Vol 1, No 1 (2013)  ISSN 2167‐8677 (online)  DOI 10.5195/d3000.2013.11     http://dentistry3000.pitt.edu This work is licensed under a Creative Commons Attribution 3.0 United States License.  This site is published by the University Library System, University of Pittsburgh as part of its D‐Scribe Digital Publishing Program and is cosponsored  by the University of Pittsburgh Press.  Non‐syndromic oral clefts and risk of cancer: a systematic review Daniela Araújo Veloso Popoff1, Mateus Prates Coelho2, Daniella Reis Barbosa Martelli1, Rajiv Saini3, Ricardo Della Cole a4, Hercílio Martelli‐ Júnior1   1 Department of Den stry, State University of Montes Claros, Montes Claros, Minas Gerais, Brazil.  2 Department of Medicine, State University of Montes Claros, Montes Claros, Minas Gerais, Brazil.  3 Department of Periodontology and Oral Implantology, Rural Dental College, Ahmednagar, Maharashtra, India.  4 Department of Oral Diagnosis, Piracicaba Dental School, University of Campinas, Piracicaba, São Paulo, Brazil.  Abstract Objec ve: To discuss the risk of cancer among rela ves of individuals with cle  lip and pal‐ ate (CL/P), as well as the risk of CL/P among rela ves of individuals with cancer, since stud‐ ies currently published  have suggested  an increased risk of cancer among rela ves of indi‐ viduals with cle s.  Methods: A systema c  literature review was carried out  in accordance with the Cochrane  Collabora on  Group  protocol,  including  literature  search  strategy,  selec on  of  papers  through  the  inclusion and exclusion criteria, data extrac on and quality assessment. Pub‐ Med, Scopus and ISI ‐ Web of Science databases were systema cally searched using the fol‐ lowing search strings: “cle   lip and palate” AND “cancer,” “oral cle s” AND “cancer” and  “orofacial cle s” AND “cancer.”  Results: From 653 studies accessed, eight comprised the final sample: six inves ga ng CL/P  index cases and their family history of cancer and two inves ga ng individuals with cancer  and their family history for CL/P. The sample sizes were not homogeneous. Oral cle s, the  type of cancer and the degree of kinship family were not categorized in all studies. Leuke‐ mia, breast cancer and colon cancer were the most cited types, even as first‐ and second‐ degree rela ves.  Conclusion: An increased risk of cancer among rela ves of individuals with cle s could not  be en rely confirmed. However,  studies with  this  specific purpose  suggest  that first‐ and  second‐degrees  rela ves of  individuals with  cle  have  some  types of  cancer more o en  than unexposed families, highligh ng that future studies should expand their samples to in‐ ves gate possible common molecular mechanisms  that allow  rela ng oral cle s and can‐ cer.  Cita on: Popoff DAV, Coelho MP, Martelli DRB, Saini R,  Cole a RD, and Martelli‐Junior H. (2013) Non‐ syndromic oral cle s and risk of cancer: a systema c  review. Den stry 3000 1:a002 doi:  10.5195/d3000.2013.11  Received: June 24, 2013  Accepted: August 12, 2013  Published: August 23, 2013  Copyright: ©2013 Popoff et al. This is an open‐access  ar cle licensed under a Crea ve Commons A ribu on  3.0 United States License.  Email: danielavelloso@yahoo.com.br Introduc on Orofacial malformations are the most com- mon form of congenital anomalies in the world [1]. Among the orofacial alterations, the most prevalent is the cleft lip with or without cleft palate (CL/P) [2-5], which may occur more commonly in an isolated and non-syndromic form as a speci ic phenotype or, more rarely, composing several associa- tions or syndromes [6,7]. Embryologically, clefts result from primary fusion defects of the craniofacial processes that form the primary and secondary palate in the irst intrauterine trimester [8]. The incidence of CL/P varies according to geographical location, racial and ethnic groups, environmental exposures, and soci- oeconomic status, affecting approximately 1/700 live births with wide variability across geographic origin. Generally, Asian and Amerindian populations have the high- est reported birth prevalence rates, often as high as 1/500, European-derived popula- tions have intermediate prevalence rates at about 1/1000, and African-derived popula- tions have the lowest prevalence rates at about 1/2500 [5,9,10]. According to Vieira (2008) [7], the last dec- ade was crucial to elucidate issues concern- ing the etiology of CL/P when compared to other defects observed at birth. As this is a multifactorial trait, environmental risk fac- tors such as smoking, alcohol, parental age, medications, birth order, interpregnancy interval, and folic acid de iciency are listed as modi iers, and risk factor identi ication is the irst step to better understanding and to prevent such craniofacial changes [7,11,12]. However, current knowledge about the causes of CL/P points particularly towards genetic risk factors, and the study of the pathogenesis of CL/P has provided ample opportunities to identify candidate genes for this disorder and even link them to the occurrence of cancer in relatives of individ- uals born with CL/P, or even in these indi- viduals themselves when at an adult age, supporting the hypothesis that common genetic factors may be present in both con- ditions [13]. Although other genes that have   Non‐syndromic oral cle s and risk of cancer: a systema c review  Vol 1, No 1 (2013)    DOI 10.5195/d3000.2013.11  