Advertisement
  • Loading metrics

Genetic anticipation in Swedish Lynch syndrome families

  • Jenny von Salomé ,

    Roles Data curation, Investigation, Project administration, Resources, Writing – original draft, Writing – review & editing

    jenny.vonsalome@sll.se

    Affiliation Department of Molecular Medicine and Surgery, Karolinska Institutet, and Department of Clinical Genetics, Karolinska University Hospital, Solna, Stockholm, Sweden

  • Philip S. Boonstra,

    Roles Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing – original draft

    Affiliation Department of Biostatistics, University of Michigan, Ann Arbor, Michigan, United States of America

  • Masoud Karimi,

    Roles Data curation, Investigation, Resources, Writing – review & editing

    Affiliation Department of Oncology, Radiumhemmet, Karolinska University Hospital, Solna, Stockholm, Sweden

  • Gustav Silander,

    Roles Resources, Writing – review & editing

    Affiliation Department of Radiation Sciences, Umeå University, Umeå, Sweden

  • Marie Stenmark-Askmalm,

    Roles Resources, Writing – review & editing

    Affiliations Department of Oncology, Linköping University, Linköping, Sweden, Department of Clinical Genetics, Office for Medical Services, Division of Laboratory Medicine, Lund, Sweden

  • Samuel Gebre-Medhin,

    Roles Resources, Writing – review & editing

    Affiliations Department of Clinical Genetics, Office for Medical Services, Division of Laboratory Medicine, Lund, Sweden, Division of Clinical Genetics, Department of Laboratory Medicine, Lund University, Lund, Sweden

  • Christos Aravidis,

    Roles Resources, Writing – review & editing

    Affiliation Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden

  • Mef Nilbert,

    Roles Writing – review & editing

    Affiliations Department of Clinical Sciences, Division of Oncology and Pathology, Lund University, Lund, Sweden, Clinical Research Centre, Hvidovre Hospital, Copenhagen University, Hvidovre, Denmark

  • Annika Lindblom,

    Roles Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing

    Affiliation Department of Molecular Medicine and Surgery, Karolinska Institutet, and Department of Clinical Genetics, Karolinska University Hospital, Solna, Stockholm, Sweden

  • Kristina Lagerstedt-Robinson

    Roles Project administration, Resources, Supervision, Writing – review & editing

    Affiliation Department of Molecular Medicine and Surgery, Karolinska Institutet, and Department of Clinical Genetics, Karolinska University Hospital, Solna, Stockholm, Sweden

Genetic anticipation in Swedish Lynch syndrome families

  • Jenny von Salomé, 
  • Philip S. Boonstra, 
  • Masoud Karimi, 
  • Gustav Silander, 
  • Marie Stenmark-Askmalm, 
  • Samuel Gebre-Medhin, 
  • Christos Aravidis, 
  • Mef Nilbert, 
  • Annika Lindblom, 
  • Kristina Lagerstedt-Robinson
PLOS
x

Abstract

Among hereditary colorectal cancer predisposing syndromes, Lynch syndrome (LS) caused by mutations in DNA mismatch repair genes MLH1, MSH2, MSH6 or PMS2 is the most common. Patients with LS have an increased risk of early onset colon and endometrial cancer, but also other tumors that generally have an earlier onset compared to the general population. However, age at first primary cancer varies within families and genetic anticipation, i.e. decreasing age at onset in successive generations, has been suggested in LS. Anticipation is a well-known phenomenon in e.g neurodegenerative diseases and several reports have studied anticipation in heritable cancer. The purpose of this study is to determine whether anticipation can be shown in a nationwide cohort of Swedish LS families referred to the regional departments of clinical genetics in Lund, Stockholm, Linköping, Uppsala and Umeå between the years 1990–2013. We analyzed a homogenous group of mutation carriers, utilizing information from both affected and non-affected family members. In total, 239 families with a mismatch repair gene mutation (96 MLH1 families, 90 MSH2 families including one family with an EPCAM–MSH2 deletion, 39 MSH6 families, 12 PMS2 families, and 2 MLH1+PMS2 families) comprising 1028 at-risk carriers were identified among the Swedish LS families, of which 1003 mutation carriers had available follow-up information and could be included in the study. Using a normal random effects model (NREM) we estimate a 2.1 year decrease in age of diagnosis per generation. An alternative analysis using a mixed-effects Cox proportional hazards model (COX-R) estimates a hazard ratio of exp(0.171), or about 1.19, for age of diagnosis between consecutive generations. LS-associated gene-specific anticipation effects are evident for MSH2 (2.6 years/generation for NREM and hazard ratio of 1.33 for COX-R) and PMS2 (7.3 years/generation and hazard ratio of 1.86). The estimated anticipation effects for MLH1 and MSH6 are smaller.

