- Email: [email protected]
Constitutional partial 1q trisomy mosaicism and Wilms tumor
Cancer Genetics and Cytogenetics 162 (2005) 166–171
Short communication
Constitutional partial 1q trisomy mosaicism and Wilms tumor Hon Fong L. Marka,b,*, Herman Wyandta,b, Agen Pana, Jeff M. Milunskya,c,d a
Center for Human Genetics, bDepartment of Pathology, cDepartment of Pediatrics, dDepartment of Genetics and Genomics, Boston University School of Medicine, 700 Albany Street, Suite 408, Boston, MA 02118 Received 4 May 2005; received in revised form 31 May 2005; accepted 31 May 2005
Abstract
We report on a female patient with severe-profound mental retardation, multiple congenital anomalies, as well as a history of mosaicism for partial 1q trisomy in the amniotic fluid and a previous Wilms tumor specimen. Peripheral blood and fibroblasts were studied and did not demonstrate the mosaicism initially detected for 1q. Array comparative genomic hybridization yielded negative results. Additional cytogenetic studies helped clarify the previous findings and revealed evidence of partial 1q trisomy mosaicism in normal kidney tissue and in a kidney lesion. GTG-banded results showing low-percentage mosaicism for the structural rearrangement der(1)t(1;1)(p36.1;q23) in both tissues were corroborated by fluorescence in situ hybridization studies. We hypothesize that the partial 1q trisomy predisposed the target tissue (in this case kidney) to neoplasia. This study provides further support for the hypothesis that certain constitutional chromosomal abnormalities can predispose to cancer. As detection of a low-percentage mosaicism may be hampered by the limits imposed by currently available technology and the constraint of a finite sample size, extra vigilance in monitoring other somatic tissues will be needed throughout the patient’s lifetime. Anticipatory clinical guidance and prognostication are meaningful only if given accurate cytogenetic diagnoses. To the best of our knowledge, this is the first reported case of Wilms tumor associated with constitutional partial 1q trisomy, either in pure or mosaic form, with the particular 1q23 breakpoint in conjunction with a break on 1p36.1. Ó 2005 Elsevier Inc. All rights reserved.
1. Introduction A constitutional cytogenetic abnormality is a pre-existing numerical or structural chromosomal abnormality that can be detected prenatally or after birth, while an acquired cytogenetic abnormality is a secondary chromosomal abnormality associated with (or ‘‘acquired’’ through) a disease process. Examples of constitutional cytogenetic abnormalities are trisomy 21, and Turner and Klinefelter syndromes. Examples of acquired cytogenetic abnormalities are the Philadelphia translocation and inversion of chromosome 16 seen in leukemia. The topic of constitutional chromosomal abnormality and cancer fascinates many because it questions the fundamental way that we view the process of carcinogenesis. For example, it has been hypothesized by us [1] as well as others [2–8] that there is an association between constitutional trisomy 8 mosaicism and neoplasia. General acceptance of this hypothesis will
* Corresponding author. Tel.: 617-638-7083. E-mail addresses: [email protected], [email protected] (H.F.L. Mark). 0165-4608/05/$ – see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cancergencyto.2005.05.012
alter our current view on the role of constitutional versus acquired chromosomal abnormalities in cancer. The possibility of constitutional mosaicism for a partial 1q trisomy was raised by a report of this abnormality in a previous amniocentesis sample and Wilms tumor specimen taken from the proband’s left kidney. Although the so-called ‘‘partial trisomy 1q syndrome’’ has been described in the literature, in our case, the breakpoints involved in the rearrangement, 1q23 and 1p36.1, are unique. The association of constitutional partial 1q trisomy with these unique breakpoints and an increased risk for neoplasia is of both theoretical and practical importance.
2. Clinical case report The female proband has been followed by one of us (J.M.M.) for approximately 5 years. She was the nonconsanguineous product of a full-term pregnancy complicated by ultrasound detection of fetal hydrocephalus and a Dandy-Walker malformation. An amniocentesis (performed elsewhere) revealed 9 out of 40 cells with partial
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
trisomy 1q by conventional cytogenetics. Parental chromosomes were normal. Delivery was uncomplicated, and birth weight was 8 pounds 5 ounces. Cord blood chromosomes revealed 20 normal cells (performed elsewhere). There was no neonatal jaundice or seizures. A cardiology evaluation initially revealed a patent ductus arteriosus, atrial septal defect, and ventricular septal defect, all of which have resolved without intervention. At 9 months of age, the proband had a left inguinal herniorrhaphy. Due to an ear malformation, a renal ultrasound was performed which revealed lesions in both kidneys. Further workup with an MRI and later a kidney biopsy revealed nephroblastomatosis, which was monitored every 2 months. Her kidney function was apparently normal. She was diagnosed with esotropia, which required surgery. A head MRI revealed agenesis of the corpus callosum and slight hydrocephalus, which did not require a shunt. The Dandy-Walker malformation was not confirmed postnatally. A skeletal survey at 4 years was normal. She has had failure to thrive. Her developmental milestones have been severely delayed. At 7 years, she is unable to stand by herself but can sit unassisted. Although she can follow some simple commands, she cannot speak any identifiable words. She had occasional bruxism and biting behavior. She did not have repetitive behaviors and may have had a seizure at 7 years. She has had an electroencephalogram with abnormal background and a normal audiology evaluation. A physical exam at 3 years revealed that weight and height were in the 3rd centile, and head circumference was below the 3rd centile. The anterior fontanel was closed. There were no overriding sutures. Dysmorphic features (Fig. 1) included a broad nasal bridge, bilateral epicanthal folds, a smooth philtrum, a thin upper vermillion border, a slightly high-arched palate, esotropia, telecanthus, and low-set, posteriorly rotated ears. Both ears were cup shaped. The chest was symmetric with wide-appearing
Fig. 1. Proband.
