Shahiba Ogilvie

Evaluating Cancer of the Central Nervous System Through Next-Generation Sequencing of Cerebrospinal Fluid

Purpose Cancer spread to the central nervous system (CNS) often is diagnosed late and is unresponsive to therapy. Mechanisms of tumor dissemination and evolution within the CNS are largely unknown because of limited access to tumor tissue. Materials and Methods We sequenced 341 cancer-associated genes in cell-free DNA from cerebrospinal fluid (CSF) obtained through routine lumbar puncture in 53 patients with suspected or known CNS involvement by cancer. Results We detected high-confidence somatic alterations in 63% (20 of 32) of patients with CNS metastases of solid tumors, 50% (six of 12) of patients with primary brain tumors, and 0% (zero of nine) of patients without CNS involvement by cancer. Several patients with tumor progression in the CNS during therapy with inhibitors of oncogenic kinases harbored mutations in the kinase target or kinase bypass pathways. In patients with glioma, the most common malignant primary brain tumor in adults, examination of cell-free DNA uncovered patterns of tumor evolution, including temozolomide-associated mutations. Conclusion The study shows that CSF harbors clinically relevant genomic alterations in patients with CNS cancers and should be considered for liquid biopsies to monitor tumor evolution in the CNS.

Survival in patients treated for anaplastic meningioma.

OBJECT While most meningiomas are benign, 1%-3% display anaplastic features, with little current understanding regarding the molecular mechanisms underlying their formation. In a large single-center cohort, the authors tested the hypothesis that two distinct subtypes of anaplastic meningiomas, those that arise de novo and those that progress from lower grade tumors, exist and exhibit different clinical behavior. METHODS Pathology reports and clinical data of 37 patients treated between 1999 and 2012 for anaplastic meningioma at Memorial Sloan-Kettering Cancer Center (MSKCC) were retrospectively reviewed. Patients were divided into those whose tumors arose de novo and those whose tumors progressed from previously documented benign or atypical meningiomas. RESULTS Overall, the median age at diagnosis was 59 years and 57% of patients were female. Most patients (38%) underwent 2 craniotomies (range 1-5 surgeries) aimed at gross-total resection (GTR; 59%), which afforded better survival when compared with subtotal resection according to Kaplan-Meier estimates (median overall survival [OS] 3.2 vs 1.3 years, respectively; p = 0.04, log-rank test). Twenty-three patients (62%) presented with apparently de novo anaplastic meningiomas. Compared with patients whose tumors had progressed from a lower grade, those patients with de novo tumors were significantly more likely to be female (70% vs 36%, respectively; p = 0.04), experience better survival (median OS 3.0 vs 2.4 years, respectively; p = 0.03, log-rank test), and harbor cerebral hemispheric as opposed to skull base tumors (91% vs 43%, respectively; p = 0.002). CONCLUSIONS Based on this single-center experience at MSKCC, anaplastic meningiomas, similar to glial tumors, can arise de novo or progress from lower grade tumors. These tumor groups appear to have distinct clinical behavior. De novo tumors may well be molecularly distinct, which is under further investigation. Aggressive GTR appears to confer an OS advantage in patients with anaplastic meningioma, and this is likely independent of tumor progression status. Similarly, those patients with de novo tumors experience a survival advantage likely independent of extent of resection.

The Role of Extent of Resection in IDH1 Wild-Type or Mutant Low-Grade Gliomas.

BACKGROUND Maximizing extent of resection (EOR) improves outcomes in adults with World Health Organization (WHO) grade II low‐grade gliomas (LGG). However, recent studies demonstrate that LGGs bearing a mutation in the isocitrate dehydrogenase 1 (IDH1) gene are a distinct molecular and clinical entity. It remains unclear whether maximizing EOR confers an equivalent clinical benefit in IDH mutated (mtIDH) and IDH wild‐type (wtIDH) LGGs. OBJECTIVE To assess the impact of EOR on malignant progression‐free survival (MPFS) and overall survival (OS) in mtIDH and wtIDH LGGs. METHODS We performed a retrospective review of 74 patients with WHO grade II gliomas and known IDH mutational status undergoing resection at a single institution. EOR was assessed with quantitative 3‐dimensional volumetric analysis. The effect of predictor variables on MPFS and OS was analyzed with Cox regression models and the Kaplan‐Meier method. RESULTS Fifty‐two (70%) mtIDH patients and 22 (30%) wtIDH patients were included. Median preoperative tumor volume was 37.4 cm3; median EOR of 57.6% was achieved. Univariate Cox regression analysis confirmed EOR as a prognostic factor for the entire cohort. However, stratifying by IDH status demonstrates that greater EOR independently prolonged MPFS and OS for wtIDH patients (hazard ratio [HR] = 0.002 [95% confidence interval {CI} 0.000‐0.074] and HR = 0.001 [95% CI 0.00‐0.108] respectively) but not for mtIDH patients (HR = 0.84 [95% CI 0.17‐4.13] and HR = 2.99 [95% CI 0.15‐61.66] respectively). CONCLUSION Increasing EOR confers oncologic and survival benefits in IDH1 wtLGGs, but the impact on IDH1 mtLGGs requires further study.

