Machine Learning Models for Pancreatic Cancer Risk Prediction Using Electronic Health Record Data-A Systematic Review and Assessment.

INTRODUCTION: Accurate risk prediction can facilitate screening and early detection of pancreatic cancer (PC). We conducted a systematic review to critically evaluate effectiveness of machine learning (ML) and artificial intelligence (AI) techniques applied to electronic health records (EHR) for PC risk prediction. METHODS: Ovid MEDLINE(R), Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, Scopus, and Web of Science were searched for articles that utilized ML/AI techniques to predict PC, published between January 1, 2012, and February 1, 2024. Study selection and data extraction were conducted by 2 independent reviewers. Critical appraisal and data extraction were performed using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist. Risk of bias and applicability were examined using prediction model risk of bias assessment tool. RESULTS: Thirty studies including 169,149 PC cases were identified. Logistic regression was the most frequent modeling method. Twenty studies utilized a curated set of known PC risk predictors or those identified by clinical experts. ML model discrimination performance (C-index) ranged from 0.57 to 1.0. Missing data were underreported, and most studies did not implement explainable-AI techniques or report exclusion time intervals. DISCUSSION: AI/ML models for PC risk prediction using known risk factors perform reasonably well and may have near-term applications in identifying cohorts for targeted PC screening if validated in real-world data sets. The combined use of structured and unstructured EHR data using emerging AI models while incorporating explainable-AI techniques has the potential to identify novel PC risk factors, and this approach merits further study.

Artificial intelligence assisted whole organ pancreatic fat estimation on magnetic resonance imaging and correlation with pancreas attenuation on computed tomography.

BACKGROUND: Fatty pancreas is associated with inflammatory and neoplastic pancreatic diseases. Magnetic resonance imaging (MRI) is the diagnostic modality of choice for measuring pancreatic fat. Measurements typically use regions of interest limited by sampling and variability. We have previously described an artificial intelligence (AI)-aided approach for whole pancreas fat estimation on computed tomography (CT). In this study, we aimed to assess the correlation between whole pancreas MRI proton-density fat fraction (MR-PDFF) and CT attenuation. METHODS: We identified patients without pancreatic disease who underwent both MRI and CT between January 1, 2015 and June 1, 2020. 158 paired MRI and CT scans were available for pancreas segmentation using an iteratively trained convolutional neural network (CNN) with manual correction. Boxplots were generated to visualize slice-by-slice variability in 2D-axial slice MR-PDFF. Correlation between whole pancreas MR-PDFF and age, BMI, hepatic fat and pancreas CT-Hounsfield Unit (CT-HU) was assessed. RESULTS: Mean pancreatic MR-PDFF showed a strong inverse correlation (Spearman -0.755) with mean CT-HU. MR-PDFF was higher in males (25.22 vs 20.87; p = 0.0015) and in subjects with diabetes mellitus (25.95 vs 22.17; p = 0.0324), and was positively correlated with age and BMI. The pancreatic 2D-axial slice-to-slice MR-PDFF variability increased with increasing mean whole pancreas MR-PDFF (Spearman 0.51; p < 0.0001). CONCLUSION: Our study demonstrates a strong inverse correlation between whole pancreas MR-PDFF and CT-HU, indicating that both imaging modalities can be used to assess pancreatic fat. 2D-axial pancreas MR-PDFF is variable across slices, underscoring the need for AI-aided whole-organ measurements for objective and reproducible estimation of pancreatic fat.

Co-expression patterns of chimeric antigen receptor (CAR)-T cell target antigens in primary and recurrent ovarian cancer.

