12951842

Standard and Molecular Cytogenetics of Endocrine Tumors

Marille Herrmann, MD, PhD, FCAP*

Key Words: Pituitary; Thyroid; Parathyroid; Adrenocortical tumors; Pheochromocytoma; Gastrointestinal carcinoid; Pancreatic endocrine tumor; Cytogenetics; Comparative genomic hybridization

DOI: 10.1309/C97C7EY00KROWYVT

Abstract

Standard cytogenetic data are reported for benign and malignant thyroid and pituitary tumors. These identified chromosomal rearrangements activating RET and PPARyl in a subset of papillary (PTCs) and follicular thyroid carcinomas (FTCs) and amplification of PKCE in FTCs. Increases in complex karyotypes accompany progression in thyroid lesions. For pituitary tumors, karyotypic abnormalities more often affect functioning than nonfunctioning tumors. Such differences extend into comparative genomic hybridization (CGH), with a similar distribution for chromosomal imbalances.

CGH data are reported on all varieties of endocrine tumors, supporting some cytogenetic findings and adding new hotspots for genomic imbalances. Increases in genomic imbalance accompany clinical progression in malignant thyroid tumors, adult adrenocortical tumors, parathyroid tumors, and some pancreatic endocrine tumors (PETs), eg, insulinomas. Pheochromocytoma and FTC show more losses than gains. Specific patterns of imbalances are emerging for gastrointestinal PETs regarding location and hormone status; for pheochromocytomas, medullary thyroid carcinomas, and parathyroid tumors regarding genetic syndrome association; and for pituitary tumors regarding hormone status.

As is well known from other organ systems, some chro- mosomal rearrangements correspond to the activation of oncogenes, such as RET, PPARyl (peroxisome proliferator- activated receptor y1), and PKCE (protein kinase C &) in papillary and follicular thyroid carcinomas.1-5 Only a few cases of endocrine gastrointestinal and adrenal tumors have been studied cytogenetically. This lack probably is due to the logistics for the study of such rare entities, and culture condi- tions have been worked out only for the more common thyroid tumor types.6,7 The majority of cytogenetic aberra- tions in pituitary tumors were reported from 3 laboratories and show particularly high detection rates of abnormal clones when direct harvest techniques, eg, culture up to 48 hours, have been applied.8-10 Fluorescence in situ hybridization (FISH) studies and, particularly, comparative genomic hybridization (CGH), which is independent of tissue culture and can be applied to archival tissues, have defined multiple regions of genomic imbalance over the broad spectrum of endocrine tumors. Most of the CGH data await the resolution into molecular detail.

Thyroid Neoplasms

Thyroid neoplasms, including follicular adenomas; hyperplasia; and papillary, follicular, anaplastic, and medullary carcinoma, have been the most extensively studied group of endocrine tumors by standard and molecular cytoge- netic techniques.11-13

Anaplastic Carcinomas

The first cytogenetic study in thyroid tumors of the follic- ular cell type analyzed 5 anaplastic thyroid carcinomas.14

Complex composite karyograms with losses and gains of chromosomes, most commonly +20, -1, -4, and -13; a large number of recurrent marker chromosomes (85 in 5 cases) with frequent isochromosomes [i(1p), i(6p), i(13q), i(14q), i(17q)]; and Robertsonian translocations involving acro- centrics or 1 acrocentric with a nonacrocentric chromosome were described. Two cases showed double minutes. Frequently involved oncogenes could not be attributed to the amplification phenomena.

Some of the relative gains and losses of genetic material detected by karyotypic analysis (-1, -4, -13, +20)1,14 have been confirmed by CGH of tumor tissue15-17 and anaplastic thyroid carcinoma cell lines.18 However, the consistent finding in these studies is complex karyotypic changes involving losses and gains of whole chromosomes, whole chromosome arms, and marker chromosomes. Only a few cases did not show changes by CGH.19 While CGH of tumor tissue (9 cases) and 2 cell lines in 1 report showed primarily gains of chromosomes 8 and 5p,15 a similar change was seen with less frequency in the karyotypic analyses.1,14 When genomic imbalances in cell lines from anaplastic carcinoma as identified by CGH were compared with those in cell lines from differentiated thyroid cancers, the only difference was a loss of material on chromosome 16p.18 Loss of heterozy- gosity (LOH) at 16p was confirmed by microsatellite analysis on material from microdissected paraffin sections in anaplastic carcinoma.20,21 Genome-wide allelotyping of 21 anaplastic thyroid carcinomas confirmed some of the cytoge- netic and CGH findings of loss of chromosomal regions 1q,

4q, 6q, 9p, 13q, 14q, and 16p.20 Congruence for gains of 3q (q27q29), 5p, 7p (p22pter), 8q (q21), and 9q (q34) was reported by three CGH studies (minimal regions of loss in parentheses).15-17 Gain of region 8q21 (MYC locus) was confirmed by FISH in one of the studies.15 Frequent gain of 17q25, which harbors the SIR-T8 gene, was found in CGH studies from tumor material and in cell lines ITable 11.15,17,18,22 The SIR-T8 gene was recently shown to be amplified in thyroid tumor tissues and cell lines.23 Regions with high-level amplification were reported in only one of the studies.17 Less congruence between allelotyping and cytogenetics is seen for other chromosomes such as 2, 11, 17, and 19, for which cytogenetic and CGH analysis identi- fied gains more frequently than losses.15-17 This apparent discrepancy may be due in part to methodological limitations of microsatellite analyses on paraffin-embedded tissues or interpretative problems (LOH vs allelic imbalance), but more likely reflects the complexity of genetic events in these regions and the variability of genetic changes encountered in these tumors.

The complexity and multitude of chromosomal changes in anaplastic thyroid carcinoma suggests accumulation of multiple events over time rather than pathogenesis de novo as was hypothesized in earlier studies.24 Definitive events charac- terizing transformation of differentiated to anaplastic carci- noma have not been defined at the molecular level. The hypoth- esis of a role for a gene on chromosome 16p is interesting and received support by the presence of LOH in 4 of 21 cases allelotyped18,20 and 7 of 7 cases in a recent Japanese study.21

ITable 1 Aberrations With Known Molecular Consequence in Papillary Thyroid Carcinoma (PTC)

Rearrange- ment	Gene(s)	Cytogenetics	Radiation Association	Morphology Association	Reference
PTC1	H4/RET	inv(10)(q11.2q21.2)	Yes	Classic PTC	Grieco et al22
PTC2	PKCR1W/RET	t(10;17)(q11.2;q23)		PTC	Sozzi et al26
PTC3	ELE1/RET	Submicroscopic	Yes	Solid variant childhood PTC	Santoro et al34
PTC3	ELE1/RET	No cytogenetic analysis	"OP post 1980, Tokyo"	Solid variant childhood and adult PTC	Motomura et al51
PTC4	ELE1/RET	Submicroscopic	Yes	PTC	Klugbauer et al35
PTC5	GOLGIN-84 (RFG5)/RET	t(10;14)(q11.2;q?)	Yes	Childhood PTC	Klugbauer et al35
PTC6	HTIF1a (RFG6)/RET	t(7;10)(q33-34;11.2)	Yes	Childhood PTC	Klugbauer and Rabes36
PTC7	ELKS/RET	t(10;12)(q11.2;p13)		PTC	Nakata et al37
PTC7	RFG7|RET	t(1;10)(p13;q11.2)	Yes	Childhood PTC	Klugbauer and Rabes36
PTC8	KINECTIN/RET (RFG8)	t(10;14)(q11.2;q22.1)	Yes	PTC	Salassidis et al38
PTC9	RFG9|RET	t(10;18)(q11.2;q21-22)	Yes	Childhood PTC	Klugbauer et al39
	PCM-1/RET	t(8;10)(p21-22;q11.2)		PTC	Corvi et al40
	No molecular analysis	t(5;10)(p15.3;q11)		PTC	Jenkins et al1
	No molecular analysis	t(1;10)(p13;q11.2)		PTC	Roque et al19
	NTRK1/TPM3	del(1)(q31q41) & add		PTC	Sozzi et al32
	NTRK1/TPM3 and	No cytogenetic analysis		PTC	Musholt et al50
	NTRK1/TPR
	SIR-T8 amplification	No cytogenetic analysis		"PTC, primary tumor and cell lines"	de Nigris et al23
	SIR-T8 amplification	17q25 amplification by		PTC poorly differentiated and	Wreesmann et al17
	suggestive of	comparative genomic hybridization		anaplastic

Papillary Thyroid Carcinomas

Cytogenetic studies on well-differentiated papillary thyroid carcinomas (PTCs) showed normal or predominantly simple karyotypes with few numeric and/or balanced struc- tural aberrations at a cytogenetic aberration frequency of 20% to 40% in more than 130 reported cases.1,2,6,19,24-33 The most prominent finding was a rearrangement of the long arm of chromosome 10, the paracentric inversion, inv(10)(q11.2q21.2).1,2,6,19 The molecular consequence of the chromosomal rearrangement is the activation of the RET oncogene by H4, a ubiquitously expressed gene.3,25 Many other genes have been identified to activate proto-RET and lead to expression of a membrane receptor tyrosine kinase, which is not expressed physiologically in follicular thyroid cells. Some of these different RET rearrangements (termed PTC1 through PTC9 rearrangements) have been identified as or are predicted to result from cytogenetically detectable rearrangements (Table 1): t(10;17) PKCR1%,26 ELE1,34 t(10;14) RFG5,35 t(7;10) HTIFla and t(1;10) RFG7,36 t(10;12) ELKS,37 t(10;14) KTN1,38 t(10;18) RFG9,39 and t(8;10) PCM-1.40

RET rearrangements PTC1 and PTC3 are inversions, which can be generated by irradiation41 and are the 2 most frequently identified ones in PTC cases associated with the Chernobyl nuclear accident.42-44 The largest multicenter study of radiation-induced PTCs, subtypes solid/follicular, classic, and diffuse sclerosing, showed an approximately 35% rate for RET-PTC1 and/or 3 rearrangements.42 Of all these chromosomal rearrangements, only PTC1 and PTC2 have been identified cytogenetically, and they were reported from different geographic locations, such as Portugal, Italy, and the United States.2,6,18,19,27 Table 1 lists known RET rearrangements and their association with radiation, occur- rence in childhood, and respective morphologic variants.

Structural Chromosomal Aberrations Other Than RET

With more than 130 PTCs analyzed to date, none other than those rearrangements involving RET have been confirmed by a second group of investigators.1,19,24,26-33 Despite the diversity of breakpoints in these rearrangements, trends for morphologic associations are evolving.

Two of 3 cases, either poorly differentiated carcinomas or with components of dedifferentiation, showed complex karyotypes and marker chromosomes.19 The single columnar variant of PTC showed several marker chromosomes of unknown origin; a translocation t3;11 with breakpoints at 3p13 and 11q13; gains of chromosomes X, 5, 7, and 9; and loss of chromosome 11.19 Of 6 cases of the tall cell variant of PTC, 3 showed gains of the long arm of chromosome 2 either as +2 or as i(2)(q10).19 The long arm of chromosome 2 also is implicated in follicular thyroid carcinoma (FTC, see “Follic- ular Thyroid Carcinoma”).4 In addition, a susceptibility locus

has been mapped to chromosome 2q21 distal to the PAX8 gene in the familial form of nonmedullary thyroid carcinoma in a Tasmanian pedigree with frequent multinodular PTCs of aggressive behavior.45

Gains of chromosomes 5, 7, and 12 or structural rearrangements involving these chromosomes have been reported for the follicular variant of PTC,19,33 as well as for follicular adenoma46 (see discussion in the “Fluorescence In Situ Hybridization” section on analyses of follicular variants of PTC). Breakpoint 5p15.3 recurs in PTC1,33 and was reported for 1 case of the follicular variant of PTC and 1 case without PTC subtype definition. Several breakpoint clusters have been identified. Next in frequency to the regions of 10q11.2 and 10q21.2 (location of RET and H4 genes) with 14 and 10 hits, 2 regions on the short arm of chromosome 1 have been identified: 1p32 to p36 with 9 hits and 1p11 to p13 with 5 hits.19 Most of the rearrangements involving these breakpoint clusters are suspected to result in deletions, because the region is affected by LOH in many tumors and harbors several genes such as P73 that are considered tumor suppressor genes.

Molecular consequences for these cytogenetic findings have not been reported. A study analyzing 2 microsatellite loci showed LOH at 1p36 in 1 of 5 anaplastic and 2 of 14 FTCs but no LOH in 16 PTCs.47 An interesting early finding was a derivative chromosome 1 with an interstitial deletion on the long arm of chromosome 1 involving band 1q31, where the NTRK1 (neurotrophic tyrosine kinase receptor 1) fusion partner gene TPM had been localized. Molecular analysis confirmed gene fusion resulting in a rearranged NTRK1 oncogene.32 Rearrangements of NTRK1 have been confirmed in several additional PTCs by molecular studies, but no further cytogenetic correlates were identified.48 Like the PTC1 and PTC3 rearrangements of RET, the NTRK1 rearrangement also is hypothesized to result from a chromo- somal inversion or interstitial deletion, produced by double- strand breaks and illegitimate reassociation as a result of radiation.49 Reverse transcriptase-polymerase chain reaction analysis of RET and NTRK rearrangements in 119 PTC cases revealed frequencies of 14.2% and 12.6%, respectively,50 and higher percentages (67%-87%) in radiation-exposed and pediatric patients.51 Increased translocation frequency in radiation-exposed (environmental or therapeutic) thyroid tissues was also shown by other approaches such as FISH painting of metaphase spreads and spectral karyotyping.52,53 Several nonrecurring translocations such as a t(1;10) and t(5;10) involve the chromosomal band 10q11.2; however, molecular correlation studies are lacking.1,19

Besides RET and NTRK1, a third tyrosine kinase has been implicated by FISH analysis of metaphase cells from a PTC with radiation history using a probe for the conserved catalytic domain of tyrosine kinase genes at 2p22-2354 and

identified 2 different inversions of chromosome 2: inv(2)(p23q35) and inv(2)(p22q35). The particular tyrosine kinase involved has not been reported in this case.

