479

Analysis of RET Protooncogene Point Mutations Distinguishes Heritable from Nonheritable Medullary Thyroid Carcinomas Paul Komminoth, M.D.,*,t Eva K . Kunz,t Xavier Matias-Guiu, M.D.,$ Olaf Hiort, M.D.,§ Gudrun Chrisfiansen,t Anna Colomer, Ph.D.,$ lurgen Roth, M.D., Ph.D.,* and Philipp U. Heitz, M.D.t

Background. The distinction of sporadic from inher- ited medullary thyroid carcinomas (MTCs) is of clinical importance because of the differences in prognosis, and the need for family screening for genetic counseling re- quired in the latter. Germline mutations in the RET pro- tooncogene are associated with multiple endocrine neo- plasia (MEN) type 2A, familial medullary thyroid carci- noma (FMTC), and MEN type 2B. Somatic point mutations in the same gene have been identified in a subset of spo- radically occurring medullary thyroid carcinomas.

Methods. A nonisotopic polymerase chain reaction- (PCR) based single strand conformation polymorphism (SSCP) analysis and heteroduplex gel electrophoresis method was used to screen DNA extracted from 32 form- aldehyde fixed and paraffin embedded MTC specimens and normal tissue or blood of the same patient for point mutations in RET exons 10,11, and 16. Point mutations were identified by nonisotopic cycle sequencing of PCR- products using an automated DNA-sequencer. Results were compared with the disease phenotype, clinical find- ings, and follow-up.

Results. Six different missense germline mutations

From the Division of *Cell and Molecular Pathology and Depart- ment of tfathology, University of Zurich, Zurich, Switzerland; the Department of $Pathology, Hospital Santa Creu i Sant Pau, Barce- lona, Spain; Department of §Pediatrics, Medical University of Lu- beck, Germany.

Supported in part by a grant of the Julius Muller-Grocholski Cancer Research Foundation, Zurich Switzerland and by FISS 94/ 1748.

The authors thank Parvin Saremaslani for technical assistance, Andrea Wodtke, Dagmar Struve, Anke Zollner, and Dieter Zimmer- mann for technical advice, Rosa Cabezas and colleagues of the De- partments of Medicine, Surgery, and Pediatrics of the University Hos- pital of Zurich, Switzerland for providing clinical data, and Norbert Wey and Hannes Nef for the photographic work.

Address for reprints: Paul Komminoth, M.D., Division of Cell and Molecular Pathology, Department of Pathology, University of Zurich, Schmelzbergstrasse 12, CH-8091 Zurich, Switzerland.

Received March 30, 1995; accepted April 13, 1995.

were identified at cysteine residues 618, 630, and 634 of the cysteine-rich extracellular RET domain encoded by exons 10 and 11 in all patients with FMTC and MEN 2A. The frequency of mutations at codon 634 was higher in patients with MEN 2A than with FMTC and a 634 Cys + Arg mutation was associated with parathyroid disease in three patients. A germline Met + Thr point mutation at codon 918 of the R E T tyrosine kinase domain was identi- fied in all three patients with MEN 2B. Two patients with clinically sporadic MTCs and negative family history ex- hibited a RET germline mutation at codon 634, indicating the presence of an nonpredicted inherited MTC. Further- more, one patient had a 618 Cys + Ser mutation in the tumor and nontumorous thyroid DNA but not in blood DNA, indicating a mosaic mutation affecting thyroid tis- sue but not blood cells. Tumor specific (somatic) Met + Thr point mutations at codon 918 were identified in 5 of 13 sporadic MTCs. The remaining eight sporadic MTCs lacked mutations in all three RET exons tested.

Conclusions. This study demonstrates that [I) the molecular methods are not only suitable to identify asymptomatic individuals at risk for MEN 2A, FMTC, and MEN 2B but also to distinguish heritable from non- heritable MTCs using archival tissue specimens, and (2) that more MTCs than clinically expected are heritable, indicating the need for genetic analysis of all patients with MTC. Cancer 1995; 76:479-89.

Key words: thyroid, molecular pathology, medullary thy- roid carcinoma, multiple endocrine neoplasia type 2 (MEN 2), paraffin embedded material, RET protoonco- gene, carcinogenesis, endocrine cancer syndrome.

Medullary thyroid carcinomas (MTCs) represent 5%- 10% of all thyroid carcinomas. Up to 80% of MTCs arise sporadically, with the remainder occurring in the inher- ited cancer syndromes multiple endocrine neoplasia (MEN) type 2A, familial MTC (FMTC), and MEN type

480 CANCER August 1, 2995, Volume 76, No. 3

2B. The MEN 2A syndrome is characterized by the com- bined occurrence of MTC, pheochromocytoma, and ad- ditional parathyroid hyperplasia in approximately 25% of patients. In FMTC, the only manifestation of the syn- drome appears to be MTCs, which usually develop later than those observed in MEN 2A or 2B. The MEN 2B syndrome is the most severe form of autosomal domi- nant MEN 2 phenotypes. It is characterized by earlier onset of MTC and pheochromocytoma, rapid progres- sion to metastatic MTC, rare involvement of parathy- roids, and occurrence of skeletal and ophthalmic abnor- malities, mucosal neuromas, and intestinal ganglioneu- romatosis. ’

The distinction of sporadically occurring and of MEN 2A- or MEN 2B-associated MTC is of clinical rel- evance because of differences in prognosis and the need for family screening, genetic counseling, and close clin- ical follow-up in the inherited forms. If family history is negative, evidence for the inherited nature of MTC can be obtained by the identification of C-cell hyperplasia in tumor free areas of thyroidectomy specimens.‘ How- ever, there is evidence that C-cell hyperplasia also can arise from causes other than MEN 2.3-A Biochemical tests including calcitonin stimulation by pentagastrin or calcium currently are used routinely to screen for early symptoms of thyroid disease in at-risk patients. Because inherited MTC may develop at a very young age, screening must start in early childhood and be repeated in regular intervals. However, besides being stressful and accompanied by unpleasant side effects, results of biochemical testing may be unreliable as normal basal and stimulated calcitonin levels in children are not well defined.’

