Post Stereotactic Radiosurgery Research Articles

Stereotactic Radiosurgery With vs. Without Prior Embolization for Brain Arteriovenous Malformations: A Propensity Score Matching Analysis

Objective: Whether partial embolization could facilitate the post-stereotactic radiosurgery (SRS) obliteration for brain arteriovenous malformations (bAVMs) remains controversial. We performed this study to compare the outcomes of SRS with and without prior embolization for bAVMs.Methods: We retrospectively reviewed the Beijing Tiantan AVMs prospective registration research database from September 2011 to October 2014. Patients were categorized into two groups, combined upfront embolization and SRS (Em+SRS group) and SRS alone (SRS group), and we performed a propensity score matching analysis based on pre-embolization baseline characteristics; the matched groups each comprised 76 patients.Results: The obliteration rate was similar between SRS and Em+SRS (44.7 vs. 31.6%; OR, 1.754; 95% CI, 0.905–3.401; p = 0.096). However, the SRS group was superior to the Em+SRS group in terms of cumulative obliteration rate at a follow-up of 5 years (HR,1.778; 95% CI, 1.017–3.110; p = 0.033). The secondary outcomes, including functional state, post-SRS hemorrhage, all-cause mortality, and edema or cyst formation were similar between the matched cohorts. In the ruptured subgroup, the SRS group could achieve higher obliteration rate than Em+SRS group (56.5 vs. 31.9%; OR, 2.773; 95% CI, 1.190–6.464; p = 0.018). The cumulative obliteration rate at 5 years was also higher in the SRS group (64.5 vs. 41.3%; HR, 2.012; 95% CI, 1.037–3.903; p = 0.038), and the secondary outcomes were also similar between the matched cohorts.Conclusion: Although there was no significant difference in the overall obliteration rate between the two strategies, this study suggested that pre-SRS embolization may have a negative effect on post-SRS obliteration. Furthermore, the obliteration rates of the SRS only strategy was significantly higher than that of the Em+SRS strategy in the ruptured cohort, while no such phenomenon was found in the unruptured cohort.

Investigating diffusion patterns of brain metastases pre- and post-stereotactic radiosurgery: a feasibility study

Purpose. Metastatic complications are responsible for 90% of cancer-associated mortality. Magnetic resonance imaging (MRI) can be used to observe the brain’s microstructure and potentially correlate changes with metastasis occurrence. Diffusion weighted imaging (DWI) is an MRI technique that utilizes the kinetics of water molecules within the body. The aim of this study is to use DWI to characterize diffusion changes within brain metastases in cancer patients pre- and post-stereotactic radiosurgery (SRS). Methods. We retrospectively analyzed 113 metastases from 13 patients who underwent SRS for brain metastasis recurrence. Longitudinal apparent diffusion coefficient (ADC) maps were registered to Gd-T1 images and CT, and clinical metastasis ROIs from all SRS treatments were retrospectively transferred onto these ADC maps for analysis. Metastases were characterized based on pre-SRS diffusion pattern, primary cancer site, and post-SRS outcome. ADC values were calculated pre- and post-SRS. Results.ADC values were significantly elevated (980.2 × 10−6 mm2 s−1 and 1040.3 × 10−6 mm2 s−1 pre- and post-SRS, respectively) when compared to healthy brain tissue (826.8 × 10−6 mm2 s−1) for all metastases. Three identified pre-SRS patterns were significantly different before SRS and within 6 months post-SRS. No significant differences were observed between different primaries pre-SRS. Post-SRS, Lung metastases ADC decreased by 86.2 × 10−6 mm2 s−1, breast metastases increased by 116.7 × 10−6 mm2 s−1, and genitourinary metastases showed no significant ADC change. SRS outcomes showed ADC variability pre-treatment but no significant differences pre- and post-SRS, except at 6–9 months post-SRS where progressing metastases were elevated when compared to other response groups. Conclusion. This study provided a unique opportunity to characterize diffusion changes in brain metastases before their manifestation on standard Gd-T1 images and post-SRS. Identified patterns may improve early detection of brain metastases as well as predict their response to treatment.

MLTI-05. Adjuvant re-irradiation improves local control of surgically resected recurrent brain metastases

