Who this is for
No inherited syndrome and no family pattern of pancreatic cancer: no screening, by any method, on both bodies' positions. Common risk factors do not change that. USPSTF: "Other factors such as new-onset diabetes, preexisting diabetes, older age, cigarette smoking, obesity, or a history of chronic pancreatitis increase risk to a lesser degree. The USPSTF considers asymptomatic persons who have these other risk factors part of the general population, and they are included in this recommendation."
An inherited syndrome or familial pancreatic cancer: USPSTF's recommendation does not apply. In its words, "this recommendation does not apply to persons at high risk of pancreatic cancer due to an inherited genetic syndrome (eg, Peutz-Jeghers syndrome, hereditary pancreatitis) or due to a history of familial pancreatic cancer." It defines familial pancreatic cancer as "a kindred with at least 2 affected first-degree relatives". USPSTF does not recommend for or against screening these groups; it calls them outside its scope. AGA (2020) does address them, and advises that screening be considered for two groups:
- first-degree relatives of a person with pancreas cancer, in a family with at least two affected genetically related relatives (best practice advice 1);
- everyone with Peutz-Jeghers syndrome, hereditary pancreatitis or a CDKN2A mutation, and people with Lynch syndrome or a BRCA1, BRCA2, PALB2 or ATM mutation who have at least one first-degree relative with pancreas cancer (best practice advice 2).
AGA also advises considering genetic testing and counseling for relatives who are eligible for surveillance, because a germline mutation is associated with a higher risk of progression and can lead to screening for other cancers (best practice advice 3).
Symptoms: a person with symptoms is outside the screening question. USPSTF describes its recommendations as being "for patients without obvious related signs or symptoms", and neither it nor AGA lists symptoms. The American Cancer Society's patient guidance (revised February 5, 2024) does. It says early pancreatic cancers often cause no signs or symptoms, and names jaundice (yellow eyes and skin) with dark urine, light-colored or greasy stools and itchy skin; unintended weight loss and poor appetite; belly or back pain; nausea and vomiting, often worse after eating; a blood clot in a large vein, often in the leg; diabetes; and fatigue or weakness. ACS notes that fatigue, jaundice, belly or back pain, and blood clots usually have causes other than pancreatic cancer. Take them to a clinician for diagnosis.
Where the recommendations diverge
- USPSTF Recommends against
- AGA Advises against
- AGA 50 and older
- AGA Ages vary
The USPSTF recommends against screening for pancreatic cancer in asymptomatic adults.
Our note: Reaffirmation of the 2004 recommendation, August 6, 2019; no ages are stated. Out of scope: people at high risk "due to an inherited genetic syndrome (eg, Peutz-Jeghers syndrome, hereditary pancreatitis) or due to a history of familial pancreatic cancer". USPSTF makes no recommendation for or against screening them. Diabetes, older age, smoking, obesity and chronic pancreatitis keep a person inside this recommendation. Not on USPSTF's in-progress list when checked October 9, 2026.
Clinicians should not screen average-risk individuals for pancreas cancer.
Our note: Best practice advice 5 of 13 in AGA's Clinical Practice Update, an expert review published on AGA's site May 13, 2020 (Gastroenterology epub May 19, 2020). AGA attaches no grade or evidence rating, and the sentence gives no reason. AGA gives no separate definition of average risk. The wording matches the PubMed abstract; the publisher's full text could not be reached, so nothing here comes from it. Not marked retired in AGA's guidance library when checked October 9, 2026.
Pancreas cancer screening should be considered in patients determined to be at high risk
Our note: Best practice advice 1, an excerpt; the sentence goes on to name this row's group. The two affected relatives must be genetically related, not necessarily first-degree relatives of each other. Start (advice 6): 50, or 10 years younger than the initial age of onset in the family; the age field holds the 50, since the second rule is not one number. Tests (advice 7): MRI and EUS together. Interval (advice 9) and stopping rule (advice 12, no age) are on this page. No grade.
Pancreas cancer screening should be considered in patients with genetic syndromes
Our note: Best practice advice 2, an excerpt; the sentence goes on to name the syndromes and genes in this row. Start ages (advice 6) differ, so no single age is entered: 50, or 10 years younger than the initial age of onset in the family; 35 in Peutz-Jeghers syndrome. Advice 6 also names 40 in a sentence on CDKN2A and PRSS1 carriers, readable as all CDKN2A carriers plus PRSS1 carriers with hereditary pancreatitis, or only people with hereditary pancreatitis carrying either mutation. AGA does not say which. Tests: MRI and EUS together. No grade.
