ICANCERRESEARCH57. 1690-1694,MayI. 19971

ABSTRACT

In this study, we compare the morphological and genetic characteristicsof 38 post-Chernobylthyroid papillary carcinomasfrom Belarussianchildren 5—18years old with those of 23 sporadic papillary carcinomasfrom the same age children without history of radiation exposure fromLos Angeles and Cincinnati. Among radiation-induced tumors, solid variant of papillary carcinoma was found in 37%, follicular in 29%, typicalpapillary in 18%, and mixed and diffuse sclerosing variants in 8% each.

In the sporadic group, a typical papillary pattern was prevalent in 70%,follicular In 17%, diffuse scierosing variant in 9%, and solid in 4%. Inboth groups, the prevalence of ret rearrangements was high, but thefrequency of specific types of rearrangement was significantly different.Among radiation-induced tumors, retlPTC3 was found in 58%, ret/PTCJin 16%, and ret/PTC2 in 3%, whereas among sporadic tumors, ret/PTCJwas found in 47% (P < 0.05), and ret/PTC3 was found in 18% (P 0.01).The morphological variants ofpapillary carcinoma showed different preyalence of the specific types of ret rearrangement. Seventy-nine % of solidvariant tumors had ret/PTC3, whereas only 7% had ret/PTCJ(P = 0.0007). Among typical papifiary tumors, ret/PTCJ was found in38%, ret/PTC3 In 19%, and ret/PTC2 in 5%. Thus, ret rearrangements arehighly prevalent in pediatric papillary carcinomas from children exposedto radiation and in those occurring sporadically. However, the types ofret/PTC vary between these two populations, with ret/PTC3 present morecommonly in post-Chernobyl tumors. Furthermore, solid variants have ahigh prevalence ofret/PTC3, whereas typical papillary carcinomas do not,suggesting that the different types of ret rearrangement confer neoplasticthyroid cells with distinct phenotypic properties.

INTRODUCTION

Thyroid cancer in children exposed to radiation as a result of theChernobyl nuclear accident in April, 1986 is now a widely acceptedparadigm of radiation-induced neoplasia. A sharp increase in theincidence of pediatric thyroid carcinoma was noted as soon as 4 yearsafter the accident (1). Thus far, several hundred cases have beenreported in children living in the contaminated areas of Belarus,Ukraine, and western regions of Russia, resulting in up to a 100-foldincreased annual incidence of childhood thyroid carcinoma as cornpared to that prior to 1986 (2, 3).

The vast majority of these tumors are thyroid carcinomas of thepapillary type. A detailed morphological analysis of 84 cases fromBelarus showed that solid and follicular variants of papillary carcinomawere prevalent in these tumors (4). Spontaneous papillary thyroid carcinomas are extremely rare in children, and to our knowledge their morphological variants have not been characterized in detail.

Our understanding of the genetic events associated with radiationinduced thyroid cancer in general, and of post-Chernobyl tumors inparticular, is far from complete. ras oncogene mutations, which are found

Received I1/4/96;accepted 3/10/97.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I This work was supported in part by Grants CA 50706 and CA 72597 and a University

of Cincinnati Cancer Research Challenge Award. J. A. F. is the recipient of an EstablishedInvestigatorship Award from the American Heart Association and Bristol Myers-Squibb.

2 To whom requests for reprints should be addressed, at Division of Endocrinology/

Metabolism, University of Cincinnati College of Medicine, P. 0. Box 670547. Cincinnati,OH 45267-0547. Phone: (513) 558-4444; Fax (513) 558-8581.

in 18—62%of papillary carcinomas in the general population, are notobserved at all in these tumors (5, 6). Mutations of the p53 gene, whichare common in sporadic undifferentiated and poorly differentiated thyroidcarcinomas (7, 8), have a very low prevalence in the Chemobyl population (5, 6), in spite of the fact that many papillary carcinomas in the latterpopulation have a solid growth pattern, indicative of a less differentiatedphenotype.

Recently, two studies of a small series of post-Chernobyl papillarythyroid carcinomas reported a 67% prevalence of the ret gene rearrangements (9, 10). ret rearrangements are formed by the fusion of theintracellular tyrosine kinase domain of the gene with different 5' genefragments, resulting in inappropriate constitutive expression of thetruncated tyrosine kinase domain of the receptor. The authors proposed that this genetic event might reflect a radiation origin of thesetumors. Indeed, the incidence of ret alterations found in pediatricpost-Chernobyl tumors is much higher than that reported previously inthe general (mostly adult) population, where three types of ret rearrangements, ret/PTCJ, ret/PTC2, and ret/PTC3 have been detected inup to 34% of papillary thyroid carcinomas (data summarized in Ref.I 1). In addition, the induction of ret/PTCJ rearrangement has beenreported after high dose X-ray irradiation in vitro (12). However, arecently published study demonstrates a similarly high incidence ofret/PTC rearrangements in spontaneous thyroid carcinomas in Italianchildren but not in adults, suggesting that this might be an agespecific, rather than radiation-specific, event (13).

