Background
Pancreatic adenocarcinoma (PC) is a leading cause of cancer-related death, with a 5-year survival rate of only 9%.1 Most patients are diagnosed with inoperable locally advanced or metastatic PC, for which systemic therapy remains largely ineffective.2 Earlier detection of PC and its high-grade precursor lesions is critical to improve patient outcomes.
Population-based screening for PC is not justified due to the low incidence of PC among unselected individuals and the lack of a proven survival benefit.3 However, surveillance may be suitable for high-risk individuals (HRIs).4 HRIs can be subdivided into those with (1) a significant family history of PC (familial pancreatic cancer [FPC]) or (2) a pathogenic germline mutation in a PC susceptibility gene (genetic syndrome [GS]), including mutations in BRCA1, BRCA2, PALB2, ATM, and CDKN2A (familial atypical multiple mole melanoma syndrome); STK11 (Peutz-Jeghers syndrome [PJS]); PRSS1 (hereditary pancreatitis); mismatch repair genes MLH1, MSH2, MSH6, and PMS2 (Lynch syndrome); and TP53 (Li-Fraumeni syndrome). Depending on the genetic syndrome, the risk of PC in HRIs varies between 5% and 36%. For patients with FPC, PC risk depends on the number of affected relatives and ranges between 6.4% and 32.0% in individuals with 2 or 3 affected first-degree relatives, respectively.5,6
The optimal surveillance strategy for HRIs is unclear, and interpretation of existing data is limited by heterogeneous inclusion criteria, imaging modalities, and surveillance intervals. NCCN, the Cancer of the Pancreas Screening (CAPS) Consortium, and the American College of Gastroenterology recommend surveillance with annual MRI and/or endoscopic ultrasound (EUS).4,7,8 In recent meta-analyses, the detection rate of high-risk lesions ranged between 0.50 and 0.74 cases per 100 patient-years screened.9,10 However, a fraction of HRIs have been diagnosed with locally advanced or metastatic PC despite annual surveillance, suggesting that some patients may exhibit accelerated progression.10 These observations provide the rationale to evaluate surveillance protocols with shorter imaging intervals that may enable earlier detection of high-risk lesions. This is the first report of a biannual PC surveillance program alternating between MRI and EUS in HRIs. The primary objectives of our program are to evaluate the detection of premalignant lesions and early-stage PCs using biannual imaging and to determine whether locally advanced or metastatic PCs develop despite biannual surveillance. Herein, we report an interim analysis of our program considering a diagnosis of an incurable PC in a patient with PJS even though we adhered to biannual imaging.
Methods
Patient Enrollment
Patients were enrolled in a prospective PC surveillance study at McGill University Health Centre (MUHC) between January 2013 and April 2020. Eligible participants were asymptomatic individuals with a ≥5-fold relative risk of PC compared with the general population. To ascertain study eligibility, a genetic counselor (A. Cuggia) obtained a personal and, at minimum, a 3-generation family history and verified clinical diagnoses through review of medical records. Participants were invited to complete a personal history questionnaire, which examined their environmental and epidemiologic risk factors. All patients were evaluated by a pancreatic surgeon (G. Zogopoulos) and a genetic counselor every 6 months. Clinical data were abstracted from a prospectively maintained database.11 All patients provided written informed consent. The study was performed in accordance with the principles of the Declaration of Helsinki and was approved by the MUHC Institutional Review Board (#2018-3171).
Inclusion Criteria
Table 1 summarizes the study inclusion criteria. These criteria underwent minor revisions over the study period, in parallel with changes in consensus guidelines.4,12 Prior versions of the inclusion criteria are outlined in supplemental eTable 1 (available with this article at JNCCN.org).
Study Inclusion Criteria (July 2019–Present)
Surveillance Protocol
Participants alternated between MRI and EUS examinations every 6 months. At each clinic visit, a serum CA 19-9 level was measured. MRI was performed on a 1.5 Tesla machine (General Electric) at McGill University Health Centre using intravenous administration of gadobutrol. Magnetic resonance cholangiopancreatography sequences were acquired. All MRI examinations were reviewed by an abdominal radiologist (A. Stanek). All EUS examinations were performed by 3 experienced endosonographers (K.A. Waschke, J. Parent, Y.I. Chen). Radial or linear instruments were used, at the discretion of the endosonographer. Pancreatic cysts were considered branch-duct intraductal papillary mucinous neoplasms (BD-IPMNs) if they indicated communication with the main pancreatic duct on MRI and/or EUS. In the absence of worrisome features, pancreatic cysts that were clinically and radiologically consistent with BD-IPMNs were not routinely aspirated for cytology or cyst fluid analysis.13 Preceding MRI/magnetic resonance cholangiopancreatography and EUS imaging results were included in the patient’s electronic medical record and were available for review by the endoscopist or radiologist at their discretion.
Germline Genetic Testing of Individuals With FPC
Patients with FPC who met institutional clinical genetic testing criteria were referred to a medical geneticist (W.D. Foulkes). These criteria included (1) >3 affected relatives on the same side of the family or (2) a family history suggestive of a genetic syndrome. The decision to pursue genetic testing and the choice of genetic testing panel were at the discretion of the medical geneticist.