By genetic linkage analyses the loci for MEN 2A, MEN 2B, and FMTC have recently been mapped to the centromeric region of chromosome lOql 1.2.’0r” This region encompasses the RET proto~ncogene,~~,’~ which is a receptor-type tyrosine kinase expressed in normal thyroid, adrenal tissue, MTC, and pheochromocy- toma.l4-I7 The RET gene consists of at least 20 ex on^'*"^ and encodes 5 major mRNA species resulting from al- ternative 3’ RNA splicing and polyadenylation sites.20 The RET protooncogene products are expressed as poly- peptides of 1072 or 11 14 amino acids,20 which are com- posed of a cytoplasmic tyrosine kinase domain, a short transmembrane domain, and a large extracellular do- main with a number of highly conserved Ca2+ binding cysteine-residues near the transmembrane region as well as a cadherin-like ligand-binding site.” Both the ligand and the exact function of the RET protein have yet to be identified. However, studies of similar receptor tyrosine kinases indicate that ligand binding results in a cascade of events that include receptor dimerization,

autophosphorylation, and tyrosine phosphorylation of substrate molecules.22

Recently, point mutations in the RET protoonco- gene have been associated with MEN 2A, FMTC, MEN 2B syndrome, and a subset of sporadic In patients with MEN 2A and FMTC, germline point mu- tations were found to be clustered in five cysteine resi- dues of the cysteine-rich extracellular domain, which is encoded by exons 10 and 11. In contrast, a single type of germline point mutation at codon 918 of the tyrosine kinase domain encoded by exon 16 has been identified in patients with MEN 2B. An identical but somatic mu- tation has been found in a subset of sporadically occur- ring MTCS.’~,~~

Recently, we demonstrated that DNA extracted from routinely processed surgical specimens is suitable for the detection of RET point mutations by nonisotopic single strand conformation polymorphism (SSCP) analysis and direct sequencing.27 In the present study, we analyzed RET protooncogene point mutations in DNA extracted from formalin fixed, paraffin embedded specimens of 16 familial and 16 clinically sporadic MTCs and compared the results with the disease phenotype, clinical findings, and follow-up. In addition, we demonstrate an improved screening method for point mutations in the RET exon 16 by heteroduplex gel electrophoresis.

Materials and Methods

Patients Samples

Tumor material of 32 patients with MTC was studied (Table 1). In 19 patients, nontumorous tissue samples or peripheral blood were also investigated. Paraffin blocks from 1983 to 1994 were drawn from the files of the De- partments of Pathology, University of Zurich, Switzer- land, and the Autonomous University of Barcelona, Spain. Tissues had been fixed by immersion in 4% formaldehyde and embedded in paraffin according to standard procedures. Clinical files of 31 patients were available. A summary of patients’ data, clinical findings, therapy, and clinical course are listed in Table 1. Tumors studied included 9 MEN 2A-, 3 MEN 2B-associated, 4 FMTC, and 16 clinically sporadic MTCs.

DNA Extraction

Genomic DNA was isolated from three 10 pm thick sec- tions prepared from each paraffin block as described p r e v i ~ u s l y . ~ ~ In brief, tissue sections were deparaffin- ized twice with xylene at 5OoC and washed in absolute ethanol, Dried samples were treated with 500 pg/ml proteinase K (Sigma P-4914) in 500 PI digestion buffer (10 mM Tris, pH 8.0, 1 mM ethylenediamine tetraacetic

RET Mutations in Medullary Thyroid Carcinoma/Kornminoth et al. 481

Table 1. Patient Data ~

Patient Age Clinical Tumor Additional no. (yrs) Sex diagnosis Stage localization Symptoms findings Therapy Follow-up

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

34 19 31 NK 68 47 45 68 53 39 26 48 33 12 8

34 64 50 47 35 45 48 45 59 45 36 67 68 68 45 46 58

F M F F F F M F F M M F F M M M F M M F F M M M F M F F M M F M

FMTC FMTC FMTC FMTC MEN 2A* MEN 2A* MEN 2A* MEN 2A* MEN 2A* MEN 2A MEN 2 A t MEN 2At MEN 2A MEN 2B MEN 28 MEN 28 spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC mar. MTC

I I I1 NK I1 I1 I1 111 I1 I I I I I1 I 1 I1 I1 11 111 I11 I11 I1 111 I1 111 111 I1 111 111 I11 111

U U €3 NK U B B B B B B U B B B B U U U U U U U U U U U U U U U U

CT CT CT NK CT CT CT CT CT CT, PTH CP, PTH CT, PTH CT CT CT CT CT CT CT CT CT CT CT CT CT CT CT CT CT CT CT CT

s S S; NUC; RT NK s S; NUC s s S; NUC S; RT s s S s S S; NUC; RT s S S S; RT; CHT S; NUC S s s s S s S S S; NUC S; NUC s

WA NK AD NK DOC WA WA WA WA WA WA WA WA AD AD AD WA WA AD DD DD NK DOC NK NK NK AD WA AD DD WA AD

NK: Not known; FMTC; familial medullary thyroid carcinoma; MEN: multiple endocrine neoplasia; spor. MTC: sporadic medullary thyroid carcinoma; I: localized tu thyroid, <1 cm; 11: localized to thyroid, >1 cm; 111: with lymph node metastases; IV: with distant metastases; U: unilateral; B: bilateral; C T elevated calcitonin levels (after stimulation); PTH: hyperparathyroidism; PCC: pheochromocytoma; GNM: ganglioneuromatosis; S: surgery; RT: radiotherapy; NUC: nuclear medicine; CHT: chemotherapy; WA: well alive; AD: alive with disease; DOC: dead of other causes; DD: dead of MTC. *Related patients. t Related patients.

