Retinoblastoma associated with chromosomal 13q14 deletion mosaicism

Retinoblastoma associated with chromosomal 13q14 deletion mosaicism

Retinoblastoma Associated with Chromosomal 13q14 Deletion Mosaicism Tero Kivela¨, MD,1 Kaija Tuppurainen, MD,2 Pekka Riikonen, MD,3 Matti Vapalahti, M...

267KB Sizes 3 Downloads 62 Views

Retinoblastoma Associated with Chromosomal 13q14 Deletion Mosaicism Tero Kivela¨, MD,1 Kaija Tuppurainen, MD,2 Pekka Riikonen, MD,3 Matti Vapalahti, MD4 Purpose: To assess the risk of retinoblastoma developing in children with microscopic chromosomal with mosaic deletions involving 13q14. Design: Case report and systematic literature review. Participants: Data on 29 patients with a mosaic and 107 patients with a nonmosaic somatic deletion of chromosome 13q14 were compared. Main Outcome Measures: Age at diagnosis, frequency, and laterality of retinoblastoma. Case Report: A dysmorphic baby, who carried a chromosomal deletion involving 13q14 in 34% of peripheral blood lymphocytes, had neuroradiologic evidence of retinoblastoma at the age of 2 weeks. She developed trilateral retinoblastoma, a pineal neuroblastic tumor, at the age of 10 months. The diagnosis of her tumor was delayed because of misjudgment of risk of retinoblastoma developing. Results: Meta-analysis revealed no difference between children with mosaic and nonmosaic chromosomal deletion of 13q14 regarding the age at diagnosis, laterality of tumor, and presence of family history for retinoblastoma. A lower percentage of somatic cells with mosaic deletion did not predict a higher age at diagnosis or unilateral tumors. No statistically significant difference was noted regarding the presence of mental retardation, dysmorphic features, and anomalies of internal organs between mosaic and nonmosaic deletions. Only 7% (95% confidence interval, 1–23) of 29 patients who had a mosaic chromosomal deletion including 13q14 were not reported to develop retinoblastoma. Conclusions: Whenever a 13q14 deletion is diagnosed, immediate ophthalmologic evaluation is recommended to ensure prompt diagnosis of retinoblastoma. Mosaic and nonmosaic chromosomal deletions of 13q14 do not differ regarding the risk and type of retinoblastoma developing. Ophthalmology 2003;110:1983–1988 © 2003 by the American Academy of Ophthalmology.

Retinoblastoma (Rb) is caused by inactivation of a dominant tumor suppressor gene, RB1, which participates in cell cycling.1 All bilateral and 10% to 15% of unilateral Rb are caused by germline mutations of RB1 and are inherited as an autosomal dominant trait. Between 5% and 10% of hereditary Rb is due to chromosome deletions that involve 13q14, the RB1 locus.2,3 These children typically are initially seen by pediatricians because of slow development and dysmorphic features such as broad nasal bridge, thin and arched upper lip, long philtrum, low-set ears, and cardiac and renal anomalies.3 Children with hereditary Rb are prone to second cancers4 and trilateral retinoblastoma,5 a midline intracranial neuroblastic tumor thought to be an independent primary focus of Rb.

If diagnosed early, the chances of salvaging the eye and remaining vision are good, and the risk of dying of Rb is small. In medically advanced countries, second primary cancers and trilateral Rb are leading causes of death among children with hereditary Rb.6,7 To save both the sight and lives of children with Rb, prompt recognition of its symptoms and signs is crucial. Especially when a child has a 13q14 deletion, or a midline tumor, the pediatrician has a key role in early diagnosis. We document that chromosomal mosaic deletions of 13q14 carry the full risk of Rb developing and emphasize the importance of being aware of the genetic and radiologic features of Rb.

Patients and Methods Originally received: May 31, 2002. Accepted: January 31, 2003. Manuscript no. 220355. 1 Pediatric Ophthalmology and Oncology Service, Department of Ophthalmology, Helsinki University Central Hospital, Helsinki, Finland. 2 Department of Ophthalmology, Kuopio University Hospital, Kuopio, Finland. 3 Department of Pediatrics, Kuopio University Hospital, Kuopio, Finland. 4 Department of Neurosurgery, Kuopio University Hospital, Kuopio, Finland. Reprint requests to Tero Kivela¨, MD, Department of Ophthalmology, Helsinki University Central Hospital, Haartmaninkatu 4 C, PL 220, FIN00029 HUS, Helsinki, Finland. © 2003 by the American Academy of Ophthalmology Published by Elsevier Inc.

