Search

The potential of artificial intelligence (AI) applied to clinical data from electronic health records (EHRs) to improve early detection for pancreatic and other cancers remains underexplored. The Kenner Family Research Fund, in collaboration with the Cancer Biomarker Research Group at the National Cancer Institute, organized the workshop entitled: "Early Detection of Pancreatic Cancer: Opportunities and Challenges in Utilizing Electronic Health Records (EHR)" in March 2021. The workshop included a select group of panelists with expertise in pancreatic cancer, EHR data mining, and AI-based modeling. This review article reflects the findings from the workshop and assesses the feasibility of AI-based data extraction and modeling applied to EHRs. It highlights the increasing role of data sharing networks and common data models in improving the secondary use of EHR data. Current efforts using EHR data for AI-based modeling to enhance early detection of pancreatic cancer show promise. Specific challenges (biology, limited data, standards, compatibility, legal, quality, AI chasm, incentives) are identified, with mitigation strategies summarized and next steps identified.

Supplemental digital content is available in the text., Objectives Pancreatic cancer (PC) and its treatments can result in pancreatic exocrine insufficiency that requires pancreatic enzyme replacement therapy (PERT). Appropriate PERT usage is during meals and snacks. The aim was to determine the frequency of appropriate use of PERT and its impact on symptom alleviation in PC through a patient-reported outcomes online platform. Methods Users in the Pancreatic Cancer Action Network's Patient Registry were prompted to answer a standalone questionnaire about their experience with PERT. Results Two hundred sixty-two users completed the PERT questionnaire (January 2016–January 2018). Patients who reported taking PERT with meals had higher alleviation of symptoms compared with those taking PERT prior to or after meals. Specifically, “feeling of indigestion,” “light-colored or orange stools,” and “visible food particles in stool” were significantly decreased. Patients taking PERT with meals reported weight gain and less weight loss. Conclusions Of the 89% of PC patients prescribed PERT, 65% were prescribed PERT appropriately with all meals and snacks. Overall compliance with PERT administration guidelines was low (50% [105/208]). Improvement in symptoms significantly correlated with appropriate use of PERT. Increase in PC patient and provider education about appropriate PERT usage and administration is warranted.

Despite considerable research efforts, pancreatic cancer is associated with a dire prognosis and a 5-year survival rate of only 10%. Early symptoms of the disease are mostly nonspecific. The premise of improved survival through early detection is that more individuals will benefit from potentially curative treatment. Artificial intelligence (AI) methodology has emerged as a successful tool for risk stratification and identification in general health care. In response to the maturity of AI, Kenner Family Research Fund conducted the 2020 AI and Early Detection of Pancreatic Cancer Virtual Summit (www.pdac-virtualsummit.org) in conjunction with the American Pancreatic Association, with a focus on the potential of AI to advance early detection efforts in this disease. This comprehensive presummit article was prepared based on information provided by each of the interdisciplinary participants on one of the 5 following topics: Progress, Problems, and Prospects for Early Detection; AI and Machine Learning; AI and Pancreatic Cancer-Current Efforts; Collaborative Opportunities; and Moving Forward-Reflections from Government, Industry, and Advocacy. The outcome from the robust Summit conversations, to be presented in a future white paper, indicate that significant progress must be the result of strategic collaboration among investigators and institutions from multidisciplinary backgrounds, supported by committed funders.

