Recommendations for a More Organized and Effective Approach to the Early Detection of Pancreatic Cancer From the PRECEDE (Pancreatic Cancer Early Detection) Consortium

Biomarkers, omics and artificial intelligence for early detection of pancreatic cancer

Pancreatic ductal adenocarcinoma (PDAC) is frequently diagnosed in its late stages when treatment options are limited. Unlike other common cancers, there are no population-wide screening programmes for PDAC. Thus, early disease detection, although urgently needed, remains elusive. Individuals in certain high-risk groups are, however, offered screening or surveillance. Here we explore advances in understanding high-risk groups for PDAC and efforts to implement biomarker-driven detection of PDAC in these groups. We review current approaches to early detection biomarker development and the use of artificial intelligence as applied to electronic health records (EHRs) and social media. Finally, we address the cost-effectiveness of applying biomarker strategies for early detection of PDAC.

Targeting pancreatic cancer screening by identification of pathogenic variants of BRCA2/ BRCA1 in healthy individuals who have no known family history of pancreatic cancer: The arguments for and against

The majority of patients with pancreatic ductal adenocarcinoma (PDAC) are no longer suitable for treatment with curable intent at the time of diagnosis resulting in a 5-year survival of less than 10 %. Imaging of asymptomatic individuals could identify early cancers, but only with a risk of falsely identifying a benign lesion as malignant. Screening of an unselected population would result in far more such false positives than true early cancers. Selection before screening is therefore essential, but there are very few populations at high enough risk to make screening more beneficial than counterproductive. These populations include carriers of specific mutations in BRCA2, and arguably BRCA1, who have a family history of PDAC. These pathogenic mutations all have a predictable effect in making loss of Homologous Recombination Repair (HRR) likely in a carrier's lifetime. In this review the impact of such loss of HRR function on the likelihood of PDAC development will be discussed. Furthermore, it will be discussed whether the identification of a germline pathogenic mutation is sufficient to justify carrier surveillance for the development of the malignancy, or whether the current practice of screening only those carriers with a close relative diagnosed with PDAC is justifiable, as only a proportion of carriers are at high risk. The review will go beyond this to discuss whether there is an essential need to better define and stratify those at high risk, so that only high-risk carriers are put on surveillance.

Identification of patients at risk for pancreatic cancer in a 3-year timeframe based on machine learning algorithms

Early detection of pancreatic cancer (PC) remains challenging largely due to the low population incidence and few known risk factors. However, screening in at-risk populations and detection of early cancer has the potential to significantly alter survival. In this study, we aim to develop a predictive model to identify patients at risk for developing new-onset PC at two and a half to three year time frame. We used the Electronic Health Records (EHR) of a large medical system from 2000 to 2021 (N = 537,410). The EHR data analyzed in this work consists of patients' demographic information, diagnosis records, and lab values, which are used to identify patients who were diagnosed with pancreatic cancer and the risk factors used in the machine learning algorithm for prediction. We identified 73 risk factors of pancreatic cancer with the Phenome-wide Association Study (PheWAS) on a matched case-control cohort. Based on them, we built a large-scale machine learning algorithm based on EHR. A temporally stratified validation based on patients not included in any stage of the training of the model was performed. This model showed an AUROC at 0.742 [0.727, 0.757] which was similar in both the general population and in a subset of the population who has had prior cross-sectional imaging. The rate of diagnosis of pancreatic cancer in those in the top 1 percentile of the risk score was 6 folds higher than the general population. Our model leverages data extracted from a 6-month window of time in the electronic health record to identify patients at nearly sixfold higher than baseline risk of developing pancreatic cancer 2.5-3 years from evaluation. This approach offers an opportunity to define an enriched population entirely based on static data, where current screening may be recommended.

