Citations

Abstract Background Neuropathologic evaluation is essential for diagnosing and treating dementia, but definitive diagnoses is only available after post‐mortem brain autopsy. Positron emission tomography (PET) offers a potential in vivo proxy for predicting neuropathology. This study proposes a semi‐supervised learning (SSL) model using a single FDG‐PET image to predict underlying neuropathology. Method The SSL model leverages a self‐organized graph structure based on a patient similarity network (PSN) created from 1,495 FDG‐PET images across the aging and neurodegenerative spectrum. Each patient is represented by a node, with edges indicating image similarity derived from principal component analysis features and cosine similarity weights. The PSN structure connects 20 nearest neighbors. Neuropathologic labels were available for 204 autopsied patients. The Poisson Learning Method was employed to propagate labels from these annotated nodes to unlabeled nodes, guided by the PSN structure. FDG‐PET images were collected from clinical and research participants at the Mayo Clinic and processed using an MRI‐free pipeline. Neuropathologic labels were assigned based on widely accepted diagnostic criteria. Leave‐one‐out cross‐validation (LOOCV) was used to assess model performance. Results The confusion matrix demonstrated high accuracy in predicting Alzheimer's disease, frontotemporal lobar degeneration with tau inclusions, and Lewy body disease pathology. A Sankey diagram and label distribution analysis revealed strong correlations between clinical and predicted neuropathologic labels. Conclusion This study highlights the potential of a graph‐based SSL pipeline to predict neuropathological changes using FDG‐PET imaging data, despite limited pathology labels. The findings underscore the value of SSL in bridging the gap between abundant neuroimaging data and scarce neuropathologic labels, contributing to improved dementia diagnosis and understanding.

Abstract Background PET visual reads for multi‐center studies are conducted centrally by one or a few experts. In CLARiTI, a broader network of raters will perform the reads, necessitating methods to ensure their accuracy and reliability. Method All ADRC sites ( N = 37) will prospectively acquire amyloid PET scans for 800 cognitively unimpaired and 1200 cognitively impaired participants. Each site can use one of four amyloid PET tracers ([ 18 F]florbetaben, [ 18 F]florbetapir, [ 11 C]PIB, [ 18 F]NAV4694). PET scans will be visually interpreted by ten PET neuroimaging experts located across eight ADRCs, using their preferred hardware, software, and file format (DICOM, NIFTI). Calibration requires readers to perform blinded, independent visual interpretation of 180 amyloid PET scans (30 unique and 15 duplicate scans per tracer), previously read and selected by an expert (Table 1). An even mix of visually non‐elevated and elevated scans were selected, with quantitation spanning ‐26‐123 Centiloids (Figure 1), and 15% were visually borderline (similar proportion to ADNI4), with quantitation spanning 4‐48 Centiloids (mean 21). Kruskal‐Wallis H test showed no significant difference in the distribution of Centiloid values for scans in each tracer group, Χ 2 (3)=2.133, p = 0.545. Result Calibration scans were read by a second expert not serving in CLARiTI. There was 97% concordance and almost perfect agreement between readers (κ=0.932, p <0.001; Figure 2). Discordantly read scans were visually borderline elevated ( N = 2), borderline non‐elevated ( N = 1), and negative ( N = 1), with quantitation of 8‐21 Centiloids (mean 14). Intra‐rater reads were 98% concordant and had almost perfect agreement (κ=0.966, p <0.001). These findings hold promise for scaling‐up visual reads in CLARiTI, and we expect ≥80% concordance and substantial to almost perfect agreement (κ≥0.7) for inter/intra‐rater reliability. Conclusion Calibration is essential for accurate/reliable amyloid PET visual reads in large multi‐center studies with multiple raters. Through the calibration process, we will develop methods to improve the adjudication of borderline/challenging scans and formalize methods for visual reads of amyloid PET tracers that are investigational (i.e., [ 11 C]PIB, [ 18 F]NAV4694). Involving multiple raters across different sites, who use diverse hardware/software/file format combinations, will mirror the natural heterogeneity present in real‐world clinical practice. This calibration set can also serve as a standardized training tool for other studies incorporating amyloid PET visual reads.