acid, 75 mM sodium chloride, and 1% sodium dodecyl sulfate) at 5OoC overnight. DNA was purified by phe- nol-chloroform extraction and quantified by spectro- photometry.” Further, DNA from peripheral blood leu- kocytes was extracted by standard procedures, Blood- derived DNA of healthy persons was used as negative controls. DNA available from our previous studies on patients carrying RET point mutations in exons 10 and 11 served as positive control^.'^ Nonisotopic Polymerase Chain Reaction-Single Strand Conformation Polymorphism and Heteroduplex Gel Electrophoresis Oligonucleotide primers (Table 2) flanking exons 10, 16 and parts of exon 11 of the RET protoonco-

were synthesized by the phosphoramid- ite method using a 392 DNA synthesizer (Applied Bio- systems, Inc., Foster City, CA) and purified according to standard procedures.” Polymerase chain reaction frag- ments ranging from 132 base pairs to 192 base pairs in length were amplified from 100-300 ng tumor or germ- line DNA in a 50 pl mixture containing 0.2 mmol dATP, dTTP, dGTP, dCTP, 20 pmol of each sense and anti- sense primer, 1.5-2.5 mmol magnesium (Table 2) , 20 mmol Tris pH 8.3,50 mmol potassium chloride, 50 p g / ml bovine serum albumin and 1 Unit Taq DNA poly- merase (Boehringer Mannheim Corp., Mannheim, Ger- many). Polymerase chain reaction amplification reac- tions were performed using HotStart 50’” reaction tubes (Catalys AG, Wallisellen, Switzerland) and a program-

gene12, 13,18.19,23

482 CANCER August 2,1995, Volume 76, No. 3

Table 2. Polymerase Chain Reaction Primers and Conditions Exon

10 10 11 11 16 16 16N 16N

Orientation Primer sequence PCR product Mg2+ (mM) Ann t ("C) Glycerol* (%)

sense 5'-CTC AGGGGGCAGCATTGTT-3' antisense 5'-CACTCACCCTGGATGTCTT-3' 132 bp 1.5 58 10 t

antisense 5'-TGTGGGCAAACTTGTGGTAGCA-3' 155 bp 2.5 69 5 t

antisense 5'-TAACCTCCACCCCAAGAG-3' 192 bp 1.5 60 0 $

sense 5'TCTCTGCGGTGCCAAGCCTC-3'

sense 5'-AGGGATAGGGCCTGGGCTT-3'

sense 5'-AGAGTTAGAGTAACTTCAATGTC-3' an tisense 5'-TAACCTCCACCCCAAGAGA-3' 151 bv 2.0 55 0 f

~~ ~ ~~~~~~

PCR: polymerase chain reaction; bp: base pairs; Mg2+: Mg2+ concentration in the polymerase chain reaction buffer; Ann t: annealing temperature. * Glycerol concentration in the gel. t Single strand conformation polymorphism-analysis. $ Heteroduplex method.

mable thermal cycler (Perkin Elmer, Norwalk, CT). Af- ter initial denaturation at 94OC for 300 seconds, 35 cy- cles of denaturation for 75 seconds at 94OC, annealing for 90 seconds at 55-69OC (Table 2), and extension for 120 seconds at 72OC were used, followed by a final ex- tension at 72OC for 300 seconds.

For the SSCP analysis of RET exons 10 and 11, ten pl of PCR products were 1:1 diluted in stop buffer (95% formamide, 20 mM EDTA, 0.05% xylene cyanol, 0.05% bromophenol blue), heat denatured at 96OC for 10 min- utes and quickly chilled in a dry ice/alcohol bath before loading onto nondenaturing 0.8 mm thick 6% poly- acrylamide gels (29: 1 acry1amide:bisacrylamide; Bio- Rad, Glattbrugg, Switzerland) containing 5% or 10% glycerol29 (Table 2). Electrophoresis was performed us- ing a sequencing gel electrophoresis apparatus (Gibco BRL, Life Technologies; Zurich, Switzerland) at 35 Watts for 6 hours with cooling by a fan at room temper- ature.

For the heteroduplex method, PCR products of RET exon 16 were denatured at 96OC for 2 minutes and then slowly cooled to room temperature (30 minutes) to al- low heteroduplex formation. Ten p1 of PCR products were 1:l diluted in stop buffer and directly loaded onto nondenaturing polyacrylamide gels (29: 1 acrylamide: bisacrylamide; BioRad, Glattbrugg, Switzerland) con- taining no glycerol. Electrophoresis was performed at 40 Watts for 4 hours with cooling by a fan at room tem- perature.

After electrophoresis the DNA was visualized by silver staining as recently described.27r30 In brief, gels were fixed for 10 minutes in 10% ethanol, treated for 3 minutes with 1 YO nitric acid, washed twice for 2 minutes in double distilled water, and incubated for 20-30 min- utes in a 0.2% silver nitrate solution (Merck, Zurich, Switzerland). After a brief wash in double distilled wa- ter, gels were developed in a 0.28M sodium carbonate (Merck, Zurich, Switzerland) solution containing 0.02%

formaldehyde, rinsed for 2 minutes in 10% acetic acid, followed by a wash for 5 minutes in 50 mM ethylenedi- amine tetraacetic acid and two washes for 5 minutes each in double distilled water.

Nonisotopic Direct Sequencing of Polymerase Chain Reaction Products

Abnormal bands from PCR-SSCP analysis were excised from additionally prepared SSCP polyacrylamide gels, stained with Sybr" Green I nucleic acid stain (Molecular probes, Inc., Eugene, OR), placed in 100 pl of 1X Tris ethylenediamine tetraacetic acid buffer, pH 8.0, and in- cubated for 10 minutes at 6OoC to elute DNA. An ali- quot (3-5 pl) of the supernatant was used as PCR tem- plate for 35 further PCR cycles as detailed above to yield PCR-products predominantly harboring the mu- tated appropriate RET sequence. Forty microliters of re- amplified DNA sequences and PCR-products showing heteroduplex formation in the electrophoresis assay were agarose gel-purified using the "freeze-squeeze" method,31 alcohol precipitated after adding 20 pg of gly- cogen (Boehringer Mannheim Corp.) and resuspended in 15 pl of lOmM Na+-free Tris EDTA buffer, pH 8.0. The DNA concentration of purified PCR-products was estimated by comparing the band intensities of 1 pl sample DNA and the quantified DNA molecular weight-marker phiXl74, Hue I11 (Boehringer Mannheim Corp.) in an ethidium bromide stained agarose gel elec- trophoresis.

DNA sequences of 30 ng PCR-products were deter- mined in sense and antisense direction by fluorescence- based dideoxy terminator cycle sequencingz7r32 using the TaqD yeDeoxy'" Terminator Cycle Sequencing kit (Applied Biosystems, Weiterstadt, Germany) followed by gel electrophoresis, data collection, and analysis on an automated DNA-sequencer (Model 373A, Applied Biosystems Inc.).