Systematic Literature Review To resolve whether Rb caused by a mosaic deletion of 13q14 differs from that caused by a nonmosaic deletion, a MEDLINE search combining key words retinoblastoma, retinocytoma, or retinoma with mutation, mosaic, mosaicism, or deletion was conducted for 1966 to 2001. The original articles were reviewed for data and references on deletions that were detected by either routine or high-resolution karyotyping of somatic cells. The deletion was accepted as being mosaic in type if the article clearly stated that somatic cells with and without the deletion were detected, regardless of the proportion of cells that carried the mutaISSN 0161-6420/03/$–see front matter doi:10.1016/S0161-6420(03)00484-6

1983

Ophthalmology Volume 110, Number 10, October 2003 tion. Data were also collected from nonmosaic deletions fulfilling the preceding criteria, provided that the article mentioned age at diagnosis, laterality of the tumor, associated anomalies, and presence or absence of family history. Data on 29 patients with a mosaic and 107 patients with a nonmosaic microscopic deletion of 13q14 in somatic cells were entered to a database. Univariate statistics and cumulative frequency distributions were calculated (StatXact-3, Cytel Software, Cambridge, MA). The data spanned 3 decades and are thus subject to changes in genetic testing methods, which can translate to differences in detection of mosaicism, and it was not possible to correct for this source of bias.

Case Report A girl was born to a 26-year-old healthy mother. A previous pregnancy ended with a stillbirth. The birth weight was 2010 g, and the Apgar score was 9. The head was large (⫹2 standard deviation). She had tendency to opisthotonus, her face was triangular, the occiput was prominent, the nasal bridge was broad, the upper lip was arched and thin, the first and second toes were widely set, and the third toe was distal. At the age of 2 weeks, a computed tomography (CT) scan ordered to rule out hydrocephalus showed a 5- ⫻ 11-mm tumor in the eye (Fig 1A), but it was read as normal. By 4 weeks, karyotype was determined as 46, XX, del(13)(q12q14)/46, XX. The deletion was found in 34% of peripheral blood lymphocytes. Karyotypes of the parents were normal. The geneticist thought that mosaicism indicated a small risk for Rb, and a nonurgent eye examination was scheduled. Six weeks later, the parents noted leukokoria in her right eye. A large tumor with vitreous seeding was diagnosed (Fig 1B). The pineal gland measured 4 ⫻ 5 ⫻ 7 mm, and the brain was considered to be normal (Fig 1C, D). Enucleation revealed a differentiated Rb that was 12 ⫻ 17 mm but did not extend into the optic nerve or choroid. At the age of 7 months, her motor development corresponded to that of a 4-month-old. Some tendency to opisthotonus remained. A 4-mm Rb close to the optic disc and a 2-mm Rb at the equator were diagnosed in the left eye, and she was referred to Helsinki University Central Hospital. The small focus was cryocoagulated and the larger one photocoagulated. On review of scans, the Rb of the right eye was noted to have been present at the age of 2 weeks (Fig 1A). Two months later, the larger focus showed regrowth and was treated with external bean radiotherapy. Proximity to optic disc precluded plaque brachytherapy, and thermotherapy was not yet available at that time. Toward the end of radiotherapy, she became lethargic, and her head circumference was increasing again. Magnetic resonance imaging showed a 25- ⫻ 27-mm pineal tumor and hydrocephalus (Fig 2E–H). A ventriculoperitoneal shunt was inserted, and the tumor was resected by means of an infratentorial, suboccipital craniotomy at the age of 11 months. A partly necrotic pineoblastoma with neuroblastic rosettes was diagnosed. Intravenous vincristine, lomustine, and prednisolone, combined with intrathecal methotrexate, cytarabine, and prednisolone, were

given. After one cycle, a residual mass of 4 ⫻ 9 mm was seen (Fig 2I). At the age of 16 months, magnetic resonance imaging revealed leptomeningeal enhancement (Fig 2J), and a lumbar puncture showed tumor cells. The residual mass had decreased to 3 ⫻ 7 mm. During the next 4 weeks, her condition deteriorated, and drop metastases were identified along the spinal cord up to the cauda equina (Fig 2K). She died at the age of 17 months in terminal care.

