Pancreatic Adenocarcinoma Research Articles

MUC17 is a potential new prognostic biomarker and promotes pancreatic cancer progression in obstructive jaundice.

Our working group has previously shown that bile acids (BAs) accelerate carcinogenic processes in pancreatic cancer (PC) in which mucin 4 (MUC4) expression has a central role. However, the role of other mucins in PC are less clear, especially in bile-induced cancer progression. The study aim was to investigate expression of MUC17 in BAs- or human serum-treated pancreatic ductal adenocarcinoma (PDAC) cell lines and use different assays with RNA silencing/overexpression to study the role of MUC17 in cancer progression. Protein expression of MUC17 was evaluated in 55 human pancreatic samples by immunohistochemistry, and Kaplan-Meier survival analysis was used to compare survival curves. Expression of MUC17 increased in PDAC patients, especially in obstructive jaundice (OJ) and the elevated MUC17 expression associated with poorer overall survival (10.66±1.99 vs. 15.05±2.03 months; Log rank: 0.0497). Treatment of Capan-1 and AsPC-1 cells with BAs or with human serum obtained from PDAC + OJ patients enhanced the expression of MUC17, as well as the proliferative potential of the cells, whereas knockdown of MUC17 alone or in combination with MUC4 decreased BAs-induced carcinogenic processes. Our results demonstrated that MUC17 has a central role in bile-induced PC progression, and in addition to MUC4, this isoform also can be used as a novel prognostic biomarker.

Oncological Outcomes of Open Versus Minimally Invasive Surgery for Ductal Adenocarcinomas of Pancreatic Head: A Propensity Score Matching Analysis.

Minimally invasive pancreatic resections (MIPRs) have been shown to be safe and feasible, but there is still a lack of high-level evidence on oncological outcomes for cephalic pancreatic ductal adenocarcinoma (PDAC). The aim of this study was to compare the oncological outcomes of patients undergoing MIPR and open pancreatic resection (OPR) for pancreatic head cancer in a single high-volume center. Data from a prospectively collected database of patients who underwent radical-intent surgery for resectable and borderline resectable PDAC of the head at our institution between January 2013 and May 2023 were retrieved and analyzed, comparing the surgical and oncological outcomes of MIPR and OPR, using a propensity score matching analysis. In the study period, 220 patients were selected. After matching, a total of 81 MIPRs and 81 OPRs were compared. No difference was found regarding R0 rate (OPR 83.9% vs. MIPR 74.1%, p = 0.122). Median overall survival (24 and 31 months for the OPR and MIPR groups, respectively; log rank p = 0.665) and disease-free survival (12 and 21 months for the OPR and MIPR groups, respectively; log rank p = 0.118) did not differ between the groups. The MIPR group was associated with a greater number of harvested lymph nodes (22 vs. 16, p = 0.0008), longer operative time (565 vs. 420 min, p < 0.0001), and shorter length of stay (12 vs. 18 days; p = 0.0001). No differences between the groups were found regarding all other postoperative and pathological outcomes. Regarding oncological outcomes, MIPR appeared to be comparable to OPR for treating patients with PDAC of the head. Despite an increased operative time, MIPR was associated with a greater number of LNs harvested and a shorter length of stay.

Global obesity epidemic and rising incidence of early-onset cancers.

