CDKN2A founder mutation in pancreatic ductal adenocarcinoma patients without cutaneous features of Familial Atypical Multiple Mole Melanoma (FAMMM) syndrome

Approximately 5% to 10% of pancreatic ductal adenocarcinoma (PDAC) has a hereditary basis with the major genes and syndromes identified thus far outlined in Table 1. In most of these syndromes, PDAC is not the predominant cancer type. Traditional criteria for publicly funded genetic testing typically require the presence of a set combination of the predominant syndrome-associated cancer types in the family history.

The genetic heterogeneity of familial pancreatic cancer (FPC), typically defined as the occurrence of pancreatic cancer in two affected first degree relatives, has been well described in the past decade, with mutations in high-penetrance genes such as BRCA2, CDKN2A, PALB2, STK11 thought to explain only 10%–15% of the familial clustering [1, 2]. In 2015, Zhen et al. reported on a comprehensive analysis of BRCA1, BRCA2, PALB2, and CDKN2A in a large cohort of FPC kindreds ascertained via the multicenter Pancreatic Cancer Genetic Epidemiology (PACGENE) Consortium and found a mutation prevalence of 8% among 515 FPC patients (41 mutations among the four genes analyzed) [3]. In that study, 2.5% of FPC PDAC patients had a mutation in the CDKN2A gene (n = 14), which is associated with the Familial Atypical Multiple Mole Melanoma (FAMMM) syndrome. Of note, half of the CDKN2A positive FPC kindreds did not have a family history of melanoma. The prevalence of CDKN2A mutations increased to 7.8% among PDAC cases with any positive family history of melanoma (n = 77).

In keeping with the finding of CDKN2A families without a history of melanoma, herein, we report the presentation of a CDKN2A pathogenic variant in a PDAC-prone family without the typical cutaneous features of multiple moles or melanoma that are characteristic of the FAMMM syndrome.

A 63 year-old male patient diagnosed with PDAC was referred to the BC Cancer Agency’s Hereditary Cancer Program for hereditary cancer syndrome assessment due to a strong family history of PDAC (Fig. 1). As the starting point for the genetic study of a family, this patient is considered the proband. The family history was noteworthy for the occurrence of PDAC in the proband and his brother, as well as an aunt and uncle on the maternal side. Diagnoses occurred between the ages of 63 and 69. Ancestry was reported as English on both sides.

The proband presented with a three-month history of weight loss and epigastric discomfort. A computerized tomography (CT) scan showed a 2.9 cm lesion in the head of the pancreas. He was deemed to have locally advanced unresectable PDAC and began treatment with a combination of 5-fluorouracil, irinotecan, and oxaliplatin (FOLFIRINOX). His medical history included type 2 diabetes, hypertension, osteoarthritis, gout and hyperlipidemia. He reported being a non-smoker who consumes 6–8 alcoholic drinks per day.

The proband’s brother presented with left upper quadrant pain and underwent an ultrasound as well as a CT scan which revealed the presence of a mass in the pancreatic head and uncinate process. He was deemed to have resectable PDAC and underwent a pancreaticoduodenectomy (Whipple) procedure. Pathology confirmed a 3.5 cm moderately differentiated PDAC. Margins were negative. He was treated with adjuvant gemcitabine for six months. He reported being a non-smoker, rarely consuming alcohol. He denied a significant prior medical history.

During the initial genetic consultation, the proband denied any history of breast, ovarian, colon cancer or melanoma in the family. There was no reported history of pancreatitis, multiple moles, lip freckling or gastrointestinal polyps. In this pancreatic-cancer prone family, none of the diagnoses occurred under the age of 50. The family history of cancer did not meet the current provincial guidelines for hereditary cancer genetic testing [7]. The patient was informed of the options of self-pay genetic testing through one of several genetic testing companies or banking DNA at the BC Cancer Agency should criteria change in the future. He elected to bank a DNA sample and consented to future research.

The frequency of CDKN2A pathogenic variants in PDAC families without melanoma has not been well established but was previously considered to be rare. Bartsch et al. (2002) investigated the frequency of CDKN2A pathogenic variants in 18 familial pancreatic cancer families and found no pathogenic variants in families without malignant melanoma, compared to two of five families with a history positive for both PDAC and melanoma [15]. McWiliams et al. (2011) describe the challenges in interpreting prevalence and penetrance of germline CDKN2A mutations in relation to PDAC given that the mutations have most commonly been described in families where the predominant lesion is cutaneous melanoma [10]. To address this, they studied an unselected PDAC population of 1537 Caucasian patients from the United States and found CKDN2A mutations in nine patients (0.6%). Of the 120 cases with a first-degree relative affected with PDAC, four (3.3%) carried mutations, which was significantly higher than in those without a family history. However, five of the nine mutation carriers did not have a first-degree affected relative. The authors estimated the lifetime risk to age 80 of PDAC to be 58% among the 59 first-degree relatives of nine pathogenic variant carriers. This is significantly higher than the 15–25% lifetime PDAC risk reported in melanoma-family studies [16, 17]. However, the authors comment that there was a high proportion of smokers in their study (41%) and that the difference in cancer risk was only evident among ever-smokers. It is interesting to note that the two brothers in this report were non-smokers.

