Leptomeningeal Progression Research Articles

Comparative analysis of combined treatment methods for patients with single brain lesions

To evaluate own experience of preoperative radiosurgery followed by surgical resection (RS+S) of metastasis regarding local control, leptomeningeal progression, surgical and radiation-induced complications; to compare treatment outcomes with surgical resection and subsequent radiotherapy (S+SRT). Retrospective study included 66 patients with solitary brain metastasis. Two groups of patients were distinguished: group 1 (n=34) - postoperative irradiation, group 2 (n=32) - preoperative irradiation. The median age was 49.5 years (range 36-75). Local 3-, 6- and 12-month control among patients with postoperative irradiation was 88.2%, 79.4% and 42.9%, in the group of preoperative irradiation - 100%, 93.3% and 66.7%, respectively (p=0.021). Leptomeningeal progression developed in 11 patients (8 and 3 ones, respectively). The one-year survival rate was 73.5% and 84.4%, respectively (p=0.33). Long-term surgical and radiation-induced complications occurred in 12 (18.2%) patients. Preoperative radiosurgery with subsequent resection provides higher local control and lower incidence of leptomeningeal progression in patients with single brain metastases.

P03.04.B EFFICACY OF NEOADJUVANT STEREOTACTIC RADIOSURGERY IN SOLID CANCER BRAIN METASTASES: A SYSTEMATIC REVIEW AND METANALYSIS

Abstract BACKGROUND Neoadjuvant stereotactic radiosurgery (NaSRS) is a novel strategy for brain metastasis (BMs) management, promising to achieve lower rates of toxicity, leptomeningeal progression, and high percentage of local control. MATERIAL AND METHODS An extensive systematic literature review and meta-analysis of proportions were performed. A total of 8 papers were eligible for final analysis after searching MEDLINE via PubMed, Web-of-science, Cochrane Wiley, and Embase databases. RESULTS A total of 551 patients undergoing NaSRS to treat 583 lesions were included. Median age was 57.7 [56.5-61.1] years, with a median tumor volume of 9.1 [8.3-14.4] cc. The most frequent histology was non-small cell lung cancer (40.7%), followed by breast (20.5%), and melanoma (15.6%). Most of the lesions had a supratentorial location (77.6%). Treatment parameters were homogenous across the studies and showed a median dose of 16 [14.9-20.5] Gy at 80% isodose.Surgery was performed after a median of 1 [1-2] day and achieved gross-total extent in 92.4% of cases. Median follow-up was 12.9 [9.9-18.9] months.Pooled proportion of patients showed an overall mortality rate of 50% (CI 37-63; I2=87%), local failure of 15% (CI 12-20%; I2=12.1%) and a rate of distant brain failure of 42% (CI 35-49%; I2=55.3%). The median local progression-free survival (PFS) was 10.4 months, while median survival without distant failure was 8 months.Rate of radionecrosis (RN) and leptomeningeal dissemination (LMD) were 4% respectively. Median overall survival (OS) was 14.4 months. CONCLUSION Existing data suggest that NaSRS achieves good local control on BMs with relatively lower dose than adjuvant treatment and low incidence of RN and other complications. These results need to be confirmed by larger studies with higher level of evidence to address formal comparisons with other regimens.

SDPS-46 CENTRAL NERVOUS SYSTEM (CNS)-RELATED MORTALITY IN PATIENTS WITH HER2-POSITIVE METASTATIC BREAST CANCER

Abstract BACKGROUND Approximately one third of patients with HER2-positive (+) metastatic breast cancer (MBC) develop brain metastases (BrM). In light of discordant control of extracranial metastases (ECM) versus BrM for standard-of-care antibody-based therapies for these patients, and the frequent use of radiation and surgical therapies particularly for patients with CNS-centric disease, we sought to determine the proportion of HER2+ MBC patients with BrM who suffered neurologic death in the CNS-only and intracranial-plus-extracranial metastasis settings, and the clinical risk factors of such. METHODS Clinicopathologic data for HER2+ MBC patients with BrM treated at an NCI-designated Comprehensive Cancer Center between August 2010 and April 2022 were collected, including disease burden at cancer presentation, the timing and pattern of BrM and ECM, and neurologic complications. CNS-related death was defined as any mortality caused by BrM, leptomeningeal progression, or BrM-associated complication. RESULTS 275 patients were included. 58/275 (21%) presented with CNS as only and first site of metastasis; the remaining (217/275, 79%) developed BrM synchronously with or subsequent to ECM. 125/275 (45%) had de novo MBC, and nearly all patients were treated with CNS therapies (254/275 [92%] received at least one radiation treatment and 88/275 [32%] underwent BrM resection). After a median follow-up of 2.5 years, 193/275 (70%) patients died, of whom 105 (54%) of CNS-related cause. The median OS was 3 years (95% CI=1.47-5.14) and 2 years (95% CI=1.74-2.34) for patients with CNS-only versus BrM concurrently/subsequent to ECM, respectively (p=0.059). The cumulative incidence of CNS-related death at 5 years was 40% (95%CI=33-46%). CONCLUSIONS Almost half of patients with HER2+ BrM suffered neurologic death. Patients with CNS-only disease had better OS than patients with synchronous/preceding ECM, suggesting that overall burden of disease affects mortality. With high rates of durable extracranial control with HER2-directed therapies, new strategies to prevent and eradicate CNS disease are urgent.