OBJECTIVE: Chimeric antigen receptor (CAR)-T cell strategies ideally target a surface antigen that is exclusively and uniformly expressed by tumors; however, no such antigen is known for high-grade serous ovarian carcinoma (HGSC). A potential solution involves combinatorial antigen targeting with AND or OR logic-gating. Therefore, we investigated co-expression of CA125, Mesothelin (MSLN) and Folate Receptor alpha (FOLRA) on individual tumor cells in HGSC. METHODS: RNA expression of CA125, MSLN, and FOLR1 was assessed using TCGA (HGSC) and GTEx (healthy tissues) databases. Antigen expression profiles and CD3+, CD8+ and CD20+ tumor-infiltrating lymphocyte (TIL) patterns were assessed in primary and recurrent HGSC by multiplex immunofluorescence and immunohistochemistry. RESULTS: At the transcriptional level, each antigen was overexpressed in >90% of cases; however, MSLN and FOLR1 showed substantial expression in healthy tissues. At the protein level, CA125 was expressed by the highest proportion of cases and tumor cells per case, followed by MSLN and FOLRA. The most promising pairwise combination was CA125 and/or MSLN (OR gate), with 51.9% of cases containing ≥90% of tumor cells expressing one or both antigens. In contrast, only 5.8% of cases contained ≥90% of tumor cells co-expressing CA125 and MSLN (AND gate). Antigen expression patterns showed modest correlations with TIL. Recurrent tumors retained expression of all three antigens and showed increased TIL densities. CONCLUSIONS: An OR-gated CAR-T cell strategy against CA125 and MSLN would target the majority of tumor cells in most cases. Antigen expression and T-cell infiltration patterns are favorable for this strategy in primary and recurrent disease.

Poly(adenosine diphosphate ribose) polymerase inhibitors induce autophagy-mediated drug resistance in ovarian cancer cells, xenografts, and patient-derived xenograft models.

BACKGROUND: Poly(adenosine diphosphate ribose) polymerase (PARP) inhibitors exhibit promising activity against ovarian cancers, but their efficacy can be limited by acquired drug resistance. This study explores the role of autophagy in regulating the sensitivity of ovarian cancer cells to PARP inhibitors. METHODS: Induction of autophagy was detected by punctate LC3 fluorescence staining, LC3I to LC3II conversion on Western blot analysis, and electron microscopy. Enhanced growth inhibition and apoptosis were observed when PARP inhibitors were used with hydroxychloroquine, chloroquine (CQ), or LYS05 to block the hydrolysis of proteins and lipids in autophagosomes or with small interfering RNA against ATG5 or ATG7 to prevent the formation of autophagosomes. The preclinical efficacy of the combination of CQ and olaparib was evaluated with a patient-derived xenograft (PDX) and the OVCAR8 human ovarian cancer cell line. RESULTS: Four PARP inhibitors (olaparib, niraparib, rucaparib, and talazoparib) induced autophagy in a panel of ovarian cancer cells. Inhibition of autophagy with CQ enhanced the sensitivity of ovarian cancer cells to PARP inhibitors. In vivo, olaparib and CQ produced additive growth inhibition in OVCAR8 xenografts and a PDX. Olaparib inhibited PARP activity, and this led to increased reactive oxygen species (ROS) and an accumulation of γ-H2AX. Inhibition of autophagy also increased ROS and γ-H2AX and enhanced the effect of olaparib on both entities. Treatment with olaparib increased phosphorylation of ATM and PTEN while decreasing the phosphorylation of AKT and mTOR and inducing autophagy. CONCLUSIONS: PARP inhibitor-induced autophagy provides an adaptive mechanism of resistance to PARP inhibitors in cancer cells with wild-type BRCA, and a combination of PARP inhibitors with CQ or other autophagy inhibitors could improve outcomes for patients with ovarian cancer.

Genes associated with bowel metastases in ovarian cancer.

