Cystic fibrosis transmembrane regulator gene carrier status is a risk factor for young onset pancreatic adenocarcinoma

Pancreatic adenocarcinoma is the fourth leading cause of cancer death in the USA. Although predominantly a cancer of the elderly, approximately 20% of patients are diagnosed under the age of 60 years. Younger patients are likely the best candidates for early surgical intervention, and patients at risk for young onset cancer comprise a logical focus for screening or prevention.

Carriers of mutations in the gene that encodes the cystic fibrosis transmembrane conductance regulator (CFTR) are associated with chronic idiopathic pancreatitis.¹ Chronic pancreatitis, in turn, increases the risk for pancreatic cancer by 26-fold.² Therefore, we hypothesised that mutations in CFTR may confer a higher risk of pancreatic cancer.

From October 2000 to April 2004, pancreatic cancer patients seen at the Mayo Clinic were ultra rapidly recruited to our study, with more than 75% of all such patients seen at the Mayo Clinic enrolled in the registry. This represents a substantial improvement over population based pancreatic cancer epidemiological studies, with participation rates ranging from 34.6%³ to 45.6%.⁴ Informed written consent and institutional review board approval were obtained.

As a pilot study, 33 patients were selected in whom a pathological diagnosis of pancreatitis was also noted at the time of pancreatic cancer surgery. The patients ranged in age from 41 to 81 years (median 65), and seven of the 33 had a diagnosis of pancreatitis made at least one year prior to cancer diagnosis. These patients were screened for variants in CFTR using the Tag-It Mutation Detection Kit, a clinically available kit testing for 40 mutations.

Of 33 samples tested, two patients (6%) were noted to have mutations in CFTR, both of which were the most common mutation identified in the CFTR gene, ΔF508. Both patients had young onset disease (ages 42 and 50 years). In total, seven patients in our pilot sample were below the age of 60 years, making the carrier rate 29% in this young onset subgroup.

Therefore, we designed a larger study to test the remainder of young onset cases in our registry, comprising a sequential unselected sample for mutations in CFTR (Cystic Fibrosis v3.0 ASR, Celera/Abbott), totalling 166 patients under the age of 60 years. Smoking status and family history were obtained from questionnaires. Personal history of chronic pancreatitis was identified by a single physician review of the medical records.

For a comparison group, the clinical database of CFTR analyses performed at the Mayo Clinic from November 2003 to May 2004 was utilised. Ethnic composition of cases and controls were highly comparable.

As shown in table 1 ▶, 14 of the 166 (8.4%) young onset pancreatic cancer cases were carriers for CFTR mutations, compared with 217 of 5349 (4.1%) patients in our control database (p = 0.006, odds ratio 2.18 (95% confidence interval 1.24–3.29)). There was no significant difference in age of onset, pancreatitis, family history of pancreatic cancer, or smoking in carriers versus non-carriers of CFTR mutations.

Table 1.

 Comparison of CFTR mutation frequencies detected in the young onset pancreatic cancer cohort versus the clinical database

	Young onset pancreatic cancer cases (<60 y old at diagnosis, n = 166)		Mayo Clinic clinical database reference group (n = 5349)
	No	%	No	%
CFTR mutation non-carriers	152	91.6	5132	95.9
CFTR mutation carriers	14	8.4	217	4.1
Mutation distribution
ΔF508	12	85.7	155	71.4
R177H	1	7.1	28	12.9
G551D			6	2.8
2789+5G>A			6	2.8
G542X			4	1.8
N1303K	1	7.1	3	1.4
1717-1G>T			2	0.9
3849+10kbC>T			2	0.9
A455E			2	0.9
R1162X			2	0.9
R347H			1	0.5
R553X			1	0.5
3905insT			1	0.5
621+1G>T			1	0.5
W1282X			1	0.5
1898+1G>A			1	0.5
R560T			1	0.5

Young onset pancreatic cancer cases were more frequent carriers of the CFTR mutations compared with patients in the control database (odds ratio 2.18 (95% confidence interval 1.24–3.29); p = 0.006).

Several cases of patients with cystic fibrosis (CF) and pancreatic adenocarcinoma have been reported,^5,6 and two cohort studies have shown an increased risk for pancreatic cancer among CF homozygotes.^7,8 Two studies have investigated CFTR mutation frequencies in pancreatic cancer patients, with negative results. However, both series only investigated one mutation (ΔF508), and neither focused on young onset patients.^9,10

Our study represents the first positive association of pancreatic cancer risk with CFTR carrier status, with mutations conferring a twofold risk for cancer before the age of 60 years. The finding that only one of the CFTR carriers had an antecedent history of pancreatitis is intriguing, as either pancreatitis is subclinical or the presence of one mutant CFTR allele may increase the risk for pancreatic cancer through a mechanism independent of chronic pancreatitis. A larger study to confirm these results is ongoing.

Acknowledgements

We thank the patients in this study and the contributions of Tammy Dahl, RN, Kathy Liffrig, Cynthia Nixa, Diane Batzel, Que Luu, Suresh Chari, MD, and Thomas Smyrk, MD.

Funding for this research was provided by the Mayo Clinic SPORE in Pancreatic Cancer (P50 CA 102701), R25T CA 92049, Lustgarten Foundation for Pancreas Cancer Research, NCI GRANT (R01 CA97075).