Abstract
The presence of pancreatic cancer (PC) in melanoma-prone families has been consistently associated with an increased frequency of CDKN2A mutations, the major high-risk susceptibility gene identified for melanoma. However, the precise relationship between CDKN2A, melanoma and PC remains unknown. We evaluated a recently identified PC susceptibility gene PALB2 using both sequencing and tagging to determine whether PALB2 might explain part of the relationship between CDKN2A, melanoma, and PC. No disease-related mutations were identified from sequencing PALB2 in multiple pancreatic cancer patients or other mutation carrier relatives of PC patients from the eight melanoma-prone families with CDKN2A mutations and PC. In addition, no significant associations were observed between 11 PALB2 tagging SNPs and melanoma risk in 23 melanoma-prone families with CDKN2A mutations or the subset of 11 families with PC or PC-related CDKN2A mutations. The results suggested that PALB2 does not explain the relationship between CDKN2A, melanoma, and pancreatic cancer in these melanoma-prone families.
Keywords: CDKN2A, PALB2, familial melanoma, pancreatic cancer, germline mutation
Introduction
The CDKN2A gene, located on chromosome 9p21, is the major known high-risk melanoma susceptibility gene identified to date. Germline mutations in CDKN2A have been observed in 20–40% of melanoma-prone families from around the world (1). The presence of pancreatic cancer (PC) in melanoma-prone families has been consistently associated with an increased frequency of CDKN2A mutations (2). However, the precise relationship between CDKN2A, melanoma and PC remains unknown. Further, only a subset of CDKN2A mutations (e.g. p.R112_L113insR, c.225_243del19, p.G101W, and p.V126D) is linked with the occurrence of PC. Even in melanoma-prone families with putative PC-related CDKN2A mutations, only a small subset of individuals with these mutations develops PC (3). Thus, factors related to these specific PC-related CDKN2A mutations or alternatively non-CDKN2A factors may be responsible for the relationship between PC and melanoma in these CDKN2A mutation-positive families.
Recently, exomic sequencing identified PALB2 as a high-risk PC susceptibility gene (4). PALB2 is a binding partner of BRCA2 and plays an important role in facilitating BRCA2’s function in repair of DNA double-strand breaks by homologous recombination (5). Previous studies have shown an increased risk of melanoma among BRCA2 mutation carriers (6). Given the importance of PALB2 in PC and the relationship between melanoma and PC, we hypothesized that PALB2 might modify the risk of PC in melanoma-prone families with CDKN2A mutations.
Materials and Methods
Study population
The 23 melanoma-prone families included in this study are part of a larger study population that has previously been described (7, 8). Briefly, American families with at least two living first-degree relatives with a history of invasive melanoma were ascertained through health care professionals or self referrals. All diagnoses of melanoma and pancreatic cancer were confirmed by histologic review of pathologic material for melanoma only, or by review of pathology reports, medical records, or death certificates for melanoma and PC. Eleven of these mutation-positive families had at least one member with pancreatic cancer (n=8) or a CDKN2A mutation that has been consistently associated with PC (n=3) [p.G101W, p.V126D, c.225-243del19] (9). The study was approved by the National Cancer Institute Clinical Center Institutional Review Board and informed consent was obtained from all participants.
PALB2 sequencing and genotyping
We sequenced 13 PALB2 exons in available PC patients (n=5) from four melanoma-prone families with CDKN2A mutations (Table 1). These 5 PC patients had previously been sequenced for BRCA2; no truncating mutations were identified. We also sequenced PALB2 in seven melanoma patients and/or CDKN2A mutation carrier relatives of PC patients from four other melanoma-prone families with CDKN2A mutations and PC but in whom DNA from PC patients was not available (Table 1). All forward and reverse sequences were assembled and variants discovered using Variant Reporter™ v1.0 (Applied Biosystems, Foster City, CA). Each variant was then visually confirmed using Sequencher™ v4.0.5 software (Gene Codes Corporation, Ann Arbor, MI).
Table 1.
Subjects sequenced for PALB2 in CDKN2A melanoma-prone families with pancreatic cancer
| Family | Affection status of subject(s) sequenced | Relationship of subject(s) sequenced to PC patient in family | No. Melanoma patients in family | No. Pancreatic Cancer patients in family | CDKN2A Mutation | ||
|---|---|---|---|---|---|---|---|
| Location | Description | Mutation Type | |||||
| P | CMM/PC | Self | 11 | 1 | Exon 2 | 240-253del14 | Frameshift-Chimera |
| F | CMM/PC | Self | 12 | 1 | Exon 2 | R87P | Missense |
| D9 | CMM; 357delG carrier | Sibling; offspring | 3 | 1 | Exon 2 | 357delG | Frameshift |
| J | CMM/PC | Self | 7 | 1 | Exon 2 | V126D | Missense |
| K | CMM/PC (n=2) | Self | 6 | 3 | Exon 2 | V126D | Missense |
| L | CMM; V126D carrier | Grandchildren (n=2) | 10 | 1 | Exon 2 | V126D | Missense |
| AP | CMM | Cousin | 3 | 1 | Intron 2 | IVS2-105a>g | Splicing |
| Q | CMM | Offspring (n=2) | 3 | 2 | Intron 2 | IVS2+1 | Splicing |
We tagged PALB2 for genotyping in the 23 melanoma-prone families with CDKN2A mutations [97 melanoma patients; 217 controls (75 spouses and 142 unaffected family members)]. We also separately examined the subset of 11 families with PC or a mutation strongly associated with the occurrence of PC [48 melanoma patients; 116 controls]. Eleven tag SNPs for PALB2 were selected for genotyping using Fluidigm or Taqman with a minimum minor allele frequency criterion of ≥5% based upon HapMap data for Caucasian (CEU) samples using Tagzilla, software that implements a tagging algorithm based on pairwise linkage disequilibrium (LD) (10). SNPs spanning 20 kb 5′ of the start of transcription (exon 1) up to 10 kb 3′ from the end of the last exon were selected.
