As the most lethal major cancer, pancreatic cancer is a global healthcare challenge. Personalized medicine utilizing cutting-edge multi-omics data holds potential for major breakthroughs in tackling this critical problem. Radiomics and deep learning, two trendy quantitative imaging methods that take advantage of data science and modern medical imaging, have shown increasing promise in advancing the precision management of pancreatic cancer via diagnosing of precursor diseases, early detection, accurate diagnosis, and treatment personalization and optimization. Radiomics employs manually-crafted features, while deep learning applies computer-generated automatic features. These two methods aim to mine hidden information in medical images that is missed by conventional radiology and gain insights by systematically comparing the quantitative image information across different patients in order to characterize unique imaging phenotypes. Both methods have been studied and applied in various pancreatic cancer clinical applications. In this review, we begin with an introduction to the clinical problems and the technology. After providing technical overviews of the two methods, this review focuses on the current progress of clinical applications in precancerous lesion diagnosis, pancreatic cancer detection and diagnosis, prognosis prediction, treatment stratification, and radiogenomics. The limitations of current studies and methods are discussed, along with future directions. With better standardization and optimization of the workflow from image acquisition to analysis and with larger and especially prospective high-quality datasets, radiomics and deep learning methods could show real hope in the battle against pancreatic cancer through big data-based high-precision personalization.

Background: Pancreatic cancer (PC) is a lethal malignancy primarily driven by activated Kras mutations and characterized by the deregulation of several genes including mucins. Previous studies on mucins have identified their significant role in both benign and malignant human diseases including PC progression and metastasis. However, the initiation of MUC expression during PC remains unknown because of lack of early stage tumor tissues from PC patients. Methods: In the present study, we have evaluated stage specific expression patterns of mucins during mouse PC progression in (Kras^G12D;Pdx1-Cre (KC)) murine PC model from pancreatic intraepithelial neoplasia (PanIN) to pancreatic ductal adenocarcinoma (PDAC) by immunohistochemistry and quantitative real-time PCR. Results: In agreement with previous studies on human PC, we observed a progressive increase in the expression of mucins particularly Muc1, Muc4 and Muc5AC in the pancreas of KC (as early as PanIN I) mice with advancement of PanIN lesions and PDAC both at mRNA and protein levels. Additionally, mucin expression correlated with the increased expression of inflammatory cytokines IFN-γ (p < 0.0062), CXCL1 (p < 0.00014) and CXCL2 (p < 0.08) in the pancreas of KC mice, which are known to induce mucin expression. Further, we also observed progressive increase in inflammation in pancreas of KC mice from 10 to 50 weeks of age as indicated by the increase in the macrophage infiltration. Overall, this study corroborates with previous human studies that indicated the aberrant overexpression of MUC1, MUC4 and MUC5AC mucins during the progression of PC. Conclusions: Our study reinforces the potential utility of the KC murine model for determining the functional role of mucins in PC pathogenesis by crossing KC mice with corresponding mucin knockout mice and evaluating mucin based diagnostic and therapeutic approaches for lethal PC.

(1) Background: Radiomics use high-throughput mining of medical imaging data to extract unique information and predict tumor behavior. Currently available clinical prediction models poorly predict treatment outcomes in pancreatic adenocarcinoma. Therefore, we used radiomic features of primary pancreatic tumors to develop outcome prediction models and compared them to traditional clinical models. (2) Methods: We extracted and analyzed radiomic data from pre-radiation contrast-enhanced CTs of 74 pancreatic cancer patients undergoing stereotactic body radiotherapy. A panel of over 800 radiomic features was screened to create overall survival and local-regional recurrence prediction models, which were compared to clinical prediction models and models combining radiomic and clinical information. (3) Results: A 6-feature radiomic signature was identified that achieved better overall survival prediction performance than the clinical model (mean concordance index: 0.66 vs. 0.54 on resampled cross-validation test sets), and the combined model improved the performance slightly further to 0.68. Similarly, a 7-feature radiomic signature better predicted recurrence than the clinical model (mean AUC of 0.78 vs. 0.66). (4) Conclusion: Overall survival and recurrence can be better predicted with models based on radiomic features than with those based on clinical features for pancreatic cancer.

