• multifocal papillary cancers;
  • papillary thyroid cancers;
  • BRAFV600E mutation


  1. Top of page
  2. Abstract


Papillary thyroid cancers often occur as multiple foci. Multifocal cancers have been considered to have a poor prognosis because they are thought to be the consequence of intrathyroidal spread of the papillary cancer. However, to the authors' knowledge there has been little investigation into whether multifocal thyroid papillary carcinomas arise from the intrathyroidal spread of a single carcinoma or from independent primary tumors. To answer this question, the BRAFV600E mutational status of individual tumor foci was examined. This approach was justified because in the Korean population a high proportion (65%) of papillary carcinomas harbor the BRAF mutation.


DNA was isolated from paraffin-embedded tissue samples of multifocal papillary thyroid carcinoma and the BRAF exon 15 was amplified by the polymerase chain reaction (PCR). The PCR product was digested with restriction endonuclease TspRI to test for the presence of the BRAFV600E (T1799A) mutation.


In all, 140 cancers from 61 patients diagnosed with multifocal papillary carcinoma were examined. The BRAF mutation was found in all the individual cancers in 29 (47.5%) of the patients (all-positive group) and the mutation was absent in all the individual cancers in 8 (13.1%) patients (all-negative group). However, in 24 (39.3%) patients, some of the individual cancers contained the BRAF mutation, whereas others did not (mixed group).


At least 39.3% of the multifocal papillary cancers in the Korean population that were examined could be attributed to independently arising papillary cancers rather than to intrathyroidal spread of single cancers. Cancer 2006. © 2006 American Cancer Society.

Papillary thyroid cancers (PTCs) often present as multifocal tumors. According to postsurgical pathologic analyses, 18% to 87% of PTCs have multiple noncontiguous tumor foci in individual glands.1–4 The decision as to whether the multiple foci arise independently or are the result of intraglandular metastasis of a single tumor is of particular interest because intraglandular metastasis of a single tumor indicates that the tumor is aggressive and suggests that a correspondingly aggressive treatment is required. In fact, it has been reported that multifocal tumors are associated with elevated risks of lymph node and distant metastases, persistent local disease after initial treatment, and regional recurrence.1, 2, 4, 5 These results support the conclusion of the National Thyroid Cancer Treatment Cooperation Study Registry that multifocal tumors have higher associated risks.5, 6 However, other reports indicate that multiple foci can arise independently, and that an individual patient's thyroid glands may simply be prone to develop papillary thyroid cancer.7, 8 Recently, Shattuck et al.9 reported that the patterns of X chromosomal inactivation in 5 of 10 multifocal PTCs were heterogeneous, indicating independent origins.

Most PTCs have unique clonal genetic alterations, such as RET/PTC or NTRK1 rearrangements or BRAFV600E mutations.10–17 An activating mutation of BRAFV600E is found in 36% to 69% of these PTCs, making it the most common genetic event in PTC.14–17 However, the genetic alterations in PTC in the Korean population are quite different from those in other populations: the RET/PTC rearrangement is found in a relatively small number of patients (9.1%–12.9%),18, 19 whereas the BRAFV600E mutation is present in many (66%–83%) (reference 20 and unpublished data). Because the BRAFV600E mutation is found in greater than two-thirds of PTCs in Korea, we reasoned that an analysis of the presence or absence of the BRAFV600E mutation in multiple tumor foci could help to define the genetic alterations in the individual tumors and provide useful information about whether these multifocal tumors arise independently or are the result of intraglandular metastasis of single tumors.

In the present study we assessed the origin of multifocal PTCs by analyzing the BRAFV600E mutation in 140 individual tumors in the multifocal PTCs of 61 patients. The status of the BRAFV600E mutation was heterogeneous in 39.3% of the multifocal PTCs, suggesting that the individual tumors arise independently in a significant subset of multifocal PTCs.


