Modality redundancy for MRI-based glioblastoma segmentation.

PURPOSE: Automated glioblastoma segmentation from magnetic resonance imaging is generally performed on a four-modality input, including T1, contrast T1, T2 and FLAIR. We hypothesize that information redundancy is present within these image combinations, which can possibly reduce a model's performance. Moreover, for clinical applications, the risk of encountering missing data rises as the number of required input modalities increases. Therefore, this study aimed to explore the relevance and influence of the different modalities used for MRI-based glioblastoma segmentation. METHODS: After the training of multiple segmentation models based on nnU-Net and SwinUNETR architectures, differing only in their amount and combinations of input modalities, each model was evaluated with regard to segmentation accuracy and epistemic uncertainty. RESULTS: Results show that T1CE-based segmentation (for enhanced tumor and tumor core) and T1CE-FLAIR-based segmentation (for whole tumor and overall segmentation) can reach segmentation accuracies comparable to the full-input version. Notably, the highest segmentation accuracy for nnU-Net was found for a three-input configuration of T1CE-FLAIR-T1, suggesting the confounding effect of redundant input modalities. The SwinUNETR architecture appears to suffer less from this, where said three-input and the full-input model yielded statistically equal results. CONCLUSION: The T1CE-FLAIR-based model can therefore be considered as a minimal-input alternative to the full-input configuration. Addition of modalities beyond this does not statistically improve and can even deteriorate accuracy, but does lower the segmentation uncertainty.

Detection of cell-free tumor DNA in cerebrospinal fluid as a diagnostic biomarker for leptomeningeal melanoma metastasis: A case series.

Leptomeningeal melanoma metastases (LMM) are associated with poor survival. Diagnosis is based on clinical presentation, brain MRI and cerebrospinal fluid (CSF) analysis. Inconclusive findings at initial presentation can delay treatment. In this single-center case series, detection of BRAFV600- and NRASQ61-mutant cell-free tumor DNA (cfDNA) in CSF was evaluated as a complementary diagnostic biomarker. In 12 patients with clinical suspicion of LMM, a retrospective analysis of MRI, CSF cytology and cfDNA analysis on 1 mL of CSF using the Idylla® platform was carried out. Nine patients displayed MRI abnormalities suggesting LMM. CSF analysis identified malignant cells in three patients (including one without MRI abnormalities). BRAFV600- or NRASQ61-mutant cfDNA was detected in CSF of nine patients (eight with and one without MRI abnormalities; all patients with positive CSF cytology). Subsequent follow-up confirmed LMM in all patients with positive and in one patient with a negative CSF cfDNA analysis (sensitivity 81.8%; specificity 100%). Our findings suggest that analyzing BRAFV600- and NRASQ61-mutant cfDNA in CSF using the Idylla® platform holds promise as a sensitive and specific complementary diagnostic biomarker for LMM, particularly in case of inconsistency between imaging and CSF cytology. The 110-min analysis can facilitate urgent treatment decisions.

Optic nerve and chiasm hemangioblastomas in von Hippel-Lindau disease: report of 12 cases and review of the literature.

Introduction: Optic nerve and chiasm hemangioblastomas are rare tumors, occurring sporadically or in the context of von Hippel-Lindau (VHL) disease. They have only been portrayed in isolated case reports and small cohorts. Their natural history and therapeutic strategies are only scarcely described. To better characterize these rare tumors, we retrospectively analyzed an optic nerve and chiasm hemangioblastoma series of 12 VHL patients. By combining our own experience to a review of all known cases in literature, we intended to create treatment recommendations for optic nerve and chiasm hemangioblastomas in VHL patients. Methods: We reviewed two electronic databases in the hospitals of our senior authors, searching for VHL patients with optic nerve or chiasm hemangioblastomas. Clinical data were summarized. Tumor size and growth rate were measured on contrast enhanced MRI. Comparable data were collected by literature review of all available cases in VHL patients (Pubmed, Trip, Google and Google Scholar). Results: Of 269 VHL patients, 12 had optic nerve or chiasm hemangioblastomas. In 10 of 12 patients, tumors were diagnosed upon annual ophthalmoscopic/MRI screening. Of 8 patients who were asymptomatic at diagnosis, 7 showed absent or very slow annual progression, without developing significant vision impairment. One patient developed moderate vision impairment. Two symptomatic patients suffered from rapid tumor growth and progressive vision impairment. Both underwent late-stage surgery, resulting in incomplete resection and progressive vision impairment. One patient presented with acute vision field loss. A watchful-waiting approach was adopted because the hemangioblastoma was ineligible for vision-sparing surgery. One patient developed progressive vision impairment after watchful waiting. In the literature we found 45 patient cases with 48 hemangioblastomas. Discussion: When optic nerve and chiasm hemangioblastomas are diagnosed, we suggest annual MRI follow-up as long as patients do not develop vision impairment. If tumors grow fast, threaten the contralateral eye, or if patients develop progressive vision deficiency; surgical resection must be considered because neurological impairment is irreversible, and resection of large tumors carries a higher risk of further visual decline.