INTRAMET: Results of a prospective, single-arm, open-label phase II trial of intraoperative radiotherapy after resection of brain metastases.

2031 Background: Brain metastases (BM) occur in about 10-20% of all cancer patients during the course of their disease. The current standard of care for lesions that cause mass effects is neurosurgical resection followed by adjuvant radiotherapy of the resection cavity. We here report efficacy data of the INTRAMET trial (NCT03226483), which tested intraoperative radiation therapy (IORT) using a high-dose, single fraction to the tumor bed. Methods: Patients ≥18 years and Karnofsky (KPS) ≥50 with suspected BM were included. A frozen-section confirmed metastasis and technically applicable IORT were mandatory. Exclusion criteria were surgical or MRI contraindications, meningeal infiltration, or IORT Dmax doses exceeding 8 Gy at risk structures. IORT was performed after tumor resection in the operating room with a mobile low energy X-ray device emitting a spherical radiation field with a sharp dose gradient. A dose of 30Gy was prescribed to the surface of the resection cavity. The primary endpoint was cumulative local control rate (LCR). Secondary endpoints were incidence of other (distant) BM (RCR), overall survival (OS), time to the next systemic treatment and adverse event rates (AE). Results: In total, 35 of the originally 50 planned patients were recruited. After a pre-specified interim analysis, the study was prematurely halted due to good safety and outcomes profiles. 54.3% were men, the mean age was 64 (45-85) years. The mean follow-up was 25.7 (0.8-64.5) month. At the time of this report, 51.4% of patients were alive. 68.6% primary histologies corresponded to lung primaries. The LCR was 94.3% (95% CI 82.9-98.8%) and RCR was 57.1% (95% CI 40.7-72.4%,). 8.6% patients or lesions developed out of field leptomeningeal progression. The mean OS was 37.4 months (95% CI 28.9-46.9). For those patients undergoing salvage whole brain RT (WBRT), the median initiation time was 147 (20-601) days with a significant survival benefit for those not requiring WBRT (p = 0.027). No significant survival differences were identified according to primary histology (p = 0.618), immunotherapy (p = 0.928), seizures at baseline (p = 0.169), KPS (p = 0.056) or radionecrosis appearance (p = 0.214). Mean time to next systemic treatment was 45.0 (95% CI 35.1-54.8) days. No G4-5 AE related to IORT occurred. The overall radionecrosis (RN) rate was 20% (n = 5 G1, n = 1 G2, n = 1 G3). No risk factors could be associated to RN. Post-surgical seizures were present in 28.6% patients, with eloquent localizations as the only predictive factor (p = 0.008). Conclusions: IORT for resected BM yields excellent local control rates and has a toxicity profile similar to those reported in post-operative SRS trials. By easing a prompt initiation of subsequent systemic treatments, IORT might shorten the overall treatment courses of many cancer patients. Clinical trial information: NCT03226483 .

Results of hypofractionated stereotactic radiotherapy for resected and intact large brain metastases

To compare the results of post-resection or isolated HF-SRT in patients with LBMs. A prospective study included 115 patients with 129 intact LBMs and 133 patients with 149 resected LBMs who underwent HF-SRT. Median baseline focal size was 22.5 and 28 mm, median target volume - 8.3 and 23.7 cm3, respectively. Median follow-up was 13.9 months, median overall survival - 19.1 months. After 12 months, local recurrences developed in 17 and 31% of patients, respectively (p=0.0078). Local recurrence after 12 months developed in 23% of patients with residual tumor in postoperative cavity compared to 16% of patients after total resection (p=0.0073). After 12 months, incidence of leptomeningeal progression was 27 and 11%, respectively (p=0.033), incidence of symptomatic radiation-induced necrosis - 4 and 23%, respectively (p=0.0006). Post-resection HF-SRT demonstrated better local control and less severe symptomatic radiation-induced necrosis compared to patients with intact LBMs. Incidence of leptomeningeal progression is significantly higher after resection of LBMs.

