Comparative genomic hybridization of primary sinonasal adenocarcinomas

335

Comparative Genomic Hybridization in Primary Sinonasal Adenocarcinomas Manuela Ariza, Ph.D.1 Jose´ Luis Llorente, M.D.1,2 Cesar Alvarez-Marcas, M.D.1,3 Lucia Baragan˜o, M.D.2 Ana Salas, Ph.D.1 Nuria Rodriguez Prado, M.D.2 Mario Hermsen, M.D.1 Carlos Sua´rez, M.D.1,2 Andres Sampedro, M.D.1 1

Instituto Universitario de Oncologia del Principado de Asturias, University of Oviedo, Oviedo, Spain.

2

Department of Otolaryngology, Hospital Central de Asturias, Oviedo, Spain.

3

Department of Otolaryngology, Hospital Valle del Nalo´n, Asturias, Spain.

BACKGROUND. Little is known about the genetic alterations that occur in sinonasal adenocarcinomas. The goal of the current study was to detect recurrent chromosomal gains and losses in a series of 21 primary sinonasal adenocarcinomas using comparative genomic hybridization (CGH). METHODS. The authors examined ethmoid sinus adenocarcinoma samples from 21 patients. All 21 adenocarcinomas were associated with work-related exposure to wood dust. CGH was used to detect chromosomal abnormalities, and the results of CGH analysis were evaluated for correlations with clinicopathologic characteristics. RESULTS. Chromosomal gains and losses were detected in all 21 adenocarcinomas. Gains were detected at high frequencies at 7q11–21 (n ⫽ 15 [71%]), 18p11 (n ⫽ 14 [66%]), 8q11–22 (n ⫽ 13 [62%]), 5p11–13 (n ⫽ 12 [57%]), 12q11–13 and 19p (n ⫽ 11 [52%]), 20q (n ⫽ 10 [47%]), X and 5p (n ⫽ 9 [43%]), and 3q26 –27 (n ⫽ 8 [38%]); and losses were detected at 8p22–23 (n ⫽ 18 [86%]), 18q22–23 (n ⫽ 17 [80%]), 17p13 (n ⫽ 12 [57%]), and 5q31– qter (n ⫽ 11 [52%]). Aside from low-level gains, 43 highlevel amplifications were observed in the current series of 21 tumors, most commonly at Xq13 (n ⫽ 7 [33%]). CONCLUSIONS. CGH revealed that ethmoid sinus adenocarcinomas carry a large number of chromosomal losses and gains, including high-level amplifications. To the authors’ knowledge, the current study represents the first attempt to investigate sinonasal adenocarcinomas on a genetic level by using CGH. The pattern of chromosomal abnormalities in these tumors was different from the pattern in other tumors within the same anatomic region (e.g., squamous cell carcinomas and salivary gland tumors); this finding may be explained by differences in etiology. Nonetheless, sinonasal adenocarcinomas appear to be genetically similar to adenocarcinomas of the stomach and colon, which also have an etiology that differs from that of sinonasal adenocarcinomas. Further study is necessary to better understand the molecular genetic basis underlying the development of sinonasal adenocarcinomas. In the near future, this type of understanding may present new possibilities for prevention and treatment of malignant disease. Cancer 2004;100:335– 41. © 2003 American Cancer Society.

Supported by a grant from the University of Oviedo (AYP-02-520).

KEYWORDS: adenocarcinoma, paranasal sinus neoplasms, comparative genomic hybridization, woodworkers.

Dr. Ariza is a postdoctoral fellow sponsored by the Obra Social Caja Astur.

S

Address for reprints: Jose´ Luis Llorente, c/ JM Caso, 14, 33006 Oviedo, Asturias, Spain; Fax: (011) 34 98 5103658; E-mail: llorentependas@ telefonica.net Received July 2, 2003; revision received September 15, 2003; accepted October 29, 2003. © 2003 American Cancer Society DOI 10.1002/cncr.11931

inonasal adenocarcinoma is a rare epithelial tumor that is well characterized morphologically by the presence of neoplastic glandular structures that arise in the nasal cavity and paranasal sinuses.1 The incidence of sinonasal tumors is less than 1 case per 100,000 individuals per year, and only 4 –20% of sinonasal tumors are adenocarcinomas.2,3 Sinonasal adenocarcinomas are located mainly in the ethmoid sinus and in the upper part of the nasal cavity.3,4 These tumors clearly are more common among men (the observed male-