http://dentistry3000.pitt.edu 2 not been explored yet can be involved in the both conditions, three genes are suspected to underlie such associations: FGF - ibro- blast signaling pathway, CDH1 - epithelial cadherin and AXIN2 - AXIS inhibition pro- tein 2 [14]. Colorectal and breast cancer are the most associated with mutations of these genes [2].Thus, given the growing number of studies published currently suggesting a common etiology for CL/P and cancer and increased risk of cancer among relatives of cleft individuals [2,5,13-18], the current study aimed to discuss the issue through a systematic literature review in order to provide additional evidence of these genetic risks, which will aid in the development of strategies that target more aggressive screening programs and preventive chemi- cal treatments. Methods The present review was carried out in ac- cordance with the Cochrane Collaboration Group protocol for systematic reviews [19], including a literature search strategy, selec- tion of papers through the inclusion and exclusion criteria, data extraction, and qual- ity assessment. Meta-analysis was not pos- sible since selected studies did not observe the same variables, methods, participants, and outcomes, which prevented compari- sons. Literature search strategy Our review was performed in February 2013 in order to obtain literature regarding cleft lip and/or palate and family history of cancer. PubMed, Scopus, and ISI - Web of Science databases were systematically searched using two basic approaches: the search strings “cleft lip and palate” AND “cancer,” “oral clefts” AND “cancer,” and “orofacial clefts” AND “cancer” for studies published up to that time. Selection of papers through the inclusion and exclusion criteria The selection of papers is diagrammatically explained in Figure 1. Of the 653 studies originally found through the literature search strategy, 235 studies were excluded (in non-English language, with no full text available, and non-cross-sectional, preva- lence, cohort, or case-control design), re- maining with 418 papers. Additional studies were found through the references of the potential studies retrieved. Then, all studies involving oral clefts (associated or not with craniofacial syndromes or any anomalies) and cancer were selected. Afterwards, only studies presenting information regarding risk of cancer in relatives of individuals born with non-syndromic oral clefts were included in the present study, as well as articles regarding the frequency of non- syndromic oral clefts among relatives of individuals diagnosed with cancer. The following speci ic inclusion criteria were used to identify relevant articles: (1) Index cases with non-syndromic cleft lip and/or palate, whose relatives were inves- tigated for any type of cancer; (2) Index cases with any type of cancer, whose rela- tives were investigated for non-syndromic cleft lip and/or palate; (3) Index cases com- pared to healthy controls for risk assess- ment; (4) The degree of relationship be- tween index cases and investigated family members is investigated in the study. All studies that included individuals with cleft lip and/or palate associated with other ab- normalities, having syndromic causes or regarding Mendelian traits were excluded. Thus, eight studies were selected. (Figure 1) Data extraction Title and abstract screening was performed by two reviewers (DAVP and HMJ) who worked independently to identify potential- ly relevant papers for which full text publi- cations were retrieved. If, however, there was any discrepancy of opinion, the review- ers reexamined the paper together and ar- rived at a joint inal decision. A standard- ized form was used to extract information, such as author and year of publication of the paper, origin of participants, study design, sample size, type of oral clefts, type of can- cer, degree of kinship of family members and the association between non-syndromic oral clefts and cancer. Quality assessment STROBE methodology (Strengthening the Reporting of Observational Studies in Epi- demiology) [20], considered the guideline for an appropriate assessment of observa- tional studies, was used to perform the quality assessment of the studies selected. The speci ic STROBE items considered were inclusion and exclusion criteria, assessment of exposure and outcome, statistical meth- ods, confounders, bias, and report limita- tion. Results The initial database search identified 653 citations published between 1980 and 2013. After screening, 235 studies were excluded (72 in non-English language, 93 with no full text available, and 70 which type of study was other than those listed in the method- ology), remaining with 418 papers. A second screening was performed on the remaining papers and other studies were excluded, this time for the following rea- sons: 51 articles in duplicity (found in more than one of the databases) and 351 articles that did not report non-syndromic cleft lip and/or palate and risk of cancer. In addi- tion, ive papers listed in the references of the selected ones were included, resulting in 21 papers for the inal screening. Figure 1: Flowchart of the studies included and excluded   Non‐syndromic oral cle s and risk of cancer: a systema c review  Vol 1, No 1 (2013)    DOI 10.5195/d3000.2013.11  http://dentistry3000.pitt.edu 3 Table 2: Most cited types of cancer among rela ves of cle  individuals and cancer index cases.  *The study specifically searched for this type of cancer only.  The full text of these papers was read and 13 were there after excluded for not report- ing risk of cancer in relatives of individuals with non-syndromic cleft lip and/or palate, as well as for not reporting the frequency of non-syndromic cleft lip and/or palate in relatives of individuals diagnosed with can- cer. Thus, eight papers were selected for the systematic review. The characteristics of the eight selected papers are presented in Table 1. Three used a case-control design, two used a cross- sectional/prevalence design and one was a retrospective cohort study. Those reporting frequency of non-syndromic cleft lip and/or palate in relatives of individuals diagnosed with cancer also used a case-control design. The sample sizes and the population inves- tigated varied considerably amongst the studies, not being homogeneous. From those investigating risk of cancer in rela- tives of individuals with non-syndromic cleft lip and/or palate, two were conducted in the United States, two in Denmark, one in Turkey, and one in Latvia. Regarding those reporting frequency of non-syndromic cleft lip and/or palate in relatives of individuals diagnosed with cancer, one was conducted in the United States and one in India (Table 1). Most of studies investigated all types of oral clefts. Type of cancer and the degree were not categorized in all studies of kinship in the family. Despite this, leukemia, testicular cancer, melanoma, colon-rectal cancer, lym- phoma, and breast cancer were the most cited types of cancer (Table 2) in both irst- and second-degree relatives (Table 1). The relationship between oral clefts and cancer within the studies was tested using chi-square and Fisher’s Exact tests [14, 18], Logistic regression [2, 17], Odds ratio [15, 21], Risk ratio [13], and Cox regression [5] (Table 1). All eight studies were submitted to quality assessment using STROBE guide- lines. Inclusion and exclusion criteria were described in all of them (Table 3). The out- comes of two studies provide suggestive evidence that families with individuals di- agnosed with some types of cancer report family history of cleft lip and/or palate more  frequently than families without cases   First author, year [reference] Country of study Study design Exposure Sample size (case/population) Cleft Classification Type of cancer Statistical test Kinship with index cases Outcome Jindal,  2012  India  Case‐control  Individuals  diagnosed  with cancer  142 (85/57)  CL/P  Leukemia  X2 and  Fischer  exact  First‐degree  Increased risk  Taioli,  2010  United  States  Case‐control  Cancer  survivors  338 (168/170)  All  Unspecified  Logistic  regression    First/second‐ degree  Increased risk  Yildirim,  2012  Turkey  Case‐control  Cleft  individuals  319 (158/161)  All  Unspecified  Logistic  regression  First‐degree  Increased risk  Menezes ,  2009  United  States  Case‐control  Cleft  individuals  168 (75/93)  Unspecified  Unspecified  X2 and  Fischer  exact  Not reported  Increased risk  Zhu,  2002  Denmark  Case‐control  Cleft  individuals  41206 (45/41161)  Unspecified  Unspecified  Odds Ratio  First Degree  Increased risk  Vieira,  2012  Latvia  Cross‐ sectional  Cleft  individuals  89  Unspecified  Unspecified  Risk Ratio  First/second/ third‐degree  Increased risk  Dietz,  2012  Denmark  Retrospec‐ tive cohort  Cleft  individuals  1809  All  Breast  Cancer  Cox regression  First‐degree  No association  when all non‐ syndromic  CL/P were  analyzed  together    Steinwachs,  2000  United  States  Cross‐ sectional  Cleft  individuals  196  CL/P and CL  Unspecified  Odds Ratio  First/second‐ degree  No increased  risk  First author, year [reference] Most commonly found types of cancer Types of cancer with increased risk at index cases rela ves Jindal, 2012 Leukemia*    Leukemia  Taioli, 2010 Breast, colon‐rectal, melanoma, tes cular  Tes cular, melanoma, colon‐ rectal  Yildirim, 2012 Unspecified    Unspecified  Menezes, 2009 Unspecified  Unspecified  Zhu, 2002 Breast, melanoma of skin, tes s, cervix uteri,  brain and nervous system    Lymphoma, Leukemia  Vieira, 2012 Unspecified    Unspecified  Dietz, 2012 Breast*    Breast  Steinwachs, 2000 Breast, lung, leukemia, colon, uterine.  None found  Table 1: Main characteris cs of the eight selected papers.    Non‐syndromic oral cle s and risk of cancer: a systema c review  Vol 1, No 1 (2013)    DOI 10.5195/d3000.2013.11  http://dentistry3000.pitt.edu 4 of cancer [2,18]. From studies investigating cancer in relatives of individuals with oral clefts, only two suggest that the irst- and second-degree relatives of the individuals with clefts are not at an increased risk for cancer [5,21] . All studies de ined the expo- sure variables (Table 1). Although all stud- ies had described potential bias/limitations, only three properly reported identi ication of bias/limitations (Table 1). Although all studies had described potential bi- as/limitations, only three properly reported identi ication of bias/limitations (Table 4). DISCUSSION This systematic review assessed available studies suggesting a common etiology for non-syndromic oral clefts and cancer and increased risk of cancer among relatives of cleft individuals, as well as increased risk of non-syndromic oral clefts among relatives of individuals diagnosed