Tumor precision medicine is an emerging approach for tumor diagnosis, treatment and prevention, which takes account of individual variability of environment, lifestyle and genetic information. Tumor precision medicine is built up on the medical imaging innovations developed during the past decades, including the new hardware, new imaging agents, standardized protocols, image analysis and multimodal imaging fusion technology. Also the development of automated and reproducible analysis algorithm has extracted large amount of information from image-based features. With the continuous development and mining of tumor clinical and imaging databases, the radiogenomics, radiomics and artificial intelligence have been flourishing. Therefore, these new technological advances bring new opportunities and challenges to the application of imaging in tumor precision medicine.

Pancreatic cancer is a highly malignant tumor and surgical resection is the only curative treatment option. In order to improve the prognosis of patients with pancreatic cancer, it is very important to diagnose and evaluate pancreatic cancer accurately and early. Imaging examinations play an important role in tumor detection, staging and surgical resectability evaluation, which can provide reliable evidences for diagnosis and treatment of pancreatic cancer. At present, the imaging techniques commonly used for diagnosis of pancreatic cancer include conventional ultrasound, endoscopic ultrasonography, PET-CT, multi-detector row CT (MDCT), magnetic resonance imaging (MRI), in which MDCT and MRI are the most widely used. In this article, the application and research progress of imaging in accurate diagnosis and evaluation of pancreatic cancer are reviewed.

Intratumor spatial heterogeneity has been ignored in evaluating tumor therapeutic effect with conventional imaging methods. The voxel-based parametric response map (PRM) derived from CT or MRI may accurately, objectively and timely evaluate the tumor response to therapy, by comparing changes of the functional parameters before and after treatment. This approach may improve the imaging monitoring and the sensitivity of spatial resolution related to tumor changes after treatment. Thus, PRM may help to improve the treatment plan and prognosis evaluation for patients. This article reviews progress on CT/MRI PRM in precision evaluation of therapeutic response of cancer patients.

In recent years, PET-CT has an increasing importance in the diagnosis and treatment of breast cancer. PET-CT scan can be used as a noninvasive method for molecular subtyping of breast cancer, and prediction of therapeutic effect and prognosis of patients. Studies have revealed that luminal A subtype has a significantly lower maximum standard intake value (SUVmax) than the other subtypes; triple-negative and human epidermal growth factor receptor 2 (HER2) positive tumors have relatively high SUVmax than luminal B subtype, but the specificity and sensitivity of SUVmax in diagnosis of molecular subtypes are very low, so its clinical application is limited. In predicting the effectiveness of the treatment and the prognosis of the patients, the decreased uptake of fluorodeoxyglucose (FDG) is correlated with better therapeutic effect. In addition, patients with high FDG uptake have worse survival outcomes. New tracers, such as ¹⁸F-fluoroestradiol (¹⁸F-FES) and[89Zr]trastuzumab play an important role in molecular subtyping of breast cancer. ¹⁸F-FES PET-CT can effectively evaluate the estrogen receptor (ER) status of breast cancer and the response to endocrine therapy.[89Zr]trastuzumab PET-CT can evaluate the expression of HER2 and localization of HER2-overexpressing tumors, but their specificities and sensitivities are also low. In this article, we review the recent advances on the correlation of PET-CT findings with molecular subtypes, treatment response and prognosis of breast cancer.