OBJECTIVE: This study is designed to identify genes and pathways that could promote metastasis to the bowel in high-grade serous ovarian cancer (OC) and evaluate their associations with clinical outcomes. METHODS: We performed RNA sequencing of OC primary tumors (PTs) and their corresponding bowel metastases (n = 21 discovery set; n = 18 replication set). Differentially expressed genes (DEGs) were those expressed at least 2-fold higher in bowel metastases (BMets) than PTs in at least 30% of patients (P < .05) with no increased expression in paired benign bowel tissue and were validated with quantitative reverse transcription PCR. Using an independent OC cohort (n = 333), associations between DEGs in PTs and surgical and clinical outcomes were performed. Immunohistochemistry and mouse xenograft studies were performed to confirm the role of LRRC15 in promoting metastasis. RESULTS: Among 27 DEGs in the discovery set, 21 were confirmed in the replication set: SFRP2, Col11A1, LRRC15, ADAM12, ADAMTS12, MFAP5, LUM, PLPP4, FAP, POSTN, GRP, MMP11, MMP13, C1QTNF3, EPYC, DIO2, KCNA1, NETO1, NTM, MYH13, and PVALB. Higher expression of more than half of the genes in the PT was associated with an increased requirement for bowel resection at primary surgery and an inability to achieve complete cytoreduction. Increased expression of LRRC15 in BMets was confirmed by immunohistochemistry and knockdown of LRRC15 significantly inhibited tumor progression in mice. CONCLUSIONS: We identified 21 genes that are overexpressed in bowel metastases among patients with OC. Our findings will help select potential molecular targets for the prevention and treatment of malignant bowel obstruction in OC.

53BP1 as a potential predictor of response in PARP inhibitor-treated homologous recombination-deficient ovarian cancer.

OBJECTIVE: Poly(ADP-ribose) polymerase (PARP) inhibitors have shown substantial activity in homologous recombination- (HR-) deficient ovarian cancer and are undergoing testing in other HR-deficient tumors. For reasons that are incompletely understood, not all patients with HR-deficient cancers respond to these agents. Preclinical studies have demonstrated that changes in alternative DNA repair pathways affect PARP inhibitor (PARPi) sensitivity in ovarian cancer models. This has not previously been assessed in the clinical setting. METHODS: Clonogenic and plasmid-based HR repair assays were performed to compare BRCA1-mutant COV362 ovarian cancer cells with or without 53BP1 gene deletion. Archival biopsies from ovarian cancer patients in the phase I, open-label clinical trial of PARPi ABT-767 were stained for PARP1, RAD51, 53BP1 and multiple components of the nonhomologous end-joining (NHEJ) DNA repair pathway. Modified histochemistry- (H-) scores were determined for each repair protein in each sample. HRD score was determined from tumor DNA. RESULTS: 53BP1 deletion increased HR in BRCA1-mutant COV362 cells and decreased PARPi sensitivity in vitro. In 36 women with relapsed ovarian cancer, responses to the PARPi ABT-767 were observed exclusively in cancers with HR deficiency. In this subset, 7 of 18 patients (39%) had objective responses. The actual HRD score did not further correlate with change from baseline tumor volume (r = 0.050; p = 0.87). However, in the HR-deficient subset, decreased 53BP1 H-score was associated with decreased antitumor efficacy of ABT-767 (r = -0.69, p = 0.004). CONCLUSION: Differences in complementary repair pathways, particularly 53BP1, correlate with PARPi response of HR-deficient ovarian cancers.

Pancreatic cancer risk is modulated by inflammatory potential of diet and ABO genotype: a consortia-based evaluation and replication study.

Diets with high inflammatory potential are suspected to increase risk for pancreatic cancer (PC). Using pooled analyses, we examined whether this association applies to populations from different geographic regions and population subgroups with varying risks for PC, including variation in ABO blood type. Data from six case-control studies (cases, n = 2414; controls, n = 4528) in the Pancreatic Cancer Case-Control Consortium (PanC4) were analyzed, followed by replication in five nested case-control studies (cases, n = 1268; controls, n = 4215) from the Pancreatic Cancer Cohort Consortium (PanScan). Two polymorphisms in the ABO locus (rs505922 and rs8176746) were used to infer participants' blood types. Dietary questionnaire-derived nutrient/food intake was used to compute energy-adjusted dietary inflammatory index (E-DII®) scores to assess inflammatory potential of diet. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multivariable-adjusted logistic regression. Higher E-DII scores, reflecting greater inflammatory potential of diet, were associated with increased PC risk in PanC4 [ORQ5 versus Q1=2.20, 95% confidence interval (CI) = 1.85-2.61, Ptrend < 0.0001; ORcontinuous = 1.20, 95% CI = 1.17-1.24], and PanScan (ORQ5 versus Q1 = 1.23, 95% CI = 0.92-1.66, Ptrend = 0.008; ORcontinuous = 1.09, 95% CI = 1.02-1.15). As expected, genotype-derived non-O blood type was associated with increased PC risk in both the PanC4 and PanScan studies. Stratified analyses of associations between E-DII quintiles and PC by genotype-derived ABO blood type did not show interaction by blood type (Pinteraction = 0.10 in PanC4 and Pinteraction=0.13 in PanScan). The results show that consuming a pro-inflammatory diet and carrying non-O blood type are each individually, but not interactively, associated with increased PC risk.