with cancer, there- by providing additional evidence of these genetic risks, which could aid in the devel- opment of strategies that target more ag- gressive screening programs and preventive chemical treatments. Our original intention was to integrate the results of included studies through a meta- analysis, however, given the differences in variables, methods, participants and out- comes among the studies, the most suitable and responsible was to address the issue through a systematic review. We demonstrated that there is a large amount of literature distinguishing non- syndromic cleft lip and palate from those cases of clefts associated with other abnor- malities, syndromes, or Mendelian traits. Our indings also demonstrated that the studies published agree that non-syndromic oral clefts are one of the most common hu- man malformations, with an average preva- lence of 1 per 700 or 1,000 live births [ 2,4,5,7,16,17,21-26], as well as that its inci- dence varies according to gender: 2:1 being the ratio of males to females for cleft lip and palate and 1:2 the approximate ratio of male to female for isolated cleft palate [1,10,16, 26,27]. Other common information between the studies is the fact that unilateral clefts are more common than bilateral clefts, and of the unilateral cases of non-syndromic cleft lip and palate, left-sided cleft lips occur more frequently than right-sided cleft lips [ 4,5,,21,23,28,29]. Many studies have also demonstrated that genetic factors may play a role in the cause of non-syndromic oral clefts in addition to certain environmental and/or stochastic factors, meaning that this malformation is a multifactorial trait [2,21,22,26,30,31]. Similarly, studies have shown that cleft lip, with or without cleft palate, is entirely different from isolated cleft palate in both embryological and path- ogenetic standpoints [32,33]. Likewise, many of the studies report rela- tionships between non-syndromic oral clefts and cancer or childhood cancer [14,18,25,34-41]. Nonetheless, the studies that presented information regarding risk of cancer in relatives of individuals born with non-syndromic oral clefts or regarding the frequency of non-syndromic oral clefts First author, year [reference] Inclusion Criteria Exclusion Criteria Jindal, 2012 All subjects were children being cared for at the General Hospital of Chandigarh, India. Cases  were  identified  from  the  Pediatric  Leukemia  Clinic  and  controls were  identified  randomly  from the Allergy/Immunology clinic.    Children with lymphoma were excluded from the  case group and the ones with positive history for  cancer, from the control group.  Taioli, 2010 Case group was composed of patients who were  treated  for cancer  in  the past, who were  considered in remission for at least six months, and who were scheduled for an appointment  for  a  routine  checkup.  Control  group was  composed  by  healthy  people  from  the  general  population who were 18 years of age or older.    Not reported.  Yildirim, 2012 All subjects were patients at the Department of Pedodontics clinic,  Istanbul University, Tur‐ key. Case group was composed of  individuals born with any type of CL/P and controls were  individuals unrelated  to  them. Case group members’  first‐degree  relatives were evaluated,  also.  Case and control siblings were excluded.  Menezes, 2009 All subjects were from white ancestry families from Pittsburgh. Cases were recruited from a  registry  maintained  by  the  Cleft‐Craniofacial  Center,  Children’s  Hospital  of  Pittsburgh  of  UPMC.  Not reported.  Zhu, 2002 Subjects were identified from the Denmark’s Central Population Registry by their kinship with  patients  diagnosed with  cancer  and  registered  at  the  Cancer  Registry.  They  also  ought  to  have been born between 1977 and 1995 in Denmark to mothers with a Danish citizenship.  Patients whose  parents  had  other  children  be‐ fore  1977,  who  were  diagnosed  with  patent  ductus  arteriosus,  undescended  testis  and  hip  dislocation.  Vieira, 2012 Subjects were picked from the Riga Cleft Lip and Palate Centre Registry from the time period  of 1980‐2009.    Not reported.  Dietz, 2012 Subjects were individuals with non‐syndromic CL/P registered at the Danish Facial Cleft Regis‐ ter.    Individuals born after 1975 were excluded.  Steinwachs, 2000 Subjects ought to have been diagnosed with cleft lip with or without cleft palate and have no  known syndromal cause for the cleft.    Not reported.  Table 3: Inclusion and exclusion criteria among the eight selected papers.    Non‐syndromic oral cle s and risk of cancer: a systema c review  Vol 1, No 1 (2013)    DOI 10.5195/d3000.2013.11  http://dentistry3000.pitt.edu 5 Table 4: Poten al bias/limita ons and the author’s efforts to iden fy them among the eight selected papers. among relatives of individuals diagnosed with cancer, besides being scarce, vary greatly in their characteristics and designs, precluding a meta-analysis of the results found. In the current systematic review, both stud- ies assessing the prevalence of CL/P in fami- ly members of cancer patients had a case- control design and suggest that this mal- formation is more frequent in families of cancer survivors in comparison with fami- lies of a population-based sample of con- trols. In the study of Taioli et al. (2010) [2], the results were not statistically signi