Bannayan-Riley-Ruvalcaba syndrome (BRR) is a hamartoma syndrome characterized by early-onset macrocephaly, lipomatosis, hemangiomas, hamartomatous polyps of the gastrointestinal tract, vascular malformations, pigmented macules of the glans penis (speckled penis) in males, Hashimoto’s thyroiditis, and mild intellectual delay (1–3). Although an anecdotal case of malignant tumor has been reported to be associated with BRR (4), malignancy has yet to be formally accepted as part of this syndrome’s phenotypic spectrum. BRR shows both clinical and genetic overlap with another hamartoma syndrome, Cowden syndrome (CS), characterized by hamartomas in multiple organ systems including the breast, skin, thyroid, central nervous system, and gastrointestinal tract (5–7). Patients with CS may suffer from malignancies and also have macrocephaly; brain tumors, especially meningiomas, have also been reported in these patients (8). Germline mutations in PTEN, a novel tumor suppressor gene mapping to 10q23.3, have been found in 13–81% of CS patients (9–12) and 57–60% of BRR cases (12, 13). Two cases of 10q deletion encompassing PTEN have also been reported in BRR (14, 15). PTEN encodes a dual specificity phosphatase with homology to the focal adhesion molecules tensin and auxilin (16–19). Mutant or decreased levels of PTEN have been shown to lead to accumulation of the phospholipid phosphotidylinositol triphosphate, which, in turn, activates the cell survival factor protein kinase B (PKB)/Akt via the phosphatidylinositol 3-kinase pathway (20–24). Thus, PTEN may have a role in cell growth, apoptosis, and adhesion. We report a case of a boy with features consistent with BRR with a novel de novo balanced translocation, 46,XY, t(10;13)(q23.2;q33), and a malignant intracranial hCG-secreting tumor resulting in precocious puberty. The pathology of these tumors can vary, but none of them has been associated with the CS or BRR phenotype or PTEN mutations to date. Although previous reports have provided evidence that germline intragenic mutation and gross deletion of PTEN can lead to BRR, we postulate that a germline balanced translocation incorporating PTEN can also lead to the BRR phenotype. After obtaining parental consent, peripheral blood lymphocyte karyotypes were generated for both the proband and his parents, and DNA was extracted from peripheral leukocytes. Cultured skin fibroblasts from the proband were grown in DMEM supplemented with 10% heat-inactivated FCS, 1 × MEM sodium pyruvate solution, penicillin/streptomycin, 10 mmol/L HEPES buffer solution and 4.4 mmol/L l-glutamine. DNA was extracted from the fibroblast cell line and from the proband’s paraffin-embedded lipoma using the QIAamp Tissue Kit (QIAGEN, Chatsworth, CA). Fine mapping of the translocation using fluorescence in situ hybridization (FISH) was performed, as described previously (25), using a digoxigenin-labeled probe, BAC46B12 (18), containing the PTEN gene, and the D10Z1 probe (Oncor, Gaithersburg, MD) specific to the centromere of chromosome 10. Hybridization signals were detected using antidigoxigenin-rhodamine supplied in the Oncor Kit (Oncor). Chromosomes were counterstained with 4,6-diamidino-2-phenylindole-dihydrochloride, and hybridization was observed using a Carl Zeiss Axiophot fluorescence microscope (New York, NY). Denaturing gradient gel electrophoresis (DGGE) was used to examine the nine PTEN exons and flanking intronic sequences in germline DNA extracted from peripheral blood leukocytes as well as from fibroblasts from the proband. Primer sequences and PCR conditions have been described previously (26–28). For purposes of haplotype analysis, exclusion of hemizygosity, and loss of heterozygosity (LOH) analysis, genotyping across 18cM in 10q22–24 was performed using 12 polymorphic microsatellite markers, D10S219-D10S551-D10S1644-D10S1744-D10S579-D10S2491-AFMa086wg9-D10S2492/IVS8+ 32T/G-D10S541-D10S1739-D10S583 (centromeric to telomeric) (29; see http://www.genome.wi. mit.edu). PCR conditions and analyses for the microsatellite markers have been described previously (12, 26, 30, 31). LOH studies were also performed to detect a second genetic hit in the proband’s lipoma tissue using intragenic PTEN markers determined to be heterozygous in the proband. Conditions for LOH studies have also been described previously (31). An 8.5-yr-old son of healthy parents presented with precocious puberty and macrocephaly. After an uneventful pregnancy, he was born at term weighing 3.86 kg (0.7 sd above the mean) with a head circumference of 37 cm (1.3 sd above the mean). By 6 weeks of age, his head