Additional breakpoint clusters have been identified on chromosomes 3 and 7. The cluster on chromosome 3p25 has been confirmed by LOH studies. However, the VHL (von Hippel-Lindau) gene, a possible candidate gene for LOH, did not show a mutation.55 The second cluster of breakpoints (5 hits) at 7q34-36 involves balanced translocations, suggesting the activation of a putative oncogene.19,28,31 Many additional breakpoints have been identified in the few cases of PTC with radiation exposure that were analyzed by cytogenetics.53

Chromosomal Gains and Amplification in PTC

A number of gains and losses of whole chromosomes have been reported as additional or single cytogenetic changes. From these numeric aberrations, the most frequent one is trisomy 7. Trisomy and monosomy 7 have been reported in PTC, medullary thyroid carcinoma (MTC), and thyroid adenoma, and their relation to the neoplastic process was questioned.31,56,57 Trisomy 7 has been identified by classic cytogenetics and by FISH analyses on tumor touch preparations from PTCs and on sections from paraffin- embedded tissue from PTCs and MTCs.56-59 These studies provided evidence that trisomy 7 is not a tissue culture arti- fact, and it is not linked to a particular cell lineage. Unlike in familial papillary kidney tumors or pheochromocytomas, in which gain of a copy of chromosome 7 or 10, respectively, was associated with an additional copy of a mutated MET or RET oncogene,60,61 molecular consequences of the trisomy have not been identified so far. Gains of chromosomes X, 5, and 7 have been reported for the follicular variant of PTC, and gain of chromosome 2 or 2q was seen in 3 cases of the tall cell variant of PTC. No molecular consequence has been pinpointed for these gains. Other chromosomal aberrations in this variant include a translocation t3;319 and gain of chro- mosome 15.62 A single case of PTC showed cytogenetic evidence of gene amplification in the form of double minutes, which hybridized to a chromosome 4 library.63

Chromosomal Losses in PTC

From the long list of clonal chromosomal losses, mono- somies 22 and 16 are of particular interest. Monosomy 22 is the second most frequent numeric aberration reported in PTC and its metastases32 and has been confirmed by LOH analysis in a case of the follicular variant of PTC manifesting with skin metastases.33 LOH at 3 microsatellite markers in the vicinity of 22q11 was demonstrated in this study and involved 2 separate foci of multifocal PTC and skin metas- tases. In addition, a single case of the follicular variant of PTC with monosomy 22 harbored additional clones with del22(q11), i(22)del(q11), or a translocation chromosome

involving breakpoint 22q11.64 Further investigation of chro- mosome 22 losses and deletions may reveal genes associated with poor outcome in a small subset of PTC and may lead to the identification of genes associated with the follicular variant of PTC. Monosomy 16 was described in PTC by 2 groups.19,24 One of the 2 cases showed loss of multiple chro- mosomes and a deletion 8q22; in the other case, a composite karyogram of 38 to 42 chromosomes with a single clonal loss of chromosome 16 was identified. The latter case showed ganglionic metastasis. Molecular data were not reported in these cases. In view of the CGH data (see “Comparative Genomic Hybridization”) comparing loss in cell lines from differentiated and undifferentiated thyroid cancers, which showed loss at the short arm of chromosome 16 as the single region with differential loss,18 the 2 PTC cases with monosomy 16 are noteworthy.

Fluorescence In Situ Hybridization

FISH studies have confirmed some of the gains and losses detected by standard cytogenetics. One study analyzed 7 PTCs, 4 of the follicular subtype and 3 of the classic type, as well as 4 benign thyroid lesions using centromere probes for chromosomes 3, 7, 9, 11, 12, 18, and X.65 This study confirmed the trend that numeric aberrations are more frequent in the follicular variant of PTC than the classic-type PTC and that chromosome 12 is involved more frequently in trisomies than chromosomes 3, 7, 11, and 18.

A second FISH study analyzed 10 malignant thyroid tumors (7 PTCs, including 5 of the follicular variant, and 3 FTCs), 30 benign nodules, and 10 normal thyroid tissue samples with centromere probes for chromosomes 7, 10, and 17.66 High frequencies in gains of these chromosomes in benign and malignant tumors (9/10 malignant; 22/30 benign) were reported, whereas loss of chromosome 17 was restricted to 2 of the PTCs.66 A search for molecular or cyto- genetic means to differentiate between benign thyroid nodules and carcinoma combined FISH analysis (chromo- somes 7, 12, 17, and X/Y) with genetic instability assess- ment by microsatellite analysis and inter-(simple sequence repeat) polymerase chain reaction on 12 benign and 10 malignant thyroid lesions.67 Chromosomal losses and gains in this study did not distinguish between benign and malig- nant conditions, nor did microsatellite analysis (using the markers BAT25, BAT26, TGFBRII, D2S123, D3S1029, and D5S346). A statistically significant difference was found using inter-(simple sequence repeat) polymerase chain reac- tion with genomic instability indices between 0 and 6.6 for PTCs and indices between 0 and 1.9 for benign conditions. If these results can be confirmed, molecular tests for genomic instability may yield good diagnostic predictive power on disease outcome in PTCs rather than specific tests for chromo- somal losses. Experience with FISH using spectral karyotyping

or probes designed to detect structural rearrangements of oncogenes involved in PTCs is very limited. One such probe spanning the breakpoint of RET was used on the PTC cell line TPC-1 and showed a 3-way translocation t(1;10;21) together with the paracentric inversion inv(10)(q11.2q21,2).68

Comparative Genomic Hybridization

CGH has been applied to more than 120 cases of PTC, and confirmed results from prior cytogenetic, FISH, and molecular studies partially, but also defined new regions of imbalance.69 PTC is associated with fewer chromosomal imbalances than FTC or poorly differentiated or anaplastic carcinomas (~30% mean, 12%-50% in PTC vs >80% in poorly differentiated or anaplastic carcinoma).5,16,17,62,70,71 With the exception of frequent loss of 22, associated with increased lymph node metastasis in one report, gains at 1q, 5q, 13, and 19 and losses at 9q chromosomal imbalances reported in PTC vary between studies.5,16,17,70,71 The comparison of reported CGH data is difficult for variability in reporting policies and cutoff values for definition of loss, gain, and amplification, and the lack of morphologic correla- tion in some of the studies. Loss of chromosome 17 or 17p in PTCs was reported by 2 CGH analyses,17,70 while 2 different CGH studies found gains of chromosomes 7 and 17 together with gain at 5q and 21q in a single case.16,71 1p34 and 1p36 were found amplified by one CGH report, which correlates with a duplication found by cytogenetics.5,24 Two other CGH reports found losses of 1p34 and 1p36 and loss of a more proximal region of 1p in poorly differentiated PTCs.17,70 Whether the effects of any of these findings correlate to those of translocation t(1;9)(p34;q21) identified in a PTC with poorly differentiated areas is unknown.19 Gain of 1q42 shown by CGH in 5 of 42 PTCs confirmed the cytogenetic data of duplications overlapping this region in 2 PTCs.16,19,30 Gain of 1q as a progression marker has been suggested by a recent CGH study associating 1q42 gain with metastases and is corroborated by its presence in cell lines derived from recurrent FTCs and PTCs.71-73 The CGH finding of amplifi- cation at 2p21 in 4 of 16 PTCs and in 1 FTC cell line led to the identification of amplification of PKCE in the FTC cell line.5 A role of PKCE in PTC tumorigenesis was suggested by findings of marked reductions in protein level of PKCE in 8 of 11 PTCs studied.74

Rarely identified gains of subregions on the long arm of chromosomes 2, 5, and 12 shown by CGH correlate with cyto- genetic data and centromere FISH evaluation.16,19,24,30-33,64,70 Gain at 2q22-24 in 2 PTCs by CGH has a counterpart in the cytogenetic isochromosome 2q resulting in gain of 2q in PTC, tall cell variant.70 However, corresponding PTC subtypes are not reported in that CGH study, and gain of chromosome 15 and loss at chromosome 13 are reported in another CGH study for the PTC tall cell variant.62 Interstudy

variation is striking, with gains or losses reported for chromosomes 4, 6, 9q34, 13, 16q22q24, 17, and 19q13.5,16,17,62,70,71 Gain at 4q12-q27 per 2 CGH studies calls back the cytogenetic amplification phenomenon associ- ated with chromosome 4 in a PTC cell line, potentially providing a marker for aggressive behavior in subsets of PTCs.17,70,72 One of the CGH studies found association of gain at 4q with poorly differentiated or anaplastic carcinoma, whereas data presentation in the other study does not allow clinical correlation of individual cases.17,70 It is impossible to predict whether molecular effects of frequent gains or losses at chromosomes 6q13-22 and 13q12-22 by CGH17,70 corre- late with molecular consequences of translocations reported for regions 6q211 and 13q12.19 Both cases with loss of 6q identified by standard cytogenetics contained either poorly differentiated components or were of higher grade with local spread.19,24 Loss of chromosome 16 as well as deletion of 16q12q13 were identified in rare cases by CGH, as they were by prior cytogenetic analyses27; however, clinical corre- lation was not presented in the CGH cases.5,70 One CGH study identified gain of chromosome 20 in 4 invasive PTC cases, a finding that correlates to 2 cases with cytogenetic analysis, one of them with poorly differentiated components. 19,23,62

Gross chromosomal imbalances as identified by CGH are uncommon and diverse in well-differentiated PTCs. They are likely not associated with initiating events in PTC carcinogenesis and may mark chromosomal instability asso- ciated with cases likely to progress. Future studies may show the association of particular imbalances with clinical outcomes.

Association of PTCs With Radiation History

Most cytogenetic data on PTCs do not specify whether patients had received radiation. The few data available for PTCs with known radiation history (either Chernobyl-associ- ated or therapeutic head and neck irradiation during child- hood) show a large spectrum of breakpoints, including several coinciding with breakpoints occurring in the absence of a history of receiving radiation and many new ones.53 Coinciding breakpoints are 10q11.2 (the location of the RET oncogene) rearranged to 1p13 (the location of the presumed partner gene, RFG7). Other breakpoints reported include 1p21, 1q42, 9q34, 15q22, and 22q12. Whether all these chromosomal rearrangements result in activated oncogenes or dysregulated genes remains to be determined.

Follicular Thyroid Carcinoma

FTCs showed clonal aberrations more frequently and with more complexity of rearrangements than PTCs.1,31,58,75,76 With more than 40 cases analyzed, the translocation t(2;3)(q13,p25) took center stage. Translocation

t(2;3) is so far the single translocation in FTC confirmed by different groups ITable 21.4,77 The molecular consequence of t2;3 has been identified as the oncogenic rearrangement of PAX8/PPARyl fusing the thyroid transcription factor PAX8 to PPARy1.4 PAX8/PPARyl was present in 5 of 8 FTCs in the original study.4 A similar frequency of 53% was found by an reverse transcriptase-polymerase chain reaction analysis,78 which demonstrated fusion in 8 of 15 FTCs and 2 follicular adenomas with trabecular growth pattern (2/25 follicular adenomas). Most of the 8 cases with PAX8/PPARyl fusion were reported as widely invasive. Interestingly not only PAX8/PPARyl fusions were described in trabecular tumors of the thyroid but also frequent RET rearrangements (75%).79 The PAX8/PPARyl rearrangement was not present in 35 papillary and 12 Hürthle cell carcinomas.78 However, 1 follicular variant of PTC was reported with a t(2;3)(q13;p25) as was 1 PTC tall cell variant with a t(3;3) involving 3p25.19 Whether these translocations involve PAX8/PPARyl or PPARyl is not known. PAX8/PPARyl was present in 3 of 3 FTCs with radiation exposure. The breakpoints 2q13 and 3p25 have been implicated in other translocations, some of

them unresolved.1,76 So far, partner genes to PAX8 other than PPARyl for participation in translocations have not been reported.

Other recurring breakpoints in FTCs found outside centromeric regions are 1p13, 3p13, 6q16, 7p15, 8q24, 12q24, and 19q13. The latter breakpoint was reported in 2 cases with either poorly differentiated components or insular features.58,77 Only molecular analysis would allow differenti- ation between the following interpretations: (1) breakpoints 19q13 may involve a different gene from the one affected in benign lesions (see below); (2) or associate the aggressive subtype with the particular partner gene; (3) or alternatively involve genetic changes not detected by cytogenetics. Frequent structural changes involving the short arm of chro- mosome 3 potentially resulting in allelic loss have been studied and confirmed by LOH analysis.24,31 A translocation t(7;8)(p15;q24) was reported in 2 widely invasive FTCs with bone and lung metastases.58 Another case with a single abnormality del11q also was widely invasive.58

Unbalanced rearrangements are predominant in FTC, and the range of genetic loss and amplification is only unfolding.

ITable 21

Aberrations With Known Molecular Consequence in Endocrine Tumors, Non-PTC

Gene(s)	Cytogenetics	Morphology Association	Methods	Reference
Follicular Thyroid Carcinoma
PAX8/PPARỵ1	t(2;3)(q13;p25)	FTC		Kroll et al4
No molecular analysis	t(2;3)(q13;p25)*	FTC invasive, metastatic, insular features		Roque et al77
PAX8/PPARy1	No cytogenetic analysis	FTC widely invasive (5/8)		Nikiforova et al78
PKC& amplification	2p21+	FTC cell line		Chen et al5
Enteropancreatic Endocrine Tumors
HER2/neu	17q11q12 high level+	PET nonfunctional	FISH	Terris et al188
HER2/neu	Amplification	Gastrinoma	Molecular	Evers et al 196
MENIN suggestive of	11q (11q13)*	PET	CGH analysis	Terris et al188
MENIN suggestive of	11q (11q13) *	PET	CGH analysis	Speel et al 192
MENIN	Mutations/no cytogenetic data	Sporadic and MEN1-related gastri- nomas and sporadic insulinomas	LOH/mutation analysis	Debelenko et al197
KREV-1	Rearrangement 1p13	Insulinoma	Blot hybridization	Iwamura et al185
GI Carcinoids
MENIN suggestive of	11q (11q13) *	GI and bronchial carcinoids	CGH analysis	Zhao et al194
	LOH 11q13	Sporadic and MEN1-related GI carcinoids	LOH analysis	D'Adda et al201
Adrenal Cortical Tumors
IGFII overexpression	Duplication 11p15 (paternal allele)/or LOH 11p15 (maternal allele)/UPD	Adrenal cortical carcinoma	Northern blot	Gicquel et al153
Pituitary Tumors
HMGA2 overexpression	12q14q151/marker chromosomes	Prolactinoma	CGH/FISH	Finelli et al223
MENIN	LOH 11q13	MEN1-related tumors	LOH/sequencing	Farrell and Clayton221
Parathyroid Tumors
PTH/cyclin D1(PRAD1)	"inv(11)(p15q13)"	Parathyroid adenoma	Allelotyping	Arnold et al123,126
MENIN, suggestive of	LOH 11q13	Parathyroid adenomas	LOH/allelotyping	Cryns et al131
MENIN	Mutations in 2 FIPH families	Parathyroid hyperplasia	Sequencing	Villablanca et al129
HPRT2	Mutations in HPRT2 families	Parathyroid adenoma	Sequencing	Carpten et al130

CGH, comparative genomic hybridization; FISH, fluorescence in situ hybridization; FTC, follicular thyroid carcinoma; GI, gastrointestinal; LOH, Loss of heterozygosity; PET, pancreatic endocrine tumor; PTC, papillary thyroid carcinoma; UPD, uniparental isodisomy.

* Additional aberrations.

* Amplification by CGH.

