This article is mentioned in:

Introduction

Nodular goiter (NG), thyroid adenoma (TA) and thyroid cancer (TC) are common thyroid diseases, and their incidence rates are rapidly increasing worldwide (1–3), and in China (4).

NG is a benign thyroid disease and, like hyperplasia, stems from recurrent attacks of simple goiter. It is diagnosed following thyroid ultrasonography and fine-needle aspiration (FNA); if the results suggest malignancy or malignant behavior, the patient is referred for a thyroidectomy (5).

TA is the most prevalent benign thyroid tumor; it primarily originates from thyroid follicular cells, and has a favorable prognosis following surgical resection (6).

TC is the most common malignant tumor among all endocrine system and head and neck neoplasms, and has a low but rapidly increasing incidence rate worldwide (7). According to the Surveillance, Epidemiology, and End Results (SEER) database (http://seer.cancer.gov/statfacts/html/ld/thyro.html), the number of new cases of TC has increased from 4.8 (per 100,000) in 1975 to 6.2 (per 100,000) in 1995, and 15.1 (per 100,000) in 2013. In China, it was estimated that the new cases and mortalities increased to 90,000 and 6,800, respectively, in 2015, according to the National Central Cancer Registry (NCCR) data of the average incidence rates from 2009–2011 in 72 population-based cancer registries (4). Differentiated TC originates from follicular thyroid cells, and includes papillary TC (PTC) and follicular TC, which account for 80–85 and 10–15% of TC cases, respectively (8).

Despite being the most commonly used diagnostic tool, the ultrasonography-guided FNA biopsy is not useful to distinguish between benign nodules, follicular adenoma and follicular carcinoma in cytology (3,9–11). Therefore, it would be beneficial in clinical practice to identify biomarkers that may distinguish NG, TA and TC in a convenient and noninvasive manner.

Fatty acids are important nutrients and bioactive molecules that are involved in energy storage, signal pathways and key biochemical activities (12,13). There are a number of studies demonstrating that fatty acids, in particular ω-3 and ω-6 fatty acids, are closely associated with the risk of certain diseases, including cancer, diabetes, and cardiovascular diseases (14,15). The thyroid is an important metabolic organ, which synthesizes thyroid hormones and controls energetic metabolism, that is closely associated with fatty acids (16,17). Schneider and Chen (3) hypothesized that an increasing body mass index (BMI), which is closely associated with fatty acids, is a possible explanation for the increasing incidence of TC. Furthermore, additional studies have reported differences in fatty acids in the serum (18–20), urine (21), and thyroid tissue samples (22,23) between TC patients and healthy controls.

Although there is a possibility that benign diseases, including NG and TA, may become malignant, it is unconfirmed whether these diseases are the precursors of TC and whether they share any common etiology, including body fat percentage and fat intake. Therefore studying the association between fatty acids and NG, TA and TC simultaneously may provide insights that would be beneficial in clinical practice. Consequently, the present study utilized a gas chromatography-flame ionization detector (GC-FID) method to measure the percentages of polyunsaturated fatty acids (PUFAs) in 122 plasma samples from patients with thyroid diseases.

Materials and methods

Participants

A total of 122 patients with thyroid diseases were recruited at 2 time points from Wenzhou Medical School Subsidiary Hospital (Wenzhou, China), including 97 patients with thyroid carcinoma (female, n=77), 11 patients with NG (female, n=7), and 14 patients with TA (female, n=9). All blood samples were collected following overnight fasting. Following centrifugation at 3,000 × g at room temperature for 15 min, plasma samples were removed and stored at −80°C until measurement. Clinical parameters, including age, sex, weight, height, fasting blood glucose, systolic/diastolic blood pressure, thyroid hormone (Thy), triiodothyronine (T3), thyroid-stimulating hormone (TSH), free tetraiodothyronine (FT4), free triiodothyronine (FT3), thyroglobulin antibody (TGA), anti-thyroperoxidase antibody (TPOA), thyroglobulin (HTG), and parathormone (PTH), were obtained from the hospital researchers. The study was approved by Institutional Review Board of the First Affiliated Hospital of Wenzhou Medical University and informed consent was obtained from all individual participants included in the study.