Why they differ
On the reader at average risk, the two bodies agree. USPSTF (2019, grade D) recommends against screening asymptomatic adults by any method. AGA (2020) says it in best practice advice 5: "Clinicians should not screen average-risk individuals for pancreas cancer." USPSTF's 2019 survey of other groups found that "No organization currently recommends screening for pancreatic cancer in the general population of asymptomatic adults."
They part on the high-risk reader, and the split is silence against advice. USPSTF makes no recommendation for people with an inherited syndrome or familial pancreatic cancer. Answering public comments that asked for one, it wrote that "the USPSTF wants to clarify that these groups are outside the scope of this recommendation and that this recommendation does not apply to these persons", and it calls screening in these groups "an active area of research". AGA gives the same groups start ages, a preferred pair of tests, an interval and a stopping rule. Its advice carries no grade, and its abstract describes the basis as reports of screening studies in high-risk individuals plus expert opinion.
The documents also draw the high-risk line in different places, so whether a family qualifies can depend on which one a clinician follows. USPSTF defines familial pancreatic cancer as a kindred with at least 2 affected first-degree relatives. AGA asks for at least two affected relatives who are genetically related, without requiring them to be first-degree relatives of each other, and adds the gene carriers listed above. The American College of Gastroenterology's 2015 hereditary syndromes guideline, still listed on ACG's index with no update marker, draws it a third way: it admits mutation carriers with Peutz-Jeghers syndrome, hereditary pancreatitis or familial atypical multiple melanoma and mole syndrome (FAMMM), members of familial kindreds who have an affected first-degree relative, and BRCA1, BRCA2, PALB2, ATM and Lynch carriers with an affected first- or second-degree relative. It says surveillance should ideally be "performed in experienced centers utilizing a multidisciplinary approach and under research conditions", and grades that advice conditional, on very low quality evidence.
One more difference a reader can trip on is new-onset diabetes. USPSTF keeps it, in a person without symptoms, inside the recommendation against screening. AGA's best practice advice 9 says new-onset diabetes in a person already at high risk should lead to more diagnostic studies or a change in the surveillance interval.
5 documents set the numbers on this page. The oldest was published in 2015, the newest in 2024.
- USPSTF recommendation
- American Gastroenterological Association recommendation
- ACG clinical guideline: Genetic testing and management of hereditary gastrointestinal cancer syndromes 2015
- Can Pancreatic Cancer Be Found Early? (last revised February 5, 2024) 2024
- Signs and Symptoms of Pancreatic Cancer (last revised February 5, 2024) 2024
We last re-read these sources on . How we verify.
What the test involves
For a person at average risk there is nothing to schedule. For a person AGA would consider for surveillance, the program is imaging. AGA's best practice advice 7 names MRI and endoscopic ultrasonography (EUS), used in combination, as the preferred tests. The schedule, from AGA's best practice advice 6, 9 and 12:
- Start: at 50, or 10 years younger than the initial age of onset in the family. AGA sets 35 for Peutz-Jeghers syndrome. Its advice 6 also names age 40 in a sentence about CDKN2A and PRSS1 mutation carriers, and that sentence can be read two ways: as all CDKN2A carriers plus PRSS1 carriers with hereditary pancreatitis, or as only people with hereditary pancreatitis who carry either mutation. AGA does not say which.
- Interval: every 12 months when no concerning lesion is found. For lesions a multidisciplinary team judges low risk, shorter intervals and/or EUS in 6 to 12 months; EUS within 3 to 6 months for indeterminate lesions, and within 3 months for high-risk lesions if surgery is not planned.
- Stop: no stopping age. AGA advises considering stopping when other illness makes death from another cause more likely than death from pancreas cancer, or when the person is not a candidate for pancreas surgery.
AGA wants the work done inside a structure: a registry or a pancreas Center of Excellence when possible (advice 4), decisions on abnormal findings made by a dedicated multidisciplinary team together with the person and family (advice 10), and surgery at high-volume centers (advice 11).
If the result is abnormal
For a person at average risk, a positive screen is where the harms USPSTF weighed begin. It bounds the harms of screening and of treating screen-detected cancer "as at least moderate, based on potential harms from false-positive results and the harms of treatment", and warns that a test used where the disease is uncommon "would potentially have a lower positive predictive value and a higher rate of false-positive results." It reports that the international consensus criteria for intraductal papillary mucinous neoplasms "had high sensitivity (98.4%) but low specificity (14.8%) to predict high-grade dysplasia or invasive cancer", and that these data "suggest the possibility that screening in the general population might lead to overdiagnosis and overtreatment."