167
nipples. She had left abnormal dermatoglyphics with tapered fingers. She had right overlapping toes IV/V. A Woods lamp demonstrated scattered hypopigmentation. She was hypotonic with increased extremity tone, upgoing toes, and increased deep tendon reflexes. She was active, alert, and did not have clonus or nystagmus. The rest of her physical exam was essentially unremarkable. At 7 years, bilateral finger camptodactyly was noted. Subsequent to a left kidney resection for stage 1 Wilms tumor at 21 months, her remaining right kidney was monitored using a kidney MRI surveillance protocol. A partial nephrectomy was performed on her remaining right kidney at 6 ½ years due to relapsed Wilms tumor. Both normal and tumor tissues from the surgery were obtained for chromosomal analysis.
3. Materials and methods Harvesting and G-banding were performed according to standard procedures. One hundred cells were analyzed to rule out low-percentage mosaicism. Fluorescent in situ hybridization (FISH) was performed as an adjunct to conventional cytogenetics according to standard protocols. Subtelomeric FISH using commercial chromosome 1– specific probes for 1p and 1q (TelVysion 1p SpectrumGreen and TelVysion 1q SpectrumOrange; Vysis, Downers Grove, IL) was performed according to the manufacturer’s instructions to confirm the structural rearrangement observed in G-banded preparations. Computer-assisted analyses were performed using a Zeiss Axioskop 2 fluorescent microscope with an integrating CCD camera (Photometrics, Tucson, AZ).
4. Results Amniotic fluid collected elsewhere at 32-weeks gestation revealed the presence of a partial 1q trisomy in 9 of 40 cells examined. A G-banded study of cord blood revealed a normal female karyotype, 46,XX, in 20 metaphases examined. To rule out a familial chromosomal rearrangement involving the same structural chromosomal abnormality detected at amniocentesis, the parents’ bloods were analyzed. Parental karyotypes were found to be normal by G-banded analyses. Subsequent analysis of the Wilms tumor tissue from the left kidney also detected the presence of the same partial 1q trisomy. Each of 20 metaphase cells contained this clonal aberration. The report also found an unusual 22q rearrangement for a Wilms tumor, 46,XX,der(1)t(1;1)(p36.1;q23),add(22)(q11.2). As can be seen in the karyotype designation, this rearrangement involves the chromosome region 22q11.2, which contains the Ewing’s (EWS) gene locus. Subsequent FISH performed elsewhere revealed that the EWS locus was deleted in each of the five metaphases cells examined. However,
168
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
there was no evidence of any other structural rearrangements, such as translocations involving the EWS locus. A blood sample was subsequently obtained from the proband, and chromosomal analysis revealed an apparently normal female karyotype, 46,XX, based on an examination of 50 cells. A skin biopsy was also performed to rule out mosaicism for partial chromosome 1q trisomy. Examination of 100 cells revealed an apparently normal female karyotype, 46,XX. In addition, this same tissue was found to be negative for cryptic rearrangement or deletion by subtelomeric FISH. Another FISH test to detect the deletion of N25 in the DiGeorge/VCFS region on 22q11.2 and the ARSA region on 22q13.3 also yielded negative results in fibroblasts. The result of array comparative genomic hybridization analysis (performed elsewhere) was negative. Surgery at a later date for the Wilms tumor on one side (nephroblastomatosis) made additional somatic tissues available for cytogenetic studies. Both normal and tumor tissues were obtained, and approximately 100 cells on each were examined. The results of these studies are as follows: Normal kidney tissue analyzed by G banding (Fig. 2) revealed the presence of a single abnormal cell with an identical structural chromosomal abnormality, as described previously (low percentage mosaicism): 46,XX,der(1) t(1;1)(p36.1;q23)[1]/46,XX[122]. Subtelomeric FISH performed on the normal tissue, however, revealed additional evidence of the abnormal cell line. Subtelomeric FISH yielded mainly normal cells with two normal chromosomes 1 with one green and one red signal found on 1p and 1q, respectively, in both homologs. Five cells were found with an abnormal signal pattern (red signals on both ends of one chromosome 1 homolog and a normal red and green signal pattern for the other homolog). Kidney tumor tissue analyzed by G banding (Fig. 2) also revealed the presence of an abnormal clone (mosaicism): 46,XX,der(1)t(1;1)(p36.1;q23)[7]/46,XX[89]. Any additional aneuploidy found was not consistent in the abnormal cells and, therefore, was not interpreted to be clonal. The result from subtelomeric FISH on the tumor tissue was confirmatory in that the subtelomere FISH yielded both normal cells with one red and one green signal on both chromosome 1 homologs and abnormal cells with red signals on both ends of one homolog and one green and one red signal on p and q, respectively, for the other homolog. Representative FISH results are given in Fig. 3. Fig. 4 summarizes the chromosomal composition of der(1) relative to the normal chromosome 1 based on the results of G-banding and FISH.
5. Discussion Although it has been known that there is an increased incidence of cancer in individuals with constitutional chromosomal abnormalities, such as trisomy 21, the
Fig. 2. Abnormal G-banded karyotypes of the proband from (a) normal kidney tissue showing the partial trisomy 1q (large arrow), 46,XX, der(1)t(1;1)(p36.1;q23), and from (b) kidney tumor showing partial trisomy 1q (large arrow). Additional aneuploidy for chromosomes 6 and 13 (small arrows) was not consistent in abnormal cells and, therefore, was not interpreted to be clonal (see text).