Chemicals and reagents

The fatty acid methyl esters (FAMEs; including 38 FAMEs) internal standard (IS) C21:0 (purity >99%), used as the calibration standard solution, was purchased from Nu-Chek Prep, Inc. (Elysian, MN, USA). High-performance liquid chromatography (HPLC)-grade methanol, dichloromethane, n-hexane, deionized H2O, and iso-octane were purchased from Honeywell (Morris Plains, NJ, USA). NaCl (purity >99.5%) was purchased from Jiangsu Hengrui Medicine Co., Ltd. (Lianyungang, China). Na2SO4 (purity >99%) and H2SO4 (purity >95%) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

Profile of fatty acids in plasma

Fatty acids were extracted from the plasma sample using methanol/dichloromethane (V/V=1:1), then dried with nitrogen, trans-methylated with methanol/concentrated sulfuric acid (V/V=25:1) and bathed in 80°C water for 1 h. FAMEs were extracted by n-hexane, then dried by nitrogen, and finally dissolved in iso-octane for GC-FID (Agilent Technologies, Inc., Santa Clara, CA, USA) equipped with a 100-m HP-88 NEFA phase column (100 m × 0.25 mm × 0.2 µm; Agilent Technologies, Inc.).

Four ω-3 and seven ω-6 fatty acids evaluated in the present study. The percentage of each individual fatty acid was calculated according to a response value of the standards using C21:0 as the IS. The standards were used to adjust the measurement deviation, and the IS was used to adjust the extraction procedure deviation. A uniform quality control (QC) sample was inserted into every 12 samples. The coefficient of variation (CV) of QC was <12% for all 11 fatty acids.

Statistics analysis

The clinical parameters were compared among three groups by non-parametric Kruskal-Wallis test for continuous variables, as the data did not follow a normal distribution. A χ2 test was used for categorical variables. The percentages or ratios of fatty acids were compared between TC, NG and TA by the non-parametric Kruskal-Wallis test. Each group comparison was performed by the rank-based ANOVA among three groups. All tests were two-sided. P<0.05 was considered to indicate a statistically significant difference. Statistical analyses were performed using SAS 9.3 (SAS Institute, Inc., Cary, NC, USA).

Results

Baseline clinical characteristics

The baseline clinical characteristics are summarized shown in Table I. In thyroid-related hormones, only the TSH was significantly different among the three groups; and the level in the TC group was the highest. However, the other hormones evaluated, including Thy, T3, FT4, FT3, TGA, TPOA, and HTG, did not show any statistically significant differences (Table I). No significant difference among the three groups was observed for PTH, and the majority of thyroid-related hormone indexes in the experimental population were in the normal range according to ‘China Thyroid Disease Diagnosis and Treatment Guidelines’ (24), except TGA and TPOA. In total, 33 out of 95 patients exhibited higher TGA, and 24 out of 93 patients exhibited higher TPOA. Out of 110 patients, 40 were in the normal range according to ‘China Thyroid Disease Diagnosis and Treatment Guidelines’ (24) for all 9 thyroid-related hormone parameters (data not shown).

Table I. Baseline characteristics and thyroid-related hormone levels in the TC, TA and NG groups.

Table I.

Baseline characteristics and thyroid-related hormone levels in the TC, TA and NG groups.

	Value
Characteristic	TC	TA	NG	P-value
Age, years	46 (38–55)	43 (38–55)	56 (41–62)	0.25
Sex, n
Female	77	9	8	0.55
Male	20	5	3
Height, cm (range)	160 (158–165)	160 (157–170)	158 (151–168)	0.34
Weight, kg (range)	60 (53.5–67)	63 (49.5–79)	57 (55–63)	0.69
BMI, kg/m2 (range)	23 (21–26)	24 (20–28)	23 (20–25)	0.91
Glu, mmol/l (range)	5.6 (4.9–6.5)	5.7 (5.5–6.1)	5.95 (5.2–7.3)	0.44
SBP, mmHg (range)	124 (117–140)	125.5 (114–134)	129 (118–135)	0.91
DBP, mmHg (range)	80 (74–87)	80.5 (70–90)	83 (79–88)	0.77
HT, %	32/97 (34)	2/11 (18)	2/10 (20)	0.56
Thy, nmol/l (range)	106 (95–117)	90 (85–111)	108 (92–130)	0.18
T3, nmol/l (range)	1.6 (1.4–1.8)	1.6 (1.3–1.7)	1.75 (1.4–1.8)	0.49
TSH, mIU/l (range)	1.4 (0.88–1.91)	1.13 (0.62–1.46)	0.67 (0.48–1.6)	0.03
FT4, pmol/l (range)	11 (9–12)	11 (10–12)	12 (9.5–13)	0.61
FT3, pmol/l (range)	4.4 (4–4.8)	4.1 (3.8–4.4)	4.4 (4.1–4.71)	0.16
TGA, IU/ml (range)	0.9 (0.9–12.8)	0.9 (0.9–0.9)	0.9 (0.9–0.9)	0.29
TPOA, IU/ml (range)	106 (95–117)	90 (85–110)	108 (92–130)	0.37
PTH, pg/ml (range)	37 (28–43)	31 (27–33)	42 (32–47)	0.20

[i] Data are presented as the median (interquartile range), unless otherwise indicated. Comparisons between three groups were performed with the Kruskal-Wallis test, except for HT and sex, which were analyzed with the Chi-squared test. P<0.05 was considered to indicate a statistically significant difference. Glu, fasting blood glucose; SBP, systolic blood pressure; DBP, diastolic blood pressure; HT, hypertension; Thy, thyroid hormone; T3, three iodine threonine; TSH, thyroid stimulating hormone; FT4, free tetraiodothyronine; FT3, free triiodothyronine; TGA, thyroglobulin antibody; TPOA, anti-thyroperoxidase antibody; HTG, thyroglobulin; PTH, parathormone; TC, thyroid cancer; TA, thyroid adenoma; NG, nodular goiter.

PUFAs among three groups

The plasma levels of 22:6n-3, ω-3 PUFAs, 22:4n-6, ω-6/ω-3 fatty acids, and total PUFAs were significantly different among the three groups (Fig. 1A-D). Total PUFA level (Fig. 1E) was significantly increased in the TA group compared with the TC and NG groups. The levels of 22:6n-3 (Fig. 1A) and total ω-3 PUFAs (Fig. 1B) were significantly increased in the TA group compared with the TC and NG groups. However, 20:5n-3 and 22:5n-3 did not demonstrate a statistically significant difference among the three groups. In addition, among ω-6 fatty acids, 22:4n-6 (Fig. 1H) was significantly increased in the NG group compared with the TC and TA groups. A number of other ω-6 fatty acids, including 18:3n-6 (Fig. 1I), 20:2n-6 (Fig. 1J), 20:3n-6 (Fig. 1K), 20:4n-6 (Fig. 1L), and 22:5n-6 (Fig. 1M), demonstrated a similar pattern. The ω-6/ω-3 (Fig. 1D) ratio was significantly decreased in the TA group compared with the TC group.

Figure 1. Percentage and the ratio of fatty acids and different kinds of fatty acids are compared among three diseases. (A) Demonstrates the percentage of total PUFA; (B) exhibits the ratio of total ω-6 to total ω-3; (C-J) were the percentage of total ω-6, 18:2n-6, 18:3n-6, 20:2n-6, 20:3n-6, 20:4n-6, 22:4n-6, and 22:5n-6; (K-O) were the percentage of total ω-3, 18:3n-3, 20:5n-3, 22:5n-3, and 22:6n-3. Fatty acids are compared by Kruskal-Wallis test among three groups, and two groups are compared using an analysis of based on rank adjusted by Bonferroni Correction. *P<0.05. 0, +, and × denote median values of NG, TA and TC, respectively, inside the interquartile range, and outliers outside the interquartile range.

