Purpose: To investigate the overall occurrence and relationship of genetic alterations in the phosphatidylinositol 3-kinase (PI3K)/Akt pathway in thyroid tumors and explore the scope of this pathway as a therapeutic target for thyroid cancer.

Experimental Design: We examined collectively the major genetic alterations and their relationship in this pathway, including PIK3CA copy number gain and mutation, Ras mutation, and PTEN mutation, in a large series of primary thyroid tumors.

Results: Occurrence of any of these genetic alterations was found in 25 of 81 (31%) benign thyroid adenoma (BTA), 47 of 86 (55%) follicular thyroid cancer (FTC), 21 of 86 (24%) papillary thyroid cancer (PTC), and 29 of 50 (58%) anaplastic thyroid cancer (ATC), with FTC and ATC most frequently harboring these genetic alterations. PIK3CA copy gain was associated with increased PIK3CA protein expression. A mutual exclusivity among these genetic alterations was seen in BTA, FTC, and PTC, suggesting an independent role of each of them through the PI3K/Akt pathway in the tumorigenesis of the differentiated thyroid tumors. However, coexistence of these genetic alterations was increasingly seen with progression from differentiated tumor to undifferentiated ATC. Their coexistence with BRAF mutation was also frequent in PTC and ATC.

Conclusions: The data provide strong genetic implication that aberrant activation of PI3K/Akt pathway plays an extensive role in thyroid tumorigenesis, particularly in FTC and ATC, and promotes progression of BTA to FTC and to ATC as the genetic alterations of this pathway accumulate. Progression of PTC to ATC may be facilitated by coexistence of PI3K/Akt pathway–related genetic alterations and BRAF mutation. The PI3K/Akt pathway may thus be a major therapeutic target in thyroid cancers.

The phosphatidylinositol-3-kinase (PI3K)/Akt signaling pathway plays an important role in the regulation of cell growth, proliferation, and survival and is involved in human tumorigenesis (13). PI3Ks are composed of heterodimers of a p85 regulatory subunit and one of the several p110 catalytic subunits. Among several isoforms of the catalytic subunits, only the α-type, PIK3CA, has been shown to harbor oncogenic mutations or amplifications in its gene in human cancers (46). The regulatory subunit of PI3K can specifically bind protein factors, including Ras, integrate various signals from membrane receptors, and activate PIK3CA (7). Upon activation, PIK3CA phosphorylates phosphatidylinositol-4,5-bisphosphate to produce phosphatidylinositol-3,4,5-trisphosphate, which localizes the Ser/Thr kinase Akt to cell membrane where it becomes activated by a phosphoinositide-dependent kinase. Activated Akt phosphorylates downstream protein effectors and amplifies the signaling cascade, enhancing cell proliferation and survival (13). Signaling of this PI3K/Akt pathway is antagonized by the tumor-suppressor gene PTEN product, PTEN, a phosphatase that dephosphorylates phosphatidylinositol-3,4,5-trisphosphate, hence terminating the signaling of the PI3K/Akt pathway (811).

Follicular epithelial cell–derived thyroid tumors are the most common endocrine neoplasms and is histologically classified into benign thyroid adenoma (BTA), papillary thyroid cancer (PTC), follicular thyroid cancer (FTC), and anaplastic thyroid cancer (ATC). PTC and FTC are collectively called differentiated thyroid cancer (DTC) as they possess differentiated features of their origin cell and generally have a good prognosis. ATC, by contrast, is undifferentiated and is the most aggressive thyroid cancer that may arise de novo or from one of the differentiated cancers. Previous studies have shown increased activities of the PI3K/Akt pathway in thyroid cancers (1216). Recently, we reported genomic copy number gain or amplification of the PIK3CA in thyroid tumors, particularly FTC and ATC (17, 18). Mutation of this gene is relatively common in ATC (16), but uncommon in DTC (16, 17). Inactivating mutations and deletions in PTEN gene also occurred in sporadic thyroid cancers, although they were infrequent (19, 20). In contrast, Ras mutation, a classic oncogenic alteration, is commonly found in thyroid tumors, particularly FTC and BTA (21). Mutated Ras may promote thyroid tumorigenesis through the classic Ras → Raf → MEK → mitogen-activated protein (MAP)/extracellular signal-regulated kinase (ERK) pathway (termed “MAP kinase pathway” hereafter) or through its interaction with the PI3K/Akt pathway.

No published genetic evidence is available to support whether these genetic alterations in the PI3K/Akt pathway can each independently be a sufficient oncogenic event in activating this pathway, a hypothesis that could be tested by examining their mutual exclusivity. There is also no published genetic evidence as to how extensive the PI3K/Akt pathway is involved in thyroid tumor development and progression. It remains to be established whether this pathway is a major therapeutic target in thyroid cancer. To explore the answers to these issues at a genetic level, we conducted the present study to investigate the patterns of these genetic alterations and their relationship in various thyroid tumors.

Human thyroid tissues and DNA isolation. Thyroid tumors were partially from a previous study (17) and additionally provided by collaborators at AmeriPath (Indiana, IN; S.Y.), Johns Hopkins Hospital (Baltimore, MD; K.S.), and M.D. Anderson Cancer Center (Houston, TX; A.K.E.), with approval by the related institutional review boards. A total of 303 tumor samples were analyzed for this study, including 86 FTC, 86 PTC, 50 ATC, and 81 BTA. Tumors were prepared and DNA was isolated from paraffin-embedded samples as previously described (17). Briefly, after a treatment for 8 to 10 h at room temperature with xylene to remove paraffin, tissues were digested with 1% SDS and 0.5 mg/mL proteinase K at 48°C for 48 h, with addition of several spiking aliquots of concentrated proteinase K to facilitate digestion. DNA was subsequently isolated after standard phenol-chloroform extraction and ethanol precipitation protocols.

