|
|
||||||||
1 Departments of Endocrinology and Diabetology and 2 Pathobiochemistry and Clinical Biochemistry, Collegium Medicum in Bydgoszcz, Nicolaus Copernicus University in Torun, ul. M. Sklodowskiej-Curie 9, 85-094 Bydgoszcz, Poland and 3 Department of Laboratory Diagnostics, Oncology Centre in Bydgoszcz, ul. dr I.Romanowskiej 2, 85-796 Bydgoszcz, Poland
(Correspondence should be addressed to J Klubo-Gwiezdzinska who is now at Klinika Endokrynologii i Diabetologii, Szpitala Uniwersyteckiego im. A. Jurasza w Bydgoszczy, ul. M. Sklodowskiej-Curie 9, 85-094 Bydgoszcz, Poland; Email: adamgw{at}poczta.onet.pl)
| Abstract |
|---|
|
|
|---|
Aim: We hypothesized that serum VEGF concentration would differ between patients with metastatic and non-metastatic thyroid cancer, multinodular goiter, and healthy subjects. We also hypothesized that endogenous TSH stimulation would affect serum VEGF level.
Subjects and methods: The study group consisted of 71 patients (62 females and 9 males), aged 44.9 ± 12.3 years, with differentiated thyroid cancer (50 papillary, 17 follicular, and 4 oxyphilic), treated in our department during the years 2003–2006. All patients had undergone total or near-total thyroidectomy and radioactive iodine treatment, that had resulted in remission in 59 patients and persistent/recurrent disease in 12 patients. The study included two control groups: 30 patients with nontoxic multinodular goiter and 30 healthy subjects.
Results: Serum VEGF concentrations were significantly higher in patients with distant metastases than those in remission or healthy patients. (423.4 vs 217.6 vs 235.55 pg/ml respectively, P < 0.05). This was not observed in patients with locoregional metastases. During endogenous TSH stimulation, VEGF decreased significantly (215.3 vs 169.6 pg/ml, P < 0.05). Patients with multinodular goiter showed significantly lower VEGF concentrations than the remaining study groups.
Conclusions: Serum VEGF concentration might be used as an additional marker of thyroid cancer with distant metastases, but its interpretation should be undertaken very cautiously. Endogenous TSH stimulation decreases VEGF levels in patients either with or without thyroid tissue, suggesting that its regulatory effects are through receptors located outside the thyrocytes.
| Introduction |
|---|
|
|
|---|
Numerous studies have shown higher VEGF expression in thyroid cancer tissues than in benign thyroid nodules and normal thyroid tissues (7–12).
Some authors have suggested using high VEGF tissue expression as a prognostic factor for locoregional or distant recurrence or poor prognosis (7, 8, 10, 11, 13, 14).
The measurement of VEGF tissue expression demands complicated immunohistochemical procedures. A more useful tool for clinicians is the estimation of serum VEGF levels. There are few studies of serum VEGF levels in patients suffering from thyroid cancer (15, 16). There are no data comparing serum VEGF levels of patients suffering from multinodular goiter to healthy subjects.
Very few studies describe the effect of recombinant human thyroid-stimulating hormone (rhTSH) stimulation on VEGF production by thyroid cancer cells. The results of in vitro studies are contradictory to those in vivo (15–19). There is no data relating to the effect of endogenous TSH stimulation on VEGF production.
Therefore, we conducted a study aimed at investigating four hypotheses: first, whether serum VEGF concentrations are elevated in metastatic thyroid cancer; secondly, whether serum VEGF concentration in patients with remission is similar to the healthy subjects; thirdly, if endogenous TSH stimulation results in a change in serum VEGF concentration in patients with differentiated thyroid cancer; and finally if serum VEGF concentrations of patients with multinodular goiter differs from patients with thyroid cancer and healthy subjects.
| Materials and methods |
|---|
|
|
|---|
The study included 91 patients treated in the Department of Endocrinology and Diabetology in Bydgoszcz, Nicolaus Copernicus University in Torun and the centre for Oncology in Bydgoszcz, during the years 2003—2006.
The inclusion criteria for the study were as follows: a well-differentiated thyroid cancer, diagnosed by an experienced histopathologist after total or near total thyroidectomy; radioactive iodine ablation; no evidence of remnant tissue in the thyroid bed uptake; no other known malignancies; normal platelet count.
From this group, we excluded 20 patients suffering from diabetes mellitus, hypertension, or heart failure because of their potential effect on the VEGF serum concentration (20).
