All patients suffered from TBN (n=79, mean age M±SD was 44±11 years, range 22-64) and from TMN (n=41, mean age M±SD was 46±15 years, range 16-75) were hospitalized in the Head and Neck Department of the Medical Radiological Research Centre (MRRC), Obninsk. Thick-needle puncture biopsy of suspicious nodules of the thyroid was performed for every patient, to permit morphological study of thyroid tissue at these sites and to estimate their TEs contents. In all cases the diagnosis has been confirmed by clinical and morphological results obtained during studies of biopsy and resected materials. Histological conclusions for TBN were: 46 colloid goiter, 19 thyroid adenoma, 8 Hashimoto's thyroiditis, and 6 Riedel s Struma, whereas for TMN were: 25 papillary adenocarcinomas, 8 follicular adenocarcinomas, 7 solid carcinomas, and 1 reticulosarcoma. Samples of nodular tissue for TEs analysis were taken from both biopsy and resected materials.

All studies were approved by the Ethical Committees of MRRC. All the procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments, or with comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

All tissue samples obtained from TBN and TMN were divided into two portions using a titanium scalpel to prevent contamination by TEs of stainless steel 52. One was used for morphological study while the other was intended for TEs analysis. After the samples intended for TEs analysis were weighed, they were freeze-dried and homogenized 53.

To determine contents of the TEs by comparison with a known standard, biological synthetic standards (BSS) prepared from phenol-formaldehyde resins were used 54. In addition to BSS, aliquots of commercial, chemically pure compounds were also used as standards. Ten sub-samples of certified reference material (CRM) IAEA H-4 (animal muscle) and five sub-samples of CRM of the Institute of Nuclear Chemistry and Technology (INCT, Warszawa, Poland) INCT-SBF-4 Soya Bean Flour, INCT-TL-1 Tea Leaves, and INCT-MPH-2 Mixed Polish Herbs were treated and analyzed in the same conditions like thyroid samples to estimate the precision and accuracy of results .

The content of I were determined by INAA-SLR using a horizontal channel equipped with the pneumatic rabbit system of the WWR-c research nuclear reactor (Branch of Karpov Institute, Obninsk). Details of used nuclear reaction, radionuclide, gamma-energies, spectrometric unit, sample preparation, and the quality control of results were presented in our earlier publications concerning the INAA-SLR of I contents in human thyroid 2728 and scalp hair 55.

A vertical channel of the same nuclear reactor was applied to determine the content of Ag, Co, Cr, Fe, Hg, Rb, Sb, Sc, Se, and Zn by INAA-LLR. Details of used nuclear reactions, radionuclides, gamma-energies, spectrometric unit, sample preparation and procedure of measurement were presented in our earlier publications concerning the INAA-LLR of TEs contents in human thyroid 2930, scalp hair 55, and prostate 56575859.

After non-destructive INAA-LLR investigation the thyroid samples were used for ICP-MS. The samples were decomposed in autoclaves and aliquots of solutions were used to determine the Ag, Al, As, Au, B, Be, Bi, Cd, Ce, Co, Cr, Cs, Dy, Er, Eu, Ga, Gd, Hg, Ho, Ir, La, Li, Lu, Mn, Mo, Nb, Nd, Ni, Pb, Pd, Pr, Pt, Rb, Sb, Se, Sm, Sn, Tb, Te, Th, Ti, Tl, Tm, U, Y, Yb, Zn, and Zr mass fractions by ICP-MS using an ICP-MS Thermo-Fisher X-7 Spectrometer (Thermo Electron, USA). Information detailing with the NAA-LLR and ICP-MS methods used and other details of the analysis were presented in our earlier publications concerning TE contents in human thyroid 293035, prostate 606162, and scalp hair 55.