Advancements in early detection of pancreatic cancer: the role of artificial intelligence and novel imaging techniques

Early detection is crucial for improving survival rates of pancreatic ductal adenocarcinoma (PDA), yet current diagnostic methods can often fail at this stage. Recently, there has been significant interest in improving risk stratification and developing imaging biomarkers, through novel imaging techniques, and most notably, artificial intelligence (AI) technology. This review provides an overview of these advancements, with a focus on deep learning methods for early detection of PDA.

Recommendations, evidence and sustainability of screening for pancreatic cancer in high-risk individuals

Pancreatic cancer is a highly lethal malignancy and is predicted to become the second leading cause of cancer-related deaths by 2030. Early detection significantly improves outcomes, but general population screening remains infeasible due to the low prevalence of the disease and lack of specific biomarkers. This review evaluates current recommendations for pancreatic cancer surveillance in high-risk individuals, synthesises evidence from recent studies and explores the sustainability of current imaging-based surveillance programmes. Challenges such as overdiagnosis, economic feasibility and disparities in access highlight the need for targeted, cost-effective strategies. Collaborative initiatives and consortia are needed to advance biomarker research and refine risk stratification. By integrating evidence-based recommendations with sustainable approaches, this review outlines pathways to improve early detection and reduce mortality from pancreatic cancer.

Practices and perspectives of genetic counselors about high-risk pancreatic cancer screening: A cross-sectional survey study

Surveillance of individuals at high-risk of pancreatic cancer using CAPS criteria and other expert consensus guidelines may result in earlier pancreatic cancer detection in some cases; therefore, clinicians are responsible for appropriately identifying and referring these individuals to appropriate high-risk pancreas cancer screening programs. This study aimed at assessing the perspective, knowledge, and clinical practices of cancer genetic counselors surveyed nationwide towards identification of individuals at high-risk of pancreatic cancer and utilization of high-risk pancreatic cancer screening programs. One hundred and eighty-nine genetic counselors who listed "Cancer" as their specialty on the NSGC website responded to the survey, which consisted of multiple practice-based, knowledge-based, and clinical vignette-based questions. Almost 70% of the genetic counselors surveyed accurately identified when an individual would be considered for high-risk pancreatic cancer screening, when using 2019 CAPS consensus guidelines as a benchmark. Access to high-risk pancreatic cancer screening programs and increased provider comfort in counseling individuals at high-risk of pancreatic cancer were found to be statistically associated in accurate identification of high-risk individuals in three of the clinical vignettes. Additionally, 60% of genetic counselors reported the majority of high-risk individuals accept a referral for pancreatic cancer screening, which shows a high uptake of patients accepting referrals from genetic counselors. Genetic counselors have high accuracy in determining who is eligible for high-risk pancreas screening; thus, they are the ideal providers for initiating referrals to high-risk pancreatic cancer screening programs. Genetic counseling programs and high-risk pancreatic cancer screening programs should establish a close working relationship to optimize the identification and subsequent referrals of high-risk individuals eligible for pancreas cancer screening.

Digestive cancers: mechanisms, therapeutics and management

Cancers of the digestive system are major contributors to global cancer-associated morbidity and mortality, accounting for 35% of annual cases of cancer deaths. The etiologies, molecular features, and therapeutic management of these cancer entities are highly heterogeneous and complex. Over the last decade, genomic and functional studies have provided unprecedented insights into the biology of digestive cancers, identifying genetic drivers of tumor progression and key interaction points of tumor cells with the immune system. This knowledge is continuously translated into novel treatment concepts and targets, which are dynamically reshaping the therapeutic landscape of these tumors. In this review, we provide a concise overview of the etiology and molecular pathology of the six most common cancers of the digestive system, including esophageal, gastric, biliary tract, pancreatic, hepatocellular, and colorectal cancers. We comprehensively describe the current stage-dependent pharmacological management of these malignancies, including chemo-, targeted, and immunotherapy. For each cancer entity, we provide an overview of recent therapeutic advancements and research progress. Finally, we describe how novel insights into tumor heterogeneity and immune evasion deepen our understanding of therapy resistance and provide an outlook on innovative therapeutic strategies that will shape the future management of digestive cancers, including CAR-T cell therapy, novel antibody-drug conjugates and targeted therapies.

