Clinical and molecular features of pediatric cancer patients with Lynch syndrome.

BACKGROUND: The association of childhood cancer with Lynch syndrome is not established compared with the significant pediatric cancer risk in recessive constitutional mismatch repair deficiency syndrome (CMMRD). PROCEDURE: We describe the clinical features, germline analysis, and tumor genomic profiling of patients with Lynch syndrome among patients enrolled in pediatric cancer genomic studies. RESULTS: There were six of 773 (0.8%) pediatric patients with solid tumors identified with Lynch syndrome, defined as a germline heterozygous pathogenic variant in one of the mismatch repair (MMR) genes (three with MSH6, two with MLH1, and one with MSH2). Tumor analysis demonstrated evidence for somatic second hits and/or increased tumor mutation burden in three of four patients with available tumor with potential implications for therapy and identification of at-risk family members. Only one patient met current guidelines for pediatric cancer genetics evaluation at the time of tumor diagnosis. CONCLUSION: Approximately 1% of children with cancer have Lynch syndrome, which is missed with current referral guidelines, suggesting the importance of adding MMR genes to tumor and hereditary pediatric cancer panels. Tumor analysis may provide the first suggestion of an underlying cancer predisposition syndrome and is useful in distinguishing between Lynch syndrome and CMMRD.

A phase 2 study of valproic acid and radiation, followed by maintenance valproic acid and bevacizumab in children with newly diagnosed diffuse intrinsic pontine glioma or high-grade glioma.

PURPOSE: To study the efficacy and tolerability of valproic acid (VPA) and radiation, followed by VPA and bevacizumab in children with newly diagnosed diffuse intrinsic pontine glioma (DIPG) or high-grade glioma (HGG). METHODS: Children 3 to 21 years of age received radiation therapy and VPA at 15 mg/kg/day and dose adjusted to maintain a trough range of 85 to 115 µg/mL. VPA was continued post-radiation, and bevacizumab was started at 10 mg/kg intravenously biweekly, four weeks after completing radiation therapy. RESULTS: From September 2009 through August 2015, 20 DIPG and 18 HGG patients were enrolled (NCT00879437). During radiation and VPA, grade 3 or higher toxicities requiring discontinuation or modification of VPA dosing included grade 3 thrombocytopenia (1), grade 3 weight gain (1), and grade 3 pancreatitis (1). During VPA and bevacizumab, the most common grade 3 or higher toxicities were grade 3 neutropenia (3), grade 3 thrombocytopenia (3), grade 3 fatigue (3), and grade 3 hypertension (4). Two patients discontinued protocol therapy prior to disease progression (one grade 4 thrombosis and one grade 1 intratumoral hemorrhage). Median event-free survival (EFS) and overall survival (OS) for DIPG were 7.8 (95% CI 5.6-8.2) and 10.3 (7.4-13.4) months, and estimated one-year EFS was 12% (2%-31%). Median EFS and OS for HGG were 9.1 (6.4-11) and 12.1 (10-22.1) months, and estimated one-year EFS was 24% (7%-45%). Four patients with glioblastoma and mismatch-repair deficiency syndrome had EFS of 28.5, 16.7, 10.4, and 9 months. CONCLUSION: Addition of VPA and bevacizumab to radiation was well tolerated but did not appear to improve EFS or OS in children with DIPG or HGG.

Therapeutic implications of CD1d expression and tumor-infiltrating macrophages in pediatric medulloblastomas.

Immunobiology of medulloblastoma (MB), the most common malignant brain tumor in children, is poorly understood. Although tumor cells in some MBs were recently shown to express CD1d and be susceptible to Vα24-invariant natural killer T (NKT)-cell cytotoxicity, the clinical relevance of CD1d expression in MB patients remains unknown. We investigated the expression of CD1d in pediatric MBs and correlated with molecular and clinical characteristics. Specifically, we explored if NKT cell therapy can be targeted at a subset of pediatric MBs with poorer prognosis. Particularly, infantile MBs have a worse outcome because radiotherapy is delayed to avoid neurocognitive sequelae. Immunohistochemistry for CD1d was performed on a screening set of 38 primary pediatric MBs. Gene expression of the membrane form of M2 macrophage marker, CD163, was studied in an expanded cohort of 60 tumors. Outcome data was collected prospectively. Thirteen of 38 MBs (34.2 %) expressed CD1d on immunohistochemistry. CD1d was expressed mainly on MB tumor cells, and on some tumor-associated macrophages. Majority (18/22, 82 %) of non sonic-hedgehog/Wingless-activated MBs (group 3 and 4) were CD1d-negative (p = 0.05). A subset of infantile MBs (4/9, 44.4 %) expressed CD1d. Macrophages infiltrating MB expressed CD163 apart from CD1d. Molecular subtypes demonstrated statistical differences in CD163 expression, SHH-tumors were the most enriched (p = 0.006). Molecular and clinical subtypes of pediatric MB exhibit distinct differences in CD1d expression, which have important therapeutic implications. High CD1d expression in infantile MBs offers potential new immunotherapeutic treatment with NKT cell therapy in infants, where treatment is suboptimal due delayed radiotherapy.

