- Email: [email protected]
Aberrant copy numbers of ALK gene is a frequent genetic alteration in neuroblastomas
Human Pathology (2009) 40, 1638–1642
www.elsevier.com/locate/humpath
Original contribution
Aberrant copy numbers of ALK gene is a frequent genetic alteration in neuroblastomas☆ Manish Mani Subramaniam MD, Marta Piqueras BS, Samuel Navarro MD, Rosa Noguera MD ⁎ Department of Pathology, Medical School, University of Valencia, 46010 Valencia, Spain Received 17 January 2009; revised 22 April 2009; accepted 4 May 2009
Keywords: Neuroblastoma; ALK; MYCN; Copy number alterations
Summary A total of 50 neuroblastomas were assessed for frequency of ALK gene copy number aberrations by interphase fluorescence in situ hybridization using a break-apart fluorescence in situ hybridization probe. The data were compared with status of MYCN, 11q, 17q, and 1p36. We observed ALK aberrations (amplification, 1 of 45; gain, 15 of 45 and loss/imbalance, 11 of 45) in a total of 27 (60%) of 45 neuroblastomas. Synchronic MYCN and ALK aberrations accounted for 23 of 45 (51%) tumors; however, MYCN alterations were also detected in 11 (60%) of 18 tumors without ALK aberrations. Our data suggest that copy number aberrations of the ALK gene is a frequent genetic event in the development of neuroblastomas. In addition, no correlation was observed between ALK aberrations and alterations of 11q, 17q, and 1p36. © 2009 Elsevier Inc. All rights reserved.
1. Introduction In advanced stages, neuroblastomas are one of the most intractable pediatric cancers, even with recent therapeutic advances. Over the years, a variety of genetic changes such as MYCN amplification (in approximately 25% of tumors) [1], loss of heterozygosity at 1p36 [2] and 11q [3], and 17q gain [4] have been implicated in the pathogenesis of neuroblastoma. In addition, expression of certain receptor tyrosine kinases such as TrkA [5] and TrkB [6] has been linked to respectively favorable and unfavorable outcomes in neuroblastomas. Recently, whole genomic DNA scan– ☆
Grant Support: From the ISCIII (RD06/0020/0102; ATC606/2007; FISPI06/1576), Fundación Inocente, Inocente (PI4/07-36) and CS (AP053/08). ⁎ Corresponding author. E-mail address: [email protected] (R. Noguera). 0046-8177/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2009.05.002
based methods and direct sequencing revealed oncogenic [7] germline and somatic [8] inactivating ALK mutations, indicating ALK to be a major predisposition gene [9] as well as a potential therapeutic target for neuroblastoma [10]. To date, genomic status of ALK has been assessed mainly using comparative genomic hybridization [11,12], comparative expressed sequence hybridization [11] and southern blot analysis [13], predominantly in tumor cell lines and some human neuroblastoma samples. To date, only a few studies have compared the genomic status of ALK with that of MYCN and other established genetic markers in archival samples of neuroblastomas [9,13]. In this study, we evaluated the copy number aberrations of ALK gene in formalin-fixed, paraffin-embedded clinical samples of neuroblastomas, as well as analyzed a possible relation between ALK aberrations and alterations of established prognostic genetic markers such as MYCN, 1p, 11q, and 17q.
ALK aberrations in neuroblastoma
2. Materials and methods 2.1. Tissue samples and fluorescence in situ hybridization probes A computer-based search of the pathology database yielded a total of 50 primary neuroblastoma samples classified as poorly differentiated (n = 28), NOS (not otherwise specified; n = 10), and undifferentiated (n = 12). Of these 50 tumors, 33 (66%) belonged to the high-risk group, and 17 (34%) to the low-risk group. MYCN gene status obtained from routine analysis of tumor touch preparations was available for all these samples. Of the 33 high-risk tumors, 27 showed MYCN amplification, 5 cases had MYCN gain, and there was 1 case with no MYCN alterations. Of the 17 low-risk tumors, 15 had no MYCN alterations, 2 were reported to be MYCN gain, and none of them harbored MYCN amplification. Representative tumor tissue cores from all these cases were subsequently assembled in a tissue microarray (TMA) with duplicate cores for each case. Sections, 4 μm thick, were cut from the TMA block and subjected to the formalin-fixed, paraffin-embedded fluorescence in situ hybridization (FISH) procedure [14] using a commercial ALK split probe (Dako, Inc, Glostrup, Denmark) composed of a Texas Red-labeled DNA probe (ALK-downstream) that binds to a 289-kb segment centromeric and fluorescein-labeled DNA probe (ALK-upstream) that binds to a 557-kb segment telomeric to the ALK breakpoint cluster region respectively on chromosome 2p23. In addition, we also confirmed MYCN gene status on the TMA tumor samples using MYCN (2p24) (labeled red) and LAF (2q11) control probe (labeled green) (Kreatech Biotechnology, Amsterdam, The Netherlands). Probe combinations for analysis of 1p36, 11q and 17q status included the following: chromosome 1p36 Midisatellite probe labeled green (MP Biomedicals, Illkirch, Cedex, France)/chromosome 1 satellite probe labeled red (Q-BIOgene, France) and ATM (11q22)specific DNA probe labeled in red/chromosome 11 satellite enumeration probe labeled with green (Kreatech Biotechnology, Amsterdam, The Netherlands) and p53 (17p13) green/ MPO (17q23) iso 17q red (Kreatech Biotechnology).