At the time of writing, all patients with FPC were retroactively tested using a 26-gene hereditary cancer panel unless they had already been tested using this panel clinically. DNA was extracted from peripheral lymphocytes using the Gentra PureGene Blood kit (Qiagen, Hilden, Germany). A clinical-grade capture-based targeted sequencing pipeline was used to identify mutations in 26 hereditary cancer genes (supplemental eTable 2), as previously described.14 Variants were interpreted by a molecular geneticist (G. Chang) and classified using the American College of Medical Genetics guidelines.15 Individuals who were initially enrolled based on their FPC pedigree and in whom a germline mutation was identified post hoc were reclassified into the GS group. Although the 26-gene panel did not include CDKN2A, patients with a personal or family history of melanoma were tested with a 9-gene melanoma panel (BAP1, BRCA2, CDK4, CDKN2A, MITF, POT1, PTEN, RB1, and TP53).
Estimation of PC Risk in Individuals With FPC
For individuals with FPC, the PancPRO risk estimation model (CancerGene V6, Dallas, TX) was used to estimate their lifetime PC risk.16 The PancPRO model was not applied to individuals in the GS group because it does not adjust for the effect of pathogenic germline mutations.
Whole-Genome Sequencing of PJS-Associated PC
For the patient with PJS (patient MUHC-PJS-1) who developed PC within the surveillance program, fresh tumor biopsy cores were cryopreserved in optimal cutting temperature. Following laser-capture microdissection to enrich for tumor cells, tumor and matched lymphocyte DNA underwent whole-genome sequencing (WGS) to a coverage of at least 40×17 and were analyzed using previously described bioinformatics pipelines.18,19
Identification of Additional Patients With PJS-Associated PC
To identify additional patients with PJS-associated PC, we reviewed the Quebec Pancreas Cancer Study (QPCS) and Ontario Pancreas Cancer Study (OPCS) databases. The QPCS and OPCS are prospective PC research registries that have been enrolling patients since 2012 and 2003, respectively.11,20 Of 3,853 patients enrolled, only 2 with PJS were identified (patients OPCS-PJS-1 and OPCS-PJS-2). A review of their family history revealed that OPCS-PJS-1 was the son of MUHC-PJS-1, whereas OPCS-PJS-2 was from an unrelated kindred.
Laser-Capture Microdissection and KRAS/GNAS Genotyping
Areas of main-duct IPMN and BD-IPMN were identified and separately laser-capture microdissected from the resection specimen of patient MUHC-FPC-2, who was in the FPC group of the surveillance program. DNA was extracted using the QIAamp DNA FFPE Tissue Kit (Qiagen) using the primers indicated in supplemental eTable 3. Sanger sequencing was analyzed using SnapGene (GSL Biotech LLC).
Identification of Patients With Colloid-Type PC From the PanCuRx Translational Research Initiative
Patients with colloid-type PC from the PanCuRx Translational Research Initiative (TRI) were identified by a gastrointestinal pathologist (S.E. Fischer). Fresh tumor tissue and matched lymphocyte DNA underwent WGS, as previously described.17 The PanCuRx TRI was approved by the University Health Network (#15-9596) and MUHC (#2018-4139) Institutional Review Boards.
Statistical Analyses
Statistical analyses were performed using SPSS Statistics, version 23 (IBM Corp). Continuous variables were expressed as medians (range) and categorical variables were expressed as number (percentage). Differences between continuous and categorical variables were compared using the Mann-Whitney U and Fisher exact tests, respectively. Multivariable binary logistic regression was used to identify factors associated with the presence of pancreatic cysts. Cohen’s kappa was used to calculate the intermodality agreement between MRI and EUS. A P value <.05 was considered statistically significant.
Results
Characteristics of HRIs
Seventy-five HRIs from 60 families completed at least one surveillance examination. The median duration of follow-up was 32 months (range, 1–87 months). In total, 218 MRIs and 185 EUSs were performed. The personal history questionnaire completion rate was 72.0%.
Thirty-two patients (42.7%) were classified in the FPC group based on their pedigree, whereas 43 (57.3%) carried a germline mutation in a PC susceptibility gene. Germline testing results of patients with FPC are summarized in supplemental eTable 4. One individual in the FPC group was found to carry a pathogenic ATM mutation and was reclassified into the GS group.
Demographic characteristics are summarized in Table 2. Individuals in the GS group were significantly younger at enrollment (median age, 55 vs 58 years; P=.03). There were no significant differences in the rates of smoking, alcohol consumption, or diabetes between the FPC and GS groups.
Demographic and Clinical Characteristics of HRIs
For the FPC group, the median number of affected relatives was 3 (range, 2–7; supplemental eTable 5), and the median PancPRO estimated lifetime PC risk was 8.8% (range, 1.8%–13.2%). Individuals in the GS group are summarized in supplemental eTable 6. BRCA2 was the most commonly mutated gene. Germline mutations in patients from the GS group are detailed in supplemental eTable 7.
Pancreatic Findings in HRIs
Of 75 HRIs, 47 (62.7%) had a pancreatic abnormality detected by MRI and/or EUS (supplemental eTable 8).
Solid Pancreatic Lesions
A solid pancreatic lesion was not detected during scheduled surveillance imaging. However, 1 patient (MUHC-PJS-1) was diagnosed with locally advanced PC between surveillance examinations. MRI identified a pancreatic lipoma or focal fatty infiltration in 3 patients and a 6-mm pancreatic neuroendocrine tumor in another patient.
Cystic Pancreatic Lesions
Of the 75 HRIs, 33 (44.0%) had at least 1 pancreatic cystic lesion detected by either MRI or EUS. Of these, 26 (78.8%) demonstrated communication with the main pancreatic duct, consistent with BD-IPMN. Twenty-eight individuals (84.8%) had multiple cysts, whereas only 10 (30.3%) had a dominant cyst >1 cm (supplemental eTable 9). Of the 27 patients with pancreatic cysts who had undergone both MRI and EUS, 19 (70.4%) had pancreatic cysts detected by both modalities. There was high intermodality agreement between MRI and EUS (Cohen’s κ, 0.714; P<.001). Most pancreatic cysts were identified at the time of the first surveillance imaging (MRI, 26/29 [90%]; EUS, 19/23 [83%]). All cysts detected on subsequent imaging were <1 cm, and 5 of 7 had been identified at the first imaging using the alternate modality.