A dedicated computer program for INAA-SLR and INAA-LLR mode optimization was used 63. All thyroid samples were prepared in duplicate, and mean values of TEs contents were used in final calculation. Mean values of TEs contents were used in final calculation for the Ag, Co, Cr, Hg, Rb, Sb, Se, and Zn mass fractions measured by INAA-LLR and ICP-MS methods. Using Microsoft Office Excel software, a summary of the statistics, including, arithmetic mean, standard deviation of mean, standard error of mean, minimum and maximum values, median, percentiles with 0.025 and 0.975 levels was calculated for TEs contents in two groups of nodular tissue (TBN and TMN). The difference in the results between two groups of samples was evaluated by the parametric Student s t-test and non-parametric Wilcoxon-Mann-Whitney U-test.

Table 1 and Table 2 depict certain statistical parameters (arithmetic mean, standard deviation, standard error of mean, minimal and maximal values, median, percentiles with 0.025 and 0.975 levels) of the Ag, Al, B, Be, Bi, Cd, Ce, Co, Cr, Cs, Fe, Ga, Hg, I, La, Li, Mn, Mo, Nd, Ni, Pb, Pr, Rb, Sb, Sc, Se, Sm, Sn, Tl, U, Y, and Zn mass fraction in two groups of samples - TBN and TMN, respectively.

The ratios of means and the comparison of mean values of Ag, Al, B, Be, Bi, Cd, Ce, Co, Cr, Cs, Fe, Ga, Hg, I, La, Li, Mn, Mo, Nd, Ni, Pb, Pr, Rb, Sb, Sc, Se, Sm, Sn, Tl, U, Y, and Zn mass fractions in pair of sample groups such as TBN and TMN is presented in Table 3.

The comparison of our results with published data for Ag, Al, B, Be, Bi, Cd, Ce, Co, Cr, Cs, Fe, Ga, Hg, I, La, Li, Mn, Mo, Nd, Ni, Pb, Pr, Rb, Sb, Sc, Se, Sm, Sn, Tl, U, Y, and Zn mass fraction in TBN 6465666768697071727374757677 and TMN 707172 is shown in Table 4 and Table 5, respectively. A number of values for TEs mass fractions were not expressed on a dry mass basis by the authors of the cited references. However, we calculated these values using published data for water (75%) 84 and ash (4.16% on dry mass basis) 85 contents in thyroid of adults.

As was shown before 27282930 good agreement of the Ag, Al, B, Be, Bi, Cd, Ce, Co, Cr, Cs, Fe, Ga, Hg, I, La, Li, Mn, Mo, Nd, Ni, Pb, Pr, Rb, Sb, Sc, Se, Sm, Sn, Tl, U, Y, and Zn contents in CRM IAEA H-4, INCT-SBF-4, INCT-TL-1, and INCT-MPH-2 samples determined by both INAA-SLR and ICP-MS methods with the certified data of these CRMs indicates acceptable accuracy of the results obtained in the study of TBN and TMN groups of tissue samples presented in Table 1, Table 2, Table 3, Table 4, Table 5

From Table 3, it is observed that in TMN tissue the mean mass fractions of Be, Fe, I, Sc, and Se are approximately 1.9, 1.7, 14, 3.1, and 1.6 times, respectively, lower while the mass fraction of Ga, Mo, and Rb were 62%, 51%, and 33%, respectively, higher than those in TBN tissue. In a general sense Ag, Al, B, Bi, Cd, Ce, Co, Cr, Cs, Hg, La, Li, Mn, Nd, Ni, Pb, Pr, Sb, Sm, Sn, Tl, U, Y, and Zn contents found in the TBN and TMN groups of tissue samples were similar (Table 3).

Mean values obtained for Ag, Al, Cd, Cr, Fe, I, Mn, Mo, Ni, Pb, Rb, Se, and Zn contents in TBN agree well with median of mean values reported by other researches (Table 4). Mean mass fractions of Co, Hg, and U in TBN obtained in present study were almost two order of magnitude lower medians of means for these TEs in published articles. No published data referring B, Be, Bi,Cs, Ga,La, Li, Nd, Pr, Sb, Sc, Sm, Sn, Tl, and Y contents of TBN were found (Table 4).