BIOM-01. GENOMIC ALTERATIONS IN ALK FUSION-POSITIVE NON-SMALL CELL LUNG CANCER (NSCLC) PATIENTS WITH BRAIN METASTASES

Abstract PURPOSE Brain metastases (BM) are common in patients with anaplastic lymphoma kinase fusion-positive NSCLC (ALK+ NSCLC). We characterized the genomics of patients with ALK+ NSCLC BM treated with alectinib with or without local therapy to identify correlations of co-mutations with CNS-specific outcomes. METHODS We retrospectively reviewed ALK+ NSCLC patients with BMs who received alectinib for the definitive treatment of BM from 1/2012 and 5/2021. Genomic characteristics of 27 specimens from 27 patients were assessed with MSK-IMPACT, a 505-gene next-generation sequencing (NGS)-based tumor sequencing assay with >700x coverage. Intracranial progression-free survival (iPFS) and overall survival (OS) from BM diagnosis were analyzed using standard statistical methods. RESULTS The median age at BM diagnosis was 57 years (range 25-83). Median iPFS was 1.94 years (95%CI: 1.58- not reached) and median OS was 6.08 years (95%CI: 1.7– not reached). All patients received alectinib for the treatment of brain metastases (78% as 1st-line TKI) and 22% received local therapy. The most frequently altered co- alterations were CDKN2A (48%), TP53 (22%), MAP2K4 (15%), SMARCA4 (11%), CREBBP (11%), and ATM (7%). CDKN2A alterations were more common in patients with intracranial progression (64% vs 31%). MAP2K4 alterations were enriched in metastatic samples (p = 0.028, q = 0.113). SMARCA4 co- alterations were associated with inferior OS (HR: 8.76, 95%CI = 1.74-44.2, p= 0.009) and trended toward association with iPFS but was insignificant (HR: 3.25, 95%CI = 0.83-12.4, p=0.089). We identified missense ALK alterations in CSF from a patient who acquired resistance to alectinib and had leptomeningeal progression (G1269A, I1171T, L1108P). CONCLUSION This study is the first analysis of CNS specific outcomes with detailed genomic annotation for ALK+ NSCLC patients who received definitive alectinib for BM. Further investigation into the role of CNS-penetrant TKIs and the genomic alterations predictive of CNS failure are needed.

CNSC-38. EGFR AMPLIFICATION IS A PUTATIVE RESISTANCE MECHANISM FOR NSCLC-LM PATIENTS WITH TKI THERAPY AND IS ASSOCIATED WITH POOR OUTCOME

Abstract BACKGROUND Leptomeningeal metastases (LM) have become increasingly common in NSCLC patients who harbor epidermal growth factor receptor (EGFR) mutation treated with EGFR-TKI and are correlated with inferior prognosis. Evidence in prior research demonstrated that EGFR amplification was more likely presented in advanced clinical stages and was associated with worse survival. However, whether the EGFR amplification is a prognostic marker in NSCLC-LM is still inconclusive. METHODS This study enrolled patients diagnosed with NSCLC-LM from June 2019 to September 2021, who had received previous EGFR-TKI at Guangdong Sanjiu Brain Hospital. Cerebrospinal fluid (CSF) samples were collected and subjected to targeted next-generation sequencing (NGS) of 168 cancer-related genes. Clinical characteristics and overall survival (OS) were compared in patients with and without EGFR amplification. RESULTS This study enrolled 53 NSCLC-LM patients, all of whom had EGFR mutations. TP53 and EGFR amplification are the two most frequent mutations in the study cohort, presenting 72% (38 of 53) and 40% (21 of 53), respectively. The rate of EGFR amplification was much higher at the time of leptomeningeal progression than at initial diagnosis (p< 0.01). Karnoskfy performance status was poorer (p=0.021), and CSF pressure was higher (p=0.0067) in patients with than without EGFR amplification. A multivariable Cox proportional hazard regression model showed that EGFR amplification was an independent prognostic factor for poorer OS (8.3 vs 15 months; p=0.017). Median OS was shorter in NSCLC-LM patients with mutated than wild-type TP53, but the differences was not statistically significant (10 vs 17.3 months, p=0.184). CONCLUSIONS EGFR gene amplification could be a potential resistance mechanism to EGFR-TKI failure in NSCLC-LM and is associated with inferior clinical outcomes.