BackgroundContinuous-infusion 5-fluorouracil (5FU) and calcium leucovorin plus nab-paclitaxel and oxaliplatin have been shown to be active in patients with pancreatic cancer. As a protracted low-dose infusion, 5FU is antiangiogenic, and has synergy with bevacizumab. As shown in the treatment of breast cancer, bevacizumab and nab-paclitaxel are also synergetic.ObjectiveIn this paper we retrospectively analyze the survival of 65 patients with advanced pancreatic cancer who were treated with low-dose continuous (metronomic) chemotherapy given in conjunction with conventional anti-VEGF therapy.Patients and methodsSince July of 2008, we have treated 65 patients with 5FU (180mg/m2/day × 14days) via an ambulatory pump. Calcium leucovorin (20mg/m2 IV), nab-paclitaxel (60mg/m2) IV as a 30-min infusion, and oxaliplatin (50mg/m2) IV as a 60-min infusion were given on days 1, 8, and 15. Bevacizumab (5mg/kg) IV over 30min was administered on days 1 and 15. Cycles were repeated every 28-35days. There were 42 women and 23 men, and the median age was 59 years. Forty-six patients had stage IV disease.ResultsThe median survival was 19months, with 82% of patients surviving 12months or longer. The overall response rate was 49%. There were 28 patients who had received prior treatment, 15 of whom responded to therapy. Fifty-two patients had elevated CA 19-9 prior to treatment. Of these, 21 patients had 90% or greater reduction in CA 19-9 levels. This cohort had an objective response rate of 71% and a median survival of 27months. Thirty patients stopped treatment due to disease progression, and an additional 22 stopped because of toxicity. One patient died while on therapy.ConclusionsThis non-gemcitabine-based regimen resulted in higher response rates and better survival than what is commonly observed with therapy given at conventional dosing schedules. Low-dose continuous (metronomic therapy) cytotoxic chemotherapy combined with antiangiogenic therapy is safe and effective.

// Michael J. Pishvaian 1,2,* , R. Joseph Bender 1,* , Lynn M. Matrisian 3 , Lola Rahib 3 , Andrew Hendifar 4 , William A. Hoos 3 , Sam Mikhail 5 , Vincent Chung 6 , Vincent Picozzi 7 , Craig Heartwell 1 , Kimberly Mason 1 , Katelyn Varieur 1 , Metasebia Aberra 1 , Subha Madhavan 1,2 , Emanuel Petricoin III 1 and Jonathan R. Brody 1,8 1 Perthera, Inc, McLean, VA, USA 2 Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, DC, USA 3 The Pancreatic Cancer Action Network, Manhattan Beach, CA, USA 4 Cedars-Sinai Medical Center, Los Angeles, CA, USA 5 Ohio State University, Columbus, OH, USA 6 City of Hope Cancer Center, Duarte, CA, USA 7 Virginia Mason Medical Center, Seattle, WA, USA 8 Jefferson Pancreatic, and Related Cancer Center, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA, USA * These authors have contributed equally to this work Correspondence to: Jonathan R. Brody, email: // Keywords : cfDNA, pancreatic cancer, blood-based NGS Received : June 15, 2016 Accepted : September 25, 2016 Published : November 08, 2016 Abstract Recent improvements in next-generation sequencing (NGS) technology have enabled detection of biomarkers in cell-free DNA in blood and may ultimately replace invasive tissue biopsies. However, a better understanding of the performance of blood-based NGS assays is needed prior to routine clinical use. As part of an IRB-approved molecular profiling registry trial of pancreatic ductal adenocarcinoma (PDA) patients, we facilitated blood-based NGS testing of 34 patients from multiple community-based and high-volume academic oncology practices. 23 of these patients also underwent traditional tumor tissue-based NGS testing. cfDNA was not detected in 9/34 (26%) patients. Overall concordance between blood and tumor tissue NGS assays was low, with only 25% sensitivity of blood-based NGS for tumor tissue NGS. Mutations in KRAS, the major PDA oncogene, were only detected in 10/34 (29%) blood samples, compared to 20/23 (87%) tumor tissue biopsies. The presence of mutations in circulating DNA was associated with reduced overall survival (54% in mutation-positive versus 90% in mutation-negative). Our results suggest that in the setting of previously treated, advanced PDA, liquid biopsies are not yet an adequate substitute for tissue biopsies. Further refinement in defining the optimal patient population and timing of blood sampling may improve the value of a blood-based test.