Incidence of early-onset cancers at multiple organ sites has increased worldwide in recent decades. We investigated whether such increasing trends could be explained by trends in obesity. We obtained incidence data for 21 common cancers among 25-49-year-olds during 2000-2012 in 42 countries from the Cancer Incidence in Five Continents database. Nine cancers we examined have been classified as obesity-related by the International Agency for Research on Cancer. Estimates of overweight and obesity prevalence came from the Non-communicable Disease Risk Factor Collaboration. Using country-level data, we examined whether changes in the prevalence of overweight and obesity combined were correlated with changes in cancer incidence, after accounting for various time lags (0-15 years) between exposure and cancer diagnosis. To test the validity of our approach, we conducted negative control analyses (using non-obesity-related cancers as the outcome variable, and per-capita gross national income as the exposure variable), and sensitivity and supplemental analyses using alternative data streams or processing. We found increased incidence for six of nine obesity-related and seven of twelve non-obesity-related cancers in 25-49-year-olds. These increases were more predominant in Western countries (particularly Australia, the USA, Canada, Norway, the Netherlands, and Lithuania). For four obesity-related cancers displaying increased incidence (colon, rectum, pancreas, kidney), changes in cancer incidence were positively correlated with changes in overweight and obesity prevalence. When accounting for a 15-year lag, the estimated correlation was 0.27 (95% confidence interval (CI) = -0.04, 0.53; P = 0.090) for colon cancer, 0.33 (95% CI = 0.02, 0.58; P = 0.036) for rectal cancer, 0.39 (95% CI = 0.08, 0.64; P = 0.018) for pancreatic cancer, and 0.22 (95% CI = -0.10, 0.50; P = 0.173) for kidney cancer. Similar correlations were found in the sensitivity and supplemental analyses. We did not find similar correlations with excess body weight for the non-obesity-related early-onset cancers, nor correlations with per-capita gross national income for any cancer types, in the negative control analyses. Worldwide increases in early-onset colon, rectal, pancreatic, and kidney cancers may have been partly driven by increases in excess body weight. The increases in other early-onset cancers, however, were likely driven by other factors deserving of further investigation.

Humanized recombinant immunotoxin targeting hCG demonstrates therapeutic potential for advanced stage difficult to treat cancers

We report the development of an immunotherapeutic molecule, a Humanized immunotoxin, for treating hCG-expressing advanced-stage cancers. PiPP, a high-affinity anti-hCG monoclonal antibody, is used in the immunotoxin for ‘homing’ hCG-positive cancer cells. The deimmunized (DI) form of α-Sarcin, a fungal-origin toxin that lacks functional T-cell epitopes, is used in the design to ensure minimal immunogenicity of the immunotoxin for repetitive use in humans. A single-chain variable fragment (scFv) of PiPP was constructed by linking the Humanized VH and VL regions of the antibody. The scFv part of the antibody was further linked to the toxin α-Sarcin (DI) at the gene level and expressed as a recombinant protein in E. coli. The immunotoxin was purified from the bacterial cell lysate and analysed for binding and cytotoxicity to hCG-secreting colorectal and pancreatic cancer cells. The results showed that the scFv(PiPP)-Sarcin immunotoxin was able to bind to colorectal and pancreatic cancer cells and killed approximately 85% of the cells. In vivo testing of the immunotoxin produced results similar to those of in vitro testing against colorectal adenocarcinoma-induced tumours. This immunotoxin could be a promising immunotherapeutic agent for treating colorectal, pancreatic and other terminal-stage hCG-expressing cancers.

Acute pancreatitis as an early sign of pancreatic cancer; a retrospective, matched cohort study

Background and aims Pancreatic ductal adenocarcinoma (PDAC) often presents as acute pancreatitis (AP). However, data on the clinical outcomes of PDAC initially presenting as AP are limited. We aimed to assess the clinical features of PDAC that manifest as AP. Methods We reviewed the PDAC database at the Asan Medical Center between 2010–2016. Our study included 77 patients with PDAC who presented with AP (PDAC-AP group) and 154 age-gender-matched PDAC patients as controls (PDAC-other group). Patients’ demographics, disease characteristics, and outcomes were compared between both groups. Results Acute pancreatitis was an initial symptom in 1.12% of the patients with PDAC (77 of 6,821). Approximately 81.8% of the patients had clinically mild pancreatitis, and 91% were diagnosed with PDAC within two months of presentation with AP. Main tumor size was significantly smaller in the PDAC-AP group than in the PDAC-other group (PDAC-AP: 2.59 ± 1.21 cm vs. PDAC-other: 3.73 ± 1.78 cm, p < 0.01). The PDAC-AP group patients were diagnosed earlier than those in the PDAC-other group (PDAC-AP: stage 1–2, 80.6% vs. PDAC-other: 46.7%, p < 0.01). The proportion of resectable PDAC was significantly higher in the PDAC-AP group (PDAC-AP: 64.9% vs. PDAC-other: 50%, p < 0.01). Overall survival was significantly longer in the PDAC-AP group than in the PDAC-other group (30.2 months vs. 19.9 months, p = 0.03). Conclusions In patients who presented with clinical AP, PDAC was identified at an earlier stage, and these patients showed better survival rates. These results suggest that AP may be an early sign of PDAC.

GL-V9 Promotes Autophagy-Mediated YAP1 Degradation and Activates Mitochondrial Apoptosis in PDAC Cells.