Author summary

Genetic anticipation is a phenomenon where symptoms of a hereditary disease appear at an earlier age and/or are more severe in successive generations. In genetic disorders such as Fragile X syndrome, Myotonic dystrophy type 1 and Huntington disease, anticipation is caused by the expansion of unstable trinucleotide repeats during meiosis. Anticipation is also reported to occur in some hereditary cancers though the underlying mechanism behind this observation is unknown. Several studies have investigated anticipation in Lynch syndrome, the most common hereditary colorectal cancer syndrome, yet there is a debate concerning whether anticipation occurs and what underlying mechanism there is. The objective of this project is to study if anticipation is part of the clinical picture in Swedish families with LS, with the long term goal to enable better prediction of age at onset in family members. Our results suggest that anticipation occurs in families with mutation in MSH2 and PMS2, while the evidence is equivocal for MLH1 and MSH6.

Introduction

Lynch syndrome (LS) is an autosomal dominant inherited syndrome that increases the risk of cancer, primarily in the colon, the rectum and the endometrial lining of the uterus, and to a lesser degree also in the stomach, the ovary, the hepatobiliary tract, the urinary tract, the small bowel and the brain [1,2]. LS is one of the most common heritable cancer syndromes, accounting for up to 4% of the total colorectal cancer burden in Europe, where patients have up to 70% lifetime risk of developing colorectal or endometrial cancer at an early age [1]. LS was formerly known as hereditary non-polyposis colorectal cancer (HNPCC), but when clinical criteria evolved to take into account not only colorectal cancer to identify families with LS [3,4] the name Lynch Syndrome became generally accepted [5]. Today the diagnosis LS is restricted to families with a known pathogenic germline mutation in one of the mismatch repair (MMR) genes MLH1, MSH2, MSH6 and PMS2 irrespective of family history [6,7]. The MMR system corrects indels or mismatches in the DNA, and is evolutionary conserved from bacteria to human [8]. In human the recognition of nucleotide mismatches is mediated by the protein heterodimers MSH2/MSH6 or MSH2/MSH3, while the removal and resynthesis of nucleotides is mediated by MLH1/PMS2 [9].

LS is a heterogeneous disease with regard to tumor spectrum and age at onset [10]. Part of this phenotypic variation has been linked to specific MMR gene mutation. For instance, MLH1 mutation carriers are suggested to have a higher risk for colorectal cancer (CRC) and earlier age of onset, compared to MSH2 and MSH6 mutation carriers [1115]. In general, MSH6 mutation carriers tend to have a later age of onset and lower penetrance for LS associated tumors, apart from endometrial cancer, compared to MLH1 and MSH2 mutation carriers [1620]. An older age of onset and a lower overall risk for CRC has also been suggested for PMS2 mutation carriers [21,22]. However, LS display phenotypic variation in age of onset also within families and between families with the same mutation [2325]. This variation is attributed to individual genetic differences modifying the effect of an inherited MMR mutation [2631]. Another factor proposed to influence age at onset is genetic anticipation, defined as progressive earlier onset and severity of disease in successive generations within a family. This phenomenon is closely related to the disease mechanisms in several genetic disorders, e.g the neurodegenerative diseases Fragile X syndrome, Myotonic dystrophy type 1 and Huntington disease, in which trinucleotide repeat expansion directly influence expressivity and penetrance of disease [32]. Anticipation has also been observed in hereditary cancer for example familial melanoma, Li-Fraumeni syndrome, breast, ovarian and pancreatic cancer, and recently in the renal cell cancer syndromes von Hippel-Lindau and HLRCC (hereditary leiomyomatosis and renal cell cancer) [3339]. In LS, a progressive decrease of age at CRC onset was proposed already in 1925 when the syndrome was first described [40,41]. However, it is complicated to estimate genetic anticipation and there are contradictory reports regarding its existence in LS, though the majority indicate anticipation [4252]. Previous studies have applied various statistical methodologies, compiled different mutations and included subjects with LS associated mutations as well as subjects with only a clinical diagnosis. In light of these studies, we analyzed affected and unaffected mutation carriers in LS families throughout Sweden, to investigate signs of anticipation using two regression models with adjustment for potential confounders, including gene-specific effects.

Materials and methods

In Sweden, families with suspected LS are referred to the regional department of Clinical genetics in Umeå, Uppsala, Stockholm, Linköping, Göteborg or Lund, for counceling and genetic testing. Out of this population-based cohort, families identified with a LS-associated MMR mutation that received genetic counseling in Lund, Stockholm, Linköping, Uppsala or Umeå between January 1990 and December 2013 were enrolled in this study. This project was approved in accordance with the Swedish legislation of ethical permission (2003:469). All patients provided oral or written informed consent for genetic diagnostics as part of their routine clinical care. This anonymized genetic information may be used for research without further consent sought from the patients if approved by an ethical review board. Accordingly, this study was approved by the Regional ethical review board in Stockholm (dnr 2014/1320-31).