Aspects and outcomes of surveillance for individuals at high-risk of pancreatic cancer

Pancreatic ductal adenocarcinoma (PDAC) is a leading cause of cancer-related deaths and is associated with a poor prognosis. The majority of these cancers are detected at a late stage, contributing to the bad prognosis. This underscores the need for novel, enhanced early detection strategies to improve the outcomes. While population-based screening is not recommended due to the relatively low incidence of PDAC, surveillance is recommended for individuals at high risk for PDAC due to their increased incidence of the disease. However, the outcomes of pancreatic cancer surveillance in high-risk individuals are not sorted out yet. In this review, we will address the identification of individuals at high risk for PDAC, discuss the objectives and targets of surveillance, outline how surveillance programs are organized, summarize the outcomes of high-risk individuals undergoing pancreatic cancer surveillance, and conclude with a future perspective on pancreatic cancer surveillance and novel developments.

The Spanish Familial Pancreatic Cancer Registry (PANGENFAM): a decade follow-up of individuals at high-risk for pancreatic cancer

The Spanish Familial Pancreatic Cancer Registry (PANGENFAM) was established in 2009 and aims to characterize the genotype and phenotype of familial pancreatic cancer (FPC). Furthermore, an early detection screening program for pancreatic ductal adenocarcinoma (PDAC) is provided to healthy high-risk individuals from FPC and hereditary pancreatic cancer families (first-degree relatives). This article describes our experience over the last 10 years in high-risk screening. Hereditary and familial pancreatic cancer families were identified through the oncology and gastroenterology units. High-risk individuals underwent annual screening with endoscopic ultrasound (EUS) and magnetic resonance (MRI) from age 40 or 10 years younger than the youngest affected family member. Results: PANGENFAM has enrolled 290 individuals from 143 families, including 52 PDAC cases and 238 high-risk individuals. All high-risk individuals eligible for screening were offered to enter the surveillance program, with 143 currently participating. Pancreatic abnormalities were detected in 94 individuals (median age 53 years (29-83), with common findings including cystic lesions and inhomogeneous parenchyma. Imaging test concordance was 66%. Surgical intervention was performed in 4 high-risk individuals following highly suspicious lesions detected by imaging. PANGENFAM is a valuable resource for science innovation, such as biobanking, with clinical and imaging data available for analysis. For high-risk families, it may offer a potential for early diagnosis. Collaboration with other national and international registries is needed to increase our understanding of the disease biology and to standardize criteria for inclusion and follow-up, optimizing cost-effectiveness and efficacy.

The Italian registry of families at risk for pancreatic cancer (IRFARPC): implementation and evolution of a national program for pancreatic cancer surveillance in high-risk individuals

Screening programs for early detection and treatment of pancreatic cancer (PC) and its precursor lesions are increasingly implemented worldwide to reduce disease-specific lethality. Given the relatively low prevalence of the disease, the ideal target of such approaches is an enriched cohort of individuals harboring a lifetime risk of developing PC significantly higher compared to the general population, given either a substantial aggregation of PC cases in their family (i.e. familial pancreatic cancer) or a genomic landscape enriched with pathogenic variants associated with pancreatic carcinogenesis (i.e. mutation carriers). In Italy, a national registry for the census and surveillance of high-risk individuals for PC was launched in 2015, enrolling some 1200 subjects as of today. In this perspective, the scientific background, multi-level structure, and evolution of IRFARPC are outlined, as well as its long-term results, future developments, and areas for improvement.

Familial pancreatic cancer: a long fruitful journey

In the early years of my GI fellowship, a healthy 40-year-old man came to my clinic and announced that he was going to die of pancreatic cancer. His brothers, father and uncles had all died of the disease; he felt his fate was inescapable. I asked whether his family members had seen doctors or had any tests. His answer was yes to both. Even so, doctors could not diagnose the pancreatic cancer at early stages. CT scans were always negative. I thought to myself, in order to help this patient-CT scans may not be reliable for early detection. Perhaps other methods of imaging the pancreas might be of more benefit. This patient opened a door that led to a 30-year journey of trying to detect pancreatic cancer at earlier stages when it is curable.