336

CANCER January 15, 2004 / Volume 100 / Number 2

to-female ratio ranges from 4:1 to 20:1), and the mean age of patients at presentation ranges from 60 to 65 years.4 The etiology of sinonasal adenocarcinomas includes work-related exposure to wood dust, mainly among carpenters and cabinetmakers (average length of exposure, 20 years; latency period, 0 – 40 years).4,5 Local recurrence is the most common cause of death among patients with sinonasal adenocarcinoma.6 Distant and lymph node metastases are rare (5–10% of all cases). Initial genetic investigations of single-gene loci by our group and others have revealed mutations and allelic losses in the ras gene family.7,8 In the last decade, more extensive analysis of chromosomal aberrations has been made possible by the introduction of new molecular cytogenetic techniques, such as comparative genomic hybridization (CGH). CGH first was described in 1992 by Kallioniemi et al.9 as a research tool for screening tumors for chromosomal abnormalities; the value of this technique has been demonstrated in a large number of studies of malignant tumors. Using CGH, we investigated a series of sinonasal adenocarcinomas to identify recurrent chromosomal alterations.

washing once in ethanol. The samples then were dried and resuspended in 1 mL sodium thiocyanate (1 M) and incubated overnight at 37 °C. Tissue samples were washed and resuspended in 400 ␮L DNA isolation buffer (75 mM sodium chloride, 25 mM etyhlenediamine tetraacetic acid, and 0.5% (tween-20). Proteinase K was added to a final concentration of 1 mg/mL, and tissue samples were incubated overnight at 55 °C. The DNA then was purified by phenol extraction. Concentration and purity were assessed by measuring absorbance at 260 and 280 nm. DNA quality was assayed on a 1% agarose gel. In general, DNA from frozen tissue samples appeared as a band at ⬎ 20 kilobases (kb), whereas DNA from paraffin-embedded tissue samples appeared as a smear between 2 and 20 kb. Tumor DNA was labeled with biotin-16-2⬘-deoxyuridine-5⬘-triphosphate (dUTP; Roche Diagnostics, Basel, Switzerland) and control DNA was labeled with digoxigenin-11-dUTP (Roche Diagnostics) in a standard nick translation reaction. The lengths of the labeled products obtained were adjusted by varying the amount of additional DNase I in the labeling reaction. For hybridization, labeled DNA (from both frozen and paraffin-embedded tissue samples) of ideal length (500 –2000 base pairs) was used.

MATERIALS AND METHODS Sample Characterization

CGH

Samples of 21 primary sinonasal adenocarcinomas (7 frozen and 14 formalin-fixed and paraffin-embedded) were obtained from the Otolaryngology Department of the Hospital Central de Asturias (Oviedo, Spain). All tumors analyzed were from male patients ages 40 – 82 years (mean, 59 years) who had been exposed to wood dust for 5–54 years (mean, 32 years). The latency period, from the end of exposure to the appearance of the tumor, ranged from 0 to 27 years (mean, 6 years). There were 3 Stage I tumors, 3 Stage II tumors, 10 Stage III tumors, and 5 Stage IV tumors.10 All tumors arose in the ethmoid sinus region or in the middle concha, and no patient had metastases at the time of diagnosis. Clinical and histopathologic data and data on follow-up were available in all cases and are summarized in Table 1. None of the patients had received any type of treatment before surgery.