Prevalence of germ-line mutations in cancer genes among pancreatic cancer patients with a positive family history.

PurposePanel-based genetic testing has identified increasing numbers of patients with pancreatic ductal adenocarcinoma (PDAC) who carry germ-line mutations. However, small sample sizes or number of genes evaluated limit prevalence estimates of these mutations. We estimated prevalence of mutations in PDAC patients with positive family history.MethodsWe sequenced 25 cancer susceptibility genes in lymphocyte DNA from 302 PDAC patients in the Mayo Clinic Biospecimen Resource for Pancreatic Research Registry. Kindreds containing at least two first-degree relatives with PDAC met criteria for familial pancreatic cancer (FPC), while the remaining were familial, but not FPC.ResultsThirty-six patients (12%) carried at least one deleterious mutation in one of 11 genes. Of FPC patients, 25/185 (14%) were carriers, while 11/117 (9%) non-FPC patients with family history were carriers. Deleterious mutations (n) identified in PDAC patients were BRCA2 (11), ATM (8), CDKN2A (4), CHEK2 (4), MUTYH/MYH (3 heterozygotes, not biallelic), BRCA1 (2), and 1 each in BARD1, MSH2, NBN, PALB2, and PMS2. Novel mutations were found in ATM, BARD1, and PMS2.ConclusionMultiple susceptibility gene testing in PDAC patients with family history of pancreatic cancer is warranted regardless of FPC status and will inform genetic risk counseling for families.

EUS-guided fine-needle injection of gemcitabine for locally advanced and metastatic pancreatic cancer.

BACKGROUND AND AIMS: Among the greatest hurdles to pancreatic cancer (PC) therapy is the limited tissue penetration of systemic chemotherapy because of tumor desmoplasia. The primary study aim was to determine the toxicity profile of EUS-guided fine-needle injection (EUS-FNI) with gemcitabine. Secondary endpoints included the ability to disease downstage leading to an R0 resection and overall survival (OS) at 6 months, 12 months, and 5 years after therapy. METHODS: In a prospective study from a tertiary referral center, gemcitabine (38 mg/mL) EUS-FNI was performed in patients with PC before conventional therapy. Initial and delayed adverse events (AEs) were assessed within 72 hours and 4 to 14 days after EUS-FNI, respectively. Patients were followed for ≥5 years or until death. RESULTS: Thirty-six patients with stage II (n = 3), stage III (n = 20), or stage IV (n = 13) disease underwent gemcitabine EUS-FNI with 2.5 mL (.7-7.0 mg) total volume of injectate per patient. There were no initial or delayed AEs reported. Thirty-five patients (97.2%) were deceased at the time of analysis with a median 10.3 months of follow-up (range, 3.1-63.9). OS at 6 months and 12 months was 78% and 44%, respectively. The median OS was 10.4 months (range, 2.7-68). Among patients with stage III unresectable disease, 4 (20%) were downstaged and underwent an R0 resection. CONCLUSIONS: Our study suggests the feasibility, safety, and potential efficacy of gemcitabine EUS-FNI for PC. Additional data are needed to verify these observations and to determine the potential role relative to conventional multimodality therapy.

Vaccinomics, adversomics, and the immune response network theory: individualized vaccinology in the 21st century.

