- Email: [email protected]
Hereditary medullary thyroid carcinoma in patients greater than 50 years old
Hereditary medullary thyroid carcinoma in patients greater than 50 years old Frank J. Quayle, MD, Ronald Benveniste, MD, Mary K. DeBenedetti, RN, Samuel A. Wells, MD, and Jeffrey F. Moley, MD, St. Louis, Mo, and Durham, NC
Background. Despite near complete penetrance and frequent early evaluation of medullary thyroid carcinoma (MTC) in multiple endocrine neoplasia 2 (MEN 2) variants, a significant minority of patients are evaluated later in life. Methods. With the aim of characterizing the expression of hereditary MTC in an older cohort, MEN 2 patients from our institutional database who were evaluated after age 50 years were identified, and clinical data were reviewed. Results. Thirty-nine patients (36 MEN 2A, 3 FMTC, and no MEN 2B) who were evaluated after age 50 years were identified; they represented 9% of all MEN 2 patients who were enrolled in our program. Most of the patients (79%) had abnormal screening examinations, and the AJCC staging was significantly higher in this cohort compared with younger patients. Overall, 43% of the patients had normal calcitonin levels after operation. There were 3 observed MTC-related deaths, all from distant metastases; the overall survival rate was 86% at 5 years and 74% at 10 years. The distribution of RET mutations in this cohort was similar to younger patients. Conclusions. These results suggest the presence of modifiers of MTC expression. Despite the tendency of older patients with hereditary MTC to have advanced stage disease at evaluation, they have high rates of biochemical cure, and the overall survival is excellent. (Surgery 2004;136:1116-21.) From the Washington University School of Medicine, St. Louis, Mo; and the Duke University School of Medicine, Durham, NC
MEDULLARY THYROID CARCINOMA (MTC) is a rare malignancy that occurs either sporadically or in the setting of the hereditary syndrome multiple endocrine neoplasia type 2 (MEN 2). MEN 2 has several clinically distinct variants: MEN 2A, characterized by MTC, pheochromocytoma, and hyperparathyroidism; MEN 2B, characterized by MTC, pheochromocytoma, characteristic mucosal ganglioneuromas, and skeletal abnormalities; and familial MTC (FMTC) characterized by MTC alone. MTC is associated with mutations in the RET proto-oncogene. Sporadic MTC tumors have an identifiable RET mutation in #80% of patients,1 and all MEN 2 variants are caused by germline mutations in the RET gene. MEN 2A is caused by Presented at the 25th Annual Meeting of the American Association of Endocrine Surgeons, Charlottesville, Virginia, April 4-6, 2004. Reprint requests: Jeffrey F. Moley, Department of Surgery, 660 S Euclid, Box 8109, St. Louis, MO 63139. 0039-6060/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.surg.2004.05.057
1116 SURGERY
missense mutations at 1 of 5 cysteine residues that are located in the extracellular domain of the RET protein.2,3 FMTC is caused by the same mutations at these cysteine residues or rarely by missense mutations at codons 768 or 804 in the intracellular domain.4-6 MEN 2B is caused by a specific methionine to threonine missense mutation at codon 918 in the intracellular tyrosine kinase domain or rarely by mutations at codon 883, also in the tyrosine kinase domain.4,7 MTC is a universal feature of all MEN 2 variants and occurs generally by the second or third decade of life. However, it has considerable variability in expression, some of which is associated with different germline mutations. For example, MTC in MEN 2B (codon 918 mutation) is characterized by extremely early onset (often within the first year of life) and an aggressive course, although MTC that is associated with the MEN 2A mutations is generally less aggressive. Despite frequent early onset, hereditary MTC is an indolent tumor with high 5- and 10-year survival rates. Recent efforts in the treatment of MEN 2 have focused on early detection and prophylactic
Quayle et al 1117
Surgery Volume 136, Number 6
Table I. Summary of mode of presentation, AJCC staging, and postoperative calcitonin levels in patients over age 50 years, broken down by RET codon mutated At evaluation
Postoperative calcitonin level
AJCC stage
Codon
N
Genetic
Calcitonin
Nodule
C-cell hyperplasia
1
2
3
4
Normal
Elevated
609 618 620 634 *Total
5 15 6 12 39
0 1 1 0 2
5 11 3 8 28
0 2 2 4 8
0 1 0 0 2
2 6 3 1 12
1 4 0 6 11
2 2 3 4 11
0 0 0 1 1
1 8 0 3 13
0 5 4 8 17
*Total row includes 1 additional patient whose mutation has not been identified. Complete information is not available for all patients with each mutation.