Objective: To investigate the feasibility of labeling endothelial progenitor cells (EPCs) with a novel dual modal contrast agent Molday ION^TM EverGreen(MIEG) and its performance in vitro MRI.Methods: EPCs were isolated from rat bone marrow and labeled with 10, 20, 50 μg/mL MIEG, respectively. The labeling rates were identified by Prussian blue staining and fluorescence microscopy. The vitality of EPCs labeled with 20 μg/mL MIEG was detected by trypan blue exclusion test at 1 d, 1 w, 2 w, and 6 w after labeling. EPCs labeled with different concentrations of MIEG were scanned by 3.0T MRI with T₁ weighted and T₂ weighted imaging.Results: The labeling rates for EPCs labeled with different concentrations of MIEG were greater than 98%,and the cytoplasm of labeled EPCs was present with Prussian blue staining. Although the green lighting level went down, the labeling rate at 6 w after labeling was greater than 90%. Trypan blue exclusion test showed that there was no significant difference in the vitality between EPCs labeled with MIEG at 1 d, 1 w, 2 w and 6 w after labeling and EPCs without labeling (all P>0.05). There was no difference in signal intensity on T₁ weighted image among EPCs labeled with different concentrations of MIEG. However, the signal intensity on T₂ weighted image was reduced in all labeled groups, and the signal reduction became more apparent with increased concentration of MIEG. Conclusions: EPCs can be effectively labeled by MIEG without interference on the cell viability at the labeled concentration of 20 μg/mL. The signal intensity change of labeled cells can be detected sensitively by T₂ weighted imaging at 3.0T MRI.

Objective: To explore the correlation between the number of previous gadolinium-based contrast agent administrations and high signal intensity (SI) in the cerebrum nucleus on unenhanced T₁-weighted magnetic resonance images.Methods: Thirty-nine patients who previously underwent at least three contrast-enhanced brain MRI examinations were enrolled in the study. The right globus pallidus, right thalamus, right dentate nucleus, pons and white matter of right frontal lobe were selected as region of interests (ROI). The mean SIs of the ROI were measured on unenhanced T₁-weighted images. The SI ratios of globus pallidus, right thalamus, right dentate nucleus and pons to white matter were calculated, respectively. The correlation of SI ratio variation with the number of previous contrast agent administration and clinical features was analyzed with Spearman analysis.Results: The dentate nucleus-to-white matter SI ratios of the last and first MRI scan were 1.113±0.136 and 1.014±0.096(P<0.01), respectively; while the SI ratio differences between the last and first MRI scan of globus pallidus-to-white matter, the thalamus-to-white matter and the pons-to-white matter were not significant (all P>0.05). The dentate nucleus-to-white matter SI ratio was correlated with the number of previous gadolinium-based contrast agent administrations (r=0.329, P<0.05), but not with age, sex, underlying disease, treatment and history of diabetes (all P>0.05).Conclusion: High SI in the dentate nucleus on unenhanced T₁-weighted images may be a consequence of the number of previous gadolinium-based contrast agent administrations.

Objective: To assess the value of contrast enhanced MRI features for predicting epidermal growth factor receptor (EGFR) gene amplification in glioblastoma multiforme (GBM) with radiomic method.Methods: Eighty patients with EGFR status examined GBM were retrospectively reviewed. The data were randomly divided into a training dataset (60%) and test dataset (40%). Texture features of each case were extracted from the enhanced region and the edema region in contrast enhanced MR images. Principal component analysis was used for dimension reduction. Random forest model, support vector machine model and neural network model were built. Area under the curve (AUC) of the receiver operating characteristics curve was used to assess the performance of models with test dataset.Results: A total of 542 features were extracted from the enhanced region and the edema region. Forty-eight principal components were obtained, which accounted for 100% accumulation contribution rate, and the first 31 principal components were selected for models building, which accounted for 98.5% accumulation contribution rate. The values of AUCs were 0.74, 0.69 and 0.63 for random forest model, support vector machine model and neural network model in the test dataset, respectively.Conclusion: Radiomic method with proper model may have a potential role in predicting the EGFR gene status with enhanced MRI features derived from the enhanced region and the edema region in patients with glioblastoma multiforme.

