Faculty Bibliography

Predictions of patient outcomes after a given therapy are fundamental to medical practice. We employ a machine learning approach towards predicting the outcomes after stereotactic radiosurgery for cerebral arteriovenous malformations (AVMs). Using three prospective databases, a machine learning approach of feature engineering and model optimization was implemented to create the most accurate predictor of AVM outcomes. Existing prognostic systems were scored for purposes of comparison. The final predictor was secondarily validated on an independent site's dataset not utilized for initial construction. Out of 1,810 patients, 1,674 to 1,291 patients depending upon time threshold, with 23 features were included for analysis and divided into training and validation sets. The best predictor had an average area under the curve (AUC) of 0.71 compared to existing clinical systems of 0.63 across all time points. On the heldout dataset, the predictor had an accuracy of around 0.74 at across all time thresholds with a specificity and sensitivity of 62% and 85% respectively. This machine learning approach was able to provide the best possible predictions of AVM radiosurgery outcomes of any method to date, identify a novel radiobiological feature (3D surface dose), and demonstrate a paradigm for further development of prognostic tools in medical care.

Cervical-petrous internal carotid artery (CP-ICA) pseudoaneurysms are rare and have different etiologies, presentations, and treatment options. A middle-aged patient with a history of chronic otitis media presented with acute otorrhagia and was found to have a left-sided CP-ICA pseudoaneurysm. The patient was a poor surgical candidate with difficult arterial access. The pseudoaneurysm was treated with stand-alone coiling via a left brachial approach with persistent contrast filling seen only in the aneurysm neck at the end of the procedure. The patient re-presented 12 days later with repeat hemorrhage and rapid enlargement of the neck remnant, and was treated with a covered stent via a transcervical common carotid artery cut-down. A covered stent may provide a more definitive treatment for CP-ICA pseudoaneurysms compared with standalone coiling.

BACKGROUND:The degree of aneurysm occlusion following coil embolization has an impact on aneurysm recanalization. OBJECTIVE:To explain the natural history of intracranial aneurysms with neck remnant, Raymond-Roy Occlusion Classification (RROC) class II. METHODS:A single-center, retrospective study of 198 patients with 209 aneurysms treated with coil embolization that were initially either RROC class I or II. The angiographic outcomes at short- and long-term follow-up were compared as well as the complication/re-treatment rates. Atypical aneurysms and those that had been previously treated were excluded. RESULTS:Ninety-nine class I aneurysms were compared with 110 class II aneurysms. There was no difference in recanalization rate between the groups (class I 3.3% vs class II 8.5%, p=0.478) at short-term follow-up (8.2 months) and at subsequent follow-ups (21.7 and 52.1 months). There was also no difference in re-treatment rates (class I 3.3% vs class II 8.5%, p=0.196) or complication rates (class I 9.1% vs class II 4.6%, p=0.12). There were no aneurysm ruptures after treatment in either group. CONCLUSIONS:The angiographic outcome of aneurysms with neck remnant following coil embolization is similar to that of completely occluded aneurysms in that most remain stable and few recanalize. This understanding could potentially help the interventional neurosurgeon avoid complications such as coil herniation, vessel compromise, and stroke in selected cases. Further investigation with a larger patient population is warranted.

BACKGROUND:The Raymond-Roy Occlusion Classification (RROC) is the standard for evaluating coiled aneurysms (Class I: complete obliteration; Class II: residual neck; Class III: residual aneurysm), but not all Class III aneurysms behave the same over time. METHODS:This is a retrospective review of 370 patients with 390 intracranial aneurysms treated with coil embolization. A Modified Raymond-Roy Classification (MRRC), in which Class IIIa designates contrast within the coil interstices and Class IIIb contrast along the aneurysm wall, was applied retrospectively. RESULTS:Class IIIa aneurysms were more likely to improve to Class I or II than Class IIIb aneurysms (83.34% vs 14.89%, p<0.001) and were also more likely than Class II to improve to Class I (52.78% vs 16.90%, p<0.001). Class IIIb aneurysms were more likely to remain incompletely occluded than Class IIIa aneurysms (85.11% vs 16.67%, p<0.001). Class IIIb aneurysms were larger with wider necks while Class IIIa aneurysms had higher packing density. Class IIIb aneurysms had a higher retreatment rate (33.87% vs 6.54%, p<0.001) and a trend toward higher subsequent rupture rate (3.23% vs 0.00%, p=0.068). CONCLUSIONS:We propose the MRRC to further differentiate Class III aneurysms into those likely to progress to complete occlusion and those likely to remain incompletely occluded or to worsen. The MRRC has the potential to expand the definition of adequate coil embolization, possibly decrease procedural risk, and help endovascular neurosurgeons predict which patients need closer angiographic follow-up. These findings need to be validated in a prospective study with independent blinded angiographic grading.

