Page 1 of 24
Accepted Preprint first posted on 16 December 2014 as Manuscript EJE-14-0818
1
RET Cys634Arg mutation confers a more aggressive multiple endocrine neoplasia type 2A
2
phenotype than Cys634Tyr mutation
3 4
Nuria Valdés1, Elena Navarro2, Jordi Mesa3, Anna Casterás3, Victoria Alcázar4, Cristina
5
Lamas5 , Javier Tébar6 , Luis Castaño7, Sonia Gaztambide8, Lluís Forga.9
6 7 8 9 10 11 12
1. Department of Endocrinology and Nutrition. Hospital Universitario Central de Asturias, Oviedo, Spain. 2. Department of Endocrinology and Nutrition. Hospital Universitario Virgen del Rocío, Sevilla, Spain. 3. Department of Endocrinology and Nutrition. Hospital de Vall d´Hebron, Barcelona, Spain.
13
4. Department of Endocrinology and Nutrition. Hospital Severo Ochoa, Madrid, Spain.
14
5. Department of Endocrinology and Nutrition. Complejo Hospitalario Universitario de
15 16 17 18 19 20 21 22
Albacete, Albacete, Spain. 6. Department of Endocrinology and Nutrition. Hospital Universitario Virgen de la Arrixaca, Murcia, Spain. 7. Endocrinology and Diabetes Research Group. Hospital Universitario de Cruces, Barakaldo, Spain. 8. Department of Endocrinology and Nutrition. Hospital Universitario de Cruces, Barakaldo, Spain. 9. Department of Endocrinology. Complejo Hospitalario de Navarra, Pamplona, Spain.
23
24
Corresponding author and requested reprints:
25
Dr. Nuria Valdés. Department of Endocrinology and Nutrition. Hospital Universitario Central
26
de Asturias. Avda. de Roma s/n, E-33011-Oviedo, Asturias, Spain. Phone: (+34)-985-108000.
27
Fax: (+34)-985-107954. E-mail:
[email protected]
28
1 Copyright © 2014 European Society of Endocrinology.
Page 2 of 24
29
Short tittle: RET C634R mutation is more adverse than C634Y.
30
Keywords: Multiple Endocrine Neoplasia Type 2 A, RET C634R mutation, RET C634Y
31
mutation, Medullary Thyroid Carcinoma, Pheochromocytoma, Hyperparathyroidism, RET-
32
proto-oncogene.
33
Word count: 4174
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
2
Page 3 of 24
57
Abstract
58 59
Objective: Specific germline mutations in the RET proto-oncogene are correlated with clinical
60
features in multiple endocrine neoplasia type 2A (MEN 2A); however, data are scarce regarding
61
differences in clinical profiles dependent on the type of nucleotide and amino acid substitution
62
at the same codon. We aimed to analyse differences in clinical risk profiles and outcomes
63
between different amino acid encoded by codon 634.
64
Design: The study was retrospective and multicentric.
65
Methods: We collected data included in the Spanish Online National Database from MEN 2A
66
patients carrying a RET-proto-oncogene mutation on codon 634. The mean follow-up time was
67
7.6 ±6.9 years (1−32).
68
Results: Patients (173) from 49 unrelated families were C634Y carriers, and 26 patients from 8
69
different families had C634R mutation. We found higher penetrance of medullary thyroid
70
carcinoma, pheochromocytoma, and hyperparathyroidism (P < 0.001, P = 0.007, P < 0.001,
71
respectively) in C634R carriers than in C634Ycarriers. The Kaplan−Meier estimate of
72
cumulative lymph node and distant metastases rates showed that these events occurred earlier in
73
individuals harbouring the C634R mutation (P < 0.001). A multivariate adjusted Cox regression
74
analysis showed that the C634R mutation was an independent factor for persistent/recurrent
75
disease (hazard ratio, 3.17; 95% confidence interval: 1.66−6.03; P < 0.001).
76
Conclusions: Our results suggest that there could be clinical differences caused by different
77
amino acid substitutions at codon 634; specifically, the C634R mutation was associated with a
78
more aggressive MEN 2A phenotype than the C634Y mutation.
79 80
3
Page 4 of 24
81
Introduction
82 83
Multiple endocrine neoplasia type 2A (MEN 2A) is a rare syndrome characterized by
84
the presence of medullary thyroid carcinoma (MTC; 95% of MEN 2A patients),
85
pheochromocytoma (PHEO: 30%–50%), and hyperparathyroidism (HPT: 10%–20%) (1,2).
86
This syndrome is caused by a germline activation mutation in the REarranged during
87
Transfection (RET) proto-oncogene transmitted by an autosomal dominant inheritance (3,4).
88
The RET proto-oncogene encodes a transmembrane tyrosine kinase receptor which comprises a
89
highly conserved, cysteine-rich, extracellular domain, 2 intracellular tyrosine kinase domains,
90
and an intracellular catalytic core (3,5). In the majority of MEN2A families (more than 90%),
91
germline mutations affect the cysteine-rich extracellular domain by converting a cysteine into
92
another amino acid, and this mutation determines RET spontaneous dimerization and activation
93
(1-5). These mutations are located on codon 634 (exon 11) or codons 609, 611, 618, and 620
94
(exon 10).
95
Genotype-phenotype correlation has been described between specific codon mutation
96
and age of onset, penetrance and clinical aggressiveness of MTC (1,2,6,7). Based on these
97
findings the American Thyroid Association (ATA) Task Force has published management
98
guidelines (8) on the timing and extension of prophylactic thyroidectomy establishing four
99
different risk categories (from A, the lowest, to D, the highest). There is also a different
100
prevalence of PHEO and HPT according to the type of specific codon mutation (7,9,10).