The PACGENE study in 515 FPC kindreds found a nearly five-fold higher prevalence of CDKN2A mutations compared to what was seen in the sporadic PDAC population described by McWilliams et al. (2.5% compared to 0.6%). Of note, 7 of 14 (50%) of the CDKN2A mutation associated FPC kindreds did not have a history of melanoma [3].

The specific pathogenic variant found in this family, p.Val126Asp, is a founder mutation in North American families estimated to have originated approximately 34 to 52 generations ago, pre-dating colonization [18]. The mutation does not appear at high frequency in any other countries besides the United States and Canada. The variant is absent from or extremely rare in population databases [19]. This mutation inhibits the catalytic activity of the cyclin D1/CDK4 and cyclin D1/CDK6 complexes in vitro [20]. Among the seven families reported with this founder mutation in the GenoMEL consortium, PDAC was observed in over 40% and all seven families had a history of melanoma by way of ascertainment [9]. The p.Val126Asp mutation was also described in two PDAC patients in the PACGENE study but details were not provided on the family history of melanoma. In a Dutch study, Harinck et al. described three CDKN2A families (two with the Dutch founder mutation c.225_243del, p.Ala76fs and one with a c.19_23dup. pSer8fs mutation) without a history of melanoma and recommended that CDKN2A testing be considered even in the absence of reported melanomas [21].

As the use of multi-gene panel testing in patients with PDAC increases, it will be interesting to compare melanoma penetrance in CDKN2A families ascertained through FPC criteria compared to CDKN2A families ascertained through FAMMM criteria. Further studies are needed to understand what other factors, in addition to the CDKN2A pathogenic variant, contribute to the development of PDAC as opposed to melanoma in certain families.

The identification of a germline CDKN2A pathogenic variant changes the management of this family in several ways. Publicly funded carrier genetic testing for the known familial CDKN2A pathogenic variant is now available to relatives and will clarify cancer risk assessment and management guidelines (Table 3). Of note, the two remaining siblings of this sibship, neither of whom had a personal history of cancer, presented for genetic counseling and tested negative for the founder mutation. From a cancer treatment perspective, an active area of research is the potential predictive role of CDKN2A genetic status in terms of estimating sensitivity to CDK4/6 inhibitors [22].

As with any description of a family history of cancer, one must recognize that the lack of a melanoma in close relatives could be due to incomplete information. Furthermore, family histories are dynamic. However, the consistent reporting from all members of the sibship regarding the lack of any known cases of melanoma in first-, second- or third-degree relatives is compelling and an important presentation to take note of. Indeed, after being encouraged to inform family members about the pathogenic variant and to enquire about any other relatives with melanoma, they became aware of only one single case of melanoma, in a maternal second cousin (fifth-degree relative), in addition to some relatives with moles. To the family’s knowledge, these relatives had not been referred for a formal dermatological assessment.

In recent dermatology literature, a cautionary tone has been expressed concerning the diagnostic pitfall of disregarding family members without nevi as potential carriers of a germline CDKN2A mutation. Indeed, this case stresses that the same diagnostic pitfall would also arise by disregarding families without melanoma (+/− or nevi) as potential carriers [23].

Although the two siblings reported here were non-smokers, it is important to stress that the single most important modifier in the risk of PDAC appears to be cigarette smoking. Smoking increases the risk of pancreatic cancer in the general population 1.5 to 5.5-fold [24]. In 2003, Rulyak and others studied pancreatic cancer-prone families and noted that pancreatic cancer developed approximately 10 years earlier in smokers compared to non-smokers [25].

In conclusion, this report describing one of the most common CDKN2A mutations reported in North America, in a family without cutaneous features of FAMMM syndrome, should remind clinicians to consider the possibility of germline CDKN2A mutations in PDAC-prone families without history of moles or melanoma. This case supports recent guidelines published by American College of Medical Genetics and Genomics (ACMG) that genetics referrals are indicated in families with three or more pancreatic cancers, even in the absence of breast, colon and melanoma cancers [26]. Furthermore, in light of the known genetic heterogeneity, a multi-gene panel approach is of particular benefit in diagnosing inherited cancer susceptibility in PDAC-only families (Table 4).