PTEN Mutations and Proteus Syndrome
Charis Eng,Hannelore Thiele,Xiao-Ping Zhou,Robert J Gorlin,Raoul CM Hennekam,Robin M Winter
DOI: https://doi.org/10.1016/s0140-6736(01)07110-0
IF: 202.731
2001-01-01
The Lancet
Abstract:Authors' replySir—We compared essential clinical features among Cowden syndrome, another difficult clinical diagnosis, Bannayan-Riley-Ruvalcaba syndrome, and Proteus syndrome. Although we were prevented by the conditions of ethical approval from disclosing very detailed clinical findings in our patients, all diagnoses of Proteus syndrome were made by very experienced clinical geneticists specialising in dys-morphology.Biesecker and colleagues claim that the diagnostic criteria might be only 50% sensitive. When a clinical diagnosis is difficult to make and the diagnostic criteria are less than robust, a more objective means to aid diagnosis is warranted. This lesson must surely be one of the most important learned from the developmental and syndromic insights accrued over the past few years—striking examples are the craniosynostosis syndromes, skeletal dysplasias, and other malformation syndromes, in which several clinically discrete phenotypes share mutations at a common locus. As a consequence, the limits of clinical definition for malformation syndromes have frequently been exposed.Ironically, Biesecker and colleagues have contributed substantially to this development in the evolution of syndrome understanding, identifying mutations in the GLI3 gene in Pallister-Hall syndrome.1Kang S Graham JM Olney AH Biesecker LG GLI3 frameshift mutations cause autosomal dominant Pallister-Hall syndrome.Nat Genet. 1997; 15: 266-268Crossref PubMed Scopus (404) Google Scholar Mutations in this gene also cause the phenotypically distinct disorders of Greig syndrome2Kalff-Suske M Wild A Topp J et al.Point mutations throughout the GLI3 gene cause Greig cephalosyndactyly syndrome.Hum Mol Genet. 1999; 8: 1769-1777Crossref PubMed Scopus (103) Google Scholar and familial hexadactyly.3Radhakrishna U Wild A Gizeschik KH Antonarakis SE Mutation in GLI3 in postaxial polydactyly type A.Nat Genet. 1997; 17: 269-271Crossref PubMed Scopus (141) Google ScholarThus, no matter what the clinical diagnosis, finding a germline PTEN mutation should reclassify the patient or family to the molecular diagnosis of PTEN hamartoma tumour syndrome (PHTS).4Marsh DJ Kum JB Lunetta KL et al.PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome.Hum Mol Genet. 1999; 8: 1461-1472Crossref PubMed Scopus (463) Google Scholar Because of genotype-phenotype associations and the risk of cancer in PHTS, we find this molecular classification useful for clinical management as well.The argument that Barker and colleagues' and their own unpublished work did not reveal germline PTEN mutations in Proteus syndrome does not detract from our observations. We suspect that Proteus syndrome is genetically heterogeneous and only a subset will harbour such germline mutations. We have never claimed that all Proteus syndrome is caused by PTEN mutation. Second, the frequency of finding PTEN mutations is highly dependent on the mutation detection technology used. We have shown that single-strand conformational polymorphism analysis for this gene is highly insensitive (detection rate was 0%, unpublished data). The gold standard of direct sequencing should detect nearly 100% of mutations. However, for Proteus syndrome, in which an unknown degree of mosaicism could exist, sequencing becomes much less sensitive.5Yan H Papadopoulos N Marra G et al.Conversion of diploidy to haploidy.Nature. 2000; 403: 723-724Crossref PubMed Scopus (20) Google Scholar The method we use, denaturing gradient gel electrophoresis has a sensitivity of around 100% and the added advantage of being able to detect mutant contributions as low as 1%.We use the term Proteus-like syndrome to denote cases likely to have Proteus syndrome but who, in the opinion of expert dysmorphologists, do not satisfy that clinical diagnosis as currently recognised. In addition, for our continuing studies, these individuals must also not meet the criteria for the diagnosis of Cowden or Bannayan-Riley-Ruvalcaba syndromes. We have noted that individuals with Proteus-like syndromes have features reminiscent of Cowden or Bannayan-Riley-Ruvalcaba, and Proteus syndromes, or a combination of these, but do not meet the respective diagnostic criteria. Up to half such individuals harbour germline PTEN mutations. Therefore, no matter what their clinical diagnoses, they should be reclassified to PHTS. Authors' reply Sir—We compared essential clinical