Genetic Testing in the Diagnosis and Management of Multiple Endocrine Neoplasia Type II

Medullary Thyroid Carcinoma

Medullary thyroid carcinoma accounts for 5% to 10% of all thyroid malignancies and occurs in four different clinical settings: sporadic medullary thyroid carcinoma, familial non-MEN medullary thyroid carcinoma, MEN IIa,and MEN IIb [5] (Table 1). Noninherited (sporadic) medullary thyroid carcinoma accounts for 80% of cases and is a unifocal tumor that often occurs as a solitary thyroid nodule. Clinically, sporadic medullary thyroid carcinoma is characterized by a negative family history in a patient whose thyroid malignancy is usually diagnosed in the fifth or sixth decade of life. Histologically, sporadic medullary thyroid carcinoma is characterized by unicentric involvement and absence of C-cell hyperplasia. Basal or stimulated calcitonin levels or both increase in all patients with sporadic medullary thyroid carcinoma, although in most cases the diagnosis of medullary thyroid carcinoma is made postoperatively rather than preoperatively. Familial forms of medullary thyroid carcinoma account for the remaining 20% of cases; most of these are cases of MEN II. Because MEN II shows autosomal dominance, a child of an affected parent has a 50% chance of inheriting the causative gene. Medullary thyroid carcinoma is the initial presentation in most patients with MEN II; medullary thyroid carcinoma remains the major cause of morbidity and accounts for 5% to 10% of deaths in these patients. A recent Mayo Clinic study (O'Riordain DS, O'Brien T, Weaver AL, et al. Personal communication) found that the youngest patient with MEN IIa who had invasive medullary thyroid carcinoma was 2.7 years old. Forty percent of those younger than 5 years of age had invasive carcinoma; approximately 100% of patients older than 20 years of age had it. Multiple endocrine neoplasia IIb develops at an even earlier age and is more aggressive than other forms of medullary thyroid carcinoma; thus, it is recommended that total thyroidectomy be done in the first year of life, if possible, or as soon as diagnosis is made [6]. Fortunately, the characteristic clinical features of MEN IIb make it possible to diagnose most affected persons by phenotype; however, diagnosis may be difficult in some newborns with the syndrome.

Familial non-MEN medullary thyroid carcinoma develops later in life and is more indolent than MEN IIb. Histologically, hereditary medullary thyroid carcinoma occurs bilaterally in the superior portion of the thyroid lobes and is multicentric in almost all cases. C-cell hyperplasia, the precursor of medullary thyroid carcinoma, is bilateral and multicentric. C-cell hyperplasia may occur early in life, and surgical treatment of disease at this stage remains curative. Biochemical screening methods have focused on diagnosis of disease at an early age when surgery still presents a high chance for cure [7].

Pheochromocytoma

Pheochromocytomas are present in 10% to 50% of patients with MEN IIa [8] (Table 1). Pathologically, they range from adrenomedullary hyperplasia to large multilobular pheochromocytomas with extra-adrenal disease. The diagnosis is made biochemically by assessing a 24-hour urine collection for metanephrine, vanillylmandelic acid, and fractionated catecholamine levels. The initial imaging study should be an abdominal computed tomographic scan; if this does not identify a lesion, magnetic resonance imaging or meta-iodobenzyl guanidine scanning may be helpful [8]. Treatment consists of unilateral or bilateral total adrenalectomy because most patients have either bilateral disease or recurrent disease in the other apparently normal adrenal gland. After surgery, patients should be followed with annual measurements of 24-hour urine metanephrine, vanillylmandelic acid, and fractionated catecholamine levels.

Hyperparathyroidism

Parathyroid disease occurs in about 20% of patients with MEN IIa; it is rarely associated with MEN IIb and not at all with familial non-MEN medullary thyroid carcinoma [9] (Table 1). Patients are usually asymptomatic; an increased serum calcium value and an increased parathyroid hormone level are found during routine screening. Parathyroid hyperplasia occurs in approximately 84% of patients, and a parathyroid adenoma is present in 16%. Current treatment is total parathyroidectomy with parathyroid autotransplantation or subtotal parathyroidectomy with removal of 3.5 glands in patients shown to have hyperplasia at surgery. If the patient has parathyroid hyperplasia, a second operation is often needed 20 to 30 years later. Biochemical screening consists of measurement of serum calcium levels and should be initiated when the patient is approximately 10 years of age.