Patient and family cancer history was reported at the time of genetic counseling and cancer diagnoses were further confirmed from medical records or pathology reports. A total of 239 families with proven pathogenic MMR variants described in [53] (96 MLH1 families, 90 MSH2 families including one EPCAM-deletion family, 39 MSH6 families, 12 PMS2 families, and 2 MLH1+PMS2 families), comprising 1029 mutation carriers, were identified in the cohort. One individual whose parents were both mutation carriers was excluded. Additionally, the sex of 11 carriers was unknown, and the age at diagnosis for an additional 14 was missing. We excluded these 25 individuals, leaving 1003 at-risk carriers with available follow-up information and sufficient pathological and medical information to be included in the study. We grouped the EPCAM-deletion family within the MSH2 families, as it is reported that a partly deleted EPCAM gene (located upstream of MSH2) cause LS through reducing the expression of MSH2 in EPCAM-expressing tissues [54]. For statistical modeling purposes, we counted two families with mutations in both MLH1 and PMS2 as having mutations in PMS2 only (unreported auxiliary analyses that excluded these families altogether or counted them as MLH1 showed that our findings are not sensitive to this decision). The follow-up period was defined as the time from birth until age at onset, and for individuals who were diagnosed with multiple Lynch-related cancers, age of onset was recorded as the time of first diagnosis. Our first analytic approach was the normal random effects model (NREM) proposed by Larsen et al. [45], which has been used previously to test for anticipation in LS [43] and BRCA-mutation related cancers [55]. Let ni denote the number of carriers in the ith family, i = 1, 2, , 239, and let j = 1, 2, , ni index the jth individual in family i. The NREM is given by (1) where Tij is the age of diagnosis in years for the jth member of family i (“person ij”), μi is the family-specific random intercept representing a typical age of diagnosis in the ith family, Zij is person ij’s generation (coded with respect to oldest observed generation in each family, as described in [41] and γ, the parameter of interest, is the mean change in age of diagnosis between consecutive generations, i.e. the anticipation effect. In the NREM, anticipation is indicated if γ < 0. Collectively the Xij term represents any other covariate(s) of interest for person ij, the effect of which is given by β. The final term εij is the residual error, assumed to be independently and normally distributed with mean zero and variance σ2. For each person who was not diagnosed with a Lynch-associated cancer during the follow-up period, the likelihood contribution is given by the normal survivor function, that is, the probability of being cancer-free at the age of last follow-up. We assume that the censoring mechanism is independent of the time to cancer diagnosis.

Our second analytic approach, which is also a regression strategy, extends the Cox proportional hazards model that was used in [56] to test for anticipation in lymphoproliferative tumors. Person ij’s hazard for cancer diagnosis at age t is modeled as: (2)

The function λ0(t) is the overall baseline hazard function. In Daugherty et al., the baseline hazard was assumed to be identical for all families, that is, was not included in the model and within-family correlations were accounted for by robust sandwich-type covariance estimates. Here we add a random family-level effect , similar to NREM, which makes the less restrictive assumption that all families’ baseline hazards are proportional to, rather than equal to, one another. We call this Cox model with family-level random effect COX-R. The remaining parameters are analogous to NREM. Specifically, gives the generational effect of anticipation as a log-hazard ratio, with indicating anticipation, and is the log-hazard ratio(s) for all other covariates.

In addition to adjusting for sex, we also included mutational status in NREM and COX-R. In Eqs (1) and (2), let person ij’s length-4 vector of covariates be given by Xij = {1[sexij = male], 1[genei = MSH2], 1[genei = MSH6], 1[genei = PMS2]}, where 1[y] is the indicator function, equal to 1 if y is true, sexij is the sex of person ij and genei is the mutational status of family i. MLH1 serves as the reference category.

We also investigated whether there were gene-specific effects of anticipation by substituting Zij in Eqs (1) and (2) with the four dimensional covariate vector.

All analysis was done in the R software package R Core Team [57]. Code for maximizing the integrated partial likelihood of model (2), marginalized over the random effects , is provided in the R package COXME [58].

Results

Table 1 presents the clinical characteristics of our data and Fig 1 plots the Kaplan-Meier estimate of the time to first Lynch-associated cancer diagnosis, to give an overview of the age at onset in our cohort. During the follow-up period, 719 carriers were diagnosed with at least one Lynch-associated cancer and 171 were diagnosed with multiple Lynch-associated cancers. Overall, the median age of first diagnosis was 51 years (95% CI: 50–53), but this varied with mutational status, being 49 years in both MLH1 and MSH2 patients and 58 and 67 years, respectively, for MSH6 and PMS2 patients.

thumbnail
Fig 1. Kaplan-Meier estimates of time to first Lynch-associated cancer diagnosis in the Swedish Registry data, both overall and stratified by mutational status.

The x axis displays age (years) and the y axis displays the probability of being free from diagnosis.