CGH was performed as described elsewhere.11 In brief, we ethanol-precipitated 1 ␮g of biotinylated tumor DNA and 1 ␮g of digoxigenin-labeled control DNA in the presence of 10 ␮g of salmon sperm DNA and 50 ␮g of human COT1 DNA (Life Technologies, Gaithersburg, MD). The probe mixture then was dried and resuspended in 10 ␮L hybridization solution (50% formamide, 2X SSC [1X SSC: 150 mM sodium chloride and 15 mM sodium citrate], and 10% dextrane sulfate). The DNA was denatured at 85 °C for 5 minutes and allowed to reanneal at 37 °C for 30 minutes. The metaphase preparations were denatured at 80 °C for 2 minutes in 70% deionized formamide and 2X SSC and dehydrated in an ethanol series (70%, 90%, and 100%). The probe mixture was applied to the denatured metaphase chromosomes under a coverslip (area, 18 mm2), and hybridization was allowed to take place for 3 days at 37 °C in a humid chamber. Tumor and control DNA competed to hybridize to the chromosomes. After washing, control DNA was stained with antidigoxigenin Fab fragments conjugated with rhodamine, and biotinylated tumor DNA sequences were detected with fluorescein isothiocyanate conjugated with avidin. Chromosomes were counterstained with 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) and embedded in an antifading agent to reduce photobleaching.

DNA Preparation Control genomic DNA (used as a reference in CGH experiments) from the blood of a healthy male donor and tumor DNA from the frozen sinonasal adenocarcinoma samples were obtained via phenol/chloroform extraction following standard DNA isolation protocols. DNA from formalin-fixed, paraffin-embedded tumor samples was isolated by cutting 50 ␮m thick tissue sections, deparaffinizing in xylene (45 °C for 15 minutes), and

Sinonasal Adenocarcinomas and CGH/Ariza et al.

Microscopy and Digital Image Analysis Gray-level images were acquired for each fluorochrome using a cooled charge-coupled device camera (COHU, San Diego, CA) coupled to a DMRXA epifluorescence microscope (Leica, Wetzlar, Germany). The images were obtained via sequential exposure through fluorochrome-specific filters for DAPI, fluorescein, and rhodamine (Leica) using the Q-FISH ACAPS imaging system (Leica). Chromosomes were identified by their DAPI banding patterns. Hybridization was evaluated quantitatively with a custom-designed computer program (QCGH software; Leica). Each average ratio profile was calculated as the mean value from at least 10 metaphases, and the results were combined to produce an average fluorescence ratio for each chromosome. Chromosomal gains or losses were indicated by a fluorescence ratio that was significantly greater than or less than 1.0, respectively, as determined by the 95% confidence interval. High-level amplification was indicated by a ratio greater than 1.5. All centromeres, as well as chromosome p35–36 and the heterochromatic regions of chromosomes Y, 16, 19, and 22, were excluded from further analysis, because these regions can yield unreliable CGH data, due to incompletely suppressed repetitive DNA sequences. Control-versus-control experiments were performed, including experiments involving control DNA obtained from paraffin-embedded tissue, and always produced normal ratio profiles. All data were analyzed using the SPSS 11.0 software package (SPSS Inc., Chicago, IL).

RESULTS CGH revealed DNA gains and losses in all 21 sinonasal adenocarcinomas investigated. Gains were more common than losses. All data are described in Table 1, and an overview of the CGH results is provided in Figure 1. The most frequently occurring gains were detected at chromosome arms 7q11–21 (n ⫽ 15 [71%]), 18p11.2 (n ⫽ 14 [66%]), 8q11–22 (n ⫽ 13 [62%]), 5p11–13 (n ⫽ 12 [57%]), 12q11–13 and 19p (n ⫽ 11 [52%]), 20q (n ⫽ 10 [47%]), X and 5p (n ⫽ 9 [43%]), and 3q26 –27 (n ⫽ 8 [38%]). Frequently occurring losses were detected at chromosome arms 8p22–23 (n ⫽ 18 [86%]), 18q22–23 (n ⫽ 17 [80%]), 17p13 (n ⫽ 12 [57%]), and 5q31– qter (n ⫽ 11 [52%]). Aside from low-level gains, 43 high-level amplifications (indicated by bold bars in Fig. 1) were detected in the series of 21 samples. The most frequently amplified segment was Xq13 (n ⫽ 7 [33%]). Other recurring amplified regions included Xq25–26 and Xq23 (n ⫽ 4 [19%] for both). Amplification also occurred at 7q11, 12p13, and 12q11–13 (n ⫽ 3 [14%]) and at 7q21–22, 8q24, 20q, and Xp22 (n ⫽ 2 [9%])