Statistical analyses
Conditional logistic regression models adjusted for age, gender, and CDKN2A mutation status were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) and the trend p-value for the association between melanoma and each SNP, using codominant coding for genotypes (0,1,2) with the homozygote of the common allele as the reference group. Conditioning on families was used to account for family ascertainment and differences in disease prevalence among families. While this approach ignores residual familial correlations among family members, it gives estimates that are attenuated toward the null and is thus conservative (11). A gene-based analysis was also performed on PALB2 to assess the significance of the joint effect of multiple SNPs genotyped. P-values were computed using a rank-truncated test statistic and a permutation-based sampling procedure (20,000 permutations) in the same regression model, taking into account the number of SNPs genotyped and their LD structure (12). All analyses were performed using SAS software, version 9.1 (SAS Institute, Inc., Cary, NC).
Results and Discussion
We did not observe disease-related mutations from sequencing PALB2 in all five available PC patients from four CDKN2A mutation-positive families. These PC patients had previously also shown no disease-related mutations from sequencing BRCA2 (unpublished data). In addition, no disease-related mutations in PALB2 were observed in seven melanoma patients and/or CDKN2A mutation carrier relatives of PC patients from four other melanoma-prone families with CDKN2A mutations and PC but in whom DNA from PC patients was not available (Table 1). In addition, none of the 11 PALB2 SNPs were significantly associated with melanoma (Table 2). Further, using the gene-based analysis, PALB2 was not associated with melanoma (p=0.34). Although based on small numbers, restricting the association analysis to the 11 CDKN2A families with PC or PC-related CDKN2A mutations again showed no significant associations between melanoma and the 11 PALB2 SNPs.
Table 2.
Associations of PALB2 tagging SNPs with melanoma in melanoma-prone families with CDKN2A mutations
| SNPs | All families (n=23) | Families with PC or PC-related mutations (n=11) | ||
|---|---|---|---|---|
| OR (95% CI)1 | p1 | OR (95% CI) 1 | p1 | |
| rs240745 | 0.58 (0.29–1.19) | 0.14 | 0.73 (0.25–2.12) | 0.57 |
| rs240744 | 1.42 (0.59–3.43) | 0.44 | 1.17 (0.39–3.53) | 0.78 |
| rs12162020 | 0.72 (0.33–1.61) | 0.43 | 1.27 (0.44–3.66) | 0.66 |
| rs420259 | 1.30 (0.73–2.32) | 0.37 | 1.60 (0.74–3.48) | 0.24 |
| rs513313 | 1.66 (0.65–4.24) | 0.29 | 1.23 (0.34–4.45) | 0.75 |
| rs16940342 | 1.14 (0.59–2.23) | 0.69 | 1.14 (0.52–2.52) | 0.74 |
| rs8058061 | 0.76 (0.18–3.26) | 0.71 | 1.30 (0.25–6.77) | 0.76 |
| rs0843812 | 0.58 (0.23–1.47) | 0.25 | 0.73 (0.15–3.55) | 0.70 |
| rs17806253 | 0.74 (0.34–1.57) | 0.43 | 0.71 (0.24–2.12) | 0.54 |
| rs34514 | 0.93 (0.49–1.75) | 0.82 | 1.02 (0.40–2.61) | 0.96 |
| rs34513 | 0.88 (0.48–1.61) | 0.68 | 1.24 (0.50–3.04) | 0.64 |
ORs and P values are obtained from likelihood ratio test in conditional logistic regression with melanoma as the outcome variable adjusting for age, gender, and CDKN2A status.
Our results are consistent with data from a recent report that found no deleterious PALB2 mutations in probands from 53 familial melanoma kindreds without CDKN2A mutations (13). In addition, recent investigations of PALB2 mutations in breast-pancreatic cancer families also found similar results showing that mutations of PALB2 in these families were rare(14, 15). Together, these results suggest that PALB2 mutations do not account for a substantial proportion of susceptibility in melanoma-prone or breast cancer-prone families with a history of PC.