Background: Pancreatic ductal adenocarcinoma (PDAC) is among the most aggressive forms of pancreatic cancer with a poor prognosis. YAP1 expression is markedly elevated in PDAC, but how it works is not clear. GL-V9, a derivative of the natural compound wogonin, effectively fights a variety of tumors; however, its effect on PDAC has not yet been studied. Methods: TCGA database analysis, Western blots, immunofluorescence, and real-time PCR were used to evaluate GL-V9's effect on YAP1 expression and mRNA levels. Immunofluorescence was used to examine the co-location of YAP1 with LAMP2 and p62. Co-immunoprecipitation was used to assess the binding of YAP1 to ubiquitin, p62, and TEAD1. A PDAC graft tumor model was used to test GL-V9's pharmacological effects. Western blots and immunohistochemistry were used to measure apoptosis- and autophagy-related protein expression. Results: GL-V9 effectively promoted the degradation of YAP1, reduced YAP1 nuclear localization, and induced mitochondrial apoptosis in PDAC cells. YAP1 overexpression led to the upregulation of Bcl-2 and attenuated the caspase cascade induced by GL-V9. Furthermore, we demonstrated that GL-V9 induced autophagosome-lysosome fusion via the AKT/mTOR/TFEB pathway, leading to mitochondrial apoptosis in PDAC cells. In vivo studies also confirmed that GL-V9 exerts anti-tumor effects by suppressing YAP1 expression, while also activating autophagy and inducing mitochondrial apoptosis in BXPC-3-bearing BALB/c nude mice. Conclusions: Our findings underscore the importance of autophagy-mediated YAP1 degradation in PDAC, providing a novel molecular rationale (GL-V9) as a promising treatment for this disease.

Efficacy and safety of the anti-IL1RAP antibody nadunolimab (CAN04) in combination with gemcitabine and nab-paclitaxel in patients with advanced/metastatic pancreatic cancer.

Interleukin (IL)-1 pathway upregulation is implicated in pancreatic ductal adenocarcinoma (PDAC) progression, therapy resistance, and survival. Nadunolimab is an IL-1 receptor accessory protein (IL1RAP)-targeting antibody with enhanced antibody-dependent cellular cytotoxicity that blocks IL-1α/IL-1β signaling. We investigated efficacy and safety of nadunolimab in PDAC, in combination with gemcitabine/nab-paclitaxel (GN). Patients with previously untreated locally advanced/metastatic PDAC received nadunolimab (1.0-7.5 mg/kg) every two weeks with standard GN. The primary objective was safety; secondary objectives were anti-tumor response, progression-free survival (PFS) and overall survival (OS). Correlations between serum and tumor biomarkers and clinical response were explored. 76 patients were enrolled, median age 63 years (range 43-89), 42% female, 97% with metastatic disease, 9% having received adjuvant chemotherapy. The most frequent Grade ≥3 adverse event was neutropenia (66%), typically during Cycle 1. Infusion-related reactions occurred in 29% (Grade 3, 3%). Only 1 of 76 patients had grade 3 or above peripheral neuropathy. No marked dose-dependent differences in safety or efficacy were observed between the four dose groups. Median OS overall was 13.2 months (95%CI 10.6-15.5) and 1-year survival was 58%. Median iPFS (iRECIST) was 7.2 months (95%CI 5.2-8.5). Treatment efficacy was higher in patients with high versus low tumor baseline IL1RAP expression (OS 14.2 vs 10.6 months, p=0.026). A reduction in serum IL-8 on treatment correlated with prolonged OS. Nadunolimab combined with GN shows promising efficacy and manageable safety in locally advanced/metastatic PDAC. Higher tumor baseline IL1RAP expression correlated with better outcome.

MUC16 Retention after Neoadjuvant Chemotherapy in Pancreatic Ductal Adenocarcinoma.