2.2. FISH scoring scheme The FISH signals were scored in 200 nonoverlapping nuclei per core, independently by two investigators (M.M. S. and R.N.), and the consensus was recorded. Four cellular groups were defined as follows [15]: group 1, cells with absent genetic alterations showing the same numbers of gene-specific and control probe signals; group 2, cells harboring genetic alterations demonstrating different copy numbers of gene-specific and control probe signals (Table 1); group 3, cells displaying nuclear truncation artifacts, for example, less MYCN/ALK signals compared to their respective control probe signals, more 1p36/11q
1639 Table 1 FISH criteria for copy number alterations of the different genetic markers Marker
Status
Description
MYCN/ALK
No Alteration
Same numbers of genespecific signals and control probe signals The number of genespecific signals is 1 to 4 times greater than the control probe signals Presence of at least 2 gene-specific signals and increased control probe signals The number of genespecific signals is more than 4 times the control probe signals Same numbers of 1 p36/11q specific signals and control probe signals At least 2 signals of 1p36/ 11q specific locus and more control probe signals One signal representing 1p36/11q region with at least 2 control probe signals Similar numbers of 17q regions signals and control probe signals The numbers of 17q signals are 1 to 4 times greater than control probe signals
Gain
Loss/Imbalance
Amplification
1p36/11q
No deletion
Loss/Imbalance
Deletion
17q
No Gain
Gain
The FISH criteria for MYCN , 11q, 1p36, and 17q alterations are according to the guidelines specified by the European Neuroblastoma Quality Assessment Group.
signals compared to their control probe signals, or less 17q23 signals compared to the control probe signals; group 4, net percentage of cells with definite genetic alteration, obtained by subtracting the percentage of cells showing truncation artifacts (group 3) from those of group 2, thereby eliminating the false positives. The FISH scoring scheme was initially evaluated on a series of 16 control tissues such as normal kidneys (n = 5) and nonneuroblastic tumors (breast cancers, n = 6 and gastric cancers, n = 5) and then on neuroblastomas in order to obtain a cutoff value for each FISH probes. The mean percentage of cells in group 4 in the control tissues were as follows: ALK (12.5%); MYCN (13%); deletion of 11q (11%) and 1p36 (11.5%); and gain of 17q (10%), thereby making an overall mean of 11.6 (SD 1.19) for all the genetic markers. The mean+3SD of the percentage of cells in group 4 for the control tissues is 15%. Thus, a tumor sample was considered positive for a genetic alteration if the percentage of cells in group 4 exceeded the mean+3SD
1640
M. M. Subramaniam et al.
(15%). The FISH criteria for the different aberrations are summarized in Table 1.
3. Results Of the 50 neuroblastomas, 45 (90%) were informative for all the FISH assays. The noninformative samples were constituted by the 5 low-risk neuroblastomas without MYCN amplification. The status of ALK, MYCN, 11q, 1p36, and 17q in neuroblastomas is summarized in Table 2. Considering only the 45 informative cases, aberrant copy numbers of ALK gene were observed in 27 (60%) samples. ALK amplification was observed in 1 (2%) of 45, ALK gain in 15 of 45 (33%), ALK loss/imbalance in 11 (24%), and no alterations of ALK in the remaining 18 (40%) cases. The solitary case of ALK amplification revealed between 25 and 30 ALK signals (Fig. 1), whereas the ALK gain cases displayed between 3 and 6 signals. MYCN alterations were observed in 34 (76%) of 45 cases. Of the 45 cases, MYCN amplification was found in 27 (60%) cases, MYCN gain in 7 (15%), and no alterations in 11 (24%) cases. Coexistent MYCN and ALK alterations were observed in 23 (51%) of 45 cases. Out of the 27 cases with ALK alterations, MYCN was altered in 23 (85%). One of the important findings is that of the 15 cases with ALK gain, MYCN amplification was found in 10 (67%) and MYCN gain in 4 (26%) cases. Nevertheless, 11 (61%) of the 18 cases without ALK alterations also showed MYCN alterations. Gain of 17q was detected in 21 of the 27 (77%) ALK-altered cases as well as in 14 (77%) of 18 cases without ALK alterations, signifying no association.