Craniospinal Reirradiation Using Proton Beam Radiotherapy in the Treatment of Pediatric Recurrent Medulloblastoma

<h3>Purpose/Objective(s)</h3> Craniospinal irradiation (CSI) is an essential part of medulloblastoma treatment even without any evidence of leptomeningeal metastasis. In case of recurrence, it is important to assess several factors to consider reirradiation course. These factors include prior radiation doses, time interval between radiotherapy courses and cumulative dose. In cases of leptomeningeal dissemination reirradiation of the entire craniospinal axis is needed, but may be accompanied by severe toxicity. Proton therapy may reduce toxicity without reducing treatment doses. <h3>Materials/Methods</h3> Proton beam craniospinal reirradiation (CSRI) course was conducted to 18 patients with recurrent medulloblastoma. Median dose to craniospinal axis during primary treatment was 30.6 Gy (24-36). Sixteen patients had additional local boost to recurrent and metastatic sites. In all cases we've made a dose gradient in the border zone to avoid irradiation of vertebral bodies. Eight patients had High Dose Chemotherapy (HDCT) before CSRI. The minimal interval between HDCT and CSRI was 2 months. Median dose at primary CSI was 23.7 Gy (23.4-36). Seven patients had reirradiation previously, two of them had CSRI with total dose of 18 Gy. <h3>Results</h3> Median time between two CSI courses was 33 months (19-126). Median follow-up period was 9 months (2-18). Six patients had disease progression, five of them died because of progression during the next 1.5 years after CSRI. Average time to progression was 13.3 ± 1.6 months (95% CI 10.1-16.4). During treatment we observed strong evidence of progressing thrombocytopenia (avg = 95 * 10⁹) with minimal values occurring during third week (p < 0.001). There was a non-significant decrease of hemoglobin and neutrophiles level (p_hemoglobin=0.4, p_neutrophiles=0.07). One patient had severe bone marrow hypoplasia during the treatment which required blood and platelet transfusion and neutrophil stimulation. The rest of the patients' hematological toxicity resolved without medical intervention. Only one patient had interruption during treatment because of COVID-19 infection. All the patients except one received chemotherapy after CSRI. None of them had any evidence of radiation necrosis or other radiological changes during the whole follow-up period. <h3>Conclusion</h3> CSRI is mostly used for medulloblastoma patients with refractory disease and leptomeningeal progression. Proton therapy may help to deliver curative doses with limited hematological toxicity and to minimize or completely avoid gastrointestinal toxicity. Probably it may lead to long-term disease control for such group of patients.

EGFR amplification is a putative resistance mechanism for NSCLC-LM patients with TKI therapy and is associated with poor outcome.

BackgroundLeptomeningeal metastases (LM) have become increasingly common in non-small cell lung cancer (NSCLC) patients who harbor epidermal growth factor receptor (EGFR) mutation treated with EGFR-TKI and are correlated with inferior prognosis. Evidence in prior research demonstrated that EGFR amplification was more likely presented in advanced clinical stages and was associated with worse survival. However, whether EGFR amplification is a prognostic marker in NSCLC–LM is still inconclusive.MethodsThis study enrolled patients diagnosed with NSCLC–LM from June 2019 to September 2021 and who had received previous EGFR-TKI at Guangdong Sanjiu Brain Hospital. Cerebrospinal fluid (CSF) samples were collected and subjected to targeted next-generation sequencing of 168 cancer-related genes. Clinical characteristics and overall survival (OS) were compared in patients with and without EGFR amplification.ResultsThis study enrolled 53 NSCLC–LM patients, all of whom had EGFR mutations. TP53 and EGFR amplifications are the two most frequent mutations in the study cohort, presenting at 72% (38 of 53) and 40% (21 of 53), respectively. The rate of EGFR amplification was much higher at the time of leptomeningeal progression than at initial diagnosis (p < 0.01). Karnoskfy performance status was poorer (p = 0.021), and CSF pressure was higher (p = 0.0067) in patients with EGFR amplification than those without. A multivariable Cox proportional hazard regression model showed that EGFR amplification was an independent prognostic factor for poorer OS (8.3 vs. 15 months; p = 0.017). The median OS was shorter in NSCLC–LM patients with mutated TP53 than those with wild-type TP53, but the difference was not statistically significant (10 vs. 17.3 months, p = 0.184).Conclusions EGFR gene amplification could be a potential resistance mechanism to EGFR-TKI failure in NSCLC–LM and is associated with inferior clinical outcomes.