337

(Table 1). Gains and losses seldom involved whole chromosomes, but whole chromosome arm events were observed in some cases. Gain of one chromosome arm together with the loss of the other, indicative of isochromosome formation, was observed at 5p/5q (n ⫽ 4 [20%]), 8q/8p (n ⫽ 3 [16%]), and 17q/17p (n ⫽ 2 [10%]). Preliminary analysis of follow-up data (mean overall survival, 2.6 years; range, 0 –14 years) indicated that all patients with Stage I disease (n ⫽ 3) had a favorable outcome, in contrast to patients with Stage IV disease (n ⫽ 5), all of whom died within 1 year after surgery. Patients with Stage II–III disease (n ⫽ 12) had mixed outcomes; 6 remained alive, and 6 died of disease. Crosstabulation of tumor stage in these three groups revealed correlations with chromosomal loss of 4p, chromosomal loss of 5q, and chromosomal gain of 20q (Pearson ␹2: P ⫽ 0.019, P ⫽ 0.051, and P ⫽ 0.019, respectively). Of these alterations, only loss of 5q was correlated with overall survival (P ⫽ 0.0124; log-rank, 6.25).

DISCUSSION To our knowledge, the current study represents the first attempt to investigate sinonasal adenocarcinomas on a genomewide level using CGH. The first conclusion that can be drawn from the data is that the pattern of chromosomal abnormalities in these tumors is different from the pattern in other tumors within the same anatomic region (e.g., squamous cell carcinomas and salivary gland tumors). For example, loss of the entire 3p arm, gain of the entire 3q arm, and high-level amplification of 11q13 are common abnormalities in squamous cell carcinomas but almost never are observed in sinonasal adenocarcinomas. Furthermore, although losses at 8p and 18q and gains at 12q and X do occur in squamous cell carcinomas, they are much more common in sinonasal adenocarcinomas. Some chromosomal alterations, such as gains at 5p, 7q11–21, and 18p, are common in both tumor types.12 In salivary gland tumors, cytogenetic similarities between adenoid cystic carcinoma and polymorphous low-grade adenocarcinoma have been reported13; however, the profile of these tumors differs from the profile of nasal adenocarcinoma.14 These genetic differences may be attributable to differences in etiology. Whereas squamous cell carcinoma is related to tobacco and alcohol use, sinonasal adenocarcinoma appears to be closely related to exposure to hardwood dust. It can be speculated that different types of carcinogens produce different patterns of chromosomal abnormalities. Nonetheless, many of the chromosomal gains and losses observed in the current series of sinonasal adenocarcinomas—

Woodworker (5)

Woodworker (15) Woodworker (NA)

Woodworker (occasional)

Woodworker (occasional)

Woodworker (45)

Woodworker (35)

Woodworker (35)

Woodworker (24)

2

3

5

6

7

8

9

10

4

Woodworker (30)

Profession (yrs of employment)

1

Patient no.

46

71

51

59

49

71

59

53

50

53

Age (yrs)

Ethmoid

Ethmoid and brain Ethmoid

Ethmoid

Middle concha

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Tumor origin

Paralateronasal

Paralateronasal

Subcranial

Subcranial

Paralateronasal

Craniofacial

Craniofacial

Craniofacial

Paralateronasal

Paralateronasal

Surgical approach

T2N0M0

T2N0M0

T4bN0M0

T3N0M0

T1N0M0

T3N0M0

T3N0M0

T3N0M0

T2N0M0

T3N0M0

TNM status

No

Yes

—

No

Yes

Yes

Yes

No

Yes

No

Recurrence

12q24, 16q23–24, 22q12–13

2p25, 4p, 4q, 5q, 8p, 10q23–26, 14q, 15q11–13, 17p, 18p, 18q, 22q 4q33–35, 5p, 5q, 6p, 6q, 8p21–23, 9p22–24, 10q25–26, 16q23–q24, 18q12–23

1p32–36, 3p21, 4p15–16, 5q, 6q23–27, 8q24, 8p23, 10p15, 10q23–26, 11q13, 11q24–25, 11p15, 13q33–34, 14q32, 15q26, 16q22–24, 17p13, 17q21–25, 18q22–23, 19q13,21q22, 22q12–13 1p32–36, 2p23–25, 8p12–23, 9q32– 34, 10p13–15, 10q26, 11q23, 12q24, 13q33–34, 14q31–32, 15q23–26, 16p, 17p, 17q, 18q21–23 4q, 9q12–13, 13q22–31, X

