Neoadjuvant chemoradiotherapy (nCRT) combined with surgery is a standard therapy for locally advanced rectal cancer (LARC). The aim of this study was to assess the expression of GOLPH3 (Golgi phosphoprotein 3), a newly found oncogene, in LARC as well as its relationship with nCRT sensitivity and prognosis. We retrospectively analyzed 148 LARC cases receiving nCRT and total mesorectal excision (TME). Immunohistochemistry was used to assess GOLPH3 and mTOR (mammalian target of rapamycin) in tumor tissues. Then, the associations of GOLPH3 with pathological characteristics and prognosis of rectal cancer were assessed. The 148 cases included 77 with high GOLPH3 expression (52.03%), which was associated with tumor invasive depth and lymphatic metastasis. Cases with high GOLPH3 expression had 2.58 and 2.71 fold higher local relapse and distant metastasis rates compared with the low expression group. Correlation analyses showed that GOLPH3 was an independent indicator for judging tumor down-staging and postoperative TRG (tumor regression grade), indicating it could predict nCRT sensitivity. In addition, GOLPH3 expression was associated with mTOR levels. Multiple-factor analysis indicated that GOLPH3 was an independent prognosis indicator for 5 year-DFS (disease free survival) and OS (overall survival) in LARC. These results reveal that GOLPH3 is an independent predictive factor for nCRT sensitivity and prognosis in LARC, with a mechanism related to mTOR.

Abstract. The aim of the present study was to investigate the association between serum tumor markers and the metabolic tumor volume (MTV) or total lesion glycolysis (TLG), as determined by fluorine-18 fluorodeoxyglucose ( 18 F-FDG) positron emission tomography-computed tomography (PET/CT) in patients with recurrent small cell lung cancer (SCLC). Data from 21 patients with recurrent SCLC were collected. The levels of neuron-specific enolase (NSE), carcinoembryonic antigen (CEA) and cytokeratin 19 fragment 21-1 were measured at the time of the 18 F-FDG PET/CT examination. The MTV and TLG of all lesions were calculated. Pearson correlation analyses were used to estimate the correlations between NSE level and PET findings. Pearson correlation analyses showed that NSE was the only tumor marker to have a strong correlation with MTV or TLG (r=0.787, P<0.001; r=0.866, P<0.001, respectively). In patients with a normal NSE level, no correlation was found between NSE and MTV or TLG (r=0.018, P=0.958; r=-0.003, P=0.92, respectively), but a significant correlation was found in patients with an abnormal NSE level (r=0.789, P<0.01; r=0.872, P=0.01, respectively). Therefore, TLG and MTV may serve as sensitive markers of tumor burden in patients with recurrent SCLC, with TLG showing greater sensitivity. In patients with an abnormal NSE level, a higher NSE level indicates greater MTV and TLG.

Background Routinely delineating of important skeletal growth centers is imperative to mitigate radiation‐induced growth abnormalities for pediatric cancer patients treated with radiotherapy. However, it is hindered by several practical problems, including difficult identification, time consumption, and inter‐practitioner variability. Purpose The goal of this study was to construct and evaluate a novel Triplet‐Attention U‐Net (TAU‐Net)‐based auto‐segmentation model for important skeletal growth centers in childhood cancer radiotherapy, concentrating on the accuracy and time efficiency. Methods A total of 107 childhood cancer patients fulfilled the eligibility criteria were enrolled in the training cohort (N = 80) and test cohort (N = 27). The craniofacial growth plates, shoulder growth centers, and pelvic ossification centers, with a total of 19 structures in the three groups, were manually delineated by two experienced radiation oncologists on axial, coronal, and sagittal computed tomography images. Modified from U‐Net, the proposed TAU‐Net has one main branch and two bypass branches, receiving semantic information of three adjacent slices to predict the target structure. With supervised deep learning, the skeletal growth centers contouring of each group was generated by three different auto‐segmentation models: U‐Net, V‐Net, and the proposed TAU‐Net. Dice similarity coefficient (DSC) and Hausdorff distance 95% (HD95) were used to evaluate the accuracy of three auto‐segmentation models. The time spent on performing manual tasks and manually correcting auto‐contouring generated by TAU‐Net was recorded. The paired t‐test was used to compare the statistical differences in delineation quality and time efficiency. Results Among the three groups, including craniofacial growth plates, shoulder growth centers, and pelvic ossification centers groups, TAU‐Net had demonstrated highly acceptable performance (the average DSC = 0.77, 0.87, and 0.83 for each group; the average HD95 = 2.28, 2.07, and 2.86 mm for each group). In the overall evaluation of 19 regions of interest (ROIs) in the test cohort, TAU‐Net had an overwhelming advantage over U‐Net (63.2% ROIs in DSC and 31.6% ROIs in HD95, p = 0.001–0.042) and V‐Net (94.7% ROIs in DSC and 36.8% ROIs in HD95, p = 0.001–0.040). With an average time of 52.2 min for manual delineation, the average time saved to adjust TAU‐Net‐generated contours was 37.6 min (p < 0.001), a 72% reduction. Conclusions Deep learning–based models have presented enormous potential for the auto‐segmentation of important growth centers in pediatric skeleton, where the proposed TAU‐Net outperformed the U‐Net and V‐Net in geometrical precision for the majority status.