High-Dose Osimertinib for CNS Progression in EGFR+ NSCLC: A Multi-Institutional Experience

IntroductionThis multicenter review evaluated the efficacy and safety of osimertinib dose escalation for central nervous system (CNS) progression developing on osimertinib 80 mg in EGFR-mutant NSCLC. MethodsRetrospective review identified 105 patients from eight institutions with advanced EGFR-mutant NSCLC treated with osimertinib 160 mg daily between October 2013 and January 2020. Radiographic responses were clinically assessed, and Kaplan-Meier analyses were used. We defined CNS disease control as the interval from osimertinib 160 mg initiation to CNS progression or discontinuation of osimertinib 160 mg. ResultsAmong 105 patients treated with osimertinib 160 mg, 69 were escalated for CNS progression, including 24 treated with dose escalation alone (cohort A), 34 who received dose-escalated osimertinib plus concurrent chemotherapy and/or radiation (cohort B), and 11 who received osimertinib 160 mg without any prior 80 mg exposure. The median duration of CNS control was 3.8 months (95% confidence interval [CI], 1.7–5.8) in cohort A, 5.1 months (95% CI, 3.1–6.5) in cohort B, and 4.2 months (95% CI 1.6–not reached) in cohort C. Across all cohorts, the median duration of CNS control was 6.0 months (95% CI, 5.1–9.0) in isolated leptomeningeal progression (n = 27) and 3.3 months (95% CI, 1.0–3.1) among those with parenchymal-only metastases (n = 23). Patients on osimertinib 160 mg experienced no severe or unexpected side effects. ConclusionAmong patients with EGFR-mutant NSCLC experiencing CNS progression on osimertinib 80 mg daily, dose escalation to 160 mg provided modest benefit with CNS control lasting approximately 3 to 6 months and seemed more effective in patients with isolated leptomeningeal CNS progression.

Analyzing the Usage of Pencil-Beam Proton Radiotherapy for Patients With Atypical Teratoid Rhabdoid Tumor of Central Nervous System

Atypical teratoid rhabdoid tumor of the central nervous system is a rare tumor affecting very young children. Treatment of ATRT includes different regimens of radiotherapy depending on age, stage and treatment protocols. Proton therapy allows to reduce acute and late toxicity. This reduction (among other benefits) helps to finish irradiation without interruptions. It matters for kids receiving craniospinal irradiation as well as for kids receiving radiotherapy after high dose chemotherapy.We reviewed data of 30 patients with ATRT treated with pencil-beam proton therapy from 06.2018 to 12.2020. We analyzed treatment characteristics, outcomes and other factors that could be prognostic for progression free survival.Median age at the start of radiotherapy was 28 months. Patients received local radiotherapy (18 patients) and craniospinal irradiation (12 patients). Total dose in case of local radiotherapy was 54 Gy. A median craniospinal dose was 29.8 Gy (24-36 Gy). Four patients got re-irradiation course because of local or leptomeningeal progression. One of them received craniospinal re-irradiation. All the patients who were 5 years old and younger received daily general anesthesia (sedation) during radiotherapy course. Two of the patients did not finish course of radiotherapy because of distant progression. Median follow-up was 10 months. No interruptions were needed for any of the patients. Inability to achieve gross tumor resection (P = 0.01) and M3-stage (P = 0.03) were the most unfavorable prognostic factors for progression-free survival.This is one of the largest reports of using pencil-beam proton therapy for children with ATRT of CNS treated in one facility. Proton therapy was well tolerated even by patients receiving re-irradiation course. Further research should be conducted in order to get information about late toxicity and survival outcomes.

A Case of Isolated Leptomeningeal Progression in a Patient with NTRK Fusion + Uterine Sarcoma (1957)

A Case of Isolated Leptomeningeal Progression in a Patient with NTRK Fusion + Uterine Sarcoma (1957)

Trastuzumab Emtansine (T-DM1) and stereotactic radiation in the management of HER2+ breast cancer brain metastases

BackgroundDue to recent concerns about the toxicity of trastuzumab emtansine (T-DM1) with stereotactic radiation, we assessed our institutional outcomes treating HER2-positive breast cancer brain metastases (BCBM) with T-DM1 and stereotactic radiation.MethodsThis is a single institution series of 16 patients with HER2-positive breast cancer who underwent 18 stereotactic sessions to 40 BCBM from 2013 to 2019 with T-DM1 delivered within 6 months. The Kaplan-Meier method was used to calculate overall survival (OS), local control (LC), distant intracranial control (DIC), and systemic progression-free survival (sPFS) from the date of SRS. A neuro-radiologist independently reviewed follow-up imaging.ResultsOne patient had invasive lobular carcinoma, and 15 patients had invasive ductal carcinoma. All cases were HER2-positive, while 10 were hormone receptor (HR) positive. Twenty-four lesions were treated with stereotactic radiosurgery (SRS) to a median dose of 21 Gy (14–24 Gy). Sixteen lesions were treated with fractionated stereotactic radiation (FSRT) with a median dose of 25 Gy (20-30Gy) delivered in 3 to 5 fractions. Stereotactic radiation was delivered concurrently with T-DM1 in 19 lesions (48%). Median follow up time was 13.2 months from stereotactic radiation. The 1-year LC, DIC, sPFS, and OS were 75, 50, 30, and 67%, respectively. There was 1 case of leptomeningeal progression and 1 case (3%) of symptomatic radionecrosis.ConclusionsWe demonstrate that stereotactic radiation and T-DM1 is well-tolerated and effective for patients with HER2-positive BCBM. An increased risk for symptomatic radiation necrosis was not noted in our series.