Copy number analysis of PIK3CA with real-time quantitative PCR. We previously used florescence in situ hybridization to validate a quantitative real-time PCR technique in evaluating copy number gain of the PIK3CA gene in thyroid tumors (17). We used this real-time PCR technique in the present study to analyze copy number of the PIK3CA gene using a PE Applied Biosystem ABI 7900HT TaqMan sequence detector (PE Applied Biosystem, Foster City, CA) following the manufacturer's instructions. Specific primers and probes were designed using a software from Applied Biosystems (Foster City, CA) to amplify both the PIK3CA and β-actin genes. For the PIK3CA gene, the TaqMan probe used was 5′-6-carboxyfluorescein-CACTGCACTGTTAATAACTCTCAGCAGGCAAA-tetramethyrhodamine-3′, and the primers were 5′-AAATGAAGCTCACTCTGGATTCC-3′ (forward) and 5′-TGTGCAATTCCTATGCAATCG-3′ (reverse). For the β-actin gene, the probe was 5′-6-carboxyfluorescein-ATGCCCTCCCCCATGCCATCC-tetramethyrhodamine-3′, and the primers were 5′-TCACCCACACTGTGCCCATCTACGA-3′ (forward) and 5′-TCGGTGAGGATCTTCATGAGGTA-3′ (reverse). Using a PCR protocol described previously (22), the samples were run in triplicate, and primers and probes to β-actin were run in parallel to standardize the input DNA. Standard curves were established using serial dilutions of normal leukocyte DNA with a quantity range of 0.01 to 20 ng. Copy number gain of the PIK3CA gene was defined by a number greater than or equal to four.

Mutational analysis for PIK3CA, Ras, PTEN, and BRAF genes. For PIK3CA gene, we selected exon 9 (for the regulatory helical domain) and exon 20 (for the kinase domain) for mutation analysis as recent large-scale analysis of PIK3CA mutations in different human tumors revealed that >80% of the mutations clustered within these domains (23, 24). Genomic DNA was amplified by PCR using the amplifying and sequencing primers for these exons of PIK3CA gene as described previously (23). The PCR was carried out in 20 μL of reaction mixture containing ∼60 ng genomic DNA, 16.6 mmol/L ammonium sulfate, 67 mmol/L Tris (pH 8.8), 5% DMSO, 2 mmol/L MgCl2, 10 mmol/L 2-mercaptoethanol, 200 μmol/L of each deoxynucleotide triphosphate (dATP, dCTP, dGTP, and dTTP), 200 nmol/L of each primer (forward and reverse), and 0.5 units platinum DNA Taq polymerase (Invitrogen Life Technologies, Inc., Gaithersburg, MD). Step-down PCR was done as follows: after a 4-min denaturation at 94°C, the PCR was run with each temperature for 1 min at six step-down steps, for two cycles each. The denaturing temperature was 95°C, and extension temperature was 72°C for each step, with the annealing temperatures of 66°C, 64°C, 62°C, 60°C, 58°C, and 56°C from the first to the last step. The PCR was finally run at 94°C, 56°C, and 72°C each for 1 min for 40 cycles, followed by an elongation at 72°C for 5 min.

As majority of the Ras mutations were found in N2-Ras, H2-Ras, and K1-Ras, particularly N2-Ras, in thyroid tumors (21), we focused our mutation analysis on these loci. Genomic DNA was amplified by PCR using the amplifying and sequencing primers for Ras genes as described previously (21). The PCR mixture contained the same components as in PCR reaction for the PIK3CA gene. The PCR was run as follows: after a 4-min denaturation at 94°C, the reaction was run 38 to 40 cycles, each comprising 30 s of denaturing at 94°C, 30 s of annealing (at 53°C for N2-Ras; 55°C for K1-Ras; 57°C for H2-Ras), and 1 min of extension at 72°C, with an extension at 72°C for 7 min as the last step.

For PTEN gene, most identified mutations were present in exons 5, 6, 7, and 8, especially exon 5, in human cancers (25, 26). Therefore, we selected these exons of PTEN gene for mutation analysis in this study. Exons 5 to 8 of PTEN gene were amplified using the primers listed in Table 1. The amplification PCR setting was the same as that for the H2-Ras gene.

Table 1.

Paired primers used to amplify PTEN genes

ExonsPrimer sequences (5′→3′)
F-CTTATTCTGAGGTTATCTTTTTACC 
 R-CTCAGAATCCAGGAAGAGGA 
F-TTGGCTTCTCTTTTTTTTCTG 
 R-ACATGGAAGGATGAGAATTTC 
F-ACAGAATCCATATTTCGTGTA 
 R-TAATGTCTCACCAATGCCA 
F-ACACATCACATACATACAAGTC 
 R-GTGCAGATAATGACAAGGAATA 
ExonsPrimer sequences (5′→3′)
F-CTTATTCTGAGGTTATCTTTTTACC 
 R-CTCAGAATCCAGGAAGAGGA 
F-TTGGCTTCTCTTTTTTTTCTG 
 R-ACATGGAAGGATGAGAATTTC 
F-ACAGAATCCATATTTCGTGTA 
 R-TAATGTCTCACCAATGCCA 
F-ACACATCACATACATACAAGTC 
 R-GTGCAGATAATGACAAGGAATA 

Abbreviations: F, forward; R, reverse.

The BRAF T1799A tranversion mutation in exon 15 of the BRAF gene was analyzed by direct DNA sequencing. Briefly, a 212-bp fragment from exon 15 of the BRAF gene containing the site where T1799A transversion mutation occurs was amplified by PCR using the primers and reaction condition as described (27).

