Purpose: Pancreatic cancer is a virtually uniformly fatal disease. We aimed to determine if screening to identify curable neoplasms is effective when offered to patients at high risk.

Experimental Design: Patients at high risk of pancreatic cancer were prospectively enrolled into a screening program. Endoscopic ultrasound (EUS), magnetic resonance imaging (MRI), and genetic testing were offered by a multidisciplinary team according to each patient's risk.

Results: Fifty-one patients in 43 families were enrolled, with mean age of 52 years, 35% of whom were male. Of these patients, 31 underwent EUS and 33 MRI. EUS revealed two patients with pancreatic cancer (one resectable, one metastatic), five with intraductal papillary mucinous neoplasms (IPMN), seven with cysts, and six with parenchymal changes. Five had pancreatic surgery (one total pancreatectomy for pancreatic cancer, three distal and one central pancreatectomy for pancreatic intraepithelial neoplasia 2 and IPMN). A total of 24 (47%) had genetic testing (19 for BRCA1/2 mutations, 4 for CDKN2A, 1 for MLH1/MSH2) and 7 were positive for BRCA1/2 mutations. Four extrapancreatic neoplasms were found: two ovarian cancers on prophylactic total abdominal hysterectomy and bilateral salpingo-oophorectomy, one carcinoid, and one papillary thyroid carcinoma. Overall, 6 (12%) of the 51 patients had neoplastic lesions in the pancreas and 9 (18%) had neoplasms in any location. All were on the initial round of screening. All patients remain alive and without complications of screening.

Conclusions: Pancreatic cancer screening for high-risk patients with a comprehensive strategy of imaging and genetics is effective and identifies curable neoplasms that can be resected. Ongoing study will better define who will benefit from screening and what screening strategy will be the most effective. Clin Cancer Res; 16(20); 5028–37. ©2010 AACR.

Translational Relevance

This article shows the efficacy of cancer screening in individuals assessed to be at high risk for pancreas cancer. Patients underwent genetic testing, imaging, and pancreas lesion resection as appropriate. This comprehensive strategy of imaging and genetic evaluation resulted in the identification of preneoplastic pancreas lesions, as well as pancreas and extrapancreatic cancers that could be resected. The study is an incremental contribution to the growing and important field of pancreas cancer risk evaluation and screening. Ongoing research will better define genetic and environmental risk factors for pancreas cancer, and improve risk stratification and identification of at-risk individuals. It will also allow us to better understand which testing modalities are most useful, and at what frequency.

Pancreatic adenocarcinoma is the fourth leading cause of cancer death in most western countries (1). In 2009, there were approximately 42,470 new cases of pancreatic cancer diagnosed in the United States, and 35,240 cancer-related deaths (2). Due to the rapid progression and almost uniform fatality of the disease, early detection through screening will be essential to improve outcomes. As premalignant stages of disease, including pancreatic intraepithelial neoplasia (PanIN; refs. 3, 4) and intraductal papillary mucinous neoplasms (IPMN; refs. 57), have been identified, and the sensitivity of pancreatic imaging has improved with endoscopic ultrasound (EUS) and high-resolution magnetic resonance imaging (MRI), early detection of small curable pancreatic cancers and premalignant lesions now seems possible. It has been shown that early-stage pancreatic cancer may be resected for cure, as evidenced by a series from Japan where 100% of patients were cured when the pancreatic cancer was <1 cm (8). Unfortunately, no current screening strategy is adequately safe, sensitive, and cost effective to be implemented in the general population, even in those with significant risk factors such as tobacco exposure or advanced age.

Pancreatic cancer screening, however, is under investigation in individuals with the highest risk of pancreatic cancer due to a hereditary predisposition (914). About 3% to 16% of pancreatic cancer is thought to be either syndromic or familial (1522), due to multiorgan cancer syndromes, other genetically driven chronic diseases, or familial groupings of pancreatic cancer with yet unidentified genetic abnormalities (familial pancreatic cancer; ref. 23). The cancer syndromes known to increase the risk of pancreatic cancer include Peutz-Jeghers syndrome (PJS; ∼132- fold increased risk; ref. 24), familial atypical multiple mole melanoma (FAMMM; ∼9- to 22-fold increased risk; refs. 2528), hereditary nonpolyposis colorectal cancer (HNPCC; up to 8.6-fold increased risk; refs. 2931), and familial breast-ovarian cancer (BRCA1 mutations with a ∼2-fold increased risk and BRCA2 mutations with ∼5% lifetime risk; refs. 3236). Other high-risk genetic diseases that increase the risk of pancreatic cancer include most prominently hereditary pancreatitis, with a 40% cumulative lifetime risk of pancreatic cancer (37, 38). Finally, familial pancreatic cancer includes groups of patients with a strong family history of pancreatic cancer but without an identified genetic syndrome. Although the definition of familial pancreatic cancer is debated, it is generally defined as at least two first-degree relatives with pancreatic cancer, without meeting criteria for one of the above syndromes. The National Familial Pancreas Tumor Registry at Johns Hopkins University showed a 4.6-, 6.4-, and 32-fold increased risk of pancreatic cancer in individuals with 1, 2, and ≥3 affected first-degree relatives, respectively (although the increased risk in those with one first-degree relative was not statistically significant; ref. 39). Careful study of these patients has led to the discovery of candidate pancreatic cancer susceptibility genes, for example the partner and localizer of BRCA2 (PALB2) gene that encodes a protein that colocalizes with and stabilizes the BRCA2 protein (4042). The Palladin gene that encodes a cytoskeletal protein and is prevalent in one well-characterized high-risk family was also thought to be a pancreatic cancer susceptibility gene (43, 44). It has recently been found to be overexpressed in tumor-associated fibroblasts and may play a role in pancreatic cancer tumor microenvironment and metastatic potential (45). Palladin may therefore be useful as a biomarker for early pancreatic cancer.