To identify factors predictive of the presence of a pancreatic cyst, we performed univariate binary logistic regression. Variables that achieved a P value <.15 in the univariate binary logistic regression model (age, sex, smoking status) were analyzed using multivariate binary logistic regression. On multivariable binary logistic regression analysis, older age was an independent predictor of the presence of a pancreatic cystic lesion (supplemental eTable 10).
Parenchymal Abnormalities
A total of 9 individuals (12.0%) had parenchymal abnormalities detected by MRI or EUS. One patient with a PRSS1 mutation and chronic pancreatitis had parenchymal changes detected by both MRI (atrophic parenchyma) and EUS (hyperechoic strands/foci). The remaining 8 patients had parenchymal changes only detected by EUS, including nonshadowing hyperechoic strands/foci (n=4), hypoechoic lobules (n=1), or both (n=3). None of these patients fulfilled Rosemont criteria for chronic pancreatitis.21
Progressive Lesions
Two patients (2.7%) developed progressive lesions while enrolled in the surveillance program. The first patient (MUHC-PJS-1) was a 65-year-old woman with PJS (Figure 1A). She had multiple BD-IPMNs, the largest of which was in the uncinate process, measured 1.8 cm, and had remained stable on biannual surveillance. During the fourth year of surveillance, she presented to the emergency department with abdominal pain. Imaging showed a locally advanced mass in the location of the previously seen BD-IPMNs, which was biopsied and confirmed to be poorly differentiated PC (Figure 1B, C). She was treated with gemcitabine and died 6 months after diagnosis. Importantly, her last EUS, performed 5 months before diagnosis, did not show worrisome features (Figure 1B).
Clinical and genomic characterization of patient MUHC-PJS-1, who developed PJS-associated PC. (A) Family history indicating the STK11 carrier status of MUHC-PJS-1 (solid arrow) and her children (including patient OPCS-PJS-1, dotted arrow). (B) (Top left) MRCP performed 11 months before PC diagnosis showing a 1.8-cm BD-IPMN at the uncinate process. Although this lesion may have corresponded to a cluster of smaller cysts, multidisciplinary review favored a multiloculated BD-IPMN. (Top right) Radial EUS performed 5 months before PC diagnosis showing the BD-IPMN without worrisome features. (Bottom left) CT scan showing locally advanced PC in the location of the previous BD-IPMNs. (Bottom right) EUS image of the PC at the time of diagnosis, showing interval progression. Note that a gastroduodenoscopy was performed at the time of PC diagnosis and confirmed the absence of any invasive duodenal lesion.
Citation: Journal of the National Comprehensive Cancer Network 20, 6; 10.6004/jnccn.2021.7107
Clinical and genomic characterization of patient MUHC-PJS-1, who developed PJS-associated PC. (A) Family history indicating the STK11 carrier status of MUHC-PJS-1 (solid arrow) and her children (including patient OPCS-PJS-1, dotted arrow). (B) (Top left) MRCP performed 11 months before PC diagnosis showing a 1.8-cm BD-IPMN at the uncinate process. Although this lesion may have corresponded to a cluster of smaller cysts, multidisciplinary review favored a multiloculated BD-IPMN. (Top right) Radial EUS performed 5 months before PC diagnosis showing the BD-IPMN without worrisome features. (Bottom left) CT scan showing locally advanced PC in the location of the previous BD-IPMNs. (Bottom right) EUS image of the PC at the time of diagnosis, showing interval progression. Note that a gastroduodenoscopy was performed at the time of PC diagnosis and confirmed the absence of any invasive duodenal lesion.
Citation: Journal of the National Comprehensive Cancer Network 20, 6; 10.6004/jnccn.2021.7107
Clinical and genomic characterization of patient MUHC-PJS-1, who developed PJS-associated PC. (A) Family history indicating the STK11 carrier status of MUHC-PJS-1 (solid arrow) and her children (including patient OPCS-PJS-1, dotted arrow). (B) (Top left) MRCP performed 11 months before PC diagnosis showing a 1.8-cm BD-IPMN at the uncinate process. Although this lesion may have corresponded to a cluster of smaller cysts, multidisciplinary review favored a multiloculated BD-IPMN. (Top right) Radial EUS performed 5 months before PC diagnosis showing the BD-IPMN without worrisome features. (Bottom left) CT scan showing locally advanced PC in the location of the previous BD-IPMNs. (Bottom right) EUS image of the PC at the time of diagnosis, showing interval progression. Note that a gastroduodenoscopy was performed at the time of PC diagnosis and confirmed the absence of any invasive duodenal lesion.
Citation: Journal of the National Comprehensive Cancer Network 20, 6; 10.6004/jnccn.2021.7107
WGS of this patient’s PC revealed biallelic inactivation of STK11 and CDKN2A (Figure 1D). The tumor retained wild-type KRAS and SMAD4. There was no evidence of chromothripsis or polyploidization. Of note, this patient’s son (OPCS-PJS-1) also had PJS and was diagnosed with metastatic PC at age 38 years (supplemental eFigure 1A, B). An MRI enterography performed for small-bowel surveillance 3 months before his PC diagnosis showed a 0.7-cm BD-IPMN in the pancreatic tail without worrisome features (supplemental eFigure 1C). Importantly, the PC developed in the same location as the BD-IPMN, suggesting that it also arose from malignant transformation of the BD-IPMN. Similar to his mother’s condition (MUHC-PJS-1), WGS of his PC demonstrated biallelic inactivation of STK11 and retention of wild-type KRAS (Figure 1D). Of note, the remaining 4 patients with PJS in our surveillance study did not develop any progressive lesions (supplemental eTable 11).