hypothesis of an association between constitutional chromosomal abnormality and cancer was put forth for gestational trophoblastic disease a decade ago [1,2]. The concurrence of constitutional trisomy 8 mosaicism and gestational trophoblastic disease provides support for a broader hypothesis of an increased predisposition to cancer in tissues with constitutional genomic imbalance, which can manifest itself as numerical chromosomal abnormalities (e.g., trisomies) or structural chromosomal abnormalities (e.g., translocations). The study mentioned above was an example of the former; the present study is an example of the latter. The general assumption has been that recurring trisomies and structural rearrangements often observed in cancer are acquired abnormalities associated with the disease process. To unequivocally demonstrate the presence of a constitutional chromosomal abnormality, a clonal abnormality needs to be found in the normal tissue. A positive finding of the derivative chromosome 1 in our proband’s normal kidney tissue in addition to the amniotic
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
Fig. 3. FISH using commercial chromosome 1–specific subtelomeric probes (Vysis) labeled with SpectrumGreen for the end of the short arm and SpectrumOrange for the end of the long arm. Cells are counterstained with DAPI. (a) Normal metaphase from normal kidney tissue with red and green signals at opposite ends of both chromosome 1 homologs. (b) Two closely adjacent metaphases from normal kidney tissue, each showing one normal homolog (n) with red and green signals at opposite ends and one homolog with duplication (dup) of 1q revealed by a red signal on each end of the chromosome. (c) Metaphase from tumor tissue with a normal chromosome 1 and the homolog with duplication of 1q (dup). Both normal and abnormal interphase cells are also evident: n 5 normal FISH pattern, dup 5 duplication of probe for 1q.
fluid and kidney tumor tissue demonstrate the constitutional origin of the partial 1q trisomy. According to published literature, partial 1q trisomy is a rare cytogenetic abnormality [9–14]. Most cases have duplications as a result of complex rearrangements involving other chromosomes. To the best of our knowledge, the breakpoints that we observed here have not been described previously. The case most similar to ours [12]
169
Fig. 4. A schematic illustration showing an International System for Human Cytogenetic Nomenclature ideogram of a normal chromosome 1 on the left and a modified ideogram representing composition of the der(1)t(1;1)(p36.1;q23) on the right. Large arrows on the normal ideogram (left) represent breakpoints. Bracketed region represents the distal segment of 1p from 1p36~pter that has been lost from the der(1). Bracketed region on the der(1) ideogram (right) represents the distal segment of 1q from q23~qter that is duplicated.
was a prenatal diagnosis of trisomy 1q21~qter resulting from a 1;1 translocation. The karyotype given was 46,XY,der(1)(qter/q21::p36.3/qter), but the malformed fetus had a pure (not mosaic) partial 1q trisomy. The presence of the unique partial 1q trisomy (1q23~qter) in our proband raises the provocative question of its etiologic relation to the observed cancer. We hypothesize that the presence of partial 1q trisomy increases the risk of cancer in the tissues possessing the abnormal clone. The tissue involved in our case originated
170
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
from the kidney; the cancer being manifested was Wilms tumor. Although an association has been reported between a novel constitutional chromosomal translocation, t(1;7)(q42;p15), and Wilms tumor (and skeletal abnormalities), the Wilms tumor gene was hypothesized to be on 7p and not on 1q [15,16]. Although partial 1q trisomy may not be a common cytogenetic abnormality in Wilms tumor, trisomy for most or all of 1q have been observed in this tumor [17]. Most of these are considered acquired abnormalities due to disease-related processes and are involved in tumor progression. Structural rearrangements of the long arm of chromosome 1, der(1q), have also been reported in other cancers besides Wilms tumor. Rearrangement of chromosome 1q has been reported as one of the most common secondary chromosomal abnormalities in Burkitt’s lymphoma and in chronic myeloproliferative disorders (e.g., myelofibrosis and polycythemia vera). Although Baumgarten et al. [18] reported a young patient with erythroleukemia with a partial trisomy of the long arm of a chromosome 1 from 1q23~qter, the patient also had two other confounding chromosomal abnormalities that made a simple interpretation difficult. In another patient with ‘‘preleukemic syndrome,’’ a duplication of 1q23~qter was described. Anderson et al. [19] speculated that partial trisomy 1q represents a nonrandom chromosomal abnormality in MDS. Translocations involving 1q have been reported in renal cell carcinoma [20]. One study reported in the literature also documented Wolff-Parkinson-White syndrome, another with anterior segment dysgenesis and congenital glaucoma, and another with camptodactyly. Schorry [21] reported growth hormone deficiency and normal intelligence. We hypothesize that the same chromosomal structural rearrangement found in different somatic tissues may have led to the development of different cancers in these cases. Our derivative chromosome 1, der(1), is missing distal 1p (1p36.1~pter) in addition to being trisomic for 1q23~qter. The exact significance of the deletion (monosomy) for the telomeric end of 1p is not known, although the del(1)(p36.1) syndrome, a contiguous gene deletion syndrome and reportedly the most common terminal deletion syndrome [22], is presumably caused by haploinsufficiency of a number of genes. The constitutional deletion of 1p36 reportedly results in a syndrome with multiple congenital anomalies and mental retardation. Chromosomal 1p36 alterations, mostly deletions, have also been reported to occur in various neoplasms, including neuroblastoma, prostate cancer, lung cancer, malignant melanoma, hepatoma, cervical carcinoma, breast cancer, colorectal adenocarcinoma, ovarian cancer, and non-Hodgkin lymphoma. The identification of deletions of 1p36 in these neoplasms may indicate that the 1p36 region contains a number of tumorsuppressor genes and that deletion of one or more of these genes is involved in the chain of events that results in malignancy in various somatic tissues. Phenotypically, our proband does not have the typical features associated with
1p36 monosomy and, hence, her phenotype likely reflects her partial trisomy 1q mosaicism. The possibility of the constitutional partial 1q trisomy playing a causative role in the initiation and development of Wilms tumor is intriguing. While the exact mechanism of carcinogenesis is not known, one can hypothesize that partial 1q trisomy in those target tissues may result in an increased rate of cell proliferation. In addition, the region 1p36.1 may harbor tumor suppressor genes. Deletion of these chromosome 1p tumor suppressor genes and extra dosage of chromosome 1q oncogenes may act singly or synergistically. Thus, in combination or alone, partial 1q trisomy and 1p36.1 monosomy may lead to cancer. The finding of constitutional chromosomal abnormalities detected in certain somatic tissues has significant clinical ramifications. Once a clonal chromosomal abnormality is detected in a target tissue, the patient must be carefully monitored for additional signs and symptoms. In our case, we recommend monitoring the kidneys periodically by abdominal imaging. For our proband, prognostication depends in part on the detection of this clonal structural chromosomal abnormality in other somatic tissues as the patient grows and develops. In conclusion, to the best of our knowledge, this is the first reported case of Wilms tumor associated with constitutional partial 1q trisomy, either in pure or mosaic form, with the particular 1q23 breakpoint in conjunction with a break on 1p36.1. Aside from its specific relevance to the pathology of Wilms tumor, the present case may hold new and important clues to the mechanism of carcinogenesis in general, as advances in human genetics continue to provide new insights for understanding various chromosomally induced diseases.