To clarify the role of ret oncogene alterations in radiation-inducedcarcinogenesis in the thyroid gland and to compare the morphology ofradiation-induced and spontaneous thyroid cancers in children, westudied a large series of post-Chernobyl pediatric papillary thyroidcarcinomas and compared them with a group of pediatric tumorswithout a history of radiation exposure from two different areas in theUnited States.

MATERIALS AND METHODS

Patient Population. Thirty-eight cases ofthyroid papillary carcinoma fromBelarussian children 5—18years of age who lived in the areas contaminated asa result of the Chernobyl nuclear accident were studied. These cases wereselected from a large consecutive series of thyroid tumors in children thatunderwent surgery at the Thyroid Tumor Center in Minsk in 1991—1992 based

on the availability of RNA extracted from paraffin-embedded tissue of sufficient integrity to allow genetic analysis. The morphological and epidemiological features of these tumors have been described previously (4, 14). Toperform a comparative analysis of radiation-induced and sporadic pediatricthyroid tumors, records of all cases of thyroid carcinoma in children up to I 8

years old were obtained from the files of the Department of Pathology at

Childrens Hospital Los Angeles (Los Angeles, CA) and the Department ofPathology at Childrens Hospital Medical Center (Cincinnati, OH). Fourteenprimary cases of papillary carcinoma were found in Los Angeles since 1974and 10 cases in Cincinnati since 1982. The medical records of each patientwere reviewed for radiation history. One patient from Cincinnati had receivedX-ray therapy in a dose of I800 rads to the brain area for acute lymphoblasticleukemia 8 years prior to removal of thyroid carcinoma. The latter case waseliminated from further analysis. Altogether, 23 spontaneous papillary thyroid

carcinomas from children 5—I8 years of age were studied.

1690

Distinct Pattern of ret Oncogene Rearrangements in Morphological Variants of

Radiation-induced and Sporadic Thyroid Papillary Carcinomas in Children'

Yun E. Nikiforov, Jon M. Rowland, Kevin E. Bove, Hector Monforte-Munoz, and James A. Fagin2

Division of Endocrinology, University of Cincinnati College of Medicine, Cincinnati. Ohio 45267-0547 (1'. E. N.. J. A. Fl: Department of Pathology, Childrens Hospital LosAngeles, Los Angeles. California 90027 Ii. M. R., H. M.M.J: and Department of Pathology. Childrens Hospital Medical Center, Cincinnati, Ohio 45229 (K. E. B.]

Research. on November 22, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table 1 Primers and probes usedfor thescreeningSizePrimer

sequences (5—3',a-sense,b-Nucleotide(bp)antisense)positionsaProbesfor hybridization(5—3')c-ms236a

-ATGACTGAGTACAAACTGGTb-AGGAAGCCTTCGCCTGTCCT727-746943-962CAAGTGGTTATAGATGGTGAc-ret,

TK155a-GGAGCCAGGGTCGGATTCCAGTTAb-CCGCTCAGGAGGAATCCCAGGATA2654-26772785-2808ACGCAAAGTGATGTATGGTCTc-ret,

EC184a -GGCGGCCCAAGTGTGCCGAACTTb-CCCAGGCCGCCACACTCCTCACA1

352-1 374

1513-1535GTAACAGTGGAGGGGTCATATGretIPTCl165a-GCTGGAGACCTACAAACTGA

b-GTTGCCTTGACCACTTTTC279-298425-443GGCACTGCAGGAGGAGAACCGCGAretIFI@C2202a-GCTTTGGAGAACTTGCTTTGATTT

b-GTTGCCTTGACCACTTTTC605-626788-806GACCGAGACAGCTATAGAAGAATCretIPTC3242a-AAGCAAACCTGCCAGTGG

b-CTTTCAGCATCTTCACGG687-704 911-928GGTCGGTGCTGGGTATGTAAGGA

ret/PTC GENE IN ThYROID TUMORS

a Nucleotide positions of the primers are based on the sequences deposited in GenBank: c-N-ras (accession no. X02751), c-ret (accession no. X12949), ret/PTCI (accession no.

M312l3), retIPTC2 (accession no. L03357), and ret/PTC3 (accession no. S7l225).

Histology. Histological slides from primary thyroid tumors stained byH&E were reviewed by at least two pathologists (Y. E. N., J. M. R., and H.M.-M.; or Y. E. N. and K. E. B.) to confirm the diagnoses and subclassify thevariants of papillary carcinomas. Simultaneously, the sections containing 70—

100% of tumor from lymph node metastasis or primary nodules were selectedfor RNA extraction from the corresponding paraffin blocks.