features among Cowden syndrome, another difficult clinical diagnosis, Bannayan-Riley-Ruvalcaba syndrome, and Proteus syndrome. Although we were prevented by the conditions of ethical approval from disclosing very detailed clinical findings in our patients, all diagnoses of Proteus syndrome were made by very experienced clinical geneticists specialising in dys-morphology. Biesecker and colleagues claim that the diagnostic criteria might be only 50% sensitive. When a clinical diagnosis is difficult to make and the diagnostic criteria are less than robust, a more objective means to aid diagnosis is warranted. This lesson must surely be one of the most important learned from the developmental and syndromic insights accrued over the past few years—striking examples are the craniosynostosis syndromes, skeletal dysplasias, and other malformation syndromes, in which several clinically discrete phenotypes share mutations at a common locus. As a consequence, the limits of clinical definition for malformation syndromes have frequently been exposed. Ironically, Biesecker and colleagues have contributed substantially to this development in the evolution of syndrome understanding, identifying mutations in the GLI3 gene in Pallister-Hall syndrome.1Kang S Graham JM Olney AH Biesecker LG GLI3 frameshift mutations cause autosomal dominant Pallister-Hall syndrome.Nat Genet. 1997; 15: 266-268Crossref PubMed Scopus (404) Google Scholar Mutations in this gene also cause the phenotypically distinct disorders of Greig syndrome2Kalff-Suske M Wild A Topp J et al.Point mutations throughout the GLI3 gene cause Greig cephalosyndactyly syndrome.Hum Mol Genet. 1999; 8: 1769-1777Crossref PubMed Scopus (103) Google Scholar and familial hexadactyly.3Radhakrishna U Wild A Gizeschik KH Antonarakis SE Mutation in GLI3 in postaxial polydactyly type A.Nat Genet. 1997; 17: 269-271Crossref PubMed Scopus (141) Google Scholar Thus, no matter what the clinical diagnosis, finding a germline PTEN mutation should reclassify the patient or family to the molecular diagnosis of PTEN hamartoma tumour syndrome (PHTS).4Marsh DJ Kum JB Lunetta KL et al.PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome.Hum Mol Genet. 1999; 8: 1461-1472Crossref PubMed Scopus (463) Google Scholar Because of genotype-phenotype associations and the risk of cancer in PHTS, we find this molecular classification useful for clinical management as well. The argument that Barker and colleagues' and their own unpublished work did not reveal germline PTEN mutations in Proteus syndrome does not detract from our observations. We suspect that Proteus syndrome is genetically heterogeneous and only a subset will harbour such germline mutations. We have never claimed that all Proteus syndrome is caused by PTEN mutation. Second, the frequency of finding PTEN mutations is highly dependent on the mutation detection technology used. We have shown that single-strand conformational polymorphism analysis for this gene is highly insensitive (detection rate was 0%, unpublished data). The gold standard of direct sequencing should detect nearly 100% of mutations. However, for Proteus syndrome, in which an unknown degree of mosaicism could exist, sequencing becomes much less sensitive.5Yan H Papadopoulos N Marra G et al.Conversion of diploidy to haploidy.Nature. 2000; 403: 723-724Crossref PubMed Scopus (20) Google Scholar The method we use, denaturing gradient gel electrophoresis has a sensitivity of around 100% and the added advantage of being able to detect mutant contributions as low as 1%. We use the term Proteus-like syndrome to denote cases likely to have Proteus syndrome but who, in the opinion of expert dysmorphologists, do not satisfy that clinical diagnosis as currently recognised. In addition, for our continuing studies, these individuals must also not meet the criteria for the diagnosis of Cowden or Bannayan-Riley-Ruvalcaba syndromes. We have noted that individuals with Proteus-like syndromes have features reminiscent of Cowden or Bannayan-Riley-Ruvalcaba, and Proteus syndromes, or a combination of these, but do not meet the respective diagnostic criteria. Up to half such individuals harbour germline PTEN mutations. Therefore, no matter what their clinical diagnoses, they should be reclassified to PHTS. PTEN mutations and Proteus syndromeIn their report, Xiao-Ping Zhou and colleagues (July 21, p 210)1 describe a study of nine patients with Proteus syndrome, of whom two had mutations in the PTEN gene. They say that patients meet published diagnostic criteria.2 We raise several issues about their conclusions. Full-Text PDF