Valerie Larouche

Immune Checkpoint Inhibition for Hypermutant Glioblastoma Multiforme Resulting From Germline Biallelic Mismatch Repair Deficiency

PURPOSE Recurrent glioblastoma multiforme (GBM) is incurable with current therapies. Biallelic mismatch repair deficiency (bMMRD) is a highly penetrant childhood cancer syndrome often resulting in GBM characterized by a high mutational burden. Evidence suggests that high mutation and neoantigen loads are associated with response to immune checkpoint inhibition. PATIENTS AND METHODS We performed exome sequencing and neoantigen prediction on 37 bMMRD cancers and compared them with childhood and adult brain neoplasms. Neoantigen prediction bMMRD GBM was compared with responsive adult cancers from multiple tissues. Two siblings with recurrent multifocal bMMRD GBM were treated with the immune checkpoint inhibitor nivolumab. RESULTS All malignant tumors (n = 32) were hypermutant. Although bMMRD brain tumors had the highest mutational load because of secondary polymerase mutations (mean, 17,740 ± standard deviation, 7,703), all other high-grade tumors were hypermutant (mean, 1,589 ± standard deviation, 1,043), similar to other cancers that responded favorably to immune checkpoint inhibitors. bMMRD GBM had a significantly higher mutational load than sporadic pediatric and adult gliomas and all other brain tumors (P < .001). bMMRD GBM harbored mean neoantigen loads seven to 16 times higher than those in immunoresponsive melanomas, lung cancers, or microsatellite-unstable GI cancers (P < .001). On the basis of these preclinical data, we treated two bMMRD siblings with recurrent multifocal GBM with the anti-programmed death-1 inhibitor nivolumab, which resulted in clinically significant responses and a profound radiologic response. CONCLUSION This report of initial and durable responses of recurrent GBM to immune checkpoint inhibition may have implications for GBM in general and other hypermutant cancers arising from primary (genetic predisposition) or secondary MMRD.

Central nervous system atypical teratoid rhabdoid tumours: The Canadian Paediatric Brain Tumour Consortium experience

BACKGROUND Atypical teratoid rhabdoid tumours (ATRT) are aggressive brain tumours mostly occurring in early childhood. Largest published series arise from registries and institutional experiences (1-4). The aim of this report is to provide population-based data to further characterise this rare entity and to delineate prognostic factors. PATIENTS AND METHODS A national retrospective study of children ⩽18years diagnosed with a central nervous system (CNS) ATRT between 1995 and 2007 was undertaken. All cases underwent central pathology review. RESULTS There were 50 patients (31 males; median age at diagnosis of 16.7months). Twelve patients were >36months. Infratentorial location accounted for 52% of all cases. Nineteen patients (38%) had metastatic disease. Fifteen (30%) underwent gross total resection (GTR). Ten patients (20%) underwent palliation. Among the 40 remaining patients, 22 received conventional chemotherapy and 18 received high dose chemotherapy regimens (HDC); nine received intrathecal chemotherapy and 15 received adjuvant radiation. Thirty of the 40 treated patients relapsed/progressed at a median time of 5.5months (0-32). The median survival time of the entire cohort was 13.5months (1-117.5months). Age, tumour location and metastatic status were not prognostic. Patients with GTR had a better survival (2years overall survival (OS): 60%±12.6 versus 21.7%±8.5, p=0.03). HDC conferred better outcome (2years OS 47.9%±12.1 versus 27.3%±9.5, p=0.036). Upfront radiation did not provide survival benefit. Six of the 12 survivors (50%) did not receive radiation. CONCLUSION The outcome of CNS ATRT remains poor. However, the use of HDC provides encouraging results. GTR is a significant prognostic factor. The role of adjuvant radiation remains unclear.

Tumors of the central nervous system in the first year of life.

Among 1,289 infants identified from this literature review, the most common histological diagnoses are astrocytoma (30.5%), medulloblastoma (12.2%), ependymoma (11.1%), and choroid plexus tumors (11%). Most tumors are supratentorial (65%). The most important prognostic factors are histology (malignant vs. benign) and extent of resection. Significant differences are noted for some tumor types by comparison with older children, for example in the aggressive behavior of low grade gliomas and the chemosensitivity of some high grade gliomas. While new techniques of radiation have been introduced in the management of infants, there is still reluctance to consider radiotherapy in this age group. Pediatr Blood Cancer 2007;49:1074–1082.