Outcomes of a 3-Year Prospective Surveillance in Individuals at High Risk of Pancreatic Cancer

INTRODUCTION

Pancreatic cancer (PC) surveillance of high-risk individuals (HRI) is becoming more common worldwide, aiming at anticipating PC diagnosis at a preclinical stage. In 2015, the Italian Registry of Families at Risk of Pancreatic Cancer was created. We aimed to assess the prevalence and incidence of pancreatic findings, oncological outcomes, and harms 7 years after the Italian Registry of Families at Risk of Pancreatic Cancer inception, focusing on individuals with at least a 3-year follow-up or developing events before.

METHODS

HRI (subjects with a family history or mutation carriers with/without a family history were enrolled in 18 centers). They underwent annual magnetic resonance with cholangiopancreatography or endoscopic ultrasound (NCT04095195).

RESULTS

During the study period (June 2015-September 2022), 679 individuals were enrolled. Of these, 524 (77.2%) underwent at least baseline imaging, and 156 (29.8%) with at least a 3-year follow-up or pancreatic malignancy/premalignancy-related events, and represented the study population. The median age was 51 (interquartile range 16) years. Familial PC cases accounted for 81.4% of HRI and individuals with pathogenic variant for 18.6%. Malignant (n = 8) and premalignant (1 PanIN3) lesions were found in 9 individuals. Five of these 8 cases occurred in pathogenic variant carriers, 4 in familial PC cases (2 tested negative at germline testing and 2 others were not tested). Three of the 8 PC were stage I. Five of the 8 PC were resectable, 3 Stage I, all advanced cases being prevalent. The 1-, 2-, and 3-year cumulative hazard of PC was 1.7%, 2.5%, and 3%, respectively. Median overall and disease-free survival of patients with resected PC were 18 and 12 months (95% CI not computable). Considering HRI who underwent baseline imaging, 6 pancreatic neuroendocrine neoplasms (1 resected) and 1 low-yield surgery (low-grade mixed-intraductal papillary mucinous neoplasm) were also reported.

DISCUSSION

PC surveillance in a fully public health care system is feasible and safe, and leads to early PC or premalignant lesions diagnoses, mostly at baseline but also over time.

Germline mutations of homologous recombination genes and clinical outcomes in pancreatic cancer: a multicenter study in Taiwan

BACKGROUND

Cancer susceptibility germline mutations are associated with pancreatic ductal adenocarcinoma (PDAC). However, the hereditary status of PDAC and its impact on survival is largely unknown in the Asian population.

METHODS

Exome sequencing was performed on 527 blood samples from PDAC individuals and analyzed for mutations in 80 oncogenic genes. Pathogenic and likely pathogenic (P/LP) germline variants were diagnosed according to the ACMG variant classification categories. The association between germline homologous recombination gene mutations (gHRmut, including BAP1, BRCA1, BRCA2, PALB2, ATM, BLM, BRIP1, CHEK2, NBN, MUTYH, FANCA and FANCC) and the treatment outcomes was explored in patients with stage III/IV diseases treated with first-line (1L) platinum-based versus platinum-free chemotherapy.

RESULTS

Overall, 104 of 527 (19.7%) patients carried germline P/LP variants. The most common mutated genes were BRCA2 (3.60%), followed by ATR (2.66%) and ATM (1.9%). After a median follow-up duration of 38.3-months (95% confidence interval, 95% CI 35.0-43.7), the median overall survival (OS) was not significantly different among patients with gHRmut, non-HR germline mutations, or no mutation (P = 0.43). Among the 320 patients with stage III/IV disease who received 1L combination chemotherapy, 32 (10%) had gHRmut. Of them, patients receiving 1L platinum-based chemotherapy exhibited a significantly longer median OS compared to those with platinum-free chemotherapy, 26.1 months (95% CI 12.7-33.7) versus 9.6 months (95% CI 5.9-17.6), P = 0.001. However, the median OS of patients without gHRmut was 14.5 months (95% CI 13.2-16.9) and 12.6 months (95% CI 10.8-14.7) for patients receiving 1L platinum-based and platinum-free chemotherapy, respectively (P = 0.22). These results were consistent after adjusting for potential confounding factors including age, tumor stage, performance status, and baseline CA 19.9 in the multivariate Cox regression analysis.

