Molecular Markers in the Diagnosis and Treatment of CancerView this Special Issue
Clinical Study | Open Access
BRAF Testing in Multifocal Papillary Thyroid Carcinoma
Background. BRAF V600E mutation is associated with poor prognosis in patients with papillary thyroid carcinoma (PTC). PTC is often multifocal, and there are no guidelines on how many tumors to test for BRAF mutation in multifocal PTC. Methods. Fifty-seven separate formalin-fixed and paraffin-embedded PTCs from twenty-seven patients were manually macrodissected and tested for BRAF mutation using a commercial allele-specific real-time polymerase chain reaction-based assay (Entrogen, Woodland Hills, CA). Data related to histologic characteristics, patient demographics, and clinical outcomes were collected. Results. All mutations detected were BRAF V600E. Seventeen patients (63%) had concordant mutation status in the largest and second-largest tumors (i.e., both were positive or both were negative). The remaining ten patients (37%) had discordant mutation status. Six of the patients with discordant tumors (22% overall) had a BRAF-negative largest tumor and a BRAF-positive second-largest tumor. No histologic feature was found to help predict which cases would be discordant. Conclusions. Patients with multifocal PTC whose largest tumor is BRAF-negative can have smaller tumors that are BRAF-positive. Therefore, molecular testing of more than just the dominant tumor should be considered. Future studies are warranted to establish whether finding a BRAF mutation in a smaller tumor has clinical significance.
Papillary thyroid carcinoma (PTC) is often a multifocal disease, with rates as high as 80% in the literature . Despite multiple studies using different methodologies [1–11] it remains an unsettled question as to whether the multifocality represents separate primary tumors or intraglandular metastasis, or a combination of both. It seems likely from prior studies that a subset of cases represent separate primaries, or at the very least have molecularly distinct tumors. The question then arises, how many tumors should be tested for BRAF mutation in multifocal PTC? A recent very large retrospective study found that patients with multifocal disease and a BRAF V600E mutation had a significantly higher rate of mortality than patients with multifocal disease without a BRAF mutation, but the study did not elaborate on how many or which tumors were tested for each patient . Our study attempts to further the data on BRAF mutation status in patients with multifocal disease and to find the optimal strategy for BRAF testing in these patients.
2. Materials and Methods
The study protocol was approved by the IRB at Tulane University. The laboratory information system was searched for cases of multifocal papillary thyroid carcinoma in patients with all thyroid tissue removed, either as a single surgery or two separate surgeries, since January 1, 2006. Thirty-three cases were identified, and the slides from each case were reviewed by the study pathologists (H.Z.K. and A.B.S.). In all but one case, the entire thyroid was submitted. The following histologic features were recorded for each tumor: size, laterality, histologic variant, and the presence or absence of additional features such as irregular border, location close to the capsule, extrathyroidal extension, satellite nodules, and isolated psammoma bodies. Satellite nodules were defined as tumor foci less than 0.5 mm that were within one section of a larger tumor nodule but not directly attached to the larger nodule. Isolated psammoma bodies were defined as psammoma bodies within the thyroid parenchyma and separate from any tumor nodules. The histologic variant was decided by each pathologist separately, based on the predominant pattern seen in the tumor; discrepancies were resolved by viewing the case together at a multiheaded scope.
The study pathologists marked the two largest tumors in each case, and the tissue was manually macrodissected and placed into tubes. DNA was extracted using the EZ1 DNA Tissue kit (Qiagen, Valencia, CA). BRAF mutational status was tested using a commercial allele-specific real-time polymerase chain reaction-based assay that can detect five point mutations in codon 600 (V600E, V600K, V600R, V600D, and V600M) when present in as little as 1% of the tissue (Entrogen, Woodland Hills, CA). Overall, six of the original thirty-three cases were excluded because of insufficient DNA (either a tumor focus was gone on deeper levels or the DNA was of poor quality), leaving a total of twenty-seven cases for the study (see Table 1). Two of the cases where both the largest and second-largest tumors were BRAF-negative had additional tumors (see Patients 24 and 25 in Table 1); three of these additional tumors had adequate DNA and were tested for BRAF mutation (two additional tumors for Patient 24 and one additional tumor for Patient 25). In all, a total of fifty-seven tumors were tested for BRAF mutation. Clinical data related to patient demographics, tumor stage, months of followup, and outcome was collected for these twenty-seven patients. Statistical calculations were performed using , with the Student’s -test and Fisher’s exact test used to calculate significance. To assess whether BRAF status was associated with positive lymph nodes independently of size, multivariate analysis was carried out using logistic regression and the corresponding likelihood ratio test. Additionally, a model was made to predict the size of each tumor (on the log scale) as a function of BRAF status and size.