Vaccines, like drugs and medical procedures, are increasingly amenable to individualization or personalization, often based on novel data resulting from high throughput "omics" technologies. As a result of these technologies, 21st century vaccinology will increasingly see the abandonment of a "one size fits all" approach to vaccine dosing and delivery, as well as the abandonment of the empiric "isolate-inactivate-inject" paradigm for vaccine development. In this review, we discuss the immune response network theory and its application to the new field of vaccinomics and adversomics, and illustrate how vaccinomics can lead to new vaccine candidates, new understandings of how vaccines stimulate immune responses, new biomarkers for vaccine response, and facilitate the understanding of what genetic and other factors might be responsible for rare side effects due to vaccines. Perhaps most exciting will be the ability, at a systems biology level, to integrate increasingly complex high throughput data into descriptive and predictive equations for immune responses to vaccines. Herein, we discuss the above with a view toward the future of vaccinology.

TP53 mutations, tetraploidy and homologous recombination repair defects in early stage high-grade serous ovarian cancer.

To determine early somatic changes in high-grade serous ovarian cancer (HGSOC), we performed whole genome sequencing on a rare collection of 16 low stage HGSOCs. The majority showed extensive structural alterations (one had an ultramutated profile), exhibited high levels of p53 immunoreactivity, and harboured a TP53 mutation, deletion or inactivation. BRCA1 and BRCA2 mutations were observed in two tumors, with nine showing evidence of a homologous recombination (HR) defect. Combined Analysis with The Cancer Genome Atlas (TCGA) indicated that low and late stage HGSOCs have similar mutation and copy number profiles. We also found evidence that deleterious TP53 mutations are the earliest events, followed by deletions or loss of heterozygosity (LOH) of chromosomes carrying TP53, BRCA1 or BRCA2. Inactivation of HR appears to be an early event, as 62.5% of tumours showed a LOH pattern suggestive of HR defects. Three tumours with the highest ploidy had little genome-wide LOH, yet one of these had a homozygous somatic frame-shift BRCA2 mutation, suggesting that some carcinomas begin as tetraploid then descend into diploidy accompanied by genome-wide LOH. Lastly, we found evidence that structural variants (SV) cluster in HGSOC, but are absent in one ultramutated tumor, providing insights into the pathogenesis of low stage HGSOC.

Pancreatic cancer: associations of inflammatory potential of diet, cigarette smoking and long-standing diabetes.

Epidemiologic studies show strong associations between pancreatic cancer (PC) and inflammatory stimuli or conditions such as cigarette smoking and diabetes, suggesting that inflammation may play a key role in PC. Studies of dietary patterns and cancer outcomes also suggest that diet might influence an individual's risk of PC by modulating inflammation. We therefore examined independent and joint associations between inflammatory potential of diet, cigarette smoking and long-standing (≥5 years) type II diabetes in relation to risk of PC. Analyses included data from 817 cases and 1756 controls. Inflammatory potential of diet was measured using the dietary inflammatory index (DII), calculated from dietary intake assessed via a 144-item food frequency questionnaire, and adjusted for energy intake. Information on smoking and diabetes were obtained via risk factor questionnaires. Associations were examined using multivariable-adjusted logistic regression. Higher DII scores, reflecting a more proinflammatory diet, were associated with increased risk of PC [odds ratio (OR)Quintile 5 versus 1 = 2.54, 95% confidence interval (CI) = 1.87-3.46, P trend < 0.0001]. Excess risk of PC also was observed among former (OR = 1.29, 95% CI = 1.07-1.54) and current (OR = 3.40, 95% CI = 2.28-5.07) smokers compared with never smokers, and among participants with long-standing diabetes (OR = 3.09, 95% CI = 2.02-4.72) compared with nondiabetics. Joint associations were observed for the combined effects of having greater than median DII score, and being a current smoker (OR = 4.79, 95% CI = 3.00-7.65) or having long-standing diabetes (OR = 6.03, 95% CI = 3.41-10.85). These findings suggest that a proinflammatory diet may act as cofactor with cigarette smoking and diabetes to increase risk of PC beyond the risk of any of these factors alone.

The composition of immune cells serves as a predictor of adaptive immunity in a cohort of 50- to 74-year-old adults.