INTRODUCTION/BACKGROUND:Although approximately one in every 10 patients undergoing coil embolization of intracranial aneurysms requires retreatment, the factors that are associated with outcome following retreatment remain to be fully elucidated. METHODS:This is a single-center, retrospective review of 43 patients with 58 intracranial aneurysms that were retreated from 2004 to 2014. Aneurysms undergoing first time or microsurgical retreatment were excluded. Retreatment types were grouped into those without permanent parent vessel support (stand-alone and balloon-assisted coiling) versus those with permanent parent vessel support (stent-assisted coiling, stand-alone stenting, and flow diversion). The Modified Raymond Roy Classification was used to group aneurysms at all angiographic follow-up points either in the successful outcome group (Class I or II) or the unsuccessful outcome group (Class IIIa or IIIb). RESULTS:Of aneurysms with follow-up, 50% were in the successful group and 50% in the unsuccessful group. In univariate analysis, small aneurysm size (p < 0.001), previous treatment type (p = 0.022), retreatment type (p = 0.001), and initial occlusion class (p = 0.005) were all associated with angiographic outcome. In multivariate analysis, small aneurysm size (p = 0.005, odds ratio (OR) 24.56, confidence interval (CI) 2.68-225.4) and retreatment type with permanent parent vessel support, namely stent-assisted coiling (p = 0.017, OR 31.1, CI 1.89-517.7), were associated with retreatment success. CONCLUSIONS:Small aneurysm size and retreatment with permanent parent vessel support, namely stent-assisted coiling, are predictors of success following endovascular retreatment of intracranial aneurysms. These findings could be useful in the effort to both prevent and predict treatment failure following endovascular retreatment.

We present a retrospective review of 357 consecutive patients with 419 aneurysms treated with coil embolization. Although incomplete occlusion and recurrence of intracranial aneurysms following coil embolization is a well-known problem, the factors that influence and predict treatment failure are still debated. For this study, we excluded non-coiling endovascular techniques (flow diversion) and non-saccular aneurysms (fusiform). The modified Raymond-Roy occlusion classification (MRRC) was used to grade the aneurysms. Treatment failure was defined as filling of the aneurysm dome (MRRC Class IIIa or IIIb) at the first angiographic follow-up (average 8 months). Univariate statistical tests were employed to select variables for incorporation into a multivariable logistic regression model. Multivariate analysis identified greater aneurysm volume (p<0.001), packing density (PD) less than 31% (p=0.007) and initial MRRC Class IIIb (p<0.001) as predictors of treatment failure. Incomplete neck coverage with coils was associated with treatment failure in univariate but not multivariate analysis. Class IIIb status was more predictive of treatment failure compared to all Class III (odds ratio 168 versus 14.4). Clinical outcomes were similar in both groups except that there were more retreatments in the treatment failure group (p<0.001). Aneurysm volume, PD and initial occlusion class are associated with angiographic outcome, consistent with prior literature. The MRRC is a powerful predictor of treatment failure. These results will be useful in the effort to both prevent and predict treatment failure after coil embolization, however, they should be verified in a prospective study.

BACKGROUND:Achieving durable local control while limiting normal tissue toxicity with definitive radiation therapy in the management of high-risk brain metastases remains a radiobiological challenge. The objective of this study was to examine the local control and toxicity of a 5-fraction stereotactic radiosurgical approach for treatment of patients with inoperable single high-risk NSCLC brain metastases. METHODS:This retrospective analysis examines 20 patients who were deemed to have "high-risk" brain metastases. High-risk tumors were defined as those with a maximum diameter greater than 2 cm and/or those located within an eloquent cortex. Patients were evaluated by a neurosurgeon prior to treatment and determined to be inoperable due to tumor or patient characteristics. Patients were treated using the CyberKnife® SRS system in 5 fractions to a total dose of 30 Gy, 35 Gy, or 40 Gy. RESULTS:Twenty patients with a median age of 65.5 years were treated from April 2010 to August 2014 in 5 fractions to a median total dose of 35 Gy. At a median follow up of 11.3 months local tumor control was observed in 18 of 20 metastases (90 %). Both local failures were observed in patients receiving a lower dose of 30 Gy. Median pre-treatment dexamethasone dose was 10 mg/day and median post-treatment nadir dose was 0 mg/day. Salvage intracranial therapy was required in 45 % of patients. Symptomatic radionecrosis was observed in 4 of 20 patients (20 %), two of which were treated to 40 Gy and the remainder to 35 Gy. Kaplan-Meier 1-year, 2-year, and median survival were calculated to be 45 %, 20 %, and 13.2 months, respectively. CONCLUSIONS:Five-fraction SRS to a total dose of 35 Gy appears to be a safe and effective management strategy for single high-risk NSCLC brain metastases, while a total dose of 40 Gy leads to an excess risk of neurotoxicity.

