Nonbenign Tumors Research Articles

Preliminary Results of Deep Learning Approach for Preoperative Diagnosis of Ovarian Cancer Based on Pelvic MRI Scans.

To predict the pathological diagnosis of ovarian tumors using preoperative MRI images, using deep learning models. A total of 185 patients were enrolled, including 40 with ovarian cancers, 25 with borderline malignant tumors, and 120 with benign tumors. Using sagittal and horizontal T2-weighted images (T2WI), we constructed the pre-trained convolutional neural networks to predict pathological diagnoses. The performance of the model was assessed by precision, recall, and F1-score on macro-average with 95% confidence interval (95%CI). The accuracy and area under the curve (AUC) were also assessed after binary transformation by the division into benign and non-benign groups. The macro-average accuracy in the three-class classification was 0.523 (95%CI=0.504-0.544) for sagittal images and 0.426 (95%CI=0.404-0.446) for horizontal images. The model achieved a precision of 0.63 (95%CI=0.61-0.66), recall of 0.75 (95%CI=0.72-0.78), and F1 score of 0.69 (95%CI=0.67-0.71) for benign tumor. Regarding the discrimination between benign and non-benign tumors, the accuracy in the binary-class classification was 0.628 (95%CI=0.592-0.662) for sagittal images and AUC was 0.529 (95%CI=0.500-0.557). Using deep learning, we could perform pathological diagnosis from preoperative MRI images.

Application of Deep Convolutional Neural Networks for Discriminating Benign, Borderline, and Malignant Serous Ovarian Tumors From Ultrasound Images.

ObjectiveThis study aimed to evaluate the performance of the deep convolutional neural network (DCNN) to discriminate between benign, borderline, and malignant serous ovarian tumors (SOTs) on ultrasound(US) images.Material and MethodsThis retrospective study included 279 pathology-confirmed SOTs US images from 265 patients from March 2013 to December 2016. Two- and three-class classification task based on US images were proposed to classify benign, borderline, and malignant SOTs using a DCNN. The 2-class classification task was divided into two subtasks: benign vs. borderline & malignant (task A), borderline vs. malignant (task B). Five DCNN architectures, namely VGG16, GoogLeNet, ResNet34, MobileNet, and DenseNet, were trained and model performance before and after transfer learning was tested. Model performance was analyzed using accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC).ResultsThe best overall performance was for the ResNet34 model, which also achieved the better performance after transfer learning. When classifying benign and non-benign tumors, the AUC was 0.96, the sensitivity was 0.91, and the specificity was 0.91. When predicting malignancy and borderline tumors, the AUC was 0.91, the sensitivity was 0.98, and the specificity was 0.74. The model had an overall accuracy of 0.75 for in directly classifying the three categories of benign, malignant and borderline SOTs, and a sensitivity of 0.89 for malignant, which was better than the overall diagnostic accuracy of 0.67 and sensitivity of 0.75 for malignant of the senior ultrasonographer.ConclusionDCNN model analysis of US images can provide complementary clinical diagnostic information and is thus a promising technique for effective differentiation of benign, borderline, and malignant SOTs.

The ATA classification and TI‐RADS ACR predict not only benignity but also the histology of nonbenign tumors in thyroid nodules with indeterminate cytology

The <scp>ATA</scp> classification and <scp>TI‐RADS ACR</scp> predict not only benignity but also the histology of nonbenign tumors in thyroid nodules with indeterminate cytology

Comparative Immunohistochemical Study of Galectins 1, 3 and 9 in Ovarian Epithelial Neoplasms

Ovarian neoplasms are among the most common in the female genital tract and often have delayed diagnosis. Tumor progression involves signalling proteins called galectins. The aim of the present study was to evaluate the immunohistochemical expression of galectins “1”, “3” and “9” in the ovarian surface epithelial neoplasia. A retrospective study involving 62 ovarian epithelial tumors (benign and non-benign) was performed with immunohistochemical polymer technique and antibodies against galectin “1”, “3” and “9”. Expression in epithelium and stroma was analysed semi-quantitatively. Fisher’s exact test was performed for statistical analysis. Galectin-“1” and “3” were strongly expressed in non-benign tumors of the epithelium. Non-benign neoplasms showed increased stromal expression of galectin-1 and increased epithelial expression of galectin-“3”. The significant increase in expression of galectin-1 and -3 in the epithelium of non- benign ovarian neoplasms suggests the participation of these galectins in ovarian carcinogenesis. We observed increased stromal expression of galectin-“1” and epithelial expression of galectin-“3” in non-benign ovarian neoplasms. These findings contribute to knowledge about the role of these galectins in the growth and spread of ovarian cancer.

A Retrospective Study of Ovarian Sex Cord Stromal Tumors at the Egyptian National Cancer Institute

Background: Ovarian sex cord-stromal tumors are an uncommon heterogeneous group of tumors with different biological behaviors and clinico-patho- logic aspects. Aim of the Work: This study will review the clinico-pathologic aspects of sex-cord stromal ovarian tumors at the National Cancer Institute (NCI), Cairo University, Egypt, as well as their management and follow-up regarding disease free survival and overall survival. Patients and Methods: This retrospective study was conducted at the National Cancer Institute Cairo University, Egypt on female patients with ovarian sex cord stromal tumors in the period from January 2008 to December 2012 with a follow-up period of 24 to 84 months. The age of the patients, different clinical presentations, radiological findings, associated uterine bleeding (need for endometrial biopsy), pre-operative CA125 levels, surgical management done, different histopathological types, different biological behaviors, presence of ascites (and its correlation with the histopathology), Adjuvant chemotherapy (according to biological behavior and pathological type), and follow-up of non-benign cases for up to 84 months will all be documented and studied. Results: The mean age at presentation was 47.34; abdominal pain and mass were the commonest presentations 54.5% and 53.2% respectively; the main radiologic findings were a pelvic mass +/- ascites which had no correlation to the pathological type (p = 0.075). Endometrial hyperplasia and endometrial carcinoma were associated with 22% and 2.5% of cases respectively. Stages I and II represented 95% of patients with non-benign tumors (48 patients). Panhysterectomy +/- infracolic omentectomy or fertility sparing surgery were done in 70.1% and 29.9% of patients respectively. AGCTs were the commonest pathological type (49.4%). Adjuvant chemotherapy was given to 14 patients (46.7%) with non-benign tumors. 6 recurrences (20%) in 30 patients with non-benign tumors on regular follow-up were documented. The median of disease free survival (DFS) was 50.5 months. The median overall survival was 49.5 months. Conclusion: Ovarian SCSTs are uncommon neoplasms with different biological behaviors where AGCTs are the commonest among Egyptian females. Hormonal manifestations are uncommon where abnormal vaginal bleeding is the commonest one. The presence of ascitic fluid has no correlation with the pathological type of the tumor. Early stages (I and II) represented about 95% of non-benign tumors. Surgical management without lymphadenectomy +/- adjuvant chemotherapy is the main line of treatment at our institute. The OS was shorter than that documented in the literature. A small number of patients, reluctance of follow-up and unavailability of some patients’ data were the main drawbacks in this study.

Quantifying predictive capability of electronic health records for the most harmful breast cancer.

Improved prediction of the "most harmful" breast cancers that cause the most substantive morbidity and mortality would enable physicians to target more intense screening and preventive measures at those women who have the highest risk; however, such prediction models for the "most harmful" breast cancers have rarely been developed. Electronic health records (EHRs) represent an underused data source that has great research and clinical potential. Our goal was to quantify the value of EHR variables in the "most harmful" breast cancer risk prediction. We identified 794 subjects who had breast cancer with primary non-benign tumors with their earliest diagnosis on or after 1/1/2004 from an existing personalized medicine data repository, including 395 "most harmful" breast cancer cases and 399 "least harmful" breast cancer cases. For these subjects, we collected EHR data comprised of 6 components: demographics, diagnoses, symptoms, procedures, medications, and laboratory results. We developed two regularized prediction models, Ridge Logistic Regression (Ridge-LR) and Lasso Logistic Regression (Lasso-LR), to predict the "most harmful" breast cancer one year in advance. The area under the ROC curve (AUC) was used to assess model performance. We observed that the AUCs of Ridge-LR and Lasso-LR models were 0.818 and 0.839 respectively. For both the Ridge-LR and Lasso-LR models, the predictive performance of the whole EHR variables was significantly higher than that of each individual component (p<0.001). In conclusion, EHR variables can be used to predict the "most harmful" breast cancer, providing the possibility to personalize care for those women at the highest risk in clinical practice.

Long-Term Clinical Outcomes of Pencil Beam Scanning Proton Therapy for Benign and Non-benign Intracranial Meningiomas

To assess and report long-term clinical outcomes regarding local control, overall survival, and toxicity-free survival after pencil beam scanning proton therapy for intracranial meningiomas at a single institution. Ninety-six patients (male/female, 29/67; median age 52.8years) with intracranial meningiomas (World Health Organization [WHO] grade 1, n=61 [63.5%]; WHO grade 2, n=33 [34.4%]; WHO grade 3, n=2 [2.1%]) were treated with pencil beam scanning proton therapy (n=53 [55.2%] at diagnosis, n=17 [17.7%] at recurrence, and n=26 [27.1%] for tumor progression). Median gross tumor volume before PBSPT was 21.4cm3 (range, 0.0-546.5cm3), with a median planning target volume of 123.4cm3 (range, 4.6-1142.0cm3). Median duration of follow-up was 56.9months (range, 12.1-207.2months). Late toxicity was graded according to the Common Terminology Criteria for Adverse Events, version 4.0. Thirteen failures (14%) (male/female, 6/7) were observed, of which the majority (n=9, 69%) were of non-benign histology. The 5-year actuarial local control and overall survival were 86.4% and 88.2%, respectively. Five-year grade ≥3 toxicity-free survival was 89.1%. On univariate analysis, local control was worse for patients with higher WHO grade (P≤.001), those treated after at least 1 recurrence (P=.006), those with non-skull base tumor location (P=.014), and males (P=.032). Significant prognosticators for 5-year overall survival were local control (P≤.001), age (P=.002), and timing of proton therapy (initial vs recurrence) (P=.002). Pencil beam scanning proton therapy is an effective and safe treatment for patients with intracranial meningiomas, resulting in high local control rates with limited toxicity. Up-front radiation likely results in improved outcomes and should be considered, especially for patients with non-benign tumors and/or for those with incomplete resections.

Diagnostic accuracy of preoperative alpha-fetoprotein as an ovarian tumor marker in children and adolescents: not as good as we thought?

To evaluate the diagnostic accuracy of preoperative serum alpha-fetoprotein (AFP) levels in predicting malignancy risk in children and adolescents presenting with ovarian neoplasms. In 110 girls aged 18 and below diagnosed with ovarian neoplasms, we retrospectively correlated preoperative serum AFP levels with histological diagnosis of germcell tumor or immature teratoma (GCT/IT) versus non-GCT/IT, and benign versus non-benign. We determined area under receiver-operating characteristic curves (AUC), sensitivity, specificity, and likelihood ratios. Twenty patients (18.2%) had non-benign ovarian neoplasms, of which 12 had GCT/IT (10.9%). In diagnosing GCT/IT versus non-GCT/IT, specificity of preoperative serum AFP was 87.8%, sensitivity 66.7%, and AUC 0.853. Excluding infants to remove the effects of increased variance in AFP in this group, specificity improved (92.0%), but not sensitivity (66.7%); AUC was 0.926. Increasing AFP cutoff to two times upper normal limit improved specificity (94.9%), but not sensitivity (66.7%). For benign versus non-benign tumors, AFP specificity was only 88.9% and sensitivity 50.0%. The diagnostic accuracy of preoperative serum AFP for detecting GCT/IT in girls was limited by poor sensitivity and positive predictive value. Excluding infants and raising cutoff levels improved specificity marginally. Clinicians should be aware of these limitations when using AFP in the preoperative evaluation of childhood ovarian neoplasms.