Low-Dose Bevacizumab for the Treatment of Focal Radiation Necrosis of the Brain (fRNB): A Single-Center Case Series.

Focal radiation necrosis of the brain (fRNB) is a late adverse event that can occur following the treatment of benign or malignant brain lesions with stereotactic radiation therapy (SRT) or stereotactic radiosurgery (SRS). Recent studies have shown that the incidence of fRNB is higher in cancer patients who received immune checkpoint inhibitors. The use of bevacizumab (BEV), a monoclonal antibody that targets the vascular endothelial growth factor (VEGF), is an effective treatment for fRNB when given at a dose of 5-7.5 mg/kg every two weeks. In this single-center retrospective case series, we investigated the effectiveness of a low-dose regimen of BEV (400 mg loading dose followed by 100 mg every 4 weeks) in patients diagnosed with fRNB. A total of 13 patients were included in the study; twelve of them experienced improvement in their existing clinical symptoms, and all patients had a decrease in the volume of edema on MRI scans. No clinically significant treatment-related adverse effects were observed. Our preliminary findings suggest that this fixed low-dose regimen of BEV can be a well-tolerated and cost-effective alternative treatment option for patients diagnosed with fRNB, and it is deserving of further investigation.

Intracerebral administration of CTLA-4 and PD-1 immune checkpoint blocking monoclonal antibodies in patients with recurrent glioblastoma: a phase I clinical trial.

BACKGROUND: Patients with recurrent glioblastoma (rGB) have a poor prognosis with a median overall survival (OS) of 30-39 weeks in prospective clinical trials. Intravenous administration of programmed cell death protein 1 and cytotoxic T-lymphocyte-associated antigen 4 inhibitors has low activity in patients with rGB. In this phase I clinical trial, intracerebral (IC) administration of ipilimumab (IPI) and nivolumab (NIVO) in combination with intravenous administration of NIVO was investigated. METHODS: Within 24 hours following the intravenous administration of a fixed dose (10 mg) of NIVO, patients underwent a maximal safe resection, followed by injection of IPI (10 mg; cohort-1), or IPI (5 mg) plus NIVO (10 mg; cohort-2) in the brain tissue lining the resection cavity. Intravenous administration of NIVO (10 mg) was repeated every 2 weeks (max. five administrations). Next generation sequencing and RNA gene expression profiling was performed on resected tumor tissue. RESULTS: Twenty-seven patients were enrolled (cohort-1: n=3; cohort-2: n=24). All patients underwent maximal safe resection and planned IC administrations and preoperative NIVO. Thirteen patients (cohort-1: n=3; cohort-2: n=10) received all five postoperative intravenous doses of NIVO. In cohort-2, 14 patients received a median of 3 (range 1-4) intravenous doses. Subacute postoperative neurological deterioration (n=2) was reversible on steroid treatment; no other central nervous system toxicity was observed. Immune-related adverse events were infrequent and mild. GB recurrence was diagnosed in 26 patients (median progression-free survival (PFS) is 11.7 weeks (range 2-152)); 21 patients have died due to progression. Median OS is 38 weeks (95% CI: 27 to 49) with a 6-month, 1-year, and 2-year OS-rate of, respectively, 74.1% (95% CI: 57 to 90), 40.7% (95% CI: 22 to 59), and 27% (95% CI: 9 to 44). OS compares favorable against a historical cohort (descriptive Log-Rank p>0.003). No significant difference was found with respect to PFS (descriptive Log-Rank test p>0.05). A higher tumor mRNA expression level of B7-H3 was associated with a significantly worse survival (multivariate Cox logistic regression, p>0.029). CONCLUSION: IC administration of NIVO and IPI following maximal safe resection of rGB was feasible, safe, and associated with encouraging OS. TRIAL REGISTRATION: NCT03233152.