Background/Objectives: Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis. Currently, surgical resection is the only potentially curative treatment. Unfortunately, less than 20% of PDAC patients are eligible for surgical resection at diagnosis. In the past few decades, neoadjuvant chemotherapy treatment (NCT) has been investigated as a way to downstage PDAC tumors for surgical resection. Fluorescence-guided surgery (FGS) is a technique that can aid in increasing complete resection rates by enhancing the tumor through passive or active targeting of a contrast agent. In active targeting, a probe (e.g., antibody) binds a protein differentially upregulated in the tumor compared to normal tissue. Mucin 16 (MUC16), a transmembrane glycoprotein, has recently been explored as an FGS target in preclinical tumor models. However, the impact of chemotherapy on MUC16 expression is unknown. Methods: To investigate this issue, immunohistochemistry was performed on PDAC patient samples. Results: We found that MUC16 expression was retained after NCT in patient samples (mean expression = 5.7) with minimal change in expression between the matched diagnostic (mean expression = 3.66) and PDAC NCT patient samples (mean expression = 4.5). Conclusions: This study suggests that MUC16 is a promising target for FGS and other targeted therapies in PDAC patients treated with NCT.

68Ga-DOTATATE Uptake in Pancreatic Ductal Adenocarcinoma.

We present a case of pancreatic ductal adenocarcinoma showing increased 68Ga-DOTATATE uptake mimicking neuroendocrine tumor on PET/CT. Immunohistochemical analyses of the resected tumor showed the tumor cells were negative for somatostatin receptor subtype 2. The nonneoplastic islets scattered throughout the tumor were positive for somatostatin receptor subtype 2, which may be responsible for the increased 68Ga-DOTATATE uptake. This case indicates that pancreatic ductal adenocarcinoma should be included in the differential diagnosis of the pancreatic lesions showing increased 68Ga-DOTATATE uptake.

Integrated analysis of microbiome and metabolome reveals signatures in PDAC tumorigenesis and prognosis.

We conducted a large sample-size pancreatic adenocarcinoma microbiome study using a novel microbiome sequencing method and two metabolomic assays. Two significant outcomes of our analysis are: (i) commensal opportunistic pathogens Staphylococcus aureus, Cutibacterium acnes, and Cutibacterium granulosum were enriched in pancreatic ductal adenocarcinoma (PDAC) tumors compared with normal adjacent tissues, and (ii) worse overall survival was found related to the presence of Ralstonia pickettii_B. Microbial species affect the tumorigenesis, metastasis, and prognosis of PDAC via unique microbe-enzyme-metabolite interaction. Thus, our study highlights the need for further investigation of the potential associations between pancreatic microbiota-derived omics signatures, which may drive the clinical transformation of microbiome-derived strategies toward therapy-targeted bacteria.

Embedded Bioprinting of Tumor-Scale Pancreatic Cancer-Stroma 3D Models for Preclinical Drug Screening.

The establishment of organotypic preclinical models that accurately resemble the native tumor microenvironment at an anatomic human scale is highly desirable to level up in vitro platforms potential for screening candidate therapies. The bioengineering of anatomic-scaled three-dimensional (3D) models that emulate native tumor scale while recapitulating their cellular and matrix components remains, however, to be fully realized. In this focus, herein, we leveraged embedded 3D bioprinting for biofabricating pancreatic ductal adenocarcinoma (PDAC) in vitro models combining gelatin-methacryloyl and hyaluronic acid methacrylate extracellular matrix (ECM)-mimetic biomaterials with human pancreatic cancer cells and cancer-associated fibroblasts to generate in vitro models capable of emulating native tumor size (∼6 mm) and stromal elements. By using a viscoelastic continuous polymeric supporting bath, tumor-scale 3D models were rapidly generated (∼50 constructs/h) and easily recovered following in-bath visible light photocrosslinking. As a proof-of-concept, tissue-scale constructs displaying physiomimetic designs were biofabricated. These models also encompass the incorporation of a stromal compartment to better emulate the cellular components of the PDAC native tumor microenvironment (TME) and its stratified spatial organization. Cell-laden tumor-size constructs remained viable for up to 14 days and were responsive to Gemcitabine in a dose-dependent mode. Cancer-stroma models also exhibited increased drug resistance compared to their monotypic counterparts, highlighting the key role of stromal cells in chemotherapeutic resistance. Overall, we report for the first time the freeform biofabrication of PDAC models exhibiting anatomic scale, different structural complexities, and engineered cancer-stromal compartments, being highly valuable for preclinical screening of therapeutics.

Intraperitoneal administration of adeno-associated virus encoding microRNA-29b for the treatment of peritoneal metastasis.