Fig. 1 Representative FISH image of the neuroblastoma cells displaying amplification of ALK gene evident by the increased numbers (25-30 approximately) of ALK upstream probe (telomeric to the ALK break point cluster region on 2p23) signals (green) compared to two signals of ALK-downstream probe (binding centromeric to ALK break point cluster region) (red). Arrows indicate two representative non-overlapping nuclei in this field exhibiting ALK amplification (DAPI counterstain, original magnification ×400).
Deletions of 11q and 1p36 were detected in 14 (31%) of 45 and 28 (62%) of 45 cases respectively. However, no correlation was observed between ALK alterations and deletions of 11q and 1p36.
4. Discussion Table 2 Frequency of copy number aberrations of ALK and its correlation with status of MYCN, 11p, 17q, and 1p36 ALK loss/ No ALK Imbalance alterations (n = 18) ALK ALK gain (n = 11) amplification (n = 15) (n = 1)
ALK status (total cases n = 45)
MYCN A G NA 17q G N 11q D ND 1p36 D ND
1 0 0 0 1 0 1 1 0
10 4 1 13 2 4 11 11 4
8 0 3 8 3 5 6 6 5
8 3 7 14 4 5 13 10 8
Abbreviations: A, Amplification; G, gain; NA, no alteration; D, deletion; ND, no deletion.
The complexity of the organization of genes in the 2p amplicon indicates that not only the MYCN locus but flanking and/or unrelated regions might be involved in the aggressive biological behavior of neuroblastomas. Besides MYCN, other genes such as DDXI [16] belonging to the family of genes encoding DEAD box proteins, ID2 (Inhibitor of DNA-binding/differentiation) [17], NAG (neuroblastoma amplified gene) [18], SNTG2, and TPO [19] have also been reported to be co-amplified in MYCN amplified neuroblastomas. A recent addition to this list is the ALK gene situated at 2p23.2. This is the first TMAbased interphase-FISH study to assess the frequency of copy number aberrations of the ALK gene and its possible association with MYCN, 17q, 11q, and 1p36 status in archival samples of neuroblastomas. Aberrant ALK copy numbers were detected in 27 (60%) of 45 neuroblastomas, with amplification constituting 1 (2%), gain 15 (33%), and loss/imbalance 11 (24%) of these
ALK aberrations in neuroblastoma cases. A straightforward comparison of our data with previously published results is difficult owing to the different methodologies employed in assessing the ALK genomic status, as well as to the data from cell lines versus that from clinical samples. Overall, our results were slightly higher when compared to previous data. However, frequency of ALK amplification detected in our study was comparable to that of the following previous reports: Osajima et al (1 [1%] of 85, in tumor tissues and 3 (12%) of 25, in cell lines by southern blot analysis) [13], Caren et al (5% in tumor tissues) [12], Mosse et al, (16 [3.3%] of 491 primary neuroblastomas by SNP-based microarrays) [9], and Stock et al (ALK amplification in one tumor and 2 (15%) of 13 cell line samples) [11]. In our study, the primary tumor with ALK amplification exhibited over 30 ALK signals in contrast to a previous report [13] in which cell lines (N30 copies) revealed increased copy numbers of ALK, as opposed to tumor tissues (2 to 10 copies). Increased frequencies of ALK amplification in tumor cell lines, a common feature reported in the above studies can be attributed to the mechanism by which cells with higher ALK copy numbers become the major population during establishment of cell lines because of their growth advantage. The ALK gain noted in our study was consistent with the partial trisomy documented by SNP-based microarrays; however, our frequencies were greater than the earlier reported frequencies: 8.2% [13], 112 (22.8% ) of 491 [9], and 2 (15%) of 13 (in cell lines) [11]. The mean number of ALK signals in cases with gain (3-6 signals) was higher than that reported earlier by Osajima et al (1.8-3.0) [13], a fact that can also be attributed to the different methods of copy number analysis. In addition, we also observed loss/imbalance of ALK gene characterized by presence of at least 2 ALK signals but an increased number of control probe signals. ALK loss/imbalance has not been reported in previous studies. Alternatively, loss/imbalance of ALK may also be explained by a possible chromosomal duplication of the 2p23.2 region with allelic loss. Taking our findings and the previous data together, aberrant copy numbers of ALK seems to be a frequent genetic event in the tumorigenesis of neuroblastomas. Concordant ALK and MYCN copy number aberrations accounted for 23 (51%) of 45 neuroblastoma cases. Furthermore, 23 (85%) of 27 cases with ALK aberrant copy numbers also revealed MYCN copy number aberrations. However, MYCN alterations were also seen in 60% of cases devoid of ALK alterations. It is important to note that the higher frequency of MYCN amplification obtained in our study was due to the inclusion of a relatively high proportion of high-risk neuroblastomas in our TMA-based analysis. Our results are consistent with previously published data wherein synchronic ALK and MYCN alterations has been reported [11,9,13]. Based on our data and the previous findings, we observe a tendency of synchronic copy number aberrations of ALK and MYCN in a majority of cases. The lack of ALK alterations in over 60% of cases with MYCN alterations observed in our study, as well as the less frequent coexistence
1641 of ALK and MYCN alterations in some of the aforementioned studies [11,13] suggest that there is no strong evidence to support the coexistence of ALK and MYCN aberrations. Moreover, the ALK gene locus (2p23.2) appears too far from the MYCN gene locus to be within a single amplicon. Therefore, we assume that copy number aberrations of ALK and MYCN might have an independent function in the pathogenesis of neuroblastomas. The frequencies of 17q, 11q, and 1p36 alterations derived from our study were consistent with published data [4,20,21]; however, no correlation was observed between them and ALK aberrations. In summary, our simple and easily reproducible FISH technique shows that copy number aberrations of ALK gene (amplification, gain and loss) is a frequent genetic event in the development of neuroblastomas.