circumference at 43 cm was 2.6 sd above the mean; a computed tomography scan at the age of 6 months did not reveal any abnormalities. At 5.5 yr, the patient presented with a swelling in the left groin that revealed a lipoma after excision. At 6.8 yr, he was referred with a few months’ history of poor weight gain, reduced appetite, and occasional effortless vomiting. Testicular enlargement was noted. Thorough investigation for failure to thrive did not reveal any abnormalities. Magnetic resonance imaging (MRI) scan of the head failed to reveal any focal lesions or ventricular dilatation (Fig. 1). After 6 months of nutritional supplements, nasogastric feeding, and behavior modification therapy, the patient started to regain weight and was discharged from follow-up. During this period, he was noticed to have symptoms of obstructive sleep apnea that improved after a tonsillectomy. At 8.4 yr, he presented to his family practitioner with a dry scaly rash that was diagnosed as eczema. Treatment with topical steroids did not lead to any improvement. He was, however, noticed to be markedly virilized, prompting his referral to the pediatric endocrine service. He had attained his developmental milestones at the appropriate age, and his educational progress in mainstream school had not been of concern to the parents or his school. The patient’s parents were probably unrelated, although both had been adopted and were not aware of their own family’s histories. They were both healthy and did not have macrocephaly. The proband’s mother reached menarche at 10.5 yr. A younger sibling was generally well. MRI scan of the head showing at age 6.8 yr no intracranial abnormality (A), at age 8.4 yr the presence of two intracranial tumors (arrows), a smaller hypothalamic lesion and a larger tumor arising from the base of the fourth ventricle (B), and at age 9.0 yr resolution of intracranial abnormalities present 6 months previously. MRI scan of the head showing at age 6.8 yr no intracranial abnormality (A), at age 8.4 yr the presence of two intracranial tumors (arrows), a smaller hypothalamic lesion and a larger tumor arising from the base of the fourth ventricle (B), and at age 9.0 yr resolution of intracranial abnormalities present 6 months previously. On examination, the patient’s height, weight, and head circumference were 1.1, 1.4, and 4.7 sd above the mean, respectively. His previous height recordings were between 0.1–0.5 sd above the mean. Blood pressure was 110/60 mm Hg. He had marked frontal bossing and scaphocephaly, and neurological examination showed mild papilloedema and evidence of left-sided cerebellar dysfunction. Pubertal status was G5 PH5 AH3, with testicular volumes of 8 mL bilaterally. He had hyperpigmented, velvety, hyperkeratotic papules consistent with acanthosis nigricans in a number of areas, but especially the neck, axillae, upper arms, and thighs. Close examination of his penis revealed speckling of the glans. His bone age, assessed by the radius-ulna score, was 13 yr. Details of biochemical investigations are listed in Table 1. MRI scan of the head confirmed the presence of two intracranial tumors: a smaller hypothalamic lesion and a larger tumor arising from the base of the fourth ventricle (Fig. 1). MRI scan of the spine, ultrasound examination of the abdomen, and a plain x-ray of the chest did not reveal any abnormalities. The patient was started on an international germ cell tumor trial consisting of multiagent chemotherapy and radiotherapy. He developed profound myelosuppression and central diabetes insipidus. During therapy, he required prolonged artificial ventilation after central respiratory failure. His treatment was then changed to a combination of radiotherapy and chemotherapy, consisting of carboplatin, etoposide, and vincristine, and he is presently in remission 2 yr after completion of this regimen. Repeat MRI of the head 6 months after treatment showed resolution of the lesions seen earlier (Fig. 1). Investigations performed at 2 yr after completion of treatment showed evidence of GH deficiency and secondary hypothyroidism (Table 1). The short Synacthen test at retesting was normal. His pubertal status was G5 PH1 AH1, with testicular volumes of 3 mL bilaterally. He still requires nocturnal positive pressure support at home and suffers from central diabetes insipidus. Biochemical details of patient at presentation and 2 yr after treatment Normal adult range, 10.5–29.5 nmol/L, multiply by 28.8 to convert to ng/dL. LHRH stimulation test with sample collection at 0, 20, and 60 mins. Normal range, 9–21 pmol/L, multiply by 0.08 to