In general, screening programs involve multidisciplinary teams of specialists and a combination of imaging techniques, usually with a focus on EUS (914, 46, 47). However, data are preliminary and there is no widely accepted screening protocol. Consensus practice recommendations (9), based largely upon expert opinion, suggest a threshold of a >10-fold increased risk of developing pancreatic cancer to select individuals who may benefit from screening, but this threshold is not validated and excludes patients at significant risk (5- to 10-fold increased). Outcomes and cost-effectiveness analyses are also not currently possible given the lack of uniform screening practices. Although genetic counseling and testing have been recommended as part of the care of select patients (9), there is little information published on the impact of genetic testing in these patients.

We therefore developed a high-risk pancreatic cancer screening program at our multidisciplinary pancreas center starting in 2005 and enrolled all patients with a family history of pancreatic cancer who were interested in screening. All patients were risk-stratified into high-, moderate-, and average-risk groups based on family history, and screened with imaging and genetic testing when appropriate. We herein report our initial experience.

Patients and protocol

All patients who were referred to the Pancreas Cancer Prevention and Genetics Program at Columbia University Medical Center/NewYork Presbyterian Hospital with a family history of pancreatic cancer and interest in their risk of disease were offered enrollment in this prospective cohort. The protocol included a detailed history and physical exam, collection of family history by a genetic counselor, a personal and family health history questionnaire completed by the patient, results of all imaging and blood testing, and storage of frozen serum and any surgically resected tissue. Blood was processed for DNA extraction and EBV immortalization. This protocol was approved by the Columbia University Medical Center Institutional Review Board and all participants provided informed consent.

Screening

This study focuses on the yield of initial screening exams for the patients in this cohort, and does not include the results of surveillance exams. Upon presentation, patients were assessed for their level of risk based upon the number of family members affected and age of onset, and whether the family history was suggestive of a known genetic cancer syndrome (Fig. 1). Patients were classified as average risk if they had only one family member affected at an age >55 years. These patients were generally not recommended for screening, but some patients were tested if significant psychological stress led them to prefer to be screened. Patients were classified as moderate risk if they had more than one family member with pancreatic cancer or one first-degree family member with the onset of disease at <55 years of age, but did not meet the criteria for high-risk classification. Depending on their age relative to the youngest affected family member, these patients were offered MRI or EUS of the pancreas as well as an oral glucose tolerance test (OGTT) and serum cancer antigen 19-9 (CA 19-9). The high-risk patients were classified based upon a family history and/or genetic testing consistent with one of the genetic cancer syndromes (PJS, HNPCC, BRCA1/2, FAMMM, hereditary pancreatitis) or who met our definition of familial pancreatic cancer (any three affected relatives, two first-degree relatives with pancreatic cancer, or one first-degree and at least one second-degree relative, one with onset at age <55). BRCA1/2 mutation carriers were placed in this group if they had at least one first- or second-degree relative with pancreatic cancer. These patients were offered EUS and MRI, as well as the OGTT and CA 19-9 testing.

Fig. 1.

Pancreatic cancer screening general algorithm used in this cohort. PC, pancreatic cancer.

Fig. 1.

Pancreatic cancer screening general algorithm used in this cohort. PC, pancreatic cancer.

Close modal

Genetic testing

Genetic testing was recommended at the discretion of the clinician and genetic counselor. The approach was to first test the youngest affected family member available, or the proband if the proband had a personal history of cancer. An exception to this approach was made in the case of suspected BRCA1/2 mutations in patients of Ashkenazi Jewish descent. In this group, patients interested in BRCA1/2 testing were offered multisite BRCA testing, even if they themselves did not have a personal history of cancer. Genetic testing for BRCA (BRCAnalysis or Multisite 3 BRCAnalysis), HNPCC (Comprehensive Colaris), and CDKN2A (Comprehensive Melaris) were commercially done through Myriad Genetic Laboratories (Salt Lake City, UT). Genetic counseling before and after testing was provided by a genetic counselor or clinician with specific expertise in pancreatic cancer genetics, and informed consent was obtained. Patients who tested positive for a germline mutation underwent multiorgan cancer screening as appropriate. In the case of documented BRCA1/2 mutations, prophylactic mastectomy and oophorectomy, chemoprevention, and enhanced screening were discussed.

Pancreatic imaging

EUS.

Standard upper endoscopy was done, followed by EUS imaging with a curvilinear echoendoscope and processor (GFUC140P and SSD-Alpha 5, Olympus Corporation). Abnormalities of interest included mass lesions, IPMNs, cysts, and chronic pancreatitis-like parenchymal changes. Fine needle aspiration (FNA) under EUS guidance was done with a linear array echoendoscope when mass lesions, cysts, or suspicious lymph nodes were encountered. In general, FNA specimens were evaluated for abnormal cells by an onsite cytopathologist, and the fluid removed from cyst aspiration was sent for carcinoembryonic antigen (CEA) and amylase. Endoscopic retrograde cholangiopancreatography (ERCP) was done at the discretion of the interventional endoscopist, generally when ductal irregularities or changes consistent with IPMN required additional assessment.

MRI and magnetic resonance cholangiopancreatography.

All MRI images were obtained at 1.5 Tesla (GE HD EXCITE 14.0 m4) using a body coil for signal transmission and an 8-channel body array coil for reception. Abdominal organs were localized on T2-weighted single-shot fast spin echo images obtained in three cardinal planes. The pancreatic duct was imaged on a 40-mm-thick slab single-shot fast spin echo sequence positioned to cover the pancreatic duct in the coronal-oblique plane using an echo time = 900 ms and also on a three-dimensional magnetic resonance cholangiopancreatography (MRCP) including the entire pancreas and duodenum. Detailed images of pancreatic architecture, liver, and abdominal vasculature were obtained using dynamic three-dimensional liver accelerated volume acquisition (LAVA) before, during, and following injection of 20 mL of gadodiamide or gadobenate dimeglumine contrast in individuals with glomerular filtration rate >30 mL/minute as estimated from serum creatinine. Images were analyzed by a radiologist experienced in pancreatic imaging, but blinded to the patient's pancreatic cancer risk factors.

