COVID-19 prognostic modeling using CT radiomic features and machine learning algorithms: Analysis of a multi-institutional dataset of 14,339 patients.

BACKGROUND: We aimed to analyze the prognostic power of CT-based radiomics models using data of 14,339 COVID-19 patients. METHODS: Whole lung segmentations were performed automatically using a deep learning-based model to extract 107 intensity and texture radiomics features. We used four feature selection algorithms and seven classifiers. We evaluated the models using ten different splitting and cross-validation strategies, including non-harmonized and ComBat-harmonized datasets. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were reported. RESULTS: In the test dataset (4,301) consisting of CT and/or RT-PCR positive cases, AUC, sensitivity, and specificity of 0.83 ± 0.01 (CI95%: 0.81-0.85), 0.81, and 0.72, respectively, were obtained by ANOVA feature selector + Random Forest (RF) classifier. Similar results were achieved in RT-PCR-only positive test sets (3,644). In ComBat harmonized dataset, Relief feature selector + RF classifier resulted in the highest performance of AUC, reaching 0.83 ± 0.01 (CI95%: 0.81-0.85), with a sensitivity and specificity of 0.77 and 0.74, respectively. ComBat harmonization did not depict statistically significant improvement compared to a non-harmonized dataset. In leave-one-center-out, the combination of ANOVA feature selector and RF classifier resulted in the highest performance. CONCLUSION: Lung CT radiomics features can be used for robust prognostic modeling of COVID-19. The predictive power of the proposed CT radiomics model is more reliable when using a large multicentric heterogeneous dataset, and may be used prospectively in clinical setting to manage COVID-19 patients.

An Overview of Oncology Researches in Iran: A Scientometric Approach (1974 - February 2019).

BACKGROUND: With the growing rate of tumors, cancer has become one of the most important health concerns in Iran. The urgency with which Iranian researchers and health professionals address this challenge leads to a load of scientific materials. METHODS: To reveal gaps in produced knowledge and suggest future research directions, applying well-validated scientometric tools, we assessed the trends of Iranian published scientific articles and citations in the field of oncology. The inclusion criteria consisted of all oncology-related articles that were data-based, and peer-reviewed; with at least an abstract published in English; and authored by at least one researcher affiliated with Iranian institutions. RESULTS: Amongst 5063525 oncology research records indexed in at least one of PubMed, Scopus, or Web of Science Core Collection (WoS) from the start to February 2019, Iranian researches accounted for about 24867 (0.49%). Published articles on all cancers by Iranian researchers had a sharp continuously ascending trend, with the same pattern for citations received. Some important topics such as complementary and alternative medicine (CAM) therapies have been missing and some such as diagnostic and pharmaceutical innovations have been less investigated. The most collaborative country was the United States, while no close collaboration was observed with China that was introduced as the most productive country in the field of oncology over the past decades. CONCLUSION: Despite the progressive trend in most oncology fields, some significant practical topics are still missing. Systematic reviews of produced theoretical innovations and translating them to functional knowledge can be of importance to fulfill the mentioned gaps.

Expression assessment of a panel of long non-coding RNAs in gastric malignancy.

BACKGROUND: Long non-coding RNAs (lncRNAs) have several important functions in the regulation of cell homeostasis and cell fate. Consequently, abnormal transcription of lncRNAs has been correlated with malignant transformation of cells. These human transcripts have been shown to participate in the progression of gastric cancer. METHODS: In the current project, we evaluated expression of a panel of lncRNAs including HULC, MALAT1, FAS-AS1, GAS5, PVT1, OIP5-AS1 and THRIL in 30 gastric cancer tissues and paired adjacent non-cancerous tissues (ANCTs) using quantitative real-time PCR. RESULTS: HULC, OIP5-AS1 and THRIL transcription quantities were significantly lower in gastric tumors compared to ANCTs (P values = .02, 0.02 and 0.007, respectively). Relative transcription quantities of HULC, MALAT1, OIP5-AS1, PVT1, FAS-AS1 and THRIL were associated with the site of the primary tumor (P values = .002, 0.003, 0.002, 0.002, 0.002, and 0.001, respectively). Moreover, relative expression levels of PVT1 were associated with history of smoking (P value = .04). Correlations were identified between transcript quantities of these lncRNAs in both tumor samples and ANCTs. Receiver operating characteristic curve assessment demonstrated that THRIL had the highest diagnostic power among the mentioned lncRNAs (area under curve (AUC) = 0.72, P value = .001). HULC and OIP5-AS1 ranked afterwards (AUC values of 0.69 and 0.68; P values = .005 and 0.007, respectively). CONCLUSION: The current investigation underscores the dysregulation of these transcripts in gastric cancer specimens and suggests a number of these transcripts for further assessments of their suitability as cancer biomarkers.