8p23, 9p11, 11p15, 12q24, 14q11, 17p, 17q21–22, 21q11 9p, 18q22–23, 20p13

8p21–23, 9q22–34, 10q26, 17p13, 21q22, 22q12–13

2q37, 2p24–25, 3p25–26, 8p22–23, 11q24–25, 12p13, 15q25–26, 18q23, 21q22

Losses

TABLE 1 Clinical and Comparative Genomic Hybridization Data from Patients with Primary Sinonasal Adenocarcinoma

1p32–36, 8q24, 9q34, 11p15, 11q13, 12p13, 14q32, 16p13, 17q12–21, 17q25, 19, 20, 22q13 2p13–22, 3p, 3q, 5p, 6p, 6q, 7p, 7q, 8q, 10p11–q22, 12p, 12q, 13q, 16p, 17q, 19q, 20p, 20q, Xp, Xq 1p13–q21, 1p36, 2q31–34, 3p21–22, 3q25–29, 7q11–31, 8q21–24, 11q12– 23, 12p, 12q11–15, 17q, 18p11–q11, 19p, 19q, 20p, 20q, 21q22, 22q11–13, Xp11–21, Xp13–q21 2p11–12, 2q32, 3p12, 3q24–26, 4p11–13, 4q, 5p11–15, 6q11–22, 7q11–31, 8q11–22, 10p11, 11p11–13, 12q11–21, 17p11–12, 18p, 19p, 21q11–21, X

1p11–q22, 4p13–q26, 5p13–p12, 7q11– 31, 8p11–q23, 9q11–21, 10q11, 13q11–12, 14q11–12, 18p, 20q11, 21q11–21, X

2q36–37, 4p15.3–16, 5q11, 7q11, 8p11– q11, 8q21, 11q13, 12p11–q12, 18p11– q11, 19p, 19q, 20p11, 22q13, X 1p31–q25, 2p12–16, 2q22–35, 3q, 4q, 5p, 6p–q16, 7p, 7q, 8q11–23, 9q32–33, 12p–q23, 13q22–31, 14q11–21, 16p– q13, 17p11–q11, 18p–q11, 20p, 20q, 21q11–21, X

1p, 1q, 3p14–21, 3q24–27, 4p11–q11, 5p11–15, 6p22–24, 6q, 7p, 7q11–31, 8p21–q23, 9p, 9q, 10q11–24, 11q12– 13, 13q11–22, 16p, 17p, 17q, 19p13, 20p, 20q, 21q21, 22, Xp-q25 4p11–15, 4q11–13, 5p11–15, 5q14–21, 5q33–35, 6q11–15, 7q11–31, 8q11–21, 9p21, 10p11–21, 11p11–15, 12p, 12q15–21, 17p11, Xp, Xq 5q11, 8q11–21, 11p11, 13q31–34, X

Gains

7q21–22, 12p13, Xq25

10q21, 12p13, 20p, 20q, Xq13

7q11, Xq

7q11–12, 7q21–22, 12q21, 15q11– 13, 18p11, 20p11–q11, Xq12–13

Xp22, Xp25, Xq

Xq12–13

9p23–21, 9q21, 14q24–32, Xq13, Xp22

Amplifications

3

1

0.2

2

4

2

1

1

3

10

Survival (yrs)

Alive

Dead

Dead

Alive

Alive

Dead

Dead

Alive

Dead

Alive

Status

Woodworker (54)

Woodworker (43)

Woodworker (30) Woodworker (30)

Woodworker (NA)

Woodworker (10)

Woodworker (40)

Woodworker (occasional)

Woodworker (occasional)

Woodworker (occasional)

12

13

14

16

17

18

19

20

21

NA: not available.