Follow-up

After completion of the first series of genetic, radiologic, and blood testing, patients were discussed at a weekly multidisciplinary Pancreas Center conference attended by surgeons, gastroenterologists, oncologists, radiologists, and geneticists. All patients with any abnormal test who had not undergone EUS as part of the initial screening were recommended for EUS. The possibility of surgical intervention was discussed by the team for patients with abnormal EUS exams for neoplastic or preneoplastic lesions in patients with high or moderate risk of pancreatic cancer. Findings such as resectable mass lesions, a high suspicion for main-duct IPMN (IPMN-M), or abnormal cytology on FNA were among those considered for surgery. Patients who did not undergo surgery were again risk-stratified based upon history and testing, and entered surveillance. Patients thought to be at highest risk due to family history or imaging findings and who were within 10 years of the youngest onset of pancreatic cancer in their family, as well as patients who underwent partial pancreatectomies, were then screened every six months with alternating MRI and EUS. Patients at moderate risk underwent annual imaging, and those at average risk returned for annual visits and further testing if they developed symptoms or new onset diabetes mellitus. The results of surveillance exams are not included in this article.

Patients

Fifty-one patients from 43 unique families were enrolled and underwent initial screening from 2005 to 2008. The mean age at presentation was 52 years (range, 29-77), 35% were male, 35% had some previous tobacco exposure, and 6% reported previous heavy alcohol use (Table 1). The majority (90%) were non-Hispanic white and 25 (49%) patients were of Ashkenazi Jewish ancestry. Fifteen patients (29%) had a personal history of cancer other than pancreatic cancer (Table 1), most commonly breast cancer (14% of the cohort).

Table 1.

Patient characteristics

Number of patients 51 
Number of unique families 43 
Mean age (SD), years 52 (12.3) 
Male gender (%) 18 (35) 
Race and ethnicity (%) 
    Non-Hispanic white 46 (90) 
    Hispanic white 3 (6) 
    Black 1 (2) 
    Native American 1 (2) 
Ashkenazi Jewish ancestry 25 (49) 
Ever smoked (%) 17 (35) 
Alcohol use (%) 
    Previous heavy use 3 (6) 
    Social or occasional use 39 (76) 
    No use 9 (18) 
Type 2 diabetes mellitus (%) 2 (4) 
Number of patients with a personal history of nonpancreatic cancer (%) 15 (29) 
    Breast cancer 
    Malignant melanoma 
    Cervical cancer 
    Endometrial cancer 
    Lung cancer 
    Squamous cell carcinoma of the face 
    Lymphoma 
Number of patients 51 
Number of unique families 43 
Mean age (SD), years 52 (12.3) 
Male gender (%) 18 (35) 
Race and ethnicity (%) 
    Non-Hispanic white 46 (90) 
    Hispanic white 3 (6) 
    Black 1 (2) 
    Native American 1 (2) 
Ashkenazi Jewish ancestry 25 (49) 
Ever smoked (%) 17 (35) 
Alcohol use (%) 
    Previous heavy use 3 (6) 
    Social or occasional use 39 (76) 
    No use 9 (18) 
Type 2 diabetes mellitus (%) 2 (4) 
Number of patients with a personal history of nonpancreatic cancer (%) 15 (29) 
    Breast cancer 
    Malignant melanoma 
    Cervical cancer 
    Endometrial cancer 
    Lung cancer 
    Squamous cell carcinoma of the face 
    Lymphoma 

Family history and risk stratification

Fifteen (29%) patients had at least two first-degree relatives with pancreatic cancer and 35 (69%) had at least two first-, second-, or third-degree relatives with pancreatic cancer (Table 2). Thirty-one (61%) patients had two or more first-, second-, or third-degree relatives with a nonpancreatic cancer. Breast cancer was the most common extrapancreatic cancer (35% of patients), followed by prostate cancer, lung cancer, and melanoma (Table 2).

Table 2.

Family cancer risk profile for the patients in our cohort

Median youngest age (range) of PC onset in family, years 61 (30-82) 
Number of patients with: 
    ≥3 first-degree relatives with PC 
    2 first-degree relatives with PC 12 
    1 first-degree relative with PC 29 
    0 first-degree relatives with PC 
Number of patients with: 
    ≥3 first-, second-, and third-degree relatives with PC 13 
    2 first-, second-, and third-degree relatives with PC 22 
    1 first-, second-, and third-degree relative with PC 16 
Number of patients with: 
    ≥5 first-, second-, and third-degree relatives with non-PC 
    4 first-, second-, and third-degree relatives with non-PC 
    3 first-, second-, and third-degree relatives with non-PC 
    2 first-, second-, and third-degree relatives with non-PC 11 
    1 first-, second-, and third-degree relative with non-PC 16 
    0 first-, second-, and third-degree relatives with non-PC 
Number of patients with first- or second-degree relatives with: 
    Breast cancer 18 
    Prostate cancer 10 
    Lung cancer 
    Malignant melanoma 
    Ovarian cancer 
    Endometrial cancer 
    Colon cancer 
    Gastric cancer 
    Head and neck cancers 
    Glioblastoma 
    Hepatocellular carcinoma 
    Cervical cancer 
Number of patients with family history suggestive of:* 
    Familial pancreatic cancer 34 
    BRCA1/2 17 
    HNPCC 
    FAMMM 
Risk category, number (%) 
    High risk 32 (63) 
    Moderate risk 14 (27) 
    Average risk 5 (10) 
Median youngest age (range) of PC onset in family, years 61 (30-82) 
Number of patients with: 
    ≥3 first-degree relatives with PC 
    2 first-degree relatives with PC 12 
    1 first-degree relative with PC 29 
    0 first-degree relatives with PC 
Number of patients with: 
    ≥3 first-, second-, and third-degree relatives with PC 13 
    2 first-, second-, and third-degree relatives with PC 22 
    1 first-, second-, and third-degree relative with PC 16 
Number of patients with: 
    ≥5 first-, second-, and third-degree relatives with non-PC 
    4 first-, second-, and third-degree relatives with non-PC 
    3 first-, second-, and third-degree relatives with non-PC 
    2 first-, second-, and third-degree relatives with non-PC 11 
    1 first-, second-, and third-degree relative with non-PC 16 
    0 first-, second-, and third-degree relatives with non-PC 
Number of patients with first- or second-degree relatives with: 
    Breast cancer 18 
    Prostate cancer 10 
    Lung cancer 
    Malignant melanoma 
    Ovarian cancer 
    Endometrial cancer 
    Colon cancer 
    Gastric cancer 
    Head and neck cancers 
    Glioblastoma 
    Hepatocellular carcinoma 
    Cervical cancer 
Number of patients with family history suggestive of:* 
    Familial pancreatic cancer 34 
    BRCA1/2 17 
    HNPCC 
    FAMMM 
Risk category, number (%) 
    High risk 32 (63) 
    Moderate risk 14 (27) 
    Average risk 5 (10) 