Expression analysis of a panel of long non-coding RNAs (lncRNAs) revealed their potential as diagnostic biomarkers in bladder cancer.

INTRODUCTION: Long non-coding RNAs (lncRNAs) have fundamental roles in cell migration, proliferation, invasion and metastasis. METHODS: In the current study, we evaluated expression of a panel of lncRNAs in bladder cancer tissues, adjacent non-cancerous tissues (ANCTs) and normal bladder tissues to evaluate their diagnostic power. RESULTS: PV1 was down-regulated in tumor tissues compared with both ANCTs and normal controls (Expression ratios of 0.48 and 0.14; P values of 0.4 and <0.001 respectively). HOTAIR, NEAT1, TUG1 and FAS-AS1 were significantly down-regulated in tumor tissues compared with normal controls (Expression ratios of 0.4, 0.68, 0.54 and 0.11; P values of 0.04, 0.02, 0.02 and <0.001 respectively). CONCLUSION: Combination of transcript levels of seven lncRNAs improved both sensitivity and specificity values to 100%. The current study shows dysregulation of lncRNAs in bladder cancer and implies their role as diagnostic markers in this malignancy.

Validation of Reference Genes for Normalization of Relative qRT-PCR Studies in Papillary Thyroid Carcinoma.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR) in thyroid tumors require accurate data normalization, however, there are no sufficient studies addressing the suitable reference genes for gene expression analysis in malignant and normal thyroid tissue specimens. The purpose of this study was to identify valid internal control genes for normalization of relative qRT-PCR studies in human papillary thyroid carcinoma tissue samples. The expression characteristics of 12 candidate reference genes (GAPDH, ACTB, HPRT1, TBP, B2M, PPIA, 18SrRNA, HMBS, GUSB, PGK1, RPLP0, and PGM1) were assessed by qRT-PCR in 45 thyroid tissue samples (15 papillary thyroid carcinoma, 15 paired normal tissues and 15 multinodular goiters). These twelve candidate reference genes were selected by a systematic literature search. GeNorm, NormFinder, and BestKeeper statistical algorithms were applied to determine the most stable reference genes. The three algorithms were in agreement in identifying GUSB and HPRT1 as the most stably expressed genes in all thyroid tumors investigated. According to the NormFinder software, the pair of genes including 'GUSB and HPRT1' or 'GUSB and HMBS' or 'GUSB and PGM1' were the best combinations for selection of pair reference genes. The optimal number of genes required for reliable normalization of qPCR data in thyroid tissues would be three according to calculations made by GeNorm algorithm. These results suggest that GUSB and HPRT1 are promising reference genes for normalization of relative qRT-PCR studies in papillary thyroid carcinoma.

Cancer-Testis Antigens as New Candidate Diagnostic Biomarkers for Transitional Cell Carcinoma of Bladder.

To evaluate the diagnostic potential of 23 candidate genes, belonging to a category of tumor-specific antigens known as cancer-testis antigens (CTAs), in transitional cell carcinoma (TCC) patients. The expression of 16 known candidate CTAs and seven testis restricted/selective genes, predominantly expressed in the testis, was evaluated by reverse transcriptase-polymerase chain reaction (RT-PCR). Urinary exfoliated cells (UECs) and cancerous tissues of 73 TCC patients were used as cases, while 25 tumor-free adjacent bladder tissue specimens along with bladder tissue specimens and UECs of five non-TCC individuals were analyzed as controls. Among the known CTAs only MAGEA3, MAGEB4, TSGA10, PIWIL2, OIP5, and ODF4 were expressed specifically in TCC tissues and UEC samples. ACTL7A, AURKC, and CGB2 were testis-restricted/selective genes that indicated specific expression in cases in comparison to controls. MAGEA3, MAGEB4, and ODF4 mRNA was detectable in more than 50% of both TCC tissues, and UEC samples. Slight differences were detected in the mRNA expression pattern of candidate genes between the UEC samples and tumor tissues. Different panels formed by combinations of these genes can show up to 95.9% and 94.5% of positivity in TCC tissues and UEC samples, respectively, suggesting their diagnostic and surveillance potential. Meanwhile the RT-PCR assay of at least MAGEA3, MAGEB4, and ODF4 may be particularly useful for diagnostic and surveillance of TCC in the form of a multi-biomarker panel.

The Association of miR-let 7b and miR-548 with PTEN in Prostate Cancer.