convert to ng/dL. TRH stimulation test with sample collection at 0, 20, 60 mins. Short Synacthen test with sample collection at 0, 30, 60 min; divide by 27.6 to convert to μg/dL. Normal range, 0–150 mU/L. Normal range, 7–50 nmol/L. Clonidine GH stimulation test with sample collection at 0, 15, 30, 60, 90, 120 mins. Normal range, <5 U/L. Normal range, <15 U/mL. Patient on thyroxine supplementation. Biochemical details of patient at presentation and 2 yr after treatment Normal adult range, 10.5–29.5 nmol/L, multiply by 28.8 to convert to ng/dL. LHRH stimulation test with sample collection at 0, 20, and 60 mins. Normal range, 9–21 pmol/L, multiply by 0.08 to convert to ng/dL. TRH stimulation test with sample collection at 0, 20, 60 mins. Short Synacthen test with sample collection at 0, 30, 60 min; divide by 27.6 to convert to μg/dL. Normal range, 0–150 mU/L. Normal range, 7–50 nmol/L. Clonidine GH stimulation test with sample collection at 0, 15, 30, 60, 90, 120 mins. Normal range, <5 U/L. Normal range, <15 U/mL. Patient on thyroxine supplementation. Cytogenetic analysis revealed a karyotype of 46,XY, t(10;13)(q23.2;q33). The karyotypes of both parents were normal. FISH mapping revealed that hybridization signals for the BAC46B12 probe specific for PTEN were present on the normal chromosome 10 and on the derivative chromosome 13[der(13)], indicating that the breakpoint on the chromosome 10 long arm was centromeric to PTEN. The D10Z1 DNA probe facilitated identification of the der(10) chromosome, and hybridization signals specific for this probe were seen, as expected, on the normal copy of chromosome 10 and the der(10) (Fig. 2). No residual hybridization signal specific to PTEN was observed on the der(10), suggesting that PTEN was not rearranged due to this translocation. FISH mapping of the balanced translocation 46,XY, t(10;13)(q23.2;q33). The PTEN-specific BAC, BAC 46B12, clearly identified two copies of PTEN. FISH mapping of the balanced translocation 46,XY, t(10;13)(q23.2;q33). The PTEN-specific BAC, BAC 46B12, clearly identified two copies of PTEN. Analysis of the 12 polymorphic markers within and flanking PTEN showed that of the four markers intragenic to PTEN, two, residing at either end of the gene, were heterozygous (Fig. 3). Four of the five centromeric flanking markers and two of the three telomeric flanking markers were also heterozygous. DGGE-based PTEN analysis in the proband did not reveal any mutations. Further, loss of heterozygosity within PTEN was not observed in lipoma tissue from the proband (data not shown). Haplotypes of the proband and his parents. Heterozygosity is clearly observed at intragenic PTEN markers as well as at markers in intervals flanking either side of PTEN in the proband. Parental haplotypes are consistent with paternity. Haplotypes of the proband and his parents. Heterozygosity is clearly observed at intragenic PTEN markers as well as at markers in intervals flanking either side of PTEN in the proband. Parental haplotypes are consistent with paternity. Unlike many other syndromes of overgrowth where considerable phenotypic overlap exists (32), the BRR phenotype has become more defined after the identification of PTEN abnormalities in affected cases (29, 33). It is unlikely that the phenotype of our case is part of any other overgrowth syndrome, such as Sotos syndrome, a condition often confused with BRR in the past (33, 34). The constellation of lipoma, early onset macrocephaly with normal ventricular size, normal childhood stature, pigmented penile lesions, absence of marked mental retardation, and the phenomenon of transient marked wasting associated with a cytogenetic abnormality in the vicinity of PTEN support the diagnosis of BRR (35, 36). However, intracranial tumors in BRR seem to be restricted to meningiomas. To our knowledge, this case is the first report of a hCG-secreting intracranial tumor in an individual with BRR. Although tumors of embryonal and nonembryonal origin occur in a number of overgrowth syndromes (32), hCG-secreting tumors have not been associated with any overgrowth syndromes, and the only reported associations seem to be with Down syndrome and Klinefelter syndrome (37, 38). In our case, signs of puberty were evident well before radiological changes were evident on MRI scan. It is possible that the intracranial lesions were metastases from an occult extracranial primary tumor. However, their location within the brain is unusual for metastases and more in keeping with the midline sites associated with germ cell tumors. The presence of two distinct lesions within the brain emphasizes