Influenza causes significant morbidity and mortality annually. Although vaccination offers a considerable amount of protection, it is far from perfect, especially in aging populations. This is due to age-related defects in immune function, a process called immunosenescence. To date, there are no assays or methods to predict or explain variations in an individual's level of response to influenza vaccination. In this study, we measured levels of several immune cell subsets at baseline (Day 0) and at Days 3 and 28 post-vaccination using flow cytometry. Statistical modelling was performed to assess correlations between levels of cell subsets and Day 28 immune responses - haemagglutination inhibition (HAI) assay, virus neutralizing antibody (VNA) assay, and memory B cell ELISPOT. Changes in several groups of cell types from Day 0 to Day 28 and Day 3 to Day 28 were found to be significantly associated with immune response. Baseline levels of several immune cell subsets, including B cells and regulatory T cells, were able to partially explain variation in memory B-cell ELISPOT results. Increased expression of HLA-DR on plasmacytoid dendritic cells after vaccination was correlated with increased HAI and VNA responses. Our data suggest that the expression of activation markers (HLA-DR and CD86) on various immune cell subsets, as well as the relative distribution of cell subsets, both have value in predicting immune responses to influenza vaccination in older individuals.

In vivo anti-tumor activity of the PARP inhibitor niraparib in homologous recombination deficient and proficient ovarian carcinoma.

OBJECTIVE: Poly(ADP-ribose) polymerase (PARP) inhibitors have yielded encouraging responses in high-grade serous ovarian carcinomas (HGSOCs), but the optimal treatment setting remains unknown. We assessed the effect of niraparib on HGSOC patient-derived xenograft (PDX) models as well as the relationship between certain markers of homologous recombination (HR) status, including BRCA1/2 mutations and formation of RAD51 foci after DNA damage, and response of these PDXs to niraparib in vivo. METHODS: Massively parallel sequencing was performed on HGSOCs to identify mutations contributing to HR deficiency. HR pathway integrity was assessed using fluorescence microscopy-based RAD51 focus formation assays. Effects of niraparib (MK-4827) on treatment-naïve PDX tumor growth as monotherapy, in combination with carboplatin/paclitaxel, and as maintenance therapy were assessed by transabdominal ultrasound. Niraparib responses were correlated with changes in levels of poly(ADP-ribose), PARP1, and repair proteins by western blotting. RESULTS: Five PDX models were evaluated in vivo. Tumor regressions were induced by single-agent niraparib in one of two PDX models with deleterious BRCA2 mutations and in a PDX with RAD51C promoter methylation. Diminished formation of RAD51 foci failed to predict response, but Artemis loss was associated with resistance. Niraparib generally failed to enhance responses to carboplatin/paclitaxel chemotherapy, but maintenance niraparib therapy delayed progression in a BRCA2-deficient PDX. CONCLUSIONS: Mutations in HR genes are neither necessary nor sufficient to predict response to niraparib. Assessment of repair status through multiple complementary assays is needed to guide PARP inhibitor therapy, design future clinical trials and identify ovarian cancer patients most likely to benefit from PARP inhibition.

Regulatory T cells, inherited variation, and clinical outcome in epithelial ovarian cancer.

The immune system constitutes one of the host factors modifying outcomes in ovarian cancer. Regulatory T cells (Tregs) are believed to be a major factor in preventing the immune response from destroying ovarian cancers. Understanding mechanisms that regulate Tregs in the tumor microenvironment could lead to the identification of novel targets aimed at reducing their influence. In this study, we used immunofluorescence-based microscopy to enumerate Tregs, total CD4 T cells, and CD8(+) cytotoxic T cells in fresh frozen tumors from over 400 patients with ovarian cancer (>80 % high-grade serous). We sought to determine whether Tregs were associated with survival and genetic variation in 79 genes known to influence Treg induction, trafficking, or function. We used Cox regression, accounting for known prognostic factors, to estimate hazard ratios (HRs) associated with T cell counts and ratios. We found that the ratios of CD8 T cells and total CD4 T cells to Tregs were associated with improved overall survival (CD8/Treg HR 0.84, p = 0.0089; CD4/Treg HR 0.88, p = 0.046) and with genetic variation in IL-10 (p = 0.0073 and 0.01, respectively). In multivariate analyses, the associations between the ratios and overall survival remained similar (IL-10 and clinical covariate-adjusted CD8/Treg HR 0.85, p = 0.031; CD4/Treg HR 0.87, p = 0.093), suggesting that this association was not driven by variation in IL-10. Thus, integration of novel tumor phenotyping measures with extensive clinical and genetic information suggests that the ratio of T cells to Tregs may be prognostic of outcome in ovarian cancer, regardless of inherited genotype in genes related to Tregs.