thyroidectomy. Most authors and recent consensus guidelines recommend genetic testing at birth for all at-risk patients. Total thyroidectomy is recommended in childhood for patients with MEN 2A and FMTC and in infancy for patients with MEN 2B .8 Early thyroidectomy in RET mutation carriers is associated with lower rates of MTC, lower rates of advanced-stage disease, and higher rates of clinical and biochemical cure.9,10 Despite these advances in the clinical treatment of MEN 2, not all at-risk patients are identified by early genetic testing. A significant number of patients are evaluated later in life, either with clinically evident disease or through screening that is prompted by disease in a family member. The aim of this study was to characterize the expression of hereditary MTC in an older cohort, particularly with regard to genotype-phenotype correlations, to better understand the observed variability in MEN 2. METHODS Under authorization from the Human Studies Committee, the MEN program at our institution maintains a prospective database of clinical and genetic information regarding patients with MEN 1 and 2. It contains genetic, biochemical, and clinical data on 383 affected members of 64 MEN 2A families, 45 affected members of 24 MEN 2B families, and 34 affected members of 8 FMTC families. From this database, patients were categorized according to age at evaluation, and the cohort of patients whose condition was diagnosed after age 50 years was analyzed. Statistical analysis was performed with SAS software (SAS Institute Inc, Cary, NC). For the purposes of comparison, patients with MEN 2B were omitted from the analysis, and MEN 2A and FMTC cohorts were examined together. Specifically, comparisons between these cohorts of older patients and younger patients regarding categoric
variables were performed with contingency tables with the use of the Fisher exact test. Survival analysis was performed with life table analysis by the Kaplan-Meier method. RESULTS A total of 39 patients (18 men and 21 women) whose condition was diagnosed after the age of 50 years were identified. These included 36 patients from 16 MEN 2A families, 3 patients from 3 FMTC families, and no patients from MEN 2B families. This represents 9% of patients with MEN 2A, 9% of patients with FMTC, and 0% of patients with MEN 2B, respectively, in our database. Five patients (13%) had codon 609 mutations; no patients (0%) had codon 611 mutations; 15 patients (39%) had codon 618 mutations; 6 patients (16%) had codon 620 mutations, and 12 patients (31%) had codon 634 mutations (Table I). One patient came from a family whose mutation has not been identified. When patients with MEN 2B are excluded from the analysis, this distribution of mutations does not differ significantly from that seen in the database as a whole. Analysis of different mutations within each codon and of different families with the same mutation did not reveal significant distortions in the likelihood of patients in this age group. Most of this cohort was identified through screening examinations, and each patient was a member of a previously recognized MEN 2 kindred. Twenty-eight patients (74%) had abnormal serum calcitonin levels, either at baseline or after calcium-pentagastrin stimulation. Two patients (5%), aged 55 and 57 years at diagnosis, had a positive genetic screen in the setting of normal stimulated calcitonin levels. The pathologic finding for each of these patients was positive for microscopic MTC. The remaining 8 patients (21%) had a palpable nodule on physical examination.
1118 Quayle et al
Surgery December 2004
Fig 1. MEN 2A and FMTC in patients who are more than 50 years old. A, Disease-free survival. B, Overall survival.
The histologic diagnosis was MTC in 37 patients and C-cell hyperplasia in 2 patients. The 2 patients with C-cell hyperplasia came from different families with different RET mutations (1 patient from a large MEN 2A kindred with a codon 618 mutation and the 1 patient from a FMTC kindred with the mutation as yet unidentified). AJCC staging was complete in 35 of the 37 patients who were diagnosed with MTC, with the following distribution: stage I, 12 patients (34%); stage II, 11 patients (31%); stage III, 11 patients (31%); and stage IV, 1 patient (3%). This represented significantly higher stage disease compared with the database as a whole (P = .016). All but 1 patient opted for surgery and underwent total thyroidectomy. Seven patients (18%) underwent central node dissection at the time of the initial operation. In these 7 patients, the number of central nodes that were sampled ranged from 3 to 89. Three patients (42%) received a diagnosis with N1 disease. An additional 8 patients were identified with node positive disease with total thyroidectomy and sampling of clinically abnormal tissue alone. The frequency of N1 disease in the cohort that was aged >50 years was similar to the database as a whole. Postoperative calcitonin testing was performed in 30 patients who were aged >50 years. Stimulated calcitonin levels were undetectable in 13 patients (43%) over a median follow-up of 77 months. Of the 17 patients (57%) who had elevated postoperative calcitonin levels, 15 patients had calcitonin elevations immediately after the operation. The remaining 2 patients had undetectable calcitonin levels until 21 and 67 postoperative months, respectively (Fig 1).