icant and the authors relate it to the sample size of the study and to the rarity of CL/P occur- rence. In the second study, relating CL/P and childhood cancer, none of the children from the unaffected group had a positive family history of cleft lip and palate, while ive children with leukemia had a positive history of oral clefts [18]. In these studies, the population investigat- ed, or their parents or guardians, answered standardized questionnaires with demo- graphic information about the respondents and their family history of CL/P. It was a consensus between the authors that infor- mation on number of second-degree rela- tives is lacking for both, which could explain their indings for the irst-degree relatives, since the more distant the relative is, less information is known about him. In the study investigating adult individuals who survived cancer, the most frequent types of cancer were melanoma, testicular, breast, colon-rectal, and lymphoma. The authors state, however, that one limitation of the study was to include cancer survivors instead of newly diagnosed cancer patients, which could exclude highly fatal cancers [2]. Despite the different characteristics of the populations investigated in the studies of Taioli et al. (2010) and Jindal and Vieira (2012) [2,18], both suggest that shared ge- netic factors may explain an association between oral clefts and cancer. Thus, inves- tigating the relationship between malfor- mations and malignancies becomes im- portant as it is speculated that they might have common causes. Endorsing that possi- bility, Chalothorn et al. (2008) and Küchler et al. (2013) [42,43] have previously pro- posed an association between tooth agene- sis and cancer. In some cases, tooth agenesis and oral cleft may share the same genetic background. This review also accessed studies [5,13- 15,17,21] regarding risk of cancer in rela- tives of individuals born with non- syndromic oral clefts. From the six studies found, three also had a case-control design and more cancer cases were reported among relatives of cleft individuals than by controls [14,15,17]. The most prevalent types of cancer have not been plainly identi- ied in all studies, which was a limitation of the studies declared by the authors. Howev- er, in those which it was possible, a clear predominance of colon cancer and leukemia cases was noticed, as well as some reports of breast cancer, lymphoma and brain can- cer [14,15]. The most affected relatives were speci ied only in the study of Zhu et al., (2002) [15], which exclusively investigated parents. A further limitation that was found is the fact that the speci ic data regarding the age of onset of cancer was not accessed, which makes it impossible to determine if family members of individuals with cleft developed cancer at earlier ages than the general population. A higher prevalence of cancer in family members of individuals born with CL/P was presented in a cross-sectional study with a Latvian population, too [13]. The risk was calculated by dividing the prevalence of cancer in the target group by the prevalence of cancer in that population. It was demon- strated that this risk is three times higher in First author, year [reference] Bias/limitations Identification of bias/limitations Jindal, 2012 Information concerning illnesses on relatives more distant than the first‐degree  may not be so reliable.  Not reported.  Taioli, 2010 Since it researched cancer survivors and not patients diagnosed with cancer per  se, the most fatal forms of cancer tend to have their association with CL/P not  reported accurately. Also, an association between the type of cancer and CL/P  becomes impossible to be made in this case because of characteristic survivor‐ ships in each type.  Not reported.  Yildirim, 2012 As information regarding specific types of cancers was not available for all indi‐ viduals, an analysis by cancer types could not be done. Also, the specific data  regarding the age of onset of cancer was not accessed, which made it impossible  to determine if family members of cleft individuals developed cancer at earlier  ages than the general population.  Not reported.  Menezes, 2009 The cancer types reported may not all have been related to the same causes.  Also, as there was no access to data that specified the age of cancer develop‐ ment in those individuals’ family members, it was no possible to determine if  family members of cleft individuals developed cancer at earlier ages than the  general population.  Proposes new research with increased sample size to  replicate findings and test the association between genetic  variants and specific types of cancer.  Zhu, 2002 There may have been some confusion from the teratogenic effects of cancer  treatment in parents who had cancer previous to the child’s birth. The outcome  of previous births also could induce some bias as well as maternal age at birth  and sex of the child. Lastly, the lack of data on stillbirths may have biased results  if fetal mortality differs between parents who will develop cancer and parents  who won’t.  It only used parents whose cancer diagnostic came up  after the birth of their indexed children. Also, the children  analyzed were only firstborns and the age of the mother at  birth and the child’s sex were reported.  