The indicates that the patient experienced a recurrence (Patient 4 at 7 months postsurgery/3 months post-radioactive iodine and Patient 9 at 19 months postsurgery/16 months post-radioactive iodine). The ‡ indicates a patient who had a possible recurrence that could not be biopsied and was treated with radioiodine. The indicates that additional tumors were tested for these patients (Patient 24: two additional nodules, Patient 25: one additional nodule); the additional nodules were all negative. Fol = follicular variant, TC = tall cell variant, Clas = classic variant, Onc = oncocytic variant, War = Warthin-like variant.|
There were 17 women and 10 men included in the study with an average age at diagnosis of 52 (range: 26 to 78). The average number of tumor nodules rounded to the nearest whole number was 3, ranging from 2 to 9. Overall, 34 of the 57 tumors (60%) were BRAF-positive and all mutations were V600E. Figure 1 illustrates the histologic variants and their respective rates of BRAF mutation. Notably, the tall cell variant had the highest rate of positivity for BRAF mutation (3 of 3 cases), and the encapsulated follicular variant had the lowest rate (1 of 6 cases). The one BRAF-positive encapsulated follicular variant showed focal invasion through the capsule and therefore should probably have been considered along with the other follicular variant tumors that showed an infiltrative growth pattern, as in Walts et al. . There was no significant correlation between BRAF-positivity and the presence of an irregular border (), location close to the capsule (), or extrathyroidal extension ().
The overall average size of the largest nodule was 1.6 cm, ranging from 0.2 cm to 3.6 cm. There was no significant difference in size between the BRAF-negative and BRAF-positive largest nodules (average sizes 1.8 cm and 1.5 cm, ). The overall average size of the second-largest nodule was 0.5 cm, ranging from 0.1 cm to 2.2 cm. There was no significant difference in size between the BRAF-negative and BRAF-positive second-largest nodules (average sizes 0.5 cm and 0.6 cm, ).
A mutation was present in at least 1 nodule in 22 of the 27 cases (81.4%). In 17 patients, both the largest and second-largest nodules had concordant BRAF status; that is, they were both either positive or negative. The remaining ten patients had discordant mutation status; that is, one tumor was positive and one tumor was negative. No histologic features were significantly different between the concordant and discordant cases. The tumors were of the same histologic variant in 11 (65%) of the concordant cases, compared with 7 (70%) of the discordant cases (), as illustrated in Figure 2. Satellite nodules were present in 53% of the concordant cases and 20% of the discordant cases (). Isolated psammoma bodies were present in 41% of the concordant cases and 20% of the discordant cases (). The largest tumor had irregular borders in 65% of concordant cases and 70% of discordant cases (). Both the largest and second-largest tumors had smooth borders in 12% of the concordant cases and 10% of the discordant cases ().
Similarly, there was no significant correlation between concordant mutation status and laterality: the largest and second-largest tumors were in the same lobe in 6 of the 17 concordant cases (35%), compared to 2 of the 10 discordant cases (20%, ). The two largest tumors were concordant in 4 of the 5 cases where the disease was unilateral (80%), compared to 13 of the 22 cases that were bilateral (59%, ).
Cases with 4 or more nodules tended to have concordant mutation status (8 of 9, or 89%) compared to cases with fewer than 4 nodules (9 of 18, or 50%) but this did not reach statistical significance ().
Lymph node dissection was performed at the discretion of the surgeon (E.K.) who was not aware of the BRAF status preoperatively. There was a significant difference in nodal stage between the patients with a BRAF-positive largest tumor and a BRAF-negative largest tumor: 11 of the 16 patients (69%) with a BRAF-positive largest tumor had cervical lymph node metastases (either N1a or N1b) compared to 2 of the 11 patients (18%) with a BRAF-negative largest tumor (). This association remained significant when a multivariate analysis was done to account for tumor size (). With regards to the largest tumors, the smallest of these were the BRAF-negative tumors with positive lymph nodes (), followed by the BRAF-positive tumors with negative lymph nodes (). There was no significant size difference between the BRAF-negative tumors with negative lymph nodes and the BRAF-positive tumors with positive lymph nodes. Of note, none of the six patients with a BRAF-negative largest tumor and BRAF-positive second-largest tumor had positive lymph nodes.