Repeat neck exploration procedures for recurrent or persistent disease were undertaken in 5 patients, 2 procedures within the first year after initial thyroidectomy and 3 procedures after >2 years. After redo neck operation, 4 patients had persistently elevated calcitonin levels. One of these patients had distant metastases and eventually died of the disease. One patient had normalization of stimulated calcitonin levels after a neck dissection that was undertaken 7 years after the initial thyroidectomy. Three patients had distant metastases in the course of follow up. These occurred in the liver and bone in 1 patient; the liver, lung, and kidney in 1 patient; and the liver and skin in the 1 patient. All 3 patients died of the disease and accounted for all of the MTC-specific deaths that were observed. Five- and 10-year MTC-specific survival was 87% and 83%, respectively, in patients aged >50 years (data not shown). Overall 5- and 15-year survival rates were 86% and 74%, respectively (Fig 1). MTC-specific (P = .04) and overall survival (P > .001) were significantly lower in the older cohort compared younger patients. Stage-specific survival was similar in older and younger patients. DISCUSSION The factors that are responsible for variability of expression in MEN 2 are incompletely understood. In this analysis, we found that within MEN 2A and FMTC kindreds, nearly 10% of patients were evaluated after the age of 50 years. These numbers suggest that a significant subset of patients have either delayed onset of MTC or indolent tumor
Surgery Volume 136, Number 6
biology because of the presence of modifiers that mediate the clinical expression of MEN 2. Some differences in expression within MEN 2 variants are clearly related to specific RET mutations. MEN 2B caused by codon 918 mutations is a distinct clinical entity in our experience, with no patients being diagnosed over age 50 years. Even within MEN 2A and FMTC, life-tables analysis reveals highly significant differences in age at evaluation between specific codons (P < .001). Specifically, codon 609 and 618 mutations seem to have later onset compared with codon 634 mutations, which corroborates other groups’ findings regarding the biologic implications of different RET mutations.4,5,7 Our current analysis demonstrates that some variability cannot be explained by RET mutation alone. When codon-specific analysis of age at evaluation is applied to patients who are aged >50 years, we find that the differences between codons disappear and that the distribution of mutations in this cohort is similar to that of the database as a whole. This suggests that a subset of patients with MEN 2A, regardless of codon mutation, has indolent disease. When different mutations at each codon (eg, C634R vs C634Y) are analyzed, no association between late onset disease and individual mutations is apparent. Similarly, no individual families had a significant preponderance of patients over age 50 years. Interestingly, within our cohort of 39 patients who were evaluated over age 50 years, 10 patients are from sibships that share $2 members who were evaluated at >50 years old. In fact, 2 sibships have 3 members, each of whom was evaluated after age 50 years. Each of these occurrences is individually quite statistically improbable (P < .001). Although ascertainment bias is a possible factor that correlates late evaluation within nuclear families, this striking phenomenon offers further evidence of mediators of MTC expression. Because members of sibships share one half of their genetic makeup, 1 possible explanation is the presence of genetic modifiers that interact with RET or otherwise mediate the expression of MTC. An alternate explanation is that members of the same nuclear family share similar surroundings, and thus environmental factors may cause differences in expression. Other investigators previously have described genetic modifiers of MEN 2 expression. Robledo et al11 reported 2 RET polymorphisms in strong linkage disequilibrium, G691S and S904S, which are associated with earlier onset of MTC in patients with MEN 2A. Borrego et al12 and Pasini et al13
Quayle et al 1119
independently postulate that Hirschsprung’s disease that occurs in individual members of MEN 2A kindreds may be due to a RET A45A polymorphism. This RET polymorphism has also been associated with familial and sporadic Hirschsprung’s disease, independent of MEN 2.14 Although these putative modifier polymorphisms all occur within the RET gene, mediator polymorphisms at other loci that interact with RET also likely exist. Remarkably, only 21% of our cohort had palpable disease at examination, although the remainder of the patients with disease was detected by abnormal screening examinations. This proportion is similar to that seen in younger patients, but given our notions of the inevitable progression of MTC over time, this finding is striking and offers further evidence that the biology of these tumors is fundamentally different. The finding of more advanced-stage disease in older patients is more consistent with expectations and likely reflects subclinical disease that had been present for some time. It is interesting that, although our older cohort had larger tumors and significantly higher T staging compared with younger patients, differences in node status were less striking and not statistically significant. This perhaps represents again a difference in tumor biology and suggests the presence of a mediator of metastatic potential. Despite the higher-stage disease that was found in patients aged >50 years, almost one half of these patients experience biochemical cure after surgery, a rate that is comparable to younger patients who undergo earlier thyroidectomy. Moreover, the long-term survival rate is quite high, near 75% at 10 years. Several treatment questions arise when the age at evaluation of a patient with MTC is considered. The first is the likelihood that a patient whose condition is assessed at an older age has a germline RET mutation. Although the current analysis does not address this question directly and none of these cases were index patients, the fact that a significant number of patients with hereditary MTC are seen after 50 years old suggests that age alone should not exclude the consideration of MEN 2A or FMTC. Additionally, these results emphasize the importance of screening older generations when an index MEN 2 case is identified. Another issue is the optimal timing of thyroidectomy in patients with MEN 2A and FMTC. Recent emphasis in the treatment of hereditary forms of MTC has been placed on early mutation analysis and prophylactic thyroidectomy. This has
1120 Quayle et al
undoubtedly led to improved outcomes for MTC, with lower rates of frank carcinoma and higher rates of long-term cure; the current data should not be construed to change this approach. Despite the overall indolent nature of these patients’ disease, we detected significantly lower overall and MTC-specific survival rates for patients who are seen later in life. Moreover, it is impossible at this point to predict which patients will have aggressive disease and which patients will have indolent disease. For these reasons, our recommendation continues to be early thyroidectomy after the identification of a germline mutation. Despite the advances of genetic mutation analysis and aggressive screening techniques, a significant proportion of patients with MEN 2A and FMTC will be assessed at a later age. Our current findings suggest that a subset of hereditary MTCs have indolent tumor biology, possibly because of the influence of genetic or environmental modifiers, and future research should be undertaken to identify and characterize such influences. Patients who are diagnosed at an advanced age have a reasonable likelihood of cure and a high likelihood of long-term survival.