This study explores a novel therapeutic approach for peritoneal metastasis (PM) using AAV-mediated delivery of tumor suppressor microRNA-29b (miR-29b) to peritoneal mesothelial cells (PMC). AAV serotypes 2 and DJ demonstrate high transduction efficiency for human and murine PMC, respectively. In vitro analysis indicates that AAV vectors encoding miR-29b precursor successfully elevate miR-29b expression in PMC and their secreted small extracellular vesicle (sEV), thereby inhibiting mesothelial mesenchymal transition and reducing subsequent attachment of tumor cells. A single intraperitoneal (IP) administration of AAV-DJ-miR-29b demonstrates robust and sustained transgene expression, suppressing peritoneal fibrosis and inhibiting the development of PM from gastric and pancreatic cancers. Additionally, AAV-DJ-miR-29b enhances the efficacy of IP chemotherapy using paclitaxel, restraining the growth of established PM. While conventional gene therapy for cancer encounters challenges targeting tumor cells directly but delivering miRNA to the tumor stroma offers a straightforward and efficient means of altering the microenvironment, leading to substantial inhibition of tumor growth. AAV-mediated miR-29b delivery to peritoneum via IP route presents a simple, minimally invasive, and promising therapeutic strategy for refractory PM.

Stress granules are not present in Kras mutant cancers and do not control tumor growth

Stress granules (SG) are membraneless ribonucleoprotein-based cytoplasmic organelles that assemble in response to stress. Their formation is often associated with an almost global suppression of translation, and the aberrant assembly or disassembly of these granules has pathological implications in neurodegeneration and cancer. In cancer, and particularly in the presence of oncogenic KRAS mutations, in vivo studies concluded that SG increase the resistance of cancer cells to stress. Hence, SG have recently been considered a promising target for therapy. Here, starting from our observations that genes coding for SG proteins are stimulated during development of pancreatic ductal adenocarcinoma, we analyze the formation of SG during tumorigenesis. We resort to in vitro, in vivo and in silico approaches, using mouse models, human samples and human data. Our analyses do not support that SG are formed during tumorigenesis of KRAS-driven cancers, at least that their presence is not universal, leading us to propose that caution is required before considering SG as therapeutic targets.

Pirfenidone antagonizes TGF-β1-mediated gabapentin resistance via reversal of desmoplasia and the ‘cold’ microenvironment in pancreatic cancer

Owing to the desmoplastic stroma constituted by cancer-associated fibroblasts (CAFs), few immune cells infiltrate the pancreatic ductal adenocarcinoma (PDAC). Gabapentin can impede the production of ketoacids by CAFs to support cancer cells. However, in our study, we discovered a dose-dependent increase in transforming growth factor β1 (TGF-β1) levels in cancer cells in response to gabapentin. This reverse increase of TGF-β1 contributes to 'Gabapentin-resistance', leading to the antitumor effects on PDAC cell lines are negatively negotiated in the presence of pancreatic stellate cells. Pirfenidone synergistically inhibited the growth and apoptosis resistance of PDAC when combined with Gabapentin. In a mouse orthotopic PDAC model, Fe3+-mediated coordination nanodrugs, which contain gabapentin, pirfenidone and the natural polyphenol (EGCG), efficiently promoted the infiltration of naïve CD8+ T cells (CD44lowCD62Lhigh) and the accumulation of inflammatory CAFs (α-SMAlowIL-6high). This led to a nearly two-fold increase in survival compared to the control. Furthermore, we identified a new subpopulation as Hmox1highiCAFs following treatment with our nanodrugs. Hmox1highiCAFs overexpressed the Cxcl10 receptor (Sdc4) and facilitated functional CD8+ T-cell infiltration through the Tnfsf9-Tnfrsf9 axis. Overall, our nanodrugs reshape the phenotype of CAFs and enhance functional CD8+ T-cell infiltration into tumors, holding the potential to be a safe and promising therapy for PDAC.