Gene signature model for breast cancer risk prediction for women with sclerosing adenosis.

Benign breast disease (BBD) is diagnosed in 1-2 million women/year in the US, and while these patients are known to be at substantially increased risk for subsequent development of breast cancer, existing models for risk assessment perform poorly at the individual level. Here, we describe a DNA-microarray-based transcriptional model for breast cancer risk prediction for patients with sclerosing adenosis (SA), which represent » of all BBD patients. A training set was developed from 86 patients diagnosed with SA, of which 27 subsequently developed cancer within 10 years (cases) and 59 remained cancer-free at 10 years (controls). An diagonal linear discriminate analysis-prediction model for prediction of cancer within 10 years (SA TTC10) was generated from transcriptional profiles of FFPE biopsy-derived RNA. This model was tested on a separate validation case-control set composed of 65 SA patients. The SA TTC10 gene signature model, composed of 35 gene features, achieved a clear and significant separation between case and control with receiver operating characteristic area under the curve of 0.913 in the training set and 0.836 in the validation set. Our results provide the first demonstration that benign breast tissue contains transcriptional alterations that indicate risk of breast cancer development, demonstrating that essential precursor biomarkers of malignancy are present many years prior to cancer development. Furthermore, the SA TTC10 gene signature model, which can be assessed on FFPE biopsies, constitutes a novel prognostic biomarker for patients with SA.

Detection of DNA damage in peripheral blood mononuclear cells from pancreatic cancer patients.

DNA repair is a key mechanism in maintaining genomic stability: repair deficiencies increase DNA damage and mutations that lead to several diseases, including cancer. We extracted DNA from peripheral blood mononuclear cells (PBMCs) of 48 pancreatic adenocarcinoma cases and 48 healthy controls to determine relative levels of nuclear DNA (nDNA) and mitochondrial DNA (mtDNA) damage by QPCR. All participants were never smokers and between the ages of 60 and 69. Average levels among cases were compared to controls using a rank sum test, and logistic regression adjusted for potential confounding factors (age, sex, and diabetes mellitus). Cases had less DNA damage, with a significant decrease in mtDNA damage (P-value = 0.03) and a borderline significant decrease in nDNA damage (P = 0.08). Across samples, we found mtDNA abundance was higher among non-diabetics compared to diabetics (P = 0.04). Our results suggest that patients with pancreatic adenocarcinoma have less DNA damage in their PBMCs, and that having diabetes, a known pancreatic cancer risk factor, is associated with lower levels of mtDNA abundance.

Exposure to environmental chemicals and heavy metals, and risk of pancreatic cancer.

