Autosegmentation based on different-sized training datasets of consistently-curated volumes and impact on rectal contours in prostate cancer radiation therapy.

Background and purpose: Autosegmentation techniques are emerging as time-saving means for radiation therapy (RT) contouring, but the understanding of their performance on different datasets is limited. The aim of this study was to determine agreement between rectal volumes by an existing autosegmentation algorithm and manually-delineated rectal volumes in prostate cancer RT. We also investigated contour quality by different-sized training datasets and consistently-curated volumes for retrained versions of this same algorithm. Materials and methods: Single-institutional data from 624 prostate cancer patients treated to 50-70 Gy were used. Manually-delineated clinical rectal volumes (clinical) and consistently-curated volumes recontoured to one anatomical guideline (reference) were compared to autocontoured volumes by a commercial autosegmentation tool based on deep-learning (v1; n = 891, multiple-institutional data) and retrained versions using subsets of the curated volumes (v32/64/128/256; n = 32/64/128/256). Evaluations included dose-volume histogram metrics, Dice similarity coefficients, and Hausdorff distances; differences between groups were quantified using parametric or non-parametric hypothesis testing. Results: Volumes by v1-256 (76-78 cm3) were larger than reference (75 cm3) and clinical (76 cm3). Mean doses by v1-256 (24.2-25.2 Gy) were closer to reference (24.2 Gy) than to clinical (23.8 Gy). Maximum doses were similar for all volumes (65.7-66.0 Gy). Dice for v1-256 and reference (0.87-0.89) were higher than for v1-256 and clinical (0.86-0.87) with corresponding Hausdorff comparisons including reference smaller than comparisons including clinical (5-6 mm vs. 7-8 mm). Conclusion: Using small single-institutional RT datasets with consistently-defined rectal volumes when training autosegmentation algorithms created contours of similar quality as the same algorithm trained on large multi-institutional datasets.

Desmoglein 3 - Influence on oral carcinoma cell migration and invasion.

Desmoglein 3 (Dsg3) is an adhesion receptor in desmosomes, but its role in carcinoma cell migration and invasion is mostly unknown. Our aim was to quantitatively analyse the motion of Dsg3-modified carcinoma cells in 2D settings and in 3D within tumour microenvironment mimicking (TMEM) matrices. We tested mutant constructs of C-terminally truncated Dsg3 (∆238 and ∆560), overexpressed full-length (FL) Dsg3, and empty vector control (Ct) of buccal mucosa squamous cell carcinoma (SqCC/Y1) cells. We captured live cell images and analysed migration velocities and accumulated and Euclidean distances. We compared rodent collagen and Matrigel® with human Myogel TMEM matrices for these parameters in 3D sandwich, in which we also tested the effects of monoclonal antibody AK23, which targets the EC1 domain of Dsg3. In monolayer culture, FL and both truncated constructs migrated faster and had higher accumulated distances than Ct cells. However, in the 3D assays, only the mutants invaded faster relative to Ct cells. Of the mutants, the shorter form (Δ238) exhibited faster migration and invasion than Δ560 cells. In the Transwell, all of the cells invaded faster through Myogel than Matrigel® coated wells. In 3D sandwich, AK23 antibody inhibited only the invasion of FL cells. We conclude that different experimental 2D and 3D settings can markedly influence the movement of oral carcinoma cells with various Dsg3 modifications.

Novel fixed z-direction (FiZD) kidney primordia and an organoid culture system for time-lapse confocal imaging.

Tissue, organ and organoid cultures provide suitable models for developmental studies, but our understanding of how the organs are assembled at the single-cell level still remains unclear. We describe here a novel fixed z-direction (FiZD) culture setup that permits high-resolution confocal imaging of organoids and embryonic tissues. In a FiZD culture a permeable membrane compresses the tissues onto a glass coverslip and the spacers adjust the thickness, enabling the tissue to grow for up to 12 days. Thus, the kidney rudiment and the organoids can adjust to the limited z-directional space and yet advance the process of kidney morphogenesis, enabling long-term time-lapse and high-resolution confocal imaging. As the data quality achieved was sufficient for computer-assisted cell segmentation and analysis, the method can be used for studying morphogenesis ex vivo at the level of the single constituent cells of a complex mammalian organogenesis model system.

A novel human leiomyoma tissue derived matrix for cell culture studies.

BACKGROUND: The composition of the matrix molecules is important in in vitro cell culture experiments of e.g. human cancer invasion and vessel formation. Currently, the mouse Engelbreth-Holm-Swarm (EHS) sarcoma-derived products, such as Matrigel®, are the most commonly used tumor microenvironment (TME) mimicking matrices for experimental studies. However, since Matrigel® is non-human in origin, its molecular composition does not accurately simulate human TME. We have previously described a solid 3D organotypic myoma disc invasion assay, which is derived from human uterus benign leiomyoma tumor. Here, we describe the preparation and analyses of a processed, gelatinous leiomyoma matrix, named Myogel. METHODS: A total protein extract, Myogel, was formulated from myoma. The protein contents of Myogel were characterized and its composition and properties compared with a commercial mouse Matrigel®. Myogel was tested and compared to Matrigel® in human cell adhesion, migration, invasion, colony formation, spheroid culture and vessel formation experiments, as well as in a 3D hanging drop video image analysis. RESULTS: We demonstrated that only 34% of Myogel's molecular content was similar to Matrigel®. All test results showed that Myogel was comparable with Matrigel®, and when mixed with low-melting agarose (Myogel-LMA) it was superior to Matrigel® in in vitro Transwell® invasion and capillary formation assays. CONCLUSIONS: In conclusion, we have developed a novel Myogel TME matrix, which is recommended for in vitro human cell culture experiments since it closely mimics the human tumor microenvironment of solid cancers.