Abbreviation: PC, pancreatic cancer.

*Some patients had family histories suggestive of more than one syndrome.

All patients evaluated were risk-stratified into average, moderate, or high risk of pancreatic cancer, based on the criteria developed for this cohort (Fig. 1). Thirty-two (63%) patients were classified as high risk, 14 (27%) as moderate risk, and 5 (10%) as average risk (Table 2).

Genetic testing

Twenty-one patients (41%) consented to genetic testing (Table 3): 16 for BRCA1/2 mutations, 1 for CDKN2A mutations (for FAMMM syndrome), 3 for both BRCA1/2 and CDKN2A mutations, and 1 for MLH1/MSH2 (for HNPCC, the patient opted not to have testing for MSH6). Of those patients tested, seven were positive for BRCA1/2 mutations (one in comprehensive testing and six with multisite testing).

Table 3.

Genetic testing done

Suspected cancer syndromeGene testedPositive tests (%)*
BRCA1/2 BRCA1/2 7/19 (37) 
FAMMM CDKN2A 0/4 (0) 
HNPCC MLH1/MSH2 0/1 (0) 
Suspected cancer syndromeGene testedPositive tests (%)*
BRCA1/2 BRCA1/2 7/19 (37) 
FAMMM CDKN2A 0/4 (0) 
HNPCC MLH1/MSH2 0/1 (0) 

*Three patients were tested for both BRCA1/2 and CDKN2A.

Of the seven patients with BRCA1/2 mutations, one had resectable pancreatic cancer on initial screening. Two patients with BRCA1/2 mutations and personal histories of breast cancer underwent prophylactic total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAH-BSO). The pathology of both of these patients was abnormal, one with ovarian papillary serous adenocarcinoma (BRCA1) and one with tubal intraepithelial carcinoma (BRCA2).

Pancreatic cancer screening

Of the 51 patients in the protocol, 31 (61%) underwent EUS and 33 (65%) underwent MRI-MRCP (Table 4). Of the 10 patients who had neither EUS nor MRI, two were in the average-risk category and were not tested. Of the three high-risk patients who did not have testing, two were very young at enrollment (>10 years younger than their youngest affected family member) and one is currently in therapy for breast and ovarian cancers and deferred further screening. Five of the moderate-risk patients did not have any imaging, four due to patient preference and one due to young age compared with affected family members.

Table 4.

Diagnostic findings in the pancreas for those patients who underwent imaging

Overall (n = 51)Average risk (n = 5)Moderate risk (n = 14)High risk (n = 32)
EUS 31 (61%) 2 (40%) 5 (36%) 24 (75%) 
    Normal* 
    Changes of chronic pancreatitis 
        ≥3 of the 9 findings 
        ≥5 of the 9 findings 
    IPMN-B 
    Other cysts 
    Solid mass lesion 
FNA 6 (18%) 0 (0%) 3 (21%) 3 (9%) 
    Of mass or parenchyma 
    Of cysts or IPMN lesions 
ERCP 7 (14%) 1 (20%) 1 (7%) 5 (16%) 
    Normal 
    IPMN-B 
    PD irregularities 
MRI/MRCP 33 (65%) 3 (60%) 7 (50%) 23 (72%) 
    Normal 22 16 
    IPMN 
    Other cysts 
    Solid mass lesion 
    Isolated PD irregularities 
No EUS or MRI 10 (20%) 2 (40%) 5 (36%) 3 (9%) 
Pancreas surgery 5 (10%) 0 (0%) 2 (14%) 3 (10%) 
Overall (n = 51)Average risk (n = 5)Moderate risk (n = 14)High risk (n = 32)
EUS 31 (61%) 2 (40%) 5 (36%) 24 (75%) 
    Normal* 
    Changes of chronic pancreatitis 
        ≥3 of the 9 findings 
        ≥5 of the 9 findings 
    IPMN-B 
    Other cysts 
    Solid mass lesion 
FNA 6 (18%) 0 (0%) 3 (21%) 3 (9%) 
    Of mass or parenchyma 
    Of cysts or IPMN lesions 
ERCP 7 (14%) 1 (20%) 1 (7%) 5 (16%) 
    Normal 
    IPMN-B 
    PD irregularities 
MRI/MRCP 33 (65%) 3 (60%) 7 (50%) 23 (72%) 
    Normal 22 16 
    IPMN 
    Other cysts 
    Solid mass lesion 
    Isolated PD irregularities 
No EUS or MRI 10 (20%) 2 (40%) 5 (36%) 3 (9%) 
Pancreas surgery 5 (10%) 0 (0%) 2 (14%) 3 (10%) 

Abbreviation: PD, pancreatic duct.

*Normal examinations included those with no abnormalities as well as those with only age-related changes such as mild atrophy or fatty infiltration.

The changes associated with chronic pancreatitis include hyperechoic foci, hyperechoic strands, lobularity, cysts, hyperechoic duct margins, dilated pancreatic duct, duct irregularity, visible side branches, and stones.