Vieira, 2012 The population studied was composed mostly by individuals in their first years of  life, and because of that, the prevalence of pediatric or earlier onset types of  cancer was relatively increased when compared with the other types. The same  is true for their relatives, as they are most likely also relatively young.  Not reported.  Dietz, 2012 A left and right truncation in this study was observed, as no information can be  gotten neither previous nor post the period studied.  Not reported.  Steinwachs, 2000 There is known lack of accuracy in information provided by patient reports about  their family members.  Proposes new and larger studies with sufficient power to  study only first‐degree relatives.    Non‐syndromic oral cle s and risk of cancer: a systema c review  Vol 1, No 1 (2013)    DOI 10.5195/d3000.2013.11  http://dentistry3000.pitt.edu 6 irst- and second-degree relatives and de- creases to 1.5 times in third-degree rela- tives. Similar to most available studies with the same proposal, the types of cancer were not consistently reported and, because of that, this study did not discriminate by can- cer types. As well as in the studies addressing preva- lence of CL/P in family members of cancer patients, indings from four [13-15,17] of six studies regarding risk of cancer in relatives of individuals born with non-syndromic oral clefts provide additional support for a common etiology between CL/P and cancer, suggesting that these families have a higher genetic load for CL/P, which also impacts their cancer risk. On the other hand, two other studies re- garding risk of cancer in relatives of indi- viduals with clefts could not conclude the same [5,21]. The irst one, performed by Steinwachs et al., (2000) [21] was emphatic in denying an increased risk for cancer in the irst- and second-degree relatives of individual affected with non-syndromic CL/P. The population investigated was composed of Caucasians, Hispanics, African Americans, and Asian individuals. The can- cer data were analyzed in aggregate by type of cancer, by degree of relation to the index case, and by ethnicity. As a result, neither irst-degree relatives nor second-degree relatives of the index cases had a signi i- cantly increased risk for cancer. No particu- lar ethnic group demonstrated an excessive- ly increased risk for cancer. Not even the most commonly reported cancers (lym- phoproliferative cancers, female cancers, and lung cancer) were increased as a whole or in irst-degree relatives only. In the same way, in a recent retrospective cohort study [5] testing whether mothers and sisters of individuals born with CL/P have an increased risk for breast cancer, the authors were not able to con irm a general increase in this risk for breast cancer among the study population. However, some asso- ciations were revealed when cleft subtype was analyzed, and having a child with iso- lated cleft palate was associated with an increased risk of breast cancer for the mother, which may be biologically plausible due to the evidence that the different cleft types have different genetic origins [44,45]. Thus, the hypothesis that parental cancer and congenital malformations in offspring may be correlated could not be entirely con irmed by the present systematic review. Despite it being known that possible mech- anisms at the basis of an association be- tween CL/P and cancer are shared genetic factors, once polymorphisms in genes in- volved in cell–cell adhesion and cellmotility are associated with both cancer and CL/P [16,18], studies in this direction are very scarce and of those existing and most inves- tigated all subtypes of oral clefts, which sets a limitation to ind out such correlation. There is agreement among almost every reviewed author [2, 14,17,18] that the in- vestigated populations in their studies need to be expanded in order to avoid any ran- dom indings in multiple testing or the po- tential impact of Type I error. It is also in agreement that future studies should ex- plore the possibility that there is a preferen- tial occurrence of speci ic types of cancer in families with individuals born with CL/P and test the hypothesis that common genet- ic and epigenetic mechanisms are playing a role in both conditions [14,17]. Through this information, future studies may be bet- ter able to identify the causes of non- syndromic orofacial clefts and ultimately to predict its occurrence and to facilitate ge- netic counseling of affected families [21]. Another issue raised by Dietz et al. (2012) is the fact that having a cleft leads to social marginalization and changes in lifestyle could predispose not only the individual, but also their family for cancer, which seems plausible from a standpoint of rela- tionship between stress and cancer, as dis- cussed in the literature. Conflict of interest: There are no conflicts of interest to report. References 1. The evolu on of human gene c studies of cle   lip  and  cle   palate. Marazita  L M.  Annu  Rev  Genomics  Hum  Genet. 2012;13:263‐83.  Epub  2012 Jun 6. PMID:22703175 [PubMed ‐ indexed  for MEDLINE]  2. Cle  Lip and Palate  in Family Members of Can‐ cer  Survivors.  Taioli  E,  Ragin  C,  Robertson  L,  Linkov  F,  Thuman  LE,  Vieira  AR.  Cancer  In‐ vest. 2010  Nov;28(9):958‐62.  PMID:20569073  [PubMed ‐ indexed for MEDLINE]  3. Iden fying  determinants  of  quality  of  life  of  children with cancer and childhood cancer sur‐ vivors: a systema c  review. Klassen AF, Antho‐ ny SJ, Khan A, Sung L, Klaassen R. Support Care  Cancer. 