A majority of breast cancer tumors express estrogen receptor (ER) and/or progesterone receptor (PR); however, the percentage of cancer cells expressing these receptors can range from 0-100%. The prognostic and therapeutic impact of the percentage of cells expressing hormone receptors in breast cancer is not fully understood. A retrospective analysis of 411 breast cancer patients who were treated at the University of Nebraska Medical Center between 2010 and 2017 was performed. Patient tumors were evaluated for percentage of cells expressing ER and PR in conjunction with clinical outcomes. Patient tumors demonstrated a highly bimodal pattern of ER and PR staining with a majority of tumors demonstrating either a high percentage (> 80% of cells) or lack of cells (0%) staining for ER or PR. An increase in the percentage of ER positivity correlated with decreased local recurrence and improved overall survival. An increase in the percentage of PR positivity demonstrated a trend towards decreased local recurrence and improved overall survival, but was not statistically significant. Results based on both continuous and categorical evaluation of ER expression revealed that increasing expression correlated with improved patient outcomes. Similar evaluation of PR expression demonstrated a trend towards improved patient outcomes though not statistically significant. These findings suggest that the degree of hormone receptor positivity and not a Boolean representation of positivity could provide additional prognostic value in the treatment and management of breast cancer.

Background: Conventionally fractionated (CF) radiation therapy (RT) has been associated with lymphopenia, leading to compromised overall survival (OS) in cancer patients. It currently remains unknown if stereotactic body (SB) RT induces lymphopenia to the same degree. The aim of this study is to determine if SBRT with either chemotherapy (CMT) (Fluorouracil (5FU) or capecitabine) or Nelfinavir (NFV) to pancreatic adenocarcinoma induces lymphopenia to the same degree as CFRT with 5FU or capecitabine and how any associated difference affects patient survival outcomes. Methods: Medical records of pancreatic adenocarcinoma patients treated with induction CMT followed by RT with concurrent CMT or NFV were reviewed. Patients with total lymphocyte counts (TLCs) available both prior to and following initiation of RT were included. Three groups were identified: CFRT/CMT, SBRT/CMT, and SBRT/NFV. Median delivered RT doses for CFRT and SBRT were 50.4 Gy in 1.8 Gy fractions and 35 Gy in 7 Gy fractions, respectively. TLCs from day 0 (the first day of RT) to 40 were recorded and analyzed using the Kruskal-Wallis test with p-values adjusted with Bonferroni’s method. Linear regressions were utilized to estimate the slope of TLCs as it changes with time and survival analysis was performed via Kaplan-Meier plots. Results: One hundred patients were identified (28 CFRT/CMT, 27 SBRT/CMT, 45 SBRT/NFV). Median pre-RT TLCs were not different among groups. Median lowest TLCs were significantly lower (p < 0.0001) and median TLCs reduction over time were significantly greater (p < 0.0001) in the CFRT group than SBRT groups. There was no difference in lowest TLCs or TLCs reduction over time between SBRT groups. Across all groups, the median time to lowest TLCs was similar. Survival analysis revealed no significant difference in median OS between SBRT and CFRT groups. However, in patients with surgery, Median OS for patients with SBRT/CMT was significantly higher than in those with SBRT/NFV (p = 0.03). Conclusions: Compared to CFRT, SBRT is associated with less lymphopenia. Further study of the effect of radiation technique on immune status is warranted.

The use of multiparametric magnetic resonance imaging (mpMRI)-derived radiomics has the potential to offer noninvasive, imaging-based biomarkers for the identification of subvisual characteristics indicative of a poor oncologic outcome. The present study, therefore, seeks to develop, validate, and assess the performance of an MRI-derived radiomic model for the prediction of prostate cancer (PC) recurrence following radical prostatectomy (RP) with curative intent. mpMRI imaging was obtained from 251 patients who had undergone an RP for the treatment of localized prostate cancer across two institutions and three surgeons. All patients had a minimum of 2 years follow-up via prostate-specific antigen serum testing. Each prostate mpMRI was individually reviewed, and the prostate was delineated as a single slice (ROI) on axial T2 high-resolution image sets. A total of 924 radiomic features were extracted and tested for stability via intraclass correlation coefficient (ICC) following image normalization via histogram matching. Fourteen important and nonredundant features were found to be predictors of PC recurrence at a mean ± SD of 3.2 ± 2.2 years post-RP. Five-fold, ten-run cross-validation of the model containing these fourteen features yielded an area under the curve (AUC) of 0.89 ± 0.04 in the training set (n = 225). In comparison, the University of California San Fransisco Cancer of the Prostate Risk Assessment score (UCSF-CAPRA) and Memorial Sloan Kettering Cancer Center (MSKCC) Pre-Radical prostatectomy nomograms yielded AUC of 0.66 ± 0.05 and 0.67 ± 0.05, respectively (p < 0.01). When the radiomic model was applied to the test set (n = 26), AUC was 0.78; sensitivity, specificity, positive predictive value, and negative predictive value were 60%, 86%, 52%, and 89%, respectively. Accuracy in predicting PC recurrence was 81%.