# Loss by CGH and LOH studies.

Beyond frequent translocations, FTC is known for loss of whole chromosomes with close to haplotypic karyograms.1,24,31,75 Whole chromosome losses most commonly involve chromosomes 6, 8, and 11 and the acro- centrics 13, 14, 15, 17, 18, and Y. Chromosomes that are the least involved by loss are 5, 7, 12, 16, 19, and X.1,24,31,75 Reported loss of these chromosomes is restricted to cytoge- netic analysis of metastases to lymph nodes or bone.1,58,75 Monosomies 2, 5, and 12 are infrequent but were detected in metastatic disease.58,75,77

FISH Studies

FISH studies using batteries of centromeric and locus- specific probes on 50 FTCs, including those with Hürthle cell differentiation, complement the published cytogenetic data.59,66,67,80,81 Polysomies of chromosomes 7 and 12 were not identified in normal thyroid tissue but were present in hyperplastic or adenomatous tissue.80 FISH analysis proved more sensitive in detection of polysomies than standard cyto- genetic analysis and identified large polysomic clones consti- tuting up to approximately 50% of tumor cells missed by cytogenetic analysis.80 Comparison of 2 FISH methods, ie, application of probes to isolated nuclei and hybridization to deparaffinized 4-um sections showed no significant differ- ence in detection of polysomies.80 Polysomies for chromo- somes 7 and 12 are detected frequently in benign, and even more frequently in malignant, follicular lesions.66,80

Increased frequency of aneusomy, especially losses of chromosomes in Hürthle cell carcinomas vs benign lesions, was reported in additional FISH analyses.59,81 In Hürthle cell lesions, monosomy of chromosome 22 was more common in malignant cases than adenomas, and monosomy 2 was restricted to malignant cases.59,81 Loss of chromosomes 8 and X were frequent in benign and malignant follicular lesions.81 Gains of chromosomes 7 and 12 were seen commonly in follicular thyroid lesions and did not help in distinguishing benign from malignant lesions.67,80 Chromo- some 17 gains were reported for benign and malignant onco- cytic thyroid lesions, while monosomy 17 was restricted to malignant oncocytic lesions in one study, but not confirmed by another.59,81 FISH using a p53 probe demonstrated loss of one p53 signal in 2 aggressive Hürthle cell carcinoma cases with 2 centromeric chromosome 17 signals.81 These 2 cases were negative by immunohistochemical p53 analysis, while 2 additional aggressive Hürthle cell carcinomas with immunohistochemical p53 accumulation did not show p53 loss by in situ hybridization.81 This study confirms the notion of a role for p53 in aggressive thyroid carcinomas and advocates for detection by the 2 complementary techniques, immunohistochemical analysis and FISH. Probes for cyclin D1, HER2/neu, N-myc, and c-myc did not show amplifica- tion in the series of Hürthle cell carcinomas studied.59,81

CGH Analysis

CGH analysis should be particularly powerful in lesions with large chromosomal imbalances such as follicular thyroid lesions and was applied to more than 40 malignant cases.16,82-84 The common gain of chromosomes 5, 7, 12, 17, and 19 in FTCs as demonstrated by classic cytogenetics and FISH analyses was found in several cases analyzed with CGH.16,82-84 The number of chromosomal gains in recurrent vs nonrecurrent carcinomas showed a significant association with tumor recurrence in a series of Hürthle cell carcinomas (mean, 9 vs 1.3; P = . 005).84 Losses of chromosome 22 were the most frequent change by CGH and were seen in mini- mally and widely invasive cases.16,82 Amplification at 1q with a minimally overlapping region at 1q42 was found in at least 6 widely invasive cases.5,16,82 Losses at 1p with a minimal region of overlap at 1p21 were reported in 5 widely invasive carcinomas.16,82 Losses of chromosome 13 were seen less frequently.16 Loss involving chromosome 2q was detected by 2 CGH studies83,84 including Hürthle cell carci- nomas that confirmed previous cytogenetic24,77 and FISH reports.59 Additional loci of amplification were identified on the short arm of chromosome 2 (2p13 and 2p21) in an FTC cell line. The latter was cloned using a refined CGH method and revealed a rearrangement of PKCE.5

In addition to the often complex chromosomal imbal- ances, several cases showed normal CGH data, and in 1 of the cases found to be normal by CGH, a mutation of codon 13 of the K-ras gene was identified.16,83

Benign Thyroid Lesions

Benign thyroid lesions have been studied extensively, with 500 cytogenetically analyzed cases (follicular adenomas, including oxyphilic type, 217; common goiter, 251; adenomatous goiter, 25; Graves disease, 5; and dyshormonogenetic goiter, 2).29,31,46,53,85-98 FISH data were published for approximately 50 cases and CGH data for 61 cases of follicular adenoma, including oncocytic and atypical adenoma.5,65,66,81-84 Clonal chromosomal changes were seen in follicular adenomas in 33% to 66% of cases; adenomatous goiters showed a similar frequency of clonal changes and a usual hyperplasia a rate of less than 10%.

Gains and Losses of Genomic Areas

Chromosomal gain was the most commonly reported clonal aberration with few additional chromosomes in hyper- plasia and 3 or more chromosomal gains in adenomas and adenomatous nodules. Frequent gains involved chromo- somes 5, 7, 9, and 12, similar to findings for the follicular variant of PTC and for FTC.46,80,95,96 These gains were confirmed in FISH and CGH studies and include findings in oxyphilic adenomas.65,66,80,81-84 Chromosomal losses have been identified by standard cytogenetic, FISH, and CGH

analysis but at much lower frequency than in FTC, and chro- mosomes X, 8, 16, and 22 were involved most frequently.5,31,59,82,86,96 Chromosomal losses were more frequent in atypical adenomas and oxyphilic adenomas.83,84 Of 6 atypical adenomas, 3 showed loss of chromosome 22, and loss at chromosome 2 or 2q was shown by CGH for Hürthle cell adenomas,82-84 which is in contrast with the findings of a study using FISH analysis in which monosomy 2 was restricted to FTC.81 In regard to loss of chromosome 22 in oxyphilic neoplasms, 2 FISH studies showed an increase of monosomy 22 in oxyphilic carcinomas over oxyphilic adenomas,59,81 similar to the CGH results for usual follicular neoplasms.82

Recurring Structural Aberrations and Breakpoints

The most common recurring structural aberration in adenomas and nodular hyperplasia involved the band 19q13 with chromosome 5q13 as a frequent translocation partner, resulting in t(5;19)(q13;q13).31,88,92,93,96,99 Alternative rearrangements of 19q13, besides 5q13, were shown to involve 1p35/6, 1q31, 2p12/13, 5p14/15, 10q11.2, 10q22, 10q24, 11p15, 16q11, 17p11.2, and Xq22.29,31,85,92,93,96 A molecular consequence of these rearrangements has not been identified; however, breakpoint 19q13.4 cloning has narrowed the region to less than 150 kb.100 An initially proposed candidate gene tentatively named RITA (rearranged in thyroid adenoma) (ZNF331 by HUGO) was shown to be close but adjacent to the 150-kb interval.101 A second recurring translocation was identified potentially involving the same breakpoint in 4 cases: t(2;3)(q12 or q13;p24 or p25).31,91,94 Whether these translocations generate a PAX8/PPARyl fusion is not known. Rearrange- ments and deletions affecting chromosome 2 have been reported by several groups,31,81,91,96,98,102 with breakpoints scattered from 2p21 to 2q33. Within that region are 3 clus- ters of breakpoints: 2p21, 2p13, and 2q13. PAX8, a thyroid transcription factor, is located at 2q13; the other 2 break- points on the short arm of chromosome 2 match amplified regions, as shown by CGH analysis of thyroid adenomas.5 The amplification region 2p21 involves PKCE in an FTC cell line (see “CGH Analysis” in FTC section). A second group is cloning the breakpoint 2p21 from cell lines derived from benign thyroid lesions, and the region has been narrowed to 450 kb.103 The third breakpoint cluster, 2p13, has not been cloned.

Beyond the breakpoint clusters on chromosomes 2, 5, and 19, only a few regions have been identified by more than one group or more than twice. (Since part of the literature does not clearly state which cases have been reported previ- ously, it is difficult to draw conclusions from breakpoints reported twice.) These regions involve chromosome 1p36 and 17p13.93,96 These 2 regions are well known for their

tumor suppressor loci, for example p73 and TP53. Other recurring breakpoints found less frequently are 1q11, 2p12, 2q31, 3p24, 11q13, 12q13, 13q33, and 16q11.29,31,53,91,95,96 Microsatellite analysis with markers at chromosomes 3p, 7q, 10q, 11q, 13q, and 17p identified bands 7q31.1 and 11q23 as frequently involved by LOH.104 Only the latter was involved in translocations t2;11 and t11;14, potentially resulting in a submicroscopic deletion.86,96 CGH showed loss of chromo- some 11 or a deletion of the long arm affecting 11q23 in 2 of 11 oxyphilic adenomas.83

Other recurring aberrations affect chromosomes 13, 8, and 22. Deletion of 13q was described as a single cytoge- netic abnormality in a follicular adenoma.88 CGH analysis identified an overlapping region of loss in an oxyphilic adenoma83 and loss of the whole chromosome 13 in an atyp- ical follicular adenoma.82 Another loss more frequently iden- tified in atypical and oxyphilic adenomas was of chromo- some 8,59 as identified by FISH analyses. In addition, there is 1 case report of a translocation t8;14 in an oxyphilic adenoma, which potentially could result in submicroscopic deletion or inactivation of partial 8q.87 10q11.2, the locus of the RET oncogene, represents a recurring breakpoint in adenomas, but despite the identical cytogenetic designation, it may not affect the RET oncogene, as was demonstrated for one adenoma with inv(10)(q11.2q21).97

Cytogenetic data on benign thyroid lesions developing in patients with known therapeutic radiation exposure in childhood are scarce in the literature. The 5 reported cases53 show a large variety of breakpoints, including many not reported previously in sporadic cases, but sharing a few breakpoints (1q11, 2p21, 3p24, 12q13, 13p13, 18p11, 19q13, and Xq22) and a previously reported deletion (2p13). Future studies will show whether there is a distinct pattern of rearrangements and whether the same genes are involved as in the sporadic cases.

Medullary Thyroid Carcinoma

Specific mutations within the RET oncogene without gene rearrangements are the genetic signature feature in MTCs and the multiple endocrine neoplasias (MENs). 105-108 Additional cytogenetic changes and chromosomal regions affected by loss and gain as detected by allelotyping and LOH studies of smaller scale have included LOH at 1p, gain of 1q, and losses at 3q, 11p, 17p, and 22.109-113 The few cytogenetic studies on tissues from primary and metastatic tumors and MTC cell lines showed a recurring aberration, isochromosome 1q or alternatively interpreted as t(1;1)(p11;q11) and/or loss at 1p, an abnormality also frequently identified by CGH.16,110,111,113,114 Genetic loss at the region 3q12q27 as postulated by LOH studies109 was confirmed by CGH findings16,114 of genetic loss at 3q26q29 and 3q23qter and is supported by the cytogenetic finding of

breakpoints potentially involving losses at 3q27 and 3q12.113

A deletion of chromosome 5 and a region on the distal long arm of chromosome 5 (5q32q33) were found in an MTC by cytogenetic analysis,113 although this might not have been a clonal abnormality, since this report mentioned single cell abnormalities. A translocation breakpoint 5q3?3 also was identified in an MTC cell line.111 CGH analysis identified loss of a large region on chromosome 5 more proximal to the previously implicated region (5p12q22).114

Cytogenetic analysis, allelotyping, and CGH identified loss of a whole copy of chromosome 22,16,109,110,115 whereas another CGH study identified cases with gain of chromo- some 22.114

Gain of chromosome 19 or 19q or 19p was the most frequent chromosomal change (7/24 MTCs) detected by CGH.114 Gain of 19 together with several other chromo- somes also was seen in 1 MTC by cytogenetic analysis.113 Interestingly, 2 of the known ligands to RET are located on 19p, and for several of the cases with gain of chromosome 19, overexpression of neurturin, one of the ligands, was demonstrated.114,116

Loss of chromosome 13 or portions of the long arm of chromosome 13 were the second most common changes in both CGH studies (2/10 and 5/24), in some cases as the sole abnormality,16,114 with a region of minimal overlap at 13q21, which is distal to the RB1 gene. In contrast, gain at 11q, reported in both CGH studies (1/10 and 3/24), was present only with several other chromosomal imbalances, and in the one study including clinical follow-up was associated with patients who died of disease.16,114 The region of minimal overlap was 11q12. Alteration at 11q also was found in an MTC cell line in the form of an insertion at 11q13.111 Aberra- tions of the short arm of chromosome 11 were reported from cytogenetic analyses and include deletion of 11p, added mate- rial at 11p, breakpoint 11p13, and telomeric associations of 11p with other chromosomes.111-113,117 Interestingly, LOH was established for a sequential region of overlap at 11pterp13.109 These more or less well-defined regions of genetic imbalances involve no more than half the cases studied. Two MEN2A- associated cases, 1 MTC and 1 C-cell hyperplasia, showed balanced translocations t(9;12) and t(10;16), respectively, a dicentric chromosome dic(2;10) resulting in loss of 2p, losses of 6p and 14 in the former, and evidence of gains by double minutes in the latter.118 None of these changes have been confirmed by cytogenetic or CGH analysis. In the few cases with gross imbalances as identified by CGH, the range varied from 1 region to 10 regions involved. The extent of genetic imbalance corresponded to clinical outcome.114 Comparing sporadic and familial cases with or without the RET mutation M918T, gross genetic imbalances were far more frequent in cases with the mutation.114

Two cytogenetic case reports, one case with a history of therapeutic radiation exposure, reported exclusively nonre- curring aberrations.31,53 These reports may indicate the exis- tence of another genetic pathway to MTC or merely reflect our sketchy knowledge of the genetic events leading to MTC initiation and progression.

MTC is an example in which the hallmark genetic feature consists of a specific RET mutation that is inherited or occurs early in the disease; however, the mutation may make the individual susceptible to rapid gain of additional genetic changes. Particularly, the RET mutation M918T present in MEN2B patients in the germ line is associated with far more genetic imbalances and a more aggressive clin- ical course. The importance of deregulated RET expression is emphasized by increased chromosomal copy number and associated expression of a RET protein ligand. Together with RET protein mutations, gain of chromosome 19 belongs to the early events, seen as a single chromosomal abnormality.

Conclusions of Thyroid Cytogenetics

Benign follicular lesions seem cytogenetically more similar to FTC than PTC. The follicular variant of PTC showed similar cytogenetic changes to follicular adenoma, in contrast with the classic variant. PTCs harbor predominantly cytogenetically balanced aberrations. Signature translocations or inversions have been identified for a subset of PTCs acti- vating the RET oncogene and for a subset of FTCs activating PPARy1. Follicular lesions, both benign and malignant, frequently show cytogenetic copy number imbalances, which can be detected by LOH analysis as allelic imbalances. Progression from clinically benign or indolent to more aggressive thyroid lesions is paralleled by increased frequency of abnormal karyograms and more complex karyo- grams. For MTCs, CGH data point toward the MEN2B-type mutation of RET as a genetic disposition for susceptibility to genetic imbalances. Chromosomal hot spots for a potential role as progression marker have been identified. However, even anaplastic carcinomas may not show genetic imbalances detectable by CGH analysis. CGH analysis may not be predictive of clinical outcomes in individual cases. These genome-wide approaches rather are pinpointing areas for further study. Allelic loss at chromosome 3p turned out to be only one of many loci affected in FTCs; however, the sugges- tion that FTCs involve pathogenetic pathways not effective in PTCs is still valid.24 Ideally, the relative low resolution of standard cytogenetics on solid tumors with complex kary- otypes would indicate complementary FISH analysis (spectral karyotyping) for identification of translocated genetic mate- rial in these complex karyotypes.119 Despite the lack of such combined techniques in most studies, many of the early conclusions are still valid and have been confirmed and better defined by molecular analyses.

Parathyroid Neoplasms

Few classic cytogenetic analyses have been reported for parathyroid tumors.120,121 One study on a sporadic parathy- roid adenoma showed a balanced translocation as a single abnormality and identified 1p22 and 5q32 as breakpoints with unknown molecular consequence. One study reported deletion of 11q distal to q13 in 4 metaphases, loss of the complete chromosome 11 in 2 cells, and loss of chromosome 21 in a single cell from a MEN1-related hyperplastic parathyroid gland.122

Genes implicated in parathyroid neoplasia are the MENIN and the cyclin D1/PRAD1 genes (Table 2). The latter is rearranged to the parathyroid hormone gene (PTH), resulting in overexpression of cyclin D1 in a subset of parathyroid adenomas (approximately 20%). This rearrange- ment supposedly resulted from a pericentric inversion, inv(11)(p15q13), which brings cyclin D1 under the regula- tory control of the PTH gene.123 The corresponding cytoge- netic aberration, however, has never been identified.