GeneticAnalysis.Inall cases,RNAwasextractedfromparaffin-embedded tissue as reported previously (15, 16). In addition, in five cases fromCincinnati, RNA was obtained from frozen tissue as described by Chomczynski and Sacchi (17). After reverse transcription at 42°Cby Superscript reversetransciptase (BRL) using a random hexanucleotide mixture as a primer, cDNAwas first tested to ensure that the integrity of RNA was sufficient for analysis.For this purpose, a 236-bp sequence of the c-N-ras cDNA was amplified withprimers spanning intron 1 of the gene (Table 1). Only those samples demonstrafing clearly detectable levels of the low abundance N-ras transcript werefurther analyzed. To screen for ret mRNA expression, we first coamplifiedfragments corresponding to the TK3 and EC domains of c-ret, respectively,using two sets of intron-spanning primers (Table 1). The cDNAs of the tumorsdemonstrating preferential expression of the TK domain were considered assuspicious for having ret rearrangements. Nevertheless, all tumor cDNAs werescreened with primers bracketing chimeric sequences corresponding to therearrangements of ret/PTCJ, ret/PTC2, and ret/PTC3 types (Table I). For each

PCR, 2 @slofreverse transcripted mixture were amplified with 10pmol of eachprimer, 200 @.sMdeoxynucleotide triphosphates, 10 mrsiTris-HC1 (pH 9.0 at25°C),50 mM KC1, 1% Triton X-lOO, I .5 mM MgC12, and 1 unit of Taq DNApolymerase (Promega Corp., Madison, WI) in a final volume of 30 p1Thirty-five cycles of denaturation (94°Cfor 1 mm), annealing (55°-63°Cfor 1rain), and extension (72°Cfor 1.5—2mm) were conducted on an automatedheat block (DNA thermal cycler; Perkin-Elmer, Norwalk, CT). Positive con

trols for amplification included RNA samples from a medullary thyroid carcinoma with constitutive expression of wild-type c-ret (for amplification of retTK and EC domains), from the TPC-l cell line with a ret/PTCJ rearrangement(18), and cloned vectors containing ret/PTC2 and ret/PTC3 rearranged geneskindly provided by Dr. S. M. Jiang (Ohio State University, Columbus, OH).

Ten pi of PCR product were electrophoresed in a 1.5% agarose gel andblotted to a nylon membrane. Each filter was then hybridized with the correspending specific probe (Table 1) for 16 h at 42°,washed, and exposed forautoradiography for 4—16h.

Statistical Analysis. A@ method was used to test homogeneity betweenthe various groupings in the radiation-induced and sporadic sets. Becausemany groupings had a small number of observations, the test of two proportions with double-sided confidence limits and the Yates correction was performed to test whether specific proportions were different between two groups.Alternatively, the two-tailed Fisher exact test was used for smaller sample sizes

(19). Two values were considered significantly different when the probabilityof P was less than 0.05.

RESULTS

Epidemiological and Morphological Characteristics. The age ofchildren in both groups ranged from 5 to 18 years. The mean age was

11.2 years among post-Chernobyl tumors, and 13.7 years among spontaneous thyroid carcinomas. In the former group, there were 18 girls and20 boys, resulting in a female:male ratio of0.9:l. In the latter group, therewere 20 girls and 3 boys with a sex ratio of 6.6:1 (P < 0.01).

Morphologically, papillary carcinomas were classified according to theprevalent growth pattern in the primary tumor nodule as either typical

papillary, follicular, or solid variant (Fig. 1). Typical papillary carcinomaswere composed of papillae with a central fibrovascular core covered byneoplastic epithelial cells. The follicular variant tumors contained predominantly follicles, which varied in size and content ofcolloid, and werelined by epitheium with cytological features of papillary carcinoma. Thesolid variant was composed exclusively or predominantly by solid sheetsof neoplastic cells surrounded by varying amounts of fibrotic stroma; thetypical nuclear features of papfflazy carcinoma were commonly retained.Mixed variant was diagnosed in cases when the tumor nodule displayedan almost equal prevalence of any two or all three patterns of growth. Inaddition, tumors were classified as diffuse sclerosing variant based on adiffuse type of growth without a distinguishable solitary nodule, presenceof severe fibrosis, dense lymphocytic infiltration, and areas of squamousmetaplasia (20). The prevalence of morphological variants in both groups

are summarized in Table 2. The distribution of histological phenotypeswas found to be significantly different between radiation-induced andsporadic tumors (@ test, P < 0.001). Typical papillary carcinoma wassignificantly more prevalent among sporadic tumors (test of proportions,P < 0.001), whereas solid variant tumors were more common among theexposed children (Fisher's Exact test, P = 0.006).

