Demographics
The study included 170 patients [female: 100 (58.8%)] with a median age of 37.3 (IQR: 27.5) years at diagnosis. One patient (0.6%) was diagnosed with a CCM2 mutation. The initial presentation was hemorrhage in 132 patients (77.6%), FND in 28 (16.5%), and seizure in 3 (1.8%). Seven patients (4.1%) were asymptomatic at the time of diagnosis.
Prior to SRS, 85 (50.0%) patients had a cranial nerve deficit, 50 (29.4%) a motor deficit, and 49 (28.8%) a sensory deficit. Table 1
Among the 170 patients, three were treated for two brainstem CCM, one patient for three CCM, and one patient for 5 CCM. Overall, 181 CCM were treated by SRS. The median time from diagnosis to SRS was 0.9 years (IQR: 4.3). Prior surgical resection occurred in 13 (7.2%) of CCM with a median delay from surgical resection to SRS of 3.2 years (IQR:5.6). A DVA was associated in 26 (14.4%) of the CCM and the CCM was adjacent to an ependymal surface in 42 (23.2%) cases. Table 2
The CCM were treated with a median margin dose of 12.0Gy (IQR:2.0) at a 50% (IQR:10) isodose line for a median volume of 0.4 cm3 (IQR:0.9). Table 2
Prior to SRS, 26 (14.4%) CCM never experiences a defined acute and/or subacute hemorrhage, 103 (56.9%) CCM bled one time, 37 (20.4%) CCM bled two times, 12 (6.6%) CCM bled three times, 1 (0.6%) four times, and 2 (1.1%) five times.
Hemorrhage analysis
Among CCMs with hemorrhage prior to SRS (n=155), the pre-SRS AHR varied based on the methodology used; 3.6 per 100 CCM-years (follow-up from birth: 6309.9 years; 227 hemorrhages), 42.0 per 100 CCM-years (follow-up from diagnosis: 540.5 years; 227 hemorrhages), and 14.8 per 100 CCM-years (follow-up from diagnosis: 540.5 years; 80 hemorrhages after the exclusion of hemorrhages leading to CCM diagnosis).
The overall post-SRS AHR was 2.3 per 100 CCM-years (follow-up post SRS: 736.8 years; 17 hemorrhages) and specificallyin the first 2 years 2.8 per 100 CCM-years (follow-up post SRS: 288.3years; 9 hemorrhages) and after the two first years 1.8 per 100 CCM-years (follow-up post SRS: 449.1 years; 8 hemorrhages). There was a statistically significant reduction in hemorrhage rate post-SRS (-6.4 per 100 CCM-years, 95% CI:3.87-11.13, p<0.0001). Using recurrent analysis, SRS was associated with a reduction of hemorrhage risk (HR= 0.17, 95% CI= 0.09-0.34, p<0.001) in multivariate analysis. Table 3
After SRS, among all the CCM (n=181), new hemorrhages occurred in 19 CCM (10.5%) with 18 (9.9%) CCMs having bled one time and one (0.6%) CCM twice. Three CCM (3/26, 11.5%) without hemorrhage prior SRS, bled after SRS. The median delay from SRS to the first bleeding post-SRS was 3.2 (IQR:6.1) years. In one case, the DVA associated with the CCM was occluded and was considered the reason for the bleeding.
Using univariate Cox-analysis, the probability of a new hemorrhages after SRS was reduced for patient older than 35 years old (HR= 0.21, 95% CI= 0.08-0.56, p=0.002) and increased with margin doses >13Gy (HR= 2.57, 95% CI=1.03-6.45, p=0.044. The number of hemorrhages prior to SRS was not associated with a risk modification of new bleeding after SRS. Table 4
The probability of new hemorrhage after SRS was 4% (95% CI= 0.8-7.2), 11.2% (95% CI= 4.6-17.3) and 20.8% (95% CI= 7.7-32.0) at 2, 5, and 10 years, respectively. Figure 1
Imaging outcomes
The median follow-up imaging was 3.4 years (IQR=5.1) after SRS. At last follow-up, the CCM volume was decreased in 77/179 (43.0%), stable in 99/179 (55.3%) and increased in 3/179 (1.7%).
The pre-SRS Zabramski score was 1 in 49/178 (27.5%), 2 in 108/178 (60.7%), 3 in 18/178 (10.1%) and 4 in 3/ 178 (1.7%). After SRS, the Zabramski score was unchanged in 100/146 (68.5%) CCM. Table 5
Adverse Radiation effects
AREs occurred in 9 CCM and 9 patients (5.3%). ARE was symptomatic in four cases (2.2%). One (0.6%) patient developed a cyst that required stereotactic aspiration and thalamotomy for the associated tremor. T2 hyperintensities occurred in 8 (4.7%) patients. They were managed conservatively in 6 and with a transient course of corticosteroids in two. No factors were associated with the occurrence of ARE using logistic analysis. Table 6
Clinical outcomes per patient
New neurological deficits occurred in 27 (15.9%) patients, with 12 (7.1%) being permanent. The cause was new post-SRS hemorrhage in 11 (6.5%) (8 (4.4%) transient deficit and 3 (1.7%) permanent deficit), ARE in 2 (1.2%) (one (0.6%) transient and one (0.6%) permanent deficit). Two (2.1%) patients (with transient deficit) presented with ARE and a new hemorrhage. In three (1.8%) patients (with permanent deficit), the symptoms were linked to the CCM despite the lack of evidence of hemorrhage or ARE on imaging. Four (2.4%) patients developed new neurologic symptom unrelated to the CCM (one (0.6%) permanent and three (1.8%) transient); two (2.1%) patients (with permanent deficit) had hemorrhage from other CCMs. In three (1.8%) patients (with 2 (1.2%) permanent deficit and one (0.6%) transient), the reason for the new deficit is unknown; one (0.6%) patient had a worsening of his clinical presentation (dysesthesia and hypoesthesia in one hand) without clear radiologic explanation.
Overall, with all events considered, 69 patients (42.3%) demonstrated improvement, 81 (49.7%) exhibited stability, and 13 (8.0%) had a worsening of their clinical neurological status at a median follow-up of 3.4 years (IQR: 5.4). Among the 13 patient with worsening of their clinical neurological status, 10 (5.6%) were potentially associated with the treated CCM (new hemorrhage, ARE, unknown). The data of previous symptom evolution were missing for 7 patients. No patient died during the follow-up period.
Additional treatment
Two patients had repeat SRS, and three had an open neurosurgical procedure with a median delay of 7.5 years (IQR=4.8) from initial SRS.