  1. Top of page
  2. Abstract

Patients and Samples

One hundred and seventy patients underwent total or subtotal thyroidectomy for PTC from May 2003 to February 2005 at Seoul National University Bundang Hospital. One hundred nine patients had solitary PTC and 61 (35.9%) patients had multifocal PTC. Of the 61 patients with multifocal PTC, 4 had multiple clinically evident carcinomas (> 1 cm), 31 had 1 clinically evident carcinoma and a single microcarcinoma or multiple microcarcinomas (≤ 1 cm), and 26 had multiple microcarcinomas.

Forty-six patients had 2 separate PTC foci, 13 patients had 3, 1 patient had 4, and another had 5. Thus, we investigated a total of 140 PTC foci in 61 patients. Of the 61 patients, 23 (37.7%) had metastases in cervical lymph nodes. The metastatic lymph nodes in 10 of the 23 patients were included in the study. Nine patients were male (15%) and 52 were female (85%), with a mean age of 51.4 ± 10.4 years (mean ± standard deviation [SD]; range, 24–74 years).

Histologic Evaluation of Tissue Specimens

Paraffin-embedded tumor samples from the 61 patients with multifocal PTCs were obtained in accordance with a protocol approved by the Institutional Review Board of Seoul National University Bundang Hospital. The patients provided informed consent as required by this protocol. Tumor and normal tissue in each sample were identified by an endocrine pathologist (S.Y.P.).

Resected surgical specimens were cut at 3-mm intervals and fixed in 10% formalin. One to 3 representative sections of the tumor and all suspicious lesions were submitted for microscopic examination. All sections were fixed in formalin, embedded in paraffin wax, and stained with hematoxylin and eosin (H & E) for histologic examination.

The diagnosis of papillary carcinoma was based on characteristic architectural features (i.e., the presence of true papillae) and/or characteristic nuclear changes, such as ground glass nuclei, nuclear pseudoinclusions, and nuclear grooves. The histologic classification was made according to the criteria of Rosai et al.21 without information regarding BRAF status. The histology of the individual tumors in multifocal PTCs was compared and multifocal PTCs were divided into 2 groups, those with tumors differing in histology and those with tumors of similar histology.

DNA Extraction

Genomic DNA was extracted from paraffin-embedded tissue. Briefly, the tumor areas were marked using 4-μm-thick H & E-stained sections as a guide. Then the marked areas were matched with dewaxed but unstained 20-μm-thick sections. The tumor areas were dissected from the unstained slides and transferred into Eppendorf tubes. After the dissection, the blocks were cut into 4-μm sections for H & E staining to confirm the tumor continuity. All samples were then digested with proteinase K for more than 24 hours at 56°C and DNA was isolated from the digested tissue using a Tissue SV mini kit (General Biosystems, Seoul, Korea).

PCR-RFLP and Sequencing

The BRAF mutation reported recently in thyroid carcinomas is confined to exons 11 and 15. We therefore amplified BRAF exon 15 by polymerase chain reaction (PCR) using the following primers: forward, 5′-GCTTGCTCTGATAGGAAAATGAG-3′; reverse, 5′-GATACTCAGCAGCATCTCAGG-3′. PCRs conditions were: initial denaturation at 95°C for 5 minutes, followed by 40 cycles of denaturation at 94°C for 20 seconds, annealing at 56°C for 20 seconds, and elongation at 72°C for 20 seconds, and a final extension at 72°C for 10 minutes. The denatured PCR products were electrophoresed and digestion of the 237-base pair (bp) PCR fragment with restriction endonuclease TspRI yielded 3 major bands of 117 bp, 87 bp, and 33 bp for the wildtype allele. The T1799A mutation abolished the restriction sites, resulting in a prominent band of 237 bp from the mutant allele and residual bands from the normal allele (Fig. 1).

thumbnail image

Figure 1. Polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) analysis and sequence of codon 600 of the BRAF gene. (A) Results of PCR-RFLP of the wild-type and mutant BRAF gene. (B) Sequence of the V600E BRAF gene. Because the 1799A mutant allele of the BRAF gene and the normal 1799T allele were both present, the sequencing result show a mixture of T and A at the 1799 site.