CONCLUSIONS

Our study showed that nearly 20% of Taiwanese PDAC patients carried germline P/LP variants. The longer survival observed in gHRmut patients treated with 1L platinum-based chemotherapy highlights the importance of germline testing for all patients with advanced PDAC at diagnosis.

Surveillance of Individuals at High Risk of Developing Pancreatic Cancer: A Prevalence Meta-analysis to Estimate the Rate of Low-yield Surgery

OBJECTIVE

To quantify the rate of low-yield surgery, defined as no high-grade dysplastic precursor lesions or T1N0M0 pancreatic cancer at pathology, during pancreatic cancer surveillance.

BACKGROUND

Global efforts have been made in pancreatic cancer surveillance to anticipate the diagnosis of pancreatic cancer at an early stage and improve survival in high-risk individuals (HRIs) with a hereditary predisposition. The negative impact of pancreatic cancer surveillance when surgery is performed for low-grade dysplasia or a non-neoplastic condition is not well quantified.

MATERIALS AND METHODS

A systematic search and prevalence meta-analysis was performed for studies reporting surgery with final diagnoses other than those defined by the Cancer of the Pancreas Screening (CAPS) goals from January 2000 to July 2023. The secondary outcome was the pooled proportion of final diagnoses matching the CAPS goals (PROSPERO: #CRD42022300408).

RESULTS

Twenty-three articles with 5027 patients (median 109 patients/study, interquartile range 251) were included. The pooled prevalence of low-yield surgery was 2.1% (95% CI: 0.9-3.7, I2 : 83%). In the subgroup analysis, this prevalence was nonsignificantly higher in studies that only included familial pancreatic cancer subjects without known pathogenic variants, compared with those enrolling pathogenic variant carriers. No effect modifiers were found. Overall, the pooled prevalence of subjects under surveillance who had a pancreatic resection that contained target lesions was 0.8% (95% CI, 0.3-1.5, I2 : 24%]. The temporal analysis showed that the rate of low-yield surgeries decreased in the last decades and stabilized at around 1% (test for subgroup differences P <0.01).

CONCLUSIONS

The risk of "low-yield" surgery during pancreatic cancer surveillance is relatively low but should be thoroughly discussed with individuals under surveillance.

Large-scale pancreatic cancer detection via non-contrast CT and deep learning

Pancreatic ductal adenocarcinoma (PDAC), the most deadly solid malignancy, is typically detected late and at an inoperable stage. Early or incidental detection is associated with prolonged survival, but screening asymptomatic individuals for PDAC using a single test remains unfeasible due to the low prevalence and potential harms of false positives. Non-contrast computed tomography (CT), routinely performed for clinical indications, offers the potential for large-scale screening, however, identification of PDAC using non-contrast CT has long been considered impossible. Here, we develop a deep learning approach, pancreatic cancer detection with artificial intelligence (PANDA), that can detect and classify pancreatic lesions with high accuracy via non-contrast CT. PANDA is trained on a dataset of 3,208 patients from a single center. PANDA achieves an area under the receiver operating characteristic curve (AUC) of 0.986-0.996 for lesion detection in a multicenter validation involving 6,239 patients across 10 centers, outperforms the mean radiologist performance by 34.1% in sensitivity and 6.3% in specificity for PDAC identification, and achieves a sensitivity of 92.9% and specificity of 99.9% for lesion detection in a real-world multi-scenario validation consisting of 20,530 consecutive patients. Notably, PANDA utilized with non-contrast CT shows non-inferiority to radiology reports (using contrast-enhanced CT) in the differentiation of common pancreatic lesion subtypes. PANDA could potentially serve as a new tool for large-scale pancreatic cancer screening.