Objective: To assess the application of the dynamic-contrast enhanced magnetic resonance imaging(DCE-MRI)pharmacokinetics models in differential diagnosis of cellular uterine leiomyoma.Methods: Sixty four patients with uterine leiomyoma confirmed by surgery and pathology were enrolled in the study between September 2015 and September 2016, including 30 cases of classical leiomyoma, 13 cases of cellular leiomyoma and 21 cases of degenerative leiomyoma. All patients underwent DCE-MRI before surgery. Reference region (RR) model, extended tofts (ET) model and exchange (EC) model were used to quantitatively analyze DCE-MRI data, and their differences among different pathological types of uterine leiomyoma were observed. Receiver operating characteristic (ROC) curve was used to evaluate the efficiency of the quantitative perfusion parameters in differential diagnosis of cellular uterine leiomyoma.Results: The values of K^trans(transfer constant), K_ep(efflux rate constant) in RR model, K^trans, K_ep, V_p (blood plasma volume ratio) in ET model and V_e(plasma volume ratio), F_p(plasma flow)in EC model of cellular uterine leiomyoma were higher than those of classical type(P<0.05 or P<0.01). The values of K^trans, K_ep in RR model,K^trans,K_ep, V_e,V_p in ET model and V_e,V_p,F_p in EC model of cellular uterine leiomyoma were higher than those of degenerative uterine leiomyoma(P<0.05 or P<0.01). There were no significant differences in other quantitative perfusion parameters among three types of uterine leiomyoma (all P>0.05). ROC curves revealed that the K^trans of the ET model was more effective in diagnosing cellular uterine leiomyoma, the area under the curve (AUC) was 0.929, and the sensitivity and specificity were 92.3% and 83.7%, respectively; meanwhile, the AUCs of F_p of the EC model, K^trans of the RR model and K_ep of the ET model in diagnosis of cellular uterine leiomyoma were 0.867, 0.849 and 0.837, the sensitivities were 91.7%, 84.6% and 92.3%, and the specificities were 78.0%, 76.0% and 73.5%, respectively. Conclusions: Three pharmacokinetics models can be used in the differentiation of cellular uterine leiomyoma from other types of uterine leiomyoma. K^trans of the ET model has higher sensitivity and specificity in differential diagnosis of cellular uterine leiomyoma.

Objective: To investigate the association of parameters in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) using reference region model with prognostic factors and molecular subtypes of breast cancer.Methods: MRI and pathological data of 50 patients with pathologically confirmed invasive ductal carcinoma of the breast were retrospectively analyzed. Reference region model was applied to analyze pharmacokinetic quantitative parameters including volume transfer constant (RR K^trans), rate constant (K_ep) and the ratio of K^trans to extracellular space volume (K^trans/V_e). The associations of the above parameters with prognostic factors and molecular subtypes of breast cancer were analyzed.Results: RR K^trans and K_ep were significantly higher in patients of histological grade 3 compared with those of histological grade 1 & 2 (all P<0.05); and the patients with estrogen receptor (ER)-negative and/or progesterone receptor (PR)-negative also had higher RR K^trans and K_ep than those with ER-positive or PR-positive (all P<0.05). For immunohistochemistry, RR K^trans and K_ep were significantly higher in triple negative breast cancer compared with luminal type breast cancer (all P<0.05).Conclusion: High RR K^trans and K_ep are associated with poor prognosis of breast cancer, and which can also be used to distinguish molecular subtypes of breast cancer.

Objective: To assess the diagnostic value of dual energy CT for lymph node metastasis in patients with non-small cell lung cancer (NSCLC).Methods: Forty NSCLC patients, including 15 cases of squamous cell carcinoma and 25 cases of adenocarcinoma, underwent dual energy CT examination in pre-contrast and venous phase contrast scans, then the CT attenuation value of the lung cancer lesions and 85 mediastinal enlarged lymph nodes (the short diameter ≥ 5 mm, 53 metastatic and 32 non-metastatic) were measured at different energy levels (40-190 keV, spacing 10 keV) in venous phase contrast. CT spectral curves of the lung cancer lesions, hilus pulmonis and mediastinal enlarged lymph nodes were produced automatically, through comparing their CT spectral curves slope to judge whether or not the lymph nodes were metastatic. Receiver operating characteristic (ROC) curve was used to evaluate the efficiency of CT spectral curve in diagnosis of lymph node metastasis.Results: The CT spectral curves slopes of the lung cancer, metastatic lymph nodes and non-metastatic lymph nodes were 1.10±0.11, 1.08±0.07 and 1.54±0.17, respectively. There was no significant difference in curve slope between metastatic lymph nodes and lung cancer (t=-1.32,P>0.05); while there was significant difference between non-metastatic lymph nodes and lung cancer (t=-2.58,P<0.05). The CT spectral curve slope ratios of metastatic and non-metastatic lymph nodes to lung cancer were 0.98±0.05 and 1.40±0.12, respectively (t=-2.86,P<0.05). ROC curve showed that taking CT spectral curve slope ratio of 1.15 as cut-off value for the diagnosis of metastatic lymph nodes, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 81.1%, 87.5%, 91.5%, 73.7% and 83.5%, respectively.Conclusion: Dual energy CT is of value in improving the diagnostic accuracy of lymph node metastasis in NSCLC patients before treatment.