None of the patients presented with or experienced distant metastases. Overall, the average amount of clinical followup was 20 months (range <1 month to 81 months). Two of the 27 patients (7.4%) had recurrence in the cervical lymph nodes, one at 7 months postsurgery/3 months post-radioactive iodine and one at 19 months postsurgery/16 months post-radioactive iodine. These two patients were BRAF-positive in both the largest and the second-largest tumors. None of the remaining patients had a definitive recurrence. One patient with a BRAF-negative largest tumor and a BRAF-positive second-largest tumor did have positive radioactive iodine uptake in the mediastinum after surgery, but it was unclear whether this represented ectopic thyroid tissue or recurrent disease and the area was not amenable to biopsy.
In this study, we attempted to find the optimal approach for BRAF mutation testing in multifocal PTC. One hypothetical strategy would be to test each tumor sequentially, starting with the largest tumor and working towards the smallest until a mutation is found; however, this would require strong coordination between the surgical pathologist and molecular pathologist and would not be the most timely or cost-effective strategy, given that each tumor would be tested on a separate run. At the other extreme, another strategy would be to test all of the tumors up front, but since twelve of our patients (44%) had three or more tumors, testing every single nodule may be costly with overuse of resources. We sought an approach somewhere in between these two extremes.
Not surprisingly, all of our tall cell variants were BRAF-positive, which is consistent with the generally accepted idea that tall cell variant has the highest rate of BRAF positivity (around 80% ). This indicates that tumors with tall cell histology should be prioritized in any testing algorithm.
The follicular variant had the lowest rate of BRAF-positivity at 42%. This percentage is high relative to the generally accepted value of around 10% , but is in line with more recent studies that used more sensitive molecular techniques and found BRAF-positivity in 21% to 54% of follicular variants [13, 15–17]. Additionally, in the current study, three of the follicular variants that were positive were in the same gland as a BRAF-positive classic variant (see Table 1) and therefore could have been intraglandular metastases. Walts et al. specifically excluded cases like this from their study of follicular variants and were very strict on the definition of follicular variant (requiring 95% or more of the tumor to have follicular architecture); their study had a rate of 33.3% BRAF-positivity (16 of 48 cases, which included both unifocal and multifocal cases) . Therefore, follicular variant tumors should not be excluded from the testing algorithm, with the caveat that tumors that are completely encapsulated have a very low chance of being positive.
In this study, 10 cases (37%) had tumors with discordant BRAF mutation status. This is in line with other studies that have reported discordant BRAF mutation status in 14% to 39% of multifocal PTC cases [2, 3, 8, 10, 11]. Ideally, there would be some histologic feature to help identify which cases are likely to have discordant mutation status and therefore indicate that more than just the largest tumor should be tested. We first looked at whether discordant tumors tended to be of different histologic variants because in Park et al. tumors were of different histologic variants in 58.3% of discordant cases compared to 32.4% of concordant cases () ; similar observations were made by Bansal et al.  and Giannini et al. . We did not find this in our study, however, as discordant tumors were statistically equally likely to be of the same or different histologic variants.
Another variable we examined was whether concordant tumors were more often in the same lobe. In Bansal et al., tumors with the same mutation (BRAF, RAS, or RET/PTC) were statistically more likely to be in the same lobe when compared to those with different mutations (60% versus 22.2%, ) . Similarly, Kuhn et al. found that tumors with concordant X chromosome inactivation patterns tended to be in the same lobe and discordant tumors were in contralateral lobes . However, Park et al. did not find a correlation between BRAF mutation concordance and laterality: 9 of 24 unilateral cases (37.5%) had mixed BRAF status compared to 15 of 37 bilateral cases (40.5%, ) . Our findings were similar to Park et al. in that tumors with concordant mutations were not more likely to be in the same lobe, and tumors with discordant mutations were not more likely to be in opposite lobes.
We next examined whether the presence of satellite nodules indicated a concordant mutation status. In theory, satellite nodules should result from intraglandular metastasis and therefore all tumors would have the same mutational status. In Bansal et al., microscopic peritumoral dissemination (which is similar to what we are calling satellite nodules) was seen at a significantly lower rate in cases with different mutations () . We found a similar trend: cases with satellite tumors did tend to have concordant mutation status, but this did not reach statistical significance. Isolated psammoma bodies, which again should indicate intraglandular metastasis, also did not correlate significantly with concordant mutation status.