RET Mutations in Medullary Thyroid Carcinoma/Komminoth et al. 483

Restriction Analysis

Ten p1 of RET exon 16 PCR-products were digested in of 50 pl of medium salt buffer containing 5 U of the restriction enzyme Fok I (Boehringer Mannheim Corp.) overnight at 37OC. Restriction fragments were analyzed by 6% polyacrylamide gel electrophoresis and silver staining.

Results

Familial Medullary Thyroid Carcinoma and Multiple Endocrine Neoplasia Type 2A-Associated Medullary Thyroid Carcinoma

We studied the sequence of RET exons 10, 11, and 16 in tumor DNA of nine individuals with MEN 2A-associ- ated MTCs and four patients with clinically confirmed FMTC. In 11 of these patients, germline DNA of either normal tissue or peripheral blood was also investigated. In all patients heterozygous germline point mutations were found and the mutations were identical iru related patients.

Polymerase chain reaction-single strand conforma- tion polymorphism analysis of RET exon 10 revealed one type of single strand conformation variant (SSCV) in 2 patients and analysis of exon 11 five different types of SSCVs in the remaining 11 patients. The hetero- duplex assays for exon 16 showed normal patterns in all patients. Examples of results obtained by nonra- dioactive PCR-SSCP analysis are shown in Figures 1 and 2.

Sequence analysis of PCR-products was performed using a cycle-sequencing procedure in which each of the four dideoxy-terminators was labeled with a different fluorochrome. This approach permits the anal- ysis of a PCR product sequence in a single reaction tube and in a single lane of the sequencing gel. To improve the detection of heterozygous point mutations, SSCV bands harboring the point mutations were isolated from SSCP gels and used as templates for an additional PCR amplification before cycle sequencing. In accordance with the SSCP results, we found six types of missense point mutations located in three different Cys residues of RET exon 10 (618 Cys + Ser) and exon 11 (630 Cys + Phe; 634 Cys + Tyr, Arg, Ser, Phe). The results of the sequence analysis are summarized in Table 3 and specific examples are shown in Figures 1 and 2. Point mutations were located at codon 634 (codon numbers derived in 10 patients, at codon 630 in 1 pa- tient, and at codon 618 in 2 patients. Three types of point mutations at codon 634 were identified in patients suffering from additional pheochromocytoma (Cys .+ Tyr, Arg, or Phe). Three patients with pheochromocy-

Exon 10

N codon 618 Figure 1. Nonradioactive polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) (left) and sequence analysis (right) of RET exon 10. Polymerase chain reaction-single strand conformation polymorphism analysis was performed using a 6% polyacrylamide gel (29:l) containing 10% glycerol at room temperature and DNA was visualized by silver staining. Arrowheads denote bands with altered migration relative to the normal control (N). Bands with altered migration in the PCR-SSCP were excised, reamplified, and sequenced. Nucleotide sequence analysis was performed using fluorescence-based dideoxynucleotide cycle sequencing and an automated DNA sequencer. The mutated base is indicated by a circle. The diagram shows a TGC -* TCC transversion causing a Cys Ser substitution at codon 618 (shadowed area).

toma and additional symptoms of hyperparathyroidism had an identical point mutation at codon 634 (TGC + CGC; Cys + Arg), which was not seen in other patients with hereditary MTC. Of four patients with FMTC, only one showed a mutation at codon 634, the others at co- don 618 and 630, respectively (Table 2).

Multiple Endocrine Neoplasia Type 2B-Associated Medullary Thyroid Carcinoma

All three patients with MEN 2B showed heteroduplex- formation in the gel electrophoresis after examination of RET exon 16 PCR-products of tumor and germline DNA. Cycle sequencing in sense and antisense direc- tion revealed a single type of missense point mutation at codon 918 (ATG + ACG) causing a substitution of a threonine for a methionine. As the mutation eliminates a Fok I restriction site, digestion of the PCR-products of

484 CANCER August I , 1995, Volume 76, No, 3

Exon 11

N TAC CGC TTC TCC l T C

Tyr Arg Phe Ser Phe

codon 634 630

codon 630 codon 634

Figure 2. Nonradioactive polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) (top left) and sequence analysis of RET exon 11. Polymerase chain reaction-single strand conformation polymorphism analysis was performed using a 6% polyacrylamide gel (29:l) containing 5% glycerol at room temperature and DNA was visualized by silver staining. Arrowheads denote bands with altered migration relative to the normal control (N). Note the 5 different types of distinct band patterns in the PCR- SSCP analysis, which corresponded to the types of point mutations at codons 630 and 634 (shadowed areas) found in the nucleotide sequence analysis (bottom left and right).

exon 16 with this restriction enzyme was used to con- firm the presence of the mutation. Indeed, the mutated DNA was resistant to Fok I digestion in all patients. Sin- gle strand conformation polymorphism analysis of exon 10 and 11 PCR-products yielded normal band patterns in all three patients with MEN 2B (results not shown).

Clinically Sporadic Medullary Thyroid Carcinoma

Sixteen tumor samples of clinically sporadic MTCs were screened for the presence of mutations in RET exons 10, 11, and 16. At the time of diagnosis, most patients pre-

sented with an advanced tumor stage with lymph node metastases in contrast to patients with inherited MTCs. In two tumor samples, we identified a SSCV in the PCR- SSCP analysis of exon 11. Identical band shifts were detected when blood or nontumorous tissue derived DNA of these patients was analyzed, indicating the presence of a germline mutation with a heritable, in- stead of a sporadic, nonheritable MTC. Sequence anal- ysis revealed a missense germline point mutation at co- don 634 (TGC + TAC) consisting of a substitution of a tyrosine for a cysteine (see “clinical” and “molecular diagnosis” in Table 3). Cross-contamination of tumor and normal-DNA was prevented by using precautions recommended by Kwok and H i g ~ c h i ~ ~ and results were consistently reproducible. The presence of tumor me- tastases in non-MTC tissue was excluded by histologic examination of samples before DNA extraction. Both patients had a negative family history and we were un- able to contact possible descendants.

By examination of RET exon 10 PCR-products, one patient exhibited SSCVs in tumor DNA and DNA ex- tracted from nontumorous thyroid tissue but not in blood-derived DNA. Nucleotide sequence analysis re- vealed a 618 Cys + Ser point mutation in tumor and nontumorous thyroid tissue.