PURPOSE: This study aims to investigate the expression level of mir-let7b-3p and mir-548, which are involved in PTEN expression in tissue samples of prostate cancer patients versus benign prostate hyperplasia (BPH) and normal adjacent tissue. MATERIALS AND METHODS: Prostate cancer tissues were obtained from patients after receiving informed consent. Total RNA extraction and cDNA synthesis were performed for determining gene expression. RESULTS: Ten patients were determined to have high Gleason scores (> 7), 36 and seven samples had intermediate Gleason scores (7?) and BPH, respectively, and 40 samples were derived from normal adjacent tissue. Downreg-ulation of mir-let7b and upregulation of mir-548 expression significantly correlated with high-risk Gleason scores. CONCLUSION: The present study showed that miR-let7b and/or mir-548 can be considered as potential targets in prostate cancer therapy.

Urinary exosomal expression of long non-coding RNAs as diagnostic marker in bladder cancer.

BACKGROUND: Long non-coding RNAs (lncRNAs) and exosomes have been regarded as components of cell signal transmission that modulate indigenous cellular microenvironments. Exosomes also participate in relocation of functional lncRNAs between cells. METHODS: In the present study, we evaluated expression of LINC00355, LINC00958, UCA1-201, UCA1-203, and MALAT1 lncRNAs in urinary exosomes isolated from transitional cell carcinoma (TCC) of bladder, non-malignant urinary disorders, and normal subjects. RESULTS: LINC00355, UCA1-203, and MALAT1 expression was significantly higher in TCC patients compared to controls (non-malignant or normal samples). However, UCA1-201 expression was significantly decreased in TCC patients compared with controls. LINC00355 and MALAT1 expression was significantly lower in cigarette smokers and opium-addicted TCC patients, respectively. On the other hand, LINC00355 expression tended to be higher in opium-addicted TCC patients. The proposed panel of lncRNAs (composed of UCA1-201, UCA1-203, MALAT1, and LINC00355) had 92% sensitivity and 91.7% specificity for diagnosis of bladder cancer from normal samples. CONCLUSION: Transcript levels of lncRNAs in urinary exosomes are potential diagnostic bio-markers in bladder cancer.

Expression profile of miRNAs in urine samples of bladder cancer patients.

AIM: miRNAs have been suggested as biomarkers for bladder cancer. We aimed to find a diagnostic panel of miRNAs based on differential expression of miRNAs in urine specimens of patient with bladder cancer compared with control group. METHODS: miR-141, miR-10b, miR-34b and miR-103 were selected to assess their expression in urine samples of 66 bladder cancer patients and 53 matched controls using quantitative real time PCR. RESULTS: miR-10b and miR-34b were upregulated in cases compared with controls. The combination of four miRNAs showed a sensitivity of 75% and specificity of 63.5% with a diagnostic power of 72%. CONCLUSION: Certain miRNAs can be used as biomarkers for early diagnosis of bladder cancer.

Urine exosome gene expression of cancer-testis antigens for prediction of bladder carcinoma.

BACKGROUND: Exosomes have been regarded as emerging tools for cancer diagnosis. Tumor-derived exosomes contain molecules that enhance cancer progression and affect immune responses. MATERIAL AND METHODS: In the present study, we evaluated expression of seven cancer-testis antigens (CTAs) that are regarded as putative biomarkers and immunotherapeutic targets along with NMP22 in urinary exosomes of bladder cancer patients, healthy subjects and patients affected with nonmalignant urinary disorders. RESULTS: Exosomal expression of MAGE-B4 was significantly higher in bladder cancer patients compared with normal samples (expression ratio=2.68, P=0.01). However, its expression was lower in bladder cancer patients compared with benign prostate hyperplasia (BPH) patients (expression ratio=0.17, P=0.01). Exosomal expression of NMP22 was significantly higher in bladder cancer patients compared with BPH patients (expression ratio=9.22, P=0.02). Expressions of other genes were not significantly different between bladder cancer patients and normal/nonmalignant samples. We found significant correlation between MAGE-A3 and MAGE-B4 expressions in exosomes obtained from controls. In addition, TSGA10 expression was correlated with expression of NMP22 in both cancer patients and controls. CONCLUSION: The present study provides evidences for differential expression of CTAs in urinary exosomes of bladder cancer patients and urogenital disorders and warrants further studies for assessment of their significance in cancer diagnosis and immunotherapeutic approaches.

A potential clinical significance of DAB2IP and SPRY2 transcript variants in prostate cancer.