2011  Sep;19(9):1275‐87.  Epub  2011  May 25. PMID:21611865 [PubMed ‐ indexed for  MEDLINE]  4. Parental  age  as  a  risk  for  non‐syndromic  oral  cle s:  a  meta‐analysis.  Herkrath  APCQ,  Her‐ krath FJ, Rebelo MAB, Ve ore MV. J Dent. 2012  Jan;40(1):3‐14.Epub  2011  Oct  13.  PMID:22019990  [PubMed  ‐  indexed  for MED‐ LINE]  5. Risk  of  breast  cancer  in  females with  cle   lip  and palate. Dietz A, Pedersen DA,  Jacobsen R,  Wehby GL, Murray  JC, Christensen K. Ann Epi‐ demiol. 2012  Jan;22(1):37‐42.  Epub  2011  Oct  28.  PMID:  22037380    [PubMed  ‐  indexed  for  MEDLINE] .  6. Prevalence  of  nonsyndromic  oral  cle s  in  a  reference hospital in the state of Minas Gerais,  Brazil,  between  2000‐2005.  Martelli‐Junior  H,  Porto LCVP, Barbosa DR, Bonan PRF, Freitas AB,  Cole a  RD.  Braz  Oral  Res. 2007  Oct‐ Dec;21(4):314‐7.  PMID:18060257  [PubMed  ‐  indexed for MEDLINE]  7. Unraveling human cle  lip and palate research.  Vieira  AR.  J  Dent  Res. 2008  Feb;87(2):119‐25.  PMID:18218836  [PubMed  ‐  indexed  for MED‐ LINE]  8. The  current  understanding  of  cle   lip malfor‐ ma ons.  Wan a  N,  Re nger  G.  Facial  Plast  Surg. 2002  Aug;18(3):147‐53.  PMID:12152133  [PubMed ‐ indexed for MEDLINE]  9. The  complex  gene cs  of  cle   lip  and  palate.  Cobourne MT. Eur  J Orthod. 2004  Feb;26(1):7‐ 16.  PMID:14994877  [PubMed  ‐  indexed  for  MEDLINE]   10. Cle   lip  and  palate:  synthesizing  gene c  and  environmental  influences. Dixon M, ML,   Beaty  TH,Murray  JC.  Nat  Rev  Genet. 2011  March; 12(3):  167–178.  PMID:21331089   [PubMed ‐ indexed for MEDLINE]   11. Zeiger  JS,  Beaty  TH  (2002) Gene–environment  interac on and risk to oral cle s. In: Wyszynski  DF editors. Cle   lip and palate:  from origin  to  treatment. New York: Oxford University Press;  pp. 283–289.  12. Frequency  of  congenital  craniofacial  malfor‐ ma ons  in  a Brazilian Reference Center. Para‐ naiba LMR, Miranda RT, Ribeiro LA, Barros LM,  Martelli‐Júnior  H.  Rev  Bras  Epidemi‐ ol. 2011;14:151‐160.  13. Risk of cancer in rela ves of children born with  isolated cle  lip and palate.  Vieira AR, Khaliq S,  Lace  B.  Am  J  Med  Genet  A. 2012  Jun;158A(6):1503‐4.  Epub  2012  May  14.  PMID:22585460  [PubMed  ‐  indexed  for MED‐ LINE]  14. AXIS  inhibi on protein 2, orofacial cle s and a  family  history  of  cancer. Menezes  R, Marazita  ML, Goldstein McHenry T, Cooper ME, Bardi K,  Brandon C, Letra A, Mar n RA, Vieira AR. J Am  Dent  Assoc. 2009  Jan;140(1):80‐4.  PMID:19119171  [PubMed  ‐  indexed  for MED‐ LINE]  15. Do parents of children with congenital malfor‐ ma ons  have  a  higher  cancer  risk?  A  na on‐ wide study  in Denmark. Zhu  JL, Basso O, Hasle  H,  Winther  JF,  Olsen  JH,  Olsen  J.  Br  J  Can‐ cer. 2002  Aug  27;87(5):524‐8.  PMID:12189550  [PubMed ‐ indexed for MEDLINE]   16. Cle   lip/palate and CDH1/E‐cadherin muta on  in  families with  hereditary  diffuse  gastric  can‐ cer. Frebourg T, Oliveira C, Hochain P, Karam R,  Manouvrier  S,  Graziadio  C,  Vekemans  M, Hartmann  A, Baert‐Desurmont  S, Alexandre  C, Lejeune  Dumoulin  S,Marroni  C, Mar n  C, Castedo  S, Love   M, Winston  J, Machado  JC, A é  T, Jabs  EW, Cai  J, Pellerin  P, Triboulet  JP, Sco e M, Le  Pessot  F, Hedouin  A, Carneiro  F, Blayau  M, Seruca  R.  J  Med  Genet. 2006  Feb;43(2):138‐42. Epub 2005 Apr 14.    Non‐syndromic oral cle s and risk of cancer: a systema c review  Vol 1, No 1 (2013)    DOI 10.5195/d3000.2013.11  http://dentistry3000.pitt.edu 7 17. Defining predictors of  cle   lip  and palate  risk.  Yildirim, Seymen F, Deeley K, Cooper ME, Vieira  AR.  J  Dent  Res. 2012  Jun;91(6):556‐61.  Epub  2012  Apr  10.  PMID:22496123  [PubMed  ‐  in‐ dexed for MEDLINE]  18. Family history of cle  lip and palate in subjects  diagnosed  with  leukemia.  Jindal  A,  Vieira  AR.  Am  J  Med  Genet  A. 2012  Mar;158A(3):678‐ 9.Epub 2012 Feb 2. PMID:22302642 [PubMed ‐  indexed for MEDLINE]  19. Alderson  P,  Green  S,  Higgins  JPT,  editors.  Cochrane Reviewers’ Handbook 4.2.2  [updated  March 2004]. In: The Cochrane Library, Issue 1,  2004. Chichester, UK: John Wiley & Sons, Ltd.  20. The  strengthening  the  repor ng  of  observa‐ onal  studies  in  epidemiology  (STROBE)  state‐ ment:  guidelines  for  repor ng  observa onal  studies. von Elm E, Altman DG, Egger M, Pocock  SJ,  Gøtzche  PC,  Vanderbroucke  JP.  BMJ. 2007  Oct  20;335(7624):806‐8.  PMID:17947786  [PubMed ‐ indexed for MEDLINE]   21. Nonsyndromic cle  lip and palate is not associ‐ ated with  cancer or other birth defects. Stein‐ wachs EF, Amos C, Johnston D, Mulliken J, Stal  S,  Hecht  JT  Am  J  Med  Genet. 2000  Jan  3;90(1):17‐24  PMID:10602112  [PubMed  ‐  in‐ dexed for MEDLINE]  22. Major  hematological  diseases  associated  with  cle   lip  and  palate.  Lin  CH,  Lo  LJ, Wang ML,  Chen YR, Noordhoff MS. Cle  Palate Craniofac  J. 2000  Sep;37(5):512‐5.  PMID:11034036  [PubMed ‐ indexed for MEDLINE]  23. Mossey  P,  Li le  J  (2002)  Epidemiology  of  oral  cle s: an interna onal perspec ve. In: Wyszyn‐ ski DF, editors. Cle   lip and palate: from origin  to  treatment.  New  York:  Oxford  University  Press.pp 127‐158.   24. Maternal use of an bio cs and the risk of oro‐ facial  cle s:  a  na onwide  cohort  study. Møl‐ gaard‐Nielsen,  Hviid  A.  Pharmacoepidemiol  Drug  Saf. 2012  Mar;21(3):246‐53.  