https://doi.org/10.1371/journal.pgen.1007012.g001

Based on this, in addition to adjusting for sex, we also included mutational status in NREM and COX-R analyses. Table 2 gives the estimates, standard errors, and Wald-type p-values for the anticipation effects only, and Table 3 provides all parameter estimates. As shown in Table 2, the estimates of γ (NREM) and (COX-R) are -2.1 (p = 0.0001) and 0.171 (p = 0.0013), respectively. Both suggest the presence of anticipation: a 2.1 year decrease in the age of diagnosis per generation and a hazard ratio of exp(0.171) = 1.19 between consecutive generations. Because mutational status appears to confound the age of diagnosis, we also investigated whether there were gene-specific effects of anticipation, yielding one estimated anticipation effect for each analyzed MMR-gene. These are given in the bottom rows of Table 2. In NREM, anticipation is estimated to be -1.8 (p = 0.044), -2.6 (p = 0.003), -1.1 (p = 0.366), and -7.3 (p = 0.014) years per generation respectively, for MLH1, MSH2, MSH6, and PMS2. In COX-R, the corresponding log-hazard ratios are 0.127 (p = 0.133), 0.284 (p = 0.001), -0.005 (p = 0.965), and 0.618 (p = 0.052), representing hazard ratios of 1.13, 1.33, 0.99, and 1.86, respectively. In both models, the confidence intervals (CIs) for the anticipation effects of MSH2 and PMS2 lie far from their null values (Table 2), whereas there is greater uncertainty with regard to any possible effect of anticipation in MLH1 and MSH6.

thumbnail
Table 2. Results corresponding to the anticipation effects for the NREM and COX-R models, without and with the gene-generation interactions (Est = Estimate, CI = 95% Confidence Interval).

https://doi.org/10.1371/journal.pgen.1007012.t002

thumbnail
Table 3. Complete results for the NREM and COX-R models, without and with the gene-generation interactions (Est = Estimate, CI = 95% Confidence Interval; SE = standard error).

https://doi.org/10.1371/journal.pgen.1007012.t003

Discussion

We investigated signs of anticipation in LS through the analysis of a large, Swedish population-based cohort and regression analyses suggest that anticipation exists in these families. The NREM analysis suggests that the age of diagnosis in families is decreasing by about 2 years per generation, and the COX-R analysis suggests a multiplicative increase in the rate of diagnosis of about 1.19 between generations. These regression analyses carry at least two important advantages over hypothesis testing approaches that compare the age of diagnosis between all parent-child pairs. First, they make use of the partial follow-up time from all at-risk carriers who have not yet been diagnosed; these individuals would otherwise be excluded from analysis. Second, the model-based structure allows for straightforward incorporation of genetic effects or other possible confounders. The underlying causal mutation evidently plays a role in the extent of anticipation, as our estimates varied between MMR genes. Among the MMR genes, the ordering of estimated anticipation effects was PMS2, MSH2, MLH1, and MSH6, with the largest effect in PMS2 (7.3 years/generation [NREM] or a hazard ratio of exp(0.618) = 1.86 [COX-R]). Although the small number of PMS2 families yielded correspondingly large uncertainty, these effects were still highly significant therefore this uncertainty does not invalidate the findings. For MSH2, the estimated effect of anticipation was 2.6 years/generation or a hazard ratio of exp(0.284) = 1.33.

The results are comparable to those reported in several earlier studies (for a review, see [41]). In an analysis of Lynch families from the Danish HNPCC Registry [45], an anticipation effect of about three years/generation was reported but no differences between mutational status was found. A version of the same data was considered again in Boonstra, et al. [43], who fit variants of both regression models considered here, reporting effects of 3.3 years/generation and hazard ratios of exp(0.22) = 1.25. Neither model in that study adjusted for mutational status. Also, the Cox model did not include family-level random effects, as we do here; our approach is arguably a more accurate, although still simplified, reproduction of the true underlying hazard process. A later report analyzed the same Danish HNPCC Registry data with Bayesian modeling techniques [59], allowing anticipation to be random between families. They estimated population-level gene-specific effects of anticipation, as performed here, for MLH1, MSH2, and MSH6. They found respective anticipation effects of 2.8, 2.5, and 1.0 years, consistent with our findings. Several other studies based on anecdotal observation or analyses of affected parent-child pairs have found effects of anticipation varying between 5.5 and 10 years [44,47,49,50]. A notable exception from previous studies is Tsai, et al. [46] who found no evidence for anticipation in 475 parent-child pairs from the Johns Hopkins Hereditary Colorectal Cancer Registry; in part this may be explained by differences in eligibility as only 14 of the 475 parent-child pairs analyzed had verified disease-predispoing germline MMR gene mutations.

The underlying mechanism for anticipation in heritable cancer is still unknown. However, it has been proposed that anticipation is caused by a progressive accumulation of germline mutations, due to the reduced DNA mismatch repair ability in mutation carriers [51]. Accordingly, haploid/monoallelic mutations in the MMR system affect the mutation load in the carrier prior to loss of the second allele, and accumulated alterations in the germ cells is transferred to the offspring [41]. Interestingly, there is an overrepresentation of mononucleotide repeats within and around the human MMR genes compared to other genomic regions, with an overrepresentation in the PMS2 gene [60,61]. It has been suggested that MMR proteins maintain the length of such microsatellites present within their own nucleotide sequences by an evolutionary mechanism operating by gene-protein interactions [60]. With the above arguments a deficient MMR system would propagate errors through generations and this would be most significant for mutations in the PMS2 gene, which is in accordance with our results. In addition, it has been shown that PMS2-deficient mice eggs forms embryos with an increased mononucleotide mutation rate, indicating that MMR mutations might affect germline mutation rate in a heterozygous state [62]. This also points to our results that PMS2 mutations carriers would display the most anticipation if the mutation load is inherited by the next generation.