Acknowledgments We thank Dr. Jonathan Fletcher for sharing some of his cytogenetic results and the staff of the Cytogenetics Laboratory at the Center for Human Genetics, Boston University School of Medicine, for lending their cytogenetic expertise. This study would not have been possible without the cooperation of the proband’s family and the family’s physician. We are grateful to all involved. The support of Dr. Roger Mark is also acknowledged. References [1] Mark HFL, Ahearn J, Lathrop JC. Constitutional trisomy 8 mosaicism and gestational trophoblastic disease. Cancer Genetics Cytogenetics 1995;80:150–4. [2] Hecht F. Trisomy and disomy in tumors (EDITORIAL). Cancer Genetics Cytogenetics 1995;80:171. [3] Gafter U, Shabtai F, Kahn Y, Halbrecht I, Djaldetti M. Aplastic anemia followed by leukemia in congenital trisomy 8 mosaicism. Clin Genet 1976;9:134–42. [4] Haas OA, Seyger M. Hypothesis: meiotic origin of trisomic neoplasms. Cancer Genet Cytogenet 1993;70:112–6.
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171 [5] Kapaun P, Kabisch H, Held KR, Walter TA, Hegewisch S, Zander AR. Atypical chronic myelogenous leukemia in a patient with trisomy 8 mosaicism syndrome. Ann Hematol 1993;66:57–8. [6] Mark HFL. Phenotypic variability in trisomy 8 mosaicism is consistent with the hypothesis of meiotic origin of trisomic neoplasms. Cancer Genet Cytogenet 1994;76:158. [7] McDonald-McGinn DM, Grace K, Spinner NB, Emanuel BS, Zackai EH. Clinical and cytogenetic variability in trisomy 8 mosaicism. Am J Hum Gen 1993;53(Suppl):222. [8] Riccardi VM, Humbert J, Peakman D. Familial cancer, reciprocal translocation [t(7p;20p)] and trisomy 8. Am J Hum Genet 1975;27: 76A. [9] Rasmussen SA, Frias JL, Lafer CZ, Eunpu DL, Zackai EH. Partial duplication 1q: Report of four patients and review of the literature. Am J Med Genet 1990;36:137–43. [10] Bartsch C, Aslan M, Kohler J, Miny P, Horst J, Holzgreve W, Rehder H, Fritz B. Duplication dup(1)(q32q44) detected by comparative genomic hybridization (CGH): further delineation of trisomies 1q. Fetal Diagn Ther 2001;16:265–73. [11] Kimya Y, Yakut T, Egeli U, Ozerkan K. Prenatal diagnosis of a fetus with pure partial trisomy 1q32-44 due to a familial balanced rearrangement. Prenat Diagn 2002;22:957–61. [12] Machlitt A, Kuepferling P, Bommer C, Koerner H, Chaoui R. Prenatal diagnosis of trisomy 1q21~qter: case report and review of literature. Amer J Med Genet 2005;134A:207–11. [13] Van Buggenhout G, De Coen L, Fryns JP. Partial trisomy 1q (1q32/1qter) in adulthood: further delineation of the phenotype. Ann Genet 1998;41:77–81. [14] Fernandez-Novoa MC, Vargas MT, Grannell MR, Carreto P. Prenatal diagnosis of de novo trisomy 1(q21-qter)der(Y)t(Y;1) in a malformed live born. Prenat Diagn 2004;24:414–7.
171
[15] Vernon EG, Malik K, Reynolds P, Powlesland R, Dallosso AR, Jackson S, Henthorn K, Green ED, Brown KW. The parathyroid hormone-responsive B1 gene is interrupted by a t(1;7)(q42;p15) breakpoint associated with Wilms tumour. Oncogene 2003;22: 1371–80. [16] Reynold PA, Powlesland RM, Keen TJ, Inglehearn CF, Cunningham AF. Localization of a novel t(1;7) translocation associated with Wilms tumor predisposition and skeletal abnormalities. Genes Chromosomes Cancer 1996;17:151–5. [17] Betts DR, Ilg EC, Oezahin H, von der Weid N, Niggli FK. Trisomy 1q generating translocations in Wilms tumor. Cancer Genet Cytogenet 1999;112:138–43. [18] Baumgarten E, Wegner RD, Fengler R, Koch H, Henze G. Partial trisomy 1q, an uncommon chromosomal aberration in erythroleukemia. Leuk Lymphoma 1993;10:237–40. [19] Anderson RL, Bagby GC. The prognostic value of chromosome studies in patients with the preleukemic syndrome (hemopoietic dysplasia). Leukemia Res 1982;6:175–81. [20] Desangles F. Kidney: t(X;1)(p11.2;q21.2) in renal cell carcinoma. Atlas Genet Cytogenet Oncol Haematol. March 1998. Available at: http://www.infobiogen.fr/services/chromcancer/Tumors/tX1ID5009. html. Accessed September 8, 2005. [21] Schorry EK, Dietrich KN, Saal HM, Blough RI, Dey S, Chernausek S, Milatovich-Cherry A. Partial trisomy 1q with growth hormone deficiency and normal intelligence. Am J Med Genet 1998;77:257–60. [22] Heilstedt HA, Ballif BC, Howard LA, Lewis RA, Stal S, Kashork CD, Bacino CA, Shapira SK, Shaffer LG. Physical map of 1p36, placement of breakpoints in monosomy 1p36, and clinical characterization of the syndrome. Am J Hum Genet 2003;72: 1200–12.