Results The analysis revealed no difference between children with somatic mosaic and nonmosaic chromosomal deletion of 13q14 regarding the age at diagnosis of Rb (Fig 2A), gender, laterality of intraocular RB, or family history for RB (Table 1). In children with mosaic deletion, the percentage of peripheral blood lymphocytes and fibroblasts that carried the deletion ranged from 5 to 98 (median, 44). No association between the percentage of cells carrying the mutation and the age at diagnosis was found (Fig 2B; P ⫽ 0.71, Spearman’s rank correlation). A higher percentage of mutated cells did not predict more frequent bilateral Rb; if anything, the correlation was inverse (Fig 2C; P ⫽ 0.016, Kendall’s ␶). Of 29 patients with mosaic deletion, 2 (7%; 95% confidence interval, 1–23) did not have Rb. No statistically significant difference between children with mosaic and nonmosaic deletion of 13q14 was noted regarding the presence of mental retardation, but the data suggest a trend toward less frequent severe retardation among children with mosaic chromosomal 13q14 deletions (Table 1). The proportion of patients reported to have growth retardation (65% vs. 52%; P ⫽ 0.27, Fisher exact test), dysmorphic features of the skull (36% vs. 28%; P ⫽ 0.44), eyes (18% vs. 11%; P ⫽ 0.47), ears (35% vs. 18%; P ⫽ 0.09), nose 32% vs. 25%; P ⫽ 0.99), and mouth (27% vs. 24%; P ⫽ 0.78), and anomaly of internal organs 11% vs. 6%; P ⫽ 0.38) were likewise comparable between the two groups. Of 112 children with trilateral Rb,5 27 were karyotyped, and a chromosomal 13q14 deletion was found in 3 of them (11%; 95% confidence interval, 2–29).8,9

Discussion A number of small case series have suggested that up to 25% of microscopic chromosomal deletions involving 13q14 of somatic cells are mosaic deletions.3,10 –27 In a large series of 203 consecutive Rb patients, 6% had a 13q14 deletion.2 Mosaicism was detected in 17% of those with 13q14 deletion, corresponding to 1% of all patients. Germ line mosaicism, in which parents with normal leukocyte DNA have more than one child with Rb, is a more recently emphasized feature of hereditary Rb.28,29 Chromosomal and genetic mosaicism consequently is a phenomenon that every physician who manages patients with Rb must understand.

™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™3

Figure 1. A, Retinoblastoma (rb) of the right eye as a coincidental finding in a noncontrast axial computed tomography (CT) of the head, age 2 weeks. B, At the age of 14 weeks, the tumor (rb) fills more than half of the vitreous space as seen in a sagittal T2-weighted magnetic resonance imaging (MRI) scan. C, Pineal gland (between arrowheads) is of normal size and without calcifications in an axial CT scan at the age of 2 weeks and (D) in a noncontrast axial T2-weighted MRI scan at the age of 14 weeks. E, An enhancing tumor diagnosed as a pineoblastoma (pb) and symptomatic hydrocephalus in an axial CT, age 10 months. F, The tumor (pb) is confined to the pineal region in a sagittal T1-weighted gadolinium MRI scan. G, Coronal scan confirms pineal localization (pb). H, Nonenhancing areas (n) suggest partial necrosis. I, Four months after resection, MRI shows a nonenhancing residual mass (between arrowheads) with leptomeningeal enhancement of brainstem (double arrowhead). J, Seven months after resection, the mass is smaller, but leptomeningeal enhancement has increased. K, The contour of spinal cord (arrowheads) and cauda equina (c) are irregular because of drop metastases.