The second patient (MUHC-FPC-2) was a 68-year-old woman with FPC (supplemental eFigure 2A). She had multiple pancreatic body BD-IPMNs, the largest of which measured 1.5 cm. During her fifth year of surveillance, MRI showed main pancreatic duct dilatation with a focal stricture in the pancreatic body, near the dominant BD-IPMN (supplemental eFigure 2B). She underwent a laparoscopic spleen-preserving subtotal pancreatectomy, and pathology revealed mixed-type IPMN (gastric-type) with moderate-grade dysplasia (supplemental eFigure 2C). The microdissected main-duct IPMN harbored a KRAS G12D mutation, whereas a microdissected BD-IPMN harbored a KRAS G12V mutation, suggesting the coexistence of distinct clones (supplemental eFigure 2D). No GNAS mutations were detected in either sample. The patient has since resumed surveillance.
Identification of Additional Patients With PJS-Associated PC
We reviewed 2 prospective PC registries (QPCS and OPCS) for additional patients with PJS-associated PC and examined their association with premalignant cystic lesions. Of 3,853 patients with PC, we identified 2 with PJS-associated PC. One patient (OPCS-PJS-1) was described earlier, and the other patient (OPCS-PJS-2) was from an unrelated kindred. This patient underwent a pancreaticoduodenectomy at age 58 years for a BD-IPMN that harbored low-grade dysplasia (Figure 2A, B). He had a subcentimetric BD-IPMN in the pancreatic tail that remained stable on follow-up for 7 years. At age 66 years, this BD-IPMN progressed in size with interval development of nodularity (Figure 2C, D). A completion pancreatectomy was recommended, but the patient declined surgery. In the subsequent 6 months, the BD-IPMN progressed to a large infiltrative mass with mesenteric lymphadenopathy (Figure 2E, F). A lymph node biopsy revealed adenocarcinoma consistent with a pancreatic primary. Notably, the adenocarcinoma displayed mucinous differentiation, further supporting the malignant transformation of the progressive BD-IPMN. He died shortly thereafter.
Evolution of patient OPCS-PJS-2. (A, B) At age 58 years, the patient underwent a pancreaticoduodenectomy for a BD-IPMN that harbored low-grade dysplasia. (C) The patient had a pancreatic-tail BD-IPMN that was stable on follow-up for the next 7 years. (D) At age 66 years, the BD-IPMN progressed in size with the development of nodularity. Completion pancreatectomy was recommended, but the patient declined surgery. (E) Over the subsequent 6 months, the BD-IPMN progressed to an infiltrative lesion involving the pancreatic tail, associated with mesenteric and retroperitoneal lymphadenopathy. (F) Biopsy of a mesenteric lymph node showed adenocarcinoma consistent with a pancreatic primary (CK7+/CK19+/CDX2+/CK20−). The presence of mucinous differentiation further supports malignant transformation from a BD-IPMN (original magnification ×10).
Abbreviations: BD-IPMN, branch-duct intraductal papillary mucinous neoplasm; d, died; dx, diagnosed; FPC, familial pancreatic cancer; IPMN, intraductal papillary mucinous neoplasm; MGD, moderate grade dysplasia; MRCP, magnetic resonance cholangiopancreatography; MUHC, McGill University Health Centre.
Citation: Journal of the National Comprehensive Cancer Network 20, 6; 10.6004/jnccn.2021.7107
Evolution of patient OPCS-PJS-2. (A, B) At age 58 years, the patient underwent a pancreaticoduodenectomy for a BD-IPMN that harbored low-grade dysplasia. (C) The patient had a pancreatic-tail BD-IPMN that was stable on follow-up for the next 7 years. (D) At age 66 years, the BD-IPMN progressed in size with the development of nodularity. Completion pancreatectomy was recommended, but the patient declined surgery. (E) Over the subsequent 6 months, the BD-IPMN progressed to an infiltrative lesion involving the pancreatic tail, associated with mesenteric and retroperitoneal lymphadenopathy. (F) Biopsy of a mesenteric lymph node showed adenocarcinoma consistent with a pancreatic primary (CK7+/CK19+/CDX2+/CK20−). The presence of mucinous differentiation further supports malignant transformation from a BD-IPMN (original magnification ×10).
Abbreviations: BD-IPMN, branch-duct intraductal papillary mucinous neoplasm; d, died; dx, diagnosed; FPC, familial pancreatic cancer; IPMN, intraductal papillary mucinous neoplasm; MGD, moderate grade dysplasia; MRCP, magnetic resonance cholangiopancreatography; MUHC, McGill University Health Centre.
Citation: Journal of the National Comprehensive Cancer Network 20, 6; 10.6004/jnccn.2021.7107
Evolution of patient OPCS-PJS-2. (A, B) At age 58 years, the patient underwent a pancreaticoduodenectomy for a BD-IPMN that harbored low-grade dysplasia. (C) The patient had a pancreatic-tail BD-IPMN that was stable on follow-up for the next 7 years. (D) At age 66 years, the BD-IPMN progressed in size with the development of nodularity. Completion pancreatectomy was recommended, but the patient declined surgery. (E) Over the subsequent 6 months, the BD-IPMN progressed to an infiltrative lesion involving the pancreatic tail, associated with mesenteric and retroperitoneal lymphadenopathy. (F) Biopsy of a mesenteric lymph node showed adenocarcinoma consistent with a pancreatic primary (CK7+/CK19+/CDX2+/CK20−). The presence of mucinous differentiation further supports malignant transformation from a BD-IPMN (original magnification ×10).