Deregulation of key signaling pathways is one of the primary phenomena in carcinogenesis. DAB2IP and SPRY2 are regulatory elements, which act as feedback inhibitors of receptor tyrosine kinases signaling in mitogen-activated protein kinase pathway. These elements have also been implicated in the pathophysiology of cancer. Therefore, this study is aimed to investigate the expression of all known splice variants of DAB2IP and SPRY2 in prostate tissue. Fresh Prostate tissue samples (50 prostate cancer/ matched normal tissue and 30 BPH) were collected and total RNA was extracted followed by cDNA synthesis. The expression of DAB2IP and SPRY2 transcript variants were evaluated using RT-PCR and quantitative Real-time PCR. The results indicated significant down-regulation of DAB2IP transcript variant 1 in cancerous tissues compared to paired normal tissues (P = 0.001) as well as SPRY2 transcript variant 2 in cancerous tissues in comparison with the normal counterparts and BPH (P = 0.008 and P = 0.025, respectively). In addition, there was a significant negative correlation between DAB2IP.1 and SPRY2.2 expression with PSA levels in prostate cancer (P = 0.039 ρ =-0.24 and P = 0.045 ρ =-0.3, respectively). Interestingly, the down-regulation of DAB2IP.1 mRNA and SPRY2.2 mRNA was positively correlated in tumor samples (P = 0.002 ρ = 0.434). For the first time, this experiment highlights the deregulation of DAB2IP and SPRY2 transcript variants in human prostate cancer. The present study confirms and extends the previous reports through indicating transcript-specific down-regulation and significant association of DAB2IP and SPRY2 in prostate tumorigenesis.

Expression analysis of a panel of cancer-testis antigens in bladder cancer.

AIM: Cancer-testis antigens (CTAs) have specific expression in gametogenic tissues and aberrant expression in cancers. Materials & methods: We assessed expression of five testis-specific genes namely KIF2B, CST8, TMEM225, RBM46, OAZ3 in bladder cancer tissues, adjacent non-neoplastic tissues and urinary cell pellets (UCPs) of bladder cancer patients compared with nonmalignant conditions. RESULTS: Expressions of all CTAs were higher in UCPs of bladder cancer patients compared with nonmalignant conditions. RBM46 expression in UCPs was higher in patients with recurrent tumors compared with primary tumors and in patients without hematuria compared with those having hematuria. TMEM225 expression in tumoral tissues was higher in high-grade tumors compared with low-grade tumors. CONCLUSION: Expression analysis of CTAs in UCP might provide diagnostic information about bladder cancer.

ODF4, MAGEA3, and MAGEB4: Potential Biomarkers in Patients with Transitional Cell Carcinoma

Background: This study aimed to evaluate the diagnostic value of outer dense fiber 4 (ODF4), melanoma-associated antigen A3 (MAGEA3), and MAGEAB4 mRNAs in transitional cell carcinoma (TCC), using a small amount of cell reverse transcriptase-polymerase chain reaction (RT-PCR) on urinary exfoliated cells. Methods: We recruited a total of 105 suspected TCC patients and 54 sex- and age-matched non-TCC controls. The candidates' genetic expression patterns were investigated with RT-PCR, while reverse transcription quantitative PCR was applied to quantify and compare each mRNA level between cases and control groups. Results: The sensitivity of ODF4, MAGEA3, and MAGEAB4 RT-PCR was 54.8%, 63%, and 53.4%, whereas the specificity was 73.7%, 86%, and 94.7%, respectively. Combining ODF4, MAGEA3, and MAGEAB4 RT-PCR offered a relatively higher sensitivity (83.6%). Conclusion: RT-PCR with ODF4, MAGEA3, and MAGEAB4 on urinary exfoliated cells could provide clinicians with a promising method to improve TCC diagnosis, especially in the case of gross hematuria and catheterization. The method used here is non-invasive, simple and convenient, and unlike cytology, it does not rely directly on expert professional opinions. These features can be of particular importance to the management of TCC patients in whom regular and lifelong surveillance is required.

Identification of new TSGA10 transcript variants in human testis with conserved regulatory RNA elements in 5'untranslated region and distinct expression in breast cancer.

Testis specific gene antigen 10 (TSGA10) is a cancer testis antigen involved in the process of spermatogenesis. TSGA10 could also play an important role in the inhibition of angiogenesis by preventing nuclear localization of HIF-1α. Although it has been shown that TSGA10 messenger RNA (mRNA) is mainly expressed in testis and some tumors, the transcription pattern and regulatory mechanisms of this gene remain largely unknown. Here, we report that human TSGA10 comprises at least 22 exons and generates four different transcript variants. It was identified that using two distinct promoters and splicing of exons 4 and 7 produced these transcript variants, which have the same coding sequence, but the sequence of 5'untanslated region (5'UTR) is different between them. This is significant because conserved regulatory RNA elements like upstream open reading frame (uORF) and putative internal ribosome entry site (IRES) were found in this region which have different combinations in each transcript variant and it may influence translational efficiency of them in normal or unusual environmental conditions like hypoxia. To indicate the transcription pattern of TSGA10 in breast cancer, expression of identified transcript variants was analyzed in 62 breast cancer samples. We found that TSGA10 tends to express variants with shorter 5'UTR and fewer uORF elements in breast cancer tissues. Our study demonstrates for the first time the expression of different TSGA10 transcript variants in testis and breast cancer tissues and provides a first clue to a role of TSGA10 5'UTR in regulation of translation in unusual environmental conditions like hypoxia.