References [1] Seeger RC, Brodeur GM, Sather H, et al. Association of multiple copies of the N-myc oncogene with rapid progression of neuroblastomas. N Engl J Med 1985;313:1111-6. [2] Brodeur GM, Sekhon G, Goldstein MN. Chromosomal aberrations in human neuroblastomas. Cancer 1977;40:2256-63. [3] McArdle L, McDermott M, Purcell R, et al. Oligonucleotide microarray analysis of gene expression in neuroblastoma displaying loss of chromosome 11q. Carcinogenesis 2004;25:1599-609. [4] Bown N, Cotterill S, Lastowska M, et al. Gain of chromosome arm 17q and adverse outcome in patients with neuroblastoma. N Engl J Med 1999;340:1954-61. [5] Nakagawara A, Arima-Nakagawara M, Scavarda NJ, et al. Association between high levels of expression of the TRK gene and favorable outcome in human neuroblastoma. N Engl J Med 1993;328:847-54. [6] Nakagawara A, Azar CG, Scavarda NJ, et al. Expression and function of TRK-B and BDNF in human neuroblastomas. Mol Cell Biol 1994; 14:759-67. [7] Chen Y, Takita J, Choi YL, et al. Oncogenic mutations of ALK kinase in neuroblastoma. Nature 2008;455:971-4. [8] Janoueix-Lerosey I, Lequin D, Brugières L, et al. Somatic and germline activating mutations of the ALK kinase receptor in neuroblastoma. Nature 2008;455:967-70. [9] Mossé YP, Laudenslager M, Longo L, et al. Identification of ALK as a major familial neuroblastoma predisposition gene. Nature 2008;455: 930-5. [10] George RE, Sanda T, Hanna M, et al. Activating mutations in ALK provide a therapeutic target in neuroblastoma. Nature 2008;455:975-8. [11] Stock C, Bozsaky E, Watzinger F, et al. Genes proximal and distal to MYCN are highly expressed in human neuroblastoma as visualized by comparative expressed sequence hybridization. Am J Pathol 2008;172: 203-14. [12] Carén H, Abel F, Kogner P, et al. High incidence of DNA mutations and gene amplifications of the ALK gene in advanced sporadic neuroblastoma tumours. Biochem J 2008;416:153-9. [13] Osajima-Hakomori Y, Miyake I, Ohira M, et al. Biological role of anaplastic lymphoma kinase in neuroblastoma. Am J Pathol 2005;167: 213-22. [14] Subramaniam MM, Noguera R, Piqueras M, et al. p16INK4A (CDKN2A) gene deletion is a frequent genetic event in synovial sarcomas. Am J Clin Pathol 2006;126:866-74. [15] Piqueras M, Navarro S, Castel V, et al. Analysis of biological prognostic factors using tissue microarrays in neuroblastic tumors. Pediatr Blood Cancer 2008;52(2):209-14.
1642 [16] Squire JA, Thorner PS, Weitzman S, et al. Co-amplification of MYCN and a DEAD box gene (DDX1) in primary neuroblastoma. Oncogene 1995;10:1417-22. [17] Lasorella A, Boldrini R, Dominici C, et al. Id2 is critical for cellular proliferation and is the oncogenic effector of N-myc in human neuroblastoma. Cancer Res 2002;62:301-6. [18] Kaneko S, Ohira M, Nakamura Y, et al. Relationship of DDX1 and NAG gene amplification/overexpression to the prognosis of patients with MYCN-amplified neuroblastoma. J Cancer Res Clin Oncol 2007; 133:185-92.
