Citations

LBA10509 Background: The MCED test (Galleri) detects a shared cancer signal from cell-free DNA in blood and predicts a cancer signal origin (CSO) to guide diagnostic (dx) workup. PF2 (NCT05155605) is a large, prospective, multicenter, interventional study of the MCED test’s performance and safety in a diverse US intended-use population. Methods: PF2 enrolled participants aged ≥50y without clinical suspicion of cancer or cancer diagnosis/treatment within 3y of enrollment. Primary objectives were MCED performance and safety in terms of dx testing triggered by a positive (cancer signal detected) MCED test result. Participants with a positive MCED test underwent targeted dx workup guided by the predicted CSO(s). Participant-reported anxiety was assessed with the State Trait Anxiety Inventory. Results: PF2 enrolled 35,878 participants; 32,007 with a 12m cancer assessment were analyzable for performance. Of 287 (0.9%) with a positive MCED test, 173 were diagnosed with cancer within 12m, for a cancer detection rate of 0.5% (95% CI 0.5-0.6%) and positive predictive value of 60.3% (54.5-65.8%). Specificity was 99.6% (99.6-99.7%). 12m episode sensitivity was 69.8% (62.8-76.0%) in a prespecified subgroup of 12 cancers responsible for ⅔ of US cancer deaths, 66.2% (55.1-75.8%) in a subgroup of 6 aggressive cancers with low 5y survival, and 39.3% (34.9-44.0%) across all cancers. Overall CSO prediction accuracy was 91.3% (86.2-94.7%) across all cancers. The MCED test detected 151 new primary and 22 recurrent cancers; additional cancers detected by screening were those with USPSTF A/B (n=31) and C (n=60) recommended screening. Of 151 MCED-detected new primaries, 80 (53.0%) were clinical stage I-II; 107 (70.9%) were stage I-III. Of 80 MCED-detected stage I-II cancers, 71.2% had no USPSTF A/B screening recommendation. Of 35,335 participants analyzable for safety, 213 (0.6%) had invasive procedure(s) to evaluate a positive MCED test result, of which 90% were nonsurgical and 1.8x more likely for participants with vs without cancer diagnosis. Most (87.6%) had a CSO-guided dx evaluation. There were 5 study-related adverse events during the time of dx workup; none were serious. Anxiety temporarily increased for participants with a positive MCED test and cancer diagnosed, and returned to baseline by 12m. Of participants with a positive MCED test, median time from test result to dx resolution was 48d (95% CI 43-56). Few participants (7/63 [11.1%]) had cancer found by dx PET/CT after a targeted dx workup that did not reveal cancer. Conclusion: The MCED test demonstrated robust performance with a favorable safety profile. When MCED testing was added to USPSTF A/B screening, 6.5x more cancers were detected by screening, most at early stages. The diverse, representative PF2 participant population supports the generalizability of these findings to the US intended-use population. Clinical trial information: NCT05155605 .

Abstract Purpose The diagnostic accuracy of imaging modalities in periprosthetic joint infections (PJIs) remains uncertain. This study assessed and compared the performance of different imaging techniques for diagnosing PJI. Methods A systematic review and meta‐analysis of studies published in PubMed and Embase (1 January 2000 to 1 March 2024) was conducted. Eligible studies included adults (≥18 years), applied a valid reference standard for PJI diagnosis, and provided sufficient data to construct 2 × 2 contingency tables. Imaging modalities included white blood cell scintigraphy (WBCS), FDG‐PET/CT, antigranulocyte scintigraphy (AGS), bone scintigraphy (BS), magnetic resonance imaging (MRI), computed tomography (CT) and related combinations. Risk of bias was assessed using the QUADAS‐2 tool. A bivariate random‐effects model estimated pooled diagnostic odds ratios, sensitivities, and specificities with 95% confidence intervals (CIs). Results Forty studies (1263 PJI cases, 4700 controls) were included. Pooled sensitivity and specificity were as follows: WBCS‐SPECT/CT: 87.1% (95% CI, 31.7–99.0) and 94.1% (95% CI, 87.5–97.3); WBCS‐BMS: 79.7% (95% CI, 58.3–91.7) and 94.2% (95% CI, 91.0–96.3); WBCS: 88.1% (95% CI, 73.8–95.1) and 80.0% (95% CI, 62.2–90.7); FDG‐PET (±PET/CT): 88.1% (95% CI, 79.9–93.2) and 87.2% (95% CI, 76.3–93.5); BS‐WBCS: 59.5% (95% CI, 43.2–73.9) and 96.6% (95% CI, 92.2–98.6); BS: 89.3% (95% CI, 80.9–94.2) and 74.1% (95% CI, 59.0–85.0); AGS: 89.1% (95% CI, 76.8–95.3) and 73.0% (95% CI, 63.1–81.0). No eligible studies were identified for MRI or CT. Substantial heterogeneity was observed. Conclusion Diagnostic performance varied substantially, and no modality consistently demonstrated superior accuracy. WBCS showed the most stable results, particularly when combined with BS or BMS. Given the heterogeneity and variability in specificity, imaging should complement comprehensive diagnostic evaluation rather than be used in isolation. Level of Evidence 2a–3a, systematic review and meta‐analysis of retrospective cohort and case‐control studies.