Molecular Genetic Analysis. Of 38 radiation-induced tumors, 33(87%) showed preferential or exclusive expression of the TK domainas opposed to the EC domain of c-ret, indicating the possible occurrence of rearrangements with a breakpoint somewhere between thesetwo regions (Fig. 2A). All of these positive cases were screened for thethree known types of ret/PTC rearrangements. Ret/PTCJ type wasidentified in 6 cases (16%), ret/PTC2 in 1 case (3%), and ret/PTC3was found in 22 cases (58%; Fig. 2B). In 4 cases with strong expression of TK domain (cases R25, R32, R33, and R35), none of theknown types of ret rearrangement was identified, in spite of the factthat the experiment was replicated twice with fresh preparations oftumor RNA. The mean age at the time of radiation exposure was 4.3years among those with ret/PTC3, 4.2 years among those who hadret/PTCI, and 5.8 years in a case with ret/PTC2 rearrangement.

In 17 of 24 sporadic tumors, the preservation of RNA was sufficientfor further screening. Of these, 12 (71%) demonstrated preferential orexclusive expression of TK domain as opposed to the EC domain of theret gene (Fig. 3A). Further screening showed the presence of ret/.PTCJ in8 (47%) cases and ret/PTC3 in 3 (18%) cases; there were no cases withthe reUPTC2 type in this group (Fig. 3B). One tumor with predominantTK expression (514) did not show any of the known types of retrearrangement. None of the cases from either group that lacked prefer3 The abbreviations used are: TK, tyrosine kinase; EC, extracellular.

1691

Research. on November 22, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table 2 Prevalence ofmorphological variantsofpapillary carcinoma amongradiation-induced and sporadic pediatricthyroidtumorsNo.

ofVariants

of papillarycarcinomaFollicular

1 1 (29%)4(17%)Typical

papillary

7 (18%)16(70%)bMixed

3 (8%)

0Diffuse

sclerosing

3 (8%)2(9%)

cases

rez/PTC GENE IN THYROID TUMORS

mors did not correlate significantly with patients' sex, despite adramatic difference in sex ratio between the two groups as a whole.On the other hand, there was a close correspondence between the typeof ret/PTC rearrangement and the morphological characteristics of thetumors (Table 4). Most notably, solid variant neoplasms had a veryhigh prevalence of ret/PTC3 rearrangement, whereas typical papillarycarcinomas did not (Fisher's Exact test, P = 0.0007).

DISCUSSION

To our knowledge, this is the largest comparative analysis of radiationinduced and sporadic thyroid carcinomas in children that demonstratesclear differences in some epidemiological characteristics, morphology,and types of the ret oncogene alteration between these two groups ofpediatric neoplasms. In addition, this is the first study documenting anassociation between different types of ret rearrangement and distinctphenotypic variants of papillary thyroid carcinoma.

There was a significant difference in sex distribution between bothgroups, with an almost equal female:male ratio among radiationinduced tumors and a dramatic female predominance among sporadiccases. Morphological analysis indicated that solid and follicular vanants of papillary carcinoma were more prevalent in radiation-inducedtumors. This was particularly obvious for the solid variant, whichconstituted 37% of radiation-induced but only 4% of sporadic tumors.In addition, three cases of mixed variant tumors (composed of equalportions of solid and follicular pattern in two cases, and of a mixtureof solid, follicular, and typical papillary patterns in one case) werefound only in the radiation-induced group, making the predominanceof solid growth pattern in this group even higher. The solid variant ofpapillary carcinoma, which is a rare finding in either children or in theadult population (21), is poorly characterized from the point of viewof biological behavior and patient prognosis but is generally considered to be a more aggressive form of the disease. In contrast toradiation-induced neoplasms, tumors displaying the typical papillarypattern were characteristic of the sporadic group. In this respect,sporadic pediatric papillary carcinomas are similar to papillary carcinomas in the adult population.

ret rearrangements were highly prevalent in both groups. However,the prevalence of specific types of rearrangement was markedlydifferent; ret/PTCI was the dominant form in sporadic tumors, andret/PTC3 was the dominant form in radiation-induced neoplasms.

Arguably, pediatric tumors from the United States are not the bestcontrol group for comparison with tumors found in Belarussian children. However, sporadic thyroid carcinomas were collected fromchildren residing in two different areas in the United States. Inaddition, similar data on predominance of ret/PTC1 over ret/PTC3types of rearrangement in sporadic pediatric thyroid carcinomas havebeen reported in Italian children and adolescents younger than 19years of age (13). In contrast, the predominance of ret/PTC3 inpost-Chernobyl pediatric papillary carcinomas has been reported before in the series of 6 and 12 cases (9, 10).

The histological variants of papillary carcinoma exhibited strongcorrespondence with the type of ret rearrangement. Thus, amongpost-Chernobyl carcinomas, solid variant tumors exhibited a striking

4@@@@ -:.@@@@@

@ ).@ C.. . 1'-‘.@- ..@..@ @% S

@ :@t-;1@'“!@:

@ ‘@‘@@‘@

@ .