Short communication
Constitutional partial 1q trisomy mosaicism and Wilms tumor Hon Fong L. Marka,b,*, Herman Wyandta,b, Agen Pana, Jeff M. Milunskya,c,d a
Center for Human Genetics, bDepartment of Pathology, cDepartment of Pediatrics, dDepartment of Genetics and Genomics, Boston University School of Medicine, 700 Albany Street, Suite 408, Boston, MA 02118 Received 4 May 2005; received in revised form 31 May 2005; accepted 31 May 2005
Abstract
We report on a female patient with severe-profound mental retardation, multiple congenital anomalies, as well as a history of mosaicism for partial 1q trisomy in the amniotic fluid and a previous Wilms tumor specimen. Peripheral blood and fibroblasts were studied and did not demonstrate the mosaicism initially detected for 1q. Array comparative genomic hybridization yielded negative results. Additional cytogenetic studies helped clarify the previous findings and revealed evidence of partial 1q trisomy mosaicism in normal kidney tissue and in a kidney lesion. GTG-banded results showing low-percentage mosaicism for the structural rearrangement der(1)t(1;1)(p36.1;q23) in both tissues were corroborated by fluorescence in situ hybridization studies. We hypothesize that the partial 1q trisomy predisposed the target tissue (in this case kidney) to neoplasia. This study provides further support for the hypothesis that certain constitutional chromosomal abnormalities can predispose to cancer. As detection of a low-percentage mosaicism may be hampered by the limits imposed by currently available technology and the constraint of a finite sample size, extra vigilance in monitoring other somatic tissues will be needed throughout the patient’s lifetime. Anticipatory clinical guidance and prognostication are meaningful only if given accurate cytogenetic diagnoses. To the best of our knowledge, this is the first reported case of Wilms tumor associated with constitutional partial 1q trisomy, either in pure or mosaic form, with the particular 1q23 breakpoint in conjunction with a break on 1p36.1. Ó 2005 Elsevier Inc. All rights reserved.
1. Introduction A constitutional cytogenetic abnormality is a pre-existing numerical or structural chromosomal abnormality that can be detected prenatally or after birth, while an acquired cytogenetic abnormality is a secondary chromosomal abnormality associated with (or ‘‘acquired’’ through) a disease process. Examples of constitutional cytogenetic abnormalities are trisomy 21, and Turner and Klinefelter syndromes. Examples of acquired cytogenetic abnormalities are the Philadelphia translocation and inversion of chromosome 16 seen in leukemia. The topic of constitutional chromosomal abnormality and cancer fascinates many because it questions the fundamental way that we view the process of carcinogenesis. For example, it has been hypothesized by us [1] as well as others [2–8] that there is an association between constitutional trisomy 8 mosaicism and neoplasia. General acceptance of this hypothesis will
* Corresponding author. Tel.: 617-638-7083. E-mail addresses: [email protected], [email protected] (H.F.L. Mark). 0165-4608/05/$ – see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cancergencyto.2005.05.012
alter our current view on the role of constitutional versus acquired chromosomal abnormalities in cancer. The possibility of constitutional mosaicism for a partial 1q trisomy was raised by a report of this abnormality in a previous amniocentesis sample and Wilms tumor specimen taken from the proband’s left kidney. Although the so-called ‘‘partial trisomy 1q syndrome’’ has been described in the literature, in our case, the breakpoints involved in the rearrangement, 1q23 and 1p36.1, are unique. The association of constitutional partial 1q trisomy with these unique breakpoints and an increased risk for neoplasia is of both theoretical and practical importance.
2. Clinical case report The female proband has been followed by one of us (J.M.M.) for approximately 5 years. She was the nonconsanguineous product of a full-term pregnancy complicated by ultrasound detection of fetal hydrocephalus and a Dandy-Walker malformation. An amniocentesis (performed elsewhere) revealed 9 out of 40 cells with partial
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
trisomy 1q by conventional cytogenetics. Parental chromosomes were normal. Delivery was uncomplicated, and birth weight was 8 pounds 5 ounces. Cord blood chromosomes revealed 20 normal cells (performed elsewhere). There was no neonatal jaundice or seizures. A cardiology evaluation initially revealed a patent ductus arteriosus, atrial septal defect, and ventricular septal defect, all of which have resolved without intervention. At 9 months of age, the proband had a left inguinal herniorrhaphy. Due to an ear malformation, a renal ultrasound was performed which revealed lesions in both kidneys. Further workup with an MRI and later a kidney biopsy revealed nephroblastomatosis, which was monitored every 2 months. Her kidney function was apparently normal. She was diagnosed with esotropia, which required surgery. A head MRI revealed agenesis of the corpus callosum and slight hydrocephalus, which did not require a shunt. The Dandy-Walker malformation was not confirmed postnatally. A skeletal survey at 4 years was normal. She has had failure to thrive. Her developmental milestones have been severely delayed. At 7 years, she is unable to stand by herself but can sit unassisted. Although she can follow some simple commands, she cannot speak any identifiable words. She had occasional bruxism and biting behavior. She did not have repetitive behaviors and may have had a seizure at 7 years. She has had an electroencephalogram with abnormal background and a normal audiology evaluation. A physical exam at 3 years revealed that weight and height were in the 3rd centile, and head circumference was below the 3rd centile. The anterior fontanel was closed. There were no overriding sutures. Dysmorphic features (Fig. 1) included a broad nasal bridge, bilateral epicanthal folds, a smooth philtrum, a thin upper vermillion border, a slightly high-arched palate, esotropia, telecanthus, and low-set, posteriorly rotated ears. Both ears were cup shaped. The chest was symmetric with wide-appearing
Fig. 1. Proband.