the invasive nature of the tumor, but could also suggest two primaries. The latter would support the true association of cranial germ cell tumors and PTEN-related BRR. Centrally mediated isosexual precocious puberty in boys is a rare occurrence, and hCG-producing tumors may account for 4–20% of these cases of precocious puberty (39, 40). Postnatally, an elevated hCG concentration indicates the presence of a germ cell tumor derived from primitive embryonal cells such as embryonic yolk sac tumors or choriocarcinomas (40, 41). In this case, the hCG-secreting tumor is probably due to a choriocarcinoma, asα -fetoprotein is often elevated in cases of yolk sac tumors. The discordance between testicular size and the stage of virilization has been described before in cases of hCG-mediated precocious puberty and may be a useful clue in the investigation of precocious puberty. The selective stimulation of the LH receptor may explain high testosterone levels with only modest testicular enlargement. Germline mutation of PTEN, including both intragenic mutation and gross deletion, has been firmly established as the primary genetic event in the development of the majority of cases of BRR and CS (9–15, 29, 31, 33) and the germline balanced de novo translocation, 46,XY, t(10;13)(q23.2;q33), in this boy was, therefore, considered to be a contributory factor in the development of the phenotype in this case. However, FISH analysis was clearly able to show the presence of two copies of PTEN in skin fibroblast cells from this patient with no detectable physical disruption of PTEN mediated by this translocation. Heterozygosity of intragenic and flanking microsatellite markers further argued against any loss of PTEN material as a result of this translocation. Methylation studies of the lipoma tissue were not attempted because the specimen had been fixed in paraffin. Lack of detectable biallelic inactivation is not uncommon for PTEN (31), and as its promoter is not characterized, methylation studies would probably have been inconclusive. DGGE analysis did not reveal any germline intragenic PTEN mutations that are commonly associated with CS or BRR. We did not proceed to DNA sequencing because, unlike SSCP analysis, DGGE is virtually 100% sensitive, and an alteration can be detected even if a mutation is represented in only 10% of the cells; in contrast, direct sequencing, which is considered the gold standard of mutation detection, requires the mutation to be present in more than 35% of the cells (42, 43). Although our FISH data suggest that the translocation breakpoint occurred centromeric to PTEN, it is possible that if it occurred at the very 5′-end of PTEN, we would have been unable to detect the signal due to limitations of FISH sensitivity. Alternatively, the translocation breakpoint may have occurred upstream of PTEN, causing the physical separation of important PTEN regulatory elements. Yet another possibility is that a gene on 13q, when brought proximal to PTEN, may dysregulate PTEN transcription or translation, leading to functional haploinsufficiency. One also needs to consider the possibility that an inactivating mutation in an as yet unidentified gene may contribute toward the BRR phenotype. Although malignancy has not been accepted formally as a feature of BRR, this case represents the second report of a malignant tumor occurring in conjunction with BRR, and the first occurring in a pure BRR patient. An anecdotal family comprising individuals with features of BRR and CS has been reported; the teenage male proband with features of BRR also had a thyroid insular follicular carcinoma and nodular hyperplasia with a focus of papillary microcarcinoma. This family was found to have the germline nonsense point mutation, R130X, in the region encoding the protein tyrosine phosphatase core motif of PTEN (4). These two cases provide anecdotal evidence of malignancy in BRR; however, the possibility that these malignant tumors arise on a background other than that of a germline PTEN mutation in these individuals cannot be excluded. Based on genetic and phenotypic evidence, it is becoming increasingly likely that BRR and CS, previously thought to be two discrete hamartoma syndromes, may, in fact, represent different manifestations of a single syndrome (29). If this is true, then BRR patients should be targeted along with CS individuals with respect to cancer surveillance. Dr Patricia Dahia is thanked for critical review of this manuscript.
No takes yet. Share an insight, caveat, or question.
Ahmed et al. (1999) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: