 2012;11(1):3–13. 16. Chapman  CH, Nagesh  V, Sundgren  PC, et  al. Diffusion tensor im- 26. Toth  P, Tarantini  S, Csiszar  A, Ungvari  Z. Functional vascular contribu- aging of normal-appearing white matter as biomarker for radiation- tions to cognitive impairment and dementia: mechanisms and conse- induced late delayed cognitive decline. Int J Radiat Oncol Biol Phys. quences of cerebral autoregulatory dysfunction, endothelial impairment, 2012;82(5):2033–2040. and neurovascular uncoupling in aging. Am J Physiol Heart Circ Physiol. 17. Tringale KR, Nguyen T, Bahrami N, et al. Identifying early diffusion imaging 2017;312(1):H1–H20. biomarkers of regional white matter injury as indicators of executive func- 27. Østergaard  L, Aamand  R, Gutiérrez-Jiménez  E, et  al. The capillary tion decline following brain radiotherapy: a prospective clinical trial in pri- dysfunction hypothesis of Alzheimer’s disease. Neurobiol Aging. mary brain tumor patients. Radiother Oncol. 2019;132(March):27–33. 2013;34(4):1018–1031. 18. Jakubovic R, Sahgal A, Ruschin M, Pejović-Milić A, Milwid R, Aviv RI. 28. Alosco ML, Gunstad J, Jerskey BA, et al. The adverse effects of reduced Non tumor perfusion changes following stereotactic radiosurgery to cerebral perfusion on cognition and brain structure in older adults with brain metastases. Technol Cancer Res Treat. 2015;14(4):497–503. cardiovascular disease. Brain Behav. 2013;3(6):626–636. 19. Taki  S, Higashi  K, Oguchi  M, et  al. Changes in regional cerebral 29. Nilsen  LB, Groevik  E, Digernes  I, et  al. Vascular responses in normal blood flow in irradiated regions and normal brain after stereotactic brain tissue after combined immunotherapy and SRS to brain metas- radiosurgery. Ann Nucl Med. 2002;16(4):273–277. tases. Radiother Oncol. 2019;133(Suppl 1):S551–S552. 20. Fahlström  M, Blomquist  E, Nyholm  T, Larsson  EM. Perfusion magnetic 30. Bastin  ME, Carpenter  TK, Armitage  PA, Sinha  S, Wardlaw  JM, resonance imaging changes in normal appearing brain tissue after radi- Whittle IR. Effects of dexamethasone on cerebral perfusion and water otherapy in glioblastoma patients may confound longitudinal evaluation diffusion in patients with high-grade glioma. AJNR Am J Neuroradiol. of treatment response. Radiol Oncol. 2018;52(2):143–151. 2006;27(2):402–408. 21. Fuss  M, Wenz  F, Scholdei  R, et  al. Radiation-induced regional ce- 31. Koedel U, Pfister HW, Tomasz A. Methylprednisolone attenuates inflam- rebral blood volume (rCBV) changes in normal brain and low-grade mation, increase of brain water content and intracranial pressure, but astrocytomas: quantification and time and dose-dependent occurrence. does not influence cerebral blood flow changes in experimental pneumo- Int J Radiat Oncol Biol Phys. 2000;48(1):53–58. coccal meningitis. Brain Res. 1994;644(1):25–31.

Journal

Neuro-Oncology AdvancesOxford University Press

Published: Jan 1, 2020

There are no references for this article.