Objective: To assess the value of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography-CT (PET-CT) in detection of primary tumor and pelvic lymph node metastasis in International Federation of Gynecology and Obstetrics (FIGO) stages ⅠA2-ⅡA cervical cancer.Methods: The clinical data of 91 patients with FIGO stagesⅠA2-ⅡA cervical cancer were retrospectively analyzed. The sensitivity of ¹⁸F-FDG PET-CT in detection of cervical cancer was calculated. The long diameter, short diameter and SUVmax were compared between metastatic lymph nodes (MLN) and non-metastatic lymph nodes (NMLN). The optimal cut-off values of different indexes were determined by receiver operating characteristic (ROC) curve and area under curve (AUC), and the sensitivity, specificity and accuracy were calculated.Results: ¹⁸F-FDG PET-CT examinations were positive in 89 patients (89/91, 96.7%). The short diameters of NMLN and MLN were (6.50±2.31)mm and (4.21±1.49)mm(t=4.855, P<0.05); the SUVmax of NMLN and MLN were 4.56±3.34 and 1.92±1.41(t=31.685, P<0.05). ROC AUCs of the short diameter and SUVmax in diagnosis of metastatic lymph nodes were 0.802 and 0.861. Taken short diameter ≥ 5.05 mm and SUVmax ≥ 2.05 as cut-off values, the corresponding sensitivity, specificity and accuracy in diagnosis of metastatic lymph nodes were 85.0%, 93.0% and 86.8%, respectively.Conclusion: ¹⁸F-FDG PET-CT is sensitive to detect primary lesion and pelvic lymph node metastases in FIGO stages ⅠA2-Ⅱ A cervical cancer, and the highest diagnostic accuracy may be obtained by taking short diameter ≥ 5.05 mm and SUVmax ≥ 2.05 as the standard.

Objective: To investigate the expression and prognostic value of memory T lymphocyte in patients with non-small cell lung cancer(NSCLC) following radiotherapy.Methods: Forty-six patients with NSCLC receiving radiotherapy in Ningbo Medical Center Lihuili Hospital from February 2010 to May 2012 were enrolled in the study and 50 healthy subjects served as the control group. The central memory T cell (T_CM) and effector memory T cell (T_EM) in peripheral blood CD4⁺, CD8⁺ cells were detected by flow cytometry. Survival of patients was analyzed by Kaplan-Meier curve, and the relationship between clinical features, memory T lymphocyte changes and overall survival was analyzed by multivariate Cox regression model.Results: CD4⁺T_CM, CD4⁺T_EM, CD8⁺T_CM levels and CD4⁺/CD8⁺ T_EM of NSCLC patients were significantly lower than those of the control group, while CD4⁺/CD8⁺ T_CM was significantly higher than that of the control group(all P<0.05). In NSCLC patients, CD4⁺T_CM, CD4⁺T_EM and CD8⁺T_CM were decreased and CD8⁺T_EM levels were increased 4 weeks after radiotherapy(all P<0.05); CD4⁺T_CM, CD4⁺T_EM and CD8⁺T_CM at 12-week after radiotherapy were increased significantly compared with those at 4-week after radiotherapy(all P<0.05). Multivariate Cox regression analysis showed that the change of CD4⁺T_CM after radiotherapy was correlated with the overall survival (95% CI:1.135-2.994, P<0.01). The survival rate and overall survival time for patients with decreasing CD4⁺ T_CM were 23.1% and 10.7 months (95% CI:0.29-12.41), while those of patients with stable CD4⁺ T_CM were 52.7% and 27.4 months (95% CI:0.00-31.26), and those of patients with increasing CD4⁺ T_CM were 66.4% and 37.4 months (95% CI:0.33-29.21), respectively. Conclusions: NSCLC patients show a significant immunosuppression at the initial stage after radiotherapy, and then a gradual improvement. Change of memory T lymphocyte after radiotherapy can be used to help predicting the prognosis of the patients.