Abbreviations: BD-IPMN, branch-duct intraductal papillary mucinous neoplasm; d, died; dx, diagnosed; FPC, familial pancreatic cancer; IPMN, intraductal papillary mucinous neoplasm; MGD, moderate grade dysplasia; MRCP, magnetic resonance cholangiopancreatography; MUHC, McGill University Health Centre.
Citation: Journal of the National Comprehensive Cancer Network 20, 6; 10.6004/jnccn.2021.7107
Genomic Characterization of Patients With Colloid-Type PC
Because both progressive lesions in our surveillance study occurred in HRIs with underlying IPMNs, we examined the genomic characteristics of colloid-type PCs, which arise from IPMNs. We performed WGS on 3 patients with colloid-type PCs from PanCuRx TRI. We found that COSMIC signature 17 was a dominant mutational signature in all 3 patients (supplemental eFigure 3). Two of the 3 patients retained wild-type KRAS. One patient had an activating GNAS mutation and the other had an activating ERBB2 mutation, highlighting alternate drivers of IPMN progression in the absence of KRAS mutation. The remaining patient harbored concurrent mutations in KRAS and GNAS.
Adverse Effects With HRI Surveillance
Two individuals developed cutaneous allergic reactions to gadobutrol and underwent subsequent noncontrast MRIs. Four patients (3 with FPC and 1 with a GS) dropped out because of anxiety (n=2) or change of interest in surveillance (n=2).
Laboratory Investigations in HRIs
Five patients (3 with FPC and 2 with a GS) had at least 1 elevated serum CA 19-9 level (>35 U/mL) during their follow-up. Of these, 3 had multiple pancreatic cysts, including 1 with a dominant cyst >1 cm. None of the 5 patients developed a progressive lesion over a median follow-up of 67 months.
Discussion
In our biannual surveillance program, one individual with PJS (MUHC-PJS-1) developed locally advanced PC. This patient’s accelerated progression highlights a limitation of PC surveillance, which may preferentially detect PCs that follow a conventional stepwise progression model. WGS of this PJS-associated PC did not reveal any chromothriptic rearrangements that are associated with accelerated progression.22 The tumor displayed biallelic inactivation of STK11. However, there was an absence of activating KRAS mutations, which are seen in >90% of PCs.23 This patient’s son (OPCS-PJS-1) also developed PC that did not show chromothriptic events but exhibited biallelic STK11 inactivation and wild-type KRAS. These findings are in keeping with a Dutch PJS cohort study, where all 5 patients who developed PC retained wild-type KRAS. 27 Together, these observations suggest that biallelic STK11 inactivation can drive pancreatic carcinogenesis in a KRAS-independent manner, with accelerated progression through mechanisms independent of chromothripsis.
Both the patient with PJS from our surveillance program (MUHC-PJS-1) and her son (OPCS-PJS-1) developed PC consistent with rapid progression of an underlying BD-IPMN in <6 months. By leveraging 2 PC registries, we identified an additional patient with PJS-associated PC (OPCS-PJS-2) that arose from malignant transformation of a BD-IPMN within 6 months. To further examine the relationship between PJS and IPMN transformation, we reviewed 12 PC surveillance studies that included patients with PJS.25–36 Of 62 patients with PJS, 9 (14.5%) underwent surgery for a pancreatic lesion, all of which were either pathologically confirmed or most consistent with an IPMN (supplemental eTable 12). Four patients had invasive carcinoma, which included 2 cases with nodal metastases, one of which progressed to metastatic disease within a 1-month interval from the last surveillance imaging examination. One patient had an IPMN with carcinoma in situ and 2 patients had IPMNs with high-grade dysplasia. This propensity among patients with PJS to develop IPMNs associated with high-grade dysplasia or invasive PC suggests that STK11 inactivation may promote IPMN progression. To this end, Sato et al37 showed that STK11 was lost in 100% of IPMNs arising in patients with PJS and in 25% of IPMNs in patients without PJS. Moreover, in transgenic mouse models, STK11 inactivation has synergized with KRAS activation to induce IPMNs.38 Together these observations suggest that the abrogation of STK11 function plays a role in driving IPMN progression and that patients with PJS are susceptible to IPMN carcinogenesis.
Surveillance studies to date have not consistently reported the time interval between PC diagnosis and the last surveillance imaging examination. Therefore, the published data do not allow for comparison of the timeline for progression of IPMN to invasive carcinoma in patients with PJS versus those without PJS. A recent systematic review and meta-analysis of observational studies estimated the rates of IPMN progression.39 Of 8,941 patients with a mean follow-up of 53 months, 11.8% underwent surgical resection, with a malignancy diagnosed in 3.5%. However, only 0.5% of patients had metastatic disease. These data suggest that in non-HRIs, most IPMNs can be safely surveilled with abdominal imaging as suggested by IPMN consensus guidelines. For HRIs, Konings et al40 recently reported on 76 patients from 11 CAPS Consortium surveillance programs who had either undergone pancreatic resection or progressed to advanced PC while under surveillance. Of these 76 HRIs, 71 had surgical resection, including 32 (45%) with a screen-detected PC or high-risk precursor lesion. Only 5 patients (6.6%) were diagnosed with advanced PC while under surveillance (PJS, n=1; FPC, n=4). These data suggest that even among HRIs, rapid progression to advanced PC between surveillance examinations is uncommon. This idea contrasts with the accelerated progression of BD-IPMNs that we have observed in patients with PJS. Thus, we are suggesting that similar to carriers of the CDKN2A mutation, patients with PJS with BD-IPMNs may warrant particular clinical attention.