The most common abnormal findings on EUS were parenchymal changes seen in chronic pancreatitis (hyperechoic foci, echogenic strands, lobularity, cysts, hyperechoic duct margins, dilated pancreatic duct, duct irregularity, visible side branches and stones). Of the 31 patients who underwent EUS, 6 (19%) had at least three of these changes, whereas 3 (10%) had at least five of these changes (Table 4). Two patients (both high risk) had mass lesions on EUS, each with adenocarcinoma on FNA (one metastatic to the liver and treated with systemic chemotherapy, the other localized and resected). FNA was also done on one patient with abnormal-appearing parenchyma (nondiagnostic) and three patients with cystic lesions (all nondiagnostic on cytology, one with a CEA of 1,169 units/mL). ERCP was done in seven patients, two of which confirmed changes consistent with branch-duct IPMN (IPMN-B). No main-duct IPMN-M lesions were found and all IPMN-B lesions were <3 cm in greatest diameter. MRI was done in 33 (65%) of the cohort, and was normal in 22 of 33 (67%; Table 4). The two masses seen on EUS were also observed on MRI. No procedural complications were encountered during the screening process.

Pancreatic surgical procedures and pathologic diagnoses

Six patients had pancreas pathology by resection or FNA available for review (Table 5). One high-risk patient had stage IV metastatic adenocarcinoma with three liver lesions at the time of presentation and underwent palliative chemotherapy. This patient had one first-degree relative with pancreatic cancer at the age of 53 and one affected second-degree relative. One high-risk patient was also diagnosed with resectable pancreatic cancer when a mass was seen on EUS, and FNA was diagnostic for adenocarcinoma. This patient underwent a total pancreatectomy, revealing a 2-cm moderately differentiated adenocarcinoma arising from IPMN-M. He had a BRCA2 mutation and two first-degree relatives with pancreatic cancer, the youngest at 46 years of age.

Table 5.

Pathologic diagnoses rendered

LocationAgeRisk categoryIndicationProcedurePathologic diagnosis
Pancreas 58 High Mass in the HOP with liver lesions on EUS/MRI FNA Stage IV PC 
61 High 2 cm mass in HOP on EUS Total pancreatectomy PC with adjacent IPMN and multifocal PanIN2 
47 High IPMN-B, irregular PD on EUS/ERCP Distal pancreatectomy IPMN-B with moderate dysplasia and multifocal PanIN2 
56 High IPMN-B on EUS/ERCP 
40 Moderate IPMN-B on EUS/MRI 
45* Moderate 1 cm cyst BOP with elevated cyst fluid CEA (1,169) on EUS Central pancreatectomy Cyst remnant, IPMN-B with moderate dysplasia, focal PanIN2 
Ovary 52 High Prophylactic, BRCA1/2 positive and personal history of breast cancer TAH-BSO Tubal intraepithelial carcinoma 
60 High   Ovarian papillary serous adenocarcinoma 
Retroperitoneal 43 High Retroperitoneal mass on MRI Resection Carcinoid 
Thyroid 45* Moderate PET thyroid uptake, FNA papillary carcinoma Thyroidectomy Papillary thyroid carcinoma 
LocationAgeRisk categoryIndicationProcedurePathologic diagnosis
Pancreas 58 High Mass in the HOP with liver lesions on EUS/MRI FNA Stage IV PC 
61 High 2 cm mass in HOP on EUS Total pancreatectomy PC with adjacent IPMN and multifocal PanIN2 
47 High IPMN-B, irregular PD on EUS/ERCP Distal pancreatectomy IPMN-B with moderate dysplasia and multifocal PanIN2 
56 High IPMN-B on EUS/ERCP 
40 Moderate IPMN-B on EUS/MRI 
45* Moderate 1 cm cyst BOP with elevated cyst fluid CEA (1,169) on EUS Central pancreatectomy Cyst remnant, IPMN-B with moderate dysplasia, focal PanIN2 
Ovary 52 High Prophylactic, BRCA1/2 positive and personal history of breast cancer TAH-BSO Tubal intraepithelial carcinoma 
60 High   Ovarian papillary serous adenocarcinoma 
Retroperitoneal 43 High Retroperitoneal mass on MRI Resection Carcinoid 
Thyroid 45* Moderate PET thyroid uptake, FNA papillary carcinoma Thyroidectomy Papillary thyroid carcinoma 

Abbreviations: HOP, head of pancreas; BOP, body of pancreas; PET, positron emission tomography.

*One patient had both pancreas and thyroid surgery.

Four other patients had partial pancreatectomies (three distal and one central; Table 5). All four of these had IPMN-B with moderate dysplasia and multifocal PanIN2 lesions on pathology. None had definitive PanIN3 or “carcinoma in situ.” Therefore, a total of 12% of the 51 patients in this cohort (10% of high-risk and 14% of the moderate-risk patients) were found to have malignant or possibly premalignant lesions of the pancreas on pathology. All of these patients are alive and all but the patient with stage IV adenocarcinoma are in surveillance without additional pathology identified.

Nonpancreatic pathology

The testing done for the purposes of this protocol to screen for pancreatic cancer also revealed lesions arising from organs other than the pancreas (Table 5). One patient had a retroperitoneal carcinoid seen on MRI, which was resected. One patient had an elevated CA 19-9 on study enrollment, and a cyst in the pancreas with an elevated CEA on fluid aspiration. This patient underwent central pancreatectomy but no pancreatic cancer was identified and the CA 19-9 remained elevated (Table 5). A positron emission tomography scan then revealed increased uptake in a thyroid nodule, which led to the diagnosis of papillary carcinoma of the thyroid. The patient underwent partial thyroidectomy, which confirmed this pathology. In addition, as mentioned above, two BRCA1/2-positive patients underwent prophylactic TAH-BSO, each revealing early cancers on pathology: one tubal intraepithelial carcinoma (BRCA2) and one ovarian papillary serous adenocarcinoma (BRCA1). Both are without evidence of metastatic disease at follow-up.

Therefore, a total of 10 preneoplastic or neoplastic lesions (pancreatic and extrapancreatic) were discovered in 9 (18%) patients through this pancreatic cancer screening program, each treated with curative intent except for one case of metastatic pancreatic cancer.