A finding can lead to surgery. USPSTF: "Pancreatectomy carries a significant risk of morbidity and mortality". It cites a German study of 428 patients undergoing pancreatectomy with a 33.6% rate of any complication, and a US study of 21,482 pancreatectomies performed between 2007 and 2010 with a 3.7% 30-day mortality rate. The diagnostic step has its own cost: "Of 150 persons in 2 studies who underwent endoscopic retrograde cholangiopancreatography as a diagnostic test, 15 (10.0%) developed acute pancreatitis, 9 of whom required hospitalization." Against those harms, USPSTF found no evidence that screening, or treatment of screen-detected pancreatic cancer, improves disease-specific or all-cause mortality.
USPSTF's summary of the high-risk screening studies shows what surveillance finds: "Among high-risk participants in all studies, a total of 18 cases of pancreatic cancer were found across all rounds of screening, for a yield of 15.6 cases per 1000 persons." Of the 57 patients who went to surgery in those studies, 14 had pancreatic cancer, 38 had precursor lesions, and 5 had neuroendocrine tumors, liver hyperplasia or a benign serous cystadenoma. AGA's target is resectable stage I pancreatic ductal adenocarcinoma and high-risk precursor lesions (advice 8), and its advice 13 is to discuss the limitations and potential risks of screening before a program starts.
More on this screening
Frequently asked questions
No one in my family has had pancreatic cancer. Should I be screened?
No. USPSTF (2019) recommends against screening adults without symptoms, by any method, and AGA (2020) advises clinicians not to screen people at average risk. USPSTF's reasons: it found no evidence that screening or treating screen-detected cancer lowers deaths, and it rates the harms, from false-positive results and from treatment, as at least moderate.
I have diabetes, I smoke, or I have chronic pancreatitis. Does that change the answer?
Not for screening. USPSTF names new-onset diabetes, preexisting diabetes, older age, cigarette smoking, obesity and chronic pancreatitis as factors that raise risk to a lesser degree, and keeps people with them inside its recommendation against screening. AGA's advice on new-onset diabetes applies to people already in high-risk surveillance, for whom it should prompt more diagnostic studies or a change in interval.
My parent had pancreatic cancer. Am I high risk?
One affected relative on its own does not meet AGA's familial criterion, which asks for a family with at least two affected genetically related relatives, or USPSTF's definition of familial pancreatic cancer, at least 2 affected first-degree relatives. If a second blood relative also had it, ask your clinician whether your family meets AGA's criterion. A single affected parent counts in combination with a gene: AGA advises considering screening for people with Lynch syndrome or a BRCA1, BRCA2, PALB2 or ATM mutation who have a first-degree relative with pancreas cancer. Whether you carry one is a genetics question, and ACS strongly recommends talking with a genetic counselor, nurse or qualified doctor before genetic testing.
Is there a blood test or scan that finds pancreatic cancer early?
Not one shown to work as a screen. USPSTF: "There currently are no accurate, validated biomarkers for early detection of pancreatic cancer." It also found no evidence on the accuracy of CT, MRI or EUS as screening tests. MRI and EUS are the tests AGA prefers for surveillance of high-risk people, and ACS writes of them: "These tests are not used to screen the general public".
What symptoms should make me see a doctor?
USPSTF and AGA do not list symptoms; their documents are about people without them. The American Cancer Society's list includes yellowing of the eyes and skin, dark urine, light-colored or greasy stools, itchy skin, unintended weight loss, poor appetite, belly or back pain, nausea and vomiting, a blood clot in a large vein, diabetes, and fatigue or weakness. ACS notes that fatigue, jaundice, belly or back pain, and blood clots usually have causes other than pancreatic cancer, and that early pancreatic cancers often cause no symptoms at all. Take any of them to a clinician for diagnosis.
I carry a BRCA2 mutation. What would surveillance look like?
Under AGA's 2020 advice, a BRCA1 or BRCA2 carrier is in the group to consider for surveillance when at least one first-degree relative had pancreas cancer. If that applies, AGA's plan is MRI and EUS together, every 12 months when nothing concerning is found, starting at 50 or 10 years younger than the initial age of onset in the family, through a registry or pancreas Center of Excellence when possible. ACG's 2015 guideline admits BRCA carriers with an affected first- or second-degree relative and prefers surveillance under research conditions. USPSTF makes no recommendation either way.