PURPOSE: Exposure to various chemicals and heavy metals has been associated with risk of different cancers; however, data on whether such exposures may increase the risk of pancreatic cancer (PC) are very limited and inconclusive. We examined PC risk with self-reported exposures to chemicals and heavy metals. METHODS: The design was a clinic-based, case-control study of data collected from 2000 to 2014 at Mayo Clinic in Rochester, Minnesota, USA. Cases were rapidly ascertained patients diagnosed with pancreatic ductal adenocarcinoma (n = 2,092). Controls were cancer-free patients in primary care clinics (n = 2,353), frequency-matched to cases on age, race, sex, and state/region of residence. Cases and controls completed identical risk factor questionnaires, which included yes/no questions about regular exposure to pesticides, asbestos, benzene, chlorinated hydrocarbons, chromium, and nickel. Unconditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CI) comparing those who affirmed exposure to each of the chemicals/heavy metals to those who reported no regular exposure, adjusting for potential confounders. RESULTS: Self-reported regular exposure to pesticides was associated with increased odds of PC (OR 1.21, 95% CI 1.02-1.44). Regular exposure to asbestos (OR 1.54, 95% CI 1.23-1.92), benzene (OR 1.70, 95% CI 1.23-2.35), and chlorinated hydrocarbons (OR 1.63, 95% CI 1.32-2.02) also was associated with higher odds of PC. Chromium and nickel exposures were not significantly associated with PC. CONCLUSIONS: These findings add to the limited data suggesting that exposure to pesticides, asbestos, benzene, and chlorinated hydrocarbons may increase PC risk. They further support the importance of implementing strategies that reduce exposure to these substances.

Plasma immune analytes in patients with epithelial ovarian cancer.

OBJECTIVES: Inflammation is a common feature of epithelial ovarian cancer (EOC), and measurement of plasma markers of inflammation might identify candidate markers for use in screening or presurgical evaluation of patients with adnexal masses. METHODS: Plasma specimens from cohorts of 100 patients with advanced EOC (AJCC Stage III and IV), 50 patients with early stage EOC (Stage I and II), and 50 patients with benign surgical conditions were assayed for concentrations of multiple cytokines, toll-like receptor agonists, and vascular growth factors via ELISA and electrochemiluminescence. Immune proteins were then analyzed for association with EOC. Differences in plasma protein levels between benign, early, and advanced EOC patient groups were assessed with and without adjustment for plasma cancer antigen 125 (CA-125) levels. RESULTS: Out of 23 proteins tested, six-including interferon gamma (IFNγ), interleukin 6 (IL-6), IL-8, IL-10, tumor necrosis factor alpha (TNFα), and placental growth factor (PlGF)-were univariately associated with EOC (all p<0.005), and one-IL-6-was associated with early stage EOC (p<0.0001). Heat shock protein 90kDa beta member 1 (HSP90B1, gp96) was associated with EOC and early stage EOC with borderline statistical significance (p=0.039 and p=0.026, respectively). However, when adjusted for (CA-125), only HSP90B1 independently predicted EOC (p=0.008), as well as early stage EOC (p=0.014). CONCLUSIONS: Multiple plasma cytokines, including IFNγ, IL-6, IL-8, IL-10, TNFα, PlGF, and HSP90B1 are associated with EOC. Of these, HSP90B1 is associated with EOC independent from the biomarker CA-125.

Assessment of published models and prognostic variables in epithelial ovarian cancer at Mayo Clinic.

OBJECTIVES: Epithelial ovarian cancer (EOC) is an aggressive disease in which first line therapy consists of a surgical staging/debulking procedure and platinum based chemotherapy. There is significant interest in clinically applicable, easy to use prognostic tools to estimate risk of recurrence and overall survival. In this study we used a large prospectively collected cohort of women with EOC to validate currently published models and assess prognostic variables. METHODS: Women with invasive ovarian, peritoneal, or fallopian tube cancer diagnosed between 2000 and 2011 and prospectively enrolled into the Mayo Clinic Ovarian Cancer registry were identified. Demographics and known prognostic markers as well as epidemiologic exposure variables were abstracted from the medical record and collected via questionnaire. Six previously published models of overall and recurrence-free survival were assessed for external validity. In addition, predictors of outcome were assessed in our dataset. RESULTS: Previously published models validated with a range of c-statistics (0.587-0.827), though application of models containing variables which are not part of routine practice were somewhat limited by missing data; utilization of all applicable models and comparison of results are suggested. Examination of prognostic variables identified only the presence of ascites and ASA score to be independent predictors of prognosis in our dataset, albeit with marginal gain in prognostic information, after accounting for stage and debulking. CONCLUSIONS: Existing prognostic models for newly diagnosed EOC showed acceptable calibration in our cohort for clinical application. However, modeling of prospective variables in our dataset reiterates that stage and debulking remains the most important predictors of prognosis in this setting.