Phase II Weekly Vinblastine for Chemotherapy-Naïve Children With Progressive Low-Grade Glioma: A Canadian Pediatric Brain Tumor Consortium Study

Purpose Vinblastine monotherapy has shown promising activity and a low-toxicity profile in patients with pediatric low-grade glioma (PLGG) who experienced treatment failure after initial treatment with chemotherapy and/or radiation. The aim of this study was to assess the activity of vinblastine in therapy-naïve children. Patients and Methods Patients < 18 years old with unresectable and/or progressive therapy-naïve PLGG were eligible. Vinblastine was administered once per week at a dose of 6 mg/m2 intravenously over a period of 70 weeks. Vision, quality of life, neurofibromatosis type 1 (NF1) status, and BRAF mutation/fusion status were also determined and correlated with outcome. Results Fifty-four patients were enrolled onto the study, with a median age of 8 years (range, 0.7 to 17.2 years). Most patients had chiasmatic/hypothalamic tumors (55.5%), and 13 patients (24.1%) had NF1. The most common histology was pilocytic astrocytoma (46.3%). Seventeen patients were diagnosed using radiologic criteria alone. Best response to chemotherapy was centrally reviewed with a response rate (complete, partial, or minor response) of 25.9%. Disease stabilization (complete, partial, or minor response or stable disease) was achieved in 47 patients (87.0%). Visual improvement was observed in 20% of patients with optic pathway glioma. Five-year overall survival and progression-free survival (PFS) rates were 94.4% (95% CI, 88.5% to 100%) and 53.2% (95% CI, 41.3% to 68.5%), respectively, for the entire cohort. Patients with NF1 had a significantly better PFS (85.1%; 95% CI, 68.0% to 100%) when compared with patients without NF1 (42.0%; 95% CI, 29.1% to 60.7%; P = .012). Age< 3 years or > 10 years was not associated with poor outcome. Treatment was well tolerated, and quality of life was not affected during treatment. In this trial, there was no correlation between BRAF alterations and outcome. Conclusion Vinblastine administered once per week is well tolerated in children with treatment naïve PLGG. Overall survival and PFS are comparable to current therapies, with a favorable toxicity profile and a maintained quality of life.

Therapeutic and prognostic implications of BRAF V600E in pediatric low-grade gliomas

Purpose BRAF V600E is a potentially highly targetable mutation detected in a subset of pediatric low-grade gliomas (PLGGs). Its biologic and clinical effect within this diverse group of tumors remains unknown. Patients and Methods A combined clinical and genetic institutional study of patients with PLGGs with long-term follow-up was performed (N = 510). Clinical and treatment data of patients with BRAF V600E mutated PLGG (n = 99) were compared with a large international independent cohort of patients with BRAF V600E mutated-PLGG (n = 180). Results BRAF V600E mutation was detected in 69 of 405 patients (17%) with PLGG across a broad spectrum of histologies and sites, including midline locations, which are not often routinely biopsied in clinical practice. Patients with BRAF V600E PLGG exhibited poor outcomes after chemotherapy and radiation therapies that resulted in a 10-year progression-free survival of 27% (95% CI, 12.1% to 41.9%) and 60.2% (95% CI, 53.3% to 67.1%) for BRAF V600E and wild-type PLGG, respectively ( P < .001). Additional multivariable clinical and molecular stratification revealed that the extent of resection and CDKN2A deletion contributed independently to poor outcome in BRAF V600E PLGG. A similar independent role for CDKN2A and resection on outcome were observed in the independent cohort. Quantitative imaging analysis revealed progressive disease and a lack of response to conventional chemotherapy in most patients with BRAF V600E PLGG. Conclusion BRAF V600E PLGG constitutes a distinct entity with poor prognosis when treated with current adjuvant therapy.