A key finding of this study is the high prevalence of pancreatic cysts (44%), which were predominantly BD-IPMNs (35%), in HRIs. Interestingly, most HRIs had multiple, millimetric cysts, consistent with previous reports.29,31 This propensity for multifocality suggests an underlying field defect that may predispose HRIs to developing cysts throughout their pancreas. One important limitation of the current Fukuoka IPMN management guidelines is that they are not specific to HRIs.41,42 The CAPS guidelines state that for HRIs who are found to have pancreatic cysts without worrisome features, annual surveillance is appropriate.4 However, the guidelines specify that for carriers of the CDKN2A mutation with newly detected pancreatic abnormalities that do not lead to immediate surgery, close follow-up imaging in 3 to 6 months is warranted. Beyond this specification for carriers of the CDKN2A mutation, there are no specific cyst management recommendations for any other HRI subgroups, including patients with PJS. Recently, Dbouk et al43 compared the diagnostic performance of the 2017 Fukuoka guidelines42 and the 2020 CAPS guidelines4 for the management of pancreatic cysts in HRIs. They found that the Fukuoka criteria had a sensitivity and specificity of 40% and 85%, respectively. The CAPS guidelines showed a similar performance, with a sensitivity and specificity of 60% and 85%, respectively. These findings highlight the limited ability of current imaging-based features to predict the biology and evolution of IPMNs in HRIs.
The integration of molecular biomarkers may improve the risk stratification of BD-IPMNs. Although oncogenic KRAS and GNAS mutations are the most common alterations associated with IPMNs, they are early events in IPMN development and do not correlate with the degree of dysplasia.44,45 To this end, we identified discordant KRAS mutations coexisting in mixed-type IPMNs, suggesting that these precursor lesions arose independently from distinct clones. In contrast, mutations in the tumor suppressor genes CDKN2A, TP53, and SMAD4 occur more frequently in high-grade IPMNs and may have greater predictive value.46 Improved risk stratification of pancreatic cysts is particularly critical in HRIs because they have a higher prevalence of cystic lesions compared with the general population.
Comparison of the MRI and EUS examinations showed a high level of concordance for the detection of pancreatic cystic lesions (Cohen’s κ, 0.714; P<.001). However, parenchymal abnormalities were seen with EUS that were not visible on MRI. This finding is consistent with previous reports suggesting that chronic pancreatitis–like changes (lobulation, echogenic foci and strands) are frequently identified in HRIs and are best captured on EUS.47
Invasive carcinomas that arise from IPMNs are of either tubular or colloid subtype, with the latter having a better prognosis and a distinct histomorphology characterized by paucicellular pools of mucin. To uncover the genomic characteristics of IPMN-associated PCs, we examined 3 patients with colloid-type PCs.48 Interestingly, we found that all colloid-type PCs harbored a dominant COSMIC signature 17. Signature 17 is not a common mutational signature in PC but has been associated with gastroesophageal and intestinal adenocarcinomas.49,50 Because colloid-type PCs typically develop from IPMNs with intestinal differentiation, this distinct intestinal pathway of carcinogenesis may drive mutational processes that enrich for signature 17.51
Interestingly, we did not identify paucicellular pools of mucin or a dominant signature 17 in the 2 patients with PJS-associated PCs (MUHC-PJS-1, OPCS-PJS-1) that arose from BD-IPMNs. Additional research is needed to determine whether PJS-associated IPMNs preferentially progress to the more aggressive tubular subtype.
Most surveillance studies have assigned patients to the FPC group based on familial PC aggregation alone, without systematically testing for germline mutations in PC susceptibility genes. Yet recent reports have found that HRIs with germline mutations are more likely to harbor progressive lesions compared with individuals with FPC, making an argument that patients in a GS group should be prioritized for surveillance protocols.31,52 We found similar rates of pancreatic cysts and progressive lesions in the GS and FPC groups, which may be a limitation of our sample size. However, this finding may also reflect the stringency of our FPC inclusion criteria compared with FPC cohorts from earlier studies. To this end, the lifetime estimated PC risk of our FPC group was 8.8%, which exceeded the 5% benchmark. Furthermore, we tested all patients with FPC for germline mutations using a 26-gene hereditary cancer panel and reclassified 1 patient into the GS group. Although we tested patients who reported a personal or family history of melanoma with a melanoma-specific gene panel, we cannot exclude the presence of CDKN2A mutations in the remaining patients with FPC, because the 26-gene panel did not include CDKN2A. Nonetheless, we suggest that germline testing at the time of enrollment may be considered to refine the classification of HRIs.
A strength of this study is the genomic characterization of the patients with PJS-associated PC. In addition, a 6-month surveillance strategy, alternating between MRI and EUS, has not been previously reported. Notably, our results suggest that advanced PCs can be missed despite a 6-month imaging interval. Furthermore, we performed germline testing on all patients with FPC to assign HRIs into the GS and FPC subgroups. Despite its strengths, the study is limited in its conclusions by its sample size, the median follow-up of 32 months, and the availability of only 3 patients with PJS-associated PC from 3,853 patients with PC across 2 research registries. Nonetheless, our data suggest that BD-IPMNs in patients with PJS may benefit from more intensive surveillance, including molecular cyst fluid analysis. However, without a head-to-head comparison between biannual surveillance and early surgical resection for BD-IPMNs in patients with PJS, there is insufficient evidence to justify a lower threshold for surgical intervention.