Noteworthy, PMS2 and MSH2 are not part of the same protein complex involved in recognition, excision and resynthesis of mismatched nucleotides [63], nor does the MSH2 gene contain the same magnitude of mononucleotide repeats as PMS2 [60]. This together suggests a different underlying mechanism generating anticipation in MSH2 mutation carriers. For example, it is shown that MMR deficiency affect telomere shortening in human fibroblasts, and that this might influence heterozygous carriers of a MSH2 mutation in particular [64]. Moreover, in a recent study telomere shortening correlated significantly with age at onset in the MSH2 carriers, whereas the MLH1 carriers displayed longer telomeres and delayed age at onset [65]. Nevertheless, MMR mutation carriers with LS-associated cancer may have specific telomere-length dynamics but telomere shortening does not alone explain anticipation, as reported by Segui et al [66], indicating that gene-specific dynamics and different mechanisms are involved.

Despite a general concurrence with earlier studies, several caveats accompany our findings. Evidently, our study and previously published evidence that performed survival analysis for genetic anticipation in LS suggests that if genetic anticipation does exist, the effect is modest [42,43,45]. This makes anticipation a difficult problem statistically and challenges some of the clinical utility of our findings. At the population level, anticipation may well also be due to reasons other than genetic. For example, cohort effects arising from changes in treatment, diagnostic or environmental factors can also result in a decline in age at diagnosis. These effects should be visible both within family trees and in the entire population (which is a mix of mutation carriers from different generations). This is in contrast to genetic anticipation, which would only be seen within each unique family tree.

Voskuil, et al. found that the hazard ratio corresponding to generation decreased considerably in magnitude after adjusting for birth cohort [42], although their final estimated hazard ratios for the effect of generation were still very close to our estimate of 1.2. Statistically, birth cohort and generation are typically highly correlated, which can cause the resulting parameter estimates to be unstable. Boonstra, et al. [59] attempted to disentangle these effects by independently estimating secular trends in age of colorectal cancer diagnosis from a cancer registry of all colorectal cancers, and adjusting the Danish HNPCC Registry data for this estimated trend before analysis. Still, the results indicated as reported earlier in this section, population-level gene-specific effects of anticipation of 2.8, 2.5, and 1.0 years, respectively, for MLH1, MSH2, and MSH6. Our estimated effects of anticipation decrease by about 0.7 years when we directly apply the secular trends estimated in Boonstra, et al. [54].

Furthermore, it has been argued that anticipation may be falsely detected due to fecundity bias [48]. Through repeated simulations of parent-child pairs in which no anticipation exists (in truth) but the fertility rate was positively correlated with age of diagnosis, Stupart, et al. demonstrated in a particular scenario that an apparent anticipation effect of about 1.8 years can manifest. Noteworthy, the greatest reduction in fertility was predominantly among those diagnosed before age 29, affecting the fertility of the cohort as a whole. In our cohort, the Kaplan-Meier estimated proportion of patients free of diagnosis at age 29 was 96.5%, which suggests that fecundity bias due to these patients is likely to be small.

Taken together, our findings are in line with those of previous studies. That being said, the study of genetic anticipation is both complex and statistically challenging. The ideal setting in the continuing assessment of fine variations in LS phenotype, such as anticipation, would be prospective, population-based datasets, together with state-of-the-art statistical methods. Still, a number of promising findings have been reported previously, yet often the statistical methods or small sample sizes have been limiting. We believe that the analyses performed in our study properly consider familial, genetic, and clinical parameters and therefore give a representative measurement of anticipation in Lynch Syndrome.

Acknowledgments

We thank the regional Oncogenetic Clinics in Lindköping, Lund, Stockholm, Umeå and Uppsala for contributing with patient data.