Screening

Biochemical Screening

Normal thyroid C cells, C-cell hyperplasia, and medullary thyroid carcinoma secrete calcitonin, and measurement of calcitonin levels has been an important diagnostic test in the evaluation of patients with familial medullary thyroid carcinoma. In patients with C-cell disease, basal plasma calcitonin levels are often increased and stimulated plasma calcitonin levels are always abnormal [10]. Because an occasional patient with C-cell disease may have a normal basal plasma calcitonin value and because basal calcitonin concentration may be increased in patients with conditions other than medullary thyroid carcinoma, provocative tests with calcium or pentagastrin have proved useful in identifying these situations. On the basis of our experience [11] and the experience of other investigators [9, 10], pentagastrin is more sensitive than calcium stimulation in diagnosing early C-cell disease. It has been common practice to screen first-degree relatives of patients with medullary thyroid carcinoma in the setting of MEN IIa with the pentagastrin test, beginning in early childhood (approximately 5 years of age) and continuing annually until 35 to 40 years of age if the intervening stimulated plasma calcitonin levels remain normal. Gagel and colleagues [12] reported that an abnormal pentagastrin test result developed in 50% of carriers of MEN IIa by the age of 12 years, in 80% by the age of 20 years, and in 95% by the age of 35 years; thyroidectomy in the preinvasive stage of medullary thyroid carcinoma was curative.

Biochemical screening using measurement of plasma calcitonin levels after pentagastrin stimulation has been extremely valuable in detecting early familial medullary thyroid carcinoma. Such screening identifies asymptomatic members in an affected kindred and permits cure with thyroidectomy. Although the value of calcitonin screening cannot be overemphasized, this screening method has problems. Adverse effects of the test include substernal discomfort, abdominal cramping, and nausea, and although these symptoms are transient, they can be disturbing and serious [10]. Children are especially reluctant to undergo regular intravenous testing, and adults are less than enthusiastic about repetitive testing. Other problems include the cost of screening programs, the necessity of lifelong testing for all members of a kindred, occasional false-positive results, and the difficulty of interpreting borderline pentagastrin test results.

Genetic Testing

A major advance in the evaluation and management of patients with medullary thyroid carcinoma has been the recent development of genetic testing. In 1987, two groups independently mapped the gene for MEN IIa to the centromeric region of chromosome 10 using genetic linkage analysis [13, 14]. Further molecular linkage studies have also localized the gene for MEN IIb and familial non-MEN medullary thyroid carcinoma to the same region of chromosome 10. Subsequently, several groups identified mutations in the RET proto-oncogene, which lies in the centromeric region on chromosome 10, in patients with MEN IIa, MEN IIb, and familial non-MEN medullary thyroid carcinoma [15-17]. RET is a member of the tyrosine kinase receptor gene family. The RET proto-oncogene on chromosome 10q11.2 is a transmembrane tyrosine kinase with a cysteine-rich extracellular domain Figure 1. It is expressed in normal human thyroid tissue, medullary thyroid carcinoma, and pheochromocytoma.

van Heyningen [18] observed that one gene is generally expected to be associated with one phenotype, but RET mutations are unique in that they cause four different syndromes: familial non-MEN medullary thyroid carcinoma, MEN IIa, MEN IIb, and Hirschsprung disease. The mutations in MEN IIa have all been found in exon 10 (codon 609, 611, 613, or 620) or exon 11 (codon 634), and all mutations to date involve a cysteine residue. It is interesting that although MEN IIa and familial non-MEN medullary thyroid carcinoma are clinically different syndromes, both share mutations at the same cysteine residues. Mulligan and colleagues [15] identified a RET mutation in exon 10 or 11 in 20 of 23 kindreds with MEN IIa and in no kindreds with MEN IIb. Donis-Keller and colleagues [16] found the RET proto-oncogene mutation in 8 of 11 kindreds with MEN IIa and in 4 of 5 kindreds with familial non-MEN medullary thyroid carcinoma. In a more recent report, Mulligan and colleagues [17] found the RET proto-oncogene mutation in 68 of 70 kindreds with MEN IIa (97%) and in 6 of 7 kindreds with familial non-MEN medullary thyroid carcinoma (86%).