Download figure to PowerPoint

To confirm the reliability of the PCR-restriction fragment length polymorphism (RFLP) results, we sequenced some 20 randomly selected samples in a PTC-225 Peltier Thermal Cycler (MJ Research, Waltham, MA) using ABI Prism BigDye Terminator Cycle Sequencing Kits (Applied Biosystems, Foster City, CA) and AmpliTaq DNA polymerase (FS enzyme) (Applied Biosystems, Foster City, CA). Single-pass sequencing was performed on each template using the forward primer. The fluorescence-labeled fragments obtained were purified from unincorporated terminators by ethanol precipitation. Samples were then resuspended in distilled water and subjected to electrophoresis in an ABI 3730xl sequencer (Applied Biosystems). All 20 sequences so determined confirmed the BRAFV600E status indicated by the PCR-RFLP analysis.

Statistical Analysis

Statistical analyses were performed with SPSS software (version 11.0; SPSS Inc., Chicago, IL). Chi-square or Fisher exact tests were used when comparing frequencies between groups. All numeric data were expressed as means ± SD and differences between group means were compared by the independent-sample Student t-test or the Mann-Whitney U-test. Probability values of < .05 were considered statistically significant.


  1. Top of page
  2. Abstract

Clinicopathologic Features of Multifocal and Solitary PTCs

To identify the clinical and pathological factors associated with PTC multifocality, we compared the clinicopathologic features of 61 patients with multifocal PTCs with those of 109 patients with solitary PTCs. Patients with multifocal PTCs were characterized by greater age at presentation (P = .001) and a higher frequency of lymph node metastasis (P = .019). In addition, a background of lymphocytic or Hashimoto thyroiditis tended to be present at a higher frequency in patients with multifocal PTCs, but this effect was not statistically significant (P = .056). No significant differences were found between the 2 groups in terms of other clinicopathologic characteristics, which included tumor size, perithyroidal extension, and morphologic variants of PTC (Table 1).

Table 1. Clinicopathologic Features of Multifocal Papillary Thyroid Carcinomas
 Multifocal PTCs (n = 61)Solitary PTCs (n = 109)P
  • PTC indicates papillary thyroid carcinoma.

  • *

    Statistically significant.

  • The largest tumor.

Age, y51.1 ± 10.445.1 ± 11.2.001*
Male:female ratio9:5213:96.598
Tumor size, cm1.30 ± 0.741.31 ± 0.76.951
Extrathyroidal extension39 (63.9%)63 (57.8%).433
Thyroiditis23 (37.7%)26 (23.9%).056
Lymph node metastasis23 (37.7%)23 (21.1%).019*
Tall cell variant3 (4.9%)3 (2.8%).668

BRAFV600E Mutation in Multifocal PTCs

The BRAFV600E mutation was detected in 52 (86.9%) of the 61 patients with multifocal PTCs and in 95 (67.9%) of 140 individual tumors. The BRAFV600E mutation occurred in 31 (78.5%) of 39 clinically evident papillary carcinomas and 64 (63.4%) of 101 microcarcinomas (P = .068). In the 39 clinically evident PTCs, the BRAFV600E mutation was seen in 25 (83.3%) of 30 conventional PTCs, all of 4 tall cell variants, 1 of 2 oncocytic variants, and 1 (33%) of 3 follicular variants. In addition, the BRAFV600E mutation was present in 59 (72.8%) of 81 microcarcinomas with the conventional pattern and 5 (26.3%) of 19 microcarcinomas with the follicular pattern (P < .001). Of the 4 tall cell variants, 2 occurred in a single gland as multifocal tumors, whereas the other 2 cases developed with microcarcinoma of the conventional type with the BRAFV600E mutation.