REFERENCES 1. Eng C, Mulligan LM. Mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2, related sporadic tumours and Hirschsprung disease. Hum Mutat 1997;9: 97-109. 2. Mulligan LM, Kwok JBJ, Healey CS, et al. Germline mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993;363:458-60. 3. Donis-Keller H, Dou S, Chi D, et al. Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet 1993;2:851-6. 4. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2: International RET Mutation Consortium Analysis. JAMA 1996;276:1575-9. 5. Mulligan LM, Eng C, Healey CS, et al. Specific mutations of the RET proto-oncogene are related to disease phenotype in MEN 2A and FMTC. Nat Genet 1994;6:70-4. 6. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2: International RET Mutation Consortium Analysis. JAMA 1996;276:1575-9. 7. Gimm O, Marsh DJ, Andrew SD, et al. Germline dinucleotide mutation in codon 883 of the RET proto-oncogene in multiple endocrine neoplasia type 2B without codon 918 mutation. J Clin Endocrinol Metab 1997;82:3902-4. 8. Brandi ML, Gagel RF, Anfeli A, Bilezikian JP, et al. Consensus: guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001;86: 5658-71.
Surgery December 2004
9. Machens A, Niccoli-Sire P, Hoegel J, et al. Early malignant progression of hereditary medullary thyroid cancer. N Engl J Med 2003;349:1517-25. 10. Wells SA Jr, Skinner MA. Prophylactic thyroidectomy, based on direct genetic testing, in patients at risk for the multiple endocrine neoplasia type 2 syndromes. Exp Clin Endocrinol Diabetes 1998;106:29-34. 11. Robledo M, Gil L, Pollan M, et al. Polymorphisms G691S/ S904S of RET as genetic modifiers of MEN 2A. Cancer Res 2003;63:1814-7. 12. Borrego S, Eng C, Sanchez B, et al. Molecular analysis of the ret and GDNF genes in a family with multiple endocrine neoplasia type 2A and Hirschsprung disease. J Clin Endocrinol Metab 1998;83:3361-4. 13. Pasini B, Rossi R, Ambrosio MR, et al. RET mutation profile and variable clinical manifestations in a family with multiple endocrine neoplasia type 2A and Hirschsprung’s disease. Surgery 2002;131:373-81. 14. Borrego S, Saez ME, Ruiz A, et al. Specific polymorphisms in the RET proto-oncogene are over-represented in patients with Hirschsprung disease and may represent loci modifying phenotypic expression. J Med Genet 1999;36:771-4.
DISCUSSION Dr Henning Dralle (Halle, Germany). In your abstract, you stated that 27% of patients had codon 634 mutations. Are these all index cases? What is the reason for the high percentage of patients with 634 mutations? You said the biochemical cure rate was the same at >50 years of age or at <50 years of age. If this is true, what is the reason for that? Did you analyze double mutations in the afflicted families? These double mutations might explain the change in tumor biology and age of evaluation. Dr Quayle. With regard to the second question about biochemical cure, we did find similar rates between the groups. Our hypothesis would be that this is due to modifiers, that these patients had indolent cancers and not advanced disease despite their late age at evaluation and therefore would be in the same time-frame that enables a cure with thyroidectomy alone. Regarding your last question, we did not analyze and do not have any experience, to my knowledge, with double mutations. Dr Dralle. So the biochemical cure rate was the same in patients who were >50 years of age compared with patients who were <50 years of age? Dr Quayle. It was approximately 40% to 50% in both groups. Dr Dralle. But the tumor stages were higher in the patients who were >50 years of age? Dr Quayle. I think it probably speaks to a lack of power. There were 30 patients in that group versus 360 patients in the other group, so I think that we probably did not have the power to determine whether that was a statistically significant difference. Dr Dralle. Is the high percentage of 634 patients, approximately one third of the patients, because of a high number of index cases or were these patients screening cases?
Surgery Volume 136, Number 6
Dr Quayle. Our cohort of patients who were >50 years of age tended to have fewer patients with 634 mutations, but our experience, and I think it corroborates with the experience that other authors have had, is that 634 mutations are the most prevalent in MEN2A. Dr Paul G. Gauger (Ann Arbor, Mich). In the process of defining your study group, did you encounter any patients that were not included because they were the only ones with that mutation and you were considering it a polymorphism? Specifically, I have a question about codon 666. I take care of a father and son who would have fit into your study group by all other criteria, but their mutation (codon 666) has not really been implicated in the same way as the others that you mention. Dr Quayle. There was one patient in whom, to date, the mutation has not been identified. This patient was examined before the extensive analysis of the other exons, so the family’s mutation has not been identified. It seems as if every year more and more mutations that are associated with this indolent disease are described. These atypical mutations are less common and harder to find. Moreover, the disease may be so indolent that it is not 100% penetrant, and we miss generations; therefore,
Quayle et al 1121
these cases are not identified as having a hereditary component. I suspect there are a lot of patients out there like your patient with less common mutations that are causing more indolent disease. Dr Douglas B. Evans (Houston, Tex). To summarize, would your data agree with the level classification in the Brandi article (Brandi et al. JCEM 2001;86:5658-5671) or would you disagree? If so, I assume it would be a level 2 mutation that you would consider changing, or were there no mutations for which you would change the level classification? Dr Quayle. I would say that our paper did not address specifically the issues of the timing of thyroidectomy when you identify a patient early. In our experience, if you recall, the penetrance plot of mutations, 609 mutations (if I remember correctly), were classified in the level 3 mutations in the Brandi conference. In our experience, we did not see cases earlier than age 20 years for those particular mutations. I do not think that there is anything to suggest that we know which patients are going to fit into this older cohort at this point. So, I think that we have no data to suggest that those recommendations are inappropriate at this time.