Heteroduplex gel electrophoresis of exon 16 PCR- products yielded additional bands with slightly re- tp-ded migration in 5 tumor samples that were not de- tectable by analyzing germline DNA available from 4 of these patients. The presence of tumor specific, somatic point mutations in these 5 MTCs was confirmed by sense and antisense sequencing and additional Fok I re- striction analysis (Fig. 3).

Heteroduplex Gel Electrophoresis Is Superior to Polymerase Chain Reaction-Single Strand Conformation Polymorphism Analysis of RET-Exon 16

Polymerase chain reaction-single strand conformation polymorphism analysis of RET exon 16 using PCR- primers as described by Hofstra et al.” (“16” in Table 2) yielded SSCVs in two cases of sporadic MTC and in one MEN 2A-associated tumor DNA (not shown). Sense and antisense sequencing, however, revealed an intronic polymorphism (CCT ---* CTT) as a possible cause of these SSCVs. Newly designed primers (desig- nated ”16N,” see Table 2) which were located closer to the intron-exon junction were used to exclude possible polymorphic intronic sequences in the PCR-products. When applied, no SSCVs were evident in all 32 samples tested even when SSCP-conditions were varied by al- tering the percentage and extent of cross-linking of the acrylamide, by adding various concentrations of glyc-

RET Mutations in Medullary Thyroid Carcinoma/Komminoth et al. 485

Table 3. Results Patient Clinical Type of RET Base Amino acid no. diagnosis mutation exon Codon* exchange substitution Molecular diagnosis

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

FMTC FMTC FMTC FMTC MEN 2AS MEN 2A$ MEN 2AS MEN 2AS MEN 2AS MEN 2A MEN 2A5 MEN 2A5 MEN 2A MEN 28 MEN 2B MEN 28 spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC spor. MTC

G t 10 G t 10 G t 11 G t 11 G t 11 G I 1 G 11 G 11 G 11 G t 11 G t 11

G t 11 G I1

G t 16 G t 16 G t 16

10 _-

II t - - -

16

11

- st

G t

-

__ -

- - - - - __ - - G t 11 st 16 st 16 st 16

32 spor. MTC st 16

FMTC: familial medullary thyroid carcinoma; MEN 2A: multiple ehdocrine neoplasia type 2A; MEN 2B: multiple endocrine neoplasia type 23; spor MTC: sporadic medullary thyroid carcinoma; G: germline mutation; S : somatic mutation. * Codon numbers according to refs. 12,13. t Additional analysis of normal tissue or blood DNA. $ Related patients. 5 Related patients. 11 Probably postygotic mutation affecting thyroid but not blood cells.

618 618 634 630 634 634 634 634 634 634 634 634 634 918 918 918 618 - -

918

634

-

- - - - - 634 918 918 918 918

TGC + TCC TGC + TCC TGC + TCC TGC + TTC TGC + TAC TGC + TAC TGC + TAC TGC + TAC TGC + TAC TGC + CGC TGC + CGC TGC + CGC TGC + TTC ATG + ACG ATG + ACG ATG -z ACG TGC + TCC

-

- ATG + ACG

-

TGC + TAC - - - - -

TGC + TAC ATG + ACG ATG + ACG ATG + ACG ATG + ACG

Cys + Ser Cys + Ser Cys + Ser Cys + Phe Cys + Tyr Cys + Tyr Cys + Tyr Cys + Tyr Cys + Tyr Cys + Arg Cys + Arg Cys + Arg Cys + Phe Met + Thr Met + Thr Met + Thr Cys + Ser - -

Met + Thr

Cys + Tyr -

-

- - - -

Cys + Tyr Met + Thr Met + Thr Met + Thr Met + Thr

hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC hereditary MTC nonheritable MTC? nonheritable MTC nonheritable MTC nonheritable MTC nonheritable MTC heriditary MTC nonheritable MTC nonheritable MTC nonheritable MTC nonheritable MTC nonheritable MTC heriditary MTC nonheritable MTC nonheritabIe MTC nonheritable MTC nonheritable MTC

erol, employing different loading buffers and by run- ning the gel electrophoresis in a cold room or overnight. This inability to identify band shifts by SSCP was prob- ably due to a long stable stem of secondary structure or loop formation of the exon 16 PCR-product sequence. However, the presence of an additional band with slightly retarded migration (when compared with dou- ble stranded PCR-products) in the SSCP gel electropho- resis of three patients with known MEN 2B provided evidence that heteroduplex formation might have oc- c ~ r r e d . ~ ~ This assumption was supported by results of the sequence analyses of excised bands which revealed the presence of wild type and mutated (ATG + ACG,

918Met --* Thr) exon 16 sequences in the same band. After modification of protocols as outlined in ”Materials and Methods,” formation of heteroduplex bands was consistently found in germline DNA of MEN 2B pa- tients as well as in tumor DNA of five sporadic MTCs.

Discussion

Our results emphasize the importance of molecular ge- netic analysis of all patients with MTC to detect clini- cally unsuspected gene carriers of MEN 2 and thus to distinguish unequivocally hereditary from nonheritable MTCs. Of 16 patients with clinically sporadic MTCs, we

486 CANCER August 1,2995, Volume 76, No. 3

Exon 16 Fok I

4

codon 918 (sense) codon 918 (antisense) Figure 3. Heteroduplex gel electrophoresis (top left) Fok I restriction analysis (top right) and sequence analysis of RET exon 16 (bottom). The heteroduplex analysis was performed using a 6% polyacrylamide gel (29:l) at room temperature and DNA was visualized by silver staining. Tumor DNA (T) of sample 2 exhibited heteroduplex formation (arrowhead), which was not present in normal DNA (N), the sample I, and the control (C). PCR-products were digested with Fok I restriction enzyme and the restriction fragments analyzed by 6% polyacrylamide gel electrophoresis and silver staining. The ATG + ACG point mutation at codon 918 removes a Fok I restriction site. Note the nondigested mutated sequence (arrowhead) of the tumor DNA PCR-product (T) in comparison to the normal DNA sequence (N) and nondigested control (C). Nucleotide sequence analysis in sense and antisense direction revealed a heterozygous (arrows) ATG + ACG point mutation at codon 918 (mutated bases are indicated by circles).

identified two harboring a point mutation (TGC + TAC; Cys --+ Tyr) at RET protooncogene codon 634 of exon 11, not only in tumor but also in nontumorous tis- sue and blood, indicating the presence of germline mu- tations. These mutations may have arisen de novo, as family history in both of our patients was negative. Our series of 16 investigated clinically sporadic MTCs is to small to determine the percentage of de novo germline mutations causing a MEN 2A or FMTC syndrome. However, our findings indicate that an important mi-

nority of patients with MTC with negative family his- tory may indeed harbor clinically unsuspected germline mutations and are therefore at risk for heritable MTC.