Clonal loss of chromosome 11 had been identified in parathyroid adenomas by molecular means a decade before the cytogenetic report.122,124 The finding of chromosomal loss in addition to a gene rearrangement started the search for a suppressor gene on chromosome 11. With the cloning of the MENIN gene, it became apparent that mutations within the gene occur in familial parathyroid hyperplasia and in sporadic parathyroid adenomas, with a frequency in the latter of approximately 20%.125 Beyond the MEN1 syndrome, parathyroid hyperplasia and single and multiple parathyroid adenomas are associated with MEN2A, familial isolated hyperparathyroidism (FIHP or HPRT1) and hyper- parathyroidism-jaw tumor syndrome (HPT-JT or HPRT2), respectively. Somatic RET mutations have not been identi- fied in sporadic or familial parathyroid proliferations.126 MENIN mutations have been identified in a few FIHP fami- lies,127-129 disclosing them as MEN1 variant. Another few FIHP families have shown linkage to 1q25q31, the region of the HPRT2 locus, suggesting that the disease in these fami- lies may represent a variant of the syndrome. In the mean- time, the HRPT2 or HPT-JT locus has been cloned, and the gene product was named parafibromin.130 Somatic inacti- vating mutations of HPRT2 were identified in parathyroid adenomas with cystic features, suggesting a tumor suppressor function. The near future will show to what extent HRPT2 mutations contribute to pathogenesis of sporadic parathyroid adenomas.

Molecular studies on parathyroid adenomas initially identified LOH at loci on chromosomes 11q and 1p.131 Allelotyping with extension of markers to a larger array of chromosomes identified the following regions as hot spots for LOH: 6q22q23 and 6q26q27, 11p, 15q11q21, and

15q26qter, in addition to 11q13 and 1p.132 These findings overlap well with the more recent CGH data.

The 204 parathyroid tumors reported include the following: sporadic carcinoma, 38; familial carcinoma, 1; familial hyperplastic glands, 17; sporadic hyperplastic glands, 16; tertiary hyperplastic glands, 3; radiation-associated adenoma, 10; and sporadic adenoma, 119.133-140 The most frequent CGH changes consisted of loss of chromosome 11 or 11q, 1p, 6 or 6q, 13, 15, or 18 and gain of 16 or 16p. These findings are concordant in most studies; only 1 study impli- cated these chromosomes less frequently and identified a large variety of other chromosomal regions and subregions, which are less frequently implicated by the majority of studies.139 The familial cases (MEN1) showed losses of 11, 6q, and 18, together with few additional losses.134,136,138,140 The 2 studies analyzing parathyroid carcinomas found simi- larities such as more frequent loss of 1p and less frequent loss of 11q in carcinomas than in adenomas.135,137 The larger study (10 cases of carcinoma)136 defined the critical region as 1p21p22, which is centromeric to the more commonly deleted region, 1p34pter, in adenomas. The more proximal location of loss at 1p in parathyroid carcinomas was repro- duced recently by LOH analysis,141 as was the observation that tumors with 1p losses showed less frequent 11q allelic losses.142 Another significant difference (P = . 04) in one study130 and shared by both CGH studies is the gain at 1q31q32,135,137 which has been implicated rarely in adenomas. 1q31 also is the locus for the HRPT2 gene, which predisposes to parathyroid adenoma or carcinoma in combi- nation with ossifying jaw fibromas and renal hamartoma.130

Another significant difference in one study was gain of Xcq13.137 Two studies135,137 showed relatively more gains, including high-level amplification in one,135 in carcinomas vs adenomas. One study revealed a larger spectrum of chro- mosomal regions involved in carcinomas vs adenomas.137 Several of the commonly affected regions in parathyroid adenomas, such as loss of 6q, 13q, 15q, and 18q and gain of 16 and 16p, are also seen in carcinomas.

For some regions, the findings are discordant for loss or gain of chromosomes 4, 7, 17, 19, and 22.133-140 This discor- dance is difficult to interpret with such limited case numbers and may be due in part to methodological problems, 143 patient cohorts, or nonuniformity of genetic pathways leading to parathyroid carcinomas. Support for the latter interpretation comes from 1 study including parathyroid adenomas with associated irradiation, which showed multiple CGH changes in the irradiated group that were seen less frequently in other adenomas. 136

Despite several shared CGH changes and the increased numbers of changes in carcinomas (0-29) over adenomas and over hyperplastic glands,137,138 a progression model is not suggested. The observation that parathyroid adenomas

that show loss at 1p often belong to the group of tumors without 11q losses supports the suggestion that a variety of pathogenetic pathways may lead to parathyroid tumors, and 1 pathway rarely used may result in the malignant tumors.

Adrenal Cortical Tumors

Chromosomal aberrations have been reported in 7 of 14 sporadic adenomas (Conn syndrome) and in 4 of 4 sporadic adrenocortical carcinomas.144-150 The number and complexity of aberrations are increased in carcinomas over adenomas and involve almost all chromosomes. Within the few cases analyzed by standard cytogenetics, a wide spectrum of aber- rations was apparent. The range of cytogenetic changes included carcinomas with only gains and others with only losses of chromosomes. None of the structural aberrations, losses, or gains of chromosomes were recurrent. However, breakpoints on chromosome 11p (p13 and p15144,147) are significant and might result in identical consequences because they affect an imprinted region, which encompasses loci for the Ha-RAS oncogene, insulin, calcitonin, and the insulin-like growth factor-II gene. Adrenocortical tumors frequently arise in Beckwith-Wiedemann syndrome, which was mapped to 11p15 (OMIM [Online Mendelian Inheritance in Man]). Henry et al151 reported the first 2 cases of Beckwith-Wiede- mann syndrome with dup11p15 and adrenocortical carci- noma. Molecular studies show that 11p15.5 rearrangements in adrenocortical carcinomas and adenomas result in reduc- tion to homozygosity and are frequently caused by uniparental isodisomy (paternal UPD) for a region 11p15.5 extending to 11p13 in some tumors (Table 2).151-153

With the advent of CGH, data on 136 adrenocortical tumors (74 adenomas, including 6 pediatric; 62 carcinomas, including 17 pediatric) became available.134,154-160 The cutoff values used in these studies for reporting gains or losses and amplifications varied. However, the results showed similar trends. The obvious question of whether clinical parameters such as size would correlate with frequency or patterns of genetic imbalances received a positive answer. In adults, the frequency of chromosomal aberrations was higher in carci- nomas than in adenomas, and some studies did not find CGH changes in very small tumors, ie, less than 5 cm154 and less than 3.5 cm.156 However, size cutoffs were questioned by other studies reporting CGH changes in a few small tumors of approximately 2 cm.155,157 The findings of all studies were in agreement about a significant difference in the number of changes per tumor in adenoma (mean, 1.1 changes) vs carci- noma (mean, 7.6 changes)154-157 (means according to Sidhu et al157). Changes seen more commonly in the adenoma group were loss of components of 1p and gains of chromosomes 4, 5, 9q, 16, 17, and 19.

All chromosomes may show imbalances in carcinomas with the following emerging trends: Similar to pituitary tumors, loss at 1p, or components thereof, and gain of 1q constitutes the pattern for chromosome 1 and also is present in an adrenocortical cell line.156 Gains were seen predomi- nantly on 5, 6p, 7, 8, 10, 12, 14, 16, 17, 20, and X. Losses and gains were frequent on chromosomes 2, with 2q more frequently involved in losses, as are 3, 4, 9, 11, 13, 15, 18, 19, and 22. It is interesting that the limited data on the few adrenocortical adenomas studied by classic cytogenetic methods had shown derivative chromosome 2 and implicated losses at chromosome 2.146,147

On chromosome 4, gain of 4p and loss of 4q were prevalent. On chromosome 9, 9p was lost and 9q34 was gained. One study evaluating background genetic changes in various tumors of 1 MEN1 case by LOH analysis reported loss of 9p and 17pterq12 in the adrenal tumor.134 Loss of 9p was common to all tumors of the MEN1 case (parathyroid, liver, ovary, stomach, lymph node), and loss of 17 was specific for the adrenal tumor.134 These findings are in agree- ment with those of a previous LOH study on MEN1-associ- ated adrenocortical carcinomas, which showed LOH at 11p, 11q, 13q, and 17p.161 Sporadic adrenocortical carcinomas harbor mostly gains,155,156 but also loss of 17 or 17p.157 Numerous regions with amplification or minimal overlap of gains have been identified; some of these overlap regions amplified in cell lines originating from adrenocortical tumors.155,156 Amplification of region 12q13 has been reported by 2 groups.155,156 Coamplification of oncogenes in this region (MDM2, CDK2) has been reported recently for 2 adrenal cortical tumors.162 Other regions of high-level ampli- fication (each identified by a single group) include 1p34.3qter, 1q22q25, 3p24pter, 3q29, 5p15, 5q12q13, 5q32qter, 7p11.2p14, 12q24.1qter, 13q34, 14q11.2q12, 14q31q32, 16p, 17q24q25, 19p13.3 and q13.4, and 22q11.2q12.155,156

The gain or amplification of 9q34 is interesting for its particular prominence in pediatric adrenocortical tumors, as reported by 2 groups.158,159 Pediatric adrenocortical tumors differed from adult tumors in their spectrum and frequency of chromosomal imbalances including adenomas (mean, 14.5 imbalances; range, 7-22 imbalances).159 In pediatric tumors, there was no difference in the average number of changes between adenomas and carcinomas. Despite the fact that 9 of 11 patients in 1 study showed germline p53 muta- tions159 (unknown p53 status in the other study158), the reported CGH changes were remarkably similar. Beyond gain or amplification of 9q34, both studies showed losses at 2q, 3, 4, 11p and 11q14q22, 13, 18, and X and gain or amplification at 5q31qter, 6p21, 11q12q13, 12q23qter, 16p, 19, and 20q. Amplification or gain of 1p32p36 was reported in 1 study (10/11 cases)159 and not in the other.158 Because

the region 1p22pter is problematic in CGH analysis, this difference may be due to methods and reporting policies rather than a true difference between the groups.143

Several of the CGH results have been confirmed by using selective probes for a subset of cases (probes for centromeres 17 and 1 and for 11q13155; probes for centromere 16 and for 9q34 and 22q12.2156). The presence or absence of gain or loss of chromosome 17 varied between studies. This apparent discordance is addressed in the litera- ture and was evaluated by using FISH with centromere- specific probes.155,163 Taken together, it seems that chromo- some 17, or components of it, shows relatively rare imbalances in carcinomas but may be gained or lost with frequent gain in some cohorts of adenomas. 155,163

Another FISH study evaluating adrenal adenomas with probes for chromosomes 3, 7, 8, 11, and 12 (all chromo- somes with infrequent CGH changes in adrenal adenomas) showed a frequency and distribution of imbalances with a statistically significant difference between aldosteronomas and Cushing syndrome-producing adenomas, with frequent gains of all chromosomes tested in the former and rare gains in the latter.164 Given the low frequency of chromosomal imbalances detected by CGH in adrenal adenomas, current CGH data on adenomas are insufficient to permit good conclusions on early steps in tumorigenesis or association with functional subtypes.

Genetic syndromes associated with adrenocortical tumors include Beckwith-Wiedemann syndrome (see “Adrenal Cortical Tumors”), Li-Fraumeni syndrome (germline p53 mutations, see earlier discussion in this section about pediatric tumor cohorts), MEN1, and Carney complex (CNC1 and CNC2). A genotyping study screening all chromosome arms for LOH in 60 patients with adreno- cortical lesions included molecular analysis of the MENIN gene and the CNC2 locus.165 The study corroborated the CGH findings of increasing chromosomal imbalance in carcinomas and detected frequent LOH for chromosomes 2 and 11q. LOH at 11q13 was present in adrenocortical carci- nomas and adenomas and confirmed previous findings of a study with aldosteronomas.166,167 LOH of the microsatellite marker PYGM close to the MENIN gene was seen in 8 of 8 informative carcinomas; however, screening for mutations within the 9 exons of MENIN by single-strand conformation polymorphism analysis was negative. Possibilities of different mechanisms of gene inactivation or involvement of another nearby gene were suggested. Likewise, the minimal region of LOH on chromosome 2 was a 1-centimorgan region at 2p16, just outside the CNC2 locus.168

Pheochromocytomas

No classic cytogenetic data on pheochromocytomas are available, to my knowledge.169 One study pioneering the

application of FISH to paraffin sections with centromere probes for chromosomes 1, 7, 15, and Y on 23 pheochromo- cytomas found rare gains of signals for chromosomes 1, 7, and Y and rare losses for 15 and Y.170 Three recent CGH studies analyzed 88 pheochromocytomas, including 36 cases associated with VHL syndrome, 5 cases with MEN2A, and 1 case with neurofibromatosis.171-173 One of these studies included an additional 11 abdominal paragangliomas.172 These studies broaden the spectrum of genetic alterations, and essentially confirm findings detected by LOH analysis of chromosomes 1, 3, 11, and 22 in sporadic and hereditary pheochromocytomas.174-180

In contrast with adrenocortical neoplasms, pheochromo- cytomas show more losses than gains by CGH and show changes through the entire genome. Losses present in sporadic, VHL, familial, and extra-adrenal tumors target predominantly chromosomes 1p, 3p, 3q, 6q, 11p, 11q, and 13q. However, losses of chromosome 3 or 3p, 11p, and 11 q are significantly (P < . 001) more frequent in VHL-related than in sporadic or MEN2A-related cases.171-173 Losses at 1p11p31 are significantly (P < . 001) more frequent in sporadic and MEN2A-associated cases than in VHL-related cases. In addition, loss of 9p is seen in all types of pheochro- mocytomas but more commonly in VHL-related tumors. The comparison of 11 extra-adrenal paragangliomas with non-VHL-related pheochromocytomas showed similar genetic losses; however, gain of 11q or 11cenq13 was seen more frequently (36% vs 4%) in the paraganglioma group.172 Gains of 9q, including the band q34, 17q, and/or loss of 17p, were present in subsets of benign and malignant sporadic pheochromocytomas.171,172 Comparison of benign (35 cases) with malignant (22 cases; metastasizing or widely invasive tumors) tumors in both non-VHL cohorts showed no promi- nent change detected by CGH. Slight tendencies of malig- nant cases seen in 1 study, such as more frequent loss of 6q, 17p, or 18 or rare gains of 5, 7, or 12,171 were not confirmed by the other.172 No amplification was detected in any of the studies.

The high frequency of loss at 1p in sporadic cases (>80%) suggests a role for a suppressor gene (or genes) in the early stages of a proposed progression model.171,172 Defining loci for losses on the distal part of 1p is problematic owing to difficulties of CGH in that region.143 Additional evaluation by FISH in selected cases or by LOH studies confirmed losses, including 1p36,171 and suggested the involvement of 3 different suppressor genes at 1p36, 1p32, and 1pcenp13.180 In contrast with the findings of previous LOH studies, CGH analyses report few cases with loss of chromosome 22. Whether this is related to methodological problems of CGH at chromosome 22 or to differently skewed case cohorts in respect to inclusion of neurofibro- matosis-related cases is unclear.

Most sporadic cases show a stepwise loss starting with 1p and 3q, followed by either loss at 3p, 6q, 11p, 11q, or 17p or gain at 9q. Only a few sporadic cases show no CGH changes or chromosomal losses without losses at 1p, 3q, or 3p. Published data support a model that defines functional loss of genes at 1p and 3q (microscopic or submicroscopic) as early events in the tumorigenesis of sporadic pheochromocy- tomas. Data are less conclusive for determining the time frame for the additional events and for defining progression markers. Whether this lack of separation of benign from malignant tumors by genetic parameters can be overcome by larger series of CGH studies, the future may tell. However, the lack of morphologic parameters to predict malignant behavior extends into definition of cohorts for the comparison of “benign” and malignant pheochromocytomas, thereby masking potential genetic parameters for progression.