Bansal et al. noted that >60% of tumors with different mutations had smooth borders, with the idea that a smooth border indicates a less aggressive tumor that would be unlikely to develop intraglandular metastasis. In other words, multiple tumors with smooth borders would more likely represent separate primaries, and tumors with irregular borders would more likely represent intraglandular metastasis . We did not find any significant association between smooth borders and discordant mutation status, or irregular borders and concordant mutation status, however.
Cases with more than three tumor foci were more likely to have discordant BRAF status in Park et al. , but our study found a trend in the opposite direction: cases with more nodules tended to be concordant, but this did not reach statistical significance.
In summary, there does not seem to be a reliable way to predict which cases will have tumors with discordant BRAF status. However, from our limited data it does seem that testing more than two tumors may not be necessary: of the five cases where the two largest tumors were both negative, two cases had additional nodules with adequate DNA for testing, and these three additional nodules were all BRAF-negative. In other words, although our numbers were limited, we did not find any cases where a third-largest or fourth-largest tumor was BRAF-positive when the largest and second-largest tumors were BRAF-negative.
Importantly, six of our cases (22%) had a BRAF mutation in the second-largest tumor when the largest tumor was BRAF-negative. This is similar to Ahn et al., where thirteen out of eighty-five patients with multifocal PTC (15%) had a BRAF mutation in a smaller nodule when the largest was BRAF-negative . Therefore, if only the largest tumor is tested, a substantial number of patients would be considered BRAF-negative even though they harbor a smaller tumor that is BRAF-positive.
The clinical significance of finding a BRAF mutation in a nondominant tumor, however, is unknown. Among our cohort, none of the six patients with a BRAF-negative largest tumor and BRAF-positive second-largest tumor presented with positive lymph nodes or had a definitive recurrence, although one patient had a small area of positivity on a radioiodine scan that was either a positive mediastinal lymph node or ectopic thyroid tissue (the area was not amenable to biopsy and the patient was given adjuvant radioiodine). Our study is too small and the followup is too short to draw further conclusions, but the two patients who experienced a definite recurrence had a BRAF mutation in both their largest and second-largest nodules. This finding is of interest, since Ahn et al. found that the patients who were BRAF-positive in all of their nodules had significantly higher rates of extrathyroidal invasion, lymph node metastases, and postoperative radioactive iodine therapy, when compared to the group with mixed mutation status. Thus, having a BRAF mutation in all nodules likely indicates more aggressive disease, probably because the tumors represent intraglandular spread from the same primary .
We note that this study had a relatively high rate of BRAF mutation (42% of the follicular variants, as described above, and 60% of the tumors overall), which may be due to two factors: first, BRAF mutation may be more frequent in multifocal disease [18, 19] and second, our BRAF assay has high analytic sensitivity. Recent studies have found that BRAF mutation is most likely not present in every cell of a given tumor and that the percentage of mutated alleles may vary regionally within the tumor. Thus, an assay that can detect a smaller percentage of mutated alleles will call more tumors positive [20–24]. One excellent example of how analytic sensitivity can affect the rate of BRAF mutations is seen in Guerra et al., who used two different assays in their study: BigDye Terminator sequencing (PE Applied Biosystems, Foster City, CA) and pyrosequencing. By BigDye Terminator sequencing, BRAF mutation was identified in 62 of 168 (36.9%) cases of unifocal papillary thyroid carcinoma and by pyrosequencing, it was identified in 90 of 168 (53.6%). This resulted in differences in the clinical variables that were significantly associated with the presence of BRAF mutation. For example, lymph node metastasis and AJCC stage I disease had positive associations with BRAF mutation by BigDye Terminator sequencing ( and , resp.) but neither had significant associations by pyrosequencing. Recurrence was 2.1 times more likely in BRAF-positive patients by BigDye Terminator sequencing (), but there was no significant difference in recurrence by pyrosequencing . Clearly, the choice of assay can affect both the rate of BRAF mutation and its clinical utility. This means that the results from our study may not be generalizable to labs that use a less sensitive assay.