Oncolytic vaccinia virus armed with 4–1BBL elicits potent and safe antitumor immunity and enhances the therapeutic efficiency of PD-1/PD-L1 blockade in a pancreatic cancer model

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal disease with a poor prognosis. Mono-immunotherapy, such as blockade of the PD-1/PD-L1 pathway, for PDAC has proven to be less effective. The systemic exertion of 4–1BB signaling enhanced antitumor immunity accompanied by hepatotoxicity, which is an obstacle for its clinical application. Our study exploits an oncolytic virus armed with 4–1BBL (VV-ΔTK-4L) to locally express 4–1BBL in the tumor microenvironment (TME), thus avoiding hepatotoxicity. VV-ΔTK-4L prolonged the survival time of a pancreatic tumor mouse model and modified the immune status of the TME and spleen. In the TME, the quantities of CD45+ cells, NK1.1+ cells, CD11c+ DCs, CD3+T, CD4+T, and CD8+T cells increased. Compared to VV-ΔTK treatment, VV-ΔTK-4L further increases the number of CD8+T cells with effector phenotypes, and downregulates exhaustion-related molecules on CD8+T cells, and does not increase the proportion of Foxp3+T cells. Thus, the TME of pancreatic cancer was converted from “cold” to “hot” by VV-ΔTK-4L. Blockade of the PD-1/PD-L1 pathway combined with VV-ΔTK-4L further significantly improves the survival ratio of a tumor-bearing mouse model. This study provides a systemic therapeutic strategy and approach for PDAC immunotherapy.

Association of dietary inflammatory index and dietary oxidative balance score with gastrointestinal cancers in NHANES 2005–2018

Background&AimsGastrointestinal (GI) cancers, including gastric, liver, esophageal, pancreatic, and colorectal cancers, represent significant global health burdens. Emerging evidence suggests that dietary patterns, particularly their inflammatory and oxidative properties, may influence cancer risk. The Dietary Inflammatory Index (DII) and Dietary Oxidative Balance Score (DOBS) assess the inflammatory and oxidative effects of diets, respectively. This study aims to explore the association between DII, DOBS, and the combined risk of GI cancers, and investigates the potential mediating roles of serum albumin and red cell distribution width (RDW).MethodsData from 26,320 participants in the NHANES 2005–2018 cycles were analyzed. DII was calculated based on 28 dietary components, and DOBS included 17 nutrients (3 pro-oxidants and 14 antioxidants). Logistic regression models assessed the associations between DII, DOBS, and GI cancers. Restricted cubic spline (RCS) models examined dose-response relationships. Mediation analysis evaluated the roles of serum albumin and RDW. Subgroup analyses explored interactions with demographic and health-related factors.ResultsHigher DII was associated with increased GI cancer risk (OR: 1.26, 95% CI: 1.07–1.49 per unit increase), while higher DOBS was associated with reduced risk (OR: 0.90, 95% CI: 0.76–0.99 per unit increase). RCS analysis indicated a significant nonlinear relationship between DII and GI cancer risk. Serum albumin and RDW partially mediated the associations between DII, DOBS, and GI cancers. Subgroup analyses showed stronger associations for DII among certain demographics, and significant interactions were found between DII and BMI. For DOBS, significant interactions were observed with age and BMI.ConclusionThis study reveals significant associations between dietary inflammatory and oxidative balance scores and GI cancer risk. Higher DII is linked to increased risk, while higher DOBS is protective. The mediating roles of serum albumin and RDW provide insights into underlying mechanisms. These findings underscore the potential of dietary modifications in GI cancer prevention and management, emphasizing the importance of anti-inflammatory and antioxidant-rich diets.

Molecular and Clinical Features of Pancreatic Acinar Cell Carcinoma: A Single-Institution Case Series

Objectives: Acinar cell carcinoma (ACC) accounts for about 1% of pancreatic cancers. The molecular and clinical features of ACC are less characterized than those of pancreatic ductal adenocarcinoma. Methods: We retrospectively evaluated the clinical and molecular features of ACC patients who underwent germline and/or somatic molecular testing at The University of Texas MD Anderson Cancer Center from 2008 to 2022 and two cases from 2023–2024 who underwent RNA and TME analysis by Boston Gene. Patient information was extracted from our institutional database with the approval of the Institutional Review Board. Results: We identified 16 patients with available molecular testing results. Fourteen patients had metastatic disease, one had borderline resectable disease, and one had localized resectable disease at diagnosis. Fifteen patients were wild type for KRAS (one patient had unknown KRAS status). Somatic/germline mutations of DNA damage repair genes (BRCA1/2, PALB2, and ATM) were present in 5 of 12 patients tested for these genes. One patient was found to have RET fusion and responded favorably to selpercatinib for over 42 months. The median overall survival (OS) was 24 months for patients with metastatic disease. One of the additional two cases who underwent BostonGene testing was found to have NTRK1 fusion. RNA and TME analysis by Boston Gene of the two cases reported immune desert features and relatively lower RNA levels of CEACAM5, CD47, CD74, and MMP1 and higher RNA levels of CDH6 compared with PDAC.