PUFAs among three groups in females

In females (Table II), the present study did not identify a statistically significant difference in any ω-3 fatty acids among the three groups examined. While the ω-6 fatty acids 20:4n-6 and 22:4n-6 were significantly different among three groups in females. In females, the levels of 18:3n-6, 20:2n-6, 20:3n-6 and 22:5n-6 in the NG group were higher than those of the other groups, and their pattern that was similar to that of the whole study population (Fig. 1).

Table II. The percentage and the ratio of fatty acids and different kinds of fatty acids in female are compared among three diseases.

Table II.

The percentage and the ratio of fatty acids and different kinds of fatty acids in female are compared among three diseases.

	Fatty acid level, median (Q1, Q3)
Fatty acid	TC (n=78)	TA (n=9)	NG (n=8)	P-value
PUFA	47 (45,49)	50 (49,51)	46 (45,48)	0.01
n-3	6.2 (5.2,6.9)	7.4 (6.5,8.1)	6.0 (4.6,7.9)	0.08
18:3n-3	0.71 (0.59,1.0)	0.53 (0.49,0.65)	0.61 (0.49,0.84)	0.13
20:5n-3	0.49 (0.4,0.61)	0.57 (0.52,0.61)	0.51 (0.45,0.55)	0.24
22:5n-3	0.54 (0.48,0.67)	0.6 (0.53,0.66)	0.64 (0.45,0.86)	0.66
22:6n-3	4.3 (3.3,5.1)	5.7 (4.5,6.3)	3.7 (3.0,6.0)	0.08
n-6	41 (39,43)	43 (42,44)	41 (37,42)	0.15
18:2n-6	32 (29,34)	33 (31,35)	30 (27,31)	0.05
18:3n-6	0.2 (0.12,0.29)	0.16 (0.13,0.23)	0.28 (0.11,0.4)	0.70
20:2n-6	0.29 (0.26,0.33)	0.27 (0.25,0.31)	0.32 (0.3,0.34)	0.18
20:3n-6	1.1 (0.87,1.5)	1.2 (0.91,1.4)	1.2 (1.1,1.6)	0.38
20:4n-6	7.3 (6.3,8.8)	7.8 (7.6,8.3)	9.1 (8.1,9.9)	0.04
22:4n-6	0.17 (0.14,0.21)	0.16 (0.14,0.18)	0.22 (0.18,0.28)	0.02
22:5n-6	0.18 (0.13,0.25)	0.2 (0.14,0.31)	0.24 (0.22,0.56)	0.06
n-6:n-3	6.7 (5.7,8.1)	5.8 (5.2,6.6)	7.2 (4.6,9.3)	0.30

[i] All variables were compared with the Kruskal-Wallis test. The data are presented as the median with the lower quartile (Q1) and the upper quartile (Q3). P<0.05 was considered to indicate a statistically significant difference. TC, thyroid cancer; TA, thyroid adenoma; NG, nodular goiter.

PUFAs among three groups with normal level of thyroid-related hormones

The percentage of fatty acids in patients with a normal level of thyroid-related hormones were compared by Kruskal-Wallis test among three groups (Table III). The results identified that 22:6n-3, total ω-3 fatty acids, 22:4n-6, and total PUFA levels were significantly different among three groups, which were consistent with the whole study population.

Table III. Percentage and ratio of fatty acids compared among three groups of patients with five types of normal level thyroid-related hormone.

Table III.

Percentage and ratio of fatty acids compared among three groups of patients with five types of normal level thyroid-related hormone.