Clinical Outcomes of Breast Cancer Brain Metastases Treated with Stereotactic Radiation and Capecitabine

Capecitabine is often used as a radiosensitizer in combination with external beam radiation therapy. However, relatively little is known about the safety and efficacy of combined treatment with capecitabine and stereotactic radiation for the treatment of brain metastases. In this study, we assess our institutional experience in treating breast cancer brain metastases (BCBM) with stereotactic radiation and capecitabine. This is a retrospective study of 23 consecutive patients with BCBM who received 31 stereotactic radiation sessions to 90 lesions from January 2005 to November 2019 with capecitabine delivered within 6 months of stereotactic radiation. Patient, tumor, and treatment characteristics were retrieved from the clinical chart. The Kruskal-Wallis and Pearson’s chi-square tests were used to test differences between cohorts as appropriate. The Kaplan-Meier method was used to calculated overall survival (OS), local control (LC), and distant intracranial control (DIC) from the date of SRS. There were 21 patients diagnosed with invasive ductal carcinoma, 1 patient with invasive lobular carcinoma, and 1 patient with metaplastic carcinoma. Nine, 10, and 9 patients had ER-positive, PR-positive, and HER2-positive disease, respectively. Fourteen patients were originally diagnosed with de novo metastatic breast cancer. Median follow up time from stereotactic radiation was 9 months (1.8 to 92.8). The median age at the time of stereotactic radiation was 51 years (40 to 75). The median number of brain metastases at presentation was 2 (1 to 12), and the initial treatment at the time of brain metastasis included stereotactic radiation (65.2%), WBRT (26.1%), and surgery with post-operative radiotherapy (8.7%). Ten lesions were treated with fractionated stereotactic radiation (FSRT) to a median prescription dose of 25 Gy (24 to 30 Gy) delivered in 3 to 5 fractions, and 80 lesions were treated with stereotactic radiosurgery (SRS) to a median prescription dose of 21 Gy (15 to 24 Gy) in a single fraction. The median planning treatment volume was 0.71cc (0.012 to 34cc). Of the 90 treated lesions, 40 (44.4%) were treated with concurrent capecitabine. The 1-year OS, LC, and DIC were 46%, 95%, and 32%, respectively. There was 1 case (1.1%) of symptomatic radiation necrosis. There were 2 cases of leptomeningeal progression. We demonstrate that stereotactic radiation and capecitabine is a well-tolerated treatment for BCBM with high LC and a low associated risk for symptomatic radiation necrosis.

Evaluation of First-line Radiosurgery vs Whole-Brain Radiotherapy for Small Cell Lung Cancer Brain Metastases

Although stereotactic radiosurgery (SRS) is preferred for limited brain metastases from most histologies, whole-brain radiotherapy (WBRT) has remained the standard of care for patients with small cell lung cancer. Data on SRS are limited. To characterize and compare first-line SRS outcomes (without prior WBRT or prophylactic cranial irradiation) with those of first-line WBRT. FIRE-SCLC (First-line Radiosurgery for Small-Cell Lung Cancer) was a multicenter cohort study that analyzed SRS outcomes from 28 centers and a single-arm trial and compared these data with outcomes from a first-line WBRT cohort. Data were collected from October 26, 2017, to August 15, 2019, and analyzed from August 16, 2019, to November 6, 2019. SRS and WBRT for small cell lung cancer brain metastases. Overall survival, time to central nervous system progression (TTCP), and central nervous system (CNS) progression-free survival (PFS) after SRS were evaluated and compared with WBRT outcomes, with adjustment for performance status, number of brain metastases, synchronicity, age, sex, and treatment year in multivariable and propensity score-matched analyses. In total, 710 patients (median [interquartile range] age, 68.5 [62-74] years; 531 men [74.8%]) who received SRS between 1994 and 2018 were analyzed. The median overall survival was 8.5 months, the median TTCP was 8.1 months, and the median CNS PFS was 5.0 months. When stratified by the number of brain metastases treated, the median overall survival was 11.0 months (95% CI, 8.9-13.4) for 1 lesion, 8.7 months (95% CI, 7.7-10.4) for 2 to 4 lesions, 8.0 months (95% CI, 6.4-9.6) for 5 to 10 lesions, and 5.5 months (95% CI, 4.3-7.6) for 11 or more lesions. Competing risk estimates were 7.0% (95% CI, 4.9%-9.2%) for local failures at 12 months and 41.6% (95% CI, 37.6%-45.7%) for distant CNS failures at 12 months. Leptomeningeal progression (46 of 425 patients [10.8%] with available data) and neurological mortality (80 of 647 patients [12.4%] with available data) were uncommon. On propensity score-matched analyses comparing SRS with WBRT, WBRT was associated with improved TTCP (hazard ratio, 0.38; 95% CI, 0.26-0.55; P < .001), without an improvement in overall survival (median, 6.5 months [95% CI, 5.5-8.0] for SRS vs 5.2 months [95% CI, 4.4-6.7] for WBRT; P = .003) or CNS PFS (median, 4.0 months for SRS vs 3.8 months for WBRT; P = .79). Multivariable analyses comparing SRS and WBRT, including subset analyses controlling for extracranial metastases and extracranial disease control status, demonstrated similar results. Results of this study suggest that the primary trade-offs associated with SRS without WBRT, including a shorter TTCP without a decrease in overall survival, are similar to those observed in settings in which SRS is already established.