We describe the results of initial screening exams in a prospective cohort of patients with a family history of pancreatic cancer. Overall, 12% of our patients were diagnosed with malignant or premalignant disease of the pancreas, and 18% were diagnosed with malignant or premalignant diseases of any organ. All patients who underwent pancreas surgery were in the moderate- or high-risk groups, and all had pathologic changes that correlated with imaging findings. No patient suffered significant complications of the screening process or surgical interventions.

Pancreas cancer has the poorest survival of any common solid malignancy, and presentation with advanced disease due to a lack of early symptoms or screening remains a major impediment to successful treatment. Although pancreatic cancer screening at a population level remains impractical at this time, screening of individuals at high risk of pancreatic cancer has begun at several centers (914, 46). Our series confirms data from these other cohorts that malignant and premalignant lesions can be identified through screening EUS and pancreas surgery may be undertaken with the goal of preventing disease progression.

The major limitation of our protocol is that it remains an observational cohort and not a controlled trial. Therefore, comparisons of imaging findings between these patients and those without known elevated risk of pancreatic cancer are not possible. We believe that it remains unknown what type of patient would represent an adequate control in this setting, and we did not expand enrollment for patients undergoing pancreatic EUS or genetic testing for other reasons. Our protocol also recruited patients more broadly than many others in the literature. For this reason, we also used a cutoff of 55 years at age of onset in affected family members to identify patients at risk. The consensus guidelines were not published at the time of study design, and as little is known about the pancreatic cancer risk in patients with a family history but without a known germline mutation, we were less stringent than other centers in the criteria needed to enter this cohort. Our strategy was to subsequently risk-stratify each participant with the criteria above, most of which are in line with the criteria for low, moderate, and high risk from the consensus conference (9). Our testing recommendations were based upon this risk stratification, with more invasive and frequent testing done in the patients at the highest risk. It is possible, however, that our rate of pancreatic pathology and germline mutations was diluted by our average and moderate risk patients. Although the two adenocarcinomas seen were in the high-risk group, it is interesting to note that a similar percentage of patients in the moderate-risk group went to surgery with preneoplastic lesions.

We stressed a multidisciplinary approach for these patients to ensure that patients and their families were evaluated for patterns of multiorgan cancer syndromes and were educated on testing. Genetic testing was used as an adjunct to imaging, and revealed BRCA1/2 germline mutations in seven patients, two of whom underwent prophylactic salpingo-oopherectomy with early ovarian malignancies. We believe this comprehensive approach is essential to any screening program as these syndromes are associated with cancers outside the pancreas. It is likely that given our predominantly non-Hispanic white population with a high proportion of individuals of Ashkenazi Jewish decent, we detected a higher proportion of BRCA1/2 mutations than will be seen in other populations, possibly limiting the generalizability of this finding.

Our cohort underwent screening with a combination of EUS and MRI imaging. It is unknown whether EUS or MRI is the best initial screening modality, or even whether radial or linear echoendoscopes are superior (47). EUS offers high sensitivity and the possibility of tissue sampling, and is therefore central to pancreatic cancer screening in most centers. The relevance of subtle EUS parenchymal findings (such as those seen in chronic pancreatitis), however, is unknown. Although some authors propose that they can correlate these parenchymal abnormalities with PanIN lesions on pathology (48), these data must be carefully assessed, and no patients in this cohort went to surgery for these findings alone. MRI has not been as widely reported in other cohorts, but permits the identification of lesions outside the purview of EUS without the radiation exposure associated with computed tomography. In this cohort, all solid and cystic lesions that went to surgery were seen on both MRI and EUS. In addition, one retroperitoneal carcinoid was seen on MRI, and given the multiorgan cancer susceptibilities in many of these syndromes, cross-sectional imaging may play an important role.

The significance of each imaging finding and the natural history of many pathologic lesions (such as IPMN and PanIN) are also not known. Therefore, there is an important and significant risk of overdiagnosis and overtreatment in people who undergo pancreatic cancer screening. Indeed, several patients in published cohorts have gone to surgery with benign findings on pathology and none of the patients in our cohort who had partial pancreatectomies had high-grade dysplasia associated with their PanIN or IPMN lesions. In addition, pancreas surgery (unlike polypectomy in colorectal cancer screening) carries significant risk of morbidity and mortality, and at least one patient cited in the literature has died as the result of prophylactic pancreas surgery (47). Therefore, although it is accepted that mass lesions found in patients at high risk should be resected when possible, the optimal management of other lesions is unknown. The natural history of IPMN remains poorly defined. The categorization is evolving, the preoperative diagnosis is often incorrect, and the management continues to be controversial. The incidence of invasive cancer in IPMN lesions is approximately 11% for IPMN-B, compared with 48% for IPMN-M, with a 5-year survival >50% for both (49). Although prognosis and survival can be affected by the number of involved lymph nodes, and familial predisposition, this is also poorly understood. Those with IPMN-M perhaps should be considered for resection given the higher malignant potential, but IPMN-B lesions that were more common in our cohort are generally managed conservatively. In the early experience with these patients, some centers did laparoscopic distal pancreatectomies to further risk-stratify these patients, and in patients with multifocal high-grade dysplasia, total pancreatectomy was considered. As the genetic susceptibility likely confers a risk throughout the entire pancreas, or a "field effect," it was thought that the abnormalities in the distal pancreatectomy specimen likely reflect changes seen throughout the entire gland. This approach is no longer widely used, and in our cohort, none of the patients who underwent distal pancreatectomy had surgery for this reason (all were done to evaluate focal abnormalities). In those patients who go to surgery for focal lesions, it is possible that multifocal PanIN lesions seen on pathology have prognostic implications for the unresected pancreas (50), and all of these patients are in surveillance.