References

  1. 1. Vasen HF, Blanco I, Aktan-Collan K, Gopie JP, Alonso A, Aretz S, et al. (2013) Revised guidelines for the clinical management of Lynch syndrome (HNPCC): recommendations by a group of European experts. Gut 62: 812–823. pmid:23408351
  2. 2. Watson P, Riley B (2005) The tumor spectrum in the Lynch syndrome. Fam Cancer 4: 245–248. pmid:16136385
  3. 3. Vasen HF, Watson P, Mecklin JP, Lynch HT (1999) New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology 116: 1453–1456. pmid:10348829
  4. 4. Umar A, Boland CR, Terdiman JP, Syngal S, de la Chapelle A, Ruschoff J, et al. (2004) Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst 96: 261–268. pmid:14970275
  5. 5. Jass JR (2006) Hereditary Non-Polyposis Colorectal Cancer: the rise and fall of a confusing term. World J Gastroenterol 12: 4943–4950. pmid:16937488
  6. 6. Boland CR (2005) Evolution of the nomenclature for the hereditary colorectal cancer syndromes. Fam Cancer 4: 211–218. pmid:16136380
  7. 7. Ligtenberg MJ, Kuiper RP, Geurts van Kessel A, Hoogerbrugge N (2013) EPCAM deletion carriers constitute a unique subgroup of Lynch syndrome patients. Fam Cancer 12: 169–174. pmid:23264089
  8. 8. Modrich P (2006) Mechanisms in eukaryotic mismatch repair. J Biol Chem 281: 30305–30309. pmid:16905530
  9. 9. Kloor M, Staffa L, Ahadova A, von Knebel Doeberitz M (2014) Clinical significance of microsatellite instability in colorectal cancer. Langenbecks Arch Surg 399: 23–31. pmid:24048684
  10. 10. You YN (2015) Anticipating phenotypic differences from genetic mutations. Dis Colon Rectum 58: 143–144. pmid:25489706
  11. 11. Plaschke J, Engel C, Kruger S, Holinski-Feder E, Pagenstecher C, Mangold E, et al. (2004) Lower incidence of colorectal cancer and later age of disease onset in 27 families with pathogenic MSH6 germline mutations compared with families with MLH1 or MSH2 mutations: the German Hereditary Nonpolyposis Colorectal Cancer Consortium. J Clin Oncol 22: 4486–4494. pmid:15483016
  12. 12. Kastrinos F, Stoffel EM, Balmana J, Steyerberg EW, Mercado R, Syngal S (2008) Phenotype comparison of MLH1 and MSH2 mutation carriers in a cohort of 1,914 individuals undergoing clinical genetic testing in the United States. Cancer Epidemiol Biomarkers Prev 17: 2044–2051. pmid:18708397
  13. 13. Goecke T, Schulmann K, Engel C, Holinski-Feder E, Pagenstecher C, Schackert HK, et al. (2006) Genotype-phenotype comparison of German MLH1 and MSH2 mutation carriers clinically affected with Lynch syndrome: a report by the German HNPCC Consortium. J Clin Oncol 24: 4285–4292. pmid:16908935
  14. 14. Vasen HF, Stormorken A, Menko FH, Nagengast FM, Kleibeuker JH, Griffioen G, et al. (2001) MSH2 mutation carriers are at higher risk of cancer than MLH1 mutation carriers: a study of hereditary nonpolyposis colorectal cancer families. J Clin Oncol 19: 4074–4080. pmid:11600610
  15. 15. Hampel H, Stephens JA, Pukkala E, Sankila R, Aaltonen LA, Mecklin JP, et al. (2005) Cancer risk in hereditary nonpolyposis colorectal cancer syndrome: later age of onset. Gastroenterology 129: 415–421. pmid:16083698
  16. 16. Wagner A, Hendriks Y, Meijers-Heijboer EJ, de Leeuw WJ, Morreau H, Hofstra R, et al. (2001) Atypical HNPCC owing to MSH6 germline mutations: analysis of a large Dutch pedigree. J Med Genet 38: 318–322. pmid:11333868
  17. 17. Peterlongo P, Nafa K, Lerman GS, Glogowski E, Shia J, Ye TZ, et al. (2003) MSH6 germline mutations are rare in colorectal cancer families. Int J Cancer 107: 571–579. pmid:14520694
  18. 18. Kolodner RD, Tytell JD, Schmeits JL, Kane MF, Gupta RD, Weger J, et al. (1999) Germ-line msh6 mutations in colorectal cancer families. Cancer Res 59: 5068–5074. pmid:10537275
  19. 19. Baglietto L, Lindor NM, Dowty JG, White DM, Wagner A, Gomez Garcia EB, et al. (2010) Risks of Lynch syndrome cancers for MSH6 mutation carriers. J Natl Cancer Inst 102: 193–201. pmid:20028993
  20. 20. Wijnen J, de Leeuw W, Vasen H, van der Klift H, Moller P, Stormorken A, et al. (1999) Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat Genet 23: 142–144. pmid:10508506
  21. 21. Gala M, Chung DC (2011) Hereditary colon cancer syndromes. Semin Oncol 38: 490–499. pmid:21810508
  22. 22. Jasperson KW, Tuohy TM, Neklason DW, Burt RW (2010) Hereditary and familial colon cancer. Gastroenterology 138: 2044–2058. pmid:20420945
  23. 23. Lynch HT, Smyrk TC, Watson P, Lanspa SJ, Lynch JF, Lynch PM, et al. (1993) Genetics, natural history, tumor spectrum, and pathology of hereditary nonpolyposis colorectal cancer: an updated review. Gastroenterology 104: 1535–1549. pmid:8482467
