Development and Validation of Prognostic Models Using Radiomic Features from Pre-Treatment Positron Emission Tomography (PET) Images in Head and Neck Squamous Cell Carcinoma (HNSCC) Patients.

High-dimensional radiomics features derived from pre-treatment positron emission tomography (PET) images offer prognostic insights for patients with head and neck squamous cell carcinoma (HNSCC). Using 124 PET radiomics features and clinical variables (age, sex, stage of cancer, site of cancer) from a cohort of 232 patients, we evaluated four survival models-penalized Cox model, random forest, gradient boosted model and support vector machine-to predict all-cause mortality (ACM), locoregional recurrence/residual disease (LR) and distant metastasis (DM) probability during 36, 24 and 24 months of follow-up, respectively. We developed models with five-fold cross-validation, selected the best-performing model for each outcome based on the concordance index (C-statistic) and the integrated Brier score (IBS) and validated them in an independent cohort of 102 patients. The penalized Cox model demonstrated better performance for ACM (C-statistic = 0.70, IBS = 0.12) and DM (C-statistic = 0.70, IBS = 0.08) while the random forest model displayed better performance for LR (C-statistic = 0.76, IBS = 0.07). We conclude that the ML-based prognostic model can aid clinicians in quantifying prognosis and determining effective treatment strategies, thereby improving favorable outcomes in HNSCC patients.

Comparison of semi-automatic and manual segmentation methods for tumor delineation on head and neck squamous cell carcinoma (HNSCC) positron emission tomography (PET) images.

Objective. Accurate and reproducible tumor delineation on positron emission tomography (PET) images is required to validate predictive and prognostic models based on PET radiomic features. Manual segmentation of tumors is time-consuming whereas semi-automatic methods are easily implementable and inexpensive. This study assessed the reliability of semi-automatic segmentation methods over manual segmentation for tumor delineation in head and neck squamous cell carcinoma (HNSCC) PET images.Approach. We employed manual and six semi-automatic segmentation methods (just enough interaction (JEI), watershed, grow from seeds (GfS), flood filling (FF), 30% SUVmax and 40%SUVmax threshold) using 3D slicer software to extract 128 radiomic features from FDG-PET images of 100 HNSCC patients independently by three operators. We assessed the distributional properties of all features and considered 92 log-transformed features for subsequent analysis. For each paired comparison of a feature, we fitted a separate linear mixed effect model using the method (two levels; manual versus one semi-automatic method) as a fixed effect and the subject and the operator as the random effects. We estimated different statistics-the intraclass correlation coefficient agreement (aICC), limits of agreement (LoA), total deviation index (TDI), coverage probability (CP) and coefficient of individual agreement (CIA)-to evaluate the agreement between the manual and semi-automatic methods.Main results. Accounting for all statistics across 92 features, the JEI method consistently demonstrated acceptable agreement with the manual method, with median values of aICC = 0.86, TDI = 0.94, CP = 0.66, and CIA = 0.91.Significance. This study demonstrated that JEI method is a reliable semi-automatic method for tumor delineation on HNSCC PET images.

Apposition to endometrial epithelial cells activates mouse blastocysts for implantation.

STUDY QUESTION: How do interactions between blastocyst-stage embryos and endometrial epithelial cells regulate the early stages of implantation in an in vitro model? SUMMARY ANSWER: Mouse blastocyst apposition with human endometrial epithelial cells initiates trophectoderm differentiation to trophoblast, which goes on to breach the endometrial epithelium. WHAT IS KNOWN ALREADY: In vitro models using mouse blastocysts and human endometrial cell lines have proven invaluable in the molecular characterisation of embryo attachment to endometrial epithelium at the onset of implantation. Genes involved in embryonic breaching of the endometrial epithelium have not been investigated in such in vitro models. STUDY DESIGN, SIZE, DURATION: This study used an established in vitro model of implantation to examine cellular and molecular interactions during blastocyst attachment to endometrial epithelial cells. PARTICIPANTS/MATERIALS, SETTING, METHODS: Mouse blastocysts developed from embryonic day (E) 1.5 in vitro were hatched and co-cultured with confluent human endometrial adenocarcinoma-derived Ishikawa cells in serum-free medium. A scale of attachment stability based on blastocyst oscillation upon agitation was devised. Blastocysts were monitored for 48 h to establish the kinetics of implantation, and optical sectioning using fluorescence microscopy revealed attachment and invasion interfaces. Quantitative PCR was used to determine blastocyst gene expression. Data from a total of 680 mouse blastocysts are reported, with 3-6 experimental replicates. T-test and ANOVA analyses established statistical significance at P < 0.05, P < 0.01 and P < 0.001. MAIN RESULTS AND THE ROLE OF CHANCE: Hatched E4.5 mouse blastocysts exhibited weak attachment to confluent Ishikawa cells over the first 24 h of co-culture, with intermediate and stable attachment occurring from 28 h (E5.5 + 4 h) in a hormone-independent manner. Attached embryos fixed after 48 h (E6.5) frequently exhibited outgrowths, characterised morphologically and with antibody markers as trophoblast giant cells (TGCs), which had breached the Ishikawa cell layer. Beginning co-culture at E5.5 also resulted in intermediate and stable attachment from E5.5 + 4 h; however, these embryos did not go on to breach the Ishikawa cell layer, even when co-culture was extended to E7.5 (P < 0.01). Blastocysts cultured from E4.5 in permeable transwell inserts above Ishikawa cells before transfer to direct co-culture at E5.5 went on to attach but failed to breach the Ishikawa cell layer by E6.5 (P < 0.01). Gene expression analysis at E5.5 demonstrated that direct co-culture with Ishikawa cells from E4.5 resulted in downregulation of trophectoderm transcription factors Cdx2 (P < 0.05) and Gata3 (P < 0.05) and upregulation of the TGC transcription factor Hand1 (P < 0.05). Co-culture with non-endometrial human fibroblasts did not alter the expression of these genes. LARGE SCALE DATA: None. LIMITATIONS, REASONS FOR CAUTION: The in vitro model used here combines human carcinoma-derived endometrial cells with mouse embryos, in which the cellular interactions observed may not fully recapitulate those in vivo. The data gleaned from such models can be regarded as hypothesis-generating, and research is now needed to develop more sophisticated models of human implantation combining multiple primary endometrial cell types with surrogate and real human embryos. WIDER IMPLICATIONS OF THE FINDINGS: This study implicates blastocyst apposition to endometrial epithelial cells as a critical step in trophoblast differentiation required for implantation. Understanding this maternal regulation of the embryonic developmental programme may lead to novel treatments for infertility. STUDY FUNDING AND COMPETING INTEREST(S): This work was supported by funds from the charities Wellbeing of Women (RG1442) and Diabetes UK (15/0005207), and studentship support for SCB from the Anatomical Society. No conflict of interest is declared.