Incidental Neuroendocrine Tumors of the Pancreas: MDCT Findings and Features of Malignancy

The objective of our study was to evaluate the MDCT features of incidentally detected neuroendocrine tumors (NETs) of the pancreas, identify features that can predict tumor biology or aggressiveness and long-term outcome, and determine the incidence of "nonbenign" behavior. In this retrospective study, 60 histologically verified pancreatic NETs incidentally detected with contrast-enhanced MDCT were included. Various MDCT features such as size, morphology, enhancement, and presence of calcifications were evaluated and were correlated with tumor biology on histopathology. The sensitivity, specificity, predictive values, and accuracy were calculated for MDCT features in predicting nonbenign biology and risk of recurrence. A total of 32 of 60 (53%) NETs were nonbenign: most were large (mean, 29.1 mm) with a solid or complex pattern. NET size of 3 cm or larger yielded a positive predictive value of 61% for nonbenign tumors and 100% when calcification was present. In 12 patients with recurrence, 92% of NETs were nonbenign. The presence of calcification, local invasion, main pancreatic duct dilatation, vascular invasion, and lymph node enlargement along with angioinvasion and a Ki-67 index greater than 2% on histology were associated with a nonbenign diagnosis and a higher risk of recurrence. Approximately 50% of incidental NETs show uncertain or malignant behavior. Solid tumors 3 cm or larger are commonly nonbenign; however, about 30% of tumors smaller than that size cutoff can be malignant. Nonbenign tumors and those with invasive features on MDCT have a higher incidence of recurrence.

Increased Co-Expression of Macrophage Migration Inhibitory Factor and Matrix Metalloproteinase 9 Is Associated with Tumor Recurrence of Meningioma

Background and Objective: We detected the expression of MIF and matrix metalloproteinase 9 (MMP9) in meningiomas to determine whether they are valuable recurrence predictor for meningioma.Methods: 67 cases of meningiomas, including 57 benign tumors (WHO grade I) and 10 non-benign tumors (WHO grade II and III), were collected, and expression of MIF and MMP9 in tissue microarray was evaluated immunohistochemically. The correlations between immunostainings and clinicopathological parameters, as well as the follow-up data of patients, were analyzed statistically.Results: Increased expressions of both MIF (58.2%, 39/67) and MMP9 (55.2%, 37/67) were significantly associated with microvessel density (MVD) of tumor, but only dual high-expression of MIF and MMP9 was in relation to tumor invasion (P=0.016) and tumor recurrence (P=0.001). Based on univariate analysis, histological grade, tumor invasion and co-expression of MIF and MMP9 were significant predictors for recurrence. However, only histological grade and co-expression of MIF and MMP9 in tumor were independent recurrence factors with a hazard ratio of 49.033 (P=0.002) and 37.766 (P=0.002) in multivariate analysis.Conclusions: Together with histological grade, increased co-expression of MIF and MMP9 in tumor might be a valuable predictor for recurrence, especially for benign meningiomas.

Pediatric risk of malignancy index for preoperative evaluation of childhood ovarian tumors

This study aimed to develop and provisionally validate a novel scoring index for preoperative cancer-risk prediction in childhood ovarian tumors. Fifty-five girls aged 18 and below underwent surgery for ovarian masses between 2004 and 2009. Benign or non-benign histological diagnoses (the latter including all malignant and borderline tumors and tumors containing immature components) were correlated with clinical and biochemical parameters, and blinded scores of ultrasound and computed-tomography using multivariate logistic regression. Regression coefficients were used as weighting factors to create an additive index. This index was validated prospectively against 23 consecutive adnexal masses operated in 2010. In total, 67 tumors were benign and 11 non-benign. Non-benign diagnosis was independently associated with the maximum diameter of the largest solid component (score = value in cm), the presence of sex hormone-related symptoms (score = +6), and enhancement or flow in a septum or solid papillary projection (score = +4). The novel scoring index was calculated as the total score of these three parameters. A cutoff score of 7 gave a specificity of 97.9% and sensitivity of 87.5% for the training data set, and specificity and sensitivity of 100% for the pilot testing set. The novel pediatric risk-of-malignancy index is able to accurately discriminate between benign and non-benign ovarian tumors in children and adolescents. Its preoperative application may guide surgical management decisions before the availability of histological confirmation.

IMMUNOHISTOCHEMICAL DETECTION OF PROGESTERONE RECEPTORS AND THE CORRELATION WITH KI-67 LABELING INDICES IN PARAFFIN-EMBEDDED SECTIONS OF MENINGIOMAS

Female sex steroids may play a role in the proliferation of meningiomas, which usually have a high level of progesterone receptors (PgRs). We aimed to investigate the presence of PgRs and the Ki-67 labeling index (Ki-67 LI) in meningioma cases (N = 110) in relation to the severity of the disease. We studied PgR immunoreactivities and the Ki-67 LI in paraffin-embedded sections from meningiomas. Immunodetection of PgRs was conducted with peroxidase-antiperoxidase complex. Im muno detection of Ki-67 antigen was achieved by streptavidin-biotin-peroxidase complex. Of 110 meningiomas, in 57 (52%) the immunostaining for PgRs was moderately to strongly positive (Group 1), in 23 (20%) weakly positive (Group 2), and in 30 (28%) negative (Group 3). The positive immunostaining rate for the PgR in the benign meningiomas (76%) was significantly higher than that in nonbenign tumors. The positive immunostaining rate for the PgR was significantly higher in women (81%) than men (55%). The results suggested that progesterone may play a role in the growth of meningiomas. Our results confirmed that the immunodetection of the PgR and Ki-67 antigens on the paraffin sections of meningiomas provides a valuable tool for estimating the biological behavior of meningiomas.

Immunohistochemical detection of female sex hormone receptors in meningiomas: correlation with clinical and histological features.

Sixty-one meningiomas from 60 patients were screened for estrogen receptors and progesterone receptors (PgR) with monoclonal antibodies in an immunohistochemical assay. In addition, 43 of the cases were evaluated for tumor size and peritumoral edema, as seen on computed tomographic scans and magnetic resonance images. Sixty-one percent of the tumors contained significant amounts of PgR, whereas no estrogen receptor-positive tumor was observed. Thirteen percent of all tumors were classified as nonbenign variants (atypical and anaplastic meningiomas) and were more frequently found in male patients (P < 0.05). Nonbenign tumors more frequently showed an absence of PgR (P < 0.05), and there was a tendency for PgR-negative tumors to be larger than PgR-positive ones. No correlation was found between PgR status and edema. It is concluded that PgR status in meningiomas is related to tumor differentiation and may be of prognostic value with regard to biological behavior and clinical outcome.

Immunohistochemical Detection of Female Sex Hormone Receptors in Meningiomas

SIXTY-ONE MENINGIOMAS FROM 60 patients were screened for estrogen receptors and progesterone receptors (PgR) with monoclonal antibodies in an immunohistochemical assay. In addition, 43 of the cases were evaluated for tumor size and peritumoral edema, as seen on computed tomographic scans and magnetic resonance images. Sixty-one percent of the tumors contained significant amounts of PgR, whereas no estrogen receptor-positive tumor was observed. Thirteen percent of all tumors were classified as nonbenign variants (atypical and anaplastic meningiomas) and were more frequently found in male patients (P < 0.05). Nonbenign tumors more frequently showed an absence of PgR (P < 0.05), and there was a tendency for PgR-negative tumors to be larger than PgR-positive ones. No correlation was found between PgR status and edema. It is concluded that PgR status in meningiomas is related to tumor differentiation and may be of prognostic value with regard to biological behavior and clinical outcome.