Although fresh frozen tissue is the optimal source of nucleic acids for sequencing, obtaining such tissue on a population scale is cost and time prohibitive; however, DNA and RNA extraction from formalin-fixed, paraffin-embedded (FFPE) tissue has been problematic, yielding low quantity and quality DNA and RNA. Through a pilot program, the National Cancer Institute is developing a Surveillance, Epidemiology and End Results (SEER)-linked Virtual Tissue Repository (VTR) for obtaining deidentified FFPE tissue and clinical history. One such pancreatic ductal adenocarcinoma (PDAC) pilot project requested FFPE tumor and normal tissue (duodenum, lymph node, or spleen) and clinical history data on 24 matched pairs, comparing PDAC patients who survived at least five years since diagnosis with patients who died within two years through six SEER registries. A pancreatic pathologist annotated H&E stained slides for tissue dissection of corresponding unstained tumor (DNA and RNA) and normal (DNA only) tissue slides. More than 30% of tumor and normal tissue specimens had DNA with high quality for sequencing (≥20% amplifiable DNA), and the majority of DNA samples had medium to high quality (amplifiable DNA ≥ 10%) for sequencing. Although less DNA (quantity) was obtained from tumor tissue than normal, the percent amplifiable DNA was similar. DNA quantity was independent of tissue age (4 to 18 years); however, DNA integrity number (DIN) was significantly decreased in older tissue. Although RNA quantity was independent of tissue age, RNA quality (DV200) was significantly decreased in older tissue. There was no significant association between RNA quantity and quality. From the 24 pairs, sufficient DNA was obtained for whole genome sequencing (WGS) for 14 pairs, whole exome sequencing (WES) for eight pairs, and methylation studies on three pairs. Four pairs that had cases with Citation Format: Yao Yuan, Alison Van Dyke, Valentina Petkov, Aatur Singhi, William Hoos, Lola Rahib, Lynn Matrisian, Lynne Penberthy. DNA and RNA from formalin-fixed, paraffin-embedded pancreatic cancer tissue for sequencing: A pilot study of the SEER-linked Virtual Tissue Repository [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-177.

278 Background: Recent studies have expanded our knowledge of the genomic landscape of PDA. While critical and in some cases, potentially actionable alterations are being identified, limited outcomes data have thus far made it difficult to validate the relevance of these observations. Methods: The Pancreatic Cancer Action Network (PanCAN) and Perthera have facilitated commercial tumor molecular profiling for over 400 PDA pts since 2014 through KYT, and have developed a database of molecular and clinical information useful for data mining of biomarker-survival correlations. The survival significance of biomarkers was assessed using standard statistical methodology including Kaplan-Meier analysis and Cox proportional hazard models. Results: Linked molecular and outcomes data were available for 360 pts, of which 173 had treatment (tx) information available. Pathogenic mutations from targeted NGS, protein expression from IHC, and protein phosphorylation from RPPA were screened for correlations with overall survival (OS) and progression-free survival (PFS) independent of tx received. As shown in the table, mutations in 3 genes were associated with a better OS; while mutations in 8 genes were associated with poorer OS. Only two mutations were correlated with PFS in 1st or 2nd-line tx ( BRCA2 and KDM6A, worse PFS). Positive expression of 7 proteins, including TS, TOP1, and RRM1, were associated with reduced OS but were not correlated with PFS. High levels of phospho-ribosomal protein S6 were associated with both poor OS (HR=10.3, p=0.001) and poor PFS (HR=9.6, p=0.006). Conclusions: Multiple biomarkers had significant correlations with OS in PDA, while fewer were correlated with PFS. Growth of this registry database will further validate tx-specific predictive biomarkers for use in pts with multi-omic profiling data. [Table: see text]