Abstract BACKGROUND While most meningiomas are classified as low-grade tumors with indolent behavior, a subset is characterized by a more aggressive clinical course and patients undergo repeated resections and courses of radiation. Once these therapies fail, prognosis is poor and progression-free survival (PFS) rates at 6 months (PFS-6) are estimated to only be 43.6% for WHO grade 1 and 38% for WHO grade 2/3 meningiomas. Effective medical therapies are limited. Somatostatin receptor 2 (SSTR2) is expressed in a majority of meningiomas and represents a suitable treatment target. [177Lu]Lu-DOTA-TATE (177Lu-Dotatate) is an SSTR2-targeted radioligand therapy, established as standard of care in the treatment of certain neuroendocrine tumors. Two ongoing non-randomized studies of 177Lu-Dotatate in meningiomas have reported encouraging efficacy compared to historical benchmarks. METHODS/STUDY DESIGN The MOMENTUM-1 study (RTOG 3523, NCT#06955169) is a multicenter, randomized, open-label study of 177Lu-Dotatate in adult patients with progressive intracranial WHO grade 1-3 meningiomas. Up to 130 eligible subjects are randomized in a 2:1 fashion to 177Lu-Dotatate at 7.4 GBq (200 mCi) every 4 weeks for 4 (+2 optional) doses or standard of care (SOC) medical therapy as defined by the protocol and deemed appropriate by the local investigator. Treatment responses are assessed by gadolinium-enhanced MRI brain and [68Ga]Ga-DOTA-TATE (68Ga-Dotatate) PET imaging. At time of progression, cross-over from the SOC to the interventional treatment arm is allowed. The primary endpoint is PFS. Secondary endpoints include PFS-6, OS-12, OS, response assessments (ORR, DCR), safety/tolerability, and patient-reported outcomes. RESULTS/CONCLUSION MOMENTUM-1 is a multicenter, randomized, open-label, phase II study of 177Lu-Dotatate in adult patients with progressive WHO grade 1-3 meningioma. This clinical trial is expected to start enrolling patients in the United States in the third quarter of 2025. Details on study design, eligibility criteria for subject enrollment and study site eligibility criteria will be presented.

Background: Post-treatment prognosis and monitoring are critical for determining the timing of salvage treatment in glioblastoma patients but has been challenging due to difficulties differentiating progression from treatment effects in conventional images. This exploratory study aimed to establish the correlation of radiomics image features from time series of amino acid tracer 18F-DOPA PET images, with outcomes, using machine learning and dimension reduction analysis. Methods: 18F-DOPA PET images were collected for a patient cohort with wild-type IDH and unmethylated MGMT who underwent dose-escalated radiation therapy. Quantitative features were derived from the high uptake region (T/N > 2.0) in pre- and post-radiation therapy follow-up 18F-DOPA PET images. A customized workflow was utilized for pre-selecting predictive features, followed by manifold learning. Machine learning algorithms were employed to establish associations between imaging features and remaining survival (RS), defined as the time between a follow-up scan and date of death. Results: The ML models exhibited 81–83% ROC_AUC in predicting RS evaluated on an independent test dataset. A RS map is proposed for monitoring tumor alterations through serial 18F-DOPA PET scans, demonstrating superior sensitivity and better correlation with survival compared to the RANO criteria. Conclusions: Our study demonstrates that ML models utilizing FU 18F-DOPA PET images have the potential to effectively predict future survival outcomes in patients with glioblastoma treated with dose-escalated radiation therapy. The capability to assess changes in tumor over time through imaging can potentially assist in patient stratification and the selection of salvage treatments, while also aiding in distinguishing treatment effects from genuine tumor progression.