Axitinib plus avelumab in the treatment of recurrent glioblastoma: a stratified, open-label, single-center phase 2 clinical trial (GliAvAx).

BACKGROUND: No treatment demonstrated to improve survival in patients with recurrent glioblastoma (rGB) in a randomized trial. Combining axitinib with the programmed cell death ligand 1 blocking monoclonal antibody avelumab may result in synergistic activity against rGB. METHODS: Adult patients with rGB following prior surgery, radiation therapy and temozolomide chemotherapy were stratified according to their baseline use of corticosteroids. Patients with a daily dose of ≤8 mg of methylprednisolone (or equivalent) initiated treatment with axitinib (5 mg oral two times per day) plus avelumab (10 mg/kg intravenous every 2 weeks) (Cohort-1). Patients with a higher baseline corticosteroid dose initiated axitinib monotherapy; avelumab was added after 6 weeks of therapy if the corticosteroid dose could be tapered to ≤8 mg of methylprednisolone (Cohort-2). Progression-free survival at 6 months (6-m-PFS%), per immunotherapy response assessment for neuro-oncology criteria, served as the primary endpoint. RESULTS: Between June 2017 and August 2018, 54 patients (27 per cohort) were enrolled and initiated study treatment (median age: 55 years; 63% male; 91% Eastern Cooperative Oncology Group Performance Status 0-1). Seventeen (63%) patients treated in Cohort-2 received at least one dose of avelumab. The 6-m-PFS% was 22.2% (95% CI 6.5% to 37.9%) and 18.5% (95% CI 3.8% to 33.2%) in Cohort-1 and Cohort-2, respectively; median overall survival was 26.6 weeks (95% CI 20.8 to 32.4) in Cohort-1 and 18.0 weeks (95% CI 12.5 to 23.5) in Cohort-2. The best objective response rate was 33.3% and 22.2% in Cohort-1 and Cohort-2, respectively, with a median duration of response of 17.9 and 19.0 weeks. The most frequent treatment-related adverse events were dysphonia (67%), lymphopenia (50%), arterial hypertension and diarrhea (both 48%). There were no grade 5 adverse events. CONCLUSION: The combination of avelumab plus axitinib has an acceptable toxicity profile but did not meet the prespecified threshold for activity justifying further investigation of this treatment in an unselected population of patients with rGB.

A focused 35-minute whole body MRI screening protocol for patients with von Hippel-Lindau disease.

BACKGROUND: Von Hippel-Lindau (VHL) disease is an autosomal dominantly inherited tumor syndrome. Affected patients develop central nervous system hemangioblastomas and abdominal tumors, among other lesions. Patients undergo an annual clinical screening program including separate magnetic resonance imaging (MRI) of the brain, whole spine and abdomen. Consequently, patients are repeatedly subjected to time-consuming and expensive MRI scans, performed with cumulative Gadolinium injections. We report our experience with a 35-min whole body MRI screening protocol, specifically designed for detection of VHL-associated lesions. METHODS: We designed an MRI protocol dedicated to the typical characteristics of VHL-associated lesions in different imaging sequences, within the time frame of 35 min. Blank imaging of the abdomen is carried out first, followed by abdominal sequences with Gadolinium contrast. Next, the full spine is examined, followed by imaging of the brain. A single dose of contrast used for abdominal imaging is sufficient for further highlighting of spine- and brain lesions, thus limiting the Gadolinium dosage. We used 1.5 Tesla equipment, dealing with fewer artifacts compared to a 3 Tesla system for spine- and abdominal imaging, while preserving acceptable quality for central nervous system images. In addition, imaging on a 1.5 Tesla scanner is slightly faster. RESULTS: From January 2016 to November 2018, we performed 38 whole body screening MRIs in 18 VHL patients; looking for the most common types of VHL lesions in the abdomen, spine, and brain, both for new lesions and follow-up. The one-step approach MRI examinations lead to 6 surgical interventions for clinically significant or symptomatic hemangioblastomas in the brain and spine. One renal cell carcinoma was treated with radiofrequency ablation. In comparison with previous conventional MRI scans of the same patients, all lesions were visible with the focused protocol. CONCLUSIONS: Annual screening in VHL disease can be done in a rapid, safe and sensitive way by using a dedicated whole body MRI protocol; saving MRI examination time and limiting Gadolinium dose.