Endocrine Tumors of the Gastrointestinal Tract and Pancreas

Classic cytogenetic data on 13 endocrine enteropancre- atic181,182 tumors are available. Cytogenetic analysis on 2 sporadic gastrinomas showed near tetraploidy, 2 marker chromosomes, one of them potentially derived from 5p and loss of chromosome 3 in a subclone in 1 case and gain of chromosome 3 in the other.183 Two studies on insulinomas, one MEN1-related and the other sporadic, reported cytoge- netic aberrations.122,184 The sporadic case showed a translo- cation, t(1;9)(p13;p22). The band 1p13 seemed interesting for its location of N-RAS and KREV-1 genes, the latter reportedly rearranged in a benign insulinoma (Table 2).185 The MEN1-associated insulinoma showed a complex karyo- type with deletions of 3q, 6q21q22, 10p, 16q, 17q, Xq27, and Y; gain of unknown material at 1p; and structural aberra- tions inv12q and a ring chromosome of unknown origin.122 One study on 9 pancreatic endocrine tumors (PETs), 2 of them MEN1-associated, showed numeric and/or few struc- tural aberrations in 5 tumors.186 More gains than losses of chromosomes with recurring gains of 5, 7, and 17 were reported. Gains may involve chromosomes 1, 3, 8, 9, 10, 18, 19, and 20 and markers and losses of 1, 2, 3, 4, 6, 11, 14, X, and Y. Structural aberrations include markers and additional material at 1p12.186 Whether the structural aberrations in 3 cases with the breakpoints 1p13, 1p12, and 1p? led to similar molecular effects is not known.122,184,186 Losses of sex chro- mosomes in 2 of 5 malignant PETs with cytogenetic aberra- tions is interesting in relation to a FISH study confirmed by LOH analysis of 40 PETs.187 That study identified signifi- cant correlation of sex chromosome loss with clinical behavior, ie, metastasis and local invasion in sporadic and familial PETs showing a frequency of 40% and 36% for

losses of X in females and Y in males, respectively (P = .001). While the association of hormone secretion was a stronger factor for association with metastasis and invasion than sex chromosome anomalies, the combination of both parameters would increase sensitivity in the detection of aggressive cases in this series.

Eight CGH studies on 214 cases evaluated genetic imbalance in sporadic and familial endocrine tumors of the pancreas and gastrointestinal (GI) tract.188-195 One study of 7 gastrinomas (familial, 1; sporadic, 6) confirmed results of a genome-wide survey by microsatellite analysis using approx- imately 400 markers with CGH analysis.190 The authors reported loss of 1p, gain of subregions on 1q, gain of 5, and loss of 11 or 11q13 detected by both methods. Four aggres- sive tumors showed an increase in allelic imbalance in several chromosomes. The majority of cases in all 8 studies showed genomic imbalances with more gains than losses. Only 8 of 44 PETs in one study and none of 12 PETs in another showed no imbalance.189

However, only a few tumors and genomic regions have been implicated by high-level amplification. These are 19q13 in 1 sporadic duodenal gastrinoma without evidence of metastasis and 17q11 in 1 nonfunctioning PET with lymph node metastasis,188 which corresponded to c-ERB-B2 ampli- fication. Amplification of c-ERB-B2 has been implicated in gastrinomas, but frequencies observed vary.189,196 Chromo- somes frequently gained were 5, 7, 14, and 17; those lost were the long arms of 6 and 11 or subregions 11q11q13 and 11q11q21 and chromosome 16.188,189

The genomic distribution of changes varied between endocrine tumors of the GI tract in regard to location, status of hormone production, specificity of hormone production, tumor size, and metastatic potential.189 These variations, which may provide therapeutic or diagnostic aids, hamper comparison between studies, which often include a wide spectrum of tumors. Nonfunctioning sporadic PETs show more CGH changes than functioning PETs (mean, 21.3 vs 7; P = . 001).189 A similar level of significance was reached for the differentiation of benign from malignant insulinomas by comparing sheer numbers of CGH changes. Also, individual chromosomal areas showed significance levels in the range .03 to .005 for association with metastasis (-6q, -3p, +14q, +Xq), with size of more than 2 cm (-3q, -6q, -10q, -11q, +12q, +14q, +17q, +Xq), and with nonfunctioning status (-3p, -8q, -10q, -11p, -11q, +5p, +5q, +7p, +12q, +14q, +17q, +Xq). Loss of sex chromosomes did not reach statis- tical significance for the aforementioned parameters. Chro- mosomal imbalances similar in both functioning and nonfunctioning PETs were losses at 1, 2, 3, 6, 8, 11, and 21 and gains at 5, 7, 9q, 14, 17, and 20. Molecular studies narrowed some of the regions of losses such as 11q13, 3p (3p21p26 and 3p23p25.3), and 6q (6q22.1 and 6q23q24) and

confirmed the association of frequencies of loss at 3p and 6q with metastatic disease.197-200

Specific subtypes of functioning PETs were associated with specific regions of imbalances.189 Malignant insuli- nomas showed loss of 6q in 6 of 6 tumors, gain of Xp and loss of Y in males, and loss of Xq in females, whereas gain of 5q was seen in 3 of 11 benign insulinomas. Gain of 9q and loss of 3p was present in 3 of 7 gastrinomas; loss of 11q, Xq, and Y in 3 of 7 VIPomas; and gain of 7q in 4 of 4 glucagonomas.189 The authors confirmed results for chromo- some 3 by FISH and microsatellite analysis.

Another study included 2 duodenal gastrinomas and 6 carcinoids (ileal and gastric) in CGH analysis of 12 PETs, 2 of them functioning.188 All cases showed imbalances with more gains than losses. The 1 calcitoninoma, 1 pancreatic gastrinoma, and 4 nonfunctioning PETs showed no losses. In the remaining nonfunctioning PETs, losses were more frequent for 16p, 1p, 11q11q21, and 17p. Gains in the PET group were similar to the study described earlier,189 with the exception of additional gains of chromosome 4. Three recent studies with less diverse case cohorts confirmed earlier find- ings such as association of increased numbers of regions affected by imbalances with metastasis in PETs.191-193 These studies refined specific associations of imbalances with clin- ical behavior and defined minimal regions of imbalances. Gains of 4 and 7 and loss of 21q were frequent changes in metastases from PETs; and gains of 5, 12q, 14q, 17q, 18q, and 20q and losses of 2, 3, 6q, 10p, and 11p were found in metastases and in their primary tumors.193 Specific associa- tion of gain of 4p and loss of 6q was seen with nonfunc- tioning tumors and gain of 9q34 and loss of 1p, 1q, 4q, 11q, Xq, and Y with insulinomas in a study comparing 24 small insulinomas with 10 nonfunctioning PETs.192 If confirmed by other groups, the finding of an association of loss of 3 and 6q and gain of 17 and 20q with malignant behavior of small PETs could provide clinical applications.192 Gains of 7 (7q11.2), 19, 14, and 20q (20q11.1q13.1) and losses of 11q21q22 and 11p13p15, listed in descending frequency with minimal regions of gain or loss in parentheses, were reported from another cohort of 25 PETs, 20 of them func- tioning (12 insulinomas).191 The 2 duodenal gastrinomas, one of them metastatic, showed only a few overlaps with PETs in their spectrum of gains and losses, which were loss of 17p and gains of 7, 17, 20, and 22.174

CGH data on 45 GI carcinoids have been reported.188,194,195 Ileal carcinoids showed gains similar to those of the pancreatic tumors, but loss of chromosome 9 or 9p was more frequent in the carcinoids.188 Similarities included losses at 11q and gains of chromosomes 5, 7, and 14. The single gastric carcinoid showed overlaps for losses (11, 18, and 1p) but differed in respect to gains, and it was nonmetastatic in contrast with the ileal group.188

Two studies comparing GI and midgut carcinoids with those from the bronchopulmonary system showed discor- dance in several aspects, suggesting different pathogenetic mechanisms for these tumors.194,195 Carcinoids in the GI tract showed few overlaps (losses at 11q) in genetic imbal- ances with those from the bronchopulmonary system. Whereas GI carcinoids show losses of chromosome 18 or subregions (minimal overlap 18q22qter) frequently, occur- ring in primary and metastatic tumors and as single CGH change, suggesting loss of 18q as an early event in GI carci- noids, loss of 18q was not seen in bronchial carcinoids.194 One study proposed a progression model based on increased numbers and acquisition of specific changes in metastases by CGH analysis of 18 cases.195 Losses of 18q (q22qter) and 11q (q22a23) were seen in primary GI carcinoids and metas- tases, whereas loss of 16q (q21qter) and gain at 4p14qter were seen exclusively or at a greater frequency in the metas- tases.195 As with genetic differences between carcinoids in different organ systems, studies reporting LOH at 11q identi- fied variable frequencies and specific location of loss at 11q in separately developed GI regions (Table 2).201 With the exceptions of MENIN mutations in a subset of PETs and carcinoids, no mutations or rearrangements of suppressor genes or oncogenes have been reported.202,203

Pituitary Neoplasms

Chromosomal aberrations were reported in 63 (37.7%) of 167 cases analyzed by classic cytogenetic tech- niques.8-10,204-210 The first cytogenetic report of pituitary adenomas,211 without available subtyping by immunohisto- chemical analysis, dates to the prebanding era.204 Near diploid and near triploid karyotypes with frequent gains of group C and F chromosomes could be discerned. Gain of chromosomes in pituitary adenomas has been confirmed by later detailed cytogenetic studies from European and US laboratories, and the gain affects functioning adenomas more often than it affects nonfunctioning adenomas.8,9,206,210 Trisomies 5, 8, and 12 in any combination were seen frequently in prolactinomas; the cytogenetic findings were confirmed by FISH analysis.8 Predominant chromosomal gain was reported for chromosomes 3, 5, 7, 8, 9, 12, 20, and X; also reported were losses for chromosomes 1 and 10 and either gain or loss for chromosomes 2, 4, 6, 11, 13, 14, 15, 16, 17, 18, 19, 21, 22, and Y.8-10,205,206,209,212

Of the few structural aberrations published, none has been confirmed by a second group of investigators. In addi- tion, a few patients in the analyzed cases received preopera- tive radiation treatment, which may have induced some of the nonrecurring structural rearrangements such as inversions (“Thyroid Neoplasms” for discussion). Recurring structural

aberrations in prolactinomas are isochromosome 1q, translo- cation t(1;3)(p13;p12), and add(12)(p13).8 The breakpoint 1q32 was involved in a duplication of partial 1q and in a translocation in prolactinomas.206,207 These findings suggest the possibility of a minimal region of overlap for amplifica- tion in prolactinomas spanning 1q32 to 1q42.8,206 By using FISH with probes for centromeres of chromosomes 1 and 9, additional signals for chromosomes 1 and 9 were detected in 1 locally invasive prolactinoma, as were loss of signal for chromosome 1 in 1 nonfunctioning adenoma and gain of 9 in 1 growth hormone (GH)-immunopositive adenoma.212 More recent CGH data confirm the tendency for gains of 1q; however, amplification of 1q or subregions including q32 to q42 was detected in pituitary adenomas positive for prolactin, GH, thyroid-stimulating hormone, or follicle-stim- ulating hormone and in immunonegative adenomas.213-216

CGH analysis, which cannot detect balanced transloca- tions, was used to detect associations of particular gains and losses for subgroups of pituitary adenomas with specific clinical or immunophenotypic features.213-219 CGH analysis has been reported on 215 cases with 111 cases (51.6%) showing chromosomal imbalances. The increased detection rate of CGH over cytogenetic analysis for chromosomal imbalances likely relates to tissue culture conditions that were not optimized for tumor cells of pituitary adenomas. This interpretation is supported by the finding of predomi- nantly cytogenetically normal clones in pituitary adenomas after short-term culture (9-10 days) compared with direct culture (24-48 hours).8,210 CGH data confirm the trend for increased frequency of chromosomal imbalances in func- tioning (average, 70%) over nonfunctioning pituitary adenomas (average, 43%).213,215-219

The compilation of CGH data reported to date shows a distribution of chromosomal gains and losses that matches the cytogenetic findings. The frequency of chromosomal aberrations as detected by CGH varies from 1 to more than 20 per case. Chromosomal gain was seen for chromosomes 3, 4, 5, 7, 8, 9, 12, 14, 16, 17, 19, 20, 22, and X, with highest frequencies for chromosomes 5, 7, 9, 12, 19, and X. Predom- inantly, losses affected chromosomes 2, 10, 11, 13, and 18.

Chromosome 1 showed losses for the short arm with a minimally overlapping region 1p32p33 and gains for the long arm most commonly affecting the whole long arm. Loss at chromosome 13 showed a region of minimal overlap at 13q21. This finding is interesting in the light of molecular studies demonstrating LOH in the region of the RB gene (at 13q14) without detecting mutations of the RB gene in pitu- itary adenomas. The presence of a second tumor suppressor gene and the association with its loss in invasive adenomas has been postulated.220 Neither cytogenetic nor CGH data relate to such an association; in contrast, these studies report loss of 13 for noninvasive and invasive adenomas with the

caveat that invasiveness is not uniformly defined in the afore- mentioned cited studies.8,213,218

The presence of pituitary adenomas with GH and prolactin production in patients with MEN1 stimulated LOH analyses with polymorphic markers in the region of the MENIN gene 11q13. LOH for the region was demon- strated for familial and for a subgroup of sporadic cases; however, mutations within MENIN were not reported for sporadic pituitary adenomas.221,222 CGH data showed loss for a region of minimal overlap 11q21qter (distal to the MEN1 locus) affecting the major subgroups of pituitary adenomas (nonfunctioning, prolactinoma, and GH- producing tumors). Another group of pituitary adenomas showed losses of 11p. Somatomammotropic pituitary adenomas may arise within the Carney complex (spotty skin pigmentation, myxomas, and endocrine overactivity). CGH analysis of 4 cases showed losses at chromosome 11, 6q, and 7q and gains at several chromosomes, but no loss for the region of the CNC gene at 2p.217 Losses affecting chro- mosomes 2, 10, and 18 affected the whole chromosome most of the time. Chromosome 4, which more often was gained than lost, showed a curious distribution of imbal- ances. Gains were seen predominantly in nonfunctioning adenomas, and the 4 cases with loss of chromosome 4 or a component of it were adenomas producing GH or adreno- corticotropic hormone.

The oncogenic effects of chromosomal gains have been explored only recently. Gain of chromosome copies harboring mutated genes was reported.60,61 A similar route was taken by Finelli et al,223 who studied gain of chromo- some 12. These authors suggested that amplification of HMGA2 (high mobility group A2), mapped to 12q14q15, may have a crucial role in the early steps of prolactinoma tumorigenesis.223 Transgenic mice overexpressing wild-type HMGA2 developed prolactin-secreting pituitary tumors.223 Bacterial artificial chromosomes containing HMGA2 showed amplification or hybridized to marker chromosomes, many of them derived from chromosome 12.223 The future will show whether the study of minimal overlaps of imbal- anced regions provides keys to understanding the tumorigen- esis of pituitary tumors.

From the Department of Gynecologic and Breast Pathology, Armed Forces Institute of Pathology, Washington, DC.