Conclusions
In this biannual surveillance program, we have shown the high prevalence of BD-IPMNs in HRIs. BD-IPMNs arising in patients with PJS are at risk for accelerated malignant transformation. In addition to biannual imaging, surveillance of BD-IPMNs in patients with PJS may benefit from biomarker-based risk stratification, such as molecular cyst fluid analysis. However, before making surgical recommendations, large-scale prospective studies are needed to evaluate whether early resection of BD-IPMNs is more effective than biannual surveillance in patients with PJS.
Acknowledgments
We acknowledge the contributions of Guillaume Bourdel and Crystal Haigh and the Department of Diagnostic Radiology and the Department of Pathology at the McGill University Health Centre. We also acknowledge the contributions of team members at the Ontario Institute for Cancer Research within the Genomics and Bioinformatics platform (https://genomics.oicr.on.ca) and the contributions of Sheng-Ben Liang at the University Health Network Oncology Biobank.
References
- 2.↑
Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med 2013;369:1691–1703.
- 3.↑
Owens DK, Davidson KW, Krist AH, et al. Screening for pancreatic cancer: U.S. Preventive Services Task Force reaffirmation recommendation statement. JAMA 2019;322:438–444.
- 4.↑
Goggins M, Overbeek KA, Brand R, et al. Management of patients with increased risk for familial pancreatic cancer: updated recommendations from the International Cancer of the Pancreas Screening (CAPS) Consortium. Gut 2020;69:7–17.
- 6.↑
Klein AP, Brune KA, Petersen GM, et al. Prospective risk of pancreatic cancer in familial pancreatic cancer kindreds. Cancer Res 2004;64:2634–2638.
- 7.↑
Syngal S, Brand RE, Church JM, et al. American College of Gastroenterology. ACG clinical guideline: genetic testing and management of hereditary gastrointestinal cancer syndromes. Am J Gastroenterol 2015;110:223–262.
- 9.↑
Corral JE, Mareth KF, Riegert-Johnson DL, et al. Diagnostic yield from screening asymptomatic individuals at high risk for pancreatic cancer: a meta-analysis of cohort studies. Clin Gastroenterol Hepatol 2019;17:41–53.
- 10.↑
Signoretti M, Bruno MJ, Zerboni G, et al. Results of surveillance in individuals at high-risk of pancreatic cancer: a systematic review and meta-analysis. United European Gastroenterol J 2018;6:489–499.
- 11.↑
Smith AL, Bascuñana C, Hall A, et al. Establishing a clinic-based pancreatic cancer and periampullary tumour research registry in Quebec. Curr Oncol 2015;22:113–121.
- 12.↑
Canto MI, Harinck F, Hruban RH, et al. International Cancer of the Pancreas Screening (CAPS) Consortium summit on the management of patients with increased risk for familial pancreatic cancer. Gut 2013;62:339–347.
- 13.↑
Elta GH, Enestvedt BK, Sauer BG, et al. ACG clinical guideline: diagnosis and management of pancreatic cysts. Am J Gastroenterol 2018;113:464–479.
- 14.↑
Schenkel LC, Kerkhof J, Stuart A, et al. Clinical next-generation sequencing pipeline outperforms a combined approach using Sanger sequencing and multiplex ligation-dependent probe amplification in targeted gene panel analysis. J Mol Diagn 2016;18:657–667.
- 15.↑
Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med 2015;17:405–424.
- 16.↑
Wang W, Chen S, Brune KA, et al. PancPRO: risk assessment for individuals with a family history of pancreatic cancer. J Clin Oncol 2007;25:1417–1422.
- 17.↑
Aung KL, Fischer SE, Denroche RE, et al. Genomics-driven precision medicine for advanced pancreatic cancer: early results from the COMPASS Trial. Clin Cancer Res 2018;24:1344–1354.
- 18.↑
Denroche RE, Mullen L, Timms L, et al. A cancer cell-line titration series for evaluating somatic classification. BMC Res Notes 2015;8:823.
- 19.↑
Alexandrov LB, Nik-Zainal S, Wedge DC, et al. Signatures of mutational processes in human cancer. Nature 2013;500:415–421.
- 20.↑
Borgida AE, Ashamalla S, Al-Sukhni W, et al. Management of pancreatic adenocarcinoma in Ontario, Canada: a population-based study using novel case ascertainment. Can J Surg 2011;54:54–60.
- 21.↑
Catalano MF, Sahai A, Levy M, et al. EUS-based criteria for the diagnosis of chronic pancreatitis: the Rosemont classification. Gastrointest Endosc 2009;69:1251–1261.
- 22.↑
Notta F, Chan-Seng-Yue M, Lemire M, et al. A renewed model of pancreatic cancer evolution based on genomic rearrangement patterns. Nature 2016;538:378–382.
- 23.↑
Cicenas J, Kvederaviciute K, Meskinyte I, et al. KRAS, TP53, CDKN2A, SMAD4, BRCA1, and BRCA2 mutations in pancreatic cancer. Cancers (Basel) 2017;9:42.
- 24.
Korsse SE, Biermann K, Offerhaus GJA, et al. Identification of molecular alterations in gastrointestinal carcinomas and dysplastic hamartomas in Peutz-Jeghers syndrome. Carcinogenesis 2013;34:1611–1619.
- 25.↑
Canto MI, Goggins M, Yeo CJ, et al. Screening for pancreatic neoplasia in high-risk individuals: an EUS-based approach. Clin Gastroenterol Hepatol 2004;2:606–621.
- 26.↑
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;4:766–781.