We recently reported our experience with linkage and RET mutation analysis in 13 kindreds (109 persons) with medullary thyroid carcinoma (9 kindreds with MEN IIa, 2 with MEN IIb, and 2 with familial non-MEN medullary thyroid carcinoma) [19]. In the 9 kindreds with MEN IIa, 41 persons (35 known affected persons and 6 gene carriers identified presymptomatically) had a mutation in the RET proto-oncogene. The RET mutation data in each case were consistent with the linkage data. No RET mutations were found in exon 10 or 11 in the two kindreds with MEN IIb or the two kindreds with familial non-MEN medullary thyroid carcinoma.

If mutations within the RET proto-oncogene are responsible for MEN IIa and familial non-MEN medullary thyroid carcinoma, then current data suggest that not all mutations responsible for this disease have been identified. Although we showed a mutation within the RET proto-oncogene in all nine of our families with MEN IIa, not all such families examined to date have had such a mutation [15-17]. Recent data [17] suggest that approximately 95% of the mutations responsible for MEN IIa involve one of the five cysteine residues in exon 10 or 11 of the RET gene. Additionally, approximately 15% of persons with familial non-MEN medullary thyroid carcinoma do not have a mutation in one of these five codons. A more complete analysis of other regions of the RET proto-oncogene is needed to identify the full complement of mutations responsible for both MEN IIa and familial non-MEN medullary thyroid carcinoma.

Although genetic linkage analyses had mapped the MEN IIb gene to the centromeric region of chromosome 10, germline mutations in the RET proto-oncogene were identified only recently. Early in 1994, Hofstra and colleagues [20] reported that a germline mutation in exon 16 (codon 918) that substitutes a threonine for a methionine in the tyrosine kinase domain was responsible for MEN IIb in nine unrelated patients. More recently, Eng and colleagues [21] reported the same germline mutation in 26 of 28 apparently distinct families with MEN IIb. Additionally, 5 of 13 sporadic medullary thyroid carcinoma tumor specimens showed a somatic alteration at this site. Data on RET mutations from several centers are shown in Table 2.

Tsai and colleagues [19] found that two persons in two different kindreds had negative results for a mutation for MEN IIa using both linkage analysis and analysis of the RET proto-oncogene but had abnormal pentagastrin tests and questionable C-cell hyperplasia at surgery. These two cases were studied in the 1970s using the older, less sensitive calcitonin assays. Similar observations have been reported by Snow and Boyd [7] and Landsvater and colleagues [22]. Landsvater and colleagues [22] identified seven persons from a kindred with MEN IIa who had abnormal calcitonin stimulation tests and C-cell hyperplasia at surgery but who were not shown by linkage analysis to have a greater than 99% likelihood of being gene carriers for MEN IIa. We found that the identification of gene carriers on the basis of both linkage analysis and RET mutation analysis makes improper identification due to a recombination event virtually impossible. Because 5% to 10% of the general population may have abnormal pentagastrin stimulation tests and C-cell hyperplasia may occur in the thyroid glands of persons with various non-neoplastic and neoplastic disorders [23], we agree with Landsvater and colleagues [22] that genetic testing in kindreds with MEN II helps avoid potentially unnecessary operations.

Current Recommendations for Screening

In the familial medullary thyroid carcinoma syndromes, the primary goal of management has been early detection through screening to identify affected presymptomatic family members and provide early treatment. Screening programs that use sensitive calcitonin assays have been effective, and most patients having thyroidectomy are selected from screening protocols. For example, at the Mayo Clinic, calcitonin screening programs have diagnosed 80% of MEN IIa and 85% of MEN IIb cases in the past two decades [24]. However, despite diligent screening programs and early diagnosis, a recent study (O'Riordain DS, O'Brien T, Weaver AL, et al. Personal communication) indicated that 82% of patients identified by screening had medullary thyroid carcinoma and 10% had metastatic disease. Obviously, earlier diagnosis would improve prognosis and offer a greater chance for cure.