Supported in part by the American Registry of Pathology, Washington, DC.

Address correspondence to Dr Herrmann: mherrm@earthlink.net.

* Under Title 17 of the US Code, Section 105, copyright protection is not available for any work of the United States Government. The opinions or assertions herein are those of the author and do not necessarily reflect the view of the Department of the Army or of the Department of Defense.

This review uses cytogenetic nomenclature of the 1995 ISCN (Mitelman F. ISCN: An International System for Human Cytogenetic Nomenclature. Basel, Switzerland: S Karger; 1995.) for designating regions amplified or deleted per comparative genomic hybridization (CGH) analysis. For example, a region shown by CGH as amplified from chromosomal band 1p13 to 1p36 would be cited as amp1p13p36. Band designations are cited as given in the literature irrespective of their existence in the ISCN 1995. Additional information on genes and chromosomal loci affected in endocrine neoplasms mentioned in this review can be obtained at gdb.org.

References

1. Jenkins RB, Hay ID, Herath JF, et al. Frequent occurrences of cytogenetic abnormalities in sporadic nonmedullary thyroid carcinoma. Cancer. 1990;66:1213-1220.

2. Herrmann ME, Mohamed A, Talpos G, et al. Cytogenetic study of a papillary thyroid carcinoma with a rearranged chromosome 10. Cancer Genet Cytogenet. 1991;57:209-217.

3. Pierotti MA, Santoro M, Jenkins RB, et al. Characterization of an inversion on the long arm of chromosome 10 juxtaposing D10S170 and RET and creating the oncogenic sequence RET/PTC. Proc Natl Acad Sci U S A. 1992;89:1616-1620.

4. Kroll T, Sarraf P, Pecciarini L, et al. PAX8-PPARgamma1 fusion oncogene in human thyroid carcinoma. Science. 2000;289:1357-1360.

5. Chen X, Knauf J, Gonsky R, et al. From amplification to gene in thyroid cancer: a high-resolution mapped bacterial- artificial-chromosome resource for cancer chromosome aberrations guides gene discovery after comparative genome hybridization. Am J Hum Genet. 1998;63:625-637.

6. Herrmann ME, Talpos GB, Mohamed AN, et al. Genetic markers in thyroid tumors. Surgery. 1991;110:941-948.

7. Herrmann ME, Gipson C, Ansari M, et al. Cytogenetic analysis of thyroid tumors after cryopreservation. Cancer Genet Cytogenet. 1995;85:20-25.

8. Finelli P, Giardino D, Rizzi N, et al. Non-random trisomies of chromosomes 5, 8 and 12 in the prolactinoma sub-type of pituitary adenomas: conventional cytogenetics and interphase FISH study. Int J Cancer. 2000;86:344-350.

9. Bello M, deCampos J, Kusak M, et al. Chromosomal abnormalities in pituitary adenomas. Cancer Genet Cytogenet. 2001;124:76-79.

10. Larsen J, Schroder H, Sorensen A, et al. Simple numerical chromosome aberrations characterize pituitary adenomas. Cancer Genet Cytogenet. 1999;114:144-149.

11. Baloch Z, LiVolsi V. Pathology of the thyroid gland. In: LiVolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:61-101.

12. LiVolsi V. Surgical Pathology of the Thyroid. Vol 22. Philadelphia, PA: Saunders; 1990.

13. Rosai J, Carcangiu ML, Delellis R. Tumors of the Thyroid Gland. Washington, DC: Armed Forces Institute of Pathology, 1992. Atlas of Tumor Pathology, Series 3, Fascicle 5.

14. Mark J, Ekedahl C, Dahlenfors R, et al. Cytogenetical observations in five human anaplastic thyroid carcinomas. Hereditas. 1987;107:163-174.

15. Wilkens L, Benten D, Tchinda J, et al. Aberrations of chromosomes 5 and 8 as recurrent cytogenetic events in anaplastic carcinoma of the thyroid as detected by fluorescence in situ hybridisation and comparative genomic hybridisation. Virchows Arch. 2000;436:312-318.

16. Hemmer S, Wasenius V, Knuutila S, et al. DNA copy number changes in thyroid carcinoma. Am J Pathol. 1999;154:1539- 1547.

17. Wreesmann V, Ghossein R, Patel S, et al. Genome-wide appraisal of thyroid cancer progression. Am J Pathol. 2002;16:1549-1556.

18. Komoike Y, Tamaki Y, Sakita I, et al. Comparative genomic hybridization defines frequent loss on 16p in human anaplastic thyroid carcinoma. Int J Oncol. 1999;14:1157-1162.

19. Roque L, Nunes V, Ribeiro C, et al. Karyotypic characteri- zation of papillary thyroid carcinomas. Cancer. 2001;92:2529- 2538.

20. Kitamura Y, Shimizu K, Tanaka S, et al. Allelotyping of anaplastic thyroid carcinoma: frequent allelic losses on 1q, 9p, 11, 17, 19p, and 22q. Genes Chromosomes Cancer. 2000;27:244-251.

21. Kadota M, Tamaki Y, Sekimoto M, et al. Loss of hetero- zygosity on chromosome 16p and 18q in anaplastic thyroid carcinoma. Oncol Rep. 2003;10:35-38.

22. Grieco M, Santoro M, Berlingieri MT, et al. PTC is a novel rearranged form of the ret proto-oncogene and is frequently detected in vivo in human thyroid papillary carcinomas. Cell. 1990;60:557-563.

23. de Nigris F, Cerutti J, Morelli C, et al. Isolation of a SIR-like gene, SIR-T8, that is overexpressed in thyroid carcinomas cell lines and tissues. Br J Cancer. 2002;86:917-923.

24. Herrmann M, Hay ID, Bartelt DH Jr, et al. Cytogenetic and molecular genetic studies of follicular and papillary thyroid cancers. J Clin Invest. 1991;88:1596-1604.

25. Herrmann M, Talpos G, Wolman S, et al. Genetic markers in thyroid tumors [letter]. Surgery. 1993;113:596-598.

26. Sozzi G, Bongarzone I, Miozzo M, et al. A t(10;17) translocation creates the RET/PTC2 chimeric transforming sequence in papillary thyroid carcinoma. Genes Chromosomes Cancer. 1994;9:244-250.

27. Roque L, Clode A, Gomez P, et al. Cytogenetic findings in papillary thyroid carcinomas. Genes Chromosomes Cancer. 1995;13:157-162.

28. Antonini P, Venuat A, Linares G, et al. A translocation (7;10)(q35;q21) in a differentiated papillary thyroid carcinoma. Cancer Genet Cytogenet. 1989;41:139-144.

29. Bondeson L, Bengtsson A, Bondeson A-G, et al. Chromo- some studies in thyroid neoplasia. Cancer. 1989;64:680-685.

30. Olah E, Balogh E, Bojan F, et al. Cytogenetic analyses of three papillary carcinomas and a follicular adenoma of the thyroid. Cancer Genet Cytogenet. 1990;44:119-129.

31. Teyssier J-R, Liautaud-Roger F, Ferre D, et al. Chromosomal changes in thyroid tumors. Cancer Genet Cytogenet. 1990;50:249-263.

32. Sozzi G, Bongarzone I, Miozzo M, et al. Cytogenetic and molecular genetic characterization of papillary thyroid carcinomas. Genes Chromosomes Cancer. 1992;5:212-218.

33. Smit JW, Van Zelderen-Bhola S, Merx R, et al. A novel chromosomal translocation t(3;5)(q12;p15.3) and loss of heterozygosity on chromosome 22 in a multifocal follicular variant of papillary carcinoma presenting with skin metastases. Clin Endocrinol (Oxf). 2001;55:543-548.

34. Santoro M, Dathan NA, Berlingieri MT, et al. Molecular characterization of RET/PTC3: a novel rearranged version of the RET proto-oncogene in a human thyroid papillary carcinoma. Oncogene. 1994;9:509-516.

35. Klugbauer S, Demidchik E, Lengfelder E, et al. Detection of a novel type of ret rearrangement (PTC5) in thyroid carcinomas after Chernobyl and analysis of the involved ret fused gene (RFG5). Cancer Res. 1998;58:198-203.

36. Klugbauer S, Rabes H. The transcription coactivator HTIF1 and a related protein are fused to the RET receptor tyrosine kinase in childhood papillary thyroid carcinomas. Oncogene. 1999;18:4388-4393.

37. Nakata T, Kitamura Y, Shimizu K, et al. Fusion of a novel gene, ELKS, to RET due to translocation t(10;12)(q11;p13) in a papillary thyroid carcinoma. Genes Chromosomes Cancer. 1999;25:97-103.

38. Salassidis K, Bruch J, Zitzelsberger H, et al. Translocation t(10;14)(q11.2;q22.1) fusing the kinetin to the ret gene creates a novel rearranged form (PTC8) of the ret proto- oncogene in radiation induced childhood papillary thyroid carcinoma. Cancer Res. 2000;60:2786-2789.

39. Klugbauer S, Jauch A, Lengfelder E, et al. A novel type of ret rearrangement (PTC8) in childhood papillary thyroid carcinomas and characterization of the involved gene (RFG8). Cancer Res. 2000;60:7028-7032.

40. Corvi R, Berger N, Balczon R, et al. RET/PCM-1: a novel fusion gene in papillary thyroid carcinomas. Oncogene. 2000;19:4236-4242.

41. Ito T, Seyama T, Iwamoto KS, et al. In vitro radiation is able to cause RET oncogene rearrangements. Cancer Res. 1993;53:2940-2943.

42. Santoro M, Thomas G, Vecchio G, et al. Gene rearrangement and Chernobyl related thyroid cancers. Br J Cancer. 2000;82:315-322.

43. Nikiforov Y, Koshoffer A, Nikiforova M, et al. Chromosomal breakpoint positions suggest a direct role for radiation in inducing illegitimate recombination between the ELE1 and RET genes in radiation-induced thyroid carcinomas. Oncogene. 1999;18:6330-6334.

44. Fugazzola L, Pilotti S, Pinchera A, et al. Oncogenic rearrangements of the RET proto-oncogene in papillary thyroid carcinomas from children exposed to the Chernobyl nuclear accident. Cancer Res. 1995;55:5617-5620.

45. McKay J, Lesueur F, Jonard L, et al. Localization of a suscepti- bility gene for familial nonmedullary thyroid carcinoma to chromosome 2q21. Am J Hum Genet. 2001;69:440-446.

46. van den Berg E, Oosterhuis J, de Jong B, et al. Cytogenetics of thyroid follicular adenomas. Cancer Genet Cytogenet. 1990;44:217-222.

47. Kleer C, Bryant B, Giordano T, et al. Genetic changes in chromosomes 1p and 17p in thyroid cancer progression. Endocr Pathol. 2000;11:137-143.

48. Bongarzone I, Vigneri P, Mariani L, et al. RET/NTRK1 rearrangements in thyroid gland tumors of the papillary carcinoma family: correlation with clinicopathological features. Clin Cancer Res. 1998;4:223-228.

49. Nikiforov YE, Rowland JM, Bove KE, et al. Distinct pattern of ret oncogene rearrangements in morphological variants of radiation-induced and sporadic papillary thyroid carcinomas in children. Cancer Res. 1997;57:1690-1694.

50. Musholt T, Musholt P, Khaladj N, et al. Prognostic significance of RET and NTRK1 rearrangements in sporadic papillary carcinoma. Surgery. 2000;128:984-993.

51. Motomura T, Nikiforov Y, Namba H, et al. RET rearrangements in Japanese pediatric and adult papillary thyroid cancers. Thyroid. 1998;8:485-489.

52. Lehmann L, Zitzelsberger H, Kellerer A, et al. Chromosome translocations in thyroid tissues from Belarussian children exposed to radioiodine from the Chernobyl accident, measured by FISH painting. Int J Radiat Biol. 1996;70:513- 516.

53. Zitzelsberger H, Lehmann L, Hieber L, et al. Cytogenetic changes in radiation-induced tumors of the thyroid. Cancer Res. 1999;59:135-140.

54. Lehmann L, Greulich K, Zitzelsberger H, et al. Cytogenetic and molecular genetic characterization of a chromosome 2 rearrangement in a case of human papillary thyroid carcinoma with radiation history. Cancer Genet Cytogenet. 1997;96:30-36.

55. Grebe SK, McIver B, Hay ID, et al. Frequent loss of heterozygosity on chromosomes 3p and 17p without VHL or p53 mutations suggests involvement of unidentified tumor suppressor genes in follicular thyroid carcinoma. J Clin Endocrinol Metab. 1997;82:3684-3691.

56. Herrmann M, Lalley P. Significance of trisomy 7 in thyroid tumors. Cancer Genet Cytogenet. 1992;62:144-149.

57. Cortezzi SS, Cesar AC, Tajara EH, et al. Does trisomy 7 occur in a particular cell type in the thyroid [letter]? Cancer Genet Cytogenet. 2002;132:77-78.

58. Roque L, Clode A, Belge G, et al. Follicular thyroid carcinoma: chromosome analysis of 19 cases. Genes Chromosomes Cancer. 1998;21:250-255.

59. Mazzucchelli L, Burckhardt E, Hirsiger H, et al. Interphase cytogenetics in oncocytic adenomas and carcinomas of the thyroid gland. Hum Pathol. 2000;31:854-859.

60. Zhuang Z, Park W, Pack S, et al. Trisomy 7-harbouring non- random duplication of the mutant MET allele in hereditary papillary renal carcinomas. Nat Genet. 1998;20:66-69.

61. Koch C, Huang S, Moley J, et al. Allelic imbalance of the mutant and wild-type RET allele in MEN2A-associated medullary thyroid carcinoma. Oncogene. 2001;20:7809- 7811.

62. Bauer A, Cavalli L, Rone J, et al. Evaluation of adult papillary thyroid carcinomas by comparative genomic hybridization and microsatellite instability analysis. Cancer Genet Cytogenet. 2002;135:182-186.

63. Herrmann M, Lalley P, Rydstedt L, et al. Double minutes in the cell line PTC-1113A. Cancer Genet Cytogenet. 1996;90:70-74.

64. Perissel B, Coupier I, DeLatour M, et al. Structural and numerical aberrations of chromosome 22 in a case of follicular variant of papillary thyroid carcinoma revealed by conventional and molecular cytogenetics. Cancer Genet Cytogenet. 2000;121:33-37.

65. Taruscio D, Carcangiu M, Ried T, et al. Numerical chromosomal aberrations in thyroid tumors detected by fluorescence in situ hybridization. Genes Chromosomes Cancer. 1994;9:180-185.

66. Barril N, Carvalho-Sales AB, Tajara EH. Detection of numerical chromosome anomalies in interphase cells of benign and malignant thyroid lesions using fluorescence in situ hybridization. Cancer Genet Cytogenet. 2000;117:50-56.

67. Stoler D, Datta R, Charles M, et al. Genomic instability measurement in the diagnosis of thyroid neoplasms. Head Neck. 2002;24:290-295.

68. Jossart G, Greulich K, Siperstein A, et al. Molecular and cytogenetic characterization of a t(1;10;21) translocation in the human papillary thyroid cancer cell line TPC-1 expressing the RET/H4 chimeric transcript. Surgery. 1995;118:1018-1023.

69. Kubo K, Yoshimoto K, Yokogoshi Y, et al. Loss of heterozygosity on chromosome 1p in thyroid adenoma and medullary carcinoma, but not in papillary carcinoma. Jpn J Cancer Res. 1991;82:1097-1103.