167
nipples. She had left abnormal dermatoglyphics with tapered fingers. She had right overlapping toes IV/V. A Woods lamp demonstrated scattered hypopigmentation. She was hypotonic with increased extremity tone, upgoing toes, and increased deep tendon reflexes. She was active, alert, and did not have clonus or nystagmus. The rest of her physical exam was essentially unremarkable. At 7 years, bilateral finger camptodactyly was noted. Subsequent to a left kidney resection for stage 1 Wilms tumor at 21 months, her remaining right kidney was monitored using a kidney MRI surveillance protocol. A partial nephrectomy was performed on her remaining right kidney at 6 ½ years due to relapsed Wilms tumor. Both normal and tumor tissues from the surgery were obtained for chromosomal analysis.
3. Materials and methods Harvesting and G-banding were performed according to standard procedures. One hundred cells were analyzed to rule out low-percentage mosaicism. Fluorescent in situ hybridization (FISH) was performed as an adjunct to conventional cytogenetics according to standard protocols. Subtelomeric FISH using commercial chromosome 1– specific probes for 1p and 1q (TelVysion 1p SpectrumGreen and TelVysion 1q SpectrumOrange; Vysis, Downers Grove, IL) was performed according to the manufacturer’s instructions to confirm the structural rearrangement observed in G-banded preparations. Computer-assisted analyses were performed using a Zeiss Axioskop 2 fluorescent microscope with an integrating CCD camera (Photometrics, Tucson, AZ).
4. Results Amniotic fluid collected elsewhere at 32-weeks gestation revealed the presence of a partial 1q trisomy in 9 of 40 cells examined. A G-banded study of cord blood revealed a normal female karyotype, 46,XX, in 20 metaphases examined. To rule out a familial chromosomal rearrangement involving the same structural chromosomal abnormality detected at amniocentesis, the parents’ bloods were analyzed. Parental karyotypes were found to be normal by G-banded analyses. Subsequent analysis of the Wilms tumor tissue from the left kidney also detected the presence of the same partial 1q trisomy. Each of 20 metaphase cells contained this clonal aberration. The report also found an unusual 22q rearrangement for a Wilms tumor, 46,XX,der(1)t(1;1)(p36.1;q23),add(22)(q11.2). As can be seen in the karyotype designation, this rearrangement involves the chromosome region 22q11.2, which contains the Ewing’s (EWS) gene locus. Subsequent FISH performed elsewhere revealed that the EWS locus was deleted in each of the five metaphases cells examined. However,
168
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
there was no evidence of any other structural rearrangements, such as translocations involving the EWS locus. A blood sample was subsequently obtained from the proband, and chromosomal analysis revealed an apparently normal female karyotype, 46,XX, based on an examination of 50 cells. A skin biopsy was also performed to rule out mosaicism for partial chromosome 1q trisomy. Examination of 100 cells revealed an apparently normal female karyotype, 46,XX. In addition, this same tissue was found to be negative for cryptic rearrangement or deletion by subtelomeric FISH. Another FISH test to detect the deletion of N25 in the DiGeorge/VCFS region on 22q11.2 and the ARSA region on 22q13.3 also yielded negative results in fibroblasts. The result of array comparative genomic hybridization analysis (performed elsewhere) was negative. Surgery at a later date for the Wilms tumor on one side (nephroblastomatosis) made additional somatic tissues available for cytogenetic studies. Both normal and tumor tissues were obtained, and approximately 100 cells on each were examined. The results of these studies are as follows: Normal kidney tissue analyzed by G banding (Fig. 2) revealed the presence of a single abnormal cell with an identical structural chromosomal abnormality, as described previously (low percentage mosaicism): 46,XX,der(1) t(1;1)(p36.1;q23)[1]/46,XX[122]. Subtelomeric FISH performed on the normal tissue, however, revealed additional evidence of the abnormal cell line. Subtelomeric FISH yielded mainly normal cells with two normal chromosomes 1 with one green and one red signal found on 1p and 1q, respectively, in both homologs. Five cells were found with an abnormal signal pattern (red signals on both ends of one chromosome 1 homolog and a normal red and green signal pattern for the other homolog). Kidney tumor tissue analyzed by G banding (Fig. 2) also revealed the presence of an abnormal clone (mosaicism): 46,XX,der(1)t(1;1)(p36.1;q23)[7]/46,XX[89]. Any additional aneuploidy found was not consistent in the abnormal cells and, therefore, was not interpreted to be clonal. The result from subtelomeric FISH on the tumor tissue was confirmatory in that the subtelomere FISH yielded both normal cells with one red and one green signal on both chromosome 1 homologs and abnormal cells with red signals on both ends of one homolog and one green and one red signal on p and q, respectively, for the other homolog. Representative FISH results are given in Fig. 3. Fig. 4 summarizes the chromosomal composition of der(1) relative to the normal chromosome 1 based on the results of G-banding and FISH.
5. Discussion Although it has been known that there is an increased incidence of cancer in individuals with constitutional chromosomal abnormalities, such as trisomy 21, the
Fig. 2. Abnormal G-banded karyotypes of the proband from (a) normal kidney tissue showing the partial trisomy 1q (large arrow), 46,XX, der(1)t(1;1)(p36.1;q23), and from (b) kidney tumor showing partial trisomy 1q (large arrow). Additional aneuploidy for chromosomes 6 and 13 (small arrows) was not consistent in abnormal cells and, therefore, was not interpreted to be clonal (see text).
hypothesis of an association between constitutional chromosomal abnormality and cancer was put forth for gestational trophoblastic disease a decade ago [1,2]. The concurrence of constitutional trisomy 8 mosaicism and gestational trophoblastic disease provides support for a broader hypothesis of an increased predisposition to cancer in tissues with constitutional genomic imbalance, which can manifest itself as numerical chromosomal abnormalities (e.g., trisomies) or structural chromosomal abnormalities (e.g., translocations). The study mentioned above was an example of the former; the present study is an example of the latter. The general assumption has been that recurring trisomies and structural rearrangements often observed in cancer are acquired abnormalities associated with the disease process. To unequivocally demonstrate the presence of a constitutional chromosomal abnormality, a clonal abnormality needs to be found in the normal tissue. A positive finding of the derivative chromosome 1 in our proband’s normal kidney tissue in addition to the amniotic
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
Fig. 3. FISH using commercial chromosome 1–specific subtelomeric probes (Vysis) labeled with SpectrumGreen for the end of the short arm and SpectrumOrange for the end of the long arm. Cells are counterstained with DAPI. (a) Normal metaphase from normal kidney tissue with red and green signals at opposite ends of both chromosome 1 homologs. (b) Two closely adjacent metaphases from normal kidney tissue, each showing one normal homolog (n) with red and green signals at opposite ends and one homolog with duplication (dup) of 1q revealed by a red signal on each end of the chromosome. (c) Metaphase from tumor tissue with a normal chromosome 1 and the homolog with duplication of 1q (dup). Both normal and abnormal interphase cells are also evident: n 5 normal FISH pattern, dup 5 duplication of probe for 1q.