Abstract INTRODUCTION In multicenter Alzheimer's disease studies, amyloid positron emission tomography (PET) visual reads are typically performed centrally by a few experts. Incorporating a broader reader network enhances scalability and generalizability. METHODS Ten neuroimaging experts from eight Alzheimer's Disease Research Centers (ADRCs) visually read 180 amyloid PET scans (30 scans and 15 duplicate scans for each of four tracers, imaged across a wide variety of scanners), using preferred reading software without anatomical imaging or quantitation. Scans were classified as elevated or non‐elevated per tracer‐specific criteria. Inter‐ and intra‐rater agreement was assessed. RESULTS Inter‐rater agreement was substantial (Fleiss’ κ = 0.78), with full consensus on 69% of scans. Inter‐rater reliability was substantial to perfect across tracers (Fleiss’ κ = 0.70–0.87). Intra‐rater agreement was substantial to perfect (Cohen's κ = 0.79‐1). Scans with intermediate (10–40 Centiloid) quantitation had lower reader agreement. DISCUSSION A multicenter expert network achieved substantial agreement classifying amyloid PET scans. These scans provide a standard for reader training and reliability assurance in future studies. Highlights Calibration methods ensure reliable amyloid positron emission tomography (PET) visual reads across multiple raters. Substantial agreement is possible across readers using their preferred tools. Agreement is also substantial regardless of the amyloid PET tracer used. Scans with intermediate (10–40 Centiloid) quantitation have lower reader agreement. The calibration set will become a training tool for amyloid PET visual read studies.

Abstract A workgroup assembled by the Alzheimer’s Association recently described a conceptual framework for Alzheimer’s disease biological staging based on amyloid and tau PET imaging. However, specific tau PET cut points were left to be determined, a step necessary prior to clinical application. We sought to operationalize and evaluate Alzheimer’s disease biological staging by identifying meaningful tau PET cut points to define the four biological stages in a well-characterized participant cohort and describe the features of individuals placed into the different biological stages. The primary analysis included 896 participants in the Mayo Clinic Study of Aging or the Mayo Clinic Alzheimer’s Disease Research Center longitudinal cohorts. A validation cohort consisted of 328 participants in the Alzheimer’s Disease Neuroimaging Initiative. Both cognitively normal and impaired individuals with positive amyloid PET and evaluable tau PET imaging were included. Tau PET cut points were identified with Gaussian mixture models to characterize Alzheimer’s disease biological stage in study participants with different clinical diagnoses and objective degrees of cognitive impairment as measured by Mini-Mental State Examination. A tau PET cut point in the medial temporal region and two cut points in the temporoparietal region were identified to produce, collectively, the four Alzheimer’s disease biological stages described in the revised criteria. Increasing stage was associated with greater likelihood of mild cognitive impairment and dementia diagnosis and worsening cognitive performance on Mini-Mental State Examination and Clinical Dementia Rating Sum of Boxes, a result that was reproduced in the independent Alzheimer’s Disease Neuroimaging Initiative cohort. This study provided empirical validation for the concept of using amyloid PET and tau PET to separate subjects with biomarker-proven Alzheimer’s disease into four biological stages with distinct characteristics.

Abstract INTRODUCTION When cerebrospinal fluid (CSF) and positron emission tomography (PET) measurements for amyloid‐beta‐peptide (Aβ) related pathology are discordant, therapeutic decision‐making becomes uncertain. METHODS Using data from patients with mild cognitive impairment ( n = 541) from the Alzheimer's Disease Neuroimaging Initiative, we examined baseline characteristics and longitudinal clinical outcomes in persons grouped according to normal/abnormal Aβ via concurrent CSF and PET determinations using standard cutpoints. RESULTS Discordant groups for brain Aβ status (CSF+/PET− and CSF−/PET+) each represented about 5% of the mild cognitive impairment (MCI) population. Longitudinally, neither discordant group declined more than the CSF−/PET− group on either a memory measure or the Clinical Dementia Rating Sum of Boxes scores over a median 4 years of observation, while the CSF+/PET+ group exhibited worsening on both measures. DISCUSSION In contrast to the clinical decline observed in the CSF+/PET+ group, persons with MCI and CSF+/PET− or CSF−/PET+ brain amyloid patterns did not exhibit incipient decline. Highlights Discrepant abnormal cerebrospinal fluid (CSF) and positron emission tomography (PET) brain amyloid indicators are uncommon in mild cognitive impairment (MCI). CSF‐PET discrepant persons with MCI tend to have less abnormal values initially. CSF‐PET discrepant persons with MCI have a benign prognosis at 4 years.