Objective: To analyze the clinical features, diagnosis and treatment of gestational trophoblastic diseases in cesarean scar.Methods: Clinical data of three cases of gestational trophoblastic diseases in cesarean scar diagnosed in Women's Hospital, Zhejiang University School of Medicine during December 2011 and December 2016 were collected. And literature search was performed in Wanfang data, VIP, CNKI, PubMed, ISI Web of Knowledge and EMbase database.Results: A total of 20 cases of gestational trophoblastic diseases were included in the analysis. Clinical features were mainly abnormal vaginal bleeding after menopause, artificial abortion or medical abortion, which might be accompanied by abdominal pain. Serum β-human chorionic gonadotropin (β-hCG) levels were increased in 19 patients. The sonographic features were increase of uterine volume, honeycomb-like abnormal intrauterine echo (or described as multiple cystic dark area, multiple anechoic area and multiple liquid dark area) or heterogeneity echo conglomeration, and no clear bound with muscular layer in some cases. There were abundant blood flow signals inside or around the lesions. The ultrasonography indicated that the lesions were located in the anterior side of the uterine isthmus with the involvement of cesarean section scar. In 12 cases with lesions in cesarean scar shown by preliminary diagnosis, 9 underwent uterine artery embolization (UAE) for pretreatment; the blood loss greater than 1500 mL was observed in only one case without UAE; no patient received hysterectomy. In 8 patients whose lesions were not shown in cesarean scar, only one case received UAE pretreatment, and hysterectomy was performed in 3 cases due to blood loss greater than 1500 mL. Two cases were lost in follow-up and no death was reported in remaining 18 cases. The serum β-hCG levels returned to normal or satisfactory level during the follow-up in 17 cases with increased β-hCG levels before treatment and no recurrence was observed. Conclusions: The misdiagnosis rate and missed diagnosis rate of gestational trophoblastic diseases in cesarean section scar are high. The identification of cesarean section scar involvement and UAE may reduce the bleeding and avoid hysterectomy.

Objective: To analyze the geographic spatial patterns and risk areas of main digestive system cancers in Yiwu city.Methods: Newly diagnosed cases of esophageal, gastric and colorectal cancer during 2010-2014 were obtained from Yiwu Center for Disease Control and Prevention (CDC). The household registration population data in 2013 were obtained from public security bureau. Hierarchy clustering and partitioning regionalization method was used to generate geographic units. Global Moran's I was used to evaluate whether cancer incidence was significantly clustered in space, Anselin Local Moran's I was used to identify statistically significant hot spots, cold spots, and spatial outliers, and Spatial Scan Statistics was implemented to analyze the relative risk of cancers in different areas.Results: The 5-year average incidence of esophageal, gastric and colorectal cancers were 9.99/100 000, 34.01/100 000 and 31.46/100 000, respectively. Males showed significantly higher incidence than females. The incidence was heterogeneous throughout the study area. Spatial Scan analysis revealed that southern Yiwu presented a significantly higher male esophageal cancer (RR=1.78) and gastric cancer (RR=1.87) risk. The central area of Yiwu showed a significantly lower female esophageal cancer risk (RR=0.00) and male stomach cancer risk (RR=0.63) and the northern Yiwu exhibited a significantly lower female colorectal cancer risk (RR=0.48).Conclusion: The incidence of main digestive tract cancers shows a heterogeneous distribution in Yiwu city.