Pancreatic cancer remains an important, deadly disease. Although premalignant lesions such as IPMN and PanIN have been characterized, efficient screening and resection of high-risk lesions is currently not possible. Although screening in the general population is therefore not feasible, it may benefit patients with the highest risk. As there is not yet, and may not be an autosomal dominant gene mutation identified for familial pancreatic cancer, this population should clinically be approached as having a multigenic predisposition. A comprehensive, multidisciplinary approach that combines imaging with genetic risk assessment should be stressed. Given the limited data available, however, screening should occur in experienced centers with prospective cohorts whenever possible.

No potential conflicts of interest were disclosed.

Grant Support: Hirshberg Foundation for Pancreatic Cancer Research grant (H. Frucht).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Jemal
A
,
Siegel
R
,
Ward
E
,
Murray
T
,
Xu
J
,
Thun
MJ
. 
Cancer statistics, 2007
.
CA Cancer J Clin
2007
;
57
:
43
66
.
2
American Cancer Society
.
Cancer facts & figures 2009
.
Atlanta
:
American Cancer Society
; 
2009
.
3
Hruban
RH
,
Adsay
NV
,
Albores-Saavedra
J
, et al
. 
Pancreatic intraepithelial neoplasia: a new nomenclature and classification system for pancreatic duct lesions
.
Am J Surg Pathol
2001
;
25
:
579
86
.
4
Real
FX
. 
A “catastrophic hypothesis” for pancreas cancer progression
.
Gastroenterology
2003
;
124
:
1958
64
.
5
Feldmann
G
,
Beaty
R
,
Hruban
RH
,
Maitra
A
. 
Molecular genetics of pancreatic intraepithelial neoplasia
.
J Hepatobiliary Pancreat Surg
2007
;
14
:
224
32
.
6
McGrath
K
,
Slivka
A
. 
Diagnosis and management of intraductal papillary mucinous neoplasia
.
Nat Clin Pract Gastroenterol Hepatol
2005
;
2
:
316
22
.
7
Tanaka
M
,
Chari
S
,
Adsay
V
, et al
. 
International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas
.
Pancreatology
2006
;
6
:
17
32
.
8
Ariyama
J
,
Suyama
M
,
Satoh
K
,
Sai
J
. 
Imaging of small pancreatic ductal adenocarcinoma
.
Pancreas
1998
;
16
:
396
401
.
9
Brand
RE
,
Lerch
MM
,
Rubinstein
WS
, et al
. 
Advances in counselling and surveillance of patients at risk for pancreatic cancer
.
Gut
2007
;
56
:
1460
9
.
10
Canto
MI
. 
Screening for pancreatic neoplasia in high-risk individuals: who, what, when, how?
Clin Gastroenterol Hepatol
2005
;
3
:
S46
8
.
11
Canto
MI
,
Goggins
M
,
Hruban
RH
, et al
. 
Screening for early pancreatic neoplasia in high-risk individuals: a prospective controlled study
.
Clin Gastroenterol Hepatol
2006
;
4
:
766
81
,
quiz 665
.
12
Canto
MI
,
Goggins
M
,
Yeo
CJ
, et al
. 
Screening for pancreatic neoplasia in high-risk individuals: an EUS-based approach
.
Clin Gastroenterol Hepatol
2004
;
2
:
606
21
.
13
Brentnall
TA
,
Bronner
MP
,
Byrd
DR
,
Haggitt
RC
,
Kimmey
MB
. 
Early diagnosis and treatment of pancreatic dysplasia in patients with a family history of pancreatic cancer
.
Ann Intern Med
1999
;
131
:
247
55
.
14
Brentnall
TA
. 
Management strategies for patients with hereditary pancreatic cancer
.
Curr Treat Options Oncol
2005
;
6
:
437
45
.
15
Klein
AP
,
Hruban
RH
,
Brune
KA
,
Petersen
GM
,
Goggins
M
. 
Familial pancreatic cancer
.
Cancer J
2001
;
7
:
266
73
.
16
Goggins
M
,
Kern
SE
,
Offerhaus
JA
,
Hruban
RH
. 
Progress in cancer genetics: lessons from pancreatic cancer
.
Ann Oncol
1999
;
10
Suppl 4
:
4
8
.
17
Ghadirian
P
,
Boyle
P
,
Simard
A
,
Baillargeon
J
,
Maisonneuve
P
,
Perret
C
. 
Reported family aggregation of pancreatic cancer within a population-based case-control study in the Francophone community in Montreal, Canada
.
Int J Pancreatol
1991
;
10
:
183
96
.
18
Lynch
HT
. 
Genetics and pancreatic cancer
.
Arch Surg
1994
;
129
:
266
8
.
19
Lynch
HT
,
Smyrk
T
,
Kern
SE
, et al
. 
Familial pancreatic cancer: a review
.
Semin Oncol
1996
;
23
:
251
75
.
20
Lynch
HT
,
Fitzsimmons
ML
,
Smyrk
TC
, et al
. 
Familial pancreatic cancer: clinicopathologic study of 18 nuclear families
.
Am J Gastroenterol
1990
;
85
:
54
60
.
21
Bartsch
DK
,
Kress
R
,
Sina-Frey
M
, et al
. 
Prevalence of familial pancreatic cancer in Germany
.
Int J Cancer
2004
;
110
:
902
6
.
22
Hemminki
K
,
Li
X
. 
Familial and second primary pancreatic cancers: a nationwide epidemiologic study from Sweden
.
Int J Cancer
2003
;
103
:
525
30
.
23
Wang
L
,
Brune
KA
,
Visvanathan
K
, et al
. 
Elevated cancer mortality in the relatives of patients with pancreatic cancer
.
Cancer Epidemiol Biomarkers Prev
2009
;
18
:
2829
34
.
24
Giardiello
FM
,
Brensinger
JD