The efficiency and quality of PCR for all these genes were confirmed by running the PCR products on a 1.5% agarose gel. The PCR products were subsequently subjected to direct sequencing PCR with Big Dye terminator V 3.0 cycle sequencing reagents (Applied Biosystems) with the following cycles: 95°C for 30 s for one cycle, and 95°C for 15 s, 50°C for 15 s, and 60°C for 4 min for 35 cycles. For mutation identification, the samples were finally analyzed on an ABI PRISM 3700 DNA Analyzer (Applied Biosystems) at the Johns Hopkins Biosynthesis and Sequencing facility.

Immunohistostaining. This procedure was pursued to investigate expression of PIK3CA protein in relation to PIK3CA copy gain in the tumor. Briefly, 15-μm paraffin-embedded tissue sections were deparaffinized, soaked in alcohol, and incubated in 3% hydrogen peroxide for 15 min to inactivate endogenous peroxidase activity with microwave treatment in an antigen unmasking solution (Vector Lab, Burlingame, CA), following the manufacturer's instructions. The tissue sections were then incubated overnight at 4°C with antibodies against the PIK3CA protein. Staining was subsequently done with Vectastain Universal Quik kit (Vector Lab), following the manufacturer's instructions. Peroxidase activity was revealed using 3,3-diaminobenzidine. The immunohistostaining score for the intensity of staining was assigned to each case in a blind fashion (i.e., without knowing the PIK3CA copy number of the case).

Statistical analysis. Sample means were compared using unpaired t test, assuming unequal variances. Two-tailed Fisher's exact test was used for data containing fewer than five cases. Linear regression analysis was done for the relationship between PIK3CA copy number and immunohistostaining score. P values <0.05 were considered significant.

High additive prevalence of genetic alterations in the PI3K/Akt pathway in thyroid tumors. Although several genetic alterations in the PI3K/Akt pathway have been individually described previously in thyroid tumors, their relative occurrence and additive prevalence, which may reflect the overall role of the PI3K/Akt pathway at a genetic level in thyroid tumorigenesis, remain unknown. We therefore collectively analyzed the major genetic alterations in the PI3K/Akt pathway–related genes in a large cohort of thyroid tumors. With a gene copy number of four or more defined as copy gain, we found PIK3CA copy gain, as shown in Table 2, in 34 of 171 (20%) DTC [24 of 85 (28%) FTC and 10 of 86 (12%) PTC], 20 of 48 (42%) ATC, and 14 of 81 (17%) BTA. The highest prevalence of PIK3CA copy gain was seen in FTC and ATC either when comparison was made for a copy number of four or above (Table 2) or for the average copy number of the PIK3CA gene among different thyroid tumors (Fig. 1A). Among all the PI3K/Akt–related genetic alterations (see below), PIK3CA copy number gain was the most prevalent in all types of thyroid tumors. This PIK3CA copy gain most likely represents amplification of the gene itself or the 3q26.3 region carrying this gene on chromosome 3 but not aneuploidy of this chromosome. This is supported by the fact that using the same detection technique, we did not see copy gain of the RASSF1A gene, which is distant from the 3q26 region on chromosome 3, either in 58 cases harboring PIK3CA copy gain or in 22 cases harboring no PIK3CA copy gain (data not shown). To investigate the effect of PIK3CA copy gain on expression of PIK3CA protein, we randomly selected 28 FTC samples with various PIK3CA copies and did immunohistostaining for PIK3CA. As illustrated by the representative samples in Fig. 2A, increased staining of PIK3CA was seen with increased PIK3CA copies. Linear regression analysis on the 28 cases revealed a significant correlation between the PIK3CA immunohistostaining score and the PIK3CA copy number (Fig. 2B; R = 0.72, P = 0.01).

Table 2.

Summary of genetic alterations in the PI3K/Akt pathway

Tumor types (total cases)PIK3CA copy gainPIK3CA mutationsRas mutationsPTEN mutations
FTC (n = 86) 24 (28)* 5 (6) 15 (17) 6 (7) 
PTC (n = 86) 10 (12) 3 (3) 7 (8) 2 (2) 
DTC (n = 172) 34 (20) 8 (5)§ 22 (13) 8 (5) 
ATC (n = 50) 20 (42) 6 (12) 4 (8) 8 (16) 
BTA (n = 81) 14 (17) 5 (6) 7 (9) 0 (0) 
Tumor types (total cases)PIK3CA copy gainPIK3CA mutationsRas mutationsPTEN mutations
FTC (n = 86) 24 (28)* 5 (6) 15 (17) 6 (7) 
PTC (n = 86) 10 (12) 3 (3) 7 (8) 2 (2) 
DTC (n = 172) 34 (20) 8 (5)§ 22 (13) 8 (5) 
ATC (n = 50) 20 (42) 6 (12) 4 (8) 8 (16) 
BTA (n = 81) 14 (17) 5 (6) 7 (9) 0 (0) 

NOTE: Data are prevalence of occurrence, n (%).

*

One case failed on copy number analysis; therefore, the prevalence is 24 of 85 (28%).

One case failed on PIK3CA mutation analysis; therefore, the prevalence is 5 of 85 (6%).

One case failed on copy number analysis; therefore, the prevalence is 34 of 171 (20%).

§

One case failed on PIK3CA mutation analysis; therefore, the prevalence is 8 of 171 (5%).

Two cases failed on copy number analysis; therefore, the prevalence is 20 of 48 (42%).

Fig. 1.