70. Singh B, Lim D, Cigudosa J, et al. Screening for genetic aberrations in papillary thyroid cancer by using genomic hybridization. Surgery. 2000;128:888-894.

71. Kjellman P, Lagercrantz S, Hoog A, et al. Gain of 1q and loss of 9q21.3-q32 are associated with a less favorable prognosis in papillary thyroid carcinoma. Genes Chromosomes Cancer. 2001;32:43-49.

72. Herrmann M, Talpos G, Abumrad N, et al. Cytogenetic and molecular characterization of the thyroid cancer cell line PTC-1113A [abstract]. Am J Hum Genet. 1995;57:A302. Abstract 1750.

73. Geldorf A, Versteegh R, vanMourik J, et al. Clonally related but phenotypically divergent human cancer cell lines derived from a single follicular thyroid cancer recurrence (TT2609). Thyroid. 2001;11:909-917.

74. Knauf J, Ward L, Nikiforov Y, et al. Isozyme-specific abnormalities of PKC in thyroid cancer: evidence for post- transcriptional changes in PKC epsilon. J Clin Endocrinol Metab. 2002;87:2150-2159.

75. van den Berg E, van Doormaal J, Oosterhuis J, et al. Chromosomal aberrations in follicular thyroid carcinoma. Cancer Genet Cytogenet. 1991;54:215-222.

76. Lui W-O, Kytola S, Anfalk L, et al. Balanced translocation (3;7)(p25;q25): another mechanism of tumorigenesis in follicular thyroid carcinoma. Cancer Genet Cytogenet. 2000;119:109-112.

77. Roque L, Castedo S, Clode A, et al. Deletion of 3p25→pter in a primary follicular thyroid carcinoma and its metastasis. Genes Chromosomes Cancer. 1993;8:199-203.

78. Nikiforova MN, Biddinger PW, Caudill CM, et al. PAX8- PPARy rearrangement in thyroid tumors: RT-PCR and immunohistochemical analyses. Am J Surg Pathol. 2002;26:1016-1023.

79. Cheung C, Boerner S, MacMillan C, et al. Hyalinizing trabecular tumor of the thyroid: a variant of papillary carcinoma proved by molecular genetics. Am J Surg Pathol. 2000;24:1622-1626.

80. Roque L, Serpa A, Clode A, et al. Significance of trisomy 7 and 12 in thyroid lesions with follicular differentiation: a cytogenetic and in situ hybridization study. Lab Invest. 1999;79:369-378.

81. Erickson L, Jalal S, Goellner J, et al. Analysis of Hürthle cell neoplasms of the thyroid by interphase fluorescence in situ hybridization. Am J Surg Pathol. 2001;25:911-917.

82. Hemmer S, Wasenius V-M, Knuutila S, et al. Comparison of benign and malignant follicular thyroid tumours by comparative genomic hybridization. Br J Cancer. 1998;78:1012-1017.

83. Tallini G, Hsueh A, Liu S, et al. Frequent chromosomal DNA unbalance in thyroid oncocytic (Hürthle cell) neoplasms detected by comparative genomic hybridization. Lab Invest. 1999;79:547-555.

84. Wada N, Duh Q, Miura D, et al. Chromosomal aberrations by comparative genomic hybridization in Hürthle cell thyroid carcinomas are associated with tumor recurrence. J Clin Endocrinol Metab. 2002;87:4595-4601.

85. Bartnitzke S, Herrmann M, Lobeck H, et al. Cytogenetic findings on 8 follicular thyroid adenomas including one with a t(10;19). Cancer Genet Cytogenet. 1989;38:65-68.

86. Antonini P, Venuat A, Linares G, et al. Cytogenetic abnormalities in thyroid adenomas. Cancer Genet Cytogenet. 1991;52:157-164.

87. Herrmann M, Hay I, Bartelt DJ, et al. Cytogenetics of six follicular thyroid adenomas including a case report of an oxyphil variant with t(8;14)(q13;q24.1). Cancer Genet Cytogenet. 1991;56:231-235.

88. Belge G, Thode B, Bullerdiek J, et al. Deletion of part of the long arm of chromosome 13 as the only karyotypic aberration in a follicular thyroid adenoma. Cancer Genet Cytogenet. 1991;56:277-280.

89. Gama N, Gama R, Thome J, et al. Cytogenetic analysis of a multinodular goiter. Cancer Genet Cytogenet. 1991;55:73-77.

90. Roque L, Castedo S, Clode A, et al. Translocation t(5;19): a recurrent change in thyroid follicular adenoma. Genes Chromosomes Cancer. 1992;4:346-347.

91. Sozzi G, Miozzo M, Cariani TC, et al. A t(2;3)(q12-13;p24- 25) in follicular thyroid adenomas. Cancer Genet Cytogenet. 1992;64:38-41.

92. Dal Cin P, Sneyers W, Aly M, et al. Involvement of 19q13 in follicular thyroid adenoma. Cancer Genet Cytogenet. 1992;60:99-101.

93. Belge G, Thode B, Bullerdiek J, et al. Aberrations of chromosome 19: do they characterize a subtype of benign thyroid adenomas? Cancer Genet Cytogenet. 1992;60:23-26.

94. Roque L, Castedo S, Gomes P, et al. Cytogenetic findings in 18 follicular thyroid adenomas. Cancer Genet Cytogenet. 1993;67:1-6.

95. Antonini P, Levy N, Caillou B, et al. Numerical aberrations, including trisomy 22 as the sole anomaly, are recurrent in follicular thyroid adenomas. Genes Chromosomes Cancer. 1993;8:63-66.

96. Belge G, Roque L, Soarez J, et al. Cytogenetic investigations of 340 thyroid hyperplasias and adenomas revealing correlations between cytogenetic findings and histology. Cancer Genet Cytogenet. 1998;101:42-48.

97. van Zelderen-Bhola S, Vink R, Smit J, et al. Translocation (X;10)(p22;q24) and (1;10)(q21;q11) in a follicular adenoma of the thyroid without apparent involvement of the RET proto-oncogene. Cancer Genet Cytogenet. 1999;112:178- 180.

98. Bol S, Belge G, Thode B, et al. Structural abnormalities of chromosome 2 in benign thyroid tumors. Cancer Genet Cytogenet. 1999;114:75-77.

99. Cigudosa J, Pedrosa Guerra A, Otero Gomez A, et al. Translocation (5;19)(q13;q13) in a multinodular thyroid goiter. Cancer Genet Cytogenet. 1995;82:67-69.

100. Belge G, Rippe V, Meiboom M, et al. Delineation of a 150-kb breakpoint cluster in benign thyroid tumors with 19q13.4 aberrations. Cytogenet Cell Genet. 2001;93:48-51.

101. Rippe V, Belge G, Meiboom M, et al. A KRAB zinc finger protein gene is the potential target of 19q13 translocation in benign thyroid tumors. Genes Chromosomes Cancer. 1999; 26:229-236.

102. Teyssier J, Ferre D. Frequent clonal chromosomal changes in human non-malignant tumors. Int J Cancer. 1989;44:828-832.

103. Bol S, Belge G, Rippe V, et al. Molecular cytogenetic investigations define a subgroup of thyroid adenomas with 2p21 breakpoint clustered to a region less than 450 kb. Cytogenet Cell Genet. 2001;95:189-201.

104. Oriola J, Halperin I, Mallofre C, et al. Screening of selected genomic areas potentially involved in thyroid neoplasms. Eur J Cancer. 2001;37:2470-2474.

105. Mulligan LM, Kwok JBJ, Healey CS, et al. Germline mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature. 1993;363:458-460.

106. Donis-Keller H, Dou S, Chi D, et al. Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet. 1993;2:851-856.

107. Carlson KM, Dou S, Chi D, et al. Single missense mutation in the tyrosine kinase catalytic domain of the RET protoonco- gene is associated with multiple endocrine neoplasia type 2B. Proc Natl Acad Sci U S A. 1994;91:1579-1583.

108. Hofstra R, Landsvater R, Ceccherini I, et al. A mutation in the RET proto-oncogene associated with multiple endocrine neoplasia type 2B and sporadic medullary thyroid carcinoma. Nature. 1994;367:375-376.

109. Mulligan L, Gardner E, Smith B, et al. Genetic events in tumor initiation and progression in multiple endocrine neo- plasia type 2. Genes Chromosomes Cancer. 1993;6:166-177.

110. Wurster-Hill D, Noll W, Bircher L, et al. A cytogenetic study of familial medullary carcinoma of the thyroid. Cancer Res. 1986;46:2134-2138.

111. Tanaka K, Baylin S, Nelkin B, et al. Cytogenetic studies of a human medullary thyroid carcinoma cell line. Cancer Genet Cytogenet. 1987;25:27-35.

112. Pfragner R, Hoefler H, Behmel A, et al. Establishment and characterization of continuous cell line MTC-SK derived from a human medullary thyroid carcinoma. Cancer Res. 1990;50:4160-4166.

113. Wurster-Hill D, Pettengill O, Noll W, et al. Hypodiploid, pseudodiploid and normal karyotypes in cytogenetic studies of medullary carcinoma of the thyroid and metastatic tissues. Cancer Genet Cytogenet. 1990;47:227-241.

114. Frisk T, Zedenius J, Lundberg J, et al. CGH alterations in medullary thyroid carcinomas in relation to the RET M918T mutation and clinical outcome. Int J Oncol. 2001;18:1219- 1225.

115. Takai A, Tateishi H, Nishisho I, et al. Loss of genes on chromosome 22 in medullary thyroid carcinoma and pheochromocytoma. Jpn J Cancer Res. 1987;78:894-898.

116. Frisk T, Farnebo F, Zedenius J, et al. Expression of RET and its ligand complexes, GDNF/GFRalpha-1 and NTN/GFRalpha- 2, in medullary thyroid carcinomas. Eur J Endocrinol. 2000;142:643-649.

117. Pfragner R, Wirnsberger G, Behmel A, et al. Biologic and cytogenetic characterization of three human medullary thyroid carcinomas in culture. Henry Ford Hosp Med J. 1992;40:299-302.

118. Scappaticci S, Arrigoni G, Capra E, et al. Cytogenetics of multiple endocrine neoplasia syndromes. Cancer Genet Cytogenet. 1992;59:51-53.

119. Veldman T, Vignon C, Schroeck E, et al. Hidden chromosome abnormalities in haematological malignancies detected by multicolour spectral karotyping. Nat Genet. 1997;15:406-410.

120. Oerndal C, Johannson M, Heim S, et al. Parathyroid adenoma with t(1;5)(p22;q32) as the sole clonal chromosome abnormality. Cancer Genet Cytogenet. 1990;48:225-228.

121. Apel R, Asa S. The parathyroid glands. In: LiVolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:103-147.

122. Sigl E, Behmel A, Henn T, et al. Cytogenetic and CGH studies of four neuroendocrine tumors and tumor-derived cell lines of a patient with multiple endocrine neoplasia type 1. Int J Oncol. 1999;15:41-51.

123. Arnold A, Kim H, Gaz R, et al. Molecular cloning and chromosomal mapping of DNA rearranged with the parathyroid hormone gene in a parathyroid adenoma. J Clin Invest. 1989;83:2034-2040.

124. Arnold A, Kim H. Clonal loss of one chromosome 11 in a parathyroid adenoma. J Clin Endocrinol Metab. 1989;69:496- 499.

125. Heppner C, Kester MB, Agarwal SK, et al. Somatic mutation of the MEN1 gene in parathyroid tumours. Nat Genet. 1997;16:375-378.

126. Arnold A, Shattuck TM, Mallya SM, et al. Molecular pathogenesis of primary hyperparathyroidism. J Bone Miner Res. 2002;17(suppl 2):N30-N36.

127. Teh B, Esapa C, Houlston R, et al. A family with isolated hyperparathyroidism segregating a missense MEN1 mutation and showing loss of the wild-type alleles in the parathyroid tumors. Am J Hum Genet. 1998;63:1544-1549.

128. Bergman L, Teh B, Cardinal J, et al. Identification of MEN1 gene mutations in families with MEN1 and related disorders. Br J Cancer. 2000;83:1009-1014.

129. Villablanca A, Wassif W, Smith T, et al. Involvement of MEN1 locus in familial isolated hyperparathyroidism. Eur J Endocrinol. 2002;147:313-322.

130. Carpten JD, Robbins CM, Villablanca A, et al. HRPT2, encoding parafibromin, is mutated in hyperparathyroidism-jaw tumor syndrome. Nat Genet. 2002;32:676-680.

131. Cryns V, Yi S, Tahara H, et al. Frequent loss of chromosome arm 1p DNA in parathyroid adenomas. Genes Chromosomes Cancer. 1995;13:9-17.

132. Tahara H, Smith A, Gaz R, et al. Genomic localization of novel candidate tumor suppressor gene loci in human parathyroid adenoma. Cancer Res. 1996;56:599-605.

133. Palanisamy N, Imanishi Y, Rao P, et al. Novel chromosomal abnormalities identified by comparative genomic hybridization in parathyroid adenomas. J Clin Endocrinol Metab. 1998;83:1766-1770.

134. Kytola S, Maekinen MJ, Kahhonen M, et al. Comparative genomic hybridization studies in tumours from a patient with multiple endocrine neoplasia type 1. Eur J Endocrinol. 1998;139:202-206.

135. Agarwal S, Schroeck E, Kester M, et al. Comparative genomic hybridization analysis of human parathyroid tumors. Cancer Genet Cytogenet. 1998;106:30-36.

136. Farnebo F, Kytola S, Teh B, et al. Alternative genetic pathways in parathyroid tumorigenesis. J Clin Endocrinol Metab. 1999;84:3775-3780.

137. Kytoelae S, Farnebo F, Obara T, et al. Patterns of chromosomal imbalances in parathyroid carcinomas. Am J Pathol. 2000;157:579-586.

138. Hemmer S, Wasenius V-M, Haglund C, et al. Deletion of 11q23 and cyclin D1 overexpression are frequent aberrations in parathyroid adenomas. Am J Pathol. 2001;158:1355-1362.

139. Garcia J, Tardio J, Gutierrez N, et al. Chromosomal imbalances identified by comparative genomic hybridization in sporadic parathyroid adenomas. Eur J Endocrinol. 2002;146:209-213.

140. Pourani J, Kaserer K, Pfragner R. Cytogenetic and molecular analyses of multiple endocrine neoplasias of the MEN1 syndrome. Int J Oncol. 2002;20:971-976.

141. Valimaki S, Forsberg L, Farnebo L, et al. Distinct target regions for chromosome 1p deletions in parathyroid adenomas and carcinomas. Int J Oncol. 2002;21:727-735.

142. Imanishi Y, Tahara H. Putative parathyroid tumor suppressor on 1p: independent molecular mechanisms of tumorigenesis from 11q allelic loss. Am J Kidney Dis. 2001;38(4 suppl 1):S165-S167.

143. Kallioniemi O-P, Kallioniemi A, Piper J, et al. Optimizing genomic hybridization for analysis of DNA sequence copy number changes in solid tumors. Genes Chromosomes Cancer. 1994;10:231-243.

144. Limon J, Dal Cin P, Gaeta J, et al. Translocation t(4;11)(q35;p13) in an adrenocortical carcinoma. Cancer Genet Cytogenet. 1987;28:343-348.

145. Limon J, Dal Cin P, Kakati S, et al. Cytogenetic findings in a primary adrenocortical carcinoma. Cancer Genet Cytogenet. 1987;26:271-277.

146. Marks J, Wyandt H, Beazley R, et al. Cytogenetic studies of an adrenocortical carcinoma. Cancer Genet Cytogenet. 1992;61:96-98.