,
Tersmette
AC
, et al
. 
Very high risk of cancer in familial Peutz-Jeghers syndrome
.
Gastroenterology
2000
;
119
:
1447
53
.
25
Goldstein
AM
,
Fraser
MC
,
Struewing
JP
, et al
. 
Increased risk of pancreatic cancer in melanoma-prone kindreds with p16INK4 mutations
.
N Engl J Med
1995
;
333
:
970
4
.
26
Lynch
HT
,
Fusaro
RM
,
Lynch
JF
,
Brand
R
. 
Pancreatic cancer and the FAMMM syndrome
.
Fam Cancer
2008
;
7
:
103
12
.
27
Parker
JF
,
Florell
SR
,
Alexander
A
,
DiSario
JA
,
Shami
PJ
,
Leachman
SA
. 
Pancreatic carcinoma surveillance in patients with familial melanoma
.
Arch Dermatol
2003
;
139
:
1019
25
.
28
Vinarsky
V
,
Fine
RL
,
Assaad
A
, et al
. 
Head and neck squamous cell carcinoma in FAMMM syndrome
.
Head Neck
2009
;
31
:
1524
7
.
29
Lynch
HT
,
Voorhees
GJ
,
Lanspa
SJ
,
McGreevy
PS
,
Lynch
JF
. 
Pancreatic carcinoma and hereditary nonpolyposis colorectal cancer: a family study
.
Br J Cancer
1985
;
52
:
271
3
.
30
Umar
A
,
Boland
CR
,
Terdiman
JP
, et al
. 
Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability
.
J Natl Cancer Inst
2004
;
96
:
261
8
.
31
Kastrinos
F
,
Mukherjee
B
,
Tayob
N
, et al
. 
Risk of pancreatic cancer in families with Lynch syndrome
.
JAMA
2009
;
302
:
1790
5
.
32
Thompson
D
,
Easton
DF
. 
Cancer Incidence in BRCA1 mutation carriers
.
J Natl Cancer Inst
2002
;
94
:
1358
65
.
33
Goggins
M
,
Schutte
M
,
Lu
J
, et al
. 
Germline BRCA2 gene mutations in patients with apparently sporadic pancreatic carcinomas
.
Cancer Res
1996
;
56
:
5360
4
.
34
Hahn
SA
,
Greenhalf
B
,
Ellis
I
, et al
. 
BRCA2 germline mutations in familial pancreatic carcinoma
.
J Natl Cancer Inst
2003
;
95
:
214
21
.
35
Murphy
KM
,
Brune
KA
,
Griffin
C
, et al
. 
Evaluation of candidate genes MAP2K4, MADH4, ACVR1B, BRCA2 in familial pancreatic cancer: deleterious BRCA2 mutations in 17%
.
Cancer Res
2002
;
62
:
3789
93
.
36
Couch
FJ
,
Johnson
MR
,
Rabe
KG
, et al
. 
The prevalence of BRCA2 mutations in familial pancreatic cancer
.
Cancer Epidemiol Biomarkers Prev
2007
;
16
:
342
6
.
37
Lowenfels
AB
,
Maisonneuve
P
,
Whitcomb
DC
,
International Hereditary Pancreatitis Study Group
. 
Risk factors for cancer in hereditary pancreatitis
.
Med Clin North Am
2000
;
84
:
565
73
.
38
Lowenfels
AB
,
Maisonneuve
P
,
DiMagno
EP
, et al
,
International Hereditary Pancreatitis Study Group
. 
Hereditary pancreatitis and the risk of pancreatic cancer
.
J Natl Cancer Inst
1997
;
89
:
442
6
.
39
Klein
AP
,
Brune
KA
,
Petersen
GM
, et al
. 
Prospective risk of pancreatic cancer in familial pancreatic cancer kindreds
.
Cancer Res
2004
;
64
:
2634
8
.
40
Jones
S
,
Hruban
RH
,
Kamiyama
M
, et al
. 
Exomic sequencing identifies PALB2 as a pancreatic cancer susceptibility gene
.
Science
2009
;
324
:
217
.
41
Slater
EP
,
Langer
P
,
Niemczyk
E
, et al
. 
PALB2 mutations in European familial pancreatic cancer families
.
Clin Genet
2010
.
Epub 18 Mar 2010. doi:
.
42
D'Andrea
AD
. 
Susceptibility pathways in Fanconi's anemia and breast cancer
.
N Engl J Med
2010
;
362
:
1909
19
.
43
Eberle
MA
,
Pfutzer
R
,
Pogue-Geile
KL
, et al
. 
A new susceptibility locus for autosomal dominant pancreatic cancer maps to chromosome 4q32–34
.
Am J Hum Genet
2002
;
70
:
1044
8
.
44
Pogue-Geile
KL
,
Chen
R
,
Bronner
MP
, et al
. 
Palladin mutation causes familial pancreatic cancer and suggests a new cancer mechanism
.
PLoS Med
2006
;
3
:
e516
.
45
Goicoechea
SM
,
Bednarski
B
,
Stack
C
, et al
. 
Isoform-specific upregulation of palladin in human and murine pancreas tumors
.
PLoS ONE
2010
;
5
:
e10347
,
PMCID: 2859948
.
46
Canto
MI
. 
Strategies for screening for pancreatic adenocarcinoma in high-risk patients
.
Semin Oncol
2007
;
34
:
295
302
.
47
Larghi
A
,
Verna
EC
,
Lecca
PG
,
Costamagna
G
. 
Screening for pancreatic cancer in high-risk individuals: a call for endoscopic ultrasound
.
Clin Cancer Res
2009
;
15
:
1907
14
.
48
Brune
K
,
Abe
T
,
Canto
M
, et al
. 
Multifocal neoplastic precursor lesions associated with lobular atrophy of the pancreas in patients having a strong family history of pancreatic cancer
.
Am J Surg Pathol
2006
;
30
:
1067
76
.
49
Crippa
S
,
Fernandez-Del Castillo
C
,
Salvia
R
, et al
. 
Mucin-producing neoplasms of the pancreas: an analysis of distinguishing clinical and epidemiologic characteristics
.
Clin Gastroenterol Hepatol
2010
;
8
:
213
9
.
50
Hwang
C
,
Rotterdam
H
,
Bill
A
, et al
. 
High-grade pancreatic intraepithelial neoplasia (PanIN) is associated with local recurrence of pancreatic adenocarcinoma after "curative" surgical resection
.
Gastroenterology
2008
;
134
:
A93
.