Outcomes of the IMMray PanCan-d Test in High-Risk Individuals Undergoing Pancreatic Surveillance: Pragmatic Data and Lessons Learned

PURPOSE

An effective blood-based test for pancreatic cancer (PC) screening has remained elusive. The IMMray PanCan-d is the first commercially available blood-based test specifically designed for early detection of PC; however, outcomes from its use in clinical practice have not been reported.

METHODS

We performed a blinded spike-in study of 100 individuals who had an IMMray PanCan-d test, including 94 high-risk individuals (HRIs) undergoing PC surveillance and six individuals with known PC. Specimens were processed blindly following the commercial laboratory's standardized operating procedure. Positive predictive value (PPV) and negative predictive value (NPV) were calculated.

RESULTS

Cohort characteristics included a median age of 63 (IQR, 55-70) years, 57% female, 96% non-Hispanic White, 57% with a pathogenic variant in a PC risk gene (BRCA2 most commonly-18%), and 83% with a family history of PC. Among IMMray PanCan-d results from 94 HRIs undergoing PC surveillance, there was one positive (1%), seven borderlines (7%), 73 negatives (78%), and 13 tests not performed because of low CA19-9 expression (14%). No PC was diagnosed among these HRIs; however, there were two sub-cm pancreatic neuroendocrine tumors, seven clinically diagnosed side branch intraductal papillary mucinous neoplasms ≥1 cm, and a sub-cm solid mass with indeterminate cytology requiring close follow-up; all these individuals had negative IMMray PanCan-d tests. Of the six spiked-in PCs, four (67%) yielded a positive and two (33%) yielded a negative. With an estimated disease prevalence of 2%, the PPV and NPV are 52% and 99%, respectively, if borderline results are considered negative and 12% and 99%, respectively, if borderline tests are considered positive.

CONCLUSION

In clinical practice, IMMray PanCan-d has a robust NPV; however, PPV is dramatically influenced by whether borderline results are characterized as a positive or negative result.

Are All Cysts Created Equal?: Pancreatic Cystic Neoplasms in Patients with Familial or Genetic Risk Factors for Pancreatic Cancer

Pancreatic cystic lesions (PCLs) have become more prevalent over time, particularly in asymptomatic individuals. Current screening guidelines for incidental PCLs offer a unified approach to surveillance and management, predicated on "worrisome features." Although PCLs are common in the general population, their prevalence may be higher in high-risk individuals (HRI, unaffected patients with specific familial and/or genetic risk factors). As more PCLs are diagnosed and more HRI identified, it is important to promote research that bridges data gaps and introduces nuance to risk assessment tools, ensuring tailoring of guidelines to the needs of HRI with varying pancreatic cancer risk factors.

Racial, Ethnic, and Sex-based Disparities among High-risk Individuals Undergoing Pancreatic Cancer Surveillance

Since its inception two years ago, the international, multicenter Pancreatic Cancer Early Detection (PRECEDE) Consortium has enrolled high-risk individuals (HRI) undergoing pancreatic ductal adenocarcinoma (PDAC) surveillance. Herein we aim to evaluate enrollment disparities in PRECEDE. Data on HRIs enrolled between May 2020 and March 2022 were collected, with HRIs defined as participants enrolled in PRECEDE meeting guideline-based criteria for PDAC surveillance. Of 1,273 HRIs enrolled, 1,113 were eligible for inclusion, with 47.2% meeting familial pancreatic cancer criteria without a known pathogenic variant (PV) and the remainder having a pathogenic variant in a PDAC-risk gene (CDKN2A, STK11, PRSS1, BRCA1, BRCA2, PALB2, ATM, MLH1, MSH2, MSH6, PMS2, or EPCAM). Study participants were predominantly from the United States (82.7%), the most common age range at enrollment was 60-69 years (37.4%), and a non-PDAC cancer was present in 32.4%. There were racial/ethnic- and sex-based disparities among enrolled subjects, as the majority of participants were female (65.9%) and self-reported white (87.7%), with only 2.9% having Hispanic ethnicity. While more than 97% of participants consented to utilize imaging data and biosamples for research, there was no difference in rate of consent based on race/ethnicity, sex, or age, thereby demonstrating uniform participation in research activities among all subgroups after enrollment. Ensuring that diversity of HRIs in PDAC surveillance programs mirrors the communities served by participating centers is important. Substantial racial/ethnic- and sex-based disparities persist among recently enrolled HRIs undergoing PDAC surveillance, and therefore reducing these disparities will be a major focus of the PRECEDE Consortium moving forward.

