High-throughput micro-CT analysis identifies sex-dependent biomarkers of erosive arthritis in TNF-Tg mice and differential response to anti-TNF therapy.

backgroundDevelopment of reliable disease activity biomarkers is critical for diagnostics, prognostics, and novel drug development. Although computed tomography (CT) is the gold-standard for quantification of bone erosions, there are no consensus approaches or rationales for utilization of specific outcome measures of erosive arthritis in complex joints. In the case of preclinical models, such as sexually dimorphic tumor necrosis factor transgenic (TNF-Tg) mice, disease severity is routinely quantified in the ankle through manual segmentation of the talus or small regions of adjacent bones primarily due to the ease in measurement. Herein, we sought to determine the particular hindpaw bones that represent reliable biomarkers of sex-dependent disease progression to guide future investigation and analysis.

methodsHindpaw micro-CT was performed on wild-type (n = 4 male, n = 4 female) and TNF-Tg (n = 4 male, n = 7 female) mice at monthly intervals from 2-5 (females) and 2-8-months (males) of age, since female TNF-Tg mice exhibit early mortality from cardiopulmonary disease at approximately 5-6-months. Further, 8-month-old WT (n = 4) and TNF-Tg males treated with anti-TNF monoclonal antibodies (n = 5) or IgG placebo isotype controls (n = 6) for 6-weeks were imaged with micro-CT every 3-weeks. For image analysis, we utilized our recently developed high-throughput and semi-automated segmentation strategy in Amira software. Synovial and osteoclast histology of ankle joints was quantified using Visiopharm.

resultsFirst, we demonstrated that the accuracy of automated segmentation, determined through analysis of ~9000 individual bones by a single user, was comparable in wild-type and TNF-Tg hindpaws before correction (79.2±8.9% vs 80.1±5.1%, p = 0.52). Compared to other bone compartments, the tarsal region demonstrated a sudden, specific, and significant bone volume reduction in female TNF-Tg mice, but not in males, by 5-months (4-months 4.3± 0.22 vs 5-months 3.4± 0.62 mm3, p<0.05). Specifically, the cuboid showed significantly reduced bone volumes at early timepoints compared to other tarsals (i.e., 4-months: Cuboid -24.1±7.2% vs Talus -9.0±5.9% of 2-month baseline). Additional bones localized to the anterolateral region of the ankle also exhibited dramatic erosions in the tarsal region of females, coinciding with increased synovitis and osteoclasts. In TNF-Tg male mice with severe arthritis, the talus and calcaneus exhibited the most sensitive response to anti-TNF therapy measured by effect size of bone volume change over treatment period.

conclusionsWe demonstrated that sexually dimorphic changes in arthritic hindpaws of TNF-Tg mice are bone-specific, where the cuboid serves as a reliable early biomarker of erosive arthritis in female mice. Adoption of automated segmentation approaches in pre-clinical or clinical models has potential to translate quantitative biomarkers to monitor bone erosions in disease and evaluate therapeutic efficacy.