Background. Despite near complete penetrance and frequent early evaluation of medullary thyroid carcinoma (MTC) in multiple endocrine neoplasia 2 (MEN 2) variants, a significant minority of patients are evaluated later in life. Methods. With the aim of characterizing the expression of hereditary MTC in an older cohort, MEN 2 patients from our institutional database who were evaluated after age 50 years were identified, and clinical data were reviewed. Results. Thirty-nine patients (36 MEN 2A, 3 FMTC, and no MEN 2B) who were evaluated after age 50 years were identified; they represented 9% of all MEN 2 patients who were enrolled in our program. Most of the patients (79%) had abnormal screening examinations, and the AJCC staging was significantly higher in this cohort compared with younger patients. Overall, 43% of the patients had normal calcitonin levels after operation. There were 3 observed MTC-related deaths, all from distant metastases; the overall survival rate was 86% at 5 years and 74% at 10 years. The distribution of RET mutations in this cohort was similar to younger patients. Conclusions. These results suggest the presence of modifiers of MTC expression. Despite the tendency of older patients with hereditary MTC to have advanced stage disease at evaluation, they have high rates of biochemical cure, and the overall survival is excellent. (Surgery 2004;136:1116-21.) From the Washington University School of Medicine, St. Louis, Mo; and the Duke University School of Medicine, Durham, NC
MEDULLARY THYROID CARCINOMA (MTC) is a rare malignancy that occurs either sporadically or in the setting of the hereditary syndrome multiple endocrine neoplasia type 2 (MEN 2). MEN 2 has several clinically distinct variants: MEN 2A, characterized by MTC, pheochromocytoma, and hyperparathyroidism; MEN 2B, characterized by MTC, pheochromocytoma, characteristic mucosal ganglioneuromas, and skeletal abnormalities; and familial MTC (FMTC) characterized by MTC alone. MTC is associated with mutations in the RET proto-oncogene. Sporadic MTC tumors have an identifiable RET mutation in #80% of patients,1 and all MEN 2 variants are caused by germline mutations in the RET gene. MEN 2A is caused by Presented at the 25th Annual Meeting of the American Association of Endocrine Surgeons, Charlottesville, Virginia, April 4-6, 2004. Reprint requests: Jeffrey F. Moley, Department of Surgery, 660 S Euclid, Box 8109, St. Louis, MO 63139. 0039-6060/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.surg.2004.05.057
1116 SURGERY
missense mutations at 1 of 5 cysteine residues that are located in the extracellular domain of the RET protein.2,3 FMTC is caused by the same mutations at these cysteine residues or rarely by missense mutations at codons 768 or 804 in the intracellular domain.4-6 MEN 2B is caused by a specific methionine to threonine missense mutation at codon 918 in the intracellular tyrosine kinase domain or rarely by mutations at codon 883, also in the tyrosine kinase domain.4,7 MTC is a universal feature of all MEN 2 variants and occurs generally by the second or third decade of life. However, it has considerable variability in expression, some of which is associated with different germline mutations. For example, MTC in MEN 2B (codon 918 mutation) is characterized by extremely early onset (often within the first year of life) and an aggressive course, although MTC that is associated with the MEN 2A mutations is generally less aggressive. Despite frequent early onset, hereditary MTC is an indolent tumor with high 5- and 10-year survival rates. Recent efforts in the treatment of MEN 2 have focused on early detection and prophylactic
Quayle et al 1117
Surgery Volume 136, Number 6
Table I. Summary of mode of presentation, AJCC staging, and postoperative calcitonin levels in patients over age 50 years, broken down by RET codon mutated At evaluation
Postoperative calcitonin level
AJCC stage
Codon
N
Genetic
Calcitonin
Nodule
C-cell hyperplasia
1
2
3
4
Normal
Elevated
609 618 620 634 *Total
5 15 6 12 39
0 1 1 0 2
5 11 3 8 28
0 2 2 4 8
0 1 0 0 2
2 6 3 1 12
1 4 0 6 11
2 2 3 4 11
0 0 0 1 1
1 8 0 3 13
0 5 4 8 17
*Total row includes 1 additional patient whose mutation has not been identified. Complete information is not available for all patients with each mutation.