- 27.↑
Poley JW, Kluijt I, Gouma DJ, et al. The yield of first-time endoscopic ultrasonography in screening individuals at a high risk of developing pancreatic cancer. Am J Gastroenterol 2009;104:2175–2181.
- 28.↑
Al-Sukhni W, Borgida A, Rothenmund H, et al. Screening for pancreatic cancer in a high-risk cohort: an eight-year experience. J Gastrointest Surg 2012;16:771–783.
- 29.↑
Canto, MI, Hruban RH, Fishman EK, et al. Frequent detection of pancreatic lesions in asymptomatic high-risk individuals. Gastroenterology 2012;142:796–804.
- 30.↑
Sud A, Wham D, Catalano M, et al. Promising outcomes of screening for pancreatic cancer by genetic testing and endoscopic ultrasound. Pancreas 2014;43:458–461.
- 31.↑
Konings ICAW, Harinck F, Poley JW, et al. Prevalence and progression of pancreatic cystic precursor lesions differ between groups at high risk of developing pancreatic cancer. Pancreas 2017;46:28–34.
- 32.↑
Canto MI, Almario JA, Schulick RD, et al. Risk of neoplastic progression in individuals at high risk for pancreatic cancer undergoing long-term surveillance. Gastroenterology 2018;155:740–751.e2.
- 33.↑
Paiella S, Capurso G, Cavestro GM, et al. Results of first-round of surveillance in individuals at high-risk of pancreatic cancer from the AISP (Italian Association for the Study of the Pancreas) registry. Am J Gastroenterol 2019;114:665–670.
- 34.↑
Barnes CA, Krzywda E, Lahiff S, et al. Development of a high risk pancreatic screening clinic using 3.0 T MRI. Fam Cancer 2018;17:101–111.
- 35.↑
DaVee T, Coronel E, Papafragkakis C, et al. Pancreatic cancer screening in high-risk individuals with germline genetic mutations. Gastrointest Endosc 2018;87:1443–1450.
- 36.↑
Overbeek KA, Levink IJM, Koopmann BDM, et al. Long-term yield of pancreatic cancer surveillance in high-risk individuals [published online April 5, 2021]. Gut, doi:10.1136/gutjnl-2020-323611
- 37.↑
Sato N, Rosty C, Jansen M, et al. STK11/LKB1 Peutz-Jeghers gene inactivation in intraductal papillary-mucinous neoplasms of the pancreas. Am J Pathol 2001;159:2017–2022.
- 38.↑
Collet L, Ghurburrun E, Meyers N, et al. Kras and Lkb1 mutations synergistically induce intraductal papillary mucinous neoplasm derived from pancreatic duct cells. Gut 2020;69:704–714.
- 39.↑
Balduzzi A, Marchegiani G, Pollini T, et al. Systematic review and meta-analysis of observational studies on BD-IPMNS progression to malignancy. Pancreatology 2021;21:1135–1145.
- 40.↑
Konings ICAW, Canto MI, Almario JA, et al. Surveillance for pancreatic cancer in high-risk individuals. BJS Open 2019;3:656–665.
- 41.↑
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.
- 42.↑
Tanaka M, Fernández-Del Castillo C, Kamisawa T, et al. Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology 2017;17:738–753.
- 43.↑
Dbouk M, Brewer Gutierrez OI, Lennon AM, et al. Guidelines on management of pancreatic cysts detected in high-risk individuals: an evaluation of the 2017 Fukuoka guidelines and the 2020 International Cancer of the Pancreas Screening (CAPS) Consortium statements. Pancreatology 2021;21:613–621.
- 44.↑
Hosoda W, Sasaki E, Murakami Y, et al. GNAS mutation is a frequent event in pancreatic intraductal papillary mucinous neoplasms and associated adenocarcinomas. Virchows Arch 2015;466:665–674.
- 45.↑
Tan MC, Basturk O, Brannon AR, et al. GNAS and KRAS mutations define separate progression pathways in intraductal papillary mucinous neoplasm-associated carcinoma. J Am Coll Surg 2015;220:845–854.e1.
- 46.↑
Omori Y, Ono Y, Tanino M, et al. Pathways of progression from intraductal papillary mucinous neoplasm to pancreatic ductal adenocarcinoma based on molecular features. Gastroenterology 2019;156:647–661.e2.
- 47.↑
Thiruvengadam SS, Chuang J, Huang R, et al. Chronic pancreatitis changes in high-risk individuals for pancreatic ductal adenocarcinoma. Gastrointest Endosc 2019;89:842–851.e1.
- 48.↑
Adsay NV, Pierson C, Sarkar F, et al. Colloid (mucinous noncystic) carcinoma of the pancreas. Am J Surg Pathol 2001;25:26–42.
- 49.↑
Ross-Innes CS, Becq J, Warren A, et al. Whole-genome sequencing provides new insights into the clonal architecture of Barrett’s esophagus and esophageal adenocarcinoma. Nat Genet 2015;47:1038–1046.
- 50.↑
Secrier M, Li X, de Silva N, et al. Mutational signatures in esophageal adenocarcinoma define etiologically distinct subgroups with therapeutic relevance. Nat Genet 2016;48:1131–1141.
- 51.↑
Adsay NV, Merati K, Basturk O, et al. Pathologically and biologically distinct types of epithelium in intraductal papillary mucinous neoplasms: delineation of an “intestinal” pathway of carcinogenesis in the pancreas. Am J Surg Pathol 2004;28:839–848.
- 52.↑
Abe T, Blackford AL, Tamura K, et al. Deleterious germline mutations are a risk factor for neoplastic progression among high-risk individuals undergoing pancreatic surveillance. J Clin Oncol 2019;37:1070–1080.