	Fatty acid level, median (Q1, Q3)
Fatty acid	TC (N=56)	TA (N=8)	NG (N=6)	P-value
PUFA	46 (44,49)	50 (49,51)	46 (45,48)	0.03
n-3	5.9 (4.9,6.7)	7.4 (6.8,8.0)	5.0 (4.5,6.0)	0.02
18:3n-3	0.68 (0.58,1.1)	0.53 (0.51,0.7)	0.52 (0.43,0.63)	0.12
20:5n-3	0.51 (0.4,0.61)	0.57 (0.54,0.61)	0.53 (0.51,0.55)	0.24
22:5n-3	0.54 (0.43,0.68)	0.6 (0.59,0.66)	0.57 (0.47,0.68)	0.65
22:6n-3	4.2 (3.1,4.8)	5.7 (5.0,6.3)	3.4 (3.0,3.7)	0.04
n-6	40 (38,43)	43 (42,43)	41 (39,43)	0.21
18:2n-6	31 (28,33)	33 (31,34)	30 (29,31)	0.20
18:3n-6	0.2 (0.12,0.29)	0.15 (0.13,0.22)	0.12 (0.11,0.32)	0.64
20:2n-6	0.3 (0.25,0.33)	0.27 (0.25,0.31)	0.33 (0.27,0.35)	0.24
20:3n-6	1.1 (0.87,1.6)	1 (0.86,1.3)	1.3 (1.2,1.4)	0.50
20:4n-6	7.2 (6.3,8.3)	7.8 (7.5,8.7)	8.4 (7.7,9.1)	0.15
22:4n-6	0.18 (0.14,0.22)	0.15 (0.13,0.17)	0.22 (0.21,0.26)	0.03
22:5n-6	0.2 (0.14,0.26)	0.16 (0.14,0.23)	0.26 (0.22,0.31)	0.06
n-6:n-3	7.0 (5.7,8.6)	5.8 (5.4,6.2)	8.1 (7.2,9.3)	0.11

[i] The hormones Thy, T3, TSH, FT4, and FT3 of all 70 cases are normal. The variables were tested with the Kruskal-Wallis test. The data are presented as the median with the lower quartile (Q1) and the upper quartile (Q3). P<0.05 was considered to indicate a statistically significant difference.

Cut-off of total ω3 fatty acids and 22:4n6 to distinguish TA and NG groups

When the 75th percentile cutoff of the ω-3 fatty acids and 22:4n-6 in the group from the first time point was applied to the group from the second time point, the patients with TA were 100% correctly classified (Tables IV and V). Consistent with the significant difference (Table II), it demonstrates that ω-3 PUFAs and 22:4n-6 are helpful to distinguish TA and NG among three diseases although sample size is limited.

Table IV. The 75% threshold value for total ω-3 fatty acids was 6.550 (75th percentile), which was used to distinguish TA from TC and NG.

Table IV.

The 75% threshold value for total ω-3 fatty acids was 6.550 (75th percentile), which was used to distinguish TA from TC and NG.

	Batch 1	Batch 2
Disease (%)	(Proportion ≥Q3)	(Proportion ≥Q3)
TC	14/67=21	17/30=57
TA	6/9=67	2/2=100
NG	2/7=29	1/3=33

[i] TC, thyroid cancer; TA, thyroid adenoma; NG, nodular goiter.

Table V. The 75% threshold value for 22:4n-6 was 0.209 (75th percentile), which was used to distinguish NG from TC and TA.

Table V.

The 75% threshold value for 22:4n-6 was 0.209 (75th percentile), which was used to distinguish NG from TC and TA.

	Batch 1	Batch 2
Disease (%)	(Proportion ≥Q3)	(Proportion ≥Q3)
TC	17/67=25	12/30=40
TA	0/9=0	0/2=0
NG	4/7=57	2/3=67

[i] TC, thyroid cancer; TA, thyroid adenoma; NG, nodular goiter.

Discussion

The present study identified that the plasma levels of 22:6n-3, total ω-3 fatty acids, and PUFAs were significantly increased in the TA group compared with those of the TC and NG groups and that 22:4n-6 level was significantly increased in NG group compared with that of TC and TA groups.

Thyroid diseases are common endocrine diseases, although some investigators hypothesize that thyroid diseases, especially TC, are over-diagnosed (1,25,26). Conversely, other researchers postulate that the incidence rate of TC is increasing worldwide (2,4); however, this increasing trend may be associated with environmental changes and the development of detection technology (3). Regardless, the clear etiological mechanism of this substantial rise requires examination. It is difficult to distinguish the different thyroid diseases of follicular based on clinical symptoms and cytology (3,9–11). Therefore, it is important to develop biomarkers and easier-to-use technologies to facilitate clinical practice.