Objective: To evaluate the application of CT scan in diagnosis of pathological types and origins of metastatic ovarian tumors.Methods: Clinical data, histopathological results and CT images of 43 patients with pathologically-proved metastatic ovarian tumor were retrospectively analyzed. Diagnostic values of CT imaging for pathological type and origin of metastatic ovarian tumors were evaluated.Results: The pathological types of metastatic ovarian tumor were related to the size of the lesion (P<0.01), while not related to the sites of lesion (unilateral or bilateral), the cystic-solid and mixed lesions with or without separation (all P>0.05). Metastatic ovarian tumors of colorectal origin were usually unilateral lesions, and showed cystic or cystic-solid masses, while those of gastric origin were usually bilateral lesions, and showed solid or solid-based masses.Conclusion: CT imaging may be of value in diagnosis of pathological types and origin of metastatic ovarian tumor.

Objective: To assess the value of ultrasonography in differential diagnosis of xanthogranulomatous cholecystitis (XGC) and wall-thickening type of gallbladder cancer (GBC).Methods: The clinical features and sonographic finding of 31 patients with XGC and 36 patients with wall-thickening type of GBC were retrospectively reviewed. The diagnosis of all cases was confirmed by pathological examination, and the ultrasonographic manifestations of the thickening of the gallbladder wall, intramural hypoechoic nodules, gallbladder mucosa line, gallbladder stones, biliary dilatation and gallbladder-liver boundary were compared between two groups.Results: There were significant differences in the intramural hypoechoic nodules, the continuous gallbladder wall mucosal line and dilatation of bile duct between XGC and GBC groups (all P<0.05), while no statistically significant differences in the other sonographic features(all P>0.05). In the six positive sonographic features of the XGC patients, the intramural hypoechoic nodules and the continuous mucosa line of the gallbladder wall had highest accuracy in the diagnosis of XGC(64.2% and 65.7%).Conclusion: Intramural hypoechoic nodule and the continuous mucosal line are characteristic sonographic features of XGC, which can be used for the diagnosis of XGC.

Objective: To compare the diagnostic value of double-balloon endoscopy (DBE) and multi-detector row computed tomography (MDCT) for small intestinal tumors.Methods: The clinical data of 88 patients with small intestinal tumor who underwent DBE and MDCT examinations during January 2010 and December 2016 was retrospectively analyzed. The diagnostic value of DBE and MDCT for small intestinal tumor was compared.Results: The positive rate of small intestinal tumors by DBE (92.0%, 81/88) was significantly higher than that of MDCT (75.0%, 66/88, P<0.01). The positive rates for malignant small intestinal tumors by DBE and MDCT were the same (91.1%), but the rate of histological qualitation was higher in DBE than that in MDCT (94.1% vs 76.5%, χ²=6.331, P<0.05); the positive rates for benign tumors or the tumors <3 cm in diameter by DBE were significantly higher than those by MDCT (96.8% vs 48.4%, 88.2% vs 47.1%, respectively, χ²=18.235 and 13.170, both P<0.01); also DBE had a higher positive rate for small intestinal tumor presenting gastrointestinal bleeding (97.2% vs 77.8%, χ²=6.222, P<0.05). Conclusions: The diagnostic value of small intestinal tumor by DBE is significantly higher than that of MDCT. The combination of two methods may improve the diagnostic accuracy for small intestinal tumor.

Pneumocystis jirovecii pneumonia is an opportunistic infection among immunocompromised people. Studies have found that the increased resistance to sulfa drugs of Pneumocystis jirovecii may be associated with the mutation of dihydropteroate synthase (DHPS) gene and dihydrofolate reductase (DHFR) genes, but the mechanism is still unclear. The mutation of DHPS and DHFR genes may be the result of sulfa drugs selection or spontaneous genetic polymorphism, and it can be acquired from person-to-person transmission. This article reviews the cause, molecular epidemiology of DHPS and DHFR gene mutation, and the relationship between DHPS and DHFR gene mutation and clinical outcomes.