Coordinate Deletion of Chromosome 3p and 11q in Neuroblastoma Detected by Comparative Genomic Hybridization

Coordinate Deletion of Chromosome 3p and 11q in Neuroblastoma Detected by Comparative Genomic Hybridization C. J. Breen, A. O’Meara, M. McDermott, M. Mullarkey, and R. L. Stallings

ABSTRACT: Neuroblastoma, the most common extracranial solid tumor of childhood, is associated with a number of genetic abnormalities that are prognostically significant. The most common abnormalities are associated with aggressive clinical behavior and include deletion of distal chromosome 1p, NMYC amplification, and unbalanced gain of the long arm of chromosome 17. There are also other recurrent, but less frequent abnormalities, the clinical significance of which is uncertain. These less common abnormalities include deletion 3p, 11q, and 14q. To gain further clinical insight into some of the less commonly observed abnormalities in neuroblastoma, we performed comparative genomic hybridization (CGH) analysis on 24 primary and metastatic neuroblastomas (6 stage 2, 5 stage 3, 11 stage 4, and 2 stage 4). Nineteen of these tumors were prechemotherapy. A total of 190 abnormalities were detected from these tumors. Four of the 24 tumors studied showed loss of 11q material, with 3 of these tumors also possessing distal chromosome 3p deletions. Our results provide confirmation that deletion of chromosome 3p is nonrandomly associated with deletion of chromosome 11q in neuroblastoma. However, analysis of our results, along with other results reported in the literature, indicate that there is no statistically significant association between 3p and 11q loss and more clinically aggressive tumors. © 2000 Elsevier Science Inc. All rights reserved.

INTRODUCTION Neuroblastoma exhibits extensive heterogeneity with respect to clinical behavior and genetic abnormalities (see Brodeur et al. [1] for review). For example, neuroblastoma occurring in patients under the age of 1 year may undergo spontaneous regression, even in the presence of extensive disease, while survival for patients presenting over the age of 2 years is poor, despite intensive therapy. Tumors occurring in the former group of patients are usually associated with hyperdiploidy, and structural chromosome abnormalities are rare. The latter group of patients, on the other hand, usually present with advanced-stage disease

From the Children’s Medical and Research Foundation (C. J. B., M. M.), the National Centre for Medical Genetics (C. J. B., M. M., R. L. S.), the Department of Oncology (A. O.), and the Department of Pathology (M. M.), Our Lady’s Hospital for Sick Children, Crumlin, Dublin, Ireland; and the Faculty of Medicine (R. L. S.), University College Dublin, Dublin, Ireland. Address reprint requests to: Dr. R. L. Stallings, Our Lady’s Hospital for Sick Children, National Centre for Medical Genetics, Crumlin, Dublin 12, Ireland. Received September 17, 1999; accepted November 29, 1999. Cancer Genet Cytogenet 120:44–49 (2000)  2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010

and are frequently characterized by NMYC amplification, deletion of chromosome 1p, and unbalanced gain of the long arm of chromosome 17. The genetic abnormalities occurring in neuroblastoma have been detected using a number of different techniques, including chromosome banding analysis [2], Southern blot analysis [3], fluorescence in situ hybridization studies [2], and loss of heterozygosity studies involving specific loci [4–9], and more recently, comparative genomic hybridization (CGH) [10–14]. These techniques, particularly LOH [4–9] and CGH [10–14], have revealed that deletion of 3p, 11q, and 14q are also recurrent, but less frequently observed, abnormalities found in neuroblastoma. The purpose of our study was to provide additional insight into the clinical significance of these less frequently occurring abnormalities and to correlate data with histopathological grading.