Genetic alterations in the PI3K/Akt pathway in various thyroid tumors. A, copy numbers of the PIK3CA gene in different thyroid tumors. Columns, average; bars, SD. **, P = 0.01; ***, P < 0.001. B, additive prevalence of all the genetic alterations in the PI3K/Akt pathway, including PIK3A copy number gain (four or more) and mutations in PIK3CA, PTEN, and Ras genes in various thyroid tumors. Values at the top of each column, prevalence (%). TF, total frequency of PI3K/Akt pathway–related genetic alterations in thyroid tumors (when any of these genetic alterations occurred). CF, coexistence frequency of two or more PI3K/Akt pathway–related genetic alterations in thyroid tumors.

Fig. 1.

Genetic alterations in the PI3K/Akt pathway in various thyroid tumors. A, copy numbers of the PIK3CA gene in different thyroid tumors. Columns, average; bars, SD. **, P = 0.01; ***, P < 0.001. B, additive prevalence of all the genetic alterations in the PI3K/Akt pathway, including PIK3A copy number gain (four or more) and mutations in PIK3CA, PTEN, and Ras genes in various thyroid tumors. Values at the top of each column, prevalence (%). TF, total frequency of PI3K/Akt pathway–related genetic alterations in thyroid tumors (when any of these genetic alterations occurred). CF, coexistence frequency of two or more PI3K/Akt pathway–related genetic alterations in thyroid tumors.

Close modal
Fig. 2.

Immunohistostaining of PIK3CA: correlation of PIK3CA copy gain with increased PIK3CA protein expression. A, representative samples of immunohistostaining on thyroid histologic slides using anti-PIK3CA antibodies. Increasing extent of specific staining (brown color) in association with increasing PIK3CA copy number (numbers inside brackets). Case# a, normal thyroid tissue; case# b to d, follicular thyroid cancer. B, linear regression analysis of the relationship between immunohistostaining score and PIK3CA copy number on 28 randomly selected samples. A significant correlation of the two is revealed (R = 0.72, P = 0.01).

Fig. 2.

Immunohistostaining of PIK3CA: correlation of PIK3CA copy gain with increased PIK3CA protein expression. A, representative samples of immunohistostaining on thyroid histologic slides using anti-PIK3CA antibodies. Increasing extent of specific staining (brown color) in association with increasing PIK3CA copy number (numbers inside brackets). Case# a, normal thyroid tissue; case# b to d, follicular thyroid cancer. B, linear regression analysis of the relationship between immunohistostaining score and PIK3CA copy number on 28 randomly selected samples. A significant correlation of the two is revealed (R = 0.72, P = 0.01).

Close modal

As summarized in Table 2, we found PIK3CA mutations in 8 of 171 (5%) DTC [5 of 85 (6%) FTC and 3 of 86 (3%) PTC], 6 of 50 (12%) ATC, and 5 of 81 (6%) BTA. These results showed a relatively high prevalence of PIK3CA mutation in ATC, but low prevalence in other types of thyroid tumors, consistent with a recent report on this mutation in thyroid tumors (16). In all types of thyroid tumors, PIK3CA copy number gain is more common than PIK3CA mutations, suggesting that copy gain of PIK3CA gene is a more relevant genetic alteration than mutation of this gene in the aberrant activation of the PI3K/Akt pathway in thyroid tumors. We found Ras mutations in 22 of 172 (13%) DTC [15 of 86 (17%) FTC and 7 of 86 (8%) PTC], 4 of 50 (8%) ATC, and 7 of 81 (9%) BTA (Table 2). Similar to previous reports (21), in this large series of thyroid tumors, we found the most common mutation to be N2-Ras mutation in codon 61, and the highest prevalence of Ras mutation to be in FTC. When distribution pattern of Ras mutations in various PTC subtypes was analyzed, they were found in 3 of 55 (5%) conventional PTC, 0 of 8 (0%) tall-cell PTC, and 4 of 23 (17%) follicular variant PTC. We found PTEN mutations in 8 of 172 (5%) DTC [6 of 86 (7%) FTC and 2 of 86 (2%) PTC], 8 of 50 (16%) ATC, and 0 of 81 (0%) BTA (Table 2). In all the exons examined for PTEN gene, we observed only one deletion in one case. The low prevalence of genetic alterations of PTEN gene in the present study was consistent with previous reports on this gene in sporadic thyroid tumors (19, 20). As with PIK3CA mutation, PTEN mutation was relatively frequent in ATC. Occurrence of any of these genetic alterations in all the three genes was seen in 68 of 172 (39.5%) DTC [47 of 86 (55%) FTC and 21 of 86 (24%) PTC], 29 of 50 (58%) ATC, and 25 of 81 (31%) BTA (Fig. 1B). It is remarkable that over 50% of FTC and ATC harbored at least one of these PI3K/Akt pathway–related genetic alterations. Coexistence of two or more of these genetic alterations was seen in 4 of 172 (2%) DTC [3 of 86 (3.5%) FTC and 1 of 86 (1%) PTC], 8 of 50 (16%) ATC, and 1 of 81 (1%) BTA (Fig. 1B), being again most common in FTC and ATC, particularly in ATC. The prevalence of the genetic alterations reported here likely represents an underestimate as we were able to only analyze exons that were known to harbor most frequently mutations in each of these genes, but not the entire coding regions of the genes. These results therefore provide strong genetic implications that the PI3K/Akt pathway plays an important role in a significant portion of thyroid tumors and, for FTC and ATC, in the majority of the cases.