PREVENTION RELEVANCE

Pancreatic cancer surveillance is critical to decreasing pancreatic cancer mortality; therefore, it is important that pancreatic cancer surveillance studies enroll diverse patients. We demonstrate that substantial racial/ethnic- and sex-based disparities exist amongst enrollment in the international PRECEDE consortium, highlighting the dire need for future efforts to reduce these disparities. See related Spotlight, p. 305.

Early Detection of Cancers in the Era of Precision Oncology

PURPOSE OF REVIEW

The increasing global incidence of cancer demands innovative cancer detection modalities. The current population-based early cancer detection approaches focus on several major types of cancers (breast, prostate, cervical, lung and colon) at their early stages, however, they generally do not target high-risk individuals at precancerous stages.

RECENT FINDINGS

Some cancers, such as pancreatic cancer, are challenging to detect in their early stages. Therefore, there is a pressing need for improved, accessible, noninvasive, and cost-effective early detection methods. Harnessing cell-free-based biomarker-driven strategies paves a new era of precision diagnosis for multicancer early detection. The majority of these tests are in the early stages and expensive, but these approaches are expected to become cost sensitive in the near future.

SUMMARY

This review provides an overview of early cancer detection strategies, highlighting the methods, challenges, and issues to be addressed to revolutionize and improve global early cancer detection.

Pancreatic Cancer: Changing Epidemiology and New Approaches to Risk Assessment, Early Detection, and Prevention

Pancreatic cancer usually results in poor survival with limited options for treatment, as most affected individuals present with advanced disease. Early detection of preinvasive pancreatic neoplasia and identifying molecular therapeutic targets provide opportunities for extending survival. Although screening for pancreatic cancer is currently not recommended for the general population, emerging evidence indicates that pancreatic surveillance can improve outcomes for individuals in certain high-risk groups. Changes in the epidemiology of pancreatic cancer, experience from pancreatic surveillance, and discovery of novel biomarkers provide a roadmap for new strategies for pancreatic cancer risk assessment, early detection, and prevention.

Serum biomarker-based early detection of pancreatic ductal adenocarcinomas with ensemble learning

BACKGROUND

Earlier detection of pancreatic ductal adenocarcinoma (PDAC) is key to improving patient outcomes, as it is mostly detected at advanced stages which are associated with poor survival. Developing non-invasive blood tests for early detection would be an important breakthrough.

METHODS

The primary objective of the work presented here is to use a dataset that is prospectively collected, to quantify a set of cancer-associated proteins and construct multi-marker models with the capacity to predict PDAC years before diagnosis. The data used is part of a nested case-control study within the UK Collaborative Trial of Ovarian Cancer Screening and is comprised of 218 samples, collected from a total of 143 post-menopausal women who were diagnosed with pancreatic cancer within 70 months after sample collection, and 249 matched non-cancer controls. We develop a stacked ensemble modelling technique to achieve robustness in predictions and, therefore, improve performance in newly collected datasets.