Abstract Background Dasatinib is an oral inhibitor of the Src kinase family, with preclinical data indicating impact on gliomagenesis, tumor invasion, and radiosensitivity. Methods North Central Cancer Treatment Group N0877 is a phase 1 dose escalation and phase II randomized study evaluating the maximum tolerated dose (MTD), safety, and efficacy of dasatinib with radiation and temozolomide (TMZ) for glioblastoma. Following identification of the MTD, adult patients with a histologic diagnosis of glioblastoma were randomized 2:1 between dasatinib given with standard concurrent and adjuvant TMZ, versus placebo with standard concurrent and adjuvant TMZ. Radiation dose was 60 Gy in 30 fractions. The primary endpoint was overall survival (OS). Secondary endpoints included progression-free survival (PFS), toxicity, and quality of life. Results Thirteen patients were enrolled in the phase I component and established the MTD and phase II dose of 150 mg given daily. A total of 204 patients were enrolled in the phase II component. OS was not different between arms (median OS 15.6 months for dasatinib compared to 19.3 months for placebo, hazard ratio:1.21 favoring placebo, 95% CI: 0.88–1.65, log-rank P-value: .238). Similarly, PFS was not significantly different between the dasatinib and placebo arms. There was a significant increase in anemia, nausea, and creatinine elevation with dasatinib, but significantly more grade 3 lymphopenia with placebo. Conclusions The addition of dasatinib to standard chemoradiation did not improve outcomes for patients with glioblastoma. Clinical Trial Identifier NCT00869401.

The translation of AI-generated brain metastases (BM) segmentation into clinical practice relies heavily on diverse, high-quality annotated medical imaging datasets. The BraTS-METS 2023 challenge has gained momentum for testing and benchmarking algorithms using rigorously annotated internationally compiled real-world datasets. This study presents the results of the segmentation challenge and characterizes the challenging cases that impacted the performance of the winning algorithms. Untreated brain metastases on standard anatomic MRI sequences (T1, T2, FLAIR, T1PG) from eight contributed international datasets were annotated in stepwise method: published UNET algorithms, student, neuroradiologist, final approver neuroradiologist. Segmentations were ranked based on lesion-wise Dice and Hausdorff distance (HD95) scores. False positives (FP) and false negatives (FN) were rigorously penalized, receiving a score of 0 for Dice and a fixed penalty of 374 for HD95. The mean scores for the teams were calculated. Eight datasets comprising 1303 studies were annotated, with 402 studies (3076 lesions) released on Synapse as publicly available datasets to challenge competitors. Additionally, 31 studies (139 lesions) were held out for validation, and 59 studies (218 lesions) were used for testing. Segmentation accuracy was measured as rank across subjects, with the winning team achieving a LesionWise mean score of 7.9. The Dice score for the winning team was 0.65 ± 0.25. Common errors among the leading teams included false negatives for small lesions and misregistration of masks in space. The Dice scores and lesion detection rates of all algorithms diminished with decreasing tumor size, particularly for tumors smaller than 100 mm3. In conclusion, algorithms for BM segmentation require further refinement to balance high sensitivity in lesion detection with the minimization of false positives and negatives. The BraTS-METS 2023 challenge successfully curated well- annotated, diverse datasets and identified common errors, facilitating the translation of BM segmentation across varied environments and providing the tools for future development of personalized volumetric reports to patients undergoing BM treatment.