Pancreatic ductal adenocarcinoma (PDAC), the most common type of pancreatic cancer, accounts for more than 90% of pancreatic tumors. While the 5-year survival rate for PDAC has increased over the last decade, it is still only 13%. However, patients with localized tumors have a 44% 5-year survival rate compared to those with nearby or distant spread whose survival rate is 17% and 3% respectively, but less than 20% of patients present with localized disease. This means that the majority of people are diagnosed with advanced disease that has limited treatment options. For this reason, early detection is one key way to increase the survival rates for people with pancreatic cancer and is a main focus for many researchers and clinicians.
One reason that PDAC is frequently diagnosed at advanced stages is because it is difficult to diagnose; the pancreas’s location in the abdomen makes it difficult to feel developing tumors and symptoms that include belly or back discomfort, fatigue, loss of appetite and mild weight loss are vague and associated with many other conditions. One question that is commonly asked is “why isn’t there screening for pancreatic cancer similar to other types of common cancers?” While there are some blood-based biomarkers currently associated with PDAC such as CA19-9 and CEA, they are also not specific to pancreatic cancer and cannot be used to screen healthy people for PDAC. Currently the only reliable way to detect pancreatic tumors is with imaging such as CT scans, MRI, or endoscopic ultrasound (EUS). Surprisingly, PDAC is also a relatively uncommon cancer type, only 13.5 people on average will be diagnosed out of 100,000 people. This low incidence of PDAC along with the lack of high sensitivity, low cost, and minimally invasive detection methods, makes screening for the general public not feasible.
In absence of screening, clinicians have implemented targeted surveillance in people with strong risk factors for developing pancreatic cancer in the hopes that early and regular monitoring can detect tumors at an early, localized stage when there are more treatment options and better outcomes.
In the 1990s Johns Hopkins established The Cancer of the Pancreas Screening (CAPS) program to monitor high-risk individuals to understand if targeted surveillance could diagnose PDAC at earlier stages and improve survival outcomes in these patients. Since that time, the CAPS consortium has demonstrated that surveillance could identify pancreatic lesions at early stages before symptoms develop and most recently that early detection results in better outcomes in high-risk patients diagnosed through surveillance.
Who is a “high-risk individual”?
There is no universal definition of who is at high-risk for developing pancreatic cancer but the following groups may qualify for targeted surveillance:
- People with mutations in the gene STK11 starting
- People with mutations in the gene CDKN2A
- People with mutations in BRCA1. BRCA2, PALB2, ATM or genes associated with Lynch Syndrome AND a 1st or 2nd degree relative with PDAC
- People with Familial Pancreatic Cancer (FPC) who have more than three 1st, 2nd, or 3rd degree relatives with PDAC
- People with mutations in BRCA2, BRCA1, ATM, or PALB2 regardless of family history
While the data show that surveillance of high-risk individuals can identify PDAC at higher rates than the general public, the vast majority of these people will not develop PDAC. One ongoing research effort is to identify additional criteria such as risk factors or biomarkers that define the highest-risk individuals. To do this, groups such as CAPS and, more recently, The Pancreatic Cancer Early Detection (PRECEDE) Consortium are following high-risk individuals through surveillance and the collection of detailed histories and biospecimens. Programs such as these have the potential to identify new risk factors and biomarkers for PDAC that may better predict who may develop the disease.
Developing AI Tools for Risk Stratification and Detection
One emerging technology being evaluated for use in identification of the highest high-risk individuals and to improve early detection of PDAC is artificial intelligence (AI). AI models are being employed in three distinct ways to identify and stratify high-risk individuals and to detect PDAC at earlier stages:
- Electronic Health Records – AI models are being developed to filter information found in electronic health records including histories, trends in lab findings, and medications to identify people who may be high-risk individuals.
- Image based detection – AI is being evaluated for its ability to detect abnormalities or changes in images of the pancreas from MRI/MRCP, CT scans, or EUS that may be undetectable to the human eye. This can aid in earlier detection of small, early-stage tumors or precancerous lesions and may provide information to distinguish which precancerous lesions or cysts are more likely to develop into PDAC in the future.
- Identification of additional biomarkers – To identify new biomarkers, large amounts of data must be collected from different sources and evaluated. AI models can more readily and quickly identify complex patterns and associations of multiple data inputs to help establish biomarkers or protein signatures that change in the earliest stages of PDAC development.