Mutual exclusivity among the genetic alterations in PIK3CA, Ras, and PTEN genes in differentiated thyroid tumors. To explore the respective role of the PI3K/Akt pathway–related genetic alterations in thyroid tumorigenesis, we examined their relationship in various types of thyroid tumors. As summarized in Table 3, we found no or rare PIK3CA mutations in differentiated thyroid tumors that gained PIK3CA copy, that is, in 0 of 34 (0%) DTC [0 of 24 (0%) FTC and 0 of 10 (0%) PTC] and 1 of 14 (7%) BTA. More frequent PIK3CA mutations were found in differentiated thyroid tumors that did not harbor PIK3CA copy gain, that is, in 8 of 137 (6%) DTC [5 of 61 (8%) FTC and 3 of 76 (4%) PTC] and 4 of 67 (6%) BTA. In contrast, the PIK3CA mutations occurred in 3 of 20 (15%) ATC that harbored PIK3CA copy gain and similarly in 3 of 28 (11%) ATC that did not (Table 3). We found Ras mutations in only 1 of 34 (3%) DTC [1 of 24 (4%) FTC and 0 of 10 (0%) PTC], 1 of 20 (5%) ATC, and 0 of 14 (0%) BTA that harbored PIK3CA copy gain. In contrast, a much higher prevalence of Ras mutations was found in tumors that did not harbor PIK3CA copy gain, including 21 of 137 (15%) DTC [14 of 61 (23%) FTC and 7 of 76 (9%) PTC], 3 of 28 (11%) ATC, and 7 of 67 (10%) BTA. We found PTEN mutations in 1 of 34 (3%) DTC [0 of 24 (0%) FTC and 1 of 10 (10%) PTC], 3 of 20 (15%) ATC, and 0 of 14 (0%) BTA that harbored PIK3CA copy number gain, and found PTEN mutations in 7 of 137 (5%) DTC [6 of 61 (10%) FTC and 1 of 76 (1%) PTC], 5 of 28 (18%) ATC, and 0 of 67 (0%) BTA that did not harbor PIK3CA copy gain (Table 3). The mutual exclusivity between PIK3CA copy gain and any of the PIK3CA, Ras, and PTEN mutations was clearly noticeable, particularly with Ras mutation, in FTC, PTC, and BTA, but not statistically significant when analyzed individually, probably due to the small number of each of these mutations (Table 3). However, a striking mutual exclusivity between PIK3CA copy gain and the total mutations of the three genes was seen in FTC (P = 0.001) and DTC (P = 0.01), but not in ATC, although the latter harbored these PI3K/Akt pathway–associated genetic alterations all with the highest prevalence (Table 3). A trend of this mutual exclusivity was also seen in BTA but did not reach statistical significance due to the relatively low frequency of these genetic alterations in this tumor (Table 3). The mutual exclusivity between PIK3CA copy gain and mutations in PI3K/Akt pathway–related genes is more clearly illustrated in Fig. 3, which shows that mutations in PIK3CA, Ras, and PTEN genes clustered mostly in the cases with low number of PIK3CA copies in all types of thyroid tumors except for ATC.

Table 3.

Relationship between the PIK3CA gene copy gain and the mutations in PI3K/Akt pathway–related genes in thyroid tumors

Tumor typesMutations (genes)PIK3CA copy gain (+)*PIK3CA copy gain (−)*P
FTC PIK3CA 0/24 (0) 5/61 (8) 0.315 
 Ras 1/24 (4) 14/61 (23) 0.057 
 PTEN 0/24 (0) 6/61 (10) 0.178 
 Total 1/24 (4) 25/61 (41) 0.001 
PTC PIK3CA 0/10 (0) 3/76 (4) 1.000 
 Ras 0/10 (0) 7/76 (9) 1.000 
 PTEN 1/10 (10) 1/76 (1) 0.220 
 Total 1/10 (10) 11/76 (14) 1.000 
DTC PIK3CA 0/34 (0) 8/137 (6) 0.359 
 Ras 1/34 (3) 21/137 (15) 0.082 
 PTEN 1/34 (3) 7/137 (5) 1.000 
 Total 2/34 (6) 36/137 (26) 0.010 
ATC PIK3CA 3/20 (15) 3/28 (11) 0.683 
 Ras 1/20 (5) 3/28 (11) 0.631 
 PTEN 3/20 (15) 5/28 (18) 1.000 
 Total 7/20 (35) 11/28 (39) 1.000 
 BRAF 5/20 (25) 9/28 (32)  
BTA PIK3CA 1/14 (7) 4/67 (6) 1.000 
 Ras 0/14 (0) 7/67 (10) 0.345 
 PTEN 0/14 (0) 0/67 (0) 1.000 
 Total 1/14 (7) 11/67 (16) 0.681 
Tumor typesMutations (genes)PIK3CA copy gain (+)*PIK3CA copy gain (−)*P
FTC PIK3CA 0/24 (0) 5/61 (8) 0.315 
 Ras 1/24 (4) 14/61 (23) 0.057 
 PTEN 0/24 (0) 6/61 (10) 0.178 
 Total 1/24 (4) 25/61 (41) 0.001 
PTC PIK3CA 0/10 (0) 3/76 (4) 1.000 
 Ras 0/10 (0) 7/76 (9) 1.000 
 PTEN 1/10 (10) 1/76 (1) 0.220 
 Total 1/10 (10) 11/76 (14) 1.000 
DTC PIK3CA 0/34 (0) 8/137 (6) 0.359 
 Ras 1/34 (3) 21/137 (15) 0.082 
 PTEN 1/34 (3) 7/137 (5) 1.000 
 Total 2/34 (6) 36/137 (26) 0.010 
ATC PIK3CA 3/20 (15) 3/28 (11) 0.683 
 Ras 1/20 (5) 3/28 (11) 0.631 
 PTEN 3/20 (15) 5/28 (18) 1.000 
 Total 7/20 (35) 11/28 (39) 1.000 
 BRAF 5/20 (25) 9/28 (32)  
BTA PIK3CA 1/14 (7) 4/67 (6) 1.000 
 Ras 0/14 (0) 7/67 (10) 0.345 
 PTEN 0/14 (0) 0/67 (0) 1.000 
 Total 1/14 (7) 11/67 (16) 0.681 

NOTE: The number of cases with the indicated gene mutation/the number of cases with the indicated status of PIK3CA copy gain (%) is shown.