The acquired group of 71 patients underwent a standard follow-up procedure: first, diagnostic whole body scintigraphy (WBS), thyroglobulin (Tg), anti-Tg antibodies and ultrasonography (USG) of the neck after endogenous TSH stimulation (TSH > 30 µIU/ml), 7–8 months after 131I treatment; secondly, the frequencyof the same procedure: in the case of full remission, after 2 years, then every 5 years; in patients with persistent/recurrent disease, 7–8 months after the additional treatment with radioiodine; and in the case of high-risk patients, every year; thirdly, systematic physical examination, TSH, and Tg measurements every 6 months during TSH suppression (TSH < 0.05 µIU/ml).
Full remission was defined as follows: Tg levels during TSH stimulation < 2 ng/ml, or spontaneous fall of stimulated Tg below this level, during control provided after endogenous stimulation; absence of anti-Tg antibodies; negative WBS with 24-h uptake < 0.1%; normal results of the standard imaging procedures: USG of the neck, X-ray of the chest, or other (computed tomography, MRI, and PET, if needed).
The study included two control groups of 30 patients each.
The inclusion criteria for the first control group (healthy subjects) were: normal USG of the neck; TSH and platelet count within normal range; no evidence of chronic or acute diseases.
The inclusion criteria for the second control group (patients with nontoxic multinodular goiter) were: nodular goiter confirmed by exact cytologic analysis after fine needle aspiration cytology (FNAC); TSH and platelet count within normal range; no evidence of chronic or acute diseases. The exclusion criteria for the second group were: rapid increase in size of the nodule, firm consistency, irregularity, adhesion to soft tissues, or neck lymphadenopathy. Patients with nontoxic multi-nodular goiter were not treated with L-thyroxine (L-T4) or iodine.
In all groups, we measured VEGF serum concentrations.
Serum samples
Blood samples were obtained as a part of routine clinical evaluations (TSH, Tg, anti-Tg antibodies, and blood morphology). They were centrifuged for 15 min at 2500 g, 30 min after donation. Serum was removed immediately and stored at < –20 °C for 1–4 months.
Serum VEGF measurements
Serum VEGF measurements were performed in batches using the Quantikine Human VEGF sandwich enzyme immunoassay technique (R&D Systems, Minneapolis, MN, USA).
The intra-assay precision, expressed as coefficients of variation, was 4.5–6.7%; the inter-assay precision –6.2–8.8%; and the sensitivity < 5 ng/ml. All assays were performed in duplicate.
Serum Tg measurements
Serum Tg was measured with an IRMA assay (LIAISON, DiaSorin, Saluggia, Italy). The detection limit was < 0.2 ng/ml. Reference values assuming patients with an intact thyroid were 0.2–70 ng/ml.
Serum anti-Tg antibodies measurements
Serum anti-Tg antibodies were measured with an IRMA assay (LIAISON, DiaSorin). Detection limit was < 5 IU/ml. Reference values were 5–100 IU/ml.
Serum TSH measurements
Serum TSH was measured with an ultrasensitive CMIA assay (ARCHITECT YOU 8200, Abbott Laboratories). Detection limit was < 0.0025 µIU/ml. Reference values were 0.35–4.94 µIU/ml.
USG and FNAC
The USG of the neck was performed by an ultrasonic probe of 7.5 MHz. The volume of the thyroid was obtained using the following formula: (21)
![]() |
where V is the thyroid volume and a, b, and c are thyroid dimensions.
FNAC was performed under USG control by an experienced cytologist. The biopsy was taken from the nodules > 1 cm, and/or solid, hypoechogenic, with microcalcifications, without peripheral halo.
Statistical analysis
Normally distributed data are presented as mean ± S.D., and not-normally distributed data are given as median. For normally distributed data, the Student t-test was used. Non-parametric tests were used when the data were not normally distributed. The groups were compared using Mann–Whitneys U test or ANOVA Kruskal–Wallis test. Relationships among variables were sought using Spearmans correlation coefficient. Paired data were analyzed using Wilcoxons test. Statistical significance was assumed when P value was
0.05.
| Results |
|---|
|
|
|---|
Histological classification revealed 50 (70.4%) papillary thyroid cancers, 17 (24%) follicular thyroid cancers, and 4 (5.6%) oxyphilic thyroid cancers.
Among the group, 59 (83.1%) patients reached the remission criteria. The remaining 12 (16.9%) patients showed biochemical and morphological evidence of metastatic disease (Table 1
).
|
Concentrations of VEGF in each group are presented in Table 2
.
|
|
|
|
Patients with nontoxic nodular goiter showed significantly lower VEGF concentrations than the healthy controls (P < 0.01; Fig. 2
) and other groups (P < 0.05).
|
During endogenous TSH stimulation, in 40 patients with thyroid cancer, VEGF significantly decreased (Fig. 3
). This was observed in metastatic as well as non-metastatic patients. There was no correlation between VEGF and Tg during suppression and stimulation, in patients with metastatic disease. The VEGF level was not dependent on the number of metastatic foci.
|
| Discussion |
|---|
|
|
|---|
There are two previous studies with a similar aims to ours. Tuttle et al. demonstrated significantly increased serum VEGF level in patients with metastatic thyroid cancer in comparison with patients in remission. The observation related to thyroid cancer with distant metastases, which was similar to our results, as well as thyroid cancer with locoregional metastases, which was not confirmed by this study. Sorvillo et al. also showed increased serum VEGF levels in patients with metastases, but they did not analyze the effect of the location of metastases (15, 16).