BackgroundThe efficacy of salvage resection (SR) of recurrent brain metastases (BrM) post-stereotactic radiosurgery (SRS) is not well described. We sought to characterize the impact of adjuvant post-salvage radiation therapy (PSRT) in this setting and identify tumor-specific variables that influence local control.MethodsRetrospective analysis of post-SRS recurrent BrM that underwent SR between 2003–2020 at Memorial Sloan Kettering Cancer Center was performed. Cases with histologically-viable malignancy were included and stratified by receipt of adjuvant PSRT within 60 days of SR (PSRT cohort) vs. observation (observation cohort). Resection-site outcomes were described using cumulative incidences and univariate and multivariate competing risks regression accounting for clustering.ResultsOne-hundred fifty-five recurrent BrM in 135 patients were included. Thirty-nine (25.2%) of the post-operative cavities were treated with adjuvant PSRT, and the remaining 116 (74.8%) cavities were initially observed. Gross- or near-total resection was associated with significantly improved local control compared to subtotal resection (p=0.007). Adjuvant PSRT was associated with a reduced rate of LR at 6 months [18.0% (95%CI: 9.8–33.1%) vs. 35.9% (95%CI: 27.9–46.2%) with initial observation] and 12 months [28.8% (95%CI: 17.0–48.8%) vs. 43.9% (95%CI: 36.2–53.4%)]. On multivariate analysis, adjuvant PSRT (p=0.095), low tumor-viability within the resected BrM (p=0.17), and first-time resection (p=0.035) all independently trended towards improved local control. BrM size at SR (≥3cm vs. <3cm, p=0.48), primary malignancy (p=0.35), and specific PSRT modality (whole or partial brain radiation vs. SRS, p=0.43) were not associated with differences in LR rate. Radiation necrosis (RN) was significantly increased in the PSRT cohort (HR 4.55, 95%CI: 1.26–16.39, p=0.02), though the total percentage with symptomatic RN remained low (PSRT cohort 5.1% vs observation cohort 0.9%).ConclusionsLocal control after SR of a recurrent BrM may be optimized with gross- or near-total resection and adjuvant post-operative re-irradiation, with low symptomatic RN.

OMRT-9. Effect of Pre-Operative Stereotactic Radiosurgery on Brain Metastasis: Analysis of DNA and RNA Genomic Profiles from Phase-II Clinical Trial NCT03398694

BackgroundWith improved systemic therapy that has limited impact on the intracranial compartment, the incidence of brain metastasis (BM) from solid cancers is rising and negatively impacting patient’s overall survival (OS). Treatment varies based on presentation, however, for patients with <4 symptomatic BMs current clinical practice involves surgical resection followed by stereotactic radiosurgery (SRS) to the resection cavity. Post-operative SRS is associated with increased risk of leptomeningeal disease (LMD) and local recurrence in the follow-up period. We hypothesize that pre-operative SRS will decrease the incidence of LMD as well as local recurrence and increase patient’s OS by delivering a lethal dose of radiation to tumor cells before they are disturbed by surgical resection. In a Phase II clinical trial (NCT03398694) we are treating patients with 1–4 symptomatic BMs with pre-operative SRS while collecting DNA and RNA sequencing data from core and peripheral edges of the resected tumor to examine the genomic effects of SRS on tumor.MethodsPost-SRS resected tumor specimens were divided into two groups: ‘center’ and ‘periphery’ with respect to the center of SRS treatment with periphery within 50% isodose line. Previously resected untreated BMs were used as control. DNA and RNA were isolated from all samples for sequencing.ConclusionsOur initial analyses show that pre-treatment with SRS, results in significant genomic changes at DNA and RNA levels throughout the tumor, in both center as well as periphery. Furthermore, significant transcriptomic differences were noted among matched samples between the central and peripheral SRS locations implicating differential effect of SRS dosing within a tumor. Initial gene ontological analysis on non-small cell lung cancer samples demonstrated an overexpression of WNT and BMP signaling pathways (p <.001, p<.01). These pathways are typically involved in neuronal development, hinting that adaptation to the brain microenvironment was occurring post SRS treatment.

A case of delayed cyst formation post brain AVM stereotactic radiosurgery for arteriovenous malformation: Case report

Arteriovenous malformations (AVMs) are a rare cause of cerebrovascular abnormality with incidence of about 1 in 100,000 people per year and point prevalence of about 0.2%. AVMs are associated with serious complications such as intracranial haemorrhage (2–4% a year, 16% and 29% at 10 and 20 years after diagnosis), seizures (10–30%), focal neurologic deficits, and headaches. The management options are surveillance, endovascular embolization, microsurgical excision and stereotactic radiosurgery (SRS).In SRS Stereotactically focused high energy beams of photons induce progressive thrombosis by fibro-intimal hyperplasia and subsequent luminal obliteration. These changes usually take one to three years known as “latency period”. Complications are reported in 8% of patients undergoing SRS, including radiographic parenchymal lesions, cranial nerve deficits, seizures, headaches, and cyst formation.Cyst formation is reported in about 1.2 % of patients undergoing SRS. While the exact mechanism of post SRS cyst formation is unclear, it is hypothesized that it might be due to damage to the blood brain barrier and increased vessel wall permeability. Delayed cyst formation is reported with latency period between 3 and 10 years after radiotherapy for treatment of cerebrovascular AVMs. However, cystic formation with longer latency periods (in one case upto 17 years) after radiotherapy for other causes such as nasopharyngeal cancers have been reported.Here we report a case of delayed cyst formation after SRS for cerebrovascular AVM with latency period of 20 years.

Five-Fraction Stereotactic Radiosurgery (SRS) For Resected Brain Metastases

Randomized trials have demonstrated improved rates of local control (LC) in patients who undergo postoperative stereotactic radiosurgery (SRS) compared to observation following resection of brain metastases. However, for larger tumors, dose reduction is commonly required for single-fraction SRS in order to minimize toxicity, potentially compromising LC. Hypofractionated SRS (HF-SRS) may improve LC with minimal toxicity, in particular radionecrosis (RN). In this IRB-approved retrospective study, patients who received 5-fraction HF-SRS at Duke University Medical Center from 2008 to 2018 were identified. GTVs were contoured per consensus guidelines; PTV expansions were 2mm. Patients were treated with 5 or 5.5 Gy per fraction, the latter typically utilized for tumor histologies considered to be radioresistant. Treatment consisted of linear-accelerator-based image-guided SRS with HD-MLC. Data collected included demographics, primary disease characteristics and treatment details. For patients with sufficient radiographic follow up, rates of LC, distant brain failure (DBF), RN, overall survival (OS), and leptomeningeal disease (LMD) were calculated. 250 patients were identified. Lung was the most common primary site (43%), followed by breast (16%) and GI (13%). 69% had evidence of active systemic disease at time of treatment. Of 144 patients with recorded Karnofsky performance status (KPS), 90% had a KPS of 70 or greater. Median age at time of craniotomy was 63 years. Median time interval between craniotomy and start of SRS was 4 weeks (range 1-16). 86% of patients had at least 1 subsequent brain MRI post SRS. 3-year LC was 84% and median OS was 12.8 months (95% CI 10.9-15.5). 24-month incidence of DBF was 50% (95% CI 44%-56%). 24-month incidence of RN was 4% (95% CI 2% to 7%). 14% of patients developed LMD. On univariate analysis (UVA), a dose of 5.5 Gy/fraction was associated with improved LC, HR 0.25 (0.08, 0.82), as was no prior treatment, HR 0.40 (0.20, 0.82). Location, receipt of immunotherapy or small molecule inhibitor, dural contact, and cavity volume were not significant for LC on UVA. Compared to historical standards of single fraction SRS, post-operative, five-fraction HF-SRS results in low rates of LC and RN. The use of 5.5 Gy per fraction and the absence of prior treatment were associated with improved LC. Rates of LMD in this patient population are comparable to other studies in the postoperative setting.

Trigeminal Neuralgia Recurrence Post-Stereotactic Radiosurgery Unaffected by Integral Dose

Trigeminal Neuralgia Recurrence Post-Stereotactic Radiosurgery Unaffected by Integral Dose

Single centre, retrospective review of gamma-knife stereotactic radiosurgery and other therapies on prevalence of seizures in patients with brain metastases

Abstract Background An increasing number of patients with brain metastases (BM) are having stereotactic radiosurgery (SRS), but it is not known whether this causes epilepsy. Methods We carried out a retrospective review of patients surviving one-year post gamma-knife SRS at the National Hospital for Neurology and Neurosurgery (NHNN) between February 2012 and April 2017. Data on seizures during the pre- and post-SRS periods were collected along with information about the primary tumour, metastasis location and previous treatments, including whole brain radiotherapy (WBRT) and surgery. Results 61 patients were treated with SRS. 6 patients had incomplete records and were excluded. Of the remaining 55, 21 had a seizure at some point. 4 had seizures both pre- and post-SRS, 7 had seizures pre-SRS but not post and 10 patients had de-novo seizures post-SRS. 34 did not have a documented seizure at any point. Of the 14 patients who had seizures post-SRS, 4 also had both WBRT and surgery, 2 had WBRT and 4 had surgery. 100% (4/4) who had WBRT, surgery and SRS went on to have a seizure. Seizures occurred in 11/25 patients who had previous surgery and 7/11 who had previous WBRT. The primary tumour and metastasis location had no obvious impact on seizure incidence. Conclusions The incidence of new seizures post-SRS is low (18%). Previous surgery and/or WBRT may increase seizure incidence post-SRS. The data is currently being reviewed for effect of tumour/ treatment volume, dose delivered, presence of significant oedema and radionecrosis. A larger prospective study is also underway.

Multifraction Stereotactic Radiosurgery to Large Brain Metastases: Is This Scheme Effective and Safe?

Many fractionation regimens have been described for the treatment of large brain metastases with stereotactic radiosurgery (SRS). The purpose of this analysis is to demonstrate the efficacy and the tolerance of multifraction stereotactic radiosurgery (mSRS), 3 fractions, in patients with large brain metastases. Were reviewed all patients with brain metastases treated with mSRS (3 fractions) at a single institution. Patients with large brain metastases were the majority population, but patients who were not appropriate candidates for single fraction SRS were treated with mSRS and included as well. Prognostic factors including Ds-GPA, Score Index for Radiosurgery (SIR), number of lesions, systemic disease status, primary site, PTV volume, dose (3 schemes were used due a constraint of the organ at risk - 7 Gy, 8 Gy or 9 Gy/fraction), use of whole brain radiotherapy (WBRT) either pre or post SRS, surgery and the use of systemic therapy were all evaluated. The toxicity was graded using the CTCAE V. 4.0. Kaplan- Mayer and Log-Rank test were used for survival curve calculations. Chi-square test was used for comparative analysis with a significance level of 0.05. Between March 2015 and November 2018, 86 patients were enrolled, of which 2 were excluded due to loss to follow-up and 144 lesions treated. The median follow-up was 9.0 months. The patient characteristics are shown in Table 1. The overall survival (OS) at 12 and 24 months was 43% and 21.6%, respectively, with a median overall survival of 10 months. Death due to neurological failure was 14.5%. Local control at 12 months was 81.3%, with 47.6% developing new lesions after the first mSRS. There wasn’t a correlation between local control and dose/fraction, PTV volume and the other factors. There were 4 patients with acute grade 3 toxicity (paresthesia), reversible with corticosteroids. The most common symptom was headache, grade 1 and 2, (23.5%). Eight patients had grade 1-2 seizures. The radionecrosis rate was 9.5%, and four patients had symptomatic radionecrosis. The PTV volume average was 20.87 cc (1-140 cc). In multivariate analysis, Ds-GPA, SIR, KPS, primary site (melanoma and breast were better) and systemic progression status had a statistically significant effect on OS. The dose/fraction and PTV volume did not correlate with toxicity or radionecrosis. This mSRS scheme of 3 fractions to large brain metastases showed good local control with low complication rates, in the studied population. In our retrospective analysis of 84 patients, we conclude that this 3 fraction mSRS is safe and effective. A prospective trial is warranted to compare and evaluate the best SRS regimens for this approach.Abstract 2160; Table 1Patients CharacteristicsAge (median)54.5 y (22-81)Numb. Lesions: 1 - 3; >3;91.6%/ 8.4%Ds-GPA 0-1; 1.5 - 2.5; >2.5;23.8%/ 29.8%/ 46.4%/Primary Lung; Breast; Melanoma; Others;39.3%/ 28.6%/ 8.3%/ 23.8%Fractionation 7 Gy x 3; 8 Gy x 3; 9 Gy x 3;8.3%/ 29.8%/ 61.9%/WBRT Pre mSRS; Pos mSRS; No;13.1%/ 23.8%/ 63.1%/ Open table in a new tab

Use of Radiomics in the Prediction of Brain Metastases Local Control Post-Stereotactic Radiosurgery

Radiomics has shown promise for predicting local recurrence in brain metastases treated with stereotactic radiosurgery (SRS). There is however uncertainty regarding the extent of overlap between information provided by radiomic and clinical features. We hypothesize that radiomics provides complementary information to what is typically available in the clinical workflow. The retrospective dataset consisted of 408 brain metastases across 87 patients with pretreatment T1c and FLAIR scans for each case. Lesion progression status was determined at the last available follow-up, with maximum axial diameter growth (>20%) on T1c indicating local failure. A total of 440 radiomic features were extracted from the tumor core region (planning GTV), as well as the surrounding region. Features were ranked based on resampled Radom Forest (RF) feature importance. Top ranked radiomic features were combined with clinical features and compared to clinical features alone. A RF classifier was trained on each set of features in a leave one patient out manner where lesions belonging to a single patient used for validation each fold. Statistical comparisons were made through a paired t-test of AUC values calculated from resampled predictions of two classifiers on the validation data. Top 10 highest ranked radiomic features were combined, one at a time, with the clinical features, resulting in a very significant (p<0.001) increase in AUC from the addition of the top 9 radiomic features, and no increase from adding the 10th best radiomic feature. An optimized combination of radiomic and clinical features resulted in a dramatically higher performance (resampled AUC: mean=0.792, 95% C.I of mean=0.790-0.793) compared to clinical features alone (0.676, 0.674-0.678). The increase in performance from incorporating radiomic features suggests that quantitative characterization of tumor structure through the use of radiomics adds complementary information to what is typically available in the clinical workflow.

MLTI-12. TIMING OF SYSTEMIC THERAPY ADMINISTRATION RELATIVE TO STEREOTACTIC RADIOSURGERY AND DEVELOPMENT OF RADIATION NECROSIS IN PATIENTS WITH BRAIN METASTASES

PURPOSE: The mainstay of oncologic therapy for patients with brain metastases involves brain-directed radiation, increasingly given via stereotactic radiosurgery (SRS), and systemic therapy for extracranial disease control. We sought to investigate the association between the timing of systemic therapy and SRS administration on development of radiation necrosis among patients with brain metastases. METHODS: We retrospectively identified 429 patients treated at Brigham and Women’s Hospital/Dana-Farber Cancer Institute with SRS for newly-diagnosed brain metastases between 2001–2015. Systemic therapy was tiered into 4 categories: chemotherapy, immunotherapy, hormonal therapy, and targeted therapy. All images were manually reviewed by two radiation oncologists specializing in brain tumors to assess the presence versus absence of radiographic necrosis. Patients with radiographic necrosis who harbored associated neurologic symptoms or were managed with steroids/bevacizumab/resection were considered to have symptomatic radiation necrosis. Data were analyzed using univariable Cox regression in SAS v9.4. The median follow-up in surviving patients was 1.79 years. RESULTS: In total, 252/429 and 361/429 patients received systemic therapy pre and/or post SRS, respectively. Patients receiving systemic therapy ≤5 days before SRS displayed higher rates of radiographic (HR 2.48, 95% CI 1.06–5.81, p=.04) and symptomatic (HR 3.74, 95% CI 1.08–12.98, p=.04) necrosis; a similar association was seen in patients receiving systemic therapy ≤5 days after SRS (HR 1.72, 95% CI 0.84–3.53, p=.14 and HR 4.42, 95% CI 1.75–11.14, p=.002, respectively). Trends towards increased necrosis risk were noted when comparing systemic therapy administration 1–5 days versus 6–10 days before/after SRS. The above 4 associations were significant when restricting the cohort to patients receiving targeted systemic therapy (HR-range 3.57–21.49, p-range 0.01–0.04). CONCLUSIONS: Our results suggest that a reasonable delay between SRS and systemic therapy administration may reduce rates of radiation necrosis, even among patients receiving targeted therapies. Validation in an independent data set would lend further support to this concept.

Impact of Integral Dose on the Maintenance of Pain Relief in Patients with Idiopathic Trigeminal Neuralgia Treated with Upfront Gamma Knife Radiosurgery

Integral dose (ID) defined as the product of mean dose and target volume, is a measure of the absorbed radiation energy. To evaluate the effect of ID on the duration of pain relief after primary Gamma knife stereotactic radiosurgery (GKSRS) in trigeminal neuralgia. A total of 78 patients who achieved initial pain response of I-III on the Barrow Neurological Institute Pain Scale (BNI-PS) following primary GKSRS for idiopathic trigeminal neuralgia were included in this study. A Cox regression model was used to compute the prognostic factor with respect to the ID within the 50% isodose line. Facial pain relief maintenance interval was defined as time interval between the day of improved BNI-PS grade and pain level back to the BNI-PS IV or V. The median duration of follow-up was 42 months (range 6-108 months). After the initial GKSRS, patients achieved pain relief at a median of 0.5 months (range, 7 days to 6 months); 28 patients developed recurrence of pain. There was a positive correlation between the pain relief maintenance and increasing ID within 50% isodose line (hazard ratio 1.85, P= 0.04) on multivariable Cox-regression analysis. Using logistic regression analysis, we found that ID was not predictive of developing post-stereotactic radiosurgery hypoesthesia (P= 0.64, hazard ratio 1.057). Stereotactic radiosurgery can be individualized based on trigeminal nerve morphology to achieve durable pain relief in patients with trigeminal neuralgia. ID calculation aids in planning an optimal radiation dose based on the nerve morphology to provide durable pain relief of idiopathic trigeminal neuralgia.

Focused Ultrasound Induced-Blood-Brain Barrier Opening in Post-Radiosurgery Mouse Brain for Locally-Enhanced Systemic Therapy

Focused ultrasound (FUS) is a novel technique that is capable of non-invasively and transiently open the blood-brain barrier (BBB). It has been show that FUS can facilitate the delivery of systemically administered therapeutic agents. Our goal here is to investigate FUF induced BBB opening in a post stereotactic radiosurgery (SRS) mouse model for novel locally-enhanced systemic therapy. We hypothesized that microvascular effects of radiation could modulate the BBB-opening induced by FUS. In this study, normal brains of C57B5 mice were treated with a fractionated SRS dose, typically given to resected tumor cavities. A single beam radiation (6Gy x 5 daily fractions) was delivered through a 10×10 mm2 collimator to half of the mouse brain using the Small Animal Radiation Research Platform (SARRP). The dose delivered to the mice was verified via two mouse-like anthropomorphic phantoms in the axial and sagittal orientations. Upon the completion of SARRP treatments, mice were divided into two groups for FUS-induced BBB opening on day 7 (acute) and day 35 (chronic), respectively. A 1.5 MHz single element FUS transducer (-6dB focal zone was 1×1×7.5 mm3) was used for all sonications. The caudate-putamen (CPu) was the FUS target and four 30s sonications were delivered on each side of the brain. The BBB opening was confirmed with contrast-enhanced T1-weighted MRI scans, while dynamic MRI scans were also performed to evaluate the blood vessel permeability. The animals were sacrificed 2 hours post sonications and lectin was injected to label the vasculature. The isodose lines from the mouse-like phantom experiment agreed well with the projected values. BBB was successfully opened after FUS sonications on both sides of the brain (with or without radiation) as evidenced by contrast-enhanced MRI. In the acute group, higher gadolinium concentration (p = 0.072) was observed on the RT positive side. MR dynamic imaging revealed a similar trend in which RT positive side had a higher (p = 0.158) vascular permeability. Although the mean vessel area in the sonicated area remained similar between the two sides, significantly less lectin labeling (blood vessel density) was observed on the RT positive hemisphere (p = 0.003) in the acute group. These observations can be attributed to increased vasculature vulnerability immediately post SRS (within 2 days) resulting in more profound BBB opening. Interestingly, this difference diminished one month later in the chronic group, where FUS elicited comparable BBB openings and minimal vascular density changes in the CPu on both sides. Localized BBB opening can be successfully induced by FUS in previously stereotactically irradiated normal brains. BBB openings were evident in both acute and chronic settings with potentially more profound outcomes if FUS is performed within 2 days of RT treatment.

Biologic Subtypes as a Predictor of Local Control of Breast Cancer Brain Metastases after Stereotactic Radiosurgery

Brain metastases (BM) are diagnosed in approximately 15% of breast cancer (BC) patients. The likelihood of developing BM varies according to the biologic subtype of BC, with up to 50% of advanced stage Her2 positive and triple negative (TN) patients likely to develop BM. Biologic subtype is predictive of loco-regional recurrence following breast conserving therapy and/or mastectomy with the highest risk in the TN subtype. Stereotactic radiosurgery (SRS) is an integral component of treatment for limited BM. We sought to determine whether biologic subtype is predictive of local control (LC) of BM treated with SRS. We conducted a retrospective analysis of patients with BC associated BM treated with SRS at our institution from 2004-2017. Patients were included if pathology data and ER/PR/Her2 status were recorded. Patients were subdivided into three biologic subtypes: ER+/Her2- (“Luminal”), Her2+ and TN (“Basal”). MRI’s were performed at 3 month intervals post SRS. Kaplan Meier method was used to estimate the overall survival (OS) distribution defined as time from first SRS to death or last follow-up with comparison among biologic subtypes made by the log-rank test. Cox proportional hazard model was used to analyze association of local failure (LF) with biologic subtypes and to estimate hazard ratios (HR) (95% CI). A total of 36 patients (66 treatment sites, median of 2 lesions per patient) were included with a median follow up of 14.3 months. The median age of the cohort (all female) was 57 years. 69.7% of lesions were supratentorial. Mean lesion size was 0.73 cc (SD 5.1 cc) and mean dose was 18.25Gy (SD 1.7Gy). 6 patients were treated with whole brain radiation therapy (WBRT) after SRS with 4 patients remaining controlled at the SRS site at the time of WBRT. 2 patients received WBRT after recurrence at the site of SRS with concomitant failure at other sites in the brain (one Her2+, one TN). 9 patients underwent resection prior to SRS. The 12 month LC and OS rates for the entire cohort were 76% and 89%, respectively (See table below for different subtypes). There was no statistically significant difference in LC and OS among the biologic subtypes. Hazard ratios for local failure and mortality comparing different subtypes were not statistically significant, although a borderline significant association was observed with increased risk of local failure of the basal vs. Her2+ subtypes (HR = 3.5, p = 0.06, CI = 0.9-12.8). In this study, biologic subtype did not impact LC of BC associated BM treated with SRS. A borderline significant association with increased LF was observed for TN patients. Consideration of treatment intensification or altered fractionation to improve LC may be indicated for the TN subtype, pending validation in a larger dataset with longer follow-up.Abstract TU_10_3413: Table 1Biologic subtypesN (%)12 Month LC (%)12 Month OS (%)TN “Basal”11 (30.56%)59.4%80%Her2+13 (36.11%)86.0%92%ER+/Her2- “Luminal”12 (33.33%)82.9%91% Open table in a new tab

P3.CR-18 Central Nervous System Activity of Checkpoint Inihibitor in Non-Small-Cell Lung Cancer

Data regarding the activity of atezolizumab in the central nervous system (CNS) and its toxicity in combination with radiotherapy in patients with brain metastases (BM) of non-small cell lung cancer (NSCLC) is limited. Most clinical trials adopt strict selection criteria excluding patients with active BM and the blood–brain barrier (BBB) seems responsible for limiting the distribution of agents into the CNS. Here we present a 51 yr-old female patient with the diagnosis of stage IV NSCLC with an isolated BM (T3 N3 M1b, non-mutated EGFR). She received stereotactic radiosurgery (SRS) to the isolated brain lesion (figure 1) and sequential palliative first-line systemic monotherapy with atezolizumab (1200mg/dose q21d, 15 doses). Control brain MRI 3-month post SRS showed a new lesion in the thalamus (figure 2) while on atezolizumab. Medical team decided to continue systemic treatment and the repeat brain MRI demonstrated complete response to the new lesion. Additionally, treated lesion showed evidence of pseudoprogression. While on follow-up, the patient continued to respond in the brain for 14 months. This case reports activity of atezolizumab in CNS. The advent of immunotherapy in the therapeutic algorithm of NSCLC has raised the need to identify patients that could mostly benefit from checkpoint inhibitors.

Hypopituitarism After Single-Fraction Pituitary Adenoma Radiosurgery: Dosimetric Analysis Based on Patients Treated Using Contemporary Techniques