Purpose Pancreatic cancer clinical trials open in the United States and their accrual were examined to identify opportunities to accelerate progress in the treatment of pancreatic cancer. Methods Pancreatic cancer–specific clinical trials open in the United States in the years 2011 and 2012 were obtained from the Pancreatic Cancer Action Network database. Accrual information was obtained from trial sponsors. Results The portfolio of pancreatic cancer clinical trials identified by type (adenocarcinoma or neuroendocrine), phase, disease stage, and treatment approach is reported. More than half of trials for patients with pancreatic ductal adenocarcinoma applied biologic insights to new therapeutic approaches, and 38% focused on optimization of radiation or chemotherapy delivery or regimens. In 2011, pancreatic cancer trials required total enrollment of 11,786 patients. Actual accrual to 93.2% of trials was 1,804 patients, an estimated 4.57% of the patients with pancreatic cancer alive in that year. The greatest need was for patients with resectable cancer. Trials open in 2011 enrolled an average of 15% of their total target accrual. Physician recommendations greatly influenced patients' decision to enroll or not enroll onto a clinical trial. Matching to a clinical trial within a 50-mile radius and identifying trials for recurrent/refractory disease were documented as challenges for patient accrual. Conclusion Overall trial enrollment indicates that pancreatic cancer trials open in 2011 would require 6.7 years on average to complete accrual. These results suggest that harmonizing patient supply and demand for clinical trials is required to accelerate progress toward improving survival in pancreatic cancer.

Commercial genomic and protein-based profiling was performed on tissue from pancreatic cancer patients who enrolled in the Pancreatic Cancer Action Network's Know Your Tumor (KYT) service. The service offers molecular profiling for patients with resectable or metastatic disease, in preparation for their next treatment plan. Patients are eligible to enroll at any institution, clinic, or cancer center in the US. Biopsies, tissue retrieval and testing is coordinated by Perthera, Inc. From 6/2014 to 11/2015 tumor biopsies, resected or archived tissue was obtained from 152 (53% male) pancreatic cancer patients and profiling including NGS/FISH (144 patients) and IHC (143 patients) was performed. The median age at biopsy or resection was 63 (25-83), 55% of the tissue came from liver, 16% from pancreas, 7% lung, 4% peritoneum, and 18% from another type of tissue. 87% of the patients were diagnosed with pancreatic ductal adenocarcinoma and the remaining with another form of confirmed or suspected pancreatic cancer. The most identified pathogenic mutations were KRAS (85%), TP53 (72%), CDKN2A (47%), CDKN2B (24%), SMAD4 (22%), ARID1A (10%), STK11 (7%), and BRCA2 (5%). The average number of genomic alterations (likely but not confirmed to be somatic) detected was five per patient from a panel of 320 genes. Molecular modifications that are linked to a treatment option were observed in 44% of patients. Twenty four of 144 patients (17%) had mutations in at least one DNA repair gene potentially benefiting from platinum or PARP inhibitor-based therapies, including BRCA2 (5%), ATM (4%), BAP1 (4%), SMARCA4 (3%), FANC (1%), CHEK2 (1%), and PALB2 ( To date, 7 of the 152 patients have enrolled in a clinical trial, 4 were treated with an off label treatment, 67 have yet to make their next treatment decision, and 53 are known to be deceased. KYT has successfully profiled tumors and presented targeted treatment options to patients and physicians in 33 states regardless of a patient's socioeconomic background and geographic location, moving the field a step closer towards a democratized, personalized approach to treating pancreatic cancer. Citation Format: Lola Rahib, Anitra Engebretson, Michael J. Pishvaian, Jonathan R. Brody, William A. Hoos, Emily E. Lyons, Joseph Bender, Emanuel F. Petricoin, Subha Madhavan, Craig Heartwell, Lynn M. Matrisian. Molecular profiling of pancreatic cancer patients from a wide geographical distribution across the US. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 93.