15

Woodworker (40)

11

40

51

63

82

66

68

61

67

69

66

61

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Ethmoid

Craniofacial

Paralateronasal

Subcranial

Subcranial

Craniofacial

Craniofacial

Craniofacial

Craniofacial

Paralateronasal

Paralateronasal

Paralateronasal

T4bN0M0

T4aN0M0

T4aN0M0

T3N0M0

T4bN0M0

T4bN0M0

T3N0M0

T3N0M0

T3N0M0

T1N0M0

T1N0M0

Yes

Yes

Yes

No

Yes

Yes

No

Yes

Yes

Yes

No

2p25–23, 2q12–24, 2q34–37, 3p, 3q, 4p16–15, 5p, 5q33–35, 6p25–23, 8p23–22, 8q24, 10p13–14, 10q21, 10q24–26, 11p15–14, 11q14–25, 13q21–34, 14q31–32, 15q, 18q12–23, 20q13 2p12–q22, 4p16–15, 5p15, 5q31– 35, 8p23–21, 8q24, 9p24–23, 10p15–13, 11p15–14, 11q24–25, 12q24–24, 13q33–34, 14q31–32, 17p, 18q23, 20p, 20q, 22q12–13

1p22–21, 2p25, 2q22–37, 3p26, 3p14–12, 4q28–35, 5q, 6q14–22, 8p23–22, 9p, 10p, 10q, 11p, 11q, 13q, 17p, 18q22–23, 21

6q12–23, 8p–q23, 11q13–23, 13q, 16p, 16q, 18q

1p36, 4p, 4q, 5q14–35, 8p23–21, 9q34, 15q, 16p, 16q, 17p, 17q, 18q21–23, 20q, 21q22, 22q 5q11–23, 8p, 13q33–34, 17q11–21, 18q12–23, Xp22–q13 1p36–32, 4p15–16, 5q, 6p25–21, 6q25–27, 8p23–12, 9q34, 12q23–24, 14q23–32, 15q, 16q22–24, 16p13–12, 17p, 17q, 18q, 19p, 20p, 21q, 22q 1p31–12, 2p24–16, 4p15–11, 4q, 5q11–33, 6q, 10p–q24, 11q13– 25, 13q11–22, 18q, 21q, Xp22– q24 4p15–q35, 5q, 8p, 9p24–21, 11p15, 14q11–23, 16q, 17p, 17q, 18p, 18q, 19p13, 22q

2p25–24, 4q28–35, 5p15.3–14, 8p23–22, 9p24–21, 10p15–14, 10q25–26, 13q21–34, 14q31–32, 15q26, 18q22–23

2p24–25, 6q25–27, 10q26, 11q24– 25, 18q21–23

1p36–33, 1q23, 1q32–41, 2q36–37, 3p, 5p, 7p, 7q, 8q24, 9q32–34, 11p15– q13, 12p,14q22–32, 18p, 19p, 19q, 20p, 20q, Xq25–28 2p–q22, 3p, 3q22–23, 3q26.3–29, 5p, 6q21–23, 7p22–q22, 7q33, 8q, 10q25– 26, 10p15–q11, 11q13–25, 11p14–12, 15q11–21, 20q, Xp11.4–q28 1p36–31, 1p13–q21, 1q32–42, 4p16, 6p25–21, 7p22–21, 9p12–q13, 9q34, 17q25, 19p13, 19q11–13, 20q 1p36–34, 1p13–q24, 1q32–44, 2q11, 3p21, 4q12, 5p, 6p22, 7q11–22, 8q11, 9q, 12p13–q14, 12q23–24, 15q23, 16p12–11, 17q, 18p11–q11, 19p, 19q, 20p, 20q, 22, Xp, Xq 1p36–31, 1p21–q41, 4q11–13, 4q31, 4q11–12, 6q24–25, 7q11–21, 7q31, 8p12–q23, 9p13–q22, 10p11–q11, 11q11–12, 12q11–13, 12q23, 13q14, 14q13, 16p12–q24, 17p11–q21, 18p11, 19p, 19q, 20p, 22q11, Xp22– q25 2q24–33, 3p22, 3p12–q13, 3q25–27, 4p13, 4q28, 4q32–34, 7q21, 8q13, 9q12–21, 11q22, 13q21–22, 15q11.2, 16q11.2–12.1, 18p11–q11, 19p13–12, 21q21, X

1p11–q21, 2p15, 2q22–35, 3p22, 3q11– 13, 3q24–27, 5p, 7p–q22, 8q211–23, 9q11, 12p11, 12q21, 13q21–32, 20q, Xq21, Xq22