Supratentorial Primitive Neuroectodermal Tumors in Young Children

TO THE EDITOR: In the April issue of the Journal of Clinical Oncology, Timmermann et al suggest that radiotherapy plays a major role in the treatment of young children with supratentorial primitive neuroectodermal tumors (stPNET) and conclude that a delay in irradiation may compromise survival. Although stPNET are undoubtedly aggressive brain tumors and results of treatments are still disappointing, the conclusions of Timmermann et al should be considered with caution. Twenty-nine children younger than 37 months were diagnosed with stPNET and were registered in these consecutive multiinstitutional protocols. The protocols used two different chemotherapy regimens, and recommendations regarding radiation were fundamentally different between HIT-SKK87 (all patients received systematic radiation at the age of 3 years, or earlier with progression) and HIT-SKK92 (radiation for patients 18 months old with recurrence/progression). The 3-year overall survival and progression-free survival of 17.2% and 14.9% of patients, respectively, are in agreement with previous reports from cooperative groups. Five children were disease-free survivors (including four who received craniospinal irradiation [CSI]), and radiation therapy was a statistically significant factor for survival. Based on their experience, the authors recommend whole CNS irradiation (54 Gy to the tumor bed, 35 Gy CSI) and also advise limiting the delay of radiation to a maximum of 6 months. The arguments of Timmermann et al are based on solid evidence that stPNET in infants has an aggressive behavior. However, their conclusions regarding the role of prophylactic radiation may be flawed, as they analyzed the impact of radiation after the fact. Almost every child whose parents, or treating physicians, refused irradiation for their children had early disease progression (14 of 15 patients). Conversely, the median time to progression was longer in the group that underwent radiation. Prophylactic radiation was offered for 10 children in remission at completion of chemotherapy. This certainly represents a subgroup of patients whose tumors behaved unusually, and one may wonder whether some of these patients who received radiation may have been overtreated. Only intent-to-treat statistics may help to refine the conclusions of this study. In addition, detailed information on the five survivors is lacking, in particular their neurocognitive assessments. The morbidity related to CSI in very young children is well documented. Walter et al reported a median IQ of 62 (44 to 86) at 5 years in a cohort of 19 young children with medulloblastoma who received CSI at a median age of 3.3 years. As stPNET are located supratentorially, concerns regarding long-term effects of radiation are even more important, as the size of these tumors is often considerable and the morbidity of a boost greater than 50 Gy is a major issue. In the absence of measurable data on neurointellectual outcome, the conclusions and recommendations of Timmermann et al should be considered with extreme caution, since the decision to avoid or delay standard-dose CSI in infants is primarily based on the unacceptable long-term morbidity of this treatment modality in this specific population. Long-term survival without radiation therapy has been reported using intensive or high-dose chemotherapy with stem cell rescue. Although the success observed with chemotherapy only is still low, new strategies are currently under investigation to improve these results. The recommended use of standard-dose CSI may represent a step back after 20 years of experience using and developing “baby brain” strategies.

Sustained complete response of recurrent glioblastoma to combined checkpoint inhibition in a young patient with constitutional mismatch repair deficiency

To the Editor: Glioblastoma multiforme (GBM) is one of the most aggressive childhood brain tumors associated with poor prognosis and less than 10% survival at 5 years.1 Importantly, at recurrence the median time to death is approximately 3 months and virtually all patients succumb to their disease. Multiple chemotherapeutic regimens have failed to induce significant improvements in survival. Novel therapies are thus urgently needed to improve the dismal survival in these patients. Immune checkpoint inhibition (ICI) is a highly promising concept in cancer therapeutics. Indeed, specific blockade of PD-1, CTLA4, PD-L1, and other immune checkpoints can be effective in arresting tumor growth and promoting tumor regression in several cancers, and induce in some cases prolonged survival in highly malignant tumors such as melanomas or lung cancer.2,3 However, only 10–30% of those patients will have sustained response to ICI, suggesting that combinational therapies with other checkpoint inhibitors, chemotherapy or other targeted agents may be needed to increase response rates. Perhaps the most important genetic factor predicting response to ICI is the tumor mutational burden. Indeed, it has been shown that tumors with a high mutational load (hypermutant) will have significantly higher response to ICI, in keepingwith positive results observed in melanomas, lung, or mismatch repair deficient colorectal cancers, which harbor the highest mutational load.4 We have shown that pediatric GBM, which arise in the context of constitutional mismatch repair deficiency (CMMRD), has the highest mutational load of all human cancers.5,6 Subsequently, we reported in Journal of Clinical Oncology7 the first response at recurrence for PD-1 blockade observed in relapsed childhood GBM. We described two siblings with CMMRD having hypermutant GBM whom at recurrence revealed objective response to the PD-1 inhibitor nivolumab.7 The index case had metastatic GBM that showed sustained response for 11 months before disease progression. Her brother was diagnosed at 3.5 years of age with a right frontoparietal GBM. The tumor was initially treated with subtotal resection followed by radiation and temozolomide. Ten months following initial diagnosis, his magnetic resonance imaging (MRI) demonstrated a multinodular regional GBM recurrence (Figure 1A) associated with left hemiparesis. He received nivolumab monotherapy infusion every 2 week obtained on a compassionate basis. After an initial symptomatic tumoral pseudoprogression at the MRI after the first dose, he showed significant clinical improvement and radiological response. Six months after immunotherapy initiation, tumor progression occurred with a new nodule in the surgical area (Figure 1B). This was accompanied by a worsening of his left hemiparesis. Based on positive results seen in patients with melanoma,8 we chose to add another CTLA-4 inhibitor (ipilimumab). Following institutional approval, the child received the combination of ipilimumab with nivolumab every 3 week for a total of four doses. The addition of ipilimumab had been well tolerated. Strikingly, MRI 3 months after initiation of the combinational therapy revealed significant improvement in his tumor status (Figure 1C) andMRI done 1-year postinitiation of ICI revealed no evidence of disease. The patient is still receivingmaintenance phase nivolumab every 2weeks and his imaging confirms sustained complete response, 30months after his first recurrence (Figure 1D). Clinically, he is going to school, has a normal life, and a normal performance for a 7-year-old male without any side effects from the ongoing checkpoint inhibition. We described a young child with CMMRD having a prolonged remission with a well-tolerated immunotherapy for a relapse GBM. This approach, reported in melanoma and other hypermutant cancers, would be beneficial for other mismatch repair deficient and pediatric hypermutant cancers for which no treatment exists. Combinational ICI was beneficial for the nodular relapse seen during therapy in our patient. Furthermore, the synergistic effect achieved by the addition of CTLA-4 inhibition upon failure of PD-1 blockade is intriguing and has not been observed in other adult hypermutant cancers when single agent fails. It is plausible that CMMRDcancers respond to different ICI due to continuous accumulation ofmutations and the unique clonal evolution of these tumors. These data strongly justify prospective trials specific to this unique population to test ICI combinations in replication repair hypermutant cancers.