M. M. Subramaniam et al. [19] Łastowska M, Viprey V, Santibanez-Koref M, et al. Identification of candidate genes involved in neuroblastoma progression by combining genomic and expression microarrays with survival data. Oncogene 2007;26:7432-44. [20] Spitz R, Hero B, Simon T, et al. Loss in chromosome 11q identifies tumors with increased risk for metastatic relapses in localized and 4S neuroblastoma. Clin Cancer Res 2006;12:3368-73. [21] Spieker N, Beitsma M, van Sluis P, et al. An integrated 5-Mb physical, genetic, and radiation hybrid map of a 1p36.1 region implicated in neuroblastoma pathogenesis. Genes Chromosomes Cancer 2000;27:143-52.
www.elsevier.com/locate/humpath
Original contribution
Aberrant copy numbers of ALK gene is a frequent genetic alteration in neuroblastomas☆ Manish Mani Subramaniam MD, Marta Piqueras BS, Samuel Navarro MD, Rosa Noguera MD ⁎ Department of Pathology, Medical School, University of Valencia, 46010 Valencia, Spain Received 17 January 2009; revised 22 April 2009; accepted 4 May 2009
Keywords: Neuroblastoma; ALK; MYCN; Copy number alterations
Summary A total of 50 neuroblastomas were assessed for frequency of ALK gene copy number aberrations by interphase fluorescence in situ hybridization using a break-apart fluorescence in situ hybridization probe. The data were compared with status of MYCN, 11q, 17q, and 1p36. We observed ALK aberrations (amplification, 1 of 45; gain, 15 of 45 and loss/imbalance, 11 of 45) in a total of 27 (60%) of 45 neuroblastomas. Synchronic MYCN and ALK aberrations accounted for 23 of 45 (51%) tumors; however, MYCN alterations were also detected in 11 (60%) of 18 tumors without ALK aberrations. Our data suggest that copy number aberrations of the ALK gene is a frequent genetic event in the development of neuroblastomas. In addition, no correlation was observed between ALK aberrations and alterations of 11q, 17q, and 1p36. © 2009 Elsevier Inc. All rights reserved.
1. Introduction In advanced stages, neuroblastomas are one of the most intractable pediatric cancers, even with recent therapeutic advances. Over the years, a variety of genetic changes such as MYCN amplification (in approximately 25% of tumors) [1], loss of heterozygosity at 1p36 [2] and 11q [3], and 17q gain [4] have been implicated in the pathogenesis of neuroblastoma. In addition, expression of certain receptor tyrosine kinases such as TrkA [5] and TrkB [6] has been linked to respectively favorable and unfavorable outcomes in neuroblastomas. Recently, whole genomic DNA scan– ☆
Grant Support: From the ISCIII (RD06/0020/0102; ATC606/2007; FISPI06/1576), Fundación Inocente, Inocente (PI4/07-36) and CS (AP053/08). ⁎ Corresponding author. E-mail address: [email protected] (R. Noguera). 0046-8177/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2009.05.002
based methods and direct sequencing revealed oncogenic [7] germline and somatic [8] inactivating ALK mutations, indicating ALK to be a major predisposition gene [9] as well as a potential therapeutic target for neuroblastoma [10]. To date, genomic status of ALK has been assessed mainly using comparative genomic hybridization [11,12], comparative expressed sequence hybridization [11] and southern blot analysis [13], predominantly in tumor cell lines and some human neuroblastoma samples. To date, only a few studies have compared the genomic status of ALK with that of MYCN and other established genetic markers in archival samples of neuroblastomas [9,13]. In this study, we evaluated the copy number aberrations of ALK gene in formalin-fixed, paraffin-embedded clinical samples of neuroblastomas, as well as analyzed a possible relation between ALK aberrations and alterations of established prognostic genetic markers such as MYCN, 1p, 11q, and 17q.
ALK aberrations in neuroblastoma
2. Materials and methods 2.1. Tissue samples and fluorescence in situ hybridization probes A computer-based search of the pathology database yielded a total of 50 primary neuroblastoma samples classified as poorly differentiated (n = 28), NOS (not otherwise specified; n = 10), and undifferentiated (n = 12). Of these 50 tumors, 33 (66%) belonged to the high-risk group, and 17 (34%) to the low-risk group. MYCN gene status obtained from routine analysis of tumor touch preparations was available for all these samples. Of the 33 high-risk tumors, 27 showed MYCN amplification, 5 cases had MYCN gain, and there was 1 case with no MYCN alterations. Of the 17 low-risk tumors, 15 had no MYCN alterations, 2 were reported to be MYCN gain, and none of them harbored MYCN amplification. Representative tumor tissue cores from all these cases were subsequently assembled in a tissue microarray (TMA) with duplicate cores for each case. Sections, 4 μm thick, were cut from the TMA block and subjected to the formalin-fixed, paraffin-embedded fluorescence in situ hybridization (FISH) procedure [14] using a commercial ALK split probe (Dako, Inc, Glostrup, Denmark) composed of a Texas Red-labeled DNA probe (ALK-downstream) that binds to a 289-kb segment centromeric and fluorescein-labeled DNA probe (ALK-upstream) that binds to a 557-kb segment telomeric to the ALK breakpoint cluster region respectively on chromosome 2p23. In addition, we also confirmed MYCN gene status on the TMA tumor samples using MYCN (2p24) (labeled red) and LAF (2q11) control probe (labeled green) (Kreatech Biotechnology, Amsterdam, The Netherlands). Probe combinations for analysis of 1p36, 11q and 17q status included the following: chromosome 1p36 Midisatellite probe labeled green (MP Biomedicals, Illkirch, Cedex, France)/chromosome 1 satellite probe labeled red (Q-BIOgene, France) and ATM (11q22)specific DNA probe labeled in red/chromosome 11 satellite enumeration probe labeled with green (Kreatech Biotechnology, Amsterdam, The Netherlands) and p53 (17p13) green/ MPO (17q23) iso 17q red (Kreatech Biotechnology).