Furthermore, we identified a 618 Cys + Ser point mutation in tumor and nontumorous thyroid DNA of a patient with clinically sporadic MTC. Because the pres- ence of tumor foci in the nontumorous thyroid was ex- cluded by histologic examination and calcitonin immu- nohistochemistry, it is likely that the MTC is due to a postzygotic de novo mutation that affects thyroid tissue but not blood cells. Similar mosaic mutations have re- cently been described in other diseases such as McCune-Albright syndrome.35

In all 16 inherited MTCs comprised of 4 FMTC, 9 MEN 2A-associated and 3 MEN 2B associated MTCs, we found RET germline point mutations, which were also detectable in nontumorous tissue or blood-derived DNA available from 11 patients and were identical in patients from the same family. Point mutations in 13 FMTC and MEN 2A patients of 11 unrelated families involved the three highly conserved cysteine residues 618, 630, and 634,l2,I3 which are encoded by exons 10 and 11, respectively. Ten of 14 mutations (72%) were found at codon 634 consisting of four different base substitutions, resulting in the exchange of a Cys residue by either Ser, Tyr, Arg, or Phe. Two of the identified point mutations involved the Cys residue 618 resulting in a substitution of Ser for Cys and one point mutation was found at Cys codon 630 consisting of a TGC + TTC base exchange resulting in the substitution of Phe for Cys. Our findings agree with recent reports about such mutations in constitutional DNA of the majority of patients with MEN 2A and FMTC but in none of more than 100 control patient^.^^'^^'^^ Alterations have been reported in all 4 Cys residues (609, 611, 618, and 620) encoded by exon 10 and in Cys residue 634 of exon 11. Carlson et a1.22 have recently reported a 6 base pair deletion in DNA isolated from a sporadic MTC that in- volved the Cys residue at codon 630. In the present study, we provide evidence that Cys codon 630 is also involved in heritable MTC because we identified a germline point mutation at this Cys residue in a patient with FMTC. Thus, mutations have been found in all six Cys residues that are located immediately adjacent to the transmembrane domain of the RET protooncogene. These Cys residues possibly play an important role in maintaining the secondary and tertiary structure of the extracellular domain. Experimental data on other tyro- sine kinases indicate that Cys-rich regions fold up to form a ligand-binding pocket. Therefore, mutations re- sulting in the replacement of Cys by other amino acids diminish or abolish the regulatory effect of ligand-bind- ing on the activity of the tyrosine kinase and result in

RET Mutations in Medullary Thyroid Carcinoma/Komminoth et al . 487

constitutive autophosphorylation in the absence of li- gand that leads to a transformed phenotype.”

The diversity of tissue involvement and disease phenotype associated with missense mutations within the same region of RET suggests that subtle difierences in amino acid sequences of the extracellular domain have different consequences for RET function or expres- sion depending on the cell type. In this study, we pro- vide evidence for such a hypothesis because all studied patients with clinically confirmed MEN 2A syndrome showed point mutations involving codon 634, includ- ing six patients with additional pheochromacytoma. This is in contrast to patients with FMTC who elthibited significantly more mutations at Cys codons other than 634. Furthermore, all three patients with MEN 2A who underwent parathyroid surgery due to hyperparathy- roidism had the same point mutation at codon 434 con- sisting of a TGC + CGC base exchange resulting in a substitution of Arg for Cys. These findings, which are consistent with those of recent reports,37 strongly indi- cate that any type of mutation at RET codon 634 is pre- dictive for pheochromocytoma whereas a specific TGC + CGC mutation at codon 634 for parathyroid disease. However, the presence of identical mutations among patients exhibiting different phenotypic features re- mains to be explained. For example, among six kindreds with the same 634 Cys + Tyr mutation only two devel- oped pheochromocytoma at the age of 45 years and 47 years, respectively, whereas the other older family members did not (Table 3). Furthermore, Ceccherini et al.3s recently reported a pedlgree of patients with MEN 2A and additional localized cutaneous lichen amy- loidosis, who showed the same type of mutation as one of the MEN 2A family studied by us (Table 3). Prelimi- nary studies suggest that no specific RET allele or hap- lotype harboring a mutation is associated with a partic- ular phen~type.’~ Differences in RET allele expression as a possible explanation have also been excluded be- cause mRNA of both the normal and mutant alleles have been detected in tumor tissue of patients with MEN 2A.3Y Another explanation has recently been given by van He~ningen,~’ who proposed that RET mu- tations found in MEN 2A, FMTC, and MEN 2B act dominantly, leading to hyperplasia in cells that nor- mally express RET and that malignancy arises in these hyperproliferative cells through mutations in other genes. This hypothesis, which remains to be confirmed, would easily explain the different phenotypes found in patients with identical RET mutations.

Germline DNA of all three patients with MEN 2B and tumor DNA of five sporadic MTCs (38%) demon- strated a single point mutation (ATG + ACGJ at codon 918. This T --* C transition results in the substitution of a threonine for a methionine residue in the catalytic core

region of the tyrosine kinase d ~ m a i n . * ~ , ~ ~ The methio- nine at codon 918 is highly conserved among other re- ceptor-type tyrosine kinases and this position is occu- pied by a threonine in nonreceptor type (cytoplasmic) tyrosine k i n a s e ~ . ~ ~ Thus, mutations at position 918 of RET appear to be crucial for function, probably resulting in increased enzyme activity and/or altered substrate specificity of RET phosphorylation.” Our findings are consistent with recent reports of such mutations in the germline DNA of patients with MEN 2B, which appear to occur de novo in approximately 23% of patients.22T26

Furthermore, our finding of the same somatic mu- tation in 5 of 12 (38%) sporadic MTCs and 1 of 7 (14%) sporadic pheochromocytoma (unpublished observa- tion) provides additional evidence for the role of RET point mutations in sporadically occurring MTCs and pheochromocytomas.z5.26 However, no difference of tu- mor onset or survival was evident in sporadic tumors with or without a RET mutation. No mutations were found in the remaining eight sporadic MTCs. This indi- cates that other yet to be identified genes may be in- volved in the formation of MTC. However, it cannot be excluded that RET mutations in other parts of the gene than those examined may play a role as well.

The results of our study show that RET protoonco- gene point mutations can be detected in archival tumor specimens of pathology files. The applied techniques not only permit the discrimination of hereditary from nonheritable MTCs but also to identify nonsymptom- atic gene carriers for MEN 2. They therefore may re- place biochemical and linkage-based screening tests. Because most of the RET mutations either create or re- move an restriction enzyme site, the detection of family members at risk is facilitated by using a fast and simple test based on PCR amplification followed by restriction d i g e s t i ~ n . ~ ~ , ~ ’ - ~ ~ Currently, the indication for thyroidec- tomy in gene carriers is controversial. Further clinical studies will reveal if preventative thyroidectomy at young age or surgery only after biochemical screening tests have become positive is the therapy of c h o i ~ e . ~ ~ , ~ ’

Note added in proof

Since the submission of our manuscript, Eng et al. have published a study on RET mutations in 71 sporadic medullary thyroid carcinomas (Genes, Chromosomes & Cancer 1995; 12:209-12). Their findings are in basic agreement with our data, although percentages and lo- cations of mutations are somewhat different. They identified RET mutations at codon 918 of exon 16 in 23%, at codons 611 and 620 of exon 10 in 3% (2/67) and none in exon 11. One of the exon 10 mutations proved to be a germline mutation since it was also pres- ent in available constitutional DNA.

488 CANCER August 2, 2995, Volume 76, No. 3

References Structural analysis of the human ret proto-oncogene using exon

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Murray D. The thyroid gland. In: K. Kovacs, S. Asa, editors. Functional endocrine pathology. Boston: Blackwell Scientific Publishers, 1991. Komminoth P, Roth J, Saremaslani P, Matias-Guiu X, Wolfe HI, Heitz PU. Polysialic acid of the neural cell adhesion molecule in the human thyroid: a marker for medullary thyroid carcinoma and primary C-cell hyperplasia. Am J Surg Pathol 1994; 18:399- 411. Albores-Saavedra J, Monforte H, Nadji M, Morales A. C-cell hy- perplasia in thyroid tissue adjacent to follicular cell tumors. Hum Pathol 1988; 19:795-9. Scopsi L, Di PS, Ferrari C, Holst JJ, Rehfeld IF, Rilke F. C-cell hyperplasia accompanying thyroid diseases other than medul- lary carcinoma: an immunocytochemical study by means of an- tibodies to calcitonin and somatostatin. ModPotho[ 1991;4:297- 304. Barbot N, Guyetant S, Beldent V, Akrass A, Cerf I, Perdrisot R, et al. Chronic autoimmune thyroiditis and C-cell hyperplasia: study of calcitonin secretion in 24 patients. Ann Endocrinol (Paris)

Biddinger PW, Brennan MF, Rosen PP. Symptomatic C-cell hy- perplasia associated with chronic lymphocytic thyroiditis. A m J Surg Pathol 1991; 15:599-604. Gibson W, Peng T-C, Croker B. Age associated C-cell hyperpla- sia in the human thyroid. A m JPuthol 1982; 106:388-93. Tomita T, Millard D. C-cell hyperplasia in secondary hyperpara- thyroidism. Histopathology 1992; 21~469-74. Gagel RF. Multiple endocrine neoplasia. In: Wilson JD, Foster DW. Williams textbook of endocrinology. Philadelphia: W.B. Saunders, 1992. Gardner E, Papi L, Easton DF, Cummings T, Jackson CE, Kaplan M, et al. Genetic linkage studies map the multiple endocrine neoplasia type 2 loci to a small interval on chromosome 10q11.2. Hum Mol Genet 1993;2:241-6. Lainnore TC, Howe JR, Korte JA, Dilley WG, Aine L, Aine E, et al. Familial medullary thyroid carcinoma and multiple endocrine neoplasia type 20 map to the same region of chromosome 10 as multiple endocrine neoplasia type 2A. Genomics 1991;9:181-92. Takahashi M, Buma Y, Iwamoto T, lnaguma Y, Ikeda H, Hiai H. Cloning and expression of the ret proto-oncogene encoding a tyrosine kinase with two potential transmembrane domains. On- cogene 1988;3:571-8. Takahashi M, Buma Y, Hiai H. Isolation of ret proto-oncogene cDNA with an amino-terminal signal sequence. Oncogene 1989; 45305-6. Miya A, Yamamoto M, Morimoto H, Tanaka N, Shin E, Kara- kawa K, et al. Expression of the ret proto-oncogene in human medullary thyroid carcinomas and pheochromocytomas of MEN 2A. Henry Ford Hosp Med 1992;40:215-9. Santoro M, Rosati R, Grieco M, Berlingieri MT, D’Amato GL, de Franciscis V, et al. The ret proto-oncogene is consistently ex- pressed in human pheochromocytomas and thyroid medullary carcinomas. Oncogene 1990;5:1595-8. Matias-Guiu X , Colomer A, Mato R, Cuatrecasas M, Komminoth P, Prat J, et al. Expression of the ret proto-oncogenein pheochro- mocytoma. An in-situ hybridization and northern blot study. ] Pathol 1995, in press. Matias-Guiu X, Komminoth P, Ricklan D, Wolfe HI. Expression of the ret proto-oncogene in human medullary thyroid carci- noma: an in-situ hybridization study [abstract]. Lab Invest 1993;68:39a. Kwok JB, Gardner E, Warner JP, Ponder BA, Mulligan LM.