Toanalyze factors associated with post-stereotactic radiosurgery (SRS) hypopituitarism among radiation-naïve patients with pituitary adenomas who underwent single-fraction SRS between 2007 and2014. This was a retrospective review of 97 patients having single-fraction SRS from 2007 until 2014. Eligible patients had no history of prior radiation, normal age- and sex-specific pituitary function before SRS, and at least 24months of endocrine follow-up. Forty patients (41%) had hormone-secreting tumors; 57 patients had nonsecreting tumors (59%). The median prescription isodose volume was 2.8cm3 (interquartile range [IQR], 1.3-4.7); the median tumor margin dose was 20Gy (IQR, 15-25Gy). The median follow-up after SRS was 48months (IQR, 34-68months). Twenty-seven patients (28%) developed pituitary insufficiency at a median of 22months (IQR, 12-36months) after SRS. The rate of new endocrine deficits was 17% at 2years (95% confidence interval [CI] 10%-25%) and 31% at 5years (95% CI 20%-42%). Male sex (hazard ratio [HR] 2.38, 95% CI 1.05-5.26, P=.04), smaller pituitary gland volume (HR 0.99, 95% CI 0.99-0.99, P=.01), and higher mean pituitary gland dose (HR 1.31, 95% CI 1.16-1.47, P<.001) were associated with post-SRS hypopituitarism in multivariable analysis. The rate of hypopituitarism for patients with a mean gland dose of <11.0Gy at 2years was 2% (95% CI 0%-4%) and at 5years was 5% (95% CI 0%-11%), whereas rate of hypopituitarism for patients with a mean gland dose of ≥11.0Gy at 2years was 31% (95% CI 17%-43%) and at 5years was 51% (95% CI 34%-65%). Hypopituitarism after pituitary adenoma SRS increases in a time- and dose-dependent manner. Reducing the radiation exposure to the identifiable gland to a mean dose < 11.0Gy whenever feasible may lower the incidence of new hormonal deficits after pituitary adenoma SRS.

Does Immunotherapy Influence the Risk of Developing Radiation Necrosis After Radiosurgery of Brain Metastases?

The risk of radiation necrosis (RN) in patients with brain metastases (BM) treated with stereotactic radiosurgery (SRS) can be estimated on the volume of normal brain receiving 12 Gy (V12). Reference RN risks at 1 year for V12 <3.3 cm3 and 3.3-5.9 cm3 are 0% and 16%, respectively. It is unclear if immunotherapy (IT), now commonly used in the treatment of various malignancies, modifies the risk of RN with long term follow-up. We have correlated V12 with radiographic appearance of RN for BMs treated with SRS +/- IT. We reviewed 71 patients harboring 112 brain lesions from breast cancer (n = 9), non-small cell lung cancer (n = 28) and melanoma (n = 34) primaries. In 63 of the lesions (56%), immunotherapy was administered during the course of the patient’s treatment, while 49 lesions (44%) only received SRS. The median SRS dose was 21 Gy (range 15-24 Gy) delivered on a Linac-based platform with a 6D robotic table. The median dosimetric data (and range) for all lesions were as follows: V12 2.0 cc (0.45-18.64), GTV 0.25 cc (0.01-4.20 cc), PTV 0.51 cc (0.07-8.55 cc), conformity index CI 1.56 (1.0-2.16), prescription IDL 79.4% (69.4-99.2%) with 99.3% (88.4-100%) coverage of the PTV. Patients were followed for a median follow-up of 13.1 months with serial MRIs and RN assessments were rendered based on multiparametric MRIs with perfusion techniques; in 6/16 cases RN was pathologically proven with surgical biopsy. The radiographic incidence of RN was then correlated with dosimetric factors and the infusion (or lack thereof) of IT. Statistical analysis was performed using Kaplan Meier analysis and Fischer’s exact chi-square test. The overall incidence of radiographic RN was 14% (16/112), peaking at a median time of 14.9 month post SRS (range 8.1-71.7 months). However only 5/16 patients had symptomatic RN necessitating treatment with bevacizumab, surgery and/or pentoxifylline/vitamin E. The 1-, 2-, and 3-year rates of RN were 2.7%, 11.6%, and 13.4%, respectively, for all patients. The risk of RN correlated with V12; lesions with V12 <5.09 cc had a significantly lower risk of RN at 3 years compared to V12 ≥5.09 cc (10% vs 32%, respectively, P = 0.044). The percentage of patients that developed RN in the presence of IT was higher (12/63, 19%) than in the control group (4/49, 8%) but this did not reach statistical significance. Finally, the V12 threshold for patients receiving IT was 5.39 cc; a V12 <5.39 cc was associated with 12.5% risk of RN at 3 years vs 46% for V12 ≥ 5.39 cc. Based on this retrospective analysis, the overall incidence of RN appears to be higher in the IT group (19% vs 8%), but this could be due to longer survival secondary to systemic response to immunotherapy. As a results, the rate of RN based on Kaplan-Meier estimates continues to increase at 3-years post SRS. This information may be useful to clinicians when trying to distinguish late recurrence vs late radiation necrosis in long-term survivors receiving SRS and immunotherapy.