Central nervous system gadolinium accumulation in patients undergoing periodical contrast MRI screening for hereditary tumor syndromes.

BACKGROUND: Patients with hereditary tumor syndromes undergo periodical magnetic resonance imaging (MRI) screening with Gadolinium contrast. Gadolinium accumulation has recently been described in the central nervous system after repeated administrations. The prevalence and rate of accumulation in different subgroups of patients are unknown. Neither are the mechanism nor clinical impact. This may cause uncertainty about the screening. To explore the prevalence and rate of Gadolinium accumulation in different subgroups, we retrospectively analyzed MRIs of patients with von Hippel-Lindau disease (VHL) and Tuberous Sclerosis Complex (TSC). METHODS: We determined the prevalence and rate of accumulation in the dentate nucleus and globus pallidus on unenhanced T1-weighted MRI from VHL and TSC patients. We compared the signal intensities of these regions to the signal intensity of the pons. We evaluated the impact of number of MRIs, kidney function and liver function on Gadolinium accumulation. RESULTS: Twenty eight VHL patients and 24 TSC patients were included. The prevalence of accumulation in the dentate nucleus and globus pallidus increased linearly according to number of Gadolinium enhanced MRIs and was higher in the VHL group (100%). A significant linear correlation between number of MRIs and increased signal intensity was observed in the VHL group. CONCLUSIONS: Gadolinium accumulation occurs in almost all patients undergoing contrast MRI screening after >5 MRIs. We advocate a screening protocol for patients with hereditary tumor syndromes that minimizes the Gadolinium dose. This can be accomplished by using a single administration to simultaneously screen for brain, spine and/or abdominal lesions, using an MRI protocol focused on either VHL- or TSC-specific lesions. Higher prevalence and rate of accumulation in VHL patients may be explained by the typical vascular leakage accompanying central nervous system hemangioblastomas.

Population Pharmacokinetic Approach Applied to Positron Emission Tomography: Computed Tomography for Tumor Tissue Identification in Patients with Glioma.

BACKGROUND AND AIMS: 18F-fluoro-ethyl-tyrosine (FET) is a radiopharmaceutical used in positron emission tomography (PET)-computed tomography in patients with glioma. We propose an original approach combining a radiotracer-pharmacokinetic exploration performed at the voxel level (three-dimensional pixel) and voxel classification to identify tumor tissue. Our methodology was validated using the standard FET-PET approach and magnetic resonance imaging (MRI) data acquired according to the current clinical practices. METHODS: FET-PET and MRI data were retrospectively analyzed in ten patients presenting with progressive high-grade glioma. For FET-PET exploration, radioactivity acquisition started 15 min after radiotracer injection, and was measured each 5 min during 40 min. The tissue segmentation relies on population pharmacokinetic modeling with dependent individuals (voxels). This model can be approximated by a linear mixed-effects model. The tumor volumes estimated by our approach were compared with those determined with the current clinical techniques, FET-PET standard approach (i.e., a cumulated value of FET signal is computed during a time interval) and MRI sequences (T1 and T2/fluid-attenuated inversion recovery [FLAIR]), used as references. The T1 sequence is useful to identify highly vascular tumor and necrotic tissues, while the T2/FLAIR sequence is useful to isolate infiltration and edema tissue located around the tumor. RESULTS: With our kinetic approach, the volumes of tumor tissue were larger than the tissues identified by the standard FET-PET and MRI T1, while they were smaller than those determined with MRI T2/FLAIR. CONCLUSION: Our results revealed the presence of suspected tumor voxels not identified by the standard PET approach.