2.2. FISH scoring scheme The FISH signals were scored in 200 nonoverlapping nuclei per core, independently by two investigators (M.M. S. and R.N.), and the consensus was recorded. Four cellular groups were defined as follows [15]: group 1, cells with absent genetic alterations showing the same numbers of gene-specific and control probe signals; group 2, cells harboring genetic alterations demonstrating different copy numbers of gene-specific and control probe signals (Table 1); group 3, cells displaying nuclear truncation artifacts, for example, less MYCN/ALK signals compared to their respective control probe signals, more 1p36/11q
1639 Table 1 FISH criteria for copy number alterations of the different genetic markers Marker
Status
Description
MYCN/ALK
No Alteration
Same numbers of genespecific signals and control probe signals The number of genespecific signals is 1 to 4 times greater than the control probe signals Presence of at least 2 gene-specific signals and increased control probe signals The number of genespecific signals is more than 4 times the control probe signals Same numbers of 1 p36/11q specific signals and control probe signals At least 2 signals of 1p36/ 11q specific locus and more control probe signals One signal representing 1p36/11q region with at least 2 control probe signals Similar numbers of 17q regions signals and control probe signals The numbers of 17q signals are 1 to 4 times greater than control probe signals
Gain
Loss/Imbalance
Amplification
1p36/11q
No deletion
Loss/Imbalance
Deletion
17q
No Gain
Gain
The FISH criteria for MYCN , 11q, 1p36, and 17q alterations are according to the guidelines specified by the European Neuroblastoma Quality Assessment Group.
signals compared to their control probe signals, or less 17q23 signals compared to the control probe signals; group 4, net percentage of cells with definite genetic alteration, obtained by subtracting the percentage of cells showing truncation artifacts (group 3) from those of group 2, thereby eliminating the false positives. The FISH scoring scheme was initially evaluated on a series of 16 control tissues such as normal kidneys (n = 5) and nonneuroblastic tumors (breast cancers, n = 6 and gastric cancers, n = 5) and then on neuroblastomas in order to obtain a cutoff value for each FISH probes. The mean percentage of cells in group 4 in the control tissues were as follows: ALK (12.5%); MYCN (13%); deletion of 11q (11%) and 1p36 (11.5%); and gain of 17q (10%), thereby making an overall mean of 11.6 (SD 1.19) for all the genetic markers. The mean+3SD of the percentage of cells in group 4 for the control tissues is 15%. Thus, a tumor sample was considered positive for a genetic alteration if the percentage of cells in group 4 exceeded the mean+3SD
1640
M. M. Subramaniam et al.
(15%). The FISH criteria for the different aberrations are summarized in Table 1.
3. Results Of the 50 neuroblastomas, 45 (90%) were informative for all the FISH assays. The noninformative samples were constituted by the 5 low-risk neuroblastomas without MYCN amplification. The status of ALK, MYCN, 11q, 1p36, and 17q in neuroblastomas is summarized in Table 2. Considering only the 45 informative cases, aberrant copy numbers of ALK gene were observed in 27 (60%) samples. ALK amplification was observed in 1 (2%) of 45, ALK gain in 15 of 45 (33%), ALK loss/imbalance in 11 (24%), and no alterations of ALK in the remaining 18 (40%) cases. The solitary case of ALK amplification revealed between 25 and 30 ALK signals (Fig. 1), whereas the ALK gain cases displayed between 3 and 6 signals. MYCN alterations were observed in 34 (76%) of 45 cases. Of the 45 cases, MYCN amplification was found in 27 (60%) cases, MYCN gain in 7 (15%), and no alterations in 11 (24%) cases. Coexistent MYCN and ALK alterations were observed in 23 (51%) of 45 cases. Out of the 27 cases with ALK alterations, MYCN was altered in 23 (85%). One of the important findings is that of the 15 cases with ALK gain, MYCN amplification was found in 10 (67%) and MYCN gain in 4 (26%) cases. Nevertheless, 11 (61%) of the 18 cases without ALK alterations also showed MYCN alterations. Gain of 17q was detected in 21 of the 27 (77%) ALK-altered cases as well as in 14 (77%) of 18 cases without ALK alterations, signifying no association.
Fig. 1 Representative FISH image of the neuroblastoma cells displaying amplification of ALK gene evident by the increased numbers (25-30 approximately) of ALK upstream probe (telomeric to the ALK break point cluster region on 2p23) signals (green) compared to two signals of ALK-downstream probe (binding centromeric to ALK break point cluster region) (red). Arrows indicate two representative non-overlapping nuclei in this field exhibiting ALK amplification (DAPI counterstain, original magnification ×400).