Mean values obtained for Cd, Cr, Fe, I, Mn, Ni, Pb, Rb, Se, and Zn contents in TMN agree well with median of mean values reported by other researches (Table 5). Mean mass fraction obtained for Co and Hg in TMN were approximately two and one order of magnitude, respectively, lower median of previously reported means. Mean mass fraction of U founded in TMN was almost one order of magnitude higher the only published result 75. No published data referring Ag, Al, B, Be, Bi, Ce, Cs, Ga, La, Li, Mo, Nd, Pr, Sb, Sc, Sm, Sn, Tl, and Y contents of TMN were found (Table 5).

The range of means of Ag, Al, Cd, Co, Cr, Fe, I, Mn, Mo, Ni, Pb, Rb, Se, U and Zn level reported in the literature for TBN and TMN vary widely (Table 3). This can be explained by a dependence of TEs content on many factors, including age, gender, ethnicity, mass of the TNs, and the stage of diseases. Not all these factors were strictly controlled in cited studies. However, in our opinion, the leading causes of inter-observer variability can be attributed to the accuracy of the analytical techniques, sample preparation methods, and inability of taking uniform samples from the affected tissues. It was insufficient quality control of results in these studies. In many scientific reports, tissue samples were ashed or dried at high temperature for many hours. In other cases, thyroid samples were treated with solvents (distilled water, ethanol, formalin etc). There is evidence that during ashing, drying and digestion at high temperature some quantities of certain TEs are lost as a result of this treatment. That concerns not only such volatile element as Hg, but also other TEs investigated in the study 868788. On the other hand, when destructive analytical techniques are used the tissue samples may be contaminated by TEs contained in chemicals used for digestion.

Trace elemental analysis of affected thyroid tissue could become a powerful diagnostic tool. To a large extent, the resumption of the search for new methods for early diagnosis of TMN was due to experience gained in a critical assessment of the limited capacity of the USS-examination 23. In addition to the US test and morphological study of needle-biopsy of the TNs, the development of other highly precise testing methods seems to be very useful. Experimental conditions of the present study were approximated to the hospital conditions as closely as possible. In all cases we analyzed a part of the material obtained from a puncture biopsy of the TNs. Therefore, our data allow us to evaluate adequately the importance of TEs content information for distinguishing TMN from TBN.

Tissue content of Ga, Be, Fe, I, Mo, Rb, Sc, and Se are different in most TMN as compared to TBN (Table 3). Level of I in nodular tissue has very promising prospects as a biomarker of malignancy, because there is a great difference between content of this TE in TBN and TMN (Table 3). It is very interest a potential possibilities of using the I/Ga, I/Mo, and I/Rb ratio as cancer biomarker, because during the thyroid malignant transformation contents of these TEs in nodular tissue change in different directions - a drastically decrease of I and an increase of Ga, Mo, Rb (Table 3). Thus, the results of study show that analysis of TEs contents in biopsy of TNs may serve as a potential tool for detection of TMN.

This study has several limitations. Firstly, analytical techniques employed in this study measure only thirty two TEs (Ag, Al, B, Be, Bi, Cd, Ce, Co, Cr, Cs, Fe, Ga, Hg, I, La, Li, Mn, Mo, Nd, Ni, Pb, Pr, Rb, Sb, Sc, Se, Sm, Sn, Tl, U, Y, and Zn) mass fractions. Future studies should be directed toward using other analytical methods which will extend the list of TEs investigated in TBN and TMN. Secondly, the sample size of TBN and TMN group was relatively small and prevented investigations of TEs contents in this group using differentials like gender, functional activity of nodules, stage of disease, and dietary habits of patients with TNs. Lastly, generalization of our results may be limited to Russian population. Despite these limitations, this study provides evidence on significant TEs level alteration in thyroid nodular tissue and shows the necessity to continue TEs research as potential biomarkers of thyroid malignant transformation.