@T@

@ —

@..;@ S . t@-.--'.@-.::@― ,,,‘,..@â€â€¢â€˜.‘. r' @4@@t'@‘,-@ . ,@.- 4 @. @@:‘-.@ •% a@';'4s@@ ‘@4SI@1@

rZ@@@ q@/I@@1P,@\c t (C@@ ‘a@@ —:@.: @.@ ‘..@.@ ,@@@ @@‘:@.@ .@ Ik@@@@@ I ‘:@‘@ •‘@

@ —@ ...—....,. ?‘ ‘@@ .@@

,-‘. ,,.s @- @. ‘ a.,. . $.@ I-@ @.@@ . - \3@;@@@ ‘z@@.'>

@ cc....@ ‘,:‘ @‘‘â€â€˜ t.p,@@.., “ ::,@@ .@ ,_;‘, .. @;@i@

@@ I@ -@;@@

,-@@ S @.

@@:‘@@ I'.@

@. :@@@ -,@@ :@ 1' : @‘“j@: •.@.- st,)4@•, .

@ ,@@@@ ‘@:

@.. •‘@:@ —.,•,@ .@ ...

. p :- •-:_‘?.,@:. ‘@ @,, .,.,....,@@ ._;@

â€..@..-•..‘,f@.,‘-.@t...; ..k@'@@@ ‘-@ “@. @: •@@ . @&.,!@ @‘:.‘-s@@.-:.@ . .:@@

@@@ ?@@ \ , ?@@ ‘;

@ â€_,‘L1 p.s. \@ .@ @‘- . ( ,..,‘ f., •‘@i:@ @?J.@ WIa .@@ .@@ , @,@ ,. . .@@ . . .@,, . @,@

@ ii―@@@@ t@@@ ..c@‘.@@@@@@@@ @‘

@@ 1@*―/@ @. ; @;e.@

@ @: ,@ e -‘@ . i...@

@@ -- I@.―@ç. ‘r@@@@ ‘ d7.-@@@

@ J;,eI @I@ \ “CI i._@%@‘•@_ —@ ‘ Sp@@@ %.

Fig. I . Morphology of the major variants of papillary thyroid carcinomas according tothe prevalent type of growth: a, typical papillary carcinoma; b, follicular variant; and c,solid variant. H&E staining. X 180.

ential TK over EC ret mRNA had evidence of either ret/PTCJ, ret/PTC2,or ret/PTC3. These findings are summarized in Table 3. The frequency ofspecific types of ret rearrangement was significantly different betweenradiation-induced and sporadic tumors (@ test, P < 0.03). Specifically,ret/PTCI was the most prevalent type in the sporadic group (test of

proportions,P ret/PTC3 was the most prevalent type in radiation-inducedtumors (Fisher's Exact test, P = 0.01). The presence of ret/PTCJ andret/PTC3 rearrangements was confirmed by Southern blotting of two

cases of sporadic carcinoma (data not shown).The observed differences in histological variants and types of

ret/PTC rearrangements between radiation-induced and sporadic tu

Solid

Radiation-induced 38 (100%) 14(37%)―Sporadic 23 (100%) 1 (4%)

ap o.oo@in comparisonwiththeprevalenceof solidvariantin sporadictumors.b ,, < 0.001 in comparison with the prevalence of typical papillary variant in radiation-induced tumors.

1692

Research. on November 22, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

cinomas both in radiation-induced and sporadic groups. Almost onethird of carcinomas with typical papillary growth had no detectablerearrangements of ret.

These data indicate that different types of ret rearrangement are associated with distinct phenotypes of papillary carcinoma. This observationwas not expected, because all types of ret rearrangement lead to theinappropriate expression of the intracellular domain of the ret receptor.This is thought to be a major mechanism of thyroid cell transformation(22,23).Thethreetypesareformedby thefusionof theTK domainofret to different 5' partners. ret/PTCJ is formed by a paracentric inversionof the long arm of chromosome 10, leading to fusion with a gene namedH4/DJOSJ7O (24). reUPTC3 is also a result of an intrachromosomalrearrangement and is formed by the fusion with the RFG/ELEJ gene alsolocated on chromosome 10 (25, 26). ret/PTC2 is formed by a reciprocaltranslocation between chromosomes 10 and 17, resulting in the fusion ofthe TK domain of c-ret with 5' terminal sequence derived from theregulatory subunit of Ria cAMP-dependent protein kinase A (27). In allcases, fusion points are located within the intron bracketed by the transmembrane domain and the TK region of ret. The truncated fragment ofret lacks the transmembrane domain and has been shown to be locatedwithin the cytosol (28). The expressionof this chimeric product is drivenby the respective promoters of the genes upstream of the rearrangement.The contrasting phenotypes reported here between tumors harboringreUPTCJ and ret/PTC3 translocation may be due to unique effects of the

sequences upstream of the breakpoint or to the levels of activity of theret/PTCI and ret/PTC3 promoters, which may result in important differ

ences in the absolute amounts of the truncated ret mRNA and protein.These data, therefore, point to important structural differences betweenthe various rearranged forms of ret, which are worthy of further study.