Epub  2011  Nov  28.  PMID:  22125260  [PubMed  ‐  indexed  for MEDLINE]  25. Cancer  risk  in persons with oral cle   ‐ a popu‐ la on‐based  study  of  8,093  cases.  Bille  C,  Winther  JF, Bautz A, Murray  JC, Olsen  J, Chris‐ tensen  K.  Am  J  Epidemiol. 2005  Jun  1;161(11):1047‐55. PMID:15901625  [PubMed  ‐  indexed for MEDLINE]   26. Nevoid  basal  cell  carcionoma  syndrome  with  cle   lip  and  palate  associated with  the  novel  PTCH  gene  muta ons.  Sasaki  R,  Saito  K,  Watanabe  Y,  Yoshinaga  T,  Fujii  K,  Agawa  K,  Miyashita  T,  Ando  T,  Akizuki  T.  J  Hum  Genet. 2009  Jul;54(7):398‐402.  Epub  2009  Jun  12.  PMID:  19521425  [PubMed  ‐  indexed  for  MEDLINE]  27. Cle   lip  and  palate  in  Puerto  Rico:  A  Thirty‐ three  year  study.  Oliver‐Padilla  G,  Mar nez‐ González V. Cle  Palate J. 1986 Jan;23(1):48‐57.  PMID:3455903  [PubMed  ‐  indexed  for  MED‐ LINE]  28. Incidence of  cle   lip,  cle  palate, and  cle   lip  and palate among races: a review. Vanderas AP.  Cle   Palate  J. 1987  Jul;24(3):216‐25.  PMID:3308178  [PubMed  ‐  indexed  for  MED‐ LINE]  29. The  gene cs  of  isolated  orofacial  cle s:  from  genotypes to subphenotypes. Jugessur A, Farlie  PG,  Kilpatrick N. Oral  Dis. 2009 Oct;15(7):437‐ 53. Epub 2009 Jul 2. PMID:19583827 [PubMed ‐  indexed for MEDLINE]  30. Hereditary  cle   lip/palate  and Will  Tumor:  A  rare associa on.  Yu CC, Wong  FH,  Lo  LJ, Chen  YR.  Cle   Palate  Craniofac  J. 2002  May;39(3):376‐9.  PMID:12019017  [PubMed  ‐  indexed for MEDLINE]  31. Gene/environment  causes  of  cle   lip  and/or  palate.  Murray  JC.  Clin  Genet. 2002  Apr;61(4):248‐56.  PMID:12030886  [PubMed  ‐  indexed for MEDLINE]  32. Cle  lip (+/‐ cle  palate) in Danish twins, 1970‐ 1990.  Christensen  K, Fogh‐Andersen  P.  Am  J  Med  Genet. 1993  Nov  1;47(6):910‐6.  PMID:8279491  [PubMed  ‐  indexed  for  MED‐ LINE]  33. Analysis of  family  incidence of  cle   lip  and/or  palate.  Kot  M, Kruk‐Jeromini  J.  Med  Sci  Monit. 2007  May;13  (5):CR231‐4.  PMID:17476195  [PubMed  ‐  indexed  for MED‐ LINE]  34. A popula on‐based study of cancer incidence in  twins  and  in  children  with  congenital malfor‐ ma ons  or  low  birth  weight,  Norway,  1967‐ 1980. Windham  GC,  Bjerkedal  T,  Langmark  F.  Am  J  Epidemiol. 1985  Jan;121(1):49‐56.  PMID:3155484  [PubMed  ‐  indexed  for  MED‐ LINE]  35. Increased  prevalence  of  minor  anomalies  in  childhood malignancy. Mehes K, Signer E, Pluss  HJ,  Müller  HJ, Stalder  G.  Eur  J  Pediatr. 1985  Sep;144(3):243‐54.  PMID:4054163  [PubMed  ‐  indexed for MEDLINE]  36. Excess  of  mild  errors  of  morphogenesis  in  childhood  lymphoblas c  leukemia.  Mehes  K,  Kajtar  P,  Sandor  G,  Scheel‐Walter  M, Niethammer D. Am  J Med Genet. 1998  Jan  6;75(1):22‐7. PMID:9450852 [PubMed ‐ indexed  for MEDLINE]  37. Maternal  and  perinatal  risk  factors  for  child‐ hood  leukemia.  Zack M, Adami HO, Ericson A.  Cancer  Res. 1991  Jul  15;51(14):3696‐701.  PMID:  2065325  [PubMed  ‐  indexed  for MED‐ LINE]  38. Risk of  childhood  cancer  for  infants with birth  defects, part  I: a  record‐linkage study, Atlanta,  Georgia, 1968‐1988. Mili F, Khoury MJ, Flanders  WD, Greenberg RS. Am  J Epidemiol. 1993 Mar  15;137(6):629‐38.  PMID:8470664  [PubMed  ‐  indexed for MEDLINE]  39. Congenital  malforma ons  and  childhood  can‐ cer. Nishi M, Miyake H, Takeda T,  Hatae Y. Med  Pediatr Oncol. 2000;34:250–254.   40. Increased  prevalence  of  minor  anomalies  in  children with hematologic malignancies. Roga‐ novic J, Radojcic‐Badovinac A, Ahel V. Med Pe‐ diatr  Oncol.  2002  Feb;38(2):128‐30.  PMID:11813182  [PubMed  ‐  indexed  for MED‐ LINE].  41. Infant leukemia and congenital abnormali es: a  children’s  oncology  group  study.  Johnson  KJ,  Roesler MA, Linabery AM, Hilden JA, Davies SM,  Ross  JA.  Pediatr  Blood  Cancer. 2010  Jul;55(1):95‐9.  PMID:20486175  [PubMed  ‐  in‐ dexed for MEDLINE]  42. Hypodon a as a risk marker for epithelial ovari‐ an  cancer: a  case‐controlled  study. Chalothorn  LA, Beeman CS, Ebersole JL, Kluemper GT, Hicks  EP, Kryscio RJ, DeSimone CP, Modesi  SC. J Am  Dent  Assoc. 2008  Feb;139(2):163‐9.   PMID:18245684  [PubMed  ‐  indexed  for MED‐ LINE]  43. Tooth agenesis associa on with self‐ reported family history of cancer.  Küchler  EC, Lips  A, Tannure  PN, Ho  B, Costa  MC, Granjeiro  JM, Vieira  AR.  J  Dent  Res. 2013  Feb;92(2):149‐55.  Epub  2012  Nov  20.  PMID:23169889  [PubMed  ‐  indexed  for MED‐ LINE]  44. Disrup on  of  an  AP‐2alpha  binding  site  in  an  IRF6  enhancer  is  associated  with  cle   lip.  Rahimov  F,  Marazita  M,  Visel  A,  Cooper  M,  Hitchler  M,  Rubini  M,  Domann  FE, Govil  M, Christensen  K, Bille  C, Melbye  M, Jugessur  A, Lie  RT, Wilcox  AJ,Fitzpatrick  DR, Green  ED, Mossey  PA, Li le  J, Steegers‐Theunissen  RP, Pennacchio  LA, Schu e  BC, Murray  JC. Nat  Genet. 2008 Nov;40(11):1341‐7. Epub 2008 Oct  5.  PMID:18836445  [PubMed  ‐  indexed  for  MEDLINE].  45. A  cohort  study  of  recurrence  pa erns  among  more than 54,000 rela ves of oral cle  cases in  Denmark:  Support  for  the  mul factorial  threshold  model  of  inheritance.  Grosen  D,  Chevrier  C,  Sky he  A,  Bille  C,  Mølsted  K,  Sivertsen  A, Murray  JC, Christensen  K.  J  Med  Genet. 2010  Mar;47(3):162‐8.Epub  2009  Sep  14.  PMID:19752161  [PubMed  ‐  indexed  for  MEDLINE]