Editorial Comment Proteus Syndrome: Misdiagnosis With PTEN Mutations

In this editorial, we briefly (1) define Proteus syndrome; (2) analyze reports of PTEN mutations claimed to have ‘‘Proteus syndrome’’ or a ‘‘Proteus-like syndrome’’; (3) demonstrate the high frequency of misdiagnosis of Proteus syndrome by clinicians less familiar with the disorder; and (4) discuss two series of patients who do meet the diagnostic criteria for Proteus syndrome among whom none have been found to have PTENmutations. Proteus syndrome is a highly variable disorder with strikingly relentless asymmetric and disproportionate overgrowth of body parts, cerebriform connective tissue nevi, epidermal nevi, disregulated adipose tissue, and vascular malformations [Cohen and Hayden, 1979; Wiedemann et al., 1983; Cohen et al., 2002]. The cause or causes are unknown. The currentworking hypothesis is that Proteus syndrome arises from a postzygotic mutation based on (1) mosaic distribution of lesions, (2) sporadic occurrence, (3) exclusively unaffected offspring born to affected individuals, and (4) discordant identical twins [Happle, 1987;Cohen, 1993;Biesecker et al., 1998, 1999; Cohen et al., 2002]. No other model has been proposed that would account for these observations. For reasons to be explained, we are of critical of (1) reported cases said to have Proteus syndrome with PTEN germ line mutations [Zhou et al., 2001; Smith et al., 2002] and (2) reported PTEN mutations, using the unhelpful and confounding clinical term ‘‘Proteuslike syndrome’’ [Zhou et al., 2000, 2001]. We do not dispute the PTEN mutations found per se, but the clinical diagnosisofProteussyndrome.Zhouetal. [2000]studied five patients said to have Proteus syndrome and one patient with a ‘‘Proteus-like syndrome.’’ No mutations were found in the five ‘‘Proteus syndrome’’ patients. A germline mutation was found and tissue samples showed loss of heterozygosity in the ‘‘Proteus-like’’ patient. Clinical features included ‘‘marked hypertrophyof the right lower extremity in girthand length, pink verrucoid epidermal naevi . . .with . . .plaques on the right side of his body, and macrocephaly.’’ He also had lipomas and ‘‘invasive arteriovenous malformations involving the muscles and bones of the entire right lower extremity, pelvis, lower abdomen, and buttocks, as well as diffuse verrucoid epidermal naevi over his hands, legs, and face.’’ He developed ‘‘progressive heart failure’’ and also had ‘‘hypothyroidism.’’ Zhou et al. [2000] specifically stated that their patient did not have Proteus syndrome but labeled him as having a ‘‘Proteuslike syndrome.’’ Zhou et al. [2001] then studied 14 patients from several academic medical centers in the United States and Europe. PTEN germline mutations were found in two of nine Proteus syndrome patients said to have met the diagnostic criteria for PS. Three of the five patients with a ‘‘Proteus-like syndrome’’ were also found to have PTEN germline mutations. Zhou et al. [2001] provide insufficient data about their two cases of ‘‘Proteus syndrome’’ to be validated. In their table of listed findings, one patient (their PS2) had insufficient findings for the diagnosis of Proteus syndrome [Biesecker et al., 2001]. More comprehensive clinical data were not available because of issues of informed consent [Biesecker et al., 2001; Eng et al., 2001]. The term ‘‘Proteus-like syndrome’’ applied to 13 patients by Zhou et al. [2001] has already been discussed above. Smith et al. [2002] reported a PTEN mutation in a patient said to have ‘‘classical Proteus syndrome.’’ Findings included an extensive epidermal nevus involving the left arm, hand, chest, and flank; widespread capillary malformations of the chest, abdomen, and right leg; andevidence of disproportionate overgrowthof the right leg. Many findings reported by Smith et al. [2002] indicate that their patient does not have Proteus syndrome. The following features of their propositus have never been seen in Proteus syndrome patients by us; lipoblastomatosis (not the same as lipomas which their patient also had); multiple sessile polypoid lesions of the jejunum and colon (characteristic of the PTEN hamartomatumor syndrome); and apparently a true hemangioma in addition to vascular malformations. We are unimpressed by the degree of ‘‘disproportionate overgrowth’’ shown in their patient. The combination of polyposis, lipomas, vascular malformations, and hemangiomas has been reported in the PTEN hamartoma-tumor syndrome [Cohen et al., 2002]. We have evaluated over 200 patients claimed to have Proteus syndrome from the literature or from referrals for consultation. This analysiswill be published in detail elsewhere [Turner et al., 2003]. Briefly, by applying the published diagnostic criteria for Proteus syndrome [Biesecker et al., 1999], only about half of these patients actually had Proteus syndrome. By using independent *Correspondence to: Dr. M. Michael Cohen, Jr., Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5. E-mail: [email protected]