  24. 24. Scott RJ, McPhillips M, Meldrum CJ, Fitzgerald PE, Adams K, Spigelman AD, et al. (2001) Hereditary nonpolyposis colorectal cancer in 95 families: differences and similarities between mutation-positive and mutation-negative kindreds. Am J Hum Genet 68: 118–127. pmid:11112663
  25. 25. Rebbeck TR, Martinez ME, Sellers TA, Shields PG, Wild CP, Potter JD (2004) Genetic variation and cancer: improving the environment for publication of association studies. Cancer Epidemiol Biomarkers Prev 13: 1985–1986. pmid:15598750
  26. 26. Reeves SG, Rich D, Meldrum CJ, Colyvas K, Kurzawski G, Suchy J, et al. (2008) IGF1 is a modifier of disease risk in hereditary non-polyposis colorectal cancer. Int J Cancer 123: 1339–1343. pmid:18623088
  27. 27. Kruger S, Silber AS, Engel C, Gorgens H, Mangold E, Pagenstecher C, et al. (2005) Arg462Gln sequence variation in the prostate-cancer-susceptibility gene RNASEL and age of onset of hereditary non-polyposis colorectal cancer: a case-control study. Lancet Oncol 6: 566–572. pmid:16054567
  28. 28. Shi Z, Johnstone D, Talseth-Palmer BA, Evans TJ, Spigelman AD, Groombridge C, et al. (2009) Haemochromatosis HFE gene polymorphisms as potential modifiers of hereditary nonpolyposis colorectal cancer risk and onset age. Int J Cancer 125: 78–83. pmid:19291797
  29. 29. Chen J, Pande M, Huang YJ, Wei C, Amos CI, Talseth-Palmer BA, et al. (2013) Cell cycle-related genes as modifiers of age of onset of colorectal cancer in Lynch syndrome: a large-scale study in non-Hispanic white patients. Carcinogenesis 34: 299–306. pmid:23125224
  30. 30. Pande M, Amos CI, Osterwisch DR, Chen J, Lynch PM, Broaddus R, et al. (2008) Genetic variation in genes for the xenobiotic-metabolizing enzymes CYP1A1, EPHX1, GSTM1, GSTT1, and GSTP1 and susceptibility to colorectal cancer in Lynch syndrome. Cancer Epidemiol Biomarkers Prev 17: 2393–2401. pmid:18768509
  31. 31. Bellido F, Guino E, Jagmohan-Changur S, Segui N, Pineda M, Navarro M, et al. (2013) Genetic variant in the telomerase gene modifies cancer risk in Lynch syndrome. Eur J Hum Genet 21: 511–516. pmid:22948024
  32. 32. Plassart E, Fontaine B (1994) Genes with triplet repeats: a new class of mutations causing neurological diseases. Biomed Pharmacother 48: 191–197. pmid:7999979
  33. 33. Goldstein AM, Clark WH Jr., Fraser MC, Tucker MA (1996) Apparent anticipation in familial melanoma. Melanoma Res 6: 441–446. pmid:9013482
  34. 34. Trkova M, Hladikova M, Kasal P, Goetz P, Sedlacek Z (2002) Is there anticipation in the age at onset of cancer in families with Li-Fraumeni syndrome? J Hum Genet 47: 381–386. pmid:12181637
  35. 35. Martinez-Borges AR, Petty JK, Hurt G, Stribling JT, Press JZ, Castellino SM (2009) Familial small cell carcinoma of the ovary. Pediatr Blood Cancer 53: 1334–1336. pmid:19621450
  36. 36. Schneider R, Slater EP, Sina M, Habbe N, Fendrich V, Matthai E, et al. (2011) German national case collection for familial pancreatic cancer (FaPaCa): ten years experience. Fam Cancer 10: 323–330. pmid:21207249
  37. 37. Ning XH, Zhang N, Li T, Wu PJ, Wang X, Li XY, et al. (2014) Telomere shortening is associated with genetic anticipation in Chinese Von Hippel-Lindau disease families. Cancer Res 74: 3802–3809. pmid:24986515
  38. 38. Wong MH, Tan CS, Lee SC, Yong Y, Ooi AS, Ngeow J, et al. (2014) Potential genetic anticipation in hereditary leiomyomatosis-renal cell cancer (HLRCC). Fam Cancer 13: 281–289. pmid:24526232
  39. 39. Guindalini RS, Song A, Fackenthal JD, Olopade OI, Huo D (2016) Genetic anticipation in BRCA1/BRCA2 families after controlling for ascertainment bias and cohort effect. Cancer 122: 1913–1920. pmid:26992017
  40. 40. Warthin AS (1925) The Further Study of a Cancer Family. The Journal of Cancer Research 9: 279–286.
  41. 41. Bozzao C, Lastella P, Stella A (2011) Anticipation in lynch syndrome: where we are where we go. Curr Genomics 12: 451–465. pmid:22547953
  42. 42. Voskuil DW, Vasen HF, Kampman E, van't Veer P (1997) Colorectal cancer risk in HNPCC families: development during lifetime and in successive generations. National Collaborative Group on HNPCC. Int J Cancer 72: 205–209. pmid:9219821
  43. 43. Boonstra PS, Gruber SB, Raymond VM, Huang SC, Timshel S, Nilbert M, et al. (2010) A review of statistical methods for testing genetic anticipation: looking for an answer in Lynch syndrome. Genet Epidemiol 34: 756–768. pmid:20878717