Implications of tumor location on subtypes of medulloblastoma.

BACKGROUND: Medulloblastoma (MB) comprises of four molecular subtypes, Sonic hedgehog (SHH), Wingless (WNT), Groups 3 and 4. WNT-subtype MBs were found to arise from midline of the brainstem occupying the fourth ventricle while SHH-subtype occupied the cerebellar hemisphere in a small subset of patients. PROCEDURE: We tested this hypothesis in a large cohort of pediatric MBs comprising of all four molecular subtypes. RESULTS: We validated in the first comprehensive analysis of tumor location of 60 human MBs representative of the four molecular subtypes, that hemispheric tumors are significantly associated with SHH-subtype MBs while midline tumors with WNT-subtype, Group 3 and 4 MBs (P < 0.001). Nearly half of SHH-subtype MBs were midline. CONCLUSIONS: Tumor location should not be generalized to MB subtypes. SHH-subtype MBs are not exclusively hemispheric and hemispheric MBs are not always SHH-activated. It is imperative to identify subtypes in conjunction with tumor location when exploring currently available targeted therapy.

Phase II study of everolimus for recurrent or progressive pediatric ependymoma.

Background: Preclinical studies have suggested that mTOR pathway signaling may be a potential therapeutic target for childhood ependymoma. Methods: A phase II clinical trial (ClinicalTrials.gov identifier: NCT02155920) of single-agent everolimus was performed to test the hypothesis that mTOR pathway inhibition would result in tumor responses for children with recurrent and/or progressive ependymomas. Results: Eleven subjects [sex: 4 females (36.4%); median age: 8 years (range: 2-15 years); race: 9 white; prior therapies: median 6 (range: 3-9)] were enrolled on the study. Ten primary tumors were located in the posterior fossa and one primary tumor was located in the spinal cord. Eight of 9 tumors were PF-A subtype epenydmomas. All subjects were treated with oral everolimus 4.5 mg/m2/day (each cycle = 28 days) that was titrated to achieve serum trough levels of 5-15 ng/ml. Overall, everolimus was well tolerated; except for a single event of grade 3 pneumonia, all adverse events were grade 1-2. No objective tumor responses were observed. Participating subjects experienced tumor progression and discontinued therapy after a median of 2 cycles of therapy (1 cycle = 2; 2 cycles = 6; 3, 4, and 8 cycles = 1 each). Conclusions: Everolimus does not appear to have activity for children with recurrent or progressive PF-A ependymoma.

Comparison of ethanol versus formalin fixation on preservation of histology and RNA in laser capture microdissected brain tissues.

Although RNA can be retrieved from formalin-fixed, paraffin-embedded (FFPE) tissues, the yield is low, and the RNA is fragmented. Recent advances in gene expression profiling underscore the importance of identifying a fixative that preserves histology and mRNA. We demonstrated that, for immersion fixation of brains, 70% ethanol is superior to formalin for mRNA preservation. RNA yield from ethanol-fixed tissues was 70% of the yield from fresh frozen specimens, but only a negligible quantity was recovered from formalin-fixed tissues. RNA from ethanol-fixed brains showed integrity comparable to RNA from fresh frozen tissues, and RT-PCR using RNA from ethanol-fixed tissues was consistently successful. RNA from FFPE tissues composed of low-molecular weight fragments, and their use in RT-PCR failed repeatedly. The yield and quality of RNA from ethanol-fixed brains were unaffected after immersion at 4 degrees C for 2 weeks. In a blinded comparison to FFPE tissues, ethanol-fixed specimens were judged to show comparable histology and superior immunostaining. After laser capture microdissection (LCM), we failed to recover mRNA from FFPE tissues but retrieved mRNA from ethanol-fixed tissues for RT-PCR and cDNA microarray analysis. We conclude that 70% ethanol preserves RNA integrity and is suitable for expression profiling of brain tissues by LCM and cDNA microarray.