1984

Kivela¨ et al 䡠 Retinoblastoma and 13q14 Deletion Mosaicism

1985

Ophthalmology Volume 110, Number 10, October 2003 Table 1. Comparison of Clinical Characteristics of Retinoblastoma Caused by Mosaic and Nonmosaic Chromosomal Deletions Involving 13q14, the RB1 Locus

Median age at diagnosis of Rb, months (range) Gender, N (%) Male Female Laterality of Rb, N (%) Unilateral Bilateral Trilateral Family history of Rb, N (%) Sporadic Familial Mental retardation, N (%) None Mild to moderate Severe

Mosaic Deletion

Nonmosaic Deletion

9 (1-40)

10 (0-84)

0.99*

12 (48) 13 (52)

40 (39) 63 (61)

0.50†

12 (46) 13 (50) 1 (4)

43 (42) 58 (56) 2 (2)

0.87‡

26 (96) 1 (4)

94 (88) 13 (12)

0.30†

15 (63) 7 (29) 2 (8)

39 (40) 50 (52) 8 (8)

0.11‡

P

Rb ⫽ retinoblastoma. *Mann–Whitney U test, two-sided, exact probability distribution. † Fisher’s exact test, two-sided. ‡ Chi-square test, two-sided, exact probability distribution.

Figure 2. Effect of 13q14 chromosomal deletion mosaicism on retinoblastoma (Rb). A, Cumulative frequency distribution plot of age at diagnosis of Rb is comparable for mosaic and nonmosaic deletions. Scattergrams of (B) age at diagnosis of Rb and (C) laterality of Rb against percentage of somatic cells carrying the 13q14 deletion reveal no correlation between these variables.

It is estimated that 90% to 95% of children with mutated RB1 in general develop intraocular Rb and that 5% to 15% of them develop trilateral Rb,5–7,30 but it is not known which proportion of those who have mosaic deletions do so. The claim that chromosomal mosaicism might prevent retardation in 13q-deletion syndrome was supported by our analysis to some extent. However, mosaic deletions were not associated with more frequent unilateral Rb, more

1986

slowly growing Rb diagnosed at a higher age, and less frequent dysmorphic features than nonmosaic deletions were. The frequency of 13q14 deletion in children with trilateral Rb was comparable with that generally reported for Rb.2 We could not show that a smaller proportion of peripheral blood lymphocytes or fibroblasts carrying the 13q14 deletion would affect the frequency of bilateral Rb and the age at diagnosis. Of note, bilateral Rb was reported in children who had deletion in only 5% to 9% of these somatic cells,12,20,27 and our patient had Rb since birth and soon developed trilateral Rb. Only 2 reports of chromosomal 13q14 mosaic deletion did not mention Rb developing, and retinocytomas were not excluded in these patients.31,32 The 93% penetrance was comparable to the 90% to 95% penetrance of hereditary Rb in general. Taken together, the evidence suggests that mosaic deletions of 13q14 should not be taken as an indication of a smaller than usual chance of Rb developing or of milder than usual form of the disease.19 We note, however, that this might not be the case in genetic mosaicism without chromosome deletion. Of 9 patients who were such somatic mosaics, 3 (33%) did not develop Rb.28 Our analysis is subject to reporting bias, and we cannot exclude the possibility that patients who have 13q14 deletion mosaicism with Rb are preferentially reported. In addition to our patient, several other children who had a 13q14 deletion diagnosed after dysmorphic features and delayed development prompted a cytogenetic analysis were later found to harbor Rb.2,23,33–35 They typically were initially seen by the pediatrician rather than the ophthalmologist. These children merit from prompt ophthalmologic examination, because Rb is the main, or perhaps the only, life-threatening association of 13q-deletion syndrome.