There are two possible reasons for the discrepant results referring to the patients with locoregional metastases: Firstly the proportion of patients with locoregional metastases might have influenced the statistical analysis: in our study, it was 10% (n = 7 out of 71); in Tuttles, it was 24% (n = 15 out of 63); and in Sorvillos, the metastases were not divided to locoregional and distant, but they were observed in 40% of patients (n = 18 out of 45).
Differences in exclusion criteria are the other possible reason. In our study, patients with hypertension, diabetes, heart failure, abnormal platelet count, and known malignancies other than thyroid cancer were excluded because of their potential effect on the VEGF level (2, 20). The authors of the studies presented above described other known malignancies and abnormal platelet count as exclusion criteria.
All studies have confirmed similar VEGF levels in patients in remission and healthy subjects.
In addition, histopathological studies are not univocal regarding VEGF expression in thyroid cancer tissues with locoregional metastases (7–10, 22) Several studies, similarly to our observations, did not demonstrate elevated VEGF expression in thyroid cancers with locoregional metastases. Interestingly, in these cases, specific isoforms of VEGF (VEGF-C and VEGF-D), responsible for lymphangiogenesis, were increased (23–25).
The assessment of serum VEGF-C and VEGF-D was not the aim of our study, but this is worth further research because it might be an explanation for the discrepancies between the presented results.
The clinical usefulness of serum VEGF level has some limitations: there are overlapping data between all the observed subgroups. This refers mainly to the interpretation of low VEGF levels – in some cases, they are observed even in patients with distant metastases (Table 3
).
Since serum VEGF level does not correlate with the Tg level and the number of metastatic foci, it cannot be used as a marker of the extension of persistent/recurrent disease.
VEGF might be used as an additional marker in patients with a thyroid cancer resistant to the standard treatment. High VEGF concentrations in these patients may be a target of a novel treatment with anti-VEGF antibodies or VEGF receptor inhibitors (26–29).
We observed a significant decrease in serum VEGF levels after endogenous TSH stimulation in patients with metastatic disease and remission. According to our knowledge, the influence of endogenous stimulation on VEGF levels is described here for the first time. The studies of Tuttle et al. and Sorvillo et al. were focused on an exogenous rhTSH stimulation and have shown different results – stable VEGF level after stimulation (15) or time-dependent significant decrease (16).
We observed a significant decrease in VEGF levels, even in patients in remission, i.e. those without clinical or biochemical evidence of disease. These observations suggest that TSH regulates VEGF synthesis through receptors not exclusively located in follicular thyroid cells. The expression of TSH-receptors in tissues other than the thyroid gland was confirmed by Sorisky et al. (30).
Some of the in vitro studies are in agreement with ours, showing that TSH down-regulates the VEGF production by thyroid cells grown in culture (31, 32). However, other results did not confirm this observation (17, 33, 34).
The other possible reason for the decrease of VEGF levels in both groups (patients with and without remission) is hypothyreosis after withdrawal of L-T4, interpreted as an additional disease which might have influenced the VEGF level.
We observed significantly lower VEGF levels in patients with multinodular goiter in comparison with the patients with thyroid cancer and the healthy ones.
The observation that VEGF levels are lower in patients with benign nodules when compared to the patients with active malignant neoplastic disease was not surprising. Several studies have documented similar observations (7, 11, 12).
What was surprising was that our study documented lower serum VEGF levels in patients with multinodular goiter than in healthy ones. Again, to the best of our knowledge, this is the first report on this theme.
We can exclude potential inhibition of VEGF synthesis by iodine, which has been documented by an in vitro study in cultured human thyroid follicles (19). None of the patients with multinodular goiter received iodine or iodine-containing L-T4.
A possible reason might be the crinopexy phenomenon, described by Gerard et al. They observed the trapping of VEGF in association with the extracellular matrix of goitrous thyroid, while the normal thyroid tissue expressed VEGF in epithelial cells (35).
This trapping might be an explanation for the decreased serum VEGF levels in patients with multi-nodular goiter observed in our study.
| Conclusions |
|---|
|
|
|---|
| Funding |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Joao Bugalho, D. Madureira, C. Espadinha, A. Paula Font, and L. G Sobrinho Serum vascular endothelial growth factor levels in patients with medullary thyroid carcinoma Eur. J. Endocrinol., August 1, 2008; 159(2): 167 - 169. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |