Purpose: To evaluate radiomics features in order to: differentiate malignant versus benign lesions; predict low versus moderate and high grading; identify positive or negative hormone receptors; and discriminate positive versus negative human epidermal growth factor receptor 2 related to breast cancer. Methods: A total of 182 patients with known breast lesions and that underwent Contrast-Enhanced Mammography were enrolled in this retrospective study. The reference standard was pathology (118 malignant lesions and 64 benign lesions). A total of 837 textural metrics were extracted by manually segmenting the region of interest from both craniocaudally (CC) and mediolateral oblique (MLO) views. Non-parametric Wilcoxon–Mann–Whitney test, receiver operating characteristic, logistic regression and tree-based machine learning algorithms were used. The Adaptive Synthetic Sampling balancing approach was used and a feature selection process was implemented. Results: In univariate analysis, the classification of malignant versus benign lesions achieved the best performance when considering the original_gldm_DependenceNonUniformity feature extracted on CC view (accuracy of 88.98%). An accuracy of 83.65% was reached in the classification of grading, whereas a slightly lower value of accuracy (81.65%) was found in the classification of the presence of the hormone receptor; the features extracted were the original_glrlm_RunEntropy and the original_gldm_DependenceNonUniformity, respectively. The results of multivariate analysis achieved the best performances when using two or more features as predictors for classifying malignant versus benign lesions from CC view images (max test accuracy of 95.83% with a non-regularized logistic regression). Considering the features extracted from MLO view images, the best test accuracy (91.67%) was obtained when predicting the grading using a classification-tree algorithm. Combinations of only two features, extracted from both CC and MLO views, always showed test accuracy values greater than or equal to 90.00%, with the only exception being the prediction of the human epidermal growth factor receptor 2, where the best performance (test accuracy of 89.29%) was obtained with the random forest algorithm. Conclusions: The results confirm that the identification of malignant breast lesions and the differentiation of histological outcomes and some molecular subtypes of tumors (mainly positive hormone receptor tumors) can be obtained with satisfactory accuracy through both univariate and multivariate analysis of textural features extracted from Contrast-Enhanced Mammography images.

Over the years, increasing information has been asked of the pathologist: we have moved from a purely morphological diagnosis to biomolecular and genetic studies, which have made it possible to implement the use of molecular targeted therapies, such as anti-epidermal growth factor receptor (EGFR) molecules in EGFR-mutated lung cancer, for example. Today, next generation sequencing (NGS) has changed the approach to neoplasms, to the extent that, in a short time, it has gained a place of absolute importance and diagnostic, prognostic and therapeutic utility. In this scenario, formaldehyde-fixed and paraffin-embedded (FFPE) biological tissue samples are a source of clinical and molecular information. However, problems can arise in the genetic material (DNA and RNA) for use in NGS due to fixation, and work is being devoted to possible strategies to reduce its effects. In this paper, we discuss the applications of FFPE tissue samples in the execution of NGS, we focus on the problems arising with the use of this type of material for nucleic acid extraction and, finally, we consider the most useful strategies to prevent and reduce single nucleotide polymorphisms (SNV) and other fixation artifacts.

In a growing number of social and clinical scenarios, machine learning (ML) is emerging as a promising tool for implementing complex multi-parametric decision-making algorithms. Regarding ovarian cancer (OC), despite the standardization of features that can support the discrimination of ovarian masses into benign and malignant, there is a lack of accurate predictive modeling based on ultrasound (US) examination for progression-free survival (PFS). This retrospective observational study analyzed patients with epithelial ovarian cancer (EOC) who were followed in a tertiary center from 2018 to 2019. Demographic features, clinical characteristics, information about the surgery and post-surgery histopathology were collected. Additionally, we recorded data about US examinations according to the International Ovarian Tumor Analysis (IOTA) classification. Our study aimed to realize a tool to predict 12 month PFS in patients with OC based on a ML algorithm applied to gynecological ultrasound assessment. Proper feature selection was used to determine an attribute core set. Three different machine learning algorithms, namely Logistic Regression (LR), Random Forest (RFF), and K-nearest neighbors (KNN), were then trained and validated with five-fold cross-validation to predict 12 month PFS. Our analysis included n. 64 patients and 12 month PFS was achieved by 46/64 patients (71.9%). The attribute core set used to train machine learning algorithms included age, menopause, CA-125 value, histotype, FIGO stage and US characteristics, such as major lesion diameter, side, echogenicity, color score, major solid component diameter, presence of carcinosis. RFF showed the best performance (accuracy 93.7%, precision 90%, recall 90%, area under receiver operating characteristic curve (AUROC) 0.92). We developed an accurate ML model to predict 12 month PFS.

In recent years, the preferentially expressed antigen in melanoma (PRAME) has also been used in the histopathological diagnosis of melanocytic lesions, in order to understand if it could constitute a valid, inexpensive, and useful resource in dermatopathological fields. We performed a double-center study to evaluate whether the data on the usefulness and possible limitations of PRAME could also be confirmed by our group. From 1 December 2021 to 29 March 2022, we collected 275 cases of melanocytic lesions that were immunostained with PRAME (Ab219650) and rabbit monoclonal antibody (Abcam). To better correlate the PRAME expression with its nature (benign, uncertain potential for malignancy, or malignant), we categorized PRAME tumor cells’ percentage positivity and intensity of immunostaining in a cumulative score obtained by adding the quartile of positive tumor cells (0, 1+, 2+, 3+, 4+) to the PRAME expression intensity in tumor cells (0, 1+, 2+, 3+). Of these 275 lesions, 136 were benign, 12 were of uncertain potential for malignancy (MELTUMP or SAMPUS or SPARK nevus), and 127 were malignant. The immunoexpression of PRAME was completely negative in 125/136 benign lesions (91.9%), with only a few positive melanocytes (1+) and intensity 1+ in the remaining 11 cases (8.1%). Of the 127 cases of melanoma (superficial spreading, lentigo maligna, and pagetoid histotypes), PRAME was strongly positive in 104/127 cases (81.8%) with intensity 4+ and 3+. In 17 cases (13.3%; melanoma spindle and nevoid cell histotypes), PRAME was positive in percentage 2+ and with intensity ranging from 2+ to 3+. In 7 cases (5.5%) of desmoplastic melanoma, PRAME was 1+ positive and/or completely negative. Of the 12 cases of lesions with uncertain potential for malignancy, the immunoexpression of PRAME was much more heterogeneous and irregularly distributed throughout the lesion. These data are perfectly in agreement with the current literature, and they demonstrate that the reliability of PRAME is quite high, but its use cannot cause physicians to disregard the morphological information and the execution of other ancillary immunohistochemical stains such as Melan-A, HMB-45, MiTF, and SOX-10.

Cervical cancer (CC) is the fourth most frequent cancer in women worldwide. HPV infection is associated with the majority of CC cases, but a small proportion of CCs actually test negative for HPV. The prevalence of HPV among CC histotypes is very different. It has been suggested that HPV-negative CC may represent a biologically distinct subset of tumors, relying on a distinct pathogenetic pathway and carrying a poorer prognosis, than HPV-positive CCs. Although, the discordance in terms of sensitivity and specificity between different HPV tests as well as the potential errors in sampling and storing tissues may be considered as causes of false-negative results. The identification of HPV-negative CCs is essential for their correct management. The aim of this narrative review is to summarize the clinical and pathological features of this variant. We also discuss the pitfalls of different HPV tests possibly leading to classification errors.

In recent years, an increasing enthusiasm has been observed towards artificial intelligence and machine learning, involving different areas of medicine. Among these, although still in the embryonic stage, the dermatopathological field has also been partially involved, with the attempt to develop and train algorithms that could assist the pathologist in the differential diagnosis of complex melanocytic lesions. In this article, we face this new challenge of the modern era, carry out a review of the literature regarding the state of the art and try to determine promising future perspectives.