Grade 2, 3 and Dedifferentiated Chondrosarcomas: A Comparative Study of Isocitrate Dehydrogenase-Mutant and Wild-Type Tumors with Implications for Prognosis and Therapy.

BACKGROUND: Grade 2 and 3 and dedifferentiated chondrosarcomas (CS) are frequently associated with isocitrate dehydrogenase (IDH) mutations and often exhibit a poor clinical outcome. Treatment is limited mainly to surgery. Defining IDH status (wild type (WT) and mutant) and the associated transcriptome may prove useful in determining other therapeutic options in these neoplasms. METHODS: Formalin-fixed paraffin-embedded material from 69 primary and recurrent grade 2, 3 and dedifferentiated CS was obtained. DNA sequencing for IDH1 and IDH2 mutations (n = 47) and RNA sequencing via Nextseq 2000 (n = 14) were performed. Differentially expressed genes (DEGs) were identified and used to predict aberrant biological pathways with Ingenuity Pathway Analysis (IPA) software (Qiagen). Gene Set Enrichment Analyses (GSEA) using subsets C3, C5 and C7 were performed. Differentially expressed genes were validated by immunohistochemistry. Outcome analysis was performed using the Wilcoxon test. RESULTS: A set of 69 CS (28 females, 41 males), average age 65, distributed among femur, pelvis, humerus, and chest wall were identified from available clinical material. After further selection based on available IDH status, we evaluated 15 IDH WT and 32 IDH mutant tumors as part of this dataset. Out of 15 IDH WT tumors, 7 involved the chest wall/scapula, while 1 of 32 mutants arose in the scapula. There were far more genes overexpressed in IDH WT tumors compared to IDH mutant tumors. Furthermore, IDH WT and IDH mutant tumors were transcriptomically distinct in the IPA and GSEA, with IDH mutant tumors showing increased activity in methylation pathways and endochondral ossification, while IDH WT tumors showed more activity in normal matrix development pathways. Validation immunohistochemistry demonstrated expression of WT1 and AR in IDH WT tumors, but not in IDH mutants. SATB2 was expressed in IDH mutant tumors and not in WT tumors. Outcome analysis revealed differences in overall survival between mutant and WT tumors (p = 0.04), dedifferentiated mutant and higher-grade (2, 3) mutant tumors (p = 0.03), and dedifferentiated mutant and higher-grade (2, 3) WT tumors (p = 0.03). The longest survival times were observed in patients with higher-grade WT tumors, while patients with dedifferentiated mutant tumors showed the lowest survival. Generally, patients with IDH WT tumors displayed longer survival in both the higher-grade and dedifferentiated groups. CONCLUSIONS: Grade 2, 3 and dedifferentiated chondrosarcomas are further characterized by IDH status, which in turn informs transcriptomic phenotype and overall survival. The transcriptome is distinct depending on IDH status, and implies different treatment targets.

Fair AI-powered orthopedic image segmentation: addressing bias and promoting equitable healthcare.

AI-powered segmentation of hip and knee bony anatomy has revolutionized orthopedics, transforming pre-operative planning and post-operative assessment. Despite the remarkable advancements in AI algorithms for medical imaging, the potential for biases inherent within these models remains largely unexplored. This study tackles these concerns by thoroughly re-examining AI-driven segmentation for hip and knee bony anatomy. While advanced imaging modalities like CT and MRI offer comprehensive views, plain radiographs (X-rays) predominate the standard initial clinical assessment due to their widespread availability, low cost, and rapid acquisition. Hence, we focused on plain radiographs to ensure the utilization of our contribution in diverse healthcare settings, including those with limited access to advanced imaging technologies. This work provides insights into the underlying causes of biases in AI-based knee and hip image segmentation through an extensive evaluation, presenting targeted mitigation strategies to alleviate biases related to sex, race, and age, using an automatic segmentation that is fair, impartial, and safe in the context of AI. Our contribution can enhance inclusivity, ethical practices, equity, and an unbiased healthcare environment with advanced clinical outcomes, aiding decision-making and osteoarthritis research. Furthermore, we have made all the codes and datasets publicly and freely accessible to promote open scientific research.