Among the 61 patients with multifocal PTCs, 29 were BRAFV600E-positive at all foci (47.6%), 8 (13.1%) were all-negative, and 24 (39.3%) were mixed (i.e., BRAFV600E-positive and BRAFV600E-negative tumors coexisted in the same patient). Among the 24 patients with BRAFV600E-mixed multifocal PTCs, 14 had 2 tumor foci, 8 had 3 (2 BRAFV600E-positive foci in 6 patients; 1 BRAFV600E-positive focus in 2 patients), 1 patient had 4 (3 BRAFV600E-positive foci), and the remaining patient had 5 (3 BRAFV600E-positive foci). When we analyzed BRAFV600E status in multifocal PTCs versus tumor number (Table 2) we found that BRAFV600E-mixed multifocal tumors were more frequent in patients with ≥ 3 tumors compared with patients with 2 tumors (P = .012).

Table 2. BRAFV600E Mutation Status as a Function of Tumor Number
No. of tumorsTotal2345
  1. P = .012; mixed BRAF cases in patients with two papillary thyroid carcinomas (PTCs) versus patients with ≥ 3 PTCs.

No. of patients61461311
Range of tumor size, cm0.05–4.6 (0.89 ± 0.72)0.05–3.5 (0.89 ± 0.63)0.2–4.6 (0.86 ± 0.83)0.4–3.7 (1.38 ± 1.56)0.2–1.1 (0.60 ± 0.35)
BRAFT1796A mutation (%)
All positive8 (13.1%)6 (13.1%)2 (15.4%)0 (0%)0 (0%)
All negative29 (47.6%)26 (56.5%)3 (23.1%)0 (0%)0 (0%)
Mixed24 (39.3%)14 (30.4%)8 (61.5%)1 (100%)1 (100%)

Clinicopathologic Features According to BRAFV600E Mutational Status

Table 3 lists the clinicopathologic features of multifocal PTCs in relation to BRAFV600E status. Patients with BRAFV600E-all-negative multifocal PTCs were younger than those with BRAFV600E-all-positive or -mixed multifocal PTCs. The histology of the individual tumors differed in 58.3% of BRAFV600E-mixed multifocal PTCs and 32.4% of BRAFV600E-all-positive or -all-negative multifocal PTCs (P = .047) (Fig. 2). No significant difference was found with regard to size, extrathyroidal extension, background of thyroiditis, or lymph node involvement between these 3 groups.

thumbnail image

Figure 2. Histology of the individual tumors of multifocal papillary thyroid carcinoma (PTC). (A, B) A case of BRAFV600E-mixed multifocal PTC with different histology. (A) A conventional PTC with the BRAFV600E mutation. (B) A follicular variant of PTC, with no BRAFV600E mutation. (C, D) A case of BRAFV600E-all-positive multifocal PTC with similar histology. (C) A conventional PTC with the BRAFV600E mutation. (D) A microcarcinoma of the conventional pattern with the BRAFV600E mutation.

Download figure to PowerPoint

Table 3. Clinicopathologic Features of Multifocal Papillary Thyroid Carcinomas in Relation to BRAFV600E Status
 BRAFV600E mutation
All negative (n = 8)All positive (n = 29)Mixed (n = 24)
  • PTC indicates papillary thyroid carcinoma.

  • *

    Small: ≤ 1 cm; large: > 1 cm.

  • † ‡, †

    P = .047.

  • P = .013.

  • §

    P = .047, BRAFV600E-mixed vs. BRAFV600E -all-positive or -all-negative.

Age, y43.3 ± 8.2† ‡51.2 ± 9.654.4 ± 10.9
Male:female ratio1:75:243:21
Size (small:mixed:large*)4:4:014:12:38:15:1
No. of involved lobes (1:2)6:29:208:16
Extrathyroidal extension5 (62.5%)17 (58.6%)17 (70.8%)
Thyroiditis5 (62.5%)9 (31.0%)9 (37.5%)
Lymph node metastasis2 (25.0%)9 (31.0%)12 (50.0%)
Different histology2 (25.0%)10 (34.5%)14 (58.3%)§

Lobal Distribution of Multifocal PTCs in Relation to the BRAFV600E Mutation

To investigate whether clonal origin can be predicted from the number of lobes involved, we analyzed BRAFV600E mutational status as a function of the lobal distribution pattern of the tumor foci in each patient. Of 24 unilaterally involved multifocal PTCs, 9 (37.5%) were BRAFV600E-mixed, and of 37 bilaterally involved multifocal PTCs, 15 (40.5%) were BRAFV600E-mixed (chi-square = 0.056, degrees of freedom [df] = 1; P > .05).

BRAFV600E Mutational Status in Metastatic Lymph Nodes

To assess the relation between the BRAF mutational status of primary tumors and of metastatic tumors in lymph nodes, we also performed BRAFV600E mutational analyses on the lymph node samples of 10 multifocal PTCs, 4 of which were BRAFV600E-all-positive and 6 BRAFV600E-mixed. The 4 BRAFV600E-all-positive multifocal PTCs also had the BRAFV600E mutation in the metastatic lymph nodes. Five of 6 BRAFV600E-mixed multifocal PTCs had the BRAFV600E mutation in their metastatic lymph nodes and 1 had wild-type BRAF (Table 4).

Table 4. Comparison of the BRAFV600E Mutational Status of Multifocal Papillary Carcinomas of the Thyroid and the Associated Metastatic Carcinomas of the Lymph Nodes
Case no.No. of tumorsBRAFV600E mutational status
Individual tumor foci of thyroid*Metastatic lymph nodes
  • *

    Arranged according to tumor size.



  1. Top of page
  2. Abstract

In the present study, we evaluated the origins of multifocal PTCs by analyzing BRAFV600E mutation status in the individual tumors in a large series of multifocal PTCs. Our results showed that BRAFV600E mutational status was heterogeneous in 39.3% of the distinct foci of multifocal PTCs. This finding suggests that individual tumors arise independently in a subset of multifocal PTCs, but that there remains a possibility of intrathyroidal metastasis in the remainder.

Multifocal PTCs are common, but the clinical significance of the multiplicity and their origin is not clear. Multiple foci may be the result of intrathyroidal metastases from a single primary tumor or, alternatively, the tumors may arise independently from separate progenitor cells. The origin of multifocal PTCs may be an important predictor of behavior, because if a given multifocal PTC has resulted from intrathyroidal metastasis, there is likely to be an increased risk of further metastases. The lymphatic drainage system of the thyroid gland supports this possibility because an abundant network of intralobular lymphatic vessels anastomoses and penetrates into the capsule throughout the gland,22 and this would favor spread to other parts of the gland. Our results also show that lymph node metastasis of multifocal PTCs is more frequent than that of solitary PTCs. Iida et al.1 reported that many small foci are histologically identical to the larger cancer nodules in the same gland, suggesting that the smaller tumors are the result of metastasis of the larger tumor. Conversely, a few multifocal PTC cases showing a distinct pattern of pathology and RET-PTC rearrangement status have been reported,7, 8 suggesting an independent origin of each tumor. Recently, Shattuck et al.9 also reported distinct patterns of X chromosome inactivation in multiple distinct foci of 17 multifocal PTCs using a microdissection technique. Their results favored the independent clonal origin of distinct foci in at least 50% of the cases examined.

To our knowledge, few methods of distinguishing tumor origins have been reported. In multicentric breast cancer a study was conducted to evaluate histologic and immunohistochemical features with a view toward determining tumor origin.23 Inactivation of the X chromosome is known to be independent of neoplastic selection and occurs before cell transformation; therefore, it is believed that it is possible to determine whether tumor cells originate from a single precursor or multifocal precursors by investigating X-chromosome inactivation status.8, 9, 24, 25

We cannot insist that all the multifocal tumors of different BRAF mutational status are of independent origin because the use of mutational status as a marker of clonality has limitations. In fact, a metastatic tumor may appear different from the primary tumor at the genetic level because of progressive accumulation of genetic alterations, and a genetic alteration is useful as a marker of clonality only to the degree that it occurs early during tumorigenesis. In a recent study, there was no overlap between PTCs with RET/PTC, BRAF, or RAS mutations, which were present in a total of 66% of cases.16 This lack of concordance provides compelling genetic evidence for a requirement for mutation of MAPK signaling components for transformation to PTC. PTC could be induced in transgenic mice by targeting expression of the BRAFV600E mutation to thyroid cells, suggesting a fundamental role of BRAF in the initiation of PTC.26 The presence of the BRAF mutation in papillary microcarcinomas also supports the idea that the BRAF mutation occurs early in the development of PTC.27–30 We therefore assumed that BRAF mutation is an early event in PTC tumorigenesis and used it as a marker of clonality.

In the present study, approximately 40% of multifocal PTCs cases had mixed BRAFV600E mutational status, which supports the notion that a large percentage of multifocal PTCs are the result of the parallel development of primary tumors. However, it cannot be said that the separate foci of BRAFV600E-all-positive (or negative) multifocal PTCs always have a unique clonal origin, because the BRAF mutation is a common genetic alteration of PTCs.

We showed that the BRAF mutation status of metastatic lymph nodes matched that of the multifocal primary tumors. It is interesting that 5 of 6 BRAFV600E-mixed multifocal PTCs had the BRAFV600E mutation in their metastatic lymph nodes. Although the largest tumor of the BRAFV600E-mixed multifocal PTCs harbored the BRAFV600E mutation, this result suggests that this mutation is associated with lymph node metastasis in BRAFV600E-mixed multifocal PTCs.

It would be useful to be able to determine whether additional PTC foci had an independent or common origin before deciding on the extent of surgery (unilateral or bilateral thyroidectomy) or other options (radioablation of remaining tissue). Therefore, we investigated whether the numbers of tumor foci or involved lobes could differentiate between these 2 types. However, when we analyzed BRAFV600E mutational status according to tumor number, we found that BRAFV600E-mixed cases were more frequent in patients with ≥ 3 tumor foci than in those with 2 foci, indicating that tumor focus number cannot be used to predict the presence of intrathyroidal metastasis. Similarly, the number of lobes involved did not appear to be helpful in predicting whether the multiple foci derived from independent events or resulted from intrathyroidal metastasis, because the frequency of BRAFV600E-mixed PTCs was not different in unilaterally or bilaterally involved multifocal PTCs.

The finding that multiple PTC foci can have independent origins has implications for pathogenesis. Because neoplastic transformation is normally a rare event, it is unlikely that many cells within the same gland undergo independent transformation in the absence of some predisposing influence such as an environmental insult. However, the reason why mutational statuses differ, and the role of the BRAFV600E mutation in tumor development, remain to be clarified.

The BRAFV600E mutation of PTCs is reported to show a clear subtype-related pattern.27–29 Our results confirmed that BRAFV600E mutation is more prevalent in tall cell and conventional PTCs than in the follicular variant of PTC, although cases of follicular and tall cell variant were too few to show statistically significant differences. Microcarcinoma with follicular pattern also had a significantly lower incidence of the BRAFV600E mutation than those with the conventional pattern, in agreement with previous reports.29, 30

We divided the multifocal PTCs into 2 groups with similar or different histology on the basis of histologic findings. As expected, the histology of the individual tumor foci was more different in BRAFV600E-mixed multifocal cases than in BRAFV600E-all-positive or -all-negative cases. However, the histology of the individual tumors was similar in as many as 41.7% of the BRAFV600E-mixed cases, indicating that comparison of histology alone cannot establish the origin of multifocal PTCs. Moreover, even when multifocal lesions are morphologically different, we cannot be sure that they are of independent origin.

The results of the current study demonstrate that the status of the BRAFV600E mutation was heterogeneous in 39.3% of multifocal PTCs, which suggests that the individual tumors arise independently in a significant subset of multifocal PTCs.


  1. Top of page
  2. Abstract