Graph neural networks (GNNs) are increasingly used for medical imaging analysis and biological data modeling, where the integration of radiomics, topology, geometry, and generative artificial intelligence (AI) may improve representation learning from medical images and related biomedical data. Across the reviewed literature, GNNs show particular value for modeling spatial relationships, multimodal interactions, graph-structured biological networks, and non-Euclidean imaging features that are difficult to capture using conventional convolutional architectures alone. Topology- and geometry-aware approaches further expand this capability by encoding multi-scale structure, higher-order relationships, curvature, geodesic organization, and equivariant spatial priors. Hybrid graph–transformer models and generative graph methods represent emerging directions for modeling long-range dependencies, augmenting scarce datasets, supporting synthetic pretraining, and improving representation learning in low-label or heterogeneous biomedical settings. However, clinical translation remains limited by variability in graph construction, limited external validation, computational cost, scalability constraints, interpretability challenges, and uncertainty regarding the biological realism of synthetic data. Overall, this review highlights that GNN-based medical imaging analysis is most likely to advance when graph construction is biologically justified, model performance is evaluated across diverse clinical cohorts, and technical gains are paired with transparent validation, interpretability, and implementation strategies.

Abstract Nonmyeloablative-related haploidentical allogeneic stem cell transplantation (NMAC-HID allo-HSCT) has emerged as an additional treatment to achieve durable remission in sickle cell disease (SCD), a prevalent blood disorder characterized by painful vaso-occlusive crises and chronic anemia. The standard of care (SOC) for SCD includes hydroxyurea, pain management, and blood transfusion, but patients with SCD still lose several decades of life expectancy. Gene therapy (GT) for SCD is the other treatment for lifelong disease amelioration in SCD, with accessibility limited by cost and manufacturing capacity in the United States and globally. Two recent prospective studies that evaluated NMAC-HID allo-HSCT validated haploidentical allotransplantation as an efficacious and accessible treatment option in the era of GT. Given the ongoing price negotiation across jurisdictions for GT implementation and the absence of cost-effectiveness data comparing NMAC-HID allo-HSCT and GT, we conducted a cost-effectiveness analysis of NMAC-HID allo-HSCT vs GT vs SOC for adults and children living with SCD. The primary outcomes were the incremental cost-effectiveness ratio and the net monetary benefits across these 3 strategies. The secondary outcome was the maximum cost-effective threshold price for GT compared with NMAC-HID allo-HSCT. Treatment with SOC, NMAC-HID allo-HSCT, and GT accrued 14.3, 20.1, and 22.1 quality-adjusted life-years at costs of $1.22 million, $1.15 million, and $2.75 million, respectively. NMAC-HID allo-HSCT was the cost-effective strategy compared with GT in 100% of 10 000 Monte Carlo iterations across the base case and all scenario analyses. The maximum cost-effective thresholds for GT vs SOC were $1.4 million in the United States and $4200 to $22 000 across India, Nigeria, and Tanzania, depending on willingness-to-pay thresholds.

ABSTRACT Objective Odontogenic sinusitis (ODS) ranges from isolated maxillary disease to pansinusitis. We compared patient characteristics, microbial and histopathological features, and symptom outcomes following endoscopic sinus surgery (ESS) between low and high radiographic burden ODS patients. Methods ODS patients undergoing ESS between 01/2013 and 08/2024 were reviewed. Preoperative CT scans were used to classify patients into high radiographic burden [with ostiomeatal complex (OMC) opacification] and low radiographic burden ODS [clear OMC]. Demographic, microbiological, and histopathological data were analyzed. Symptom improvement was assessed using pre‐ and post‐operative SNOT‐22 scores. Results Eighty‐seven patients were identified (high CT burden: 70; low CT burden: 17). High CT burden ODS was associated with older age ( p = 0.014), higher overall degree of inflammation ( p = 0.002), hyperplastic/papillary epithelial changes ( p = 0.034), and neutrophil infiltration ( p = 0.003). Culture of classic ODS bacteria ( Fusobacterium spp., Prevotella spp., mixed anaerobes, Streptococcus anginosus group) did not correlate with CT disease burden ( p = 0.110). No independent predictors of radiographic high‐burden disease were identified. Post‐ESS, significant reduction in total and rhinologic SNOT‐22 scores was only observed in the radiographic high‐burden ODS group ( n = 39; median reductions: 23 and 12 respectively; p < 0.001). Conclusion High radiographic burden was associated with older patients and with tissue inflammation level. Classic ODS bacterial profile did not associate with increased CT burden. While these findings remain exploratory, the observed differences in inflammatory and symptomatic profiles between groups suggest that radiographic burden warrants further study as a potential factor in ODS clinical decision‐making. Level of Evidence 3.