1q23–q31, 7p, 7q, 8q, 18p, Xq21–q28

1p21–q22, 7q11–21, 8q11–13, 12p–q14, 14q11, 16p11–q13, 17q11–12, 17q21, 19p, 19q, 21q11, 22q11–12, X 1p31–36, 1p13–q22, 3p21, 4p12–q12, 7q11, 7q22, 8p11–q11, 9q11–21, 9q34, 10p11.2, 11p11–q13, 12p11– q13, 12q22–24, 16p, 16q, 17p–q22, 18p11, 19p, 19q, 20q11–12, 22, X 2q33–36, 3p, 3q, 7p15–q34, 8q22–23, 12p13–11, 13q21–34, 14q22, 14q24

7q11.2, 16q22, 19p13, Xq23

1q41, 12p12–13, 19p, 20q, Xq

8q24

5p15.3–13

8q24

1p36, 12p11–q13

8q11–12, 12p12, 12q12–13

0.75

0.6

0.6

0.75

0.5

2

2

2

4

14

3

Dead

Dead

Dead

Alive

Dead

Dead

Alive

Dead

Dead

Alive

Alive

340

CANCER January 15, 2004 / Volume 100 / Number 2

FIGURE 1. Summary of DNA copy number alterations in sinonasal adenocarcinoma (n ⫽ 21). Bars on the left side of each chromosome ideogram denote losses of sequences in the tumor genome, and bars on the right side denote gains. Bold bars indicate high-level increases in copy number (i.e., amplifications). Multiple aberrations were mapped, including gains of chromosomal segments on 5p, 7q, 8q, and 19p and losses on 5q, 8p, 17p, and 18q. High-level amplifications were common.

e.g., loss of 8p, 17p, and 18q and gain of 8q11–22 and 20q15,16—also are observed in adenocarcinomas of the stomach and colon, which have an etiology that differs from that of sinonasal adenocarcinomas. A large number of chromosomal gains and losses were detected in all 21 adenocarcinomas examined. Some loci exhibiting gains are known to harbor genes that code for growth factors and growth factor receptors, such as the hepatocyte growth factor gene (HGF), at 7q21–22, and MOS and MYCC, at 8q. In addition, tumor suppressor genes; such as DCC and SMAD4, at 18q, APC, at 5q, and p53, at 17p13; map to regions of chromosomal loss. Given the important role of these genes in the development of many types of tumors (APC is altered in the earliest stages of colorectal adenocarcinoma, and p53 is one of the most commonly altered tumor suppressor genes in sporadic malignancies), the results of the current study support the use of CGH as a reliable tool for detecting alterations in large, critical chromosomal regions in sinonasal adenocarcinomas.17 Further analysis of genes mapping to such regions is underway in our laboratory, with the goal of determining the specific contributions of these genes in the development of sinonasal adenocarcinomas. High-level amplifications also were detected at various chromosomal locations in the current series of adenocarcinomas. These amplifications occurred almost exclusively in regions that also commonly were involved in low-level gains– e.g., 7q (HGF), 8q (MYCC), 9q (GSM1), 12p (KRAS2, KRAS oncogene—associated gene), 12q (RAB5B), and Xq. This finding suggests that low-level gains may precede the amplification of

genes involved in sinonasal adenocarcinoma tumorigenesis. The limited number of cases analyzed prevented us from uncovering significant differences regarding the distribution of genetic alterations among patients with different stages of disease and clinical characteristics (Table 1). Nonetheless, preliminary results indicate that loss of 4p, loss of 5q, and gain of 20p are associated with disease stage and that loss of 5q may be related to poorer clinical outcome. Because the majority of 5q losses involved the entire arm, it is not possible at present to speculate on which gene might be implicated. The latency period between carcinogen (i.e., wood dust) exposure and the appearance of sinonasal adenocarcinoma may be as long as 40 years. Before an invasive phenotype develops, a number of genetic alterations are believed to accumulate in the exposed mucus, as has been described in other tumors, such as head and neck squamous cell carcinoma and colorectal carcinoma.18,19 The current study contributes to understanding of the molecular genetic basis underlying the development of sinonasal adenocarcinomas. In the near future, this understanding may facilitate the early detection of patients who are at risk of developing sinonasal adenocarcinoma and, consequently, present new possibilities for prevention.