Table 3 Prevalence ofret rearrangements inradiation-induced and sporadicpediatric thyroidtumorsTK

positive1K

negativeret/PTCJret/PTC2ret/PTC3Novel ret/PTC?Radiation-induced

Sporadic6(16%)

8 (47%)b1(3%)

022(58%)―

3 (18%)4(ll@%)

1 (6%)5(13%) 5 (29%)

ret/PTC GENE IN ThYROID TUMORS

- res/P@rC3

preponderance of ret/PTC3, which was detected in 79% of cases. Inaddition, all three mixed variant tumors, consisting of 30—50%ofsolid growth pattern also had ret/PTC3 rearrangements. In contrast,ret/PTCI was associated more commonly with typical papillary car

A

B

a p 0.01 in comparison with the prevalence of ret/PTC3 in sporadic tumors.

b ,@ < o.os in comparison with the prevalence of ret/PTC1 in radiation-induced tumors.

1693

0 — N “@@ ‘r@@ON 00 0' 0 — r'i “@@ v@ @ON@ O@ 0 — —i@@ @iso N@ ,,,

—@@@ .-.@ -ECA

.—.

— —@ . .ret/PTC1

B@

= -:@ _ — @—.. — @-@ 0@@ .

Fig. 2. Screening of radiation-induced tumors for ret gene rearrangements. cDNA was amplified with the indicated primers followed by Southern blotting and hybridization withspecific probes (Table I). A. amplification ofthe fragments corresponding to the TK and EC domains ofc-re:. Positive control (Pos. C.) is medullary thyroid carcinoma with expressionof wild-type, full-length c-ret mRNA. Negative control (Neg. C.) is in the absence of template. Tumors demonstrating disproportionate expression of TK as opposed to EC domainwere considered to be positive for ret rearrangement. B, amplification of ret/FTC!, ret/PTC2, and ret/PTC3 in each tumor sample, positive controls (indicated in “MaterialsandMethods―),and negative control. ret/PTCI, ret/PTC2, or ret/PTC3 were found in all TK-positive cases except for R25, R32, R33, and R35. As expected, tumor samples that werenegative for expression of TK (RiO, R16. R34, R36, and R38) did not show evidence of either ret/PTCI, ret/PTC2, or ret/PTC3.

@@Lj&5 @i@2@ ;3@ t;i@ °‘B@ “w@ @or'

•-EC

=@-r@@ — @. @i@_@ .@:.@@ -@ —

@@@ —@ •-ret/PTCI

- rct/PTC2

S S -ret/PTC3

Fig. 3. Screening of sporadic tumors for ret gene rearrangements. A, amplification ofthe fragments corresponding to the TK and EC domains of c-ret. Tumors demonstratingdisproportionate expression of TK as opposed to the EC domain were considered to bepositive for ret rearrangement. Negative cases had no expression of either the TK or ECdomain of ret (54, 56, 57, and 512) or exhibited low abundance of both fragments (511)proportional to the expression of 1K and EC in the C-cell tumor (positive control; Pos.C.). In B, ret/PTCI or ret/PTC3 rearrangementswere found in all cases positive for TKexpression except for case SI4.

Research. on November 22, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table4 Prevalence of ret rearrangements in morphological variantsofpapillarycarcinomaret!Morphological

variantRadiat.―ret!PTCISpored.PTC2 Radiat.Radiat.ret/PTC3Sporad.Novel

ret/PTC ?

Radiat. Sporad.TK

negative

Radiat.Sporad.Typical

papillary2/76/141/71/73/141/7 1/142/74/14Follicular

Solid2/Il 1/140 00 05/11 11114b0 01/110

2/14 03/111/1

0Mixed0003/30000Diffuse

sclerosing1/32/202/300 00

rez/PTC GENE IN THYROID TUMORS

a Radiat.. radiation-induced; Sporad.. sporadic.

b ,, o.oooi in comparison with the prevalence of ret/PTCI in the solid variant tumors.

Five of the cases reported here had clear evidence of preferential

expression of the TK fragment of ret, which was not due to either

ret/PTCJ, ret/PTC2, or ret/PTC3, suggesting that it may have resultedfrom novel forms of ret rearrangement. While this report was inpreparation, a subtype of ret/PTC3 containing an additional 93 bp wasreported and termed ret/PTC4 (29). However, it appears that at best

only one of the five tumors may be explained by ret/PTC4 (data notshown), indicating that additional rearrangements of ret are present intumors from this pediatric population.