We also note that the study cohort contained a high number of male patients (10 out of 27, or 37%). One possibility for this is that the percentage of men might be higher among patients with multifocal disease as was found in Huang et al. , who studied 648 patients with PTC in China. In their study, 49 of 168 patients with multifocal disease were male (29%) compared to 89 of 480 patients with unifocal disease (19%, ). However, other studies have found nearly equal percentages of males in both multifocal and unifocal cases, ranging from 16% to 24% [11, 26, 27]. Therefore, it is more likely that the high number of male patients reflects some bias in the types of patients referred to the study surgeon. Since the percentage of cases with discordant BRAF mutations is in line with previous studies, as discussed above, it is unlikely that this bias had a great effect on our results.
Patients with multifocal PTC whose largest tumor is BRAF-negative can have smaller tumors that are BRAF-positive. Therefore, molecular testing of more than just the dominant nodule should be considered, especially if a smaller tumor has tall cell features. Future larger studies are warranted to establish whether finding a BRAF mutation in a smaller tumor correlates with more aggressive disease.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors would like to thank Tulane University’s Department of Pathology and Laboratory Medicine for funding this study, and Darl D. Flake II, MS, for additional statistical calculations. The abstract was previously presented as a poster at the American Thyroid Association’s annual meeting in 2014 (no. 2036822).
- M. Bansal, M. Gandhi, R. L. Ferris et al., “Molecular and histopathologic characteristics of multifocal papillary thyroid carcinoma,” The American Journal of Surgical Pathology, vol. 37, no. 10, pp. 1586–1591, 2013.
- R. Giannini, C. Ugolini, C. Lupi et al., “The heterogeneous distribution of BRAF mutation supports the independent clonal origin of distinct tumor foci in multifocal papillary thyroid carcinoma,” Journal of Clinical Endocrinology and Metabolism, vol. 92, no. 9, pp. 3511–3516, 2007.
- L. Jovanovic, B. Delahunt, B. McIver, N. L. Eberhardt, and S. K. G. Grebe, “Most multifocal papillary thyroid carcinomas acquire genetic and morphotype diversity through subclonal evolution following the intra-glandular spread of the initial neoplastic clone,” Journal of Pathology, vol. 215, no. 2, pp. 145–154, 2008.
- E. Kuhn, L. Teller, S. Piana, J. Rosai, and M. J. Merino, “Different clonal origin of bilateral papillary thyroid carcinoma, with a review of the literature,” Endocrine Pathology, vol. 23, no. 2, pp. 101–107, 2012.
- X. Lin, S. D. Finkelstein, B. Zhu, and J. F. Silverman, “Molecular analysis of multifocal papillary thyroid carcinoma,” Journal of Molecular Endocrinology, vol. 41, no. 3-4, pp. 195–203, 2008.
- R. P. McCarthy, M. Wang, T. D. Jones, R. W. Strate, and L. Cheng, “Molecular evidence for the same clonal origin of multifocal papillary thyroid carcinomas,” Clinical Cancer Research, vol. 12, no. 8, pp. 2414–2418, 2006.
- S. Moniz, A. L. Catarino, A. R. Marques, B. Cavaco, L. Sobrinho, and V. Leite, “Clonal origin of non-medullary thyroid tumours assessed by non-random X-chromosome inactivation,” European Journal of Endocrinology, vol. 146, no. 1, pp. 27–33, 2002.
- S. Y. Park, Y. J. Park, Y. J. Lee et al., “Analysis of differential BRAF V600E mutational status in multifocal papillary thyroid carcinoma: evidence of independent clonal origin in distinct tumor foci,” Cancer, vol. 107, no. 8, pp. 1831–1838, 2006.
- T. M. Shattuck, W. H. Westra, P. W. Ladenson, and A. Arnold, “Independent clonal origins of distinct tumor foci in multifocal papillary thyroid carcinoma,” The New England Journal of Medicine, vol. 352, no. 23, pp. 2406–2412, 2005.
- W. Wang, H. Wang, X. Teng et al., “Clonal analysis of bilateral, recurrent, and metastatic papillary thyroid carcinomas,” Human Pathology, vol. 41, no. 9, pp. 1299–1309, 2010.
- X. Zheng, T. Xia, L. Lin et al., “BRAF V600E status and clinical characteristics in solitary and multiple papillary thyroid carcinoma: experience of 512 cases at a clinical center in China,” World Journal of Surgical Oncology, vol. 10, article 104, 2012.
- M. M. Xing, A. S. Alzahrani, K. A. Carson et al., “Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer,” Journal of the American Medical Association, vol. 309, no. 14, pp. 1493–1501, 2013.
- A. E. Walts, J. M. Mirocha, and S. Bose, “Follicular variant of papillary thyroid carcinoma (FVPTC): histological features, BRAF V600E mutation, and lymph node status,” Journal of Cancer Research and Clinical Oncology, 2015.
- Y. E. Nikiforov, “Molecular diagnostics of thyroid tumors,” Archives of Pathology and Laboratory Medicine, vol. 135, no. 5, pp. 569–577, 2011.
- Y. Y. Jung, J. H. Yoo, E. S. Park et al., “Clinicopathologic correlations of the BRAFV600E mutation, BRAF V600E immunohistochemistry, and BRAF RNA in situ hybridization in papillary thyroid carcinoma,” Pathology—Research and Practice, vol. 211, pp. 162–170, 2015.
- R. K. Virk, A. L. van Dyke, A. Finkelstein et al., “BRAFV600E mutation in papillary thyroid microcarcinoma: a genotype-phenotype correlation,” Modern Pathology, vol. 26, no. 1, pp. 62–70, 2013.
- H. S. Min, C. Lee, and K. C. Jung, “Correlation of immunohistochemical markers and BRAF mutation status with histological variants of papillary thyroid carcinoma in the Korean population,” Journal of Korean Medical Science, vol. 28, no. 4, pp. 534–541, 2013.
- H. Y. Ahn, Y. J. Chung, B. S. Kim et al., “Clinical significance of the BRAF V600E mutation in multifocal papillary thyroid carcinoma in Korea,” Surgery, vol. 155, no. 4, pp. 689–695, 2014.
- C. Li, K. C. Lee, E. B. Schneider, and M. A. Zeiger, “BRAF V600E mutation and its association with clinicopathological features of papillary thyroid cancer: a meta-analysis,” Journal of Clinical Endocrinology and Metabolism, vol. 97, no. 12, pp. 4559–4570, 2012.
- S. P. Cheng, Y. C. Hsu, C. L. Liu et al., “Significance of allelic percentage of BRAF c.1799T > A (V600E) mutation in papillary thyroid carcinoma,” Annals of Surgical Oncology, vol. 21, no. 4, pp. 619–626, 2014.
- D. de Biase, V. Cesari, M. Visani et al., “High-sensitivity BRAF mutation analysis: BRAF V600E is acquired early during tumor development but is heterogeneously distributed in a subset of papillary thyroid carcinomas,” The Journal of Clinical Endocrinology & Metabolism, vol. 99, no. 8, pp. E1530–E1538, 2014.
- G. Gandolfi, V. Sancisi, F. Torricelli et al., “Allele percentage of the BRAF V600E mutation in papillary thyroid carcinomas and corresponding lymph node metastases: no evidence for a role in tumor progression,” The Journal of Clinical Endocrinology and Metabolism, vol. 98, no. 5, pp. E934–E942, 2013.
- A. Guerra, M. R. Sapio, V. Marotta et al., “The primary occurrence of BRAFV600E is a rare clonal event in papillary thyroid carcinoma,” The Journal of Clinical Endocrinology & Metabolism, vol. 97, no. 2, pp. 517–524, 2012.
- A. Guerra, L. Fugazzola, V. Marotta et al., “A high percentage of BRAFV600E alleles in papillary thyroid carcinoma predicts a poorer outcome,” Journal of Clinical Endocrinology and Metabolism, vol. 97, no. 7, pp. 2333–2340, 2012.
- G. Huang, X. Tian, Y. Li, and F. Ji, “Clinical characteristics and surgical resection of multifocal papillary thyroid carcinoma: 168 cases,” International Journal of Clinical and Experimental Medicine, vol. 7, no. 12, pp. 5802–5807, 2014.
- H. Mazeh, Y. Samet, D. Hochstein et al., “Multifocality in well-differentiated thyroid carcinomas calls for total thyroidectomy,” The American Journal of Surgery, vol. 201, no. 6, pp. 770–775, 2011.
- K.-J. Kim, S.-M. Kim, Y. S. Lee, W. Y. Chung, H.-S. Chang, and C. S. Park, “Prognostic significance of tumor multifocality in papillary thyroid carcinoma and its relationship with primary tumor size: a retrospective study of 2,309 consecutive patients,” Annals of Surgical Oncology, vol. 22, no. 1, pp. 125–131, 2015.
Copyright © 2015 Hillary Z. Kimbrell et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.