Liquid Biopsy Tests for Early Detection
Imaging technologies such as MRI and EUS are commonly used for surveillance of high-risk people for PDAC but a screening method that is relatively low cost, sensitive and accurate, as well as non-invasive could act as a bridge to these more intensive and costly procedures. While CA 19-9 is used to monitor response to treatment or recurrence, it is not sufficient on its own to diagnose PDAC. The generation of blood-based tests, also called “liquid biopsies,” is an area of active research and development.
There are multiple approaches for detecting the presence of PDAC using blood samples. The following are some of the methods being developed and some tests look at multiple types of markers.
- Protein markers: Measures the expression of proteins (like CA 19-9) that are higher or lower in patients with PDAC than in healthy people. These tests are likely to include CA 19-9 and a panel of other proteins to determine if PDAC is present.
- Cell free DNA (cfDNA) or circulating tumor DNA (ctDNA): Detects the presence of tumor DNA in the blood by looking for DNA with tumor-derived mutations or cancer-associated modifications.
- microRNAs (miRNAs): miRNA are small pieces of RNA that do not code for a gene but help regulate gene expression. Similar to proteins, there are miRNAs expressed in PDAC patients differently than in healthy people and can be detected in blood.
While there are currently no blood tests that are FDA approved for PDAC diagnosis, there are some commercial tests being evaluated in high-risk patients such as PancreaSure (from Immunovia) and Avantec (Clearnote). For a test to be used to detect and diagnose PDAC, it would need to be highly sensitive and specific. Ideally, a blood-based test could detect changes that happen early in the PDAC disease process and be able to diagnose early-stage patients. In the future, a highly sensitive and specific test could be used as a screening tool to determine those who may need additional testing or imaging.
Prevention and Interception
There are known risk factors that are associated with an increased risk for developing PDAC that include smoking, longstanding type 2 diabetes, obesity, and chronic pancreatitis. These specific risk factors can be reduced or eliminated by behavior changes such as quitting smoking, limiting alcohol, and eating a balanced diet full of nutrient rich food.
However, some of the biggest risk factors such as inherited genetic mutations and family history are outside of our control. For these individuals the concept of cancer interception is emerging. Cancer interception aims to identify precancerous lesions or protein pathway alterations early in tumor formation and disrupt (or intercept) the process to keep cancer from developing at all. The key for cancer interception to work is early identification of high-risk individuals and early detection of precancerous cysts or lesions.
An emerging approach for interception is to treat high-risk individuals with cancer vaccines. Recently, an early stage clinical trial treated high-risk individuals with the cancer vaccine mKRAS-VAX to train their T cells to identify and destroy tumor cells that express mutant KRAS protein which is present in over 90% of PDAC tumors. Though this trial showed promising results, it still needs validation in a study of a larger population of people with more long-term follow up. Additionally, researchers are working to identify signaling pathway alterations that occur during cancer initiation. An area of active investigation, the hope is to design drugs to modulate those pathways to enforce “healthy” signaling and prevent tumors from developing.
While developing more treatment options for patients with PDAC is a vital need, it’s becoming apparent that an additional way to increase survival rates for pancreatic cancer is to focus on early detection. Identification of who to enroll in targeted surveillance programs through better understanding of the characteristics defining a “high-risk individual” and more sensitive, less invasive detection methods can lead to the discovery of patients in the earliest stages of disease progression. For patients with localized PDAC, the possibility of tumor resection leads to better prognoses but for patients with pre-cancerous lesions or pathway alterations, the promise of interception could one day cure PDAC before it develops.
With new treatments, such as RAS inhibitors, now giving patients with PDAC more options, early detection offers a second path to better outcomes. In addition, research into disease interception, for patients with pre-cancerous lesions or pathway alterations, interception could one day prevent PDAC from developing.
The Hirshberg Foundation is proud to fund the catalytic research that will hopefully change the course of pancreatic cancer. We back bold science to advance early detection and cancer interception, so clinicians can find this disease in the earliest stages and halt it before tumors form. Through The Agi Hirshberg Center for Pancreatic Diseases at UCLA, we proudly support the PRECEDE Consortium and other research efforts working to turn that vision into reality.