The prevalence of ret rearrangements found in both pediatric groups ismuch higher than in adults. This invites the speculation that almost allthyroid carcinomas in children may be radiation induced. If this were thecase, the differences observed in this study may be due to the fact that inthe documented radiation-induced cases, the high-dose exposure ledprimarily to ret/PTC3 types of rearrangement and a solid growth pattern.The so-called sporadic tumors may have also been induced by radiationbut by inadvertent exposure to lower doses that led to tumor developmentin children with a higher susceptibility to radiation-induced DNA damage, and are characterized primarily by ret/PTCJ types of rearrangementand a typicalpapillarygrowth pattern.Alternatively,only post-Chernobyltumors may be truly radiation induced, and the high prevalence of retrearrangements in the sporadic cases are due to alternative mechanisms of

tumor initiation. Regardless of the manner in which these mutations take

place, this study shows that the different types of ret rearrangementsresult in distinct morphological tumor phenotypes and possibly in cancers

with different biological behaviors.

ACKNOWLEDGMENTS

We are grateful to Dr. Nobuo Satoh (Kanazawa University, Japan) forproviding the TPC-l cell line, to Dr. S. M. Jiang (Ohio State University,Columbus, OH) for providing clones of ret/PTC2 and ret/PTC3 rearrangements, and to Dr. H. Kalkwarf (University of Cincinnati) for advice onstatistical analysis.

REFERENCES

I. Kazakov, V. S., Demidchik, E. P., and Astakhova, L. N. Thyroid cancer afterChemobyl. Nature (Lund.). 359: 21. 1992.

2. Stsjazhko, V. A., Tsyb, A. F., Tronko, N. D., Ssouchkevitch, G., and Baverstock, K. F.Childhood thyroid cancer since accident at Chemobyl. Br. Med. J., 310: 801, 1995.

3. BaIter, M. Chernobyl's thyroid cancer toll. Science (Washington DC), 270: 1758—1759, 1995.

4. Nikiforov, Y., and Gnepp, D. R. Pediatric thyroid cancer after the Chemobyl disaster.Pathomorphologic study of 84 cases (1991—1992) from the Republic of Belarus.Cancer (Phila.). 74: 748—766,1994.

5. Nikiforov, Y. E., Nikiforova, M., Gnepp, D. R., and Fagin, J. A. Prevalence ofmutations of i'as and p53 in thyroid tumors from children exposed to radiation afterChemobyl. Oncogene. 13: 687—693.1996.

6. Williams, E. Pathology and molecular biology of post Chernobyl thyroid carcinoma.J. Endocrinol. Invest., 19: 4, 1996.

7. Fagin, J. A., Matsuo, K., Karmakar, A., Chen, D. L., Tang, S., and Koeffler, H. P.High prevalence of mutations of the p53 gene in poorly differentiated human thyroidcarcinomas. J. Clin. Invest., 91: 179—184,1993.

8. Donghi, R., Longoni, A., Pilotti, S., Michieli, P., Ports, G. P., and Pieroui, M. A.

Gene p53 mutations are restricted to poorly differentiated and undifferentiated carcinomas of the thyroid Gland. J. Clin. Invest., 91: 1753—1760,1993.

9. Fugazzola, L., Pilotti, S., Pinchera, A., Vorontsova, T. V., Mondellini, P.,Bongarzone, I., Greco, A., Astakhova, L., Butti, M. G., Demidchik, E. P., Pacini, F.,and Pierotti, M. A. Oncogenic rearrangements of the RETproto-oncogene in papillarythyroid carcinomas from children exposed to the Chernobyl nuclear accident. CancerRes., 55: 5617—5620,1995.

10. Klugbauer, S., Lengfelder, E., Demidchik, E. P., and Rabes, H. M. High prevalenceof RET rearrangement in thyroid tumors of children from Belarus after the Chernobylreactor accident. Oncogene, I 1: 2459—2461, 1995.

I I. Sugg, S. L., Theng, L., Rosen, I. B., Freeman, J. L., Ezzat, S., and Asa, S. L.ret/P'FC-I, -2, and -3 oncogene rearrangements in human thyroid carcinomas: implications for metastatic potential? J. Clin. Endocrinol. Metab., 8!: 3360—3365,1996.

12. Ito, T., Seyama, T.. Iwamoto, K. S., Hayashi, T., Mizuno, T., Tsuyama, N., Dohi,Nakamura, N., and Akiyama, M. In vitro irradiation is able to cause RET'oncogenerearrangement. Cancer Res., 53: 2940—2943, 1993.

13. Bongarzone. I., Fugazzola, L., Vigneri, P., Mariani, L., Mondellini, P., Pacini, F.,Basolo, F., Pinchera, A., Pilotti, S., and Pieroni, M. A. Age-related activation of thetyrosine kinase receptor protooncogenes RET and NTRKI in papillary thyroid carcinoma. J. Clin. Endocrinol. Metab., 81: 2006—2009, 1996.

14. Nikiforov, Y., Gnepp, D. R., and Fagin. J. A. Thyroid lesions in children andadolescents after the Chemobyl disaster: implications for the study of radiationtumorigenesis. i. Clin. Endocrinol. Metab., 81: 9—14,1996.