The major limitation of the current study was the relatively small sample size available for investigation of PALB2. Additional examination of PALB2 in larger samples will be required to conclusively exclude a relationship between PALB2, melanoma and CDKN2A. Finally, the relationship between CDKN2A, melanoma and PC remains unexplained and additional studies are needed to determine the cause(s) for the observed associations.
Acknowledgments
We are indebted to the participating families, whose generosity and cooperation have made this study possible. We also acknowledge the contributions to this work that were made by Virginia Pichler, Deborah Zametkin, and Mary Fraser. This research was supported by the Intramural Research Program of the NIH, NCI, DCEG.
Footnotes
Conflict of interest
The authors declare no conflict of interest.
References
- 1.Hayward NK. Genetics of melanoma predisposition. Oncogene. 2003 May 19;22(20):3053–62. doi: 10.1038/sj.onc.1206445. [DOI] [PubMed] [Google Scholar]
- 2.Goldstein AM, Chan M, Harland M, et al. High-risk melanoma susceptibility genes and pancreatic cancer, neural system tumors, and uveal melanoma across GenoMEL. Cancer Res. 2006 Oct 15;66(20):9818–28. doi: 10.1158/0008-5472.CAN-06-0494. [DOI] [PubMed] [Google Scholar]
- 3.Goldstein AM, Chan M, Harland M, et al. Features associated with germline CDKN2A mutations: a GenoMEL study of melanoma-prone families from three continents. J Med Genet. 2007 Feb;44(2):99–106. doi: 10.1136/jmg.2006.043802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jones S, Hruban RH, Kamiyama M, et al. Exomic sequencing identifies PALB2 as a pancreatic cancer susceptibility gene. Science. 2009 Apr 10;324(5924):217. doi: 10.1126/science.1171202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Xia B, Sheng Q, Nakanishi K, et al. Control of BRCA2 cellular and clinical functions by a nuclear partner, PALB2. Mol Cell. 2006 Jun 23;22(6):719–29. doi: 10.1016/j.molcel.2006.05.022. [DOI] [PubMed] [Google Scholar]
- 6.Cancer risks in BRCA2 mutation carriers. The Breast Cancer Linkage Consortium. J Natl Cancer Inst. 1999 Aug 4;91(15):1310–6. doi: 10.1093/jnci/91.15.1310. [DOI] [PubMed] [Google Scholar]
- 7.Goldstein AM, Landi MT, Tsang S, Fraser MC, Munroe DJ, Tucker MA. Association of MC1R variants and risk of melanoma in melanoma-prone families with CDKN2A mutations. Cancer Epidemiol Biomarkers Prev. 2005 Sep;14(9):2208–12. doi: 10.1158/1055-9965.EPI-05-0321A. [DOI] [PubMed] [Google Scholar]
- 8.Goldstein AM, Struewing JP, Chidambaram A, Fraser MC, Tucker MA. Genotype-phenotype relationships in U.S. melanoma-prone families with CDKN2A and CDK4 mutations. J Natl Cancer Inst. 2000 Jun 21;92(12):1006–10. doi: 10.1093/jnci/92.12.1006. [DOI] [PubMed] [Google Scholar]
- 9.Goldstein AM. Familial melanoma, pancreatic cancer and germline CDKN2A mutations. Hum Mutat. 2004 Jun;23(6):630. doi: 10.1002/humu.9247. [DOI] [PubMed] [Google Scholar]
- 10.Carlson CS, Eberle MA, Rieder MJ, Yi Q, Kruglyak L, Nickerson DA. Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet. 2004 Jan;74(1):106–20. doi: 10.1086/381000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Pfeiffer RM, Hildesheim A, Gail MH, et al. Robustness of inference on measured covariates to misspecification of genetic random effects in family studies. Genet Epidemiol. 2003 Jan;24(1):14–23. doi: 10.1002/gepi.10191. [DOI] [PubMed] [Google Scholar]
- 12.Dudbridge F, Koeleman BP. Rank truncated product of P-values, with application to genomewide association scans. Genet Epidemiol. 2003 Dec;25(4):360–6. doi: 10.1002/gepi.10264. [DOI] [PubMed] [Google Scholar]
- 13.Sabbaghian N, Kyle R, Hao A, Hogg D, Tischkowitz M. Mutation analysis of the PALB2 cancer predisposition gene in familial melanoma. Fam Cancer. 2010 Dec 11; doi: 10.1007/s10689-010-9405-y. [DOI] [PubMed] [Google Scholar]
- 14.Hofstatter EW, Domchek SM, Miron A, Garber J, Wang M, Componeschi K, Boghossian L, Miron PL, Nathanson KL, Tung N. PALB2 mutations in familial breast and pancreatic cancer. Fam Cancer. 2011 March 2; doi: 10.1007/s10689-011-9426-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Stadler ZK, Salo-Mullen E, Sabbaghian N, Simon JA, Zhang L, Olson SH, Kurtz R, Offit K, Foulkes WD, Robson ME, Tischkowitz M. Germline PALB2 mutation analysis in breast-pancreas cancer families. J Med Genet. 2011 Mar 17; doi: 10.1136/jmg.2010.087379. [DOI] [PubMed] [Google Scholar]