279 Background: There has been a significant increase in the utilization of MoP to facilitate the selection of therapies for pts with advanced cancers. However, coordinated efforts to perform uniform MoP and collect clinical outcomes data have only successfully been executed in limited academic settings. In an effort to “democratize” the availability of MoP for PDA patients in a broad spectrum of clinical settings, The Pancreatic Cancer Action Network (PanCAN) and Perthera initiated the KYT initiative, an academic, industry, and advocacy group collaboration. Methods: PDA pts self-referred to the Patient Central call center at PanCAN. Pts interested in enrolling in an IRB-approved registry trial were then referred to Perthera to facilitate consent, tissue acquisition, and commercial MoP of the pt’s tumor. Test results were reviewed by a team of PDA specialists in the context of the pt’s treatment (tx) history. A report detailing tx options was delivered to the pt’s oncologist. Pts are being actively followed longitudinally to assess physician acceptance of the tx options and to track outcomes. Results: From 06/2014 to 09/2015, 382 pts were enrolled into KYT. Of these, 273 pts from 192 physicians (community and academic) in 38 states were enrolled. Reasons for non-enrollment were: pts no longer interested (67%); pts too ill or not appropriate for KYT (17%); the pt’s physician would not sanction MoP (16%). To date, 162 pt tumor samples have been obtained, and 117 reports delivered. 75% of samples had adequate tissue for genomic profiling (NGS), with 87% of samples harboring a KRAS mutation. Actionable findings (i.e. linked to a specific tx option) identified by NGS included mutations in BRCA2 (5%), PALB2 (1%), ATM (4%); BRAF (2%), PIK3C/PIK3R (7%), STK11 (5%), amplification of ERBB2 (3%), FGFR (2%), PDGFR (2%); and RET fusions (2%). Of the 117 profiles, 43% revealed actionable findings; 58% presented off-label therapy options; and 48% led towards high priority clinical trials. Conclusions: MoP of pts with PDA is feasible irrespective of their geographical location or access to an academic center. Updated information on tx selection and patient outcomes will be provided.

268 Background: Obtaining an adequate tissue sample for molecular profiling to guide therapy selection for patients with advanced cancers can be clinically difficult, and/or patients may not want to undergo a biopsy. There has been a growing interest in the use of BB tests, including cell free DNA (cfDNA) and exosome/circulating tumor cell based-analyses as surrogates for tumor tissue (TT) testing. Validation of BB tests in PDA is possible because > 90% of PDAs harbor KRAS mutations – thus providing a reliable “internal control.” Methods: The Pancreatic Cancer Action Network (PanCAN) and Perthera initiated an IRB-approved registry trial for patients with pancreatic cancer wherein we facilitated commercially available, CLIA certified multi-Omic profiling including next generation DNA sequencing (NGS), immunohistochemistry, and phosphoproteomics on patient tumor samples. In a subset of these patients, we incorporated BB testing. Results: A KRAS mutation has been identified in 87% of KYT patients based on TT NGS in general. As of this report, 17 BB test results (cfDNA NGS) were available. In 8 patients we were able to compare the cfDNA NGS directly with TT NGS. BB testing identified a KRAS mutation in 1/8 compared to 8/8 from the tumor tissue. Of the 9/17 patients who did not have corresponding TT NGS, a KRAS mutation was found on BB testing in 3/9 samples. 4/17 BB cases overall revealed KRAS mutations, but when cases were filtered for patients with extensive metastatic disease and progressive disease at the time of blood sampling, the KRAS mutation rate increased to 2/5 overall. Actionable findings (i.e. linked to a specific therapeutic option) were identified in 6/17 cases, of which 3/17 had both an actionable mutation and a KRAS mutation. Conclusions: Although we are aware of the limitations of this study, we recommend that for patients who have biopsiable disease, a TT test should still be the gold standard for molecular profiling. For those without biopsiable disease, the KYT program presents an opportunity to determine the parameters that influence the probability of obtaining reliable molecular profiling information from a BB sample.