  44. 44. Nilbert M, Timshel S, Bernstein I, Larsen K (2009) Role for genetic anticipation in Lynch syndrome. J Clin Oncol 27: 360–364. pmid:19075283
  45. 45. Larsen K, Petersen J, Bernstein I, Nilbert M (2009) A parametric model for analyzing anticipation in genetically predisposed families. Stat Appl Genet Mol Biol 8: Article26.
  46. 46. Tsai YY, Petersen GM, Booker SV, Bacon JA, Hamilton SR, Giardiello FM (1997) Evidence against genetic anticipation in familial colorectal cancer. Genet Epidemiol 14: 435–446. pmid:9271715
  47. 47. Westphalen AA, Russell AM, Buser M, Berthod CR, Hutter P, Plasilova M, et al. (2005) Evidence for genetic anticipation in hereditary non-polyposis colorectal cancer. Hum Genet 116: 461–465. pmid:15772852
  48. 48. Stupart D, Goldberg P, Algar U, Vorster A, Ramesar R (2014) No evidence of genetic anticipation in a large family with Lynch syndrome. Fam Cancer 13: 29–34. pmid:23771324
  49. 49. Vasen HF, Taal BG, Griffioen G, Nagengast FM, Cats A, Menko FH, et al. (1994) Clinical heterogeneity of familial colorectal cancer and its influence on screening protocols. Gut 35: 1262–1266. pmid:7959234
  50. 50. Rodriguez-Bigas MA, Lee PH, O'Malley L, Weber TK, Suh O, Anderson GR, et al. (1996) Establishment of a hereditary nonpolyposis colorectal cancer registry. Dis Colon Rectum 39: 649–653. pmid:8646951
  51. 51. Stella A, Surdo NC, Lastella P, Barana D, Oliani C, Tibiletti MG, et al. (2007) Germline novel MSH2 deletions and a founder MSH2 deletion associated with anticipation effects in HNPCC. Clin Genet 71: 130–139. pmid:17250661
  52. 52. Ponti G, Ruini C, Tomasi A (2015) Mismatch repair gene deficiency and genetic anticipation in Lynch syndrome: myth or reality? Dis Colon Rectum 58: 141–142. pmid:25489705
  53. 53. Lagerstedt-Robinson K, Rohlin A, Aravidis C, Melin B, Nordling M, Stenmark-Askmalm M, et al. (2016) Mismatch repair gene mutation spectrum in the Swedish Lynch syndrome population. Oncol Rep 36: 2823–2835. pmid:27601186
  54. 54. Ligtenberg MJ, Kuiper RP, Chan TL, Goossens M, Hebeda KM, Voorendt M, et al. (2009) Heritable somatic methylation and inactivation of MSH2 in families with Lynch syndrome due to deletion of the 3' exons of TACSTD1. Nat Genet 41: 112–117. pmid:19098912
  55. 55. Litton JK, Ready K, Chen H, Gutierrez-Barrera A, Etzel CJ, Meric-Bernstam F, et al. (2012) Earlier age of onset of BRCA mutation-related cancers in subsequent generations. Cancer 118: 321–325. pmid:21913181
  56. 56. Daugherty SE, Pfeiffer RM, Mellemkjaer L, Hemminki K, Goldin LR (2005) No evidence for anticipation in lymphoproliferative tumors in population-based samples. Cancer Epidemiol Biomarkers Prev 14: 1245–1250. pmid:15894680
  57. 57. Team RC (2014) R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.
  58. 58. Therneau TM (2015) coxme: Mixed Effects Cox Models. pp. R package version 2.2–4.
  59. 59. Boonstra PS, Mukherjee B, Taylor JM, Nilbert M, Moreno V, Gruber SB (2011) Bayesian modeling for genetic anticipation in presence of mutational heterogeneity: a case study in Lynch syndrome. Biometrics 67: 1627–1637. pmid:21627626
  60. 60. Falster DS, Nakken S, Bergem-Ohr M, Rodland EA, Breivik J (2010) Unstable DNA repair genes shaped by their own sequence modifying phenotypes. J Mol Evol 70: 266–274. pmid:20213140
  61. 61. Senter L, Clendenning M, Sotamaa K, Hampel H, Green J, Potter JD, et al. (2008) The clinical phenotype of Lynch syndrome due to germ-line PMS2 mutations. Gastroenterology 135: 419–428. pmid:18602922
  62. 62. Larson JS, Stringer SL, Stringer JR (2004) Impact of mismatch repair deficiency on genomic stability in the maternal germline and during early embryonic development. Mutat Res 556: 45–53. pmid:15491631
  63. 63. Li G-M (2008) Mechanisms and functions of DNA mismatch repair. Cell Res 18: 85–98. pmid:18157157
  64. 64. Mendez-Bermudez A, Royle NJ (2011) Deficiency in DNA mismatch repair increases the rate of telomere shortening in normal human cells. Hum Mutat 32: 939–946. pmid:21538690
  65. 65. Bozzao C, Lastella P, Ponz de Leon M, Pedroni M, Di Gregorio C, D'Ovidio FD, et al. (2011) Analysis of telomere dynamics in peripheral blood cells from patients with Lynch syndrome. Cancer 117: 4325–4335. pmid:21387278
  66. 66. Segui N, Pineda M, Guino E, Borras E, Navarro M, Bellido F, et al. (2013) Telomere length and genetic anticipation in Lynch syndrome. PLoS One 8: e61286. pmid:23637804