147. Herrmann ME, Rydstedt LL, Talpos GB, et al. Chromosomal aberrations in two adrenocortical tumors, one with a rearrangement at 11p15. Cancer Genet Cytogenet. 1994;75:111-116.

148. Gordon R, Stowasser M, Martin N, et al. Karyotypic abnormalities in benign adrenocortical tumors producing aldosterone. Cancer Genet Cytogenet. 1993;68:78-81.

149. Bettio D, Rizzi N, Ciardino D, et al. Translocation t(7;17)(q22;p13) as a sole karyotypic change in an adrenal adenoma. Cancer Genet Cytogenet. 1998;103:180-181.

150. McNicol A. The adrenal cortex. In: LiVolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:171-204.

151. Henry I, Bonaiti-Pellie C, Chehensse V, et al. Uniparental paternal disomy in a genetic cancer-predisposing syndrome. Nature. 1991;351:665-667.

152. Gicquel C, Bertagna X, Schneid H, et al. Rearrangements at the 11p15 locus and overexpression of insulin-like growth factor-II gene in sporadic adrenocortical tumors. J Clin Endocrinol Metab. 1994;78:1444-1453.

153. Gicquel C, Raffin-Sanson M-L, Gaston V, et al. Structural and functional abnormalities at 11p15 are associated with the malignant phenotype in sporadic adrenocortical tumors: study on a series of 82 tumors. J Clin Endocrinol Metab. 1997;82:2559-2565.

154. Kjellman M, Kallioniemi O-P, Karhu R, et al. Genetic aberrations in adrenocortical tumors detected using comparative genomic hybridization correlate with tumor size and malignancy. Cancer Res. 1996;56:4219-4223.

155. Zhao J, Speel E, Muletta-Feurer S, et al. Analysis of genomic alterations in sporadic adrenocortical lesions. Am J Pathol. 1999;155:1039-1045.

156. Dohna M, Reincke M, Mincheva A, et al. Adrenocortical carcinoma is characterized by a high frequency of chromosomal gains and high-level amplifications. Genes Chromosomes Cancer. 2000;28:145-152.

157. Sidhu S, Marsh D, Theodosopoulos G, et al. Comparative genomic hybridization analysis of adrenocortical tumors. J Clin Endocrinol Metab. 2002;87:3467-3474.

158. Figueiredo B, Stratakis C, Sandrini R, et al. Comparative genomic hybridization analysis of adrenocortical tumors of childhood. J Clin Endocrinol Metab. 1999;84:1116-1121.

159. James L, Kelsey A, Birch J, et al. Highly consistent genetic alterations in childhood adrenocortical tumors detected by comparative genomic hybridization. Br J Cancer. 1999;81:300-304.

160. Figueiredo B, Ribeiro R, Zambetti G, et al. Amplification of 9q34 in childhood adrenocortical tumors: a specific feature unrelated to ethnic origin or living conditions. Braz J Med Biol Res. 2000;33:1217-1224.

161. Skogeid B, Larsson C, Lindgren P, et al. Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type 1. J Clin Endocrinol Metab. 1992;75:76-81.

162. Zhao J, Roth J, Bode-Lesniewska B, et al. Combined comparative genomic hybridization and genomic microarray for detection of gene amplifications in pulmonary artery intimal sarcomas and adrenocortical tumors. Genes Chromosomes Cancer. 2002;34:48-57.

163. Russel A, Sibbald J, Haak H, et al. Increasing genome instability in adrenocortical carcinoma progression with involvement of chromosomes 3, 9 and X at the adenoma stage. Br J Cancer. 1999;81:684-689.

164. Shono T, Sakai H, Takehara K, et al. Analysis of numerical chromosomal aberrations in adrenal cortical neoplasms by fluorescence in situ hybridization. J Urol. 2002;168:1370- 1373.

165. Kjellman M, Roshani L, Teh BT, et al. Genotyping of adrenocortical tumors: very frequent deletions of the MEN1 locus in 11q13 and of a 1-centimorgan region in 2p16. J Clin Endocrinol Metab. 1999;84:730-735.

166. Gordon R, Gartside M, Tunny T, et al. Different allelic patterns at chromosome 11q13 in paired aldosterone- producing tumors and blood DNA. Clin Exp Pharmacol Physiol. 1996;23:594-596.

167. Iida A, Blake K, Tunny T, et al. Allelic losses on chromosome band 11q13 in aldosterone-producing adrenal tumors. Genes Chromosomes Cancer. 1995;12:73-75.

168. Kirschner L, Carney J, Pack S, et al. Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex. Nat Genet. 2000;26:89-92.

169. Komminoth P, De Krueger R, Tischler A. Paraganglia and the adrenal medulla. In: LiVolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:149-169.

170. van Dekken H, Bosman F, Teijgeman R, et al. Identification of numerical chromosome aberrations in archival tumors by in situ hybridization to routine paraffin sections: evaluation of 23 pheochromocytomas. J Pathol. 1993;171:161-171.

171. Dannenberg H, Speel E, Zhao J, et al. Losses of chromosomes 1p and 3q are early genetic events in the development of sporadic pheochromocytomas. Am J Pathol. 2000;157:353-359.

172. Edstroem E, Mahlamaeki E, Nord B, et al. Comparative genomic hybridization reveals frequent losses of chromosomes 1p and 3q in pheochromocytomas and abdominal paragangliomas, suggesting a common etiology. Am J Pathol. 2000;156:651-659.

173. Lui W-O, Chen J, Glaesker S, et al. Selective loss of chromosome 11 in pheochromocytomas associated with the VHL syndrome. Oncogene. 2002;21:1117-1122.

174. Yokogoshi Y, Yoshimoto K, Saito S. Loss of heterozygosity on chromosomes 1 and 11 in sporadic pheochromocytomas. Jpn J Cancer Res. 1990;81:632-638.

175. Tanaka N, Nishisho I, Yamamoto M, et al. Loss of hetero- zygosity on the long arm of chromosome 22 in pheochromo- cytoma. Genes Chromosomes Cancer. 1992;5:399-403.

176. Shin E, Fujita S, Takami K, et al. Deletion mapping of chromosome 1p and 22q in pheochromocytoma. Jpn J Cancer Res. 1993;84:402-408.

177. Zeiger M, Zbar B, Keiser H, et al. Loss of heterozygosity on the short arm of chromosome 3 in sporadic, von Hippel-Lindau and familial pheochromocytoma. Genes Chromosomes Cancer. 1995;13:151-156.

178. Vargas MP, Zhuang Z, Wang C, et al. Loss of heterozygosity on the short arm of chromosomes 1 and 3 in sporadic pheochromocytoma and extra-adrenal paraganglioma. Hum Pathol. 1997;28:411-415.

179. Benn DE, Dwight T, Richardson AL, et al. Sporadic and familial pheochromocytomas are associated with loss of at least two discrete intervals on chromosome 1p. Cancer Res. 2000;60:7048-7051.

180. Edstrom Elder E, Nord B, Carling T, et al. Loss of heterozygosity on the short arm of chromosome 1 in pheochromocytoma and abdominal paraganglioma. World J Surg. 2002;26:965-971.

181. Furth E. Gastrointestinal tract. In: Livolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:237-260.

182. Lubensky I. Endocrine pancreas. In: LiVolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:205-235.

183. Herrmann M, Rydstedt L, Talpos G, et al. Chromosomal aberrations in two sporadic gastrinomas. Cancer Genet Cytogenet. 1993;67:44-49.

184. Bugalho MJ, Roque L, Sobrinho LG, et al. Calcitonin- producing insulinoma: clinical, immunocytochemical and cytogenetic study. Clin Endocrinol (Oxf). 1994;41:257-260.

185. Iwamura Y, Futagawa T, Kaneko M, et al. Co-deletions of the retinoblastoma and Wilms’ tumor gene and rearrangement of the Krev-1 gene in human insulinoma. Jpn J Clin Oncol. 1992;22:6-9.

186. Long P, Hruban R, Lo R, et al. Chromosome analysis of nine endocrine neoplasms of the pancreas. Cancer Genet Cytogenet. 1994;77:55-59.

187. Missiaglia E, Moore P, Williamson J, et al. Sex chromosome anomalies in pancreatic tumors. Int J Cancer. 2002;98:532-538.

188. Terris B, Meddeb M, Marchio A, et al. Comparative genomic hybridization analysis of sporadic neuroendocrine tumors of the digestive system. Genes Chromosomes Cancer. 1998;22:50- 56.

189. Speel E, Richter J, Moch H, et al. Genetic differences in endocrine pancreatic tumor subtypes detected by comparative genomic hybridization. Am J Pathol. 1999;155:1787-1794.

190. Yu F, Jensen R, Lubensky I, et al. Survey of genetic alterations in gastrinomas. Cancer Res. 2000;60:5536-5542.

191. Stumpf E, Aalto Y, Hoog A, et al. Chromosomal alterations in human pancreatic endocrine tumors. Genes Chromosomes Cancer. 2000;29:83-87.

192. Speel E, Scheidweiler A, Zhao J, et al. Genetic evidence for early divergence of small functioning and nonfunctioning endocrine pancreatic tumors: gain of 9Q34 is an early event in insulinomas. Cancer Res. 2001;61:5186-5192.

193. Zhao J, Moch H, Scheidweiler A, et al. Genomic imbalances in the progression of endocrine pancreatic tumors. Genes Chromosomes Cancer. 2001;32:364-372.

194. Zhao J, de Krijger R, Meier D, et al. Genomic alterations in well-differentiated gastrointestinal and bronchial neuroendocrine tumors (carcinoids): marked differences indicating diversity in molecular pathogenesis. Am J Pathol. 2000;157:1431-1438.

195. Kytola S, Hoog A, Nord B, et al. Comparative genomic hybridization identifies loss of 18q22-qter as an early and specific event in tumorigenesis of midgut carcinoids. Am J Pathol. 2001;158:1803-1808.

196. Evers B, Rady P, Sandoval K, et al. Gastrinomas demonstrate amplification of the HER2/neu proto-oncogene. Ann Surg. 1994;219:596-604.

197. Debelenko LV, Zhuang Z, Emmert-Buck MR, et al. Allelic deletions on chromosome 11q13 in multiple endocrine neoplasia type 1-associated and sporadic gastrinomas and pancreatic endocrine tumors. Cancer Res. 1997;57:2238- 2243.

198. Chung D, Smith A, Louis D, et al. A novel pancreatic endocrine tumor suppressor gene locus on chromosome 3p with clinical prognostic implications. J Clin Invest. 1997;100:404-410.

199. Barghorn A, Komminoth P, Bachmann D, et al. Deletion at 3p25.3-p23 is frequently encountered in endocrine pancreatic tumours and is associated with metastatic progression. J Pathol. 2001;194:451-458.

200. Barghorn A, Speel E, Farspour B, et al. Putative tumor suppressor loci at 6q22 and 6q23-q24 are involved in the malignant progression of sporadic endocrine pancreatic tumors. Am J Pathol. 2001;158:1903-1911.

201. D’Adda T, Pizzi S, Azzoni C, et al. Different patterns of 11q allelic losses in digestive endocrine tumors. Hum Pathol. 2002;33:322-329.

202. Marx S, Agarwal S, Heppner C, et al. The gene for multiple endocrine neoplasia type 1: recent findings. Bone. 1999;25:119-122.

203. Gortz B, Roth J, Krahenmann A, et al. Mutations and allelic deletions of the MEN1 gene are associated with a subset of sporadic endocrine pancreatic and neuroendocrine tumors and not restricted to foregut neoplasms. Am J Pathol. 1999;154:429-436.

204. Mark J. Chromosomal characteristics of human pituitary adenomas. Acta Neuropathol (Berl). 1971;19:99-109.

205. Rey J, Bello M, De Campos J, et al. A case of pituitary adenoma with 58 chromosomes. Cancer Genet Cytogenet. 1986;23:171-174.

206. Rock J, Babu U, Drumheller T, et al. Cytogenetic findings in pituitary adenomas: results of a pilot study. Surg Neurol. 1993;40:224-229.

207. Papi L, Baldassari G, Montali E, et al. Cytogenetic studies in sporadic and multiple endocrine neoplasia type I-associated pituitary adenomas. Genes Chromosomes Cancer. 1993;7:63- 65.

208. Dietrich C, Pandis N, Bjerre P, et al. Simple numerical chromosome aberrations in two pituitary adenomas. Cancer Genet Cytogenet. 1993;69:118-121.

209. Capra E, Rindi G, Santi G, et al. Chromosome abnormalities in a case of pituitary adenoma. Cancer Genet Cytogenet. 1993;68:140-142.

210. Bettio D, Rizzi N, Giardino D, et al. Cytogenetic study of pituitary adenomas. Cancer Genet Cytogenet. 1997;98:131-136.

Herrmann / STANDARD AND MOLECULAR CYTOGENETICS OF ENDOCRINE TUMORS

211. Cheung C, Asa S. Pathology of the pituitary gland. In: Livolsi V, Asa S, eds. Endocrine Pathology. Philadelphia, PA: Churchill Livingstone; 2002:31-59.

212. Herrmann M, Wojcik EM, Siziopikou K, et al. FISH evaluation of archival specimens of pituitary and breast tumors for chromosome copy number and oncogene amplification [abstract]. Mod Pathol. 2000;13:223A.

213. Metzger A, Mohapatra G, Minn Y, et al. Multiple genetic aberrations including evidence of chromosome 13 rearrangement detected in pituitary adenomas by comparative genomic hybridization. J Neurosurg. 1999;90:306-314.

214. Daniely M, Aviram A, Adams E, et al. Comparative genomic hybridization analysis of nonfunctioning pituitary tumors. J Clin Endocrinol Metab. 1998;83:1801-1805.

215. Trautmann K, Thakker R, Ellison D, et al. Chromosomal aberrations in sporadic pituitary hormones. Int J Cancer. 2001;91:809-814.

216. Harada K, Nishizaki T, Ozaki S, et al. Cytogenetic alterations in pituitary adenomas detected by comparative genomic hybridization. Cancer Genet Cytogenet. 1999;112:38-41.

217. Pack SD, Kirschner LS, Pak E, et al. Genetic and histologic studies of somatomammotropic pituitary tumors in patients with the “complex of spotty skin pigmentation, myxomas, endocrine overactivity and schwannomas” (Carney complex). J Clin Endocrinol Metab. 2000;85:3860-3865.

218. Hui A, Pang J, Ko C-W, et al. Detection of chromosomal imbalances in growth hormone-secreting pituitary tumors by comparative genomic hybridization. Hum Pathol. 1999;30:1019-1023.

219. Fan X, Paetau A, Aalto Y, et al. Gain of chromosome 3 and loss of 13q are frequent alterations in pituitary adenomas. Cancer Genet Cytogenet. 2001;128:97-103.

220. Pei L, Melmed S, Scheithauer B, et al. Frequent loss of heterozygosity at the retinoblastoma susceptibility gene (RB) locus in aggressive pituitary tumors: evidence for a chromosome 13 tumor suppressor gene other than RB. Cancer Res. 1995;55:1613-1616.

221. Farrell W, Clayton R. Molecular biology of human pituitary adenomas. Ann Med. 1998;30:192-198.

222. Boggild M, Jenkinson S, Pistorello M, et al. Molecular genetic studies of sporadic pituitary tumors. J Clin Endocrinol Metab. 1994;78:387-392.

223. Finelli P, Pierantoni G, Giardino D, et al. The high mobility group A2 gene is amplified and overexpressed in human prolactinoma. Cancer Res. 2002;62:2398-2405.