Medullary thyroid cancer (MTC) accounts for 1–3% of all cases of thyroid carcinoma and arises from parafollicular cells (C cells). About 20% of the cases are hereditary, as part of type 2 multiple endocrine neoplasia syndrome. At diagnosis, about 70–80% of the cases present with cervical lymph node (LN) metastases and 10% of cases also present with distant metastases [
All patients with a preoperative diagnosis of MTC should undergo neck ultrasound (US) and measurement of serum calcitonin (CT) [
There are a few other additional systemic and local therapies used to achieve tumor control, but their efficacy is limited. Therefore, complete surgical resection is of utmost importance for all MTCs and successful surgery with clear surgical margins, and removal of all LN metastases is a prerequisite for cure [
During follow-up of MTC patients and in case of raising levels of serum CT, imaging exams should be performed in order to diagnose local recurrence, LN metastases, or distant metastases [
Patients with DFTC also have an additional tool for the diagnosis of LN metastases: measurement of thyroglobulin (TG) in FNA washout fluid. The ATA guidelines for the management of differentiated thyroid cancer recommend using both FNA cytology and measurement of TG in the washout fluid for confirmation of malignancy in suspicious LN, with a cut-off of <1 ng/mL for the exclusion of LN metastases [
A retrospective analysis of our institutional database identified 69 consecutive FNA cytology and CT measurement in FNA washout of LN (from 42 patients) between 2012 and 2017. Measurement of FNA-CT in the washout of LN was performed in nonthyroidectomized patients evaluated in our department for suspicious neck LN and in patients with personal history of MTC, suspicious neck LN, and progressive increase in serum CT levels. In nonthyroidectomized patients, both TG and CT were measured in the LN washout fluid to exclude LN metastases from a thyroid carcinoma, and in patients with personal history of MTC, only CT was measured as an adjunct to cytologic examination. We did not measure FNA-CT in the washout fluid of thyroid nodules.
All patients underwent a thorough neck examination, neck US, and US-guided FNA of suspicious LN for conventional cytology and measurement of FNA-CT. Neck US was performed using a Canon XarioTM 100 US machine, with a 4.2–14 mHz linear electronic transducer, by an endocrinologist with at least 5 years of experience. All LNs and cervical masses were identified and localized, and their diameters were measured. Suspicious LNs were identified according to standard criteria (loss of echogenic hilum, hyperechogenicity, cystic changes, calcifications, round shape, and abnormal vascularity) and submitted to one FNA cytology under US visual guidance with a 22 or 25-gauge needle attached to a 10 mL syringe [
Cytological examination was performed by two thyroid pathologists with at least 10 years of experience, and the diagnosis of MTC was made according to standard criteria [
After smear preparation, the syringe and needle were sent to the laboratory department to be flushed with 1 mL of normal saline solution (NaCl 0.9% m/V). The needle washout was centrifuged (1500 g, 2 min), and the supernatant, which is not the cellular material, was used for analysis. The solution was processed for FNA-CT measurement with a solid-phase, enzyme-labeled, two-site chemiluminescent immunometric assay (Siemens Healthcare Diagnostics, Llanberis, Gwynedd LL55 4EL, United Kingdom, Immulite 2000XPI Calcitonin), calibrated to the World Health Organization 2nd International Reference Preparation 89/620 standard. The limits of detection and quantification were 2.0 and 5.0 pg/mL, respectively, whereas the upper reference limits were 11.5 pg/mL for females and 18.2 pg/mL for males [
Histological analysis of surgical specimens was made using standard pathological techniques, including detection of CT expression by immunohistochemistry.
In this study, we included demographic (age and gender), laboratory (serum CT and FNA-CT), clinical (age and gender), and pathologic data (results from FNA cytology and histological analysis of LN) extracted from patients’ charts.
All statistical analysis was performed using SPSS Statistics 23.0 software (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY) Parametric statistical tests (
This study’s population had a mean age of 60.7 ± 15.9 years and was mainly comprised of male patients (
Calcitonin was detectable in needle washout in 23 (33.3%) FNA. In patients with detectable levels of FNA-CT, median was 635 pg/mL (interquartile range 277–2524). These patients had a median serum CT level of 2014 (490–15111) pg/mL and a mean LN diameter of 19.7 ± 10.7 mm. These results are shown in Tables
Patients characteristics according to FNA-CT measurement.
FNA-CT measurements |
|
|||
---|---|---|---|---|
Detectable |
Undetectable | |||
Gender | Female | 4 | 15 | NSS |
Male | 6 | 17 | ||
|
||||
Age (years) | 63.2 ± 16.7 | 57.9 ± 14.2 | NSS | |
Personal history of MTC | 7 | 6 | NSS | |
FNA-CT levels (pg/mL) |
2014 (490 – 15111) | <2 (0) | <0.001 | |
Serum CT (pg/mL) |
635 (277 – 2524) | 1 (1 – 277) | 0.016 | |
LN diameter (mm) | 19.7 ± 10.7 | 16.2 ± 6.1 | 0.014 |
FNA-CT levels and serum CT levels expressed as median (interquartile range); age and LN diameter expressed as mean ± standard deviation; NSS = no statistical significance.
Patients with detectable levels of FNA-CT in LN washout fluid (available as a supplementary file).
Personal history | LN (size, mm) | FNAC | FNA-CT (pg/mL) | Serum CT (pg/mL) | Histology | |
---|---|---|---|---|---|---|
Patient 1 | MTC | LN1 (18) | MTC LN metastasis | 200000 | 86.9 | MTC LN metastasis |
Patient 2 | MTC | LN 1 (17.4) | Nondiagnostic | 704 | 932 | MTC LN metastasis |
Patient 3 | MTC | LN 1 (15.5) | Nondiagnostic | 1086 | 533 | MTC LN metastasis |
Patient 4 | MTC | LN1 (15) | Nondiagnostic | 312 | 9704 | MTC LN metastasis |
Patient 5 | MTC | LN1 (16) | MTC LN metastasis | 85621 | 635 | MTC LN metastasis |
Patient 6 | MTC | LN 1 (25) | Benign | <2 | 794 | — |
<2 | ||||||
LN 2 (40) | MTC LN metastasis | 11065 | 3796 | MTC LN metastasis | ||
40226 | ||||||
LN3 (15) | MTC LN metastasis | 4913 | 121 | MTC LN metastasis | ||
LN 4 (12) | Benign | <2 | 340 | — | ||
<2 | ||||||
Patient 7 | Nonthyroidectomized | LN 1 (45) | MTC LN metastasis | 15111 | 1871 | MTC LN metastasis |
MTC | LN 2 (14) | MTC LN metastasis | 6779 | 1224 | MTC LN metastasis | |
60.2 | 1224 | |||||
Patient 8 | MTC | LN 1 (13.2) | MTC LN metastasis | 117 | 277 | MTC LN metastasis |
2014 | ||||||
LN 2 (18.8) | MTC LN metastasis | 4653 | MTC LN metastasis | |||
LN 3 (12) | Benign | <2 | 545 | — | ||
<2 | ||||||
LN 4 (14) | Benign | <2 | — | |||
LN 5 (18) | MTC LN metastasis | 40.4 | 598 | MTC LN metastasis | ||
490 | ||||||
LN 6 (13) | MTC LN metastasis | 838 | 515 | MTC LN metastasis | ||
1686 | ||||||
Patient 9 | Nonthyroidectomized | LN 1 (25.5) | DTFC | 5.6 | 7 | DFTC LN metastasis |
Patient 10 | Nonthyroidectomized | LN 1 (25) | MTC LN metastasis | 7364 | 5007 | MTC LN metastasis |
66104 | ||||||
MTC | LN 2 (28.4) | MTC LN metastasis | 200000 | 1355 | MTC LN metastasis | |
LN 3 (27) | MTC LN metastasis | 1488 | 2524 |
|
Distribution of MTC LN metastases according to FNA-CT measurements.
Very high levels of FNA-CT (>200 pg/mL) were seen in most cases (
Lymph node resection was performed in 17 patients (36 lesions, 52.1%): 9 patients (24 lesions, 34.7%) with detectable levels of FNA-CT and 8 patients (12 lesions, 17.3%) with undetectable FNA-CT. Patients with undetectable FNA-CT underwent surgery for Bethesda V on thyroid nodule cytology and LN cytology suspicious for DFTC LN metastasis. Lymph node excisional biopsy was additionally performed in 7 patients with suspicious LN on neck US but undetectable levels of FNA-CT and benign cytology. Histological examination of these LN proved them to be granulomatous disease.
Histology reported LN metastases of MTC in 8 patients (19 lesions, 52.8%) and LN metastases of DFTC in 10 patients (12 lesions, 33.3%). All but one patient with DFTC had very low levels of FNA-CT, and it was the only false-positive (FP) in our study. He had a FNA-CT level of 5.6 pg/mL, a FNA cytology suspicious for papillary thyroid carcinoma, and a serum CT of 7 pg/mL. There was only one case of positive FNA-CT (1488 pg/mL) and MTC LN metastases on cytologic examination, without histological confirmation. It was a patient with known progressive MTC with two previous lateral-cervical dissections. He was diagnosed with LN recurrence and pulmonary metastases, and it was decided not to perform an additional LN dissection, but to start systemic therapy instead. These results are shown in Figure
Histologic examination of LN, according to FNA-CT levels.