MATERIALS AND METHODS Neuroblastoma Tissue Comparative genomic hybridization analysis was carried out using DNA isolated from biopsy material (14 primary

0165-4608/00/$–see front matter PII S0165-4608(99)00252-6

45

CGH Analysis of Neuroblastoma Table 1 Analysis of neuroblastomas INSS Joshi Case stage grade

Site of disease

CGH result

Pretreatment tumors: 1 2 I Cervical 2 2b II Thoracic 3 2a I Thoracoabdominal 4 2a I Thoracic 5 2a II Thoracic 6 2b II Abdominal 7 8 9 10

3 3 4 4

II II III III

11

4

II

12 13 14 15 16

4s 4 4 4 4

II

17 18 19

4 2 4s

III III II I II II

enh 2, 7, 17, 18 Normal enh 1, 2, 6, 7, 10, 12, 17, 18; dim 19 enh 6, 7, 13, 17, 18; dim 19 enh 1, 2, 5, 7, 9, 10, 12, 15, 17, 20, 22 enh 2, 2p24, 3q22, 6p, 7q21, 12q13–q15, 17q; dim 2q33–q35, 10 Abdominal enh 1, 2, 7, 8 Abdominal Normal Thoracoabdominal enh 2p24, 11q12–q14, 17q, 18q; dim 1p32 Thoracoabdominal enh 1q, 2, 2p24, 3p14–qter, 4q, 6, 7, 7p, 8, 9q, 10q, 11p, 12, 12q23, 13, 14, 16q, 17q, 18q, 20, 21, 22, Xp; dim 3p21–pter, 4p, 11q22–qter Abdominal enh 1q, 2p24, 4q31–q31, 5p15–p15, 5q31–q31, 7q, 12, 16, 17q, 18q, 20q, 22, Xp22 Abdominal enh 1, 2, 9, 13, 15, 17 Abdominal enh 6p, 9q11, 15, Xp22 Neck, Thoracoabdominal enh 2p13–pter, 2p24, 17q; dim 1p32 Abdominal enh 2p24, 17q; dim 17p Abdominal enh 1p31–qter, 2, 2p24, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17p, 17q, 18q, 19, 20, 21, 22 Thoracoabdominal enh 2p21–qter; dim 1p31–pter, 3p21–pter, 11q22–qter, 19 Cervicothoracic enh 5, 6, 12, 17, 18 Abdominal enh 1p32–qter, 7, 12

Post-treatment tumors: 20 3 Abdominal 21 22 23 24

4 4 4 3

Abdominal Abdominal Abdominal Pelvic

enh1q, 2, 2p24, 5, 12q21, 13, 14, 17, 17q, 18, 20, 21, 22; dim 1p32, 10q24–qter enh 1, 5, 9, 14, 15, 17, 20, 22 enh 7, 8p22, 17q, 18; dim 3p24, 11q22 enh 1, 2, 9q, 10, 12q, 13, 14, 16, 17q, 18q, 20, 21, 22, Xp enh 11q13–q14, 17q, 18p11.3; dim 4p13, 9q34, 11q22

FR 2p24 Agea

Current status

0.54 0.31 0.75 2.41 0.04 1.25 4.51

NED 11 yrs 7 mths NED 2 yrs 3 mths NED 1 yr 8 mths NED 9 yrs 5 mths NED 11 yrs DOD

2.0 1.2

0.12 0.50 1.15 2.56

NED 4 yrs 8 mths NED 10 yrs DOD NED 2 yrs 1 mth

1.25 4.95 DOD 0.15 2.70 2.0 2.05 1.75 1.98 3.0 0.94

NED 2 yrs DOD DOD REL 20 mths DOD

3.30 NED 9 mths 0.03 NED 7mths 0.10 PR 6 mths 3.25 1.92 NED 5 yrs 11 mths 1.55 3.0 7.41 3.94

DOD DOD DOD NED 1 yr 3 mths

Abbreviations: NED, no evidence of disease; DOD, died of disease; PR, partial remission; REL, relapse; yrs, years; mths, months; enh, enhanced fluorescence ratio; dim, diminished fluorescence ratio; FR, fluorescence ratio. Neuroblastomas obtained from primary sites are designated as primary. a

Age at diagnosis.

tumors) obtained from 19 patients with untreated neuroblastoma attending the Oncology Department of Our Lady’s Hospital for Sick Children, Dublin, who were staged according to INSS classification [15] (Table 1). Histology from the same specimen was reviewed and tumors graded according to the Joshi system (Table 1) [16, 17]. All pretreatment tumors except case 4 were stroma-poor and undifferentiated according to the Shimada classification [18]. The tumor from case 4 was stroma-poor and differentiated. Comparative genomic hybridization analysis was also undertaken from tissue obtained from a further 5 patients who still had viable disease at the time of elective resection; histopathological grading of such specimens was not appropriate. Comparative Genomic Hybridization Comparative genomic hybridization experiments were carried out as previously described [19].

Statistical Analysis The statistical significance of various findings was assessed by standard Chi square analysis. RESULTS A total of 24 neuroblastoma were analyzed by CGH, with the results summarized in Table 1 and Figures 1 and 2. Nineteen of these neuroblastoma were pretreatment biopsies, while 5 tumors were obtained from patients who still had viable disease at elective resection, following intensive chemotherapy. Tumors from two patients (cases 2 and 8) had normal CGH profiles, and both are alive with no evidence of disease. Abnormalities were detected in 6 of 7 patients with stage 2 disease and in both infants with stage 4s neuroblastoma (Fig. 1). Five of the stage 2 neuroblastomas (cases 1, 3, 4, 15, and 18) showed only entire chromosome gains and losses, without any detectable partial gain

46

C. J. Breen et al.

Figure 1 Summary of abnormalities occuring in eight stage 1, 2, and 4s neuroblastomas.

or loss of specific chromosome regions. One patient (case 6) initially presented with opisthotonus and cerebellar ataxia associated with abdominal paraspinal stage 2b disease and ultimately died of disease. This tumor had amplification of the NMYC region at 2p24, unbalanced gain of 17q, and amplification of the 12q13–q15 region, as well as partial gains and losses of other chromosomal regions. As illustrated in Figure 1, overall chromosome gains were nonrandom, with gain of entire chromosomes 1, 2, 6, 7, 12, 17, and 18 being the most frequent. Data from 15 patients with stage 3 and 4 neuroblastomas are summarized in Figure 2. Seven tumors had an increase in fluorescence ratio (FR) for the NMYC region at 2p24, while three of these tumors (10, 16, and 20) showed an increase in FR for the entire chromosome 2, with a further increase in FR at 2p24. Considerable hyperdiploidy was present in these tumors. Three of the tumors (cases 9, 14, and 20) that showed NMYC amplification also had deletion of the region 1p32 to pter. Unbalanced gain of chromosome 17q was detected in 10 of the 15 stage 3 and 4 tumors, making this the most frequent abnormality found in the higher grade tumors of this study. Two (cases 16 and 20) of these 11 tumors also showed a significant increase in FR on the p arm of chromosome 17, with an even greater increase on the long arm. Four (cases 10, 17, 22, and 24) of the stage 3 and 4 tumors had deletions of the 11q22 to qter region. Three of these tumors (cases 10, 17, and 22) also showed deletion of distal 3p. Two tumors (cases 9 and 24) showed amplification of the 11q12–q14 region. In addition to unbalanced

gain of chromosome 17q, there was unbalanced gain of 1q and 9q in three cases, and unbalanced gain of 18q in five cases. No loss of chromosome 14q material was detected. The stage 3 and 4 tumors also had a considerable number of whole chromosome gains. The whole chromosomes that were gained in stage 3 and 4 tumors were similar to the gains detected in stage 2 neuroblastomas. As the numbers of patients in individual histological categories was small, comparison of histological data with CGH data from the same patient did not identify statistically significant correlations. However, it is noteworthy that no patients with NMYC-amplified tumors were histological grade 1, or exhibited prominent histological features of ganglion cell differentiation. Similarly, only one of nine tumors with unbalanced gain of 17q displayed prominent gangliocytic features, although two of the nine were grade 1 tumors. DISCUSSION The most frequently occurring chromosome abnormalities detected in this study and in five related published studies [10–14] are summarized in Table 2. Unbalanced gain of 17q (43%), NMYC amplification (32.4%), and distal 1p deletion (31.7%) are the most frequently occurring abnormalities detectable by CGH and are associated with higher stage tumors and unfavorable prognosis. The next most common structural chromosome abnormalities involve deletions of distal 14q (16.9% of tumors), 11q (15.5%), and 3p (13.5%), and gain of chromosome 1q (14.9%). The

47

CGH Analysis of Neuroblastoma

Figure 2 Summary of abnormalities occuring in 14 stage 3 and 4 neuroblastomas.

gain of 1q could be related to an unbalanced t(1;17) commonly detected in neuroblastoma [20]. All of the tumors showing gain of 1q in this study also showed unbalanced gain of 17q. It is the unbalanced gain of 17q that is associated with a poor prognosis [21]. Some studies have indicated an association between 11q loss and prognostically unfavorable tumors [7, 11, 12], while others [14] have documented 60% probability of greater than 3-year survival in patients with this anomaly.

Of a total of 15 tumors reported with 11q deletions [12, 14, and this study], 11 had stage 4 disease. This is not a statistically significant correlation. In this study, three of four tumors noted to have distal 11q deletions also had deletions involving 3p. Nine patients had deletions of 11q together with 3p, from a total of 15 reported deletions of chromosome 11q. This is a highly significant correlation (P ⬍ 0.001) because the overall frequency of 3p deletions in these studies is only

Table 2 Overview of abnormalities detected in neuroblastomas by CGH

Reference Lastowska [13] Brinkschmidt [11] Plantaz [12] Vandescompele [14] Altura [10] Present study Total a

No. stage 3 and 4 tumors 12 22 12 23 3 15 87

Total no. tumors 20 35 29 36 6 24 150

No. of tumors detected with each type of abnormality 1p⫺ 8 13 10 9 3 4 47

1q⫹ 1 7 1 8 2 3 22

NMYC amplification 6 18 5 9 2 8 48

3p⫺

11q⫺

4 3 1 5 3 3 19

Information about coordinated deletion of 3p and 11q could not be ascertained from these publications.

2 6 3 8 0 4 23

3p⫺ and 11q⫺ a a

1 5 0 3 9

⫺14 or 14q⫺

17q⫹

2 9 11 11 0 0 33

6 20 11 11 4 10 62

48 13.5%. An interchromosomal rearrangement between 11q and 3p seems unlikely to account for this correlation because no such rearrangement has ever been reported. It seems more plausible that independent deletions on both 3p and 11q confer a growth advantage to the neuroblastoma in some synergistic manner. Eight of nine patients whose tumors showed loss of 3p and 11q had stage 4 disease. Vandescompele et al. [14] documented loss of 3p and 11q in one patient with stage 1 disease. This tumor was atypical of this category in that it showed considerable partial gains and losses of chromosomal regions, as opposed to whole chromosome gains. Further data are required to substantiate whether co-deletion of 3p and 11q is correlated with higher stage disease. Six of 15 tumors reported elsewhere [12, 14] had an 11q deletion together with deletion of 14q. Loss of heterozygosity studies have also detected loss of 11q and 14q regions [6, 7]. This correlation is not as strong as that which was observed between the 11q and 3p; however, it is still significant at a 95–97.5% confidence level. Chromosome 14q deletions were not observed in our study. Based on the overall frequency of 14q deletions (Table 2), one might have expected the detection of 3 or 4 tumors with these deletions. Perhaps differences in CGH techniques, software programs, or criteria used for defining abnormalities account for this discrepancy. Vandescompele et al. [14] also noted discrepancies in the frequencies of observed abnormalities in different studies. Others have noted that 3p and 11q deletions, and 3p and 14q deletions, tend to occur in a high proportion of tumors that do not posses 1p deletions or NMYC amplification, and may form a genetically distinct subgroup of tumors [14]. One of nine tumors had NMYC amplification [14], and one of nine had loss of chromosome 1p [this report]. At present, there is insufficient evidence to conclude that co-deletion 3p⫺ 11q⫺ is an alternative genetic pathway for neuroblastoma progression distinct from deletion 1p and NMYC amplification. It would be of considerable interest to isolate the specific genes located on chromosomes 3p, 11q, and 14q that are involved with neoplasia. Hyperdiploidy was noted in a number of patients in this study and other CGH studies [10–14]. It is unknown whether early-stage neuroblastoma displaying hyperdiploidy ever evolves into more aggressive disease [1]. Many lower stage neuroblastomas occurring in infancy undergo spontaneous regression; conversely, some higher stage neuroblastomas with NMYC amplification also display hyperdiploidy (e.g., see cases 10, 16, and 20). The specific chromosomes involved with whole chromosome gains are usually similar in lower and higher stage tumors. In conclusion, this study provides further data that deletion of chromosome 3p is nonrandomly associated with 11q deletion in neuroblastoma. The prognostic significance of this observation remains to be elucidated.

This work was supported in part by grants from the Irish Cancer Society and the Children’s Medical and Research Foundation, Our Lady’s Hospital for Sick Children (Dublin, Ireland).

C. J. Breen et al. REFERENCES 1. Brodeur GM, Maris JM, Yamashiro DJ, Hogarty MD, White PS (1997): Biology and genetics of human neuroblastomas. J Pediatr Hematol Oncol 19:93–101. 2. Lastowska M, Roberts P, Pearson AD, Lewis I, Wolstenholme J, Bown N (1997): Promiscuous translocations of chromosome arm 17q in human neuroblastomas. Genes Chromosom Cancer 19:143–149. 3. Emanuel BS, Balaban G, Boyd JP, Grossman A, Negishi M, Pariter A, Glick MC (1985): N-myc amplification in multiple homogeneously staining regions in two human neuroblastomas. Proc Natl Acad Sci USA 82:3736–3740. 4. Ejeskar K, Aburatani H, Abrahamsson J, Kogner P, Martinsson T (1998): Loss of heterozygosity of 3p markers in neuroblastoma tumors implicate a tumor-suppressor locus distal to the FHIT gene. Br J Cancer 77:1787–1791. 5. Fong CT, Dracopoli NC, White PS, Merrill PT, Griffith RC, Housman DE, Brodeur GM (1989): Loss of heterozygosity for the short arm of chromosome 1 in human neuroblastomas: correlation with N-myc amplification. Proc Natl Acad Sci USA. 86:3753–3757. 6. Hallstensson K, Thulin S, Aburatani H, Hippo Y, Martinsson T (1997): Representational difference analysis and loss of heterozygosity studies detect 3p deletions in neuroblastoma. Eur J Cancer 33:1966–1970. 7. Srivatsan ES, Ying KL, Seeger RC (1993): Deletion of chromosome 11 and 14q sequences in neuroblastoma. Genes Chromosom Cancer 7:32–37. 8. Srivatsan ES, Murali V, Seeger RC (1991): Loss of heterozygosity for alleles on chromosomes 11q and 14q in neuroblastoma. Prog Clin Biol Res 366:91–98. 9. Suzuki T, Yokata J, Mugishima H, Okabe I, Ookuni M, Sugimura T, Terada M. (1989): Frequent loss of heterozygosity on chromosome 14q in neuroblastoma. Cancer Res 49:1095– 1098. 10. Altura RA, Maris JM, Li H, Boyett JM, Brodeur GM, Look AT (1997): Novel regions of chromosomal loss in familial neuroblastoma by comparative genomic hybridization. Genes Chromosom Cancer 19:176–184. 11. Brinkschmidt C, Christiansen H, Terpe HJ, Simon R, Boecker W, Lampert F, Stoerkel S (1997): Comparative genomic hybridization (CGH) analysis of neuroblastomas—an important methodological approach in paediatric tumor pathology. J Pathol 181:394–400. 12. Plantaz D, Mohapatra G, Matthay KK, Pellarin M, Seeger RC, Feuerstein BG (1997): Gain of chromosome 17 is the most frequent abnormality detected in neuroblastoma by comparative genomic hybridization. Am J Pathol 150:81–89. 13. Lastowska M, Nacheva E, McGuckin A, Curtis A, Grace C, Pearson A, Bown N (1997): Comparative genomic hybridization study of primary neuroblastoma tumors. Genes Chromosom Cancer 18:162–169. 14. Vandescompele J, Van Roy N, Van Gele M, Laureys G, Ambros P, Heimann P, Devalck C, Schuuring E, Brock P, Otten J, Gyselinck J, De Paepe A, Speleman F (1998): Genetic heterogeneity of neuroblastoma studied by comparative genomic hybridization. Genes Chromosom Cancer 23:141– 152. 15. Brodeur GM, Pritchard J, Berthold F, Carlsen NLT, Caster V, Castleberry RP, DeBernardi B, Evans AE, Faurot M, Hedborg F, Kaneko M, Kemshead J, Lampert F, Lee R, Look T, Pearson AD, Philip T, Roald B, Sawada T, Seeger RC, Tsuchida Y, Voute PA (1993): Revisions of the International Criteria for Neuroblastoma Diagnosis, Staging and Response to Treatment. J Clin Oncol 11:1466–1477. 16. Joshi VV, Cantor AB, Altshuler G, Larkin EW, Neill JSA, Shuster JJ, Holbrook CT, Hayes FA, Castleberry RP (1992):

CGH Analysis of Neuroblastoma Recommendations for modification of terminology of neuroblastic tumors and prognostic significance of Shimada classification. Cancer 69:2183–2196. 17. Joshi VV, Cantor AB, Altshuler G, Larkin EW, Neill JSA, Shuster JJ, Holbrook CT, Hayes FA, Nitschke R, Duncan MH, Shocat SJ, Talbert J, Smith EI, Castleberry RP (1992): Agelinked prognostic categorization based on a new histologic grading system of neuroblastomas. Cancer 69:2197–2211. 18. Shimada H, Chatten J, Newton WA, Sachs N, Hamoudi AB, Chiba T, Marsden HB, Misugi K (1984): Histopathologic prognostic factors in neuroblastic tumors: definition of subtypes of ganglioneuroblastoma and an age-linked classification of neuroblastomas. J Natl Cancer Inst 73:405–413.

49 19. Breen CJ, Barton L, Carey A, Dunlop A, Glancy M, Hall K, Hegarty AM, Khokhar MT, Power M, Ryan K, Green AJ, Stallings RL (1999): Applications of comparative genomic hybridization in constitutional chromosome studies. J Med Genet 36:511–517. 20. Caron H, van Sluis P, van Roy N, Beks L, Maes P, Pereira do Tanque R, Slater R, de Kraker J, Speleman F, Voute PA (1994): Chromosome 1p allelic loss in neuroblastoma: prognosis, genomic imprinting and 1;17 translocations. Prog Clin Biol Res 385:35–42. 21. Lastowska M, Cotterill S, Pearson ADJ, Roberts P, McGuckin A, Lewis I, Bown N (1997): Gain of chromosome arm 17q predicts unfavourable outcome in neuroblastoma patients. Eur J Cancer 33:1627–1633.