Prostate cancer screening

Prediction of Gleason score in prostate cancer patients based on radiomic features of transrectal ultrasound images

OBJECTIVES

The aim of this study was to develop a model for predicting the Gleason score of patients with prostate cancer based on ultrasound images.

METHODS

Transrectal ultrasound images of 838 prostate cancer patients from The Cancer Imaging Archive database were included in this cross-section study. Data were randomly divided into the training set and testing set (ratio 7:3). A total of 103 radiomic features were extracted from the ultrasound image. Lasso regression was used to select radiomic features. Random forest and broad learning system (BLS) methods were utilized to develop the model. The area under the curve (AUC) was calculated to evaluate the model performance.

RESULTS

After the screening, 10 radiomic features were selected. The AUC and accuracy of the radiomic feature variables random forest model in the testing set were 0.727 (95% CI, 0.694-0.760) and 0.646 (95% CI, 0.620-0.673), respectively. When PSA and radiomic feature variables were included in the random forest model, the AUC and accuracy of the model were 0.770 (95% CI, 0.740-0.800) and 0.713 (95% CI, 0.688-0.738), respectively. While the BLS method was utilized to construct the model, the AUC and accuracy of the model were 0.726 (95% CI, 0.693-0.759) and 0.698 (95% CI, 0.673-0.723), respectively. In predictions for different Gleason grades, the highest AUC of 0.847 (95% CI, 0.749-0.945) was found to predict Gleason grade 5 (Gleason score ≥9).

CONCLUSIONS

A model based on transrectal ultrasound image features showed a good ability to predict Gleason scores in prostate cancer patients.

ADVANCES IN KNOWLEDGE

This study used ultrasound-based radiomics to predict the Gleason score of patients with prostate cancer.

Mutational Analysis of Prostate-Specific Antigen Defines the Intrinsic Proteolytic Activity of the proPSA Zymogen

BACKGROUND

Prostate-specific antigen (PSA) is an important prostate cancer biomarker. It is also a protease expressed at high concentrations by the normal and malignant prostate. PSA is secreted as a zymogen (proPSA) with an inhibitory prodomain that must be removed for full activity. ProPSA variants, assumed to be inactive, are found in the blood of prostate cancer patients, and are indicative of poor clinical outcome. Despite the abundance of clinical reports, our understanding of PSA's enzymology is limited, in part due to a lack of appropriate experimental systems. We sought to develop a series of PSA-derived mutants that would help to enhance our understanding of the gene.

METHODS

Sixteen rPSA variants were generated and characterized by a variety of biochemical methods.

RESULTS

The wildtype cDNA (WT) provided the template for generating a panel of recombinants. These included variants that abolished removal of the prodomain (R24A), disabled its enzymatic activity (S213A), and/or facilitated a cell-based conversion to the active conformation (FR). The purified variants' proteolytic activity was examined using a fluorogenic substrate, known PSA-cleavable proteins, and physiologically relevant inhibitors. Upon demonstrating our successful generation and purification of the PSA variants, we characterized proPSA activity, describing cleavage of synthetic and biologic substrates, but not serum protease inhibitors. This finding was exploited in the development of a self-activating mutant (PSA_QY) that exhibited the greatest enzymatic activity of all the variants.

CONCLUSIONS

The system described herein will prove useful for varied applications. ProPSA is partially functional with relatively high activity compared to the mature enzyme. In demonstrating the zymogen's intrinsic activity, we suggest that the proPSA in prostate cancer patient serum is not inert. This may have implications for our understanding of the disease. Prostate 76:1203-1217, 2016. © 2016 Wiley Periodicals, Inc.

Maturation of the developing human fetal prostate in a rodent xenograft model

BACKGROUND

Prostate cancer is the most commonly diagnosed nonskin cancer in men. The etiology of prostate cancer is unknown, although both animal and epidemiologic data suggest that early life exposures to various toxicants, may impact DNA methylation status during development, playing an important role.

METHODS

We have developed a xenograft model to characterize the growth and differentiation of human fetal prostate implants (gestational age 12-24 weeks) that can provide new data on the potential role of early life stressors on prostate cancer. The expression of key immunohistochemical markers responsible for prostate maturation was evaluated, including p63, cytokeratin 18, α-smooth muscle actin, vimentin, caldesmon, Ki-67, prostate-specific antigen, estrogen receptor-α, and androgen receptor. Xenografts were separated into epithelial and stromal compartments using laser capture microdissection (LCM), and the DNA methylation status was assessed in >480,000 CpG sites throughout the genome.

RESULTS

Xenografts demonstrated growth and maturation throughout the 200 days of post-implantation evaluation. DNA methylation profiles of laser capture microdissected tissue demonstrated tissue-specific markers clustered by their location in either the epithelium or stroma of human prostate tissue. Differential methylated promoter region CpG-associated gene analysis revealed significantly more stromal than epithelial DNA methylation in the 30- and 90-day xenografts. Functional classification analysis identified CpG-related gene clusters in methylated epithelial and stromal human xenografts.

CONCLUSION

This study of human fetal prostate tissue establishes a xenograft model that demonstrates dynamic growth and maturation, allowing for future mechanistic studies of the developmental origins of later life proliferative prostate disease.

The spectrum of circulating RNA: a window into systems toxicology

Adverse effects caused by therapeutic drugs are a serious and costly health concern. Despite the body's systemic responses to therapeutics, the liver is often the focus of damage and is usually the focus of studies of toxic effects due to its active roles in the metabolism of xenobiotics. It is extremely difficult, however, to assess systemic responses with currently available methods. Comprehensive cataloging of cell-free circulating RNAs using next-generation sequencing technology may open a window to assess drug-associated adverse effects at the systems level. To explore this potential, we conducted an RNA profiling study using the well-characterized acetaminophen overdose mouse model on liver and plasma with microarray and next-generation sequencing platforms, respectively. After drug treatment, the levels of a number of transcripts, both endogenous and exogenous RNAs, showed significant changes in plasma, reflecting not only the classical liver injury induced by acetaminophen overdose but also damage in tissues other than the liver. The changes in exogenous RNAs also reflect alteration on dieting behavior after acetaminophen overdose. Besides reporting an extensive list of circulating RNA-based biomarker candidates, this study illustrates the possibility of using circulating RNAs to assess global effects of therapeutics. This could also lead to a new approach for a more comprehensive assessment of the efficacy and safety of therapeutics.

Presence of antigen-specific somatic allelic mutations and splice variants do not predict for immunological response to genetic vaccination

Background

Antigen-specific anti-tumor vaccines have demonstrated clinical efficacy, but immunological and clinical responses appear to be patient-dependent. We hypothesized that naturally-occurring differences in amino acid sequence of a host’s target antigen might predict for immunological outcome from genetic vaccination by presentation of epitopes different from the vaccine.

Methods

Using peripheral blood cells from 33 patients who had been treated with a DNA vaccine encoding prostatic acid phosphatase (PAP), we sequenced the exons encoding PAP and PSA genes from somatic DNA to identify single nucleotide polymorphisms. In addition, mRNA was collected to detect alternative splice variants of PAP.

Results

We detected four synonymous coding mutations of PAP among 33 patients; non-synonymous coding mutations were not identified. Alternative splice variants of PAP were detected in 22/27 patients tested. The presence of detectable splice variants was not predictive of immunological outcome from vaccination. Immune responses to peptides encoded by these splice variants were common (16/27) prior to immunization, but not associated with immune responses elicited with vaccination.

Conclusions

These results suggest that antigen-specific immune responses detectable after treatment with this genetic vaccine are specific for the host-encoded antigen and not due to epitope differences between the vaccine and a particular individual’s somatic coding sequence.