Survival Following Tumor Recurrence in Children With Medulloblastoma

Medulloblastoma is the most common malignant brain tumor in children. Published survival rates for this tumor are ∼70%; however, there is limited published information on outcome after disease recurrence. This was an observational study which included all persons under the age of 18 years diagnosed with medulloblastoma from 1990 to 2009 inclusive in Canada. Data collected included date of diagnosis, age at diagnosis, sex, stage, pathology, treatment, recurrence, and current status. Survival rates were determined. In total, 550 cases were ascertained meeting the study criteria. The overall survival rate at 1 year was 83.6%±1.7%, at 3 years 77.2%±1.9%, and at 5 years 72.5%±20%. The progression-free survival rates were 78%±1.9%, 70%±2.1%, and 69±2.1% at 1, 3, and 5 years from initial diagnosis. In total, 173 (31.2%) were reported to have had tumor recurrence and 23 (11.4%) of them were alive at the time of survey with an overall survival rate at 1 year of 38.3%±4%, at 2 years of 16.9%±3.3%, and at 5 years of 12.4%±2.8%. Our data confirm that children with recurrent medulloblastoma have a poor prognosis, supporting the need for novel treatment approaches for this group.

Elevated blood levels of Dickkopf-1 are associated with acute infections: DKK1 and infections

Dickkopf‐1 (DKK1) is a soluble protein and antagonist of the Wnt/β‐catenin signaling pathway. DKK1 is found elevated in serum from patients affected with various types of cancers and in some instances, it is considered a diagnostic and prognostic biomarker. Elevated serum levels of DKK1 have also been detected in animal models of chronic inflammatory diseases. Previous work from our laboratory has demonstrated upregulation of DKK1 in cells and mouse models of the bone marrow failure (BMF) and cancer‐prone disease Fanconi anemia (FA). The present study aimed to investigate whether DKK1 blood levels in patients are associated with FA or inflammatory responses to acute infections.

Embryonal tumors in Canadian children less than 36 months of age: results from the Canadian Pediatric Brain Tumor Consortium (CPBTC)

Embryonal tumors are a heterogeneous group of central nervous system (CNS) tumors whose subgroups have varying incidence and outcome. Despite these differences, they are often grouped as a single entity for study purposes. To date, there are no Canadian multi-institutional studies examining the incidence and outcome of all embryonal subtypes. The current study is an observational study reviewing embryonal tumors in all patients less than 36 months of age diagnosed with a CNS tumor in Canada from 1990 to 2005. Embryonal tumors accounted for 26.9% of all CNS tumors. Medulloblastomas were the highest proportion of the embryonal tumors at 61.5%. Atypical teratoid/rhabdoid tumors (AT/RT) had the second highest proportion of embryonal tumors at 18%. The proportion of primitive neuroectodermal tumors (PNET) was 16%, with 2.6 and 1.9% for congenital medulloepithelioma and ependymoblastoma tumors, respectively. AT/RT and PNET were more common in younger age groups. Medulloblastoma became more prevalent with increasing age, with its highest prevalence in the 25 to 36 month age group. Survival rates for our Canadian population at 18 and 24 months were 0.74 and 0.68 for medulloblastoma, 0.64 and 0.60 for PNET, and 0.36 and 0.29 for AT/RT, respectively. Overall, our data are comparable with published international rates for embryonal tumors. These incidence and outcome figures can guide future research into these rare tumors.