thyroidectomy. Most authors and recent consensus guidelines recommend genetic testing at birth for all at-risk patients. Total thyroidectomy is recommended in childhood for patients with MEN 2A and FMTC and in infancy for patients with MEN 2B .8 Early thyroidectomy in RET mutation carriers is associated with lower rates of MTC, lower rates of advanced-stage disease, and higher rates of clinical and biochemical cure.9,10 Despite these advances in the clinical treatment of MEN 2, not all at-risk patients are identified by early genetic testing. A significant number of patients are evaluated later in life, either with clinically evident disease or through screening that is prompted by disease in a family member. The aim of this study was to characterize the expression of hereditary MTC in an older cohort, particularly with regard to genotype-phenotype correlations, to better understand the observed variability in MEN 2. METHODS Under authorization from the Human Studies Committee, the MEN program at our institution maintains a prospective database of clinical and genetic information regarding patients with MEN 1 and 2. It contains genetic, biochemical, and clinical data on 383 affected members of 64 MEN 2A families, 45 affected members of 24 MEN 2B families, and 34 affected members of 8 FMTC families. From this database, patients were categorized according to age at evaluation, and the cohort of patients whose condition was diagnosed after age 50 years was analyzed. Statistical analysis was performed with SAS software (SAS Institute Inc, Cary, NC). For the purposes of comparison, patients with MEN 2B were omitted from the analysis, and MEN 2A and FMTC cohorts were examined together. Specifically, comparisons between these cohorts of older patients and younger patients regarding categoric
variables were performed with contingency tables with the use of the Fisher exact test. Survival analysis was performed with life table analysis by the Kaplan-Meier method. RESULTS A total of 39 patients (18 men and 21 women) whose condition was diagnosed after the age of 50 years were identified. These included 36 patients from 16 MEN 2A families, 3 patients from 3 FMTC families, and no patients from MEN 2B families. This represents 9% of patients with MEN 2A, 9% of patients with FMTC, and 0% of patients with MEN 2B, respectively, in our database. Five patients (13%) had codon 609 mutations; no patients (0%) had codon 611 mutations; 15 patients (39%) had codon 618 mutations; 6 patients (16%) had codon 620 mutations, and 12 patients (31%) had codon 634 mutations (Table I). One patient came from a family whose mutation has not been identified. When patients with MEN 2B are excluded from the analysis, this distribution of mutations does not differ significantly from that seen in the database as a whole. Analysis of different mutations within each codon and of different families with the same mutation did not reveal significant distortions in the likelihood of patients in this age group. Most of this cohort was identified through screening examinations, and each patient was a member of a previously recognized MEN 2 kindred. Twenty-eight patients (74%) had abnormal serum calcitonin levels, either at baseline or after calcium-pentagastrin stimulation. Two patients (5%), aged 55 and 57 years at diagnosis, had a positive genetic screen in the setting of normal stimulated calcitonin levels. The pathologic finding for each of these patients was positive for microscopic MTC. The remaining 8 patients (21%) had a palpable nodule on physical examination.
1118 Quayle et al
Surgery December 2004
Fig 1. MEN 2A and FMTC in patients who are more than 50 years old. A, Disease-free survival. B, Overall survival.
The histologic diagnosis was MTC in 37 patients and C-cell hyperplasia in 2 patients. The 2 patients with C-cell hyperplasia came from different families with different RET mutations (1 patient from a large MEN 2A kindred with a codon 618 mutation and the 1 patient from a FMTC kindred with the mutation as yet unidentified). AJCC staging was complete in 35 of the 37 patients who were diagnosed with MTC, with the following distribution: stage I, 12 patients (34%); stage II, 11 patients (31%); stage III, 11 patients (31%); and stage IV, 1 patient (3%). This represented significantly higher stage disease compared with the database as a whole (P = .016). All but 1 patient opted for surgery and underwent total thyroidectomy. Seven patients (18%) underwent central node dissection at the time of the initial operation. In these 7 patients, the number of central nodes that were sampled ranged from 3 to 89. Three patients (42%) received a diagnosis with N1 disease. An additional 8 patients were identified with node positive disease with total thyroidectomy and sampling of clinically abnormal tissue alone. The frequency of N1 disease in the cohort that was aged >50 years was similar to the database as a whole. Postoperative calcitonin testing was performed in 30 patients who were aged >50 years. Stimulated calcitonin levels were undetectable in 13 patients (43%) over a median follow-up of 77 months. Of the 17 patients (57%) who had elevated postoperative calcitonin levels, 15 patients had calcitonin elevations immediately after the operation. The remaining 2 patients had undetectable calcitonin levels until 21 and 67 postoperative months, respectively (Fig 1).