Kivela¨ et al 䡠 Retinoblastoma and 13q14 Deletion Mosaicism The intraocular and trilateral Rb of our patient were diagnosed at the age of 3 months and 10 months, respectively. However, a CT scan taken at the age of 2 weeks showed an advanced Rb. The fact that she had macrocephaly and opisthotonus after birth raises the possibility that the trilateral Rb might have been congenital. However, neither the pineal tumor nor the hydrocephalus was seen by CT and magnetic resonance imaging at the age of 2 weeks and 14 weeks. Even though microcephaly and hypotony are more typical of 13q deletion, both macrocephaly21,36,37 and seizures10,33,35,38 have been noted, and it is possible that they were initially related to the deletion rather than pineoblastoma. Pineal trilateral Rb is diagnosed, on average, 25 months later than suprasellar and parasellar trilateral Rb, and it is rarely detectable at the time when intraocular Rb is diagnosed.5 In addition to encouraging immediate ophthalmologic assessment whenever a deletion that involves 13q14 is found, we would like to stress the importance of systematic reading of neuroradiologic scans by all specialists. The first CT scan taken to rule out hydrocephalus was not diagnostic of intracranial disease but showed an occult Rb that escaped the attention of the radiologist, pediatrician, geneticist, and pediatric neurologist who had an opportunity to review the scan and who knew about the 13q14 deletion. We have seen additional instances of occult intraocular tumors being missed in neuroradiologic scans. This could be avoided if all scans were to be evaluated in a systematic manner by all specialists, always paying attention both to the brain and the orbits.

11. 12.

13.

14.

15.

16.

17.

18.

19.

20. 21.

References 1. Herwig S, Strauss M. The retinoblastoma protein: a master regulator of cell cycle, differentiation and apoptosis. Eur J Biochem 1997;246:581– 601. 2. Bunin GR, Emanuel BS, Meadows AT, et al. Frequency of 13q abnormalities among 203 patients with retinoblastoma. J Natl Cancer Inst 1989;81:370 – 4. 3. Munier F, Pescia G, Jotterand-Bellomo M, et al. Constitutional karyotype in retinoblastoma. Case report and review of literature. Ophthalmic Paediatr Genet 1989;10:129 –50. 4. Moll AC, Imhof SM, Bouter LM, Tan KE. Second primary tumors in patients with retinoblastoma. A review of the literature. Ophthalmic Genet 1997;18:27–34. 5. Kivela¨ T. Trilateral retinoblastoma: a meta-analysis of hereditary retinoblastoma associated with primary ectopic intracranial retinoblastoma. J Clin Oncol 1999;17:1829 –37. 6. Kingston JE, Plowman PN, Hungerford JL. Ectopic intracranial retinoblastoma in childhood. Br J Ophthalmol 1985;69: 742– 8. 7. Blach LE, McCormick B, Abramson DH, Ellsworth RM. Trilateral retinoblastoma—incidence and outcome: a decade of experience. Int J Radiat Oncol Biol Phys 1994;29:729 –33. 8. Malik RK, Friedman HS, Djang WT, et al. Treatment of trilateral retinoblastoma with vincristine and cyclophosphamide. Am J Ophthalmol 1986;102:650 – 6. 9. Amare P, Jose J, Chitalkar P, et al. Trilateral retinoblastoma with an RB1 deletion inherited from a carrier mother: a case report. Cancer Genet Cytogenet 1999;111:28 –31. 10. Thompson H, Lyons RB. Retinoblastoma and multiple con-

22.

23.

24. 25.

26.

27. 28.

29.

30.