Transcriptome analysis of the cancer/testis genes, DAZ1, AURKC, and TEX101, in breast tumors and six breast cancer cell lines.

Breast cancer is the most frequent cancer with second mortality rate in women worldwide. Lack of validated biomarkers for early detection of breast cancer to warranty the diagnosis and effective treatments in early stages has directed to the new therapeutic approach. Cancer/testis antigens which have restricted normal expression in testis and aberrant expression in different cancers are promising targets for generating cancer vaccines, monoclonal antibodies, or dendritic cell-based immunotherapy. In this context, we investigated the expression of two known cancer testis genes, Aurora kinase C (AURKC) and testis expressed 101 (TEX101), and one new candidate, deleted in azoospermia 1 (DAZ1), in six breast cancer cell lines including two ductal carcinomas, T47D and BT-474, and four adenocarcinomas, MDA-MB-231, MDA-MB-468, MCF7, and SKBR3 as well as 50 breast cancer tumors in comparison to normal mammary epithelial cells using quantitative real-time reverse transcription PCR (RT-PCR). Results showed significant overexpression (p = 0.000) of all three genes in BT474, DAZ1 in MDA-MB-231, and AURKC and DAZ1 in SKBR3 and significant downregulation (p = 0.000) of AURKC in MCF7 cell line relative to normal breast epithelial cells. Breast tumors showed significant overexpression of AURKC in comparison to normal breast tissues (p = 0.016). The results are noticeable especially in the case of AURKC; however, there is a little knowledge about the nature, causes, consequences, and effects of cancer/testis antigens activation in different cancers. It is suggested that AURKC has effects on cell division via its serin/threonin kinases activity and organizing microtubules in relation to centrosome/spindle function during mitosis.

Molecular assessment of atpase6 mutations associated with artemisinin resistance among unexposed and exposed Plasmodium falciparum clinical isolates to artemisinin-based combination therapy.

BACKGROUND: Artemisinin-based combination therapy (ACT) is the mainstay of global efforts for treatment of Plasmodium falciparum malaria, but decline in its efficacy is the most important obstacle towards malaria control and elimination. Therefore, the present molecular analysis provides information on putative mutations associated with artemisinin resistance in P. falciparum clinical population unexposed and exposed to artesunate 4 years after adoption of ACT as the first-line anti-malarial therapy in Iran. METHODS: In this study, blood samples (n = 226) were collected from uncomplicated P. falciparum-infected patients from different health centers of Chabahar district in Sistan and Baluchistan province in the south-eastern part of Iran, during 2003 to 2010. All collected isolates were analysed for putative candidate mutations (TTA) L263E (GAA), (GAA) E431K (AAA), (GCA) A623E (GAA) and (AGT) S769N (AAT) of pfatpase6 gene using nested PCR/RFLP, followed by sequencing. Furthermore, the gene copy number was assessed by real-time quantitative PCR (RT-qPCR) in the presence of SYBR green. RESULTS: Neither the pfatpase6 L263E nor the A623E mutation was detected among all examined isolates. The E431K mutation was found in 23% of the analysed samples unexposed to ACT; however, it was detected in 17.8% (34/191) of P. falciparum isolates exposed to artesunate after 2007. High frequency of this single nucleotide polymorphisms (SNP) (overall 18.6%) among both examined groups (X2 test, P>0.05) indicated that this SNP should be considered as an unrelated mutation to artemisinin resistance. In contrast, S769N mutation was not detected in unexposed isolates; however, it was found in 2.6% (5/191), four years after introduction of ACT in this malaria setting. Also, detected SNPs were not significantly frequent in both unexposed and exposed examined isolates (X2 test, P> 0.05). Investigation in the copy number of pfatpase6 gene revealed a similar number of copy (n = 1) as in an isolate sensitive to artemisinin. CONCLUSION: Taken together, the results suggest, in particular, that pfatpase6 S769N gene needs more consideration for its possible association with artesunate resistance among P. falciparum isolates.