The estimated numbers of new cases of TC in 2015 were 40,200 in east, 14,300 in central, 10,700 in northeast and <10,000 in other regions of China (4). A number of studies reported that the fish and seafood-rich dietary intake was associated with the increased risk of TC (27,28), in particular dried or salted fish that specifically occur among Asian seasonings (29). Wenzhou is an eastern coastal city where people consume an increased amount of seafood compared with the national average (30), and seafood is rich in ω-3 fatty acids. Therefore, evaluating the link between thyroid disease and fatty acids may provide beneficial insights; hence the patient sample was obtained from a Wenzhou-based hospital.

Activation of de novo lipogenesis may result in TC tumorigenesis through fatty acid synthase (FASN) catalyzing the synthesis of 16-carbon saturated fatty acids and AKT pathway signaling (31). However, consistent with the results of previous studies (18,21,23), the present study identified that the ω-3 and ω-6 fatty acids with chains longer than 16 carbons were not increased in patients with cancer. We hypothesize that this is due to long chain ω-3 and ω-6 fatty acids being primarily from diet, and produced through other enzymatic reactions.

ω-3 and ω-6 fatty acids, including 18:3n-3, 20:5n-3, 22:5n-3, 22:6n-3 and 18:2n-6, 18:3n-6, 20:2n-6, 20:3n-6, 20:4n-6, 22:4n-6, 22:5n-6, are involved in two different metabolic pathways in KEGG. It is reported that ω-3 fatty acids are protective in the majority of diseases, while ω-6 fatty acids do not exhibit this activity type (32). The present study identified that ω-3 and ω-6 fatty acid profiles were exhibited differently among three diseases: ω-3 fatty acids, including 20:5n-3, 22:5n-3, and 22:6n-3, were highest in the TA group; and ω-6 fatty acids, including 18:3n-6, 20:2n-6, 20:3n-6, 20:4n-6, 22:4n-6, and 22:5n-6, were highest in the NG group. We hypothesize that ω-3 and ω-6 fatty acids may have different metabolic pathways in TA and NG compared with TC, which may result in the varying etiological characteristics of those diseases.

The incidence of nodules generally increases with age (33). The present study corroborated this, as NG patients of the study sample were older than in the other two groups. TGA and TPOA are related to autoimmune thyroid disease and thyroiditis (34,35). TSH level, which is a risk factor for TC (3,36), was highest in the TC group followed by the TA group in our study.

Total PUFA levels were significantly increased in the TA group compared with the TC and NG groups in the present study. Berg et al (18) reported that the sum of arachidonic acid (20:4n-6) and docosahexaenoic acid (DHA) concentrations and the sum of arachidonic acid, EPA, and DHA were significant decreased in patients with cancer compared with healthy controls. Guo et al (23) reported C20:4, C22:4 and C22:5 were significantly decreased in six types of cancer tissues (breast, lung, colorectal, esophageal and gastric cancer and TC); however, C22:4 and C22:5 were increased in TC tissue compared with normal tissue. Furthermore, urinary PUFAs are increased in TC compared with healthy controls (21). To date, the results of previous studies on total PUFA in cancer have been inconclusive, suggesting that total PUFA may not serve as an appropriate indicator for thyroid disease assessment.

The present study demonstrated that DHA and total ω-3 fatty acids were significantly increased in TA compared with TC and NG. If confirmed through further study, this exhibits potential to aid in distinguishing TA from other thyroid diseases. Although statistically significant differences were not identified for other ω-3 fatty acids, they exhibited similar trends. Therefore a study with a larger cohort would be required to examine this. Consistent with our results, Xu et al (37) reported that DHA levels were increased in TA compared with TC tissue. Yao et al (20) reported serum DHA levels were decreased in PTC and NG compared with healthy controls; however, they did not identify a difference between PTC and NG. Zhang et al (19) reported that the serum DHA and 18:3n-3 were significantly increased in TC patients compared with those with benign thyroid diseases, and the difference may be due to the different types of fatty acids. Zhang et al (19) measured the free fatty acids (FFAs) while Yao et al (20) measured the total fatty acids, including FFA and esterified fatty acids. Furthermore, Zhang et al (19) used concentration (µM), whereas the present study merely obtained percentages of fatty acids, which may be responsible for the differences observed.