Abstract A09: Cerebrospinal fluid (CSF) as a source of ctDNA for diagnosis and monitoring of pediatric patients with midline diffuse gliomas

Abstract Background: Midline diffuse gliomas are aggressive pediatric tumors, characterized molecularly by the presence of H3K27 mutation. These tumors are extremely malignant, with median overall survival of less than a year. Localized mainly in the pons or thalamic regions, they are not amenable to a gross total resection; however, a biopsy is needed for pathologic and molecular diagnosis. In this scenario, liquid biopsy is of utmost importance and has the potential to spare the risk of morbidity with surgical procedures, determine diagnosis and prognosis, as well as serve as a tool to identify potential targetable alterations. The objective of this pilot feasibility study is to explore the use of CSF ctDNA using ddPCR and next-generation sequencing tools to diagnose and monitor patients with midline diffuse gliomas. Methods: We extracted CSF ct DNA collected from external ventricular drain in two patients diagnosed with midline diffuse glioma (one thalamic glioblastoma, one DIPG). ctDNA was extracted from 2 mL of CSF, quantification was done by quibit, and samples were processed first with ddPCR for detection of H3K27M point mutation followed by comprehensive profiling by whole-exome sequencing using Illumina HiSeq2500 sequencer. We assessed the feasibility of identifying diagnostic point mutations, as well as potential evolution of the tumor molecular landscape at disease progression. Results: The first case, a thalamic GBM, had tissue available for molecular analysis; targeted RNA sequencing was performed on a clinical basis detecting H3K27M mutation (VAF=43%) and a PDGFRA N659K mutation (VAF=43%). ddPCR analysis of the CSF ctDNA identified the H3K27M mutation with VAF of 14%. Whole-exome sequencing (WES) confirmed the presence of H3K27M point mutation at VAF 11% and PDGFRA mutation at VAF 13%. This patient received radiation and chemotherapy but progressed rapidly and passed 7 months after the diagnosis. The second case, a DIPG diagnosed upon radiologicl and clinical criteria, was previously treated with conventional focal radiation and needed an EVD upon massive leptomeningeal progression with associated meningitis. CSF was then processed for ctDNA analysis. ddPCR confirmed the presence of H3K27M mutation with VAF of 48%. WES confirmed the presence of H3K27M mutation (VAF 14%) and additional TP53 R273C point mutation with VAF of 87%. This patient passed roughly 11 months post diagnosis. Conclusion: We have demonstrated the feasibility of detection of molecular markers in CSF through ddPCR and NGS platform to not only to assist in diagnosis of pediatric midline diffuse gliomas, but also to monitor these patients for additional/new molecular events that may be targetable. Building on these results, we are currently implementing a targeted sequencing panel to establish the specificity of this method in a large pediatric brain tumor cohort. Citation Format: Liana Nobre, Robert Siddaway, Monique Johnson, Scott Ryall, Michal Zapotocky, Julie Bennet, Uri Tabori, Cynthia Hawkins. Cerebrospinal fluid (CSF) as a source of ctDNA for diagnosis and monitoring of pediatric patients with midline diffuse gliomas [abstract]. In: Proceedings of the AACR Special Conference on Advances in Liquid Biopsies; Jan 13-16, 2020; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(11_Suppl):Abstract nr A09.

Management of advanced uni- or bilateral retinoblastoma with macroscopic optic nerve invasion.

Retinoblastoma with macroscopic optic nerve (ON) invasion depicted by imaging at diagnosis remains a major problem and carries a poor prognosis. We sought to describe the treatment and outcome of these high-risk patients. Retrospective mono-institutional clinical, radiological, and histological review of patients with uni- or bilateral retinoblastoma with obvious ON invasion, defined by radiological optic nerve enlargement (RONE) depicted by computed tomography scan or magnetic resonance imaging (MRI), was performed. Between 1997 and 2014, among the 936 patients with retinoblastoma treated at Institut Curie, 11 had detectable RONE. Retinoblastoma was unilateral in 10 and bilateral in one. Median age at diagnosis was 28 months (range, 11-96). ON enlargement extended to the orbital portion in three patients, to the optic canal in five, to the prechiasmatic portion in two, and to the optic chiasm in one. Nine patients received neoadjuvant chemotherapy and partial response was obtained in all. Enucleation was performed in 10/11 patients-by an anterior approach in three and by anterior and subfrontal approaches in seven. Three patients had a positive ON resection margin (2/3 after primary enucleation). All enucleated patients received adjuvant treatment (conventional chemotherapy: 10, high-dose chemotherapy: seven, radiotherapy: five). Leptomeningeal progression occurred in four patients. Seven are in first complete remission (median follow up: 8 years [3.5-19.4]). Neoadjuvant chemotherapy and microscopic complete resection have a pivotal role in the management of retinoblastoma with RONE. MRI is recommended for initial and pre-operative accurate staging. Surgery should be performed by neurosurgeons in case of posterior nerve invasion. Radiotherapy is required in case of incomplete resection.

OS5.4 Characteristics of adult diffuse H3K27M-mutant gliomas at recurrence

Abstract BACKGROUND Adult diffuse H3K27M-mutant gliomas are rare and associated with a poor prognosis but could benefit in the next future from specific therapeutic strategies. In this context, the aim of the present study was to describe the characteristics of these tumors at recurrence. MATERIAL AND METHODS We retrospectively analyzed the characteristics of a series of 27 adult diffuse H3K27M-mutant gliomas at recurrence RESULTS Median age at diagnosis was 37 years. Initial treatment consisted of temozolomide radiochemotherapy (n=17, 62%), radiotherapy alone (n=1, 4%), chemotherapy alone (n=4, 15%), wait and see (n=1, 4%) and palliative care (n=4, 15%). Median PFS and median OS were 11 and 22 months in the whole series and 15 and 29 months in the patients who were treated with temozolomide radiochemotherapy. The pattern of recurrence could be analyzed in 19 patients. Ten patients (50%) demonstrated a local recurrence, five patients a local and distant recurrence (25%), two patients only a distant recurrence (10%) and two patients a leptomeningeal progression (10%). At recurrence, 15 patients received an oncological treatment that consisted of an alkylating chemotherapy (n=5), a bevacizumab based chemotherapy regimen (n=9) and of radiotherapy (n=1). Median PFS and OS after first recurrence in these patients were 6 and 14 months, respectively. An activating FGFR1 mutation was identified in 4 out of the 9 patients in whom it was assessed. CONCLUSION At recurrence, adult diffuse H3K27M-mutant gliomas are associated with a high rate of distant locations. A subset of patients harbor targetable FGFR1 activating mutations.

Could the extreme conformality achieved with proton therapy in paranasal sinuses cancers accidentally results in a high rate of leptomeningeal progression?

See article on page 3736–3736, in this issue

Isolated leptomeningeal progression from sinonasal carcinomas: Implications for staging workup and treatment.

To evaluate the rate and risk factors of isolated leptomeningeal progression in sinonasal carcinomas. We retrospectively reviewed imaging and clinical records to determine progression patterns, and estimated rates using the Kaplan-Meier method. We evaluated risk factors using proportional hazard regression. We analyzed 120 patients who received adjuvant or primary radiotherapy for sinonasal carcinomas. Most patients had T4 disease (68%) and underwent surgery (84%) and chemotherapy (72%). Twenty-seven (23%) patients developed distant metastases (DM), including 20 (17%) with isolated DMs. Leptomeningeal progression was the most common site of isolated DMs (n = 9; 45%) with an average disease-free interval of 1.2 years (0.1-4.3 years). High-grade histology (P = 0.0003), intracranial invasion (P < 0.0001), and neuroendocrine histology (P = 0.06) were associated with increased risk. Isolated leptomeningeal progression is a common pattern of DM in advanced sinonasal carcinomas. We recommend adding cerebrospinal fluid cytology and contrast-enhanced spine MRI to routine staging evaluations for high-risk patients.