Regarding the diagnosis of MTC LN metastases, FNA-CT showed sensitivity of 100%, specificity of 97.9%, positive predictive value (PPV) of 95.6%, and negative predictive value (NPV) of 100%. There was only one FP, as was previously mentioned. On the other hand, FNA cytology had a sensitivity of 81.8%, specificity of 97.9%, PPV of 94.7%, and NPV of 92%. This method had only one FP, which was a patient with personal history of MTC with FNA-CT of <2 pg/mL, in which the pathology report confirmed no signs of malignancy.
FNA cytology was considered nondiagnostic in four cases (5.8%). Three of them had personal history of MTC and positive FNA-CT levels (312 pg/mL, 704 pg/mL, and 1086 pg/mL, respectively). They also had rising levels of serum CT, which increased the suspicion of metastatic disease. In these cases, high levels of FNA-CT in the washout lead to the diagnosis of LN metastases. Surgery was later performed, which confirmed the diagnosis of MTC LN metastases by histological analysis.
Combining both diagnostic methods showed the highest diagnostic power (sensitivity of 100%, specificity of 97.9%, PPV of 95.6%, and NPV of 100%). The lowest value of FNA-CT with histologically proven MTC LN metastases was 40.4 pg/mL. The optimal cut-off value of FNA-CT for determination MTC LN metastases through ROC analysis was 23.0 pg/mL (area under the curve = 99%;
Diagnostic accuracy of FNA cytology and FNA-CT.
MTC | Non-MTC | Nondiagnostic | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | |
---|---|---|---|---|---|---|---|
FNA cytology | 19 (27.5%) | 46 (66.6%) | 4 (5.8%) | 81.8 | 97.9 | 94.7 | 92.0 |
FNA-CT | 23 (33.3%) | 46 (66.6%) | — | 100 | 97.9 | 95.6 | 100 |
FNA cytology + FNA-CT | 23 (33.3%) | 46 (66.6%) | — | 100 | 97.9 | 95.6 | 100 |
Cytologic evaluation of suspicious LN is the most used diagnostic tool for detection of MTC LN metastases [
Our results showed that high levels of FNA-CT were present in all patients with histologically proven MTC with a sensitivity of 100% and specificity of 97.9%. The lowest value of FNA-CT in a histologically proven metastatic LN was 40.4 pg/mL, and the optimal cut-off, according to ROC analysis is 23.0 pg/mL, which is similar to other cut-offs proposed by other studies [
There was a wide range of FNA-CT measurements between patients, as well as high variability between measurements in the same LN. This can be explained by the fact that each fine-needle aspiration is a different procedure in and of itself and bound to get a different amount of LN fluid, each time it is performed. The wide range of FNA-CT values in different LN might be due to the inherent differences in calcitonin secretion by different tumors. We would like to point out that, despite of the broad range in FNA-CT measurements, the diagnostic accuracy is very high and we had no false-negatives and only one false-positive.
Fine-needle aspiration cytology also showed high accuracy for the diagnosis of LN metastases and higher than previous studies (sensitivity of 81.8% and specificity of 97.9%) [
Some studies also evaluated the potential utility of the FNA-CT/serum CT ratio for the diagnosis of MTC in thyroid nodules and LN, which could be useful in order to avoid false elevations of CT in the washout fluid if serum CT is significantly elevated. In order to avoid this potential bias, we excluded cases that might have had blood in the needle and all of them were centrifuged to ensure that only the supernatant was analyzed. We also need to keep in mind that the aspirate leftover volume is usually very small and subsequently rinsed/diluted with 1 mL of normal saline solution to overcome possible interferences of serum CT in FNA-CT measurement. Thus, we did not find the ratio FNA-CT/serum CT useful in this situation. In fact, 22 of the 23 detectable measurements were true-positives and only 7 of them had higher levels of serum CT than FNA-CT. All of these cases had histologically proven MTC LN metastases.
As with most clinical studies of thyroid cancer, this is a retrospective study and potential selection bias should be considered in the interpretation of the results. MTC is a rare entity and this study has a heterogeneous cohort. On the other hand, there is still limited evidence regarding CT measurement in FNA washout fluid. Therefore, in this context, we believe that it is important to evaluate and describe the results from clinical practice.
This is the largest study regarding the diagnostic utility of FNA-CT for detection of MTC exclusively in LN. It further supports the value of FNA-CT as a reliable and inexpensive diagnostic tool, along with FNA cytology, that should be included in the clinical workup of neck adenopathies in patients with thyroid nodules or medical history of MTC. However, there is still need for further studies in order to assess the actual relevance of this technique in the management of MTC and validate the optimal cut-off value for the diagnosis of LN metastases.
The clinical data used to support the findings of this study are included within the article.
Preliminary results from this study were presented at the 40th Annual Meeting of the European Thyroid Association and the abstract can be read at
There are no conflicts of interest that could be perceived as prejudicing the impartiality of the research reported.