*

“+” and “−” indicate presence and absence of PIK3CA copy gain, respectively.

Per two-tailed Fisher's exact test.

Fig. 3.

Distribution of various genetic alterations associated with the PI3K/Akt pathway in individual cases of thyroid tumors. Y axis, copy number of the PIK3CA gene corresponding to each individual case of thyroid tumor (circle). Additional genetic events (i.e., mutations) are denoted with different colors: red, green, and blue, Ras, PTEN, and PIK3CA mutations, respectively. Blank circles, lack of mutations; dual-color circles, coexistence of two types of mutations denoted by the corresponding colors.

Fig. 3.

Distribution of various genetic alterations associated with the PI3K/Akt pathway in individual cases of thyroid tumors. Y axis, copy number of the PIK3CA gene corresponding to each individual case of thyroid tumor (circle). Additional genetic events (i.e., mutations) are denoted with different colors: red, green, and blue, Ras, PTEN, and PIK3CA mutations, respectively. Blank circles, lack of mutations; dual-color circles, coexistence of two types of mutations denoted by the corresponding colors.

Close modal

Table 4 summarizes all the individual cases of differentiated thyroid tumors that were positive for any of the genetic alterations in the PI3K/Akt pathway. A clear pattern of mutual exclusivity also existed among the mutations of PIK3CA, Ras, and PTEN genes in differentiated thyroid tumors. Overlap of two or more genetic alterations in the PI3K/Akt pathway was seen only in 5 of 93 (5%) cases that were positive for any of these genetic alterations. We also examined the BRAF mutation status in the 19 PTC that harbored PI3K/Akt pathway–related genetic alterations and in randomly selected 59 PTC that did not harbor these genetic alterations. We found BRAF mutation in 7 of 19 (37%) cases in the former group versus 22 of 59 (37%) in the latter group, with no significant difference (P = 0.969), suggesting that there is no mutual exclusivity between BRAF mutation and the PI3K/Akt pathway–related genetic alterations in PTC, consistent with the fact that they belong to unrelated signaling pathways. This result also supports the idea that the mutual exclusivity among the PI3K/Akt pathway–related genetic alterations seen in differentiated thyroid tumors was a specific phenomenon.

Table 4.

Summary of individual cases of differentiated thyroid tumors with genetic alterations in the PI3K/Akt pathway: mutual exclusivity of the genetic alterations

NOTE: Gray regions, coexistence of two or more genetic alterations; bold letters, nucleotide substitution in the mutated codon.

Abbreviation: NI, no information.

Overlap of genetic alterations in PIK3CA, Ras, PTEN, and BRAF genes in ATC. As for the overlap of PIK3CA copy gain with the PI3K/Akt pathway–related gene mutations in ATC (Tables 3 and 5), overlap among gene mutations themselves in the PI3K/Akt pathway was common in ATC (Table 5). As described above for PTC, overlap between BRAF mutation and the PI3K/Akt pathway–related genetic alterations was also common in ATC (Table 5). BRAF mutation was found in 14 of 49 (29%) ATC, and 6 of 14 (43%) ATC that harbored BRAF mutation also harbored PI3K/Akt pathway–related genetic alterations. When all these genetic alterations were pooled, coexistence of two or more of them was seen in 12 of 37 (32%) ATC that were positive for any of these genetic alterations. There was a clear trend of association of coexistence of these genetic alterations with increased aggressiveness of thyroid tumors from BTA to FTC to ATC (Fig. 1B).

Table 5.

Summary of individual cases of ATC with genetic alterations in the PI3K/Akt pathway and BRAF gene: overlap of the genetic alterations

NOTE: Gray regions, coexistence of two or more genetic alterations; bold letter, nucleotide substitution in the mutated codon.

In this study, we showed a high additive prevalence of genetic alterations in the PIK3/Akt pathway and a mutual exclusivity among them in various thyroid tumors. These data strongly suggest that the PI3K/Akt pathway, through these genetic alterations, plays an extensive role in thyroid tumorigenesis and that each of these genetic alterations is independently a sufficient oncogenic event in thyroid tumorigenesis through this pathway. Of particular interest is the PIK3CA copy gain. Its high prevalence in the present study, particularly in FTC and ATC, is consistent with our previous findings in a smaller series of thyroid tumors (17, 18). This, together with its mutual exclusivity with other PI3K/Akt pathway–related genetic alterations in differentiated thyroid tumors and its associated increase in PIK3CA expression, strongly suggests that PIK3CA copy gain is a significant oncogenic genetic event in thyroid tumorigenesis. Thus, it seems that this genetic alteration is as important in tumorigenesis of thyroid cancer as in that of various other human cancers (2831).

We also revealed a mutually exclusive relationship of Ras mutations with other genetic alterations in the PI3K/Akt pathway in differentiated thyroid tumors, hence providing first strong genetic evidence supporting an independent role of Ras mutations in thyroid tumorigenesis through the PI3K/Akt pathway. Previous studies on PTEN mutations in sporadic thyroid tumors were mostly conducted in DTC and BTA and showed a low prevalence (19, 20). In the present study, we analyzed PTEN mutations also in a large number of ATC and observed a relatively high prevalence of mutations in this tumor-suppressor gene in ATC. We similarly found a relatively high prevalence of PIK3CA mutations in ATC, confirming a previous report (16). These mutations were all each mutually exclusive from other PI3K/Akt pathway–related genetic alterations in differentiated thyroid tumors, suggesting that each of them, in tandem in the PI3K/Akt pathway, is individually sufficient to play a significant oncogenic role in thyroid tumorigenesis.

Although the additive prevalence of the PI3K/Akt pathway–related genetic alterations in thyroid tumors revealed in the present study was already high, it likely represents an underestimate. As not all coding regions of the genes were examined, one can expect even a larger percentage of thyroid tumors to harbor genetic alterations in the PI3K/Akt pathway. All the Ras mutations identified in the present study have been reported previously. Some of the PIK3CA and PTEN mutations in the present study are new. Although their function has not been characterized, they are most likely functionally relevant as they are close to the sites of the previously reported mutations in the related genes. This notion is consistent with their mutual exclusivity from other genetic alterations in the PI3K/Akt pathway. The occurrence of some of these genetic alterations, such as PIK3CA copy gain and Ras mutations, even in BTA suggests that the PI3K/Akt pathway plays a role at an early stage of thyroid tumorigenesis. The additive prevalence and coexistence of these genetic alterations were most remarkable in ATC, second in FTC, and least in PTC and BTA (Fig. 1B). This is consistent with the increased activities of the PI3K/Akt pathway most often seen in aggressive thyroid cancers, such as ATC (14, 16), particularly in the invasive areas (14). It is also consistent with the proangiogenic role of the PI3K/Akt pathway in tumorigenesis (2, 3) and the well-known aggressiveness order of ATC ≫ FTC > PTC ≫ … > BTA.

BRAF mutation is seen in nearly half of the cases of PTC and is completely excluded from FTC (32). Transgenic mouse model showed mutant BRAF-induced development of PTC and its progression to ATC (33). Large clinical studies showed an association of this mutation with poorer clinicopathologic outcomes of PTC (32). Therefore, BRAF mutation, through the MAP kinase pathway, plays a fundamental role in the tumorigenesis of PTC. The high overall prevalence of the genetic alterations in the PI3K/Akt pathway in FTC supports a fundamental role of the PI3K/Akt pathway in the tumorigenesis of FTC. Therefore, PTC and FTC are driven, respectively, by these two independent signaling pathways, which likely accounts for the different clinicopathologic characteristics of the two types of DTC and provide different therapeutic targets for these cancers. As shown previously (32) and in the present study, BRAF mutation is a relatively common event in ATC. The overlap of BRAF mutation and PI3K/Akt pathway–related genetic alterations was common in ATC (Table 5), suggesting that both the MAP kinase pathway and the PI3K/Akt pathway play an important role in the tumorigenesis and aggressiveness of a significant portion of ATC tumors and both pathways should be targeted for effective treatment of this cancer.

In summary, we investigated collectively the major genetic alterations in the PI3K/Akt pathway and their relationship in a large series of thyroid tumors. We found a high additive prevalence of these genetic alterations in thyroid tumors, particularly in FTC and ATC, and showed their mutual exclusivity in differentiated thyroid tumors and overlap (with BRAF mutation as well) in undifferentiated ATC. Our data are consistent with a model (Fig. 4) in which oncogenic activation of the PI3K/Akt pathway drives transformation of BTA to FTC and then to ATC. Moreover, PTC develops de novo after activation of the MAP kinase pathway by oncogenic genetic alterations, such as BRAF mutation, and its transformation to ATC may be facilitated by PI3K/Akt pathway. Our demonstration of extensive involvement of the PI3K/Akt pathway in thyroid tumorigenesis provides strong implication that this pathway is a major therapeutic target in thyroid cancers, particularly in FTC and ATC.

Fig. 4.

A simplified model for the role of PI3K/Akt pathway in thyroid tumorigenesis and progression. The model illustrates the two major signaling pathways, the PI3K/Akt and the MAP kinase pathways, which, respectively, play a major role in the tumorigenesis of FTC and PTC. ↑ to ↑↑ to ↑↑↑, the increasing role of the PI3K/Akt pathway. The increasing color intensity of the circle represents the increasing progression and aggressiveness of the tumor. MAPK pathway, the Ras→Raf→MEK→MAP/ERK pathway.

Fig. 4.

A simplified model for the role of PI3K/Akt pathway in thyroid tumorigenesis and progression. The model illustrates the two major signaling pathways, the PI3K/Akt and the MAP kinase pathways, which, respectively, play a major role in the tumorigenesis of FTC and PTC. ↑ to ↑↑ to ↑↑↑, the increasing role of the PI3K/Akt pathway. The increasing color intensity of the circle represents the increasing progression and aggressiveness of the tumor. MAPK pathway, the Ras→Raf→MEK→MAP/ERK pathway.

Close modal

Grant support: NIH RO-1 grant CA113507-01 and American Cancer Society grant RSG-05-199-01-CCE (M. Xing).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Dr. Elizabeth Mambo for technical advice.