Although the correlation between pentagastrin test results and the results of genetic analysis is excellent, it is clear that in some gene carriers, calcitonin levels are normal with C-cell disease in early stages of development. We have followed gene carrier members of families with MEN IIa, members with normal calcitonin results who were operated on when calcitonin levels became abnormal and who were found to have C-cell hyperplasia or microscopic medullary thyroid carcinoma. Similarly, Wells and colleagues [25] reported that surgery confirmed C-cell disease in all 13 members of a family with MEN IIa and pentagastrin test results confirmed it in only 7 of the 13. Clearly, thyroidectomy based on the results of genetic analysis is a rational course of action.

Our current management strategy, and that of other groups [24, 26], for patients with medullary thyroid carcinoma includes use of genetic testing. Every patient with medullary thyroid carcinoma, sporadic or familial, now undergoes direct DNA analysis to identify possible germline mutations in the RET proto-oncogene (Figure 2). If a RET mutation is identified, all family members are tested for the same RET proto-oncogene mutation as early as possible. Once gene carriers (those with RET mutation) are identified, those in whom medullary thyroid carcinoma will develop are identified, and management with prophylactic thyroidectomy is discussed with these patients. The timing of surgery depends on the type of syndrome present. The algorithm shown in Figure 2 applies to patients with MEN IIa, familial non-MEN medullary thyroid carcinoma, and MEN IIb in whom a RET proto-oncogene mutation has already been identified.

In a kindred in which a RET mutation is identified, a family member without the mutation is considered to have essentially the same risk as the general population for future medullary thyroid carcinoma. These persons are informed of the results, and no further tests are necessary for them or their children. In the absence of a detectable mutation in the patient, genetic linkage analysis can be used to evaluate family members at risk (ure 2). If markers are adequate, the accuracy of the test can be as high as 95% to 99%, and if linkage analysis results are negative, we follow patients with pentagastrin stimulation tests in 3 to 5 years.

Presymptomatic testing using linkage analysis (rather than direct mutation analysis) has the disadvantage of requiring blood samples from at least two affected family members. Furthermore, false results due to recombination can occur at a rate as high as 2% to 5%, depending on the DNA markers used (recombination rate is related to the distance of the DNA markers used from the RET gene). Direct DNA analysis, on the other hand, can be done on persons in whom DNA from affected relatives is unavailable or on persons with negative family histories. Its accuracy is not affected by meiotic recombinations. If a mutation is identified that has previously been associated with familial medullary thyroid carcinoma or MEN, the diagnosis is secure. If a mutation is discovered that has not previously been reported, its significance is much less certain. Fortunately, there appear to be several common mutations in these disorders and interpretation of results is seldom an issue. Table 3 compares DNA analysis and linkage analysis.

With the rapid expansion of genetic knowledge and the new information that has become available, no consensus has been reached on how to use genetic testing in the management of patients with medullary thyroid carcinoma. For example, Raue and colleagues [27] suggested biochemical testing for all MEN II gene carriers and thyroidectomy when stimulated plasma calcitonin levels are abnormal. However, we agree with investigators at M.D. Anderson Hospital [24] and Washington University [26] that direct RET mutation analysis is reliable in patients with MEN IIa and that gene carriers should have prophylactic thyroidectomy once genetic testing has been completed. If, however, the primary physician caring for the patient, the patient, or the patient's family is not eager to proceed with immediate thyroid surgery, then an alternative policy is to follow up with pentagastrin stimulation tests and surgery when plasma calcitonin results are abnormal (ure 2). No case has been reported of a family member from a kindred with MEN IIa who tested negative for the RET mutation in a kindred with an identifiable mutation but in whom clinical MEN IIa then developed. This statement does not necessarily apply to linkage-based (indirect) DNA tests because a small error rate due to genetic recombination is recognized.