RESULTS

Here we show that with ensemble learning we can predict PDAC status with an AUC of 0.91 (95% CI 0.75-1.0), sensitivity of 92% (95% CI 0.54-1.0) at 90% specificity, up to 1 year prior to diagnosis, and at an AUC of 0.85 (95% CI 0.74-0.93) up to 2 years prior to diagnosis (sensitivity of 61%, 95% CI 0.17-0.83, at 90% specificity).

CONCLUSIONS

The ensemble modelling strategy explored here outperforms considerably biomarker combinations cited in the literature. Further developments in the selection of classifiers balancing performance and heterogeneity should further enhance the predictive capacity of the method.

Tissue methylated DNA markers for sporadic pancreatic cancer are strongly associated with familial and genetically predisposed pancreatic cancer

High-risk individuals (HRIs) with familial and genetic predisposition to pancreatic ductal adenocarcinoma (PDAC) are eligible for screening. There is no accurate biomarker for detecting early-stage PDAC. We previously demonstrated that a panel of methylated DNA markers (MDMs) accurately detect sporadic PDAC. In this study we compared the distribution of MDMs in DNA extracted from tissue of PDAC cases who carry germline mutations and non-carriers with family history, with control tissue and demonstrate high discrimination like that seen in sporadic PDAC. These results provide scientific rationale for examining plasma MDMs in HRIs with the goal of developing a minimally-invasive early detection test.

Germline Aberrations in Pancreatic Cancer: Implications for Clinical Care

Pancreatic ductal adenocarcinoma (PDAC) has an extremely poor prognosis and represents a major public health issue, as both its incidence and mortality are expecting to increase steeply over the next years. Effective screening strategies are lacking, and most patients are diagnosed with unresectable disease precluding the only chance of cure. Therapeutic options for advanced disease are limited, and the treatment paradigm is still based on chemotherapy, with a few rare exceptions to targeted therapies. Germline variants in cancer susceptibility genes-particularly those involved in mechanisms of DNA repair-are emerging as promising targets for PDAC treatment and prevention. Hereditary PDAC is part of the spectrum of several syndromic disorders, and germline testing of PDAC patients has relevant implications for broad cancer prevention. Germline aberrations in BRCA1 and BRCA2 genes are predictive biomarkers of response to poly(adenosine diphosphate-ribose) polymerase (PARP) inhibitor olaparib and platinum-based chemotherapy in PDAC, while mutations in mismatch repair genes identify patients suitable for immune checkpoint inhibitors. This review provides a timely and comprehensive overview of germline aberrations in PDAC and their implications for clinical care. It also discusses the need for optimal approaches to better select patients for PARP inhibitor therapy, novel therapeutic opportunities under clinical investigation, and preclinical models for cancer susceptibility and drug discovery.

Hereditary pancreatic cancer

Pancreatic cancer is one of the deadliest malignancies. Therefore, there is an urgent need to detect pancreatic cancer in an earlier stage to improve outcomes. A variety of hereditary cancer syndromes have been associated with an increased risk of developing pancreatic cancer, and these individuals may benefit from surveillance programs. Surveillance programs have shown potential to improve outcomes, but have important risks such as overtreatment. In this review we will discuss the definitions and epidemiology of hereditary pancreatic cancer, recommendations for genetic testing and participation in surveillance. Important aspects are differences in surveillance strategies, target lesions, and potential benefits and harms of surveillance. Lastly we will highlight future directions for research and improvement of care for individuals at high-risk of pancreatic cancer.

Role of artificial intelligence in early detection and screening for pancreatic adenocarcinoma

Pancreatic adenocarcinoma remains to be one of the deadliest malignancies in the world despite treatment advancement over the past few decades. Its low survival rates and poor prognosis can be attributed to ambiguity in recommendations for screening and late symptom onset, contributing to its late presentation. In the recent years, artificial intelligence (AI) as emerged as a field to aid in the process of clinical decision making. Considerable efforts have been made in the realm of AI to screen for and predict future development of pancreatic ductal adenocarcinoma. This review discusses the use of AI in early detection and screening for pancreatic adenocarcinoma, and factors which may limit its use in a clinical setting.