15. Redline, R. W., Genest, D. R., and Tycko, B. Detection of enteroviral infection inparaffin-embedded tissue by the RNA polymerase chain reaction technique. Am. J.Clin. Pathol., 96: 568—571, 1991.

16. Thiang, S., Caniso, D. R., Tuttle, R. M., Gomez, R. R., Peele, M. E., and Mazzafem, E LDetectionofthe FFC chimerictranscriptsin humanparaffin-embeddedand frozenthyroidtumors. l'he 75th Annual Meeting of the EndocrineSociety, p. 157, 1993.

17. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acidguanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156—159, 1987.

18. Ishizaka, Y., Ushijima, T., Sugimura, T., and Nagao, M. cDNA: cloning and char

acterization of ret activated in a human papillary thyroid carcinoma cell line. Biochem. Biophys. Ret. Commun.. 168: 402—408.1990.

19. GIants, S. A. Primer of Biostatistics,pp. 121-148. New York: McGraw-Hill.Inc., 1992.20. Vickery, A. L., Jr., Carcangiu, M. L., Johannessen, J. V., and Sobrinho-Simoes, M.

Session I. Papillary carcinoma. Semin. Diagn. Pathol., 2: 90—100, 1985.21. LiVolsi, V. A. Surgical pathology of the thyroid. in: J. L. Bennington (ed)., Major

Problems in Pathology. Philadelphia: W. B. Saunders, 1990.22. Jhiang, S. M., Sagartz, J. E., Tong, Q., Parker-Thomburg, J., Capen, C. C., Cho, J. Y.,

Xing, S., and Ledent, C. Targeted expression of the ret/PTCJ oncogene inducespapillary thyroid carcinomas. Endocrinology, 137: 375—378,1996.

23. Santoro, M., Ghiappetta, G., Cerrato, A., Salvatore, D., Zhang, L., Vecchio, G., andFusco, A. Development of thyroid papillary carcinomas secondary to tissue-specificexpression ofthe RET/PTCI oncogene in transgenic mice. Oncogene, 12: 1821—1826,1996.

24. Bongarzone, I., Monzini. N.. Borrello, G., Carcano, C., Ferraresi, G., Alrighi, E.,Mondellini, P., DelIa Ports, G., and Pierotti, M. A. Molecular characterization of athyroid tumor-specific transfonning sequence fonned by the fusion of ret tyrosinekinase and the regulatory subunit ma of cyclic AMP-dependent protein kinase A.Mol. Cell. Biol., 13: 358—366,1996.

25. Grieco, M., Santoro, M., Berlingieri. M. T.. Melillo, R. M., Donghi, R., Bongarzone,I., Pierotti, M. A., Della Porta, G., Fusco, A., and Vecchio, G. PTC is a novelrearranged form of the ret proto-oncogene and is frequently detected in vivo in humanthyroid papillary carcinomas. Cell, 60: 557—563,1990.

26. Santoro, M., Dathan, N. A., Berlingieri, M. T., Bongarzone, I., Paulin, C., Grieco, M.,Pierotti, M. A., Vecchio, G., and Fusco, A. Molecular characterization of RET/PTC3,a novel rearranged version of the RET proto-oncogene in a human thyroid papillarycarcinomas. Oncogene. 9: 509—516, 1994.

27. Minolem, F., Butti, M. G., Coronelli,S., Miozzo, M., Sozzi, G., Pilotti, S., Tunnacliffe,A., Pierotti, M. A., and Bongarzone, I. The two genes generating RET/PTC3 are localizedin chromosomal band lOqI 1.2. Genes Chromosomes Cancer, 11: 51—57,1994.

28. Lanzi, C., Borrello, M. G., Bongarzone, I., Migliazza, A., Fusco, A., Grieco, M.,Santoro, M., Gambetta, R. A., Zumno, F., Della Ports, G., and Pierotti, M. A.Identification of the product of two oncogenic rearranged forms of the RET protooncogene in papillary thyroid carcinomas. Oncogene, 7: 2189—2194,1992.

29. Fugazzola. L., Pierotti. M. A., Vigano, E., Pacini, F., Vorontsova, T. V., andBongarzone. I. Molecular and biochemical analysis of RET/PTC4,a novel oncogenicrearrangement between RET and ELE! genes, in a post-Chernobyl papillary thyroidcancer. Oncogene, 13: 1093—1097,1996.

1694

Research. on November 22, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

1997;57:1690-1694. Cancer Res   Yuri E. Nikiforov, Jon M. Rowland, Kevin E. Bove, et al.   Thyroid Papillary Carcinomas in ChildrenMorphological Variants of Radiation-induced and Sporadic

Oncogene Rearrangements inretDistinct Pattern of

  Updated version

  http://cancerres.aacrjournals.org/content/57/9/1690

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/57/9/1690To request permission to re-use all or part of this article, use this link

Research. on November 22, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from