In patients or family members who have positive genetic results (either direct or indirect) for MEN IIa or IIb, annual screening for pheochromocytoma and hyperparathyroidism begins in childhood because these abnormalities manifest later than does medullary thyroid carcinoma in the MEN II syndromes [9]. If pheochromocytoma is present, thyroidectomy should be done after adrenal surgery. Our policy for postoperative follow-up of patients with C-cell hyperplasia or medullary thyroid carcinoma is a pentagastrin stimulation test and plasma calcitonin measurement on an annual basis for a minimum of 3 to 5 years. If test results remain negative during this period, no future tests are planned; if plasma calcitonin levels are abnormal, further evaluation with imaging studies is done.

Surveillance is not recommended for the person with a negative direct mutation analysis in a family with an identified mutation, and prophylactic therapy is not indicated for persons with either pheochromocytoma or hyperparathyroidism. In a person with a negative DNA test result based on family linkage data, some ongoing biochemical surveillance may be advisable, given the small percentage error inherent in this method. On the other hand, a positive test result, even linkage-based, is so highly predictive of the ultimate development of medullary thyroid carcinoma that aggressive management, including thyroidectomy, seems warranted.

In the young patient found to be RET positive by genetic testing, the question is always the risk of surgery in childhood compared with the risk posed by the subsequent development of medullary thyroid carcinoma and the possibility that thyroid malignancy will have gone beyond the curative stage at initial operation [24]. The risk of thyroid surgery and its complications of hypoparathyroidism, as well as recurrent laryngeal nerve damage, are particularly difficult to manage and tolerate in children, yet these risks are minimal in the hands of experienced thyroid surgeons. The issues of thyroid hormone replacement and child development can be best managed with minimal or no long-term sequelae by a competent pediatric endocrinologist. We agree with Thomas and Gagel [24] and recommend thyroid surgery for children at centers with experience and expertise in the operative and perioperative care of children.

Conclusion

Molecular genetic studies have increased our understanding of the pathogenesis of MEN tumors, have improved screening practices in the MEN II syndromes, and have modified management strategies. For now, genetic testing is recommended for patients with medullary thyroid carcinoma and all family members of these patients; thyroidectomy is offered to gene carriers. Genetic testing raises issues about the confidentiality of genetic information, genetic testing in minors, prenatal diagnosis, and other social and psychological aspects of genetic disease [28]. Pretest genetic counseling may be helpful to families considering DNA-based testing. We remain confident that with time these concerns will be resolved.

Appendix: Glossary

Codon: A triplet of three bases in a DNA or RNA molecule that specifies a single amino acid.

Autosomal dominant: Those conditions that are expressed in heterozygotes (that is, individuals with one copy of the mutant gene and one copy of the normal allele) in which the gene is on a nonsex chromosome.

Exon: Transcribed regions of a gene that are present in mature messenger RNA and usually contain coding information.

Germline mutation: A change in a gene that was present in the zygote and in all subsequent cells arising from that zygote.

Somatic mutation: A change in a gene that occurred after conception and that involves only some of a person's cells.

Mutation: Any permanent heritable change in the sequence of genomic DNA.

Genome: The complete DNA sequence of an organism; it contains the organism's complete genetic information.

Proto-oncogene: Normal genes that are found in normal eukaryotic cells concerned with various aspects of cell division. If amplified, mutated, or rearranged, they may give rise to carcinogenic oncogenes.

Linkage: Coinheritance of two or more nonallelic genes; because their loci are in such close proximity on the same chromosome, they remain associated after meiosis more often than 50% of the time, the expected probability for physically unlinked genes.

Sporadic disease: A disease arising from a somatic rather than an inherited mutation.

Familial disease: A disease occurring as a result of inheritance of a germline mutation.

Tyrosine kinase gene family: A group of genes involved in catalyzing the transfer of a phosphate group from adenosine triphosphate to a specific tyrosine residue in the enzyme, resulting in an increase or decrease of enzyme activity.