101
The most common mutation, accounting for 34% of European MEN2A families, affects
102
codon 634 (11). “In vitro” assays have demonstrated that the codon 634 mutation had a higher
103
probability of transforming NIH3T3 cells than other RET codon mutations (12). This biological
104
characteristic of the highest transforming activity correlates with phenotype because codon 634
105
mutations are associated with the more severe forms of the MEN 2A phenotype and it is
106
classified as class C in the ATA guidelines (8).
107
There are six different amino acid substitutions for the same cysteine on codon 634,
108
and all of them displayed comparable transforming activity in one “in vitro” assay (12),
4
Page 5 of 24
109
indicating that the degree of the activity depends on the position of cysteine mutations rather
110
than the substituted amino acids. However, the assay was performed only in fibroblasts; thus the
111
biological effects of the substituted amino acids are not known in thyroid C cells, adrenal
112
medullary cells, and parathyroid cells. From a clinical viewpoint, few reports have indicated
113
whether there is any difference in the clinical profiles based on amino acid substitutions on
114
codon 634 (4,6,9,10,13).
115
The aim of our study was to evaluate any potential differences in clinical risk profiles
116
and outcomes in Spanish patients caused by different mutations on codon 634. This information
117
might be useful for tailoring more accurate clinical care.
118
Cys634Arg (C634R) is the most common RET mutation in European countries (14),
119
except in Italy where Val804Met is more common (15). However, the most common RET
120
mutation in MEN 2A Spanish patients is Cys634Tyr (C634Y) (16). Consequently, we examined
121
the largest number of patients ever reported with this specific mutation on codon 634, which
122
provided a firmer basis for statistical analyses.
123 124
Subjects and methods
125 126
The Spanish Group for the Study of MEN and Pheochromocytomas/Paragangliomas,
127
which depends on the Spanish Endocrinology and Nutrition Society, designed an online
128
national database to collect extensive clinical and genetic information about Spanish MEN 1,
129
MEN 2A, MEN 2B, and pheochromocytomas/paragangliomas patients. Informed consent was
130
obtained for all diagnostic and therapeutic procedures and for the inclusion of family and
131
personal clinical data in the Register. The anonymity of patients was not breached; therefore,
132
the study met the conditions required under Spanish law for making medical and or personal
133
data available for scientific research.
134
Data from patients diagnosed with MEN 2A, according to international clinical
135
guidelines (1), were included in the database from January 2009 to December 2011 and were
5
Page 6 of 24
136
analysed in the course of 2012. All of these patients were diagnosed and clinically examined by
137
the authors. The study was retrospective and multicentric.
138
Briefly, the data includes gender, birth date, family history, and type of diagnosis (index
139
cases and diagnosis by genetic or biochemical screening), the type of RET mutation, and the
140
dates of diagnosis for the syndrome and for each lesion. The age at MEN2A syndrome
141
diagnosis was considered the age at the second manifestation of any clinical features in index
142
cases and the first manifestation or genetic diagnosis in familial cases. The age at diagnosis of
143
clinical disease was considered the age at histological diagnosis for MTC and biochemical
144
diagnosis for PHEO and HPT.
145
For thyroid disease, the following characteristics were recorded: calcitonin (CT) basal
146
and/or pentagastrin-stimulated levels (positive or negative), extent of surgery, and outcome after
147
thyroidectomy. With respect to thyroid histopathology, we recorded the presence or absence of
148
C-cell hyperplasia (CCH) and MTC. For lymph node status, we collected data on lymph node
149
dissection as N0, N1, and Nx (non-conclusive data). For analysis, the N0 classification was used
150
to describe N0 or Nx data. The same approach was applied for distant metastases, which were
151
recorded at thyroid surgery. Tumour stage was assessed according to the 2002 UICC TNM
152
classification (17). Patients were considered biochemically free of the disease at the last follow-
153
up, according to international guidelines.
154 155 156 157
For PHEO, we collected data about the presence or absence of symptoms at diagnosis, uni- or bilateral involvement, type of surgery, and clinical outcome. For HPT, we included the presence or absence of clinical symptoms, type of surgery, and clinical outcome.
158 159
Statistical analysis
160 161
Categorical variables were summarized as frequency counts and percentages.
162
Continuous variables were summarized as the mean ± standard deviation unless otherwise
163
specified, and range. For continuous variables, normality was assessed with the Kolmogorov-
6
Page 7 of 24
164
Smirnov test, and logarithmic transformations were applied as necessary to ensure normal
165
distribution. Baseline characteristics were compared using the 2 test or Fisher’s exact test for
166
qualitative variables, or the Student’s t test or Mann-Whitney’s U test for quantitative variables.
167
The Kaplan−Meier estimate for age-related penetrance of MTC, PHEO, HPT, lymph nodes and
168
distant metastases, and for biochemically free of disease were compared among type of amino
169
acid substitution using the long rank test. Risk factors for disease-free survival were assessed
170
via the chi-square test or Fisher´s exact test and we also performed a Cox regression model.
171
Five candidate risk factors were assessed: gender, age at the time of surgery, basal and/or
172
pentagastrin-stimulated calcitonin levels, lymph nodes metastases, and type of amino acid
173
substitution. Associations were expressed as hazard ratios (HR) with their 95% confidence
174
intervals (95% CI). Results were considered statistically significant at P