Repeat neck exploration procedures for recurrent or persistent disease were undertaken in 5 patients, 2 procedures within the first year after initial thyroidectomy and 3 procedures after >2 years. After redo neck operation, 4 patients had persistently elevated calcitonin levels. One of these patients had distant metastases and eventually died of the disease. One patient had normalization of stimulated calcitonin levels after a neck dissection that was undertaken 7 years after the initial thyroidectomy. Three patients had distant metastases in the course of follow up. These occurred in the liver and bone in 1 patient; the liver, lung, and kidney in 1 patient; and the liver and skin in the 1 patient. All 3 patients died of the disease and accounted for all of the MTC-specific deaths that were observed. Five- and 10-year MTC-specific survival was 87% and 83%, respectively, in patients aged >50 years (data not shown). Overall 5- and 15-year survival rates were 86% and 74%, respectively (Fig 1). MTC-specific (P = .04) and overall survival (P > .001) were significantly lower in the older cohort compared younger patients. Stage-specific survival was similar in older and younger patients. DISCUSSION The factors that are responsible for variability of expression in MEN 2 are incompletely understood. In this analysis, we found that within MEN 2A and FMTC kindreds, nearly 10% of patients were evaluated after the age of 50 years. These numbers suggest that a significant subset of patients have either delayed onset of MTC or indolent tumor
Surgery Volume 136, Number 6
biology because of the presence of modifiers that mediate the clinical expression of MEN 2. Some differences in expression within MEN 2 variants are clearly related to specific RET mutations. MEN 2B caused by codon 918 mutations is a distinct clinical entity in our experience, with no patients being diagnosed over age 50 years. Even within MEN 2A and FMTC, life-tables analysis reveals highly significant differences in age at evaluation between specific codons (P < .001). Specifically, codon 609 and 618 mutations seem to have later onset compared with codon 634 mutations, which corroborates other groups’ findings regarding the biologic implications of different RET mutations.4,5,7 Our current analysis demonstrates that some variability cannot be explained by RET mutation alone. When codon-specific analysis of age at evaluation is applied to patients who are aged >50 years, we find that the differences between codons disappear and that the distribution of mutations in this cohort is similar to that of the database as a whole. This suggests that a subset of patients with MEN 2A, regardless of codon mutation, has indolent disease. When different mutations at each codon (eg, C634R vs C634Y) are analyzed, no association between late onset disease and individual mutations is apparent. Similarly, no individual families had a significant preponderance of patients over age 50 years. Interestingly, within our cohort of 39 patients who were evaluated over age 50 years, 10 patients are from sibships that share $2 members who were evaluated at >50 years old. In fact, 2 sibships have 3 members, each of whom was evaluated after age 50 years. Each of these occurrences is individually quite statistically improbable (P < .001). Although ascertainment bias is a possible factor that correlates late evaluation within nuclear families, this striking phenomenon offers further evidence of mediators of MTC expression. Because members of sibships share one half of their genetic makeup, 1 possible explanation is the presence of genetic modifiers that interact with RET or otherwise mediate the expression of MTC. An alternate explanation is that members of the same nuclear family share similar surroundings, and thus environmental factors may cause differences in expression. Other investigators previously have described genetic modifiers of MEN 2 expression. Robledo et al11 reported 2 RET polymorphisms in strong linkage disequilibrium, G691S and S904S, which are associated with earlier onset of MTC in patients with MEN 2A. Borrego et al12 and Pasini et al13
Quayle et al 1119
independently postulate that Hirschsprung’s disease that occurs in individual members of MEN 2A kindreds may be due to a RET A45A polymorphism. This RET polymorphism has also been associated with familial and sporadic Hirschsprung’s disease, independent of MEN 2.14 Although these putative modifier polymorphisms all occur within the RET gene, mediator polymorphisms at other loci that interact with RET also likely exist. Remarkably, only 21% of our cohort had palpable disease at examination, although the remainder of the patients with disease was detected by abnormal screening examinations. This proportion is similar to that seen in younger patients, but given our notions of the inevitable progression of MTC over time, this finding is striking and offers further evidence that the biology of these tumors is fundamentally different. The finding of more advanced-stage disease in older patients is more consistent with expectations and likely reflects subclinical disease that had been present for some time. It is interesting that, although our older cohort had larger tumors and significantly higher T staging compared with younger patients, differences in node status were less striking and not statistically significant. This perhaps represents again a difference in tumor biology and suggests the presence of a mediator of metastatic potential. Despite the higher-stage disease that was found in patients aged >50 years, almost one half of these patients experience biochemical cure after surgery, a rate that is comparable to younger patients who undergo earlier thyroidectomy. Moreover, the long-term survival rate is quite high, near 75% at 10 years. Several treatment questions arise when the age at evaluation of a patient with MTC is considered. The first is the likelihood that a patient whose condition is assessed at an older age has a germline RET mutation. Although the current analysis does not address this question directly and none of these cases were index patients, the fact that a significant number of patients with hereditary MTC are seen after 50 years old suggests that age alone should not exclude the consideration of MEN 2A or FMTC. Additionally, these results emphasize the importance of screening older generations when an index MEN 2 case is identified. Another issue is the optimal timing of thyroidectomy in patients with MEN 2A and FMTC. Recent emphasis in the treatment of hereditary forms of MTC has been placed on early mutation analysis and prophylactic thyroidectomy. This has
1120 Quayle et al
undoubtedly led to improved outcomes for MTC, with lower rates of frank carcinoma and higher rates of long-term cure; the current data should not be construed to change this approach. Despite the overall indolent nature of these patients’ disease, we detected significantly lower overall and MTC-specific survival rates for patients who are seen later in life. Moreover, it is impossible at this point to predict which patients will have aggressive disease and which patients will have indolent disease. For these reasons, our recommendation continues to be early thyroidectomy after the identification of a germline mutation. Despite the advances of genetic mutation analysis and aggressive screening techniques, a significant proportion of patients with MEN 2A and FMTC will be assessed at a later age. Our current findings suggest that a subset of hereditary MTCs have indolent tumor biology, possibly because of the influence of genetic or environmental modifiers, and future research should be undertaken to identify and characterize such influences. Patients who are diagnosed at an advanced age have a reasonable likelihood of cure and a high likelihood of long-term survival.