1991;52:109-12.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

trapping. Oncogene 1993; 8:2575-82.- Ceccherini I, Bocciardi R, Luo Y, Pasini B, Hofstra R, Takahashi M, et al. Exon structure and flanking intronic sequences of the human RET proto-oncogene. Biochem Biophys Res Commun 1993; 196:1288-95. Tahira T, Ishizaka Y, Itoh F, Sugimura T, Nagao M. Character- ization of ret proto-oncogene mRNAs encoding two isoforms of the protein product in a human neuroblastoma cell line. Onco- gene 1990;5:97-102. Kuma K , Iwabe N, Miyata T. Motifs of cadherin- and fibronectin type 111-related sequences and evolution of the receptor-type- protein tyrosine kinases: sequence similarity between proto-on- cogene ret and cadherin family. Mol Biol Evol 1993; 10:539-51. Carlson KM, Dou S, Chi D, Scavarda N, Toshima K, Jackson CE, et al. Single missense mutation in the tyrosine kinase catalytic domain of the ret protooncogene is associated with multiple en- docrine neoplasia type 28. Proc Nuti Acud Sci USA 1994;91:

Donis-Keller H, Dou S, Chi D, Carlson KM, Toshima K, Lairmore TC, et al. Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hun2 Mol Genet 1993;2:851-6. Mulligan LM, Kwok JB, Healey CS, Elsdon MJ, Eng C, Gardner E, et al. Germline mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993; 363:458-60. Hofstra RMW, Landsvater RM, Ceccherini I, Stulp RP, Stelwa- gen T, Luo Y, et al. A mutation in the RET proto-oncogene asso- ciated with multiple endocrine neoplasia type 2B and sporadic medullary thyroid carcinoma. Nature 1994; 367:375-6. Eng C, Smith DP, Mulligan LM, Nagai MA, Healey CS, Ponder MA, et al. Point mutations within the tyrosine kinase domain of the ret proto-oncogene in multiple endocrine neoplasia type 2B and related sporadic tumours. Hum Mol Genet 1994;3:237-41. Komminoth P, Kunz E, Hiort 0, Schroder S, Matias-Guiu X, Christiansen G, et al. Detection of RET proto-oncogene point mutations in paraffin-embedded pheochromocytoma specimens by nonradioactive single strand conformation polymorphysm analysis and direct sequencing. A m ] Pathol 1994; 145:922-9. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a labora- tory manual. New York: Cold Spring Harbor Laboratory Press, 1989. Hiort 0, Huang Q, Sinnecker GHG, Sadeghi-Nejad A, Kruse K, Wolfe HJ, et al. Single strand conformation polymorphism anal- ysis of androgen receptor gene mutations in patients with an- drogen insensitivity syndromes: application for diagnosis, ge- netic counseling and therapy. J Clin Endocrinol Metab 1993; 7 7 262-6. Hiort 0, Wodtke A, Struve D, Zijllner A, Sinnecker GHG. De- tection of point mutations in the androgen receptor gene using non-isotopic single strand conformation polymorphism analy- sis. Hum Mol Genet 1994;1163-6. Tautz D, Renz M. An optimized freeze-squeeze method for the recovery of DNA fragments from agarose gels. Anal Biochem 1983; 132:14-9. Halloran N, Du 2, Wilson RK. Sequencing reactions for the Ap- plied Biosystems 373A automated DNA sequencer. In: H.G. Griffin and A. M. Griffin. Methods in molecular biology, Vol.: 10: DNA sequencing: Laboratory protocols. Clifton, NJ: The Hu- mana Press, 1992. Kwok S, Higuchi R. Avoiding false positives with PCR. Nature

White MB, Carvalho M, Derse D, O’Brien SJ, Dean M. Detecting single base substitutions as heteroduplex polymorphisms. Geno- mics 1992; 12301-6.

v

1579-83.

1989; 339123743,

RET Mutations in Medullary Thyroid Carcinoma/Komminoth et al. 489

35.

36.

37.

38.

39.

Schwindinger WF, Francomano CA, Levine MA. Ideqtification of a mutation in the gene encoding the alpha subunit of: the stim- ulatory G protein of adenylyl cyclase in McCune-Albgight syn- drome. Proc Null Acad Sci USA 1992;89:5152-6. Lips CJM, Landsvater RM, Hoppener JWM, Geerdink RA, Blij- ham G, Jansen-Schillhorn von Veen JM, et al. Clinical screening as compared with DNA analysis in families with multi le endo- crine neoplasia type 2A. N EnglIMed 1994;331:828-3 %! . Mulligan LM, Eng C, Healey CS, Clayton D, Kwok JBJ, Gardner E, et al. Specific mutations of the ret proto-oncogene are related to disease phenotype in MEN 2A and FMTC. Nature Genet

Ceccherini I, Romei C, Barone V, Pacini F, Martino E, Loviselli A, et al. Identification of the Cys634 + Tyr mutation of the RET proto-oncogene in a pedigree with multiple endocrine neoplasia type 2A and localized cutaneous lichen amyloidosis. J Endocrinol Invest 1994; 17201-4. Quadro L, Panariello L, Salvatore D, Carlomagno F, Del Prete M, Nunziata V, et a]. Frequent RET protooncogene mutations in multiple endocrine neoplasia type 2A. Clin Endocrinol Metab 1994; 79:590-4.

1994; 6:70-4.

40.

41.

42.

43.

44.

45.

van Heyningen V. One gene-four syndromes. Nature 1994;367: 319-20. Hanks SK, Quinn AM, Hunter T. The protein kinase family: con- served features and deduced phylogeny of the catalytic do- mains. Science 1988;241:42-52. McMahon R, Mulligan LM, Healy CS, Payne SJ, Ponder M, Fer- guson-Smith MA, et al. Direct, non-radioactive detection of mu- tations in multiple endocrine neoplasia type 2A families. Hum Mol Genet 1994;3:643-6. Xue F, Yu H, Maurer LH, Memoli VA, Nutile-McMenemy N, Schuster MK, el al. Germline ret mutations in MEN 2A and FMTC and their detection by simple DNA diagnostic tests. Hum Mol Genet 1994;3:635-8. Chi DD, Toshima K, Donis-Keller H, Wells SAJ. Predictive test- ing for multiple endocrine neoplasia type 2A (MEN 2A) based on the detection of mutations in the RET protooncogene. Surgery 1994; 11 6:124-32. Utiger RD. Medullary thyroid carcinoma, genes, and the preven- tion of cancer [editorial]. N Engl I M r d 1994;331:870-1.