The Evolutionary Forest of Pancreatic Cancer.

The genomic features of pancreatic ductal adenocarcinoma (PDAC) have been well described, yet the evolutionary contexts within which those features occur remains unexplored. We studied the genome landscapes, phylogenies and clonal compositions of 91 PDACs in relation to clinicopathologic features. There was no difference in the number of driver mutations or the evolutionary timing that each mutation occurred. High truncal density, a metric of the accumulation of somatic mutations in the lineage that gave rise to each PDAC, was significantly associated with worse overall survival. Polyclonal, monoclonal or mixed polyclonal/monoclonal metastases were identified across the cohort highlighting multiple forms of inter-tumoral heterogeneity. Advanced stage and treated PDACs had higher odds of being polyclonal, whereas oligometastatic PDACs had fewer driver alterations, a lower fractional allelic loss and increased likelihood of being monoclonal. In sum, our findings reveal novel insights into the dynamic nature of the PDAC genome beyond established genetic paradigms.

Prediction of homologous recombination deficiency from routine histology with attention-based multiple instance learning in nine different tumor types.

Homologous recombination deficiency (HRD) is recognized as a pan-cancer predictive biomarker that potentially indicates who could benefit from treatment with PARP inhibitors (PARPi). Despite its clinical significance, HRD testing is highly complex. Here, we investigated in a proof-of-concept study whether Deep Learning (DL) can predict HRD status solely based on routine hematoxylin & eosin (H&E) histology images across nine different cancer types. We developed a deep learning pipeline with attention-weighted multiple instance learning (attMIL) to predict HRD status from histology images. As part of our approach, we calculated a genomic scar HRD score by combining loss of heterozygosity (LOH), telomeric allelic imbalance (TAI), and large-scale state transitions (LST) from whole genome sequencing (WGS) data of n = 5209 patients across two independent cohorts. The model's effectiveness was evaluated using the area under the receiver operating characteristic curve (AUROC), focusing on its accuracy in predicting genomic HRD against a clinically recognized cutoff value. Our study demonstrated the predictability of genomic HRD status in endometrial, pancreatic, and lung cancers reaching cross-validated AUROCs of 0.79, 0.58, and 0.66, respectively. These predictions generalized well to an external cohort, with AUROCs of 0.93, 0.81, and 0.73. Moreover, a breast cancer-trained image-based HRD classifier yielded an AUROC of 0.78 in the internal validation cohort and was able to predict HRD in endometrial, prostate, and pancreatic cancer with AUROCs of 0.87, 0.84, and 0.67, indicating that a shared HRD-like phenotype occurs across these tumor entities. This study establishes that HRD can be directly predicted from H&E slides using attMIL, demonstrating its applicability across nine different tumor types.

Development of a Machine‐Learning Model for Diagnosis of Pancreatic Cancer from Serum Samples Analyzed by Thermal Liquid Biopsy

Pancreatic ductal adenocarcinoma (PDAC) poses a considerable diagnostic and therapeutic challenge due to the lack of specific biomarkers and late diagnosis. Early detection is crucial for improving prognosis, but current techniques are insufficient. An innovative approach based on differential scanning calorimetry (DSC) of blood serum samples, thermal liquid biopsy (TLB), combined with machine‐learning (ML) analysis, may offer a more efficient method for diagnosing PDAC. Serum samples from a cohort of 212 PDAC patients and 184 healthy controls are studied. DSC thermograms are analyzed using ML models. The generated models are built applying algorithms based on penalized regression, resampling, categorization, cross validation, and variable selection. The ML‐based model demonstrates outstanding ability to discriminate between PDAC patients and control subjects, with a sensitivity of 90% and an area under the ROC receiver operating characteristic curve of 0.83 in the training and test groups. Application of the model to an independent validation cohort of 113 PDAC patients confirms its robustness and utility as a diagnosis tool. The application of ML to serum TLB data emerges as a promising methodology for early diagnosis, representing a significant advance for detecting and managing PDAC, envisaging a minimally invasive and more efficient methodology for identifying biomarkers.