Patient-reported outcomes after surgical stabilization of spinal tumors: symptom-based validation of the Spinal Instability Neoplastic Score (SINS) and surgery

BACKGROUND CONTEXT Neoplastic spinal instability is movement-related pain or neurologic compromise under physiologic loads with the Spinal Instability Neoplastic Score (SINS) developed to facilitate diagnosis. There is a paucity of evidence that mechanical instability correlates with patient-reported symptoms and that surgical stabilization significantly improves these patient-reported outcomes (PROs). PURPOSE The objective of this study was to determine if SINS correlates with patient-reported preoperative pain and disability, and if surgical stabilization significantly improves PRO. STUDY DESIGN A single-institution prospective cohort study was carried out. PATIENT SAMPLE A total of 131 patients who underwent stabilization for metastatic spinal tumor treatment between July 2014 and August 2016 were included. OUTCOMES MEASURES Preoperative baseline and mean difference in perioperative PROs as assessed by the Brief Pain Inventory (BPI) and MD Anderson Symptom Inventory (MDASI) were the outcome measures. METHODS The SINS was analyzed as a continuous, ordinal, and categorical variable (Stable: 0-6, Indeterminate: 7-12, Unstable: 13-18). Statistical analysis was performed using Spearman rank coefficient (rho), the Kruskal-Wallis test, and an extension of the Cochran-Armitage trend test. The SINS and association between the mean differences in post- and preoperative PRO scores was analyzed using the Wilcoxon signed-rank test. RESULTS There was a statistically significant positive correlation between increasing SINS and severity of preoperative pain with BPI average pain (rho=0.20; p=.03) and MDASI pain (rho=0.19; p=.03). Increasing SINS correlated with severity of preoperative disability with BPI walking (rho=0.19; p=.04), MDASI activity (rho=0.24; p=.006), and MDASI walking (rho=0.20; p=.03). Similar associations were noted when SINS was analyzed as an ordinal categorical variable. Stabilization significantly improved nearly all PRO measures for patients with indeterminate and unstable SINS. Significant correlations persisted when controlling for neurologic status and were not affected based on the technique of surgical stabilization used. CONCLUSIONS Patient-related outcome-based validation of SINS confirms this scoring system for diagnosing neoplastic spinal instability and provides surgeons with a tool to determine which patients will benefit from stabilization. Surgical stabilization of cancer patients with SINS consistent with mechanical instability provides significant reduction in pain and improves patient mobility independent of neurologic status and stabilization technique.

Tissue microarray analysis reveals protein expression patterns and potential biomarkers of clinical benefit to bortezomib in relapsed/refractory non‐Hodgkin lymphoma

Patients with mantle cell lymphoma (MCL) and follicular lymphoma treated with bortezomib have consistent response rates of 30–50% across several clinical trials, suggesting a common tumour biology that may predict response. It remains unknown which processes affected by proteasome inhibitors (PI) are most important in their activity in nonHodgkin Lymphoma (NHL). Leading theories include inhibition of the cell cycle, nuclear factor jB (NFKB1; NF-jB) signalling, angiogenesis and decreased degradation of antiapoptotic proteins. Between June 2001 and December 2006, 103 patients enrolled in a multicentre Cancer Therapy Evaluation Program-sponsored Phase II trial and were treated with singleagent bortezomib. Eligibility criteria included: (i) confirmed indolent NHL or MCL, (ii) <3 prior cytotoxic treatment regimens, and (iii) adequate organ reserve (O’Connor et al, 2005; Gerecitano et al, 2009). Tissue microarrays (TMA) were stained for a panel of targets selected for their possible prognostic associations with NHL or mechanisms of action (MOA) of PI (Table S1). We correlated outcome with pretreatment tumour protein expression patterns in this unique patient population. Tissue microarrays were constructed from 55 pre-treatment tissue blocks as previously published (Koreishi, 2010). Stains for CDKN1A (p21), CDKN1B (p27), BIRC5 (Survivin), TP53 (p53), BCL2, MCL1, CFLAR (c-FLIP), Caspase, VCAM1, REL/RELA (p65), PSMB1 (proteasome subunit b1), PSMA5 (proteasome subunit a5), TOP2A(TOPOa), MIB1/ MKI67 (Ki67), CCND1 (cyclin D1), MUM1, BCL6 and CTAG1B (NYEso) were graded by two independent pathologists, and a consensus grade was determined for each marker. Cases were considered positive if >20% of cells showed staining. In cases of discrepancy, the higher value was reported. Stains were also graded on a scale from 0 to 4, representing the proportion of cells per high power field. Nuclear RELA was estimated using continuous percentages, and then separated into quartiles for analysis. In those cases where the diagnostic samples did not run through the entire TMA block, shavings from the initial block were graded separately. Fisher’s exact test was used to assess the association between response rates [progressive disease (PD) versus complete response (CR)/partial response (PR)/stable disease (SD), CR/PR versus SD/PD] and the expression of each marker. Progression-free survival (PFS) was defined from the start of treatment to the date of death or progression, whichever occurred first. Patients were censored at their last date of follow-up if they were alive and progression-free. For patients who responded to therapy (CR/PR), duration of response (DOR) was calculated from the date of the best response to the date of progression, and patients who remained progression-free at the last follow-up were censored. None of the patients included in this analysis died prior to disease progression. The Log-rank test or permutation log-rank test was used to compare PFS and DOR between levels of stain expression. Multiple comparison adjustment was not applied considering the exploratory nature of the study. All tests were completed in SAS 9.2 (SAS Institute, Inc., Cary, NC, USA) and R version 2.9.2 (http:// www.r-project.org/). Demographic and response/survival information for patients included in this analysis (n = 55) reflect the total population (n = 103, Table 1) (Gerecitano et al, 2009; O’Connor et al, 2010). Two proteins showed significant association with response: High expression (50–75%) of MCL1 conferred a decreased chance of achieving SD or better compared with 25–50% or 0–25% expression of MCL (54%, 100% and 73%, respec-