REFERENCES 1.

Hyams VJ. Nonepidermoid epithelial neoplasms of the upper respiratory tract. In: Hyams VJ, Batsakis JG, Michels LM. Tumors of the upper respiratory tract and ear. Washington, DC: Armed Forces Institute of Pathology, 1988:83–111.

Sinonasal Adenocarcinomas and CGH/Ariza et al. 2.

Alessi DM, Trapp TK, Fu YS, Calcaterra TC. Nonsalivary sinonasal adenocarcinoma. Arch Otolaryngol Head Neck Surg. 1998;114:996 –999. 3. Nu ´n ˜ ez F, Sua´rez C, Alvarez I, Losa JL, Barthe P, Fresno M. Sino-nasal adenocarcinoma: epidemiological and clinicopathological study of 34 cases. J Otolaryngol. 1993;22:86 –90. 4. Kleinsasser O, Schroeder HG. Adenocarcinomas of the inner nose after exposure to wood dust. Morphological findings and relationships between histopathology and clinical behavior in 79 cases. Arch Otorhinolaryngol. 1998;245:1–15. 5. Hayes RB, Gerin M, Raatgever JW, de Bruyn A. Wood related occupations, wood dust exposure, and sinonasal cancer. Am J Epidemiol. 1986;124:569 –577. 6. Barnes L. Intestinal-type adenocarcinoma of the nasal cavity and paranasal sinuses. Am J Surg Pathol. 1986;10:192–202. 7. Pe´rez P, Domı´nguez O, Gonza´lez S, Gonza´lez S, Trivin ˜ o A, Sua´rez C. ras gene mutations in ethmoid sinus adenocarcinoma: prognostic implications. Cancer. 1999;86:255–264. 8. Barbacid M. ras genes. Annu Rev Biochem. 1987;56:779 – 827. 9. Kallioniemi A, Kallioniemi OP, Sudar D, et al. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science. 1992;285:818 – 821. 10. Sobin LH, Wittekind C, editors. TNM classification of malignant tumors (6th edition). New York: John Wiley & Sons, 2002. 11. Ried T, Knutzen R, Steinbeck R, et al. Comparative genomic hybridization reveals a specific pattern of chromosomal gains and losses during the genesis of colorectal tumors. Genes Chromosomes Cancer. 1996;15:234 –245.

341

12. Hermsen M, Guervos MA, Meijer G, et al. New chromosomal regions with high-level amplifications in squamous cell carcinomas of the larynx and pharynx, identified by comparative genomic hybridization. J Pathol. 2001;194:177–182. 13. Martins C, Fonseca I, Roque L, Ribeiro C, Soares J. Cytogenetic similarities between two types of salivary gland carcinomas: adenoid cystic carcinoma and polymorphous lowgrade adenocarcinoma. Cancer Genet Cytogenet. 2001;128: 130 –136. 14. Toida M, Balazs M, Mori T, et al. Analysis of genetic alterations in salivary gland tumors by comparative genomic hybridization. Cancer Genet Cytogenet. 2001;127:34 –37. 15. Hermsen M, Postma C, Baak JP, et al. Colorectal adenoma to carcinoma progression follows multiple pathways of chromosomal instability. Gastroenterology. 2002;123:1109 –1119. 16. van Grieken NC, Weiss MM, Meijer GA, et al. Helicobacter pylori-related and -non-related gastric cancers do not differ with respect to chromosomal aberrations. J Pathol. 2000; 192:301–306. 17. Stoler DL, Chen N, Basik M, et al. The onset and extent of genomic instability in sporadic colorectal tumor progression. Proc Natl Acad Sci U S A. 1999;96:15121–15126. 18. Califano J, van der Riet P, Westra W, et al. Genetic progression model for head and neck cancer: implications for field cancerization. Cancer Res. 1996;56:2488 –2492. 19. Vogelstein B, Fearon ER, Hamilton SR. Genetic alteration during colorectal tumor development. N Engl J Med. 1988; 319:525–532.