Early Posttreatment Assessment of MRI Perfusion Biomarkers Can Predict Long-Term Response of NSCLC Brain Metastases to SRS: A Longitudinal Analysis

Optimizing patient selection and identification of patients who may not respond to brain metastasis treatment is critical in personalizing therapy. MRI-based imaging biomarkers using perfusion MRI (pMRI) may satisfy this crucial unmet need. The aim of this study is to identify early post-treatment perfusion MRI changes and correlate to long-term outcomes after treatment of lung cancer brain metastases (BM) with stereotactic radiosurgery (SRS). Pre-treatment and post-treatment pMRI within 8 weeks was obtained in patients treated with SRS for an intact non-small cell lung cancer (NSCLC) brain metastasis. Time-dependent leakage (Ktrans), blood plasma volume (Vp), and extravascular extracellular volume (Ve) were calculated on regions of interest delineated on pMRI. Patients were followed longitudinally with serial MRI until death, progression, or intervention (whole brain radiation or surgery). Genotyping was performed on a biopsy of each patient’s primary or metastatic lesion with the Sequenom MassARRAY system. We hypothesize that early post SRS quantitative analysis of perfusion imaging can predict whether or not lung cancer BMs will ultimately respond or fail treatment Fifty-three lesions were included from 41 total patients. All lesions were treated with SRS with median dose 21 Gy (range = 18-22 Gy). Median follow-up after treatment was 11 months. Actuarial local control at 1 year was 85%. Univariate analysis showed that there is a significant difference in post-RT Ktrans standard deviation (STD) between patients with progressive disease (PD) (mean = 0.0317) and without progressive disease (mean = 0.0219) (P = 0.032). We additionally determined a cutoff value for predicting PD and non-PD using Ktrans STD. ROC curve analysis resulted in the area under the curve (AUC) = 0.73, sensitivity = 89%, and specificity = 67%, with a cutoff value 0.017. Pre-treatment Ve was significantly lower in KRAS mutants vs nonmutants (P = 0.0428). Using a median value of 0.1334 as a cutoff, Ve was able to significantly predict KRAS mutant status compared to the Sequenom reference standard (AUC = 0.71, sensitivity = 81.8%, specificity = 61.9%). Post-RT Ktrans STD may be used as an early post treatment (within 8 weeks) imaging biomarker to help predict long-term response of lung BMs to SRS. This can help identify patients who will ultimately fail SRS in a timelier manner and allow for early adjustment in treatment approach. We determine that Ve from pMRI can identify KRAS mutation status with >80% sensitivity. This study demonstrates promising evidence of the utility of early assessment of post-treatment imaging biomarkers. These data are to be validated in another cohort of patients and also in other histologies.

Temporal evolution of perfusion parameters in brain metastases treated with stereotactic radiosurgery: comparison of intravoxel incoherent motion and dynamic contrast enhanced MRI

Intravoxel incoherent motion (IVIM) is a magnetic resonance imaging (MRI) technique that is seeing increasing use in neuro-oncology and offers an alternative to contrast-enhanced perfusion techniques for evaluation of tumor blood volume after stereotactic radiosurgery (SRS). To date, IVIM has not been validated against contrast enhanced techniques for brain metastases after SRS. In the present study, we measure blood volume for 20 brain metastases (15 patients) at baseline, 1 week and 1 month after SRS using IVIM and dynamic contrast enhanced (DCE)-MRI. Correlation between blood volume measurements made with IVIM and DCE-MRI show poor correlation at baseline, 1 week, and 1 month post SRS (r = 0.33, 0.14 and 0.30 respectively). At 1 week after treatment, no significant change in tumor blood volume was found using IVIM or DCE-MRI (p = 0.81 and 0.41 respectively). At 1 month, DCE-MRI showed a significant decrease in blood volume (p = 0.0002). IVIM, on the other hand, demonstrated the opposite effect and showed a significant increase in blood volume at 1 month (p = 0.03). The results of this study indicate that blood volume measured with IVIM and DCE-MRI are not equivalent. While this may relate to differences in the type of perfusion information each technique is providing, it could also reflect a limitation of tumor blood volume measurements made with IVIM after SRS. IVIM measurements of tumor blood volume in the month after SRS should therefore be interpreted with caution.

Assessment of stereotactic radiosurgery treatment response for brain metastases using MRI based diffusion index

IntroductionTo investigate the clinical predictive values of the apparent diffusion coefficient (ADC) as a biomarker in radiation response of brain metastases. MethodForty-one patients with brain metastases treated with stereotactic radiosurgery (SRS) were imaged at baseline, one month post SRS, and six months post SRS using diffusion weighted MRI. The mean of ADC for metastases and tumor volume was calculated. A diffusion index (DI) was generated using the sum of 1/ADC among all the voxels in a tumor. Tumor response status was determined by lesion volume measured at six month post-SRS, or the last available follow-up MRI. Logistic regression analysis was used to account for factors associated with tumor response at baseline and one month post SRS. ResultsA higher ADC mean distinguished responders from non-responders only at six month post SRS (p<0.05). However, a lower DI distinguished a responder from non-responders on the baseline, one month post SRS and six months post SRS, indicating better diagnostic performance of the DI with regard to a non-invasive biomarker in monitoring SRS treatment response. A multivariate logistic regression analysis identified the DI as a predictor of tumor response at baseline and one month post SRS (p=0.002 and p=0.001, respectively). However, logistic regression analysis identified the ADC mean as a predictor of tumor response only at six months post SRS (p=0.019). ConclusionOur results support the hypothesis that ADC and tumor volume generated DT at baseline, one month and six months post SRS may be a promising biomarker predicting brain metastases’ response. Specifically, a lower DI at baseline and one month distinguished responders from non-responders.