BACKGROUND: Embolization before stereotactic radiosurgery (SRS) for cerebral arteriovenous malformations (AVM) has been shown to negatively affect obliteration rates, but its impact on the risks of radiosurgery-induced complications and latency period hemorrhage is poorly defined. OBJECTIVE: To determine, in a case-control study, the effect of prior embolization on AVM SRS outcomes. METHODS: We evaluated a database of AVM patients who underwent SRS. Propensity score analysis was used to match the case (embolized nidi) and control (nonembolized nidi) cohorts. AVM angioarchitectural complexity was defined as the sum of the number of major feeding arteries and draining veins to the nidus. Multivariate Cox proportional hazards regression analyses were performed on the overall study population to determine independent predictors of obliteration and radiation-induced changes. RESULTS: The matching process yielded 242 patients in each cohort. The actuarial obliteration rates were significantly lower in the embolized (31%, 49% at 5, 10 years, respectively) compared with the nonembolized (48%, 64% at 5, 10 years, respectively) cohort (P = .003). In the multivariate analysis for obliteration, lower angioarchitectural complexity (P < .001) and radiologically evident radiation-induced changes (P = .016) were independent predictors, but embolization was not significant (P = .744). In the multivariate analysis for radiologic radiation-induced changes, lack of prior embolization (P = .009) and fewer draining veins (P = .011) were independent predictors. CONCLUSION: The effect of prior embolization on AVM obliteration after SRS may be significantly confounded by nidus angioarchitectural complexity. Additionally, embolization could reduce the risk of radiation-induced changes. Thus, combined embolization and SRS may be warranted for appropriately selected nidi. ABBREVIATIONS: AVM, arteriovenous malformationRBAS, radiosurgery-based AVM scoreSRS, stereotactic radiosurgeryVRAS, Virginia Radiosurgery AVM Scale.

Cervical-petrous internal carotid artery (CP-ICA) pseudoaneurysms are rare and have different etiologies, presentations, and treatment options. A middle-aged patient with a history of chronic otitis media presented with acute otorrhagia and was found to have a left-sided CP-ICA pseudoaneurysm. The patient was a poor surgical candidate with difficult arterial access. The pseudoaneurysm was treated with stand-alone coiling via a left brachial approach with persistent contrast filling seen only in the aneurysm neck at the end of the procedure. The patient re-presented 12 days later with repeat hemorrhage and rapid enlargement of the neck remnant, and was treated with a covered stent via a transcervical common carotid artery cut-down. A covered stent may provide a more definitive treatment for CP-ICA pseudoaneurysms compared with standalone coiling.

PURPOSE/OBJECTIVE:Stereotactic body radiation therapy (SBRT) is increasingly utilized as primary treatment for clinically localized prostate cancer. While acute post-SBRT urinary symptoms are well recognized, the late genitourinary toxicity of SBRT has not been fully described. Here, we characterize the clinical features of late urinary symptom flare and recommend conservative symptom management approaches that may alleviate the associated bother. METHODS:Between February 2008 and August 2011, 216 men with clinically localized prostate cancer were treated definitively with SBRT at Georgetown University Hospital. Treatment was delivered using the CyberKnife with doses of 35-36.25 Gy in five fractions. The prevalence of each of five Common Terminology Criteria for Adverse Events (CTCAE) graded urinary toxicities was assessed at each follow-up visit. Medication usage was documented at each visit. Patient-reported urinary symptoms were assessed using the American Urological Association (AUA) symptom score and the Expanded Prostate Cancer Index Composite (EPIC)-26 at 1, 3, 6, 9, 12, 18, and 24 months. Late urinary symptom flare was defined as an increase in the AUA symptom score of ≥5 points above baseline with a degree of severity in the moderate to severe range (AUA symptom score ≥15). The relationship between the occurrence of flare and pre-treatment characteristics were examined. RESULTS:For all patients, the AUA symptom score spiked transiently at 1 month post-SBRT. Of the 216 patients, 29 (13.4%) experienced a second transient increase in the AUA symptom score that met the criteria for late urinary symptom flare. Among flare patients, the median age was 66 years compared to 70 for those without flare (p = 0.007). In patients who experienced flare, CTCAE urinary toxicities including dysuria, frequency/urgency, and retention peaked at 9-18 months, and alpha-antagonist utilization increased at 1 month post-treatment, rose sharply at 12 months post-treatment, and peaked at 18 months (85%) before decreasing at 24 months. The EPIC urinary summary score of flare patients declined transiently at 1 month and experienced a second, more protracted decline between 6 and 18 months before returning to near baseline at 2-year post-SBRT. Statistically and clinically significant increases in patient-reported frequency, weak stream, and dysuria were seen at 12 months post-SBRT. Among flare patients, 42.9% felt that urination was a moderate to big problem at 12 months following SBRT. CONCLUSION/CONCLUSIONS:In this study, we characterize late urinary symptom flare following SBRT. Late urinary symptom flare is a constellation of symptoms including urinary frequency/urgency, weak stream, and dysuria that transiently occurs 6-18 months post-SBRT. Provision of appropriate anticipatory counseling and the maintenance of prophylactic alpha-antagonists may limit the bother associated with this syndrome.