REFERENCES 1. Eng C, Mulligan LM. Mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2, related sporadic tumours and Hirschsprung disease. Hum Mutat 1997;9: 97-109. 2. Mulligan LM, Kwok JBJ, Healey CS, et al. Germline mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993;363:458-60. 3. Donis-Keller H, Dou S, Chi D, et al. Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet 1993;2:851-6. 4. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2: International RET Mutation Consortium Analysis. JAMA 1996;276:1575-9. 5. Mulligan LM, Eng C, Healey CS, et al. Specific mutations of the RET proto-oncogene are related to disease phenotype in MEN 2A and FMTC. Nat Genet 1994;6:70-4. 6. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2: International RET Mutation Consortium Analysis. JAMA 1996;276:1575-9. 7. Gimm O, Marsh DJ, Andrew SD, et al. Germline dinucleotide mutation in codon 883 of the RET proto-oncogene in multiple endocrine neoplasia type 2B without codon 918 mutation. J Clin Endocrinol Metab 1997;82:3902-4. 8. Brandi ML, Gagel RF, Anfeli A, Bilezikian JP, et al. Consensus: guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001;86: 5658-71.
Surgery December 2004
9. Machens A, Niccoli-Sire P, Hoegel J, et al. Early malignant progression of hereditary medullary thyroid cancer. N Engl J Med 2003;349:1517-25. 10. Wells SA Jr, Skinner MA. Prophylactic thyroidectomy, based on direct genetic testing, in patients at risk for the multiple endocrine neoplasia type 2 syndromes. Exp Clin Endocrinol Diabetes 1998;106:29-34. 11. Robledo M, Gil L, Pollan M, et al. Polymorphisms G691S/ S904S of RET as genetic modifiers of MEN 2A. Cancer Res 2003;63:1814-7. 12. Borrego S, Eng C, Sanchez B, et al. Molecular analysis of the ret and GDNF genes in a family with multiple endocrine neoplasia type 2A and Hirschsprung disease. J Clin Endocrinol Metab 1998;83:3361-4. 13. Pasini B, Rossi R, Ambrosio MR, et al. RET mutation profile and variable clinical manifestations in a family with multiple endocrine neoplasia type 2A and Hirschsprung’s disease. Surgery 2002;131:373-81. 14. Borrego S, Saez ME, Ruiz A, et al. Specific polymorphisms in the RET proto-oncogene are over-represented in patients with Hirschsprung disease and may represent loci modifying phenotypic expression. J Med Genet 1999;36:771-4.
DISCUSSION Dr Henning Dralle (Halle, Germany). In your abstract, you stated that 27% of patients had codon 634 mutations. Are these all index cases? What is the reason for the high percentage of patients with 634 mutations? You said the biochemical cure rate was the same at >50 years of age or at <50 years of age. If this is true, what is the reason for that? Did you analyze double mutations in the afflicted families? These double mutations might explain the change in tumor biology and age of evaluation. Dr Quayle. With regard to the second question about biochemical cure, we did find similar rates between the groups. Our hypothesis would be that this is due to modifiers, that these patients had indolent cancers and not advanced disease despite their late age at evaluation and therefore would be in the same time-frame that enables a cure with thyroidectomy alone. Regarding your last question, we did not analyze and do not have any experience, to my knowledge, with double mutations. Dr Dralle. So the biochemical cure rate was the same in patients who were >50 years of age compared with patients who were <50 years of age? Dr Quayle. It was approximately 40% to 50% in both groups. Dr Dralle. But the tumor stages were higher in the patients who were >50 years of age? Dr Quayle. I think it probably speaks to a lack of power. There were 30 patients in that group versus 360 patients in the other group, so I think that we probably did not have the power to determine whether that was a statistically significant difference. Dr Dralle. Is the high percentage of 634 patients, approximately one third of the patients, because of a high number of index cases or were these patients screening cases?
Surgery Volume 136, Number 6
Dr Quayle. Our cohort of patients who were >50 years of age tended to have fewer patients with 634 mutations, but our experience, and I think it corroborates with the experience that other authors have had, is that 634 mutations are the most prevalent in MEN2A. Dr Paul G. Gauger (Ann Arbor, Mich). In the process of defining your study group, did you encounter any patients that were not included because they were the only ones with that mutation and you were considering it a polymorphism? Specifically, I have a question about codon 666. I take care of a father and son who would have fit into your study group by all other criteria, but their mutation (codon 666) has not really been implicated in the same way as the others that you mention. Dr Quayle. There was one patient in whom, to date, the mutation has not been identified. This patient was examined before the extensive analysis of the other exons, so the family’s mutation has not been identified. It seems as if every year more and more mutations that are associated with this indolent disease are described. These atypical mutations are less common and harder to find. Moreover, the disease may be so indolent that it is not 100% penetrant, and we miss generations; therefore,
Quayle et al 1121
these cases are not identified as having a hereditary component. I suspect there are a lot of patients out there like your patient with less common mutations that are causing more indolent disease. Dr Douglas B. Evans (Houston, Tex). To summarize, would your data agree with the level classification in the Brandi article (Brandi et al. JCEM 2001;86:5658-5671) or would you disagree? If so, I assume it would be a level 2 mutation that you would consider changing, or were there no mutations for which you would change the level classification? Dr Quayle. I would say that our paper did not address specifically the issues of the timing of thyroidectomy when you identify a patient early. In our experience, if you recall, the penetrance plot of mutations, 609 mutations (if I remember correctly), were classified in the level 3 mutations in the Brandi conference. In our experience, we did not see cases earlier than age 20 years for those particular mutations. I do not think that there is anything to suggest that we know which patients are going to fit into this older cohort at this point. So, I think that we have no data to suggest that those recommendations are inappropriate at this time.