genital anomalies associated with complex mosaicism with deletion of D chromosome and probable D/C translocation. Hum Chrom Newslett 1965;15:21. Taylor AI. Dq-, Dr and retinoblastoma. Humangenetik 1970; 10:209 –17. Howard RO, Warburton D, Breg WR, et al. Retinoblastoma and partial deletion of the long arm of chromosome 13. Trans Am Ophthalmol Soc 1978;76:172– 83. Sparkes RS, Sparkes MC, Wilson MG, et al. Regional assignment of genes for human esterase D and retinoblastoma to chromosome band 13q14. Science 1980;208:1042– 4. Motegi T. Lymphocyte chromosome survey in 42 patients with retinoblastoma: effort to detect 13q14 deletion mosaicism. Hum Genet 1981;58:168 –73. Orye E, Benoit Y, Coppieters R, et al. A case of retinoblastoma, associated with histiocytosis-X and mosaicism of a deleted D-group chromosome (13q14 leads to q31). Clin Genet 1982;22:37–9. Motegi T. High rate of detection of 13q14 deletion mosaicism among retinoblastoma patients (using more extensive methods). Hum Genet 1982;61:95–7. Verdaguer J, Lacassie I, Cubillos TME. Syndrome 13q (interstitial deletion of chromosome 13) [in Spanish]. Arch Chiliana Oftalmol 1983;40:97–101. Ward P, Packman S, Loughman W, et al. Location of the retinoblastoma susceptibility gene(s) and the human esterase D locus. J Med Genet 1984;21:92–5. Motegi T, Minoda K. A decreasing tendency for cytogenetic abnormality in peripheral lymphocytes of retinoblastoma patients with 13q14 deletion mosaicism. Hum Genet 1984;66: 186 –9. Dudin G, Nasr A, Traboulsi E, et al. Hereditary retinoblastoma and 13q-mosaicism. Cytogenet Cell Genet 1984;38:235–7. Liberfarb RM, Bustos T, Miller WA, Sang D. Incidence and significance of a deletion of chromosome band 13q14 in patients with retinoblastoma and in their families. Ophthalmology 1984;91:1695–9. Lalande M, Donlon T, Petersen RA, et al. Molecular detection and differentiation of deletions in band 13q14 in human retinoblastoma. Cancer Genet Cytogenet 1986;23:151–7. Seidman DJ, Shields JA, Augsburger JJ, et al. Early diagnosis of retinoblastoma based on dysmorphic features and karyotype analysis. Ophthalmology 1987;94:663– 6. Turleau C, de Grouchy J. Constitutional karyotypes in retinoblastoma. Ophthalmic Pediatr Genet 1987;8:11–7. Ejima Y, Sasaki MS, Kaneko A, Tanooka H. Types, rates, origin and expressivity of chromosome mutations involving 13q14 in retinoblastoma patients. Hum Genet 1988;79:118 – 23. Ribeiro MC, Andrade JA, Erwenne CM, Brunoni D. Bilateral retinoblastoma associated with 13q-mosaicism. Possible manifestation of a germinal mutation. Cancer Genet Cytogenet 1988;32:169 –75. Liu XX, Sun YJ, Zhang L. A genetic study of retinoblastoma. J Tongji Med Univ 1991;11:220 – 4. Sippel KC, Fraioli RE, Smith GD, et al. Frequency of somatic and germ-line mosaicism in retinoblastoma: implications for genetic counseling. Am J Hum Genet 1998;62:610 –9. Smith JH, Murray TG, Fulton L, O’Brien JM. Siblings of retinoblastoma patients: are we underestimating their risk? Am J Ophthalmol 2000;129:396 – 8. DePotter P, Shields CL, Shields JA. Clinical variations of trilateral retinoblastoma: a report of 13 cases. J Pediatr Ophthalmol Strabismus 1994;31:26 –31.

1987

Ophthalmology Volume 110, Number 10, October 2003 31. Michalova´ K, Kloucek F, Musilova´ J. Deletion of 13q in two patients with retinoblastoma, one probably caused by 13qmosaicism in the mother. Hum Genet 1982;61:264 – 6. 32. Gao Y. Studies on deletions on chromosome 13 and their transmission in patients with retinoblastoma [in Chinese]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao 1991;13:157– 60. 33. Lungarotti MS, Mariotti G, Quarta C, et al. Chromosome 13 deletion syndrome: report of a new case and discussion of the different etiologic patterns of retinoblastoma. Ophthalmologica 1980;181:245–50. 34. Kingston JE, Clark J, Santos H, et al. Failure to thrive leading to early detection of retinoblastoma. Pediatr Hematol Oncol 1990;7:199 –203. 35. Kennerknecht I, Barbi G, Greher J. Diagnosis of retinoblas-

1988

toma in a presymptomatic stage after detection of interstitial chromosomal deletion 13q. Ophthalmic Genet 1994;15:19 – 24. 36. Rivera H, Turleau C, de Grouchy J, et al. Retinoblastomadel(13q14): report of two patients, one with a trisomic sib due to maternal insertion. Gene-dosage effect for esterase D. Hum Genet 1981;59:211– 4. 37. Wilson WG, Campochiaro PA, Conway BP, et al. Deletion (13)(q14.1q14.3) in two generations: variability of ocular manifestations and definition of the phenotype. Am J Med Genet 1987;28:675– 83. 38. Serena-Lungarotti M, Calabro A, Mariotti G, et al. Interstitial deletion 13q syndromes. a report on two unrelated patients. Hum Genet 1979;52:269 –74.