Deletions of 11q and 1p36 were detected in 14 (31%) of 45 and 28 (62%) of 45 cases respectively. However, no correlation was observed between ALK alterations and deletions of 11q and 1p36.
4. Discussion Table 2 Frequency of copy number aberrations of ALK and its correlation with status of MYCN, 11p, 17q, and 1p36 ALK loss/ No ALK Imbalance alterations (n = 18) ALK ALK gain (n = 11) amplification (n = 15) (n = 1)
ALK status (total cases n = 45)
MYCN A G NA 17q G N 11q D ND 1p36 D ND
1 0 0 0 1 0 1 1 0
10 4 1 13 2 4 11 11 4
8 0 3 8 3 5 6 6 5
8 3 7 14 4 5 13 10 8
Abbreviations: A, Amplification; G, gain; NA, no alteration; D, deletion; ND, no deletion.
The complexity of the organization of genes in the 2p amplicon indicates that not only the MYCN locus but flanking and/or unrelated regions might be involved in the aggressive biological behavior of neuroblastomas. Besides MYCN, other genes such as DDXI [16] belonging to the family of genes encoding DEAD box proteins, ID2 (Inhibitor of DNA-binding/differentiation) [17], NAG (neuroblastoma amplified gene) [18], SNTG2, and TPO [19] have also been reported to be co-amplified in MYCN amplified neuroblastomas. A recent addition to this list is the ALK gene situated at 2p23.2. This is the first TMAbased interphase-FISH study to assess the frequency of copy number aberrations of the ALK gene and its possible association with MYCN, 17q, 11q, and 1p36 status in archival samples of neuroblastomas. Aberrant ALK copy numbers were detected in 27 (60%) of 45 neuroblastomas, with amplification constituting 1 (2%), gain 15 (33%), and loss/imbalance 11 (24%) of these
ALK aberrations in neuroblastoma cases. A straightforward comparison of our data with previously published results is difficult owing to the different methodologies employed in assessing the ALK genomic status, as well as to the data from cell lines versus that from clinical samples. Overall, our results were slightly higher when compared to previous data. However, frequency of ALK amplification detected in our study was comparable to that of the following previous reports: Osajima et al (1 [1%] of 85, in tumor tissues and 3 (12%) of 25, in cell lines by southern blot analysis) [13], Caren et al (5% in tumor tissues) [12], Mosse et al, (16 [3.3%] of 491 primary neuroblastomas by SNP-based microarrays) [9], and Stock et al (ALK amplification in one tumor and 2 (15%) of 13 cell line samples) [11]. In our study, the primary tumor with ALK amplification exhibited over 30 ALK signals in contrast to a previous report [13] in which cell lines (N30 copies) revealed increased copy numbers of ALK, as opposed to tumor tissues (2 to 10 copies). Increased frequencies of ALK amplification in tumor cell lines, a common feature reported in the above studies can be attributed to the mechanism by which cells with higher ALK copy numbers become the major population during establishment of cell lines because of their growth advantage. The ALK gain noted in our study was consistent with the partial trisomy documented by SNP-based microarrays; however, our frequencies were greater than the earlier reported frequencies: 8.2% [13], 112 (22.8% ) of 491 [9], and 2 (15%) of 13 (in cell lines) [11]. The mean number of ALK signals in cases with gain (3-6 signals) was higher than that reported earlier by Osajima et al (1.8-3.0) [13], a fact that can also be attributed to the different methods of copy number analysis. In addition, we also observed loss/imbalance of ALK gene characterized by presence of at least 2 ALK signals but an increased number of control probe signals. ALK loss/imbalance has not been reported in previous studies. Alternatively, loss/imbalance of ALK may also be explained by a possible chromosomal duplication of the 2p23.2 region with allelic loss. Taking our findings and the previous data together, aberrant copy numbers of ALK seems to be a frequent genetic event in the tumorigenesis of neuroblastomas. Concordant ALK and MYCN copy number aberrations accounted for 23 (51%) of 45 neuroblastoma cases. Furthermore, 23 (85%) of 27 cases with ALK aberrant copy numbers also revealed MYCN copy number aberrations. However, MYCN alterations were also seen in 60% of cases devoid of ALK alterations. It is important to note that the higher frequency of MYCN amplification obtained in our study was due to the inclusion of a relatively high proportion of high-risk neuroblastomas in our TMA-based analysis. Our results are consistent with previously published data wherein synchronic ALK and MYCN alterations has been reported [11,9,13]. Based on our data and the previous findings, we observe a tendency of synchronic copy number aberrations of ALK and MYCN in a majority of cases. The lack of ALK alterations in over 60% of cases with MYCN alterations observed in our study, as well as the less frequent coexistence
1641 of ALK and MYCN alterations in some of the aforementioned studies [11,13] suggest that there is no strong evidence to support the coexistence of ALK and MYCN aberrations. Moreover, the ALK gene locus (2p23.2) appears too far from the MYCN gene locus to be within a single amplicon. Therefore, we assume that copy number aberrations of ALK and MYCN might have an independent function in the pathogenesis of neuroblastomas. The frequencies of 17q, 11q, and 1p36 alterations derived from our study were consistent with published data [4,20,21]; however, no correlation was observed between them and ALK aberrations. In summary, our simple and easily reproducible FISH technique shows that copy number aberrations of ALK gene (amplification, gain and loss) is a frequent genetic event in the development of neuroblastomas.
References [1] Seeger RC, Brodeur GM, Sather H, et al. Association of multiple copies of the N-myc oncogene with rapid progression of neuroblastomas. N Engl J Med 1985;313:1111-6. [2] Brodeur GM, Sekhon G, Goldstein MN. Chromosomal aberrations in human neuroblastomas. Cancer 1977;40:2256-63. [3] McArdle L, McDermott M, Purcell R, et al. Oligonucleotide microarray analysis of gene expression in neuroblastoma displaying loss of chromosome 11q. Carcinogenesis 2004;25:1599-609. [4] Bown N, Cotterill S, Lastowska M, et al. Gain of chromosome arm 17q and adverse outcome in patients with neuroblastoma. N Engl J Med 1999;340:1954-61. [5] Nakagawara A, Arima-Nakagawara M, Scavarda NJ, et al. Association between high levels of expression of the TRK gene and favorable outcome in human neuroblastoma. N Engl J Med 1993;328:847-54. [6] Nakagawara A, Azar CG, Scavarda NJ, et al. Expression and function of TRK-B and BDNF in human neuroblastomas. Mol Cell Biol 1994; 14:759-67. [7] Chen Y, Takita J, Choi YL, et al. Oncogenic mutations of ALK kinase in neuroblastoma. Nature 2008;455:971-4. [8] Janoueix-Lerosey I, Lequin D, Brugières L, et al. Somatic and germline activating mutations of the ALK kinase receptor in neuroblastoma. Nature 2008;455:967-70. [9] Mossé YP, Laudenslager M, Longo L, et al. Identification of ALK as a major familial neuroblastoma predisposition gene. Nature 2008;455: 930-5. [10] George RE, Sanda T, Hanna M, et al. Activating mutations in ALK provide a therapeutic target in neuroblastoma. Nature 2008;455:975-8. [11] Stock C, Bozsaky E, Watzinger F, et al. Genes proximal and distal to MYCN are highly expressed in human neuroblastoma as visualized by comparative expressed sequence hybridization. Am J Pathol 2008;172: 203-14. [12] Carén H, Abel F, Kogner P, et al. High incidence of DNA mutations and gene amplifications of the ALK gene in advanced sporadic neuroblastoma tumours. Biochem J 2008;416:153-9. [13] Osajima-Hakomori Y, Miyake I, Ohira M, et al. Biological role of anaplastic lymphoma kinase in neuroblastoma. Am J Pathol 2005;167: 213-22. [14] Subramaniam MM, Noguera R, Piqueras M, et al. p16INK4A (CDKN2A) gene deletion is a frequent genetic event in synovial sarcomas. Am J Clin Pathol 2006;126:866-74. [15] Piqueras M, Navarro S, Castel V, et al. Analysis of biological prognostic factors using tissue microarrays in neuroblastic tumors. Pediatr Blood Cancer 2008;52(2):209-14.
1642 [16] Squire JA, Thorner PS, Weitzman S, et al. Co-amplification of MYCN and a DEAD box gene (DDX1) in primary neuroblastoma. Oncogene 1995;10:1417-22. [17] Lasorella A, Boldrini R, Dominici C, et al. Id2 is critical for cellular proliferation and is the oncogenic effector of N-myc in human neuroblastoma. Cancer Res 2002;62:301-6. [18] Kaneko S, Ohira M, Nakamura Y, et al. Relationship of DDX1 and NAG gene amplification/overexpression to the prognosis of patients with MYCN-amplified neuroblastoma. J Cancer Res Clin Oncol 2007; 133:185-92.
M. M. Subramaniam et al. [19] Łastowska M, Viprey V, Santibanez-Koref M, et al. Identification of candidate genes involved in neuroblastoma progression by combining genomic and expression microarrays with survival data. Oncogene 2007;26:7432-44. [20] Spitz R, Hero B, Simon T, et al. Loss in chromosome 11q identifies tumors with increased risk for metastatic relapses in localized and 4S neuroblastoma. Clin Cancer Res 2006;12:3368-73. [21] Spieker N, Beitsma M, van Sluis P, et al. An integrated 5-Mb physical, genetic, and radiation hybrid map of a 1p36.1 region implicated in neuroblastoma pathogenesis. Genes Chromosomes Cancer 2000;27:143-52.