In the present study, 22:4n-6 was significantly decreased in the TC and TA groups compared with NG group; a number of other ω-6 fatty acids, including 18:3n-6, 20:2n-6, 20:3n-6, 20:4n-6, 22:4n-6, and 22:5n-6, exhibited similar, though statistically insignificant trends. ω-6 fatty acids also can be helpful to differentiate NG from other thyroid diseases. In a study by Kim et al (21), the urinary concentrations of ω-6 fatty acids, including 18:2n-6 and 20:4n-6, were decreased in TC compared with healthy controls. However, Yao et al (20) reported that serum 18:2n-6, 20:4n-6 and 22:4n-6 were increased in PTC compared with NG. Furthermore, Zhang et al (19) reported the serum 20:4n-6 was significantly increased in TC patients compared with those with benign thyroid diseases. However, the differences may be related to the types of diseases examined and the units used.

When the analysis was restricted to female subjects, the DHA and total ω-3 fatty acids did not demonstrate a significant difference. This may be due to the smaller sample size. The 20:4n-6 was significantly decreased in the TC and TA groups compared with the NG group in female subjects, which may suggest that more 20:4n-6 would be transferred into the downstream product. This is due to arachidonic acid being the substrate of prostaglandin E2, which may inhibit the anti-tumor reaction of the immune system (38). Therefore, 20:4n-6 may be associated with a high incidence of TC in female subjects.

To the best of our knowledge, the present study is amongst the first to explore the plasma fatty acid profiles among three common thyroid diseases. TA was identified to be different from TC and NG in ω-3 fatty acids, while NG was different from TC and TA in ω-6 fatty acids. FNA is the most common tool for thyroid disease diagnosis; however, its sampling accuracy is limited. While the mutational analyses of BRAF, RAS, RET/PTC, and PAX8-PPARγ rearrangement contribute to distinguishing follicular lesions (39,40), the tissue damage resulting from surgery is not negligible. Furthermore, this method also has uncertainty due to the dependency of accurate sampling on the equipment and the operator. However, if validated, fatty acids have the potential to serve as diagnostic biomarkers, which may facilitate the diagnosis of thyroid diseases.

The present study is not without limitations. It had a small sample size, which limited the ability to detect moderate differences. Second, fatty acids in the plasma only reflect short-term dietary exposure, and the absence of analysis of PUFAs in a healthy population is a limitation. However, the present study focused on comparing three thyroid diseases, as opposed to a case-control study with healthy controls, in order to provide insight on the etiological evolution of the diseases. This could aid in the understanding of whether NG or TA are precancerous forms of TC from the aspect of fatty acid metabolism. The results obtained may also provide data for the prevention/treatment of TC; however, further studies with somatic tissues are required to investigate the long-term effects of the exposure to ω-3 and ω-6 PUFAs.

In summary, the present study demonstrated the differences in clinical parameters and plasma ω-3 and ω-6 PUFA profiles among three different common thyroid diseases, namely NG, TA and TC. The results of the present study suggest that ω-3 fatty acids, especially 22:6n-3, are advantageous for distinguishing TA from NG and TC; and ω-6 fatty acids, especially 22:4n-6, are effective for distinguishing NG from TA and TC. However, further study is required with a larger cohort and prospective design, including used of somatic tissues.

Acknowledgements

Not applicable.

Funding

This research was supported by funds from the Key Laboratory of Nutrition and Metabolism (awarded to OW and YG) and the 100 Talented Plan of Chinese Academy of Sciences (awarded to YG).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

YG and FC designed the current study. OW, LJ, and FY obtained biological samples. XL, JZ, QD and CH performed the experiments. XL and HL analyzed the data. All authors were involved in the interpretation of the results. XL, FC and YG wrote the manuscript. All authors read, gave comments, and approved the final version of the manuscript. All authors had taken responsibility for the integrity of the data and the accuracy of the data analysis.

Ethics approval and consent to participate

Institutional Review Board of the First Affiliated Hospital of Wenzhou Medical University approved the study and informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References