1
Leevers SJ, Vanhaesebroeck B, Waterfield MD. Signalling through phosphoinositide 3-kinases: the lipids take centre stage.
Curr Opin Cell Biol
1999
;
11
:
219
–25.
2
Vivanco I, Sawyers CL. The phosphatidylinositol 3-kinase AKT pathway in human cancer.
Nat Rev Cancer
2002
;
2
:
489
–501.
3
Fresno Vara JA, Casado E, de Castro J, et al. PI3K/Akt signalling pathway and cancer.
Cancer Treat Rev
2004
;
30
:
193
–204.
4
Shayesteh L, Lu Y, Kuo WL, et al. PIK3CA is implicated as an oncogene in ovarian cancer.
Nat Genet
1999
;
21
:
99
–102.
5
Samuels Y, Ericson K. Oncogenic PI3K and its role in cancer.
Curr Opin Oncol
2006
;
18
:
77
–82.
6
Karakas B, Bachman KE, Park BH. Mutation of the PIK3CA oncogene in human cancers.
Br J Cancer
2006
;
94
:
455
–9.
7
Hennessy BT, Smith DL, Ram PT, et al. Exploiting the PI3K/AKT pathway for cancer drug discovery.
Nat Rev Drug Discov
2005
;
4
:
988
–1004.
8
Cantley LC, Neel BG. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway.
Proc Natl Acad Sci U S A
1999
;
96
:
4240
–5.
9
Weng LP, Smith WM, Dahia PL, et al. PTEN suppresses breast cancer cell growth by phosphatase activity-dependent G1 arrest followed by cell death.
Cancer Res
1999
;
59
:
5808
–14.
10
Lu Y, Lin YZ, LaPushin R, et al. The PTEN/MMAC1/TEP tumor suppressor gene decreases cell growth and induces apoptosis and anoikis in breast cancer cells.
Oncogene
1999
;
18
:
7034
–45.
11
Sansal I, Sellers WR. The biology and clinical relevance of the PTEN tumor suppressor pathway.
J Clin Oncol
2004
;
22
:
2954
–63.
12
Ringel MD, Hayre N, Saito J, et al. Overexpression and overactivation of Akt in thyroid carcinoma.
Cancer Res
2001
;
61
:
6105
–11.
13
Miyakawa M, Tsushima T, Murakami H, et al. Increased expression of phosphorylated p70S6 kinase and Akt in papillary thyroid cancer tissues.
Endocr J
2003
;
50
:
77
–83.
14
Vasko V, Saji M, Hardy E, et al. Akt activation and localisation correlate with tumour invasion and oncogene expression in thyroid cancer.
J Med Genet
2004
;
41
:
161
–70.
15
Kada F, Saji M, Ringel MD. Akt: a potential target for thyroid cancer therapy.
Curr Drug Targets Immune Endocr Metabol Disord
2004
;
4
:
181
–5.
16
Garcia-Rostan G, Costa AM, Pereira-Castro I, et al. Mutation of the PIK3CA gene in anaplastic thyroid cancer.
Cancer Res
2005
;
65
:
10199
–207.
17
Wu G, Mambo E, Guo Z, et al. Uncommon mutation, but common amplifications, of the PIK3CA gene in thyroid tumors.
J Clin Endocrinol Metab
2005
;
90
:
4688
–93.
18
Liu D, Mambo E, Ladenson PW, et al. Letter re: uncommon mutation but common amplifications of the PIK3CA gene in thyroid tumors.
J Clin Endocrinol Metab
2005
;
90
:
5509
.
19
Dahia PL, Marsh DJ, Zheng Z, et al. Somatic deletions and mutations in the Cowden disease gene, PTEN, in sporadic thyroid tumors.
Cancer Res
1997
;
57
:
4710
–3.
20
Halachmi N, Halachmi S, Evron E, et al. Somatic mutations of the PTEN tumor suppressor gene in sporadic follicular thyroid tumors.
Genes Chromosomes Cancer
1998
;
23
:
239
–43.
21
Vasko V, Ferrand M, Di Cristofaro J, et al. Specific pattern of RAS oncogene mutations in follicular thyroid tumors.
J Clin Endocrinol Metab
2003
;
88
:
2745
–52.
22
Mambo E, Gao X, Cohen Y, et al. Electrophile and oxidant damage of mitochondrial DNA leading to rapid evolution of homoplasmic mutations.
Proc Natl Acad Sci U S A
2003
;
100
:
1838
–43.
23
Samuels Y, Wang Z, Bardelli A, et al. High frequency of mutations of the PIK3CA gene in human cancers.
Science
2004
;
304
:
554
.
24
Campbell IG, Russell SE, Choong DY, et al. Mutation of the PIK3CA gene in ovarian and breast cancer.
Cancer Res
2004
;
64
:
7678
–81.
25
Minaguchi T, Yoshikawa H, Oda K, et al. PTEN mutation located only outside exons 5, 6, and 7 is an independent predictor of favorable survival in endometrial carcinomas.
Clin Cancer Res
2001
;
7
:
2636
–42.
26
Konopka B, Paszko Z, Janiec-Jankowska A, et al. Assessment of the quality and frequency of mutations occurrence in PTEN gene in endometrial carcinomas and hyperplasias.
Cancer Lett
2002
;
178
:
43
–51.
27
Xing M, Westra WH, Tufano RP, et al. BRAF mutation predicts a poorer clinical prognosis for papillary thyroid cancer.
J Clin Endocrinol Metab
2005
;
90
:
6373
–9.
28
Ma YY, Wei SJ, Lin YC, et al. PIK3CA as an oncogene in cervical cancer.
Oncogene
2000
;
19
:
2739
–44.
29
Singh B, Reddy PG, Goberdhan A, et al. p53 regulates cell survival by inhibiting PIK3CA in squamous cell carcinomas.
Genes Dev
2002
;
16
:
984
–93.
30
Byun DS, Cho K, Ryu BK, et al. Frequent monoallelic deletion of PTEN and its reciprocal association with PIK3CA amplification in gastric carcinoma.
Int J Cancer
2003
;
104
:
318
–27.
31
Wu G, Xing M, Mambo E, et al. Somatic mutation and gain of copy number of PIK3CA in human breast cancer.
Breast Cancer Res
2005
;
7
:
R609
–16.
32
Xing M. BRAF mutation in thyroid cancer.
Endocr Relat Cancer
2005
;
12
:
245
–62.
33
Knauf JA, Ma X, Smith EP, et al. Targeted expression of BRAFV600E in thyroid cells of transgenic mice results in papillary thyroid cancers that undergo dedifferentiation.
Cancer Res
2005
;
65
:
4238
–45.