Has the Time Come to Screen Routinely for Pancreatic Cancer in High-Risk Individuals?

Much progress has been made since the first studies of pancreatic screening were published almost 20 years ago. We have a better understanding of the risk factors that define high-risk individuals, and we have begun to understand the benefits and risks of pancreatic screening using endoscopic ultrasound (EUS) and other imaging tests. However, we have not generated the necessary evidence to state that pancreatic cancer screening should be offered routinely for high-risk individuals.

The best evidence indicating that pancreatic screening is beneficial for high-risk individuals is from studies that have reported the results of long-term surveillance; these studies find that most pancreatic cancers detected by screening are stage I/II cancers, an improvement compared to the approximately 80 percent of pancreatic cancers diagnosed outside of screening that present with stage III/IV disease, and also that many of the patients who undergo pancreatic resection for lesions identified by screening have high-grade dysplasia (i.e. as pancreatic intraepithelial neoplasia-3, PanIN-3 and/or intraductal papillary mucinous neoplasia [IPMN]) as their highest-grade lesion, indicating that their screening and surgical resection was probably worthwhile.1 The major advantage of pancreatic imaging tests over blood tests is that pancreatic imaging can identify precancerous lesions that help define future cancer risk and can provide an opportunity to intervene and resect precursor lesions before they progress to pancreatic cancer, where cure is more elusive. Most of the precancerous lesions identified by pancreatic imaging in high-risk individuals are small, often multiple, pancreatic cysts (small IPMNs) of low malignant potential. Although these pancreatic cysts reflect an increased susceptibility to pancreatic neoplasia, the vast majority of high-risk individuals with these cysts will not develop to invasive pancreatic cancer. The limitation of current pancreatic imaging tests is the inability to detect PanIN. PanIN are the most common precursor lesion to pancreatic cancer, but these lesions usually do not create a discrete lesion visible by pancreatic imaging and so often the earliest lesion detected by imaging is a solid pancreatic cancer mass that emerged from a PanIN not a pancreatic cyst. Approaches to detect evidence of PanIN such as collecting and analyzing pancreatic juice samples for biomarkers, or novel molecular imaging approaches, needs further investigation.

The major advantage of pancreatic imaging tests over blood tests is that pancreatic imaging can identify precancerous lesions that help define future cancer risk.

The goal of pancreatic screening is to reduce the mortality and morbidity from pancreatic cancer. This will only be achieved once we have highly accurate screening tests that we can offer to individuals at sufficiently high-risk to warrant screening. Identifying individuals at sufficiently high-risk reduces the likelihood of having false positive results of any subsequent screening tests. The average lifetime risk of developing pancreatic cancer in the general U.S. population is approximately 1.4 percent; this risk is spread out over many decades. Individuals identified as high-risk for developing pancreatic cancer usually have a lifetime risk that is roughly five to 20 times higher than average risk. The risk of pancreatic cancer can be estimated by identifying the number of first-degree relatives who have developed the disease. Pancreatic screening guidelines recommend screening based on family history of individuals who have at least one first-degree and one second-degree relative with pancreatic cancer; if the individual carries a germline mutation in a pancreatic cancer susceptibility gene (in BRCA2, ATM, CDKN2A, PALB2, BRCA1, MLH1, MSH2, STK11), their risk is increased even without a family history of pancreatic cancer, but a detailed family history remains important to estimate their pancreatic cancer risk. The average age of pancreatic cancer at diagnosis for patients with a familial clustering of pancreatic cancer is somewhat younger than it is for those without a family history; for those diagnosed with pancreatic cancer carrying a deleterious germline pancreatic cancer susceptibility gene mutation, the average age is younger still (early 60s). For this reason, we recommend initiating pancreatic screening at age 55 for those who have a family history of pancreatic cancer in multiple close blood relatives and age 50 for most individuals who carry a germline mutation in a pancreatic cancer susceptibility gene, although more evidence is needed to better define risk in these populations.2 With direct-to-consumer gene tests becoming available, more patients are undergoing gene tests to identify the cause of their susceptibility to pancreatic and other cancers. There is also an emerging debate about the value of offering gene testing to all newly diagnosed individuals with pancreatic cancer because roughly 5 percent of them carry a deleterious germline mutation in a pancreatic cancer susceptibility gene and identification of these mutation carriers may not only inform treatment decisions for their pancreatic cancer, it can provide an opportunity to identify mutation-carrying relatives who might benefit from screening and other cancer prevention strategies.3 There is also a risk of harm from providing these gene tests so their value needs further study.

A blood test for pancreatic screening has a lot of appeal, but such tests aim to detect pancreatic cancer rather than pancreatic cancer precursors, and in order for it to be effective, such a test has to be able to detect potentially curable (stage I) cancers. Because so few patients (currently less than 1 percent of patients) present with stage I disease, biomarker diagnostic accuracy is poorly evaluated in this setting. Instead, circulating biomarkers are generally evaluated in patients with higher tumor burden where they appear to have better diagnostic performance. For now, annual EUS and MRI/MRCP remain the standard tests for pancreatic screening.

Patients enrolled into a pancreatic screening program need regular long-term surveillance. Although most high-risk individuals are willing to undergo this surveillance, there are still many unanswered questions about the benefits and risks of pancreatic screening in high-risk populations. For this reason, these patients should continue to be enrolled in research studies that evaluate the long-term risks and benefits of pancreatic screening.

Dr. Goggins has a licensing agreement with Myriad Genetics for discovery of PALB2 as a pancreatic cancer susceptibility gene.

1. Vasen, H., Ibrahim, I., Ponce, C.G. et al, Benefit of Surveillance for Pancreatic Cancer in High-Risk Individuals: Outcome of Long-Term Prospective Follow-Up Studies From Three European Expert Centers. J Clin Oncol. 2016; 34:2010-9.
2. Canto, M.I., Harinck, F., Hruban, R.H. et al, International Consensus Recommendations on the Management of Patients with Increased Risk for Familial Pancreatic Cancer (The Cancer of the Pancreas Screening (CAPS) Consortium Summit). Gut. 2013; 62:339-47.
3. Shindo, K., Yu, J., Shah, S. et al, Germline mutations in pancreatic cancer susceptibilty genes in patients with apparently sporadic cancer. J Clin Oncol. 2017; in press.

Join the discussion

Your email address will not be published. Required fields are marked *