fluid and kidney tumor tissue demonstrate the constitutional origin of the partial 1q trisomy. According to published literature, partial 1q trisomy is a rare cytogenetic abnormality [9–14]. Most cases have duplications as a result of complex rearrangements involving other chromosomes. To the best of our knowledge, the breakpoints that we observed here have not been described previously. The case most similar to ours [12]
169
Fig. 4. A schematic illustration showing an International System for Human Cytogenetic Nomenclature ideogram of a normal chromosome 1 on the left and a modified ideogram representing composition of the der(1)t(1;1)(p36.1;q23) on the right. Large arrows on the normal ideogram (left) represent breakpoints. Bracketed region represents the distal segment of 1p from 1p36~pter that has been lost from the der(1). Bracketed region on the der(1) ideogram (right) represents the distal segment of 1q from q23~qter that is duplicated.
was a prenatal diagnosis of trisomy 1q21~qter resulting from a 1;1 translocation. The karyotype given was 46,XY,der(1)(qter/q21::p36.3/qter), but the malformed fetus had a pure (not mosaic) partial 1q trisomy. The presence of the unique partial 1q trisomy (1q23~qter) in our proband raises the provocative question of its etiologic relation to the observed cancer. We hypothesize that the presence of partial 1q trisomy increases the risk of cancer in the tissues possessing the abnormal clone. The tissue involved in our case originated
170
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171
from the kidney; the cancer being manifested was Wilms tumor. Although an association has been reported between a novel constitutional chromosomal translocation, t(1;7)(q42;p15), and Wilms tumor (and skeletal abnormalities), the Wilms tumor gene was hypothesized to be on 7p and not on 1q [15,16]. Although partial 1q trisomy may not be a common cytogenetic abnormality in Wilms tumor, trisomy for most or all of 1q have been observed in this tumor [17]. Most of these are considered acquired abnormalities due to disease-related processes and are involved in tumor progression. Structural rearrangements of the long arm of chromosome 1, der(1q), have also been reported in other cancers besides Wilms tumor. Rearrangement of chromosome 1q has been reported as one of the most common secondary chromosomal abnormalities in Burkitt’s lymphoma and in chronic myeloproliferative disorders (e.g., myelofibrosis and polycythemia vera). Although Baumgarten et al. [18] reported a young patient with erythroleukemia with a partial trisomy of the long arm of a chromosome 1 from 1q23~qter, the patient also had two other confounding chromosomal abnormalities that made a simple interpretation difficult. In another patient with ‘‘preleukemic syndrome,’’ a duplication of 1q23~qter was described. Anderson et al. [19] speculated that partial trisomy 1q represents a nonrandom chromosomal abnormality in MDS. Translocations involving 1q have been reported in renal cell carcinoma [20]. One study reported in the literature also documented Wolff-Parkinson-White syndrome, another with anterior segment dysgenesis and congenital glaucoma, and another with camptodactyly. Schorry [21] reported growth hormone deficiency and normal intelligence. We hypothesize that the same chromosomal structural rearrangement found in different somatic tissues may have led to the development of different cancers in these cases. Our derivative chromosome 1, der(1), is missing distal 1p (1p36.1~pter) in addition to being trisomic for 1q23~qter. The exact significance of the deletion (monosomy) for the telomeric end of 1p is not known, although the del(1)(p36.1) syndrome, a contiguous gene deletion syndrome and reportedly the most common terminal deletion syndrome [22], is presumably caused by haploinsufficiency of a number of genes. The constitutional deletion of 1p36 reportedly results in a syndrome with multiple congenital anomalies and mental retardation. Chromosomal 1p36 alterations, mostly deletions, have also been reported to occur in various neoplasms, including neuroblastoma, prostate cancer, lung cancer, malignant melanoma, hepatoma, cervical carcinoma, breast cancer, colorectal adenocarcinoma, ovarian cancer, and non-Hodgkin lymphoma. The identification of deletions of 1p36 in these neoplasms may indicate that the 1p36 region contains a number of tumorsuppressor genes and that deletion of one or more of these genes is involved in the chain of events that results in malignancy in various somatic tissues. Phenotypically, our proband does not have the typical features associated with
1p36 monosomy and, hence, her phenotype likely reflects her partial trisomy 1q mosaicism. The possibility of the constitutional partial 1q trisomy playing a causative role in the initiation and development of Wilms tumor is intriguing. While the exact mechanism of carcinogenesis is not known, one can hypothesize that partial 1q trisomy in those target tissues may result in an increased rate of cell proliferation. In addition, the region 1p36.1 may harbor tumor suppressor genes. Deletion of these chromosome 1p tumor suppressor genes and extra dosage of chromosome 1q oncogenes may act singly or synergistically. Thus, in combination or alone, partial 1q trisomy and 1p36.1 monosomy may lead to cancer. The finding of constitutional chromosomal abnormalities detected in certain somatic tissues has significant clinical ramifications. Once a clonal chromosomal abnormality is detected in a target tissue, the patient must be carefully monitored for additional signs and symptoms. In our case, we recommend monitoring the kidneys periodically by abdominal imaging. For our proband, prognostication depends in part on the detection of this clonal structural chromosomal abnormality in other somatic tissues as the patient grows and develops. In conclusion, to the best of our knowledge, this is the first reported case of Wilms tumor associated with constitutional partial 1q trisomy, either in pure or mosaic form, with the particular 1q23 breakpoint in conjunction with a break on 1p36.1. Aside from its specific relevance to the pathology of Wilms tumor, the present case may hold new and important clues to the mechanism of carcinogenesis in general, as advances in human genetics continue to provide new insights for understanding various chromosomally induced diseases.
Acknowledgments We thank Dr. Jonathan Fletcher for sharing some of his cytogenetic results and the staff of the Cytogenetics Laboratory at the Center for Human Genetics, Boston University School of Medicine, for lending their cytogenetic expertise. This study would not have been possible without the cooperation of the proband’s family and the family’s physician. We are grateful to all involved. The support of Dr. Roger Mark is also acknowledged. References [1] Mark HFL, Ahearn J, Lathrop JC. Constitutional trisomy 8 mosaicism and gestational trophoblastic disease. Cancer Genetics Cytogenetics 1995;80:150–4. [2] Hecht F. Trisomy and disomy in tumors (EDITORIAL). Cancer Genetics Cytogenetics 1995;80:171. [3] Gafter U, Shabtai F, Kahn Y, Halbrecht I, Djaldetti M. Aplastic anemia followed by leukemia in congenital trisomy 8 mosaicism. Clin Genet 1976;9:134–42. [4] Haas OA, Seyger M. Hypothesis: meiotic origin of trisomic neoplasms. Cancer Genet Cytogenet 1993;70:112–6.
H.F.L. Mark et al. / Cancer Genetics and Cytogenetics 162 (2005) 166–171 [5] Kapaun P, Kabisch H, Held KR, Walter TA, Hegewisch S, Zander AR. Atypical chronic myelogenous leukemia in a patient with trisomy 8 mosaicism syndrome. Ann Hematol 1993;66:57–8. [6] Mark HFL. Phenotypic variability in trisomy 8 mosaicism is consistent with the hypothesis of meiotic origin of trisomic neoplasms. Cancer Genet Cytogenet 1994;76:158. [7] McDonald-McGinn DM, Grace K, Spinner NB, Emanuel BS, Zackai EH. Clinical and cytogenetic variability in trisomy 8 mosaicism. Am J Hum Gen 1993;53(Suppl):222. [8] Riccardi VM, Humbert J, Peakman D. Familial cancer, reciprocal translocation [t(7p;20p)] and trisomy 8. Am J Hum Genet 1975;27: 76A. [9] Rasmussen SA, Frias JL, Lafer CZ, Eunpu DL, Zackai EH. Partial duplication 1q: Report of four patients and review of the literature. Am J Med Genet 1990;36:137–43. [10] Bartsch C, Aslan M, Kohler J, Miny P, Horst J, Holzgreve W, Rehder H, Fritz B. Duplication dup(1)(q32q44) detected by comparative genomic hybridization (CGH): further delineation of trisomies 1q. Fetal Diagn Ther 2001;16:265–73. [11] Kimya Y, Yakut T, Egeli U, Ozerkan K. Prenatal diagnosis of a fetus with pure partial trisomy 1q32-44 due to a familial balanced rearrangement. Prenat Diagn 2002;22:957–61. [12] Machlitt A, Kuepferling P, Bommer C, Koerner H, Chaoui R. Prenatal diagnosis of trisomy 1q21~qter: case report and review of literature. Amer J Med Genet 2005;134A:207–11. [13] Van Buggenhout G, De Coen L, Fryns JP. Partial trisomy 1q (1q32/1qter) in adulthood: further delineation of the phenotype. Ann Genet 1998;41:77–81. [14] Fernandez-Novoa MC, Vargas MT, Grannell MR, Carreto P. Prenatal diagnosis of de novo trisomy 1(q21-qter)der(Y)t(Y;1) in a malformed live born. Prenat Diagn 2004;24:414–7.
171
[15] Vernon EG, Malik K, Reynolds P, Powlesland R, Dallosso AR, Jackson S, Henthorn K, Green ED, Brown KW. The parathyroid hormone-responsive B1 gene is interrupted by a t(1;7)(q42;p15) breakpoint associated with Wilms tumour. Oncogene 2003;22: 1371–80. [16] Reynold PA, Powlesland RM, Keen TJ, Inglehearn CF, Cunningham AF. Localization of a novel t(1;7) translocation associated with Wilms tumor predisposition and skeletal abnormalities. Genes Chromosomes Cancer 1996;17:151–5. [17] Betts DR, Ilg EC, Oezahin H, von der Weid N, Niggli FK. Trisomy 1q generating translocations in Wilms tumor. Cancer Genet Cytogenet 1999;112:138–43. [18] Baumgarten E, Wegner RD, Fengler R, Koch H, Henze G. Partial trisomy 1q, an uncommon chromosomal aberration in erythroleukemia. Leuk Lymphoma 1993;10:237–40. [19] Anderson RL, Bagby GC. The prognostic value of chromosome studies in patients with the preleukemic syndrome (hemopoietic dysplasia). Leukemia Res 1982;6:175–81. [20] Desangles F. Kidney: t(X;1)(p11.2;q21.2) in renal cell carcinoma. Atlas Genet Cytogenet Oncol Haematol. March 1998. Available at: http://www.infobiogen.fr/services/chromcancer/Tumors/tX1ID5009. html. Accessed September 8, 2005. [21] Schorry EK, Dietrich KN, Saal HM, Blough RI, Dey S, Chernausek S, Milatovich-Cherry A. Partial trisomy 1q with growth hormone deficiency and normal intelligence. Am J Med Genet 1998;77:257–60. [22] Heilstedt HA, Ballif BC, Howard LA, Lewis RA, Stal S, Kashork CD, Bacino CA, Shapira SK, Shaffer LG. Physical map of 1p36, placement of breakpoints in monosomy 1p36, and clinical characterization of the syndrome. Am J Hum Genet 2003;72: 1200–12.