Minimal Access Surgery for Spinal Metastases: Prospective Evaluation of a Treatment Algorithm Using Patient-Reported Outcomes

BACKGROUND Minimal access surgery (MAS) allows for an early return to systemic and radiation therapy in patients with cancer, leading to its increasing usage in the treatment of spinal metastases. Systematic examination of surgical indications resulted in the development of an algorithm for implementation of MAS in the treatment of spinal metastases. The objective of the present study was to evaluate a spine tumor MAS treatment algorithm using patient-reported outcomes for patients with cancer undergoing treatment of spinal metastases. METHODS We performed a prospective cohort study of patients who had undergone spinal percutaneous instrumented stabilization with the addition of MAS spinal cord or nerve root decompression and/or kyphoplasty when indicated at a tertiary cancer center from December 2013 to August 2016. Validated patient-reported outcome measures, including the Brief Pain Inventory and the MD Anderson Symptom Inventory-spine module, were used. The patient-reported outcome measures were collected and compared at baseline, 3 months, and long-term follow-up (range, 4.5-12 months). RESULTS A total of 51 patients were included. MAS resulted in a statistically significant decrease in the severity of pain and improved activity, ability to work, and enjoyment of life (P < 0.001). The improvement was reported at the short- and long-term follow-up points. CONCLUSIONS We present our treatment algorithm for MAS implementation in the treatment of thoracolumbar spinal metastases. Prospectively collected data have demonstrated that using this algorithm, MAS surgery for the treatment of spinal metastases results in significant decreases in pain severity and symptom interference with daily activities.

Prior malignancies in patients harboring glioblastoma: an institutional case-study of 2164 patients

More patients are surviving long-term following a cancer diagnosis and as such are at risk for second malignancies. As the most common primary brain tumor, glioblastoma (GBM) will not infrequently occur in this population. No study has examined the incidence of prior cancer (PC) in patients harboring GBM. Here we evaluate the epidemiological features, as well as the molecular and clinical characteristics of GBM as a second cancer. Utilizing a web-based cancer data management system at our institution, we identified 2164 patients harboring GBM from 2007 to 2014. We collected baseline demographic, molecular, and clinical data. Univariate analysis was performed to compare the cohort of GBM patients with and without PC diagnosis. Survival differences were analyzed with Kaplan–Meier and log-rank testing. A Cox-proportional hazards model was fit for multivariable analysis. 170 patients (7.9%) harboring GBM had a PC diagnosis. The median interval between diagnoses was 79 months. The most common pathologies were breast (18.8%) and prostate (18.8%) cancer. Patients with a PC were older at the time of GBM diagnosis than those without PC (66 vs. 59 years, p < 0.001) and were more likely to be white (88.2 vs. 72.8%, p < 0.001). Patients with PC were more likely to harbor an EGFR (20 vs. 12.3%, p < 0.001) or MGMT mutation (17.6 vs. 11.6%, p < 0.001). Median survival was 13 months in the PC cohort and 15 months in the cohort without PC (p = NS). Age, KPS, and diagnosis year were the only factors which influenced outcome in multivariable analysis. Patients who develop GBM following a prior malignancy constitute ~8% of patients with GBM. Despite significant molecular differences these two cohorts appear to have a similar overall prognosis and clinical course. Thus, whether or not a patient harbors a malignancy prior to diagnosis of GBM should not exclude him or her from aggressive treatment or for consideration of novel investigational therapies.