BRAF^V600E Mutation and Papillary Thyroid Cancer: Chicken or Egg?

The BRAF^V600E mutation, resulting from a T1799A transversion nucleotide change in the BRAF gene, was discovered in papillary thyroid cancer (PTC) initially in 2003 (1). Many other studies quickly showed that, among different thyroid cancers, this was a PTC-specific genetic event, occurring in about 45% of cases on average (2); the only other thyroid cancer in which this mutation occurs sometimes is the presumably PTC-derived type of anaplastic thyroid cancer (3, 4). Although research in this area has seen explosive progress since then, the issue of whether BRAF^V600E mutation is a primary or secondary genetic event in PTC remains debatable. The recent study on the clonal status of BRAF^V600E in PTC by Guerra et al. (5) from Professor Mario Vitale's group and their subsequent sister study (6) published in this issue of the JCEM have shone new light on this issue.

The oncogenic function and clinicopathological role of the BRAF^V600E mutation in PTC have been extensively investigated. Data today overwhelmingly support that, as the most common oncogene known in PTC, BRAF^V600E, through overactivation of the MAPK pathway, plays a fundamental role in the tumorigenesis and progression of PTC harboring it (7–10). Such a prominent PTC-specific role of BRAF^V600E mutation may favor the idea of its being a primary initiating genetic event for the tumorigenesis of PTC. However, a puzzling issue that remains to be explained is why PTC harbors the BRAF^V600E mutation in some cases but not in others of identical histological type. If the MAPK pathway plays a primary and fundamental role in the tumorigenesis of PTC, as is well supported by numerous data (7–10), this phenomenon can be explained by the possibility that alternative activating genetic alterations that hit a component close to RAF in the MAPK pathway signaling also exist. Driven by this idea, a search for mutations in such components as MEK1 and ERK2 genes in PTC, which sometimes occur in other cancers, was pursued but proved to be fruitless (11). RET/PTC and RAS mutations may instead account for a portion of PTC because they are classical activators of the MAPK pathway. In fact, some early studies reported mutual exclusivity among BRAF^V600E, RET/PTC, and RAS mutations in well-differentiated PTC (12, 13), suggesting that these genetic alterations may all be primary genetic events that each may independently initiate PTC tumorigenesis when occurring naturally in thyroid follicular cells. However, other studies showed coexistence of these genetic alterations in thyroid cancer, particularly in aggressive cases (14–17), suggesting that they may occur as secondary genetic alterations that drive tumorigenesis while they accumulate during the progression of thyroid cancer.

The concept that BRAF^V600Ecould be a second genetic event in the tumorigenesis of PTC was first formally entertained in the demonstration that BRAF^V600E mutation was detected in lymph node metastases from primary PTC tumors that did not harbor this mutation, prompting Vasko et al. (18) to propose that this mutation could occur de novo as a secondary genetic event. On the other hand, however, strong oncogenicity and tumorigenicity of BRAF^V600E was well demonstrated in cell line and xenograft tumor systems (19–21) and in transgenic mouse models (22–24). In the latter, development and aggressive progression of PTC could be initiated by the expression of BRAF^V600E alone in mouse thyroid glands. These data demonstrate that as a single genetic event, BRAF^V600E is capable of initiating oncogenesis and development of PTC. Taking all these studies together, it appears that BRAF^V600E and PTC may be capable of giving birth to each other. This creates a chicken–egg relationship for BRAF^V600E and PTC in their occurrence, leaving as a puzzle the issue of which is the first to occur naturally in the human thyroid gland.

The two studies by Guerra et al. (5, 6) have brought this chicken–egg puzzle closer to resolution. As opposed to applying methods used in previous studies that mainly test the presence of BRAF^V600E in the entire tumor, the studies by Guerra et al. (5, 6) used pyrosequencing PCR to quantitatively analyze the allelic percentage of BRAF^V600E in PTC tumors. With this unique approach, the authors were able to demonstrate in the first study (5) that most of BRAF^V600E-positive PTC tumors harbored this mutation in less than 50% of alleles; in fact, it was in less than 25% of alleles in most cases, suggesting that only less than 50% of PTC cells of the tumor harbored this mutation, whereas the remaining PTC cells harbored the wild-type BRAF. Several approaches were taken to minimize normal cell contamination and ensure the purity of PTC cells, including selection of tumor samples with minimal lymphoreticular infiltrating cells, laser-captured harvest of PTC cells, and subcloning confirmation of the BRAF^V600E/wild-type BRAF ratios. By selecting different regions of pure PTC tumors for BRAF^V600E analysis, the authors also demonstrated a clear heterogeneous distribution of BRAF^V600E in the tumor, very similar to the recently reported intratumor genetic heterogeneity, including that of mTOR mutation, in renal carcinoma (25). Thus, by showing that BRAF^V600E mostly occurs only in a subpopulation of PTC cells, these results are consistent with the notion that this genetic alteration occurs as a late secondary genetic event in PTC. Interestingly, subclonality of BRAF^V600E was previously similarly demonstrated in melanoma tumors (26), suggesting that being a secondary genetic event is likely a common mechanism for the occurrence of BRAF^V600E in human cancers. This is reminiscent of the case of RET/PTC, which can also be a heterogeneous subclonal event in PTC (27, 28). Use of pyrosequencing analysis in the Guerra et al. (5) study also revealed coexistence of BRAF^V600E and RAS mutations in the same tumor in some cases. Although it is yet to be proven, this may represent different thyroid cancer cell clones for the two mutations, as in the case of melanoma in which the two mutations may occur in different cells of the same tumor (26). Thus, it appears that the classical somatic genetic alterations in PTC, as in melanoma, can all occur as secondary genetic events. It is unlikely that PTC originates as a genetically polyclonal tumor; it is more likely that PTC originates as a monoclonal tumor that can subsequently progress to a polyclonal tumor as secondary genetic alterations occur and accumulate.

In the second study of Guerra et al. (6), the authors took a further step to demonstrate a strong association of the higher allelic percentage of BRAF^V600E with several high-risk clinicopathological characteristics of PTC, including disease recurrence, with an odds ratio of 5.31 (P = 0.002) for PTC recurrence when comparing BRAF^V600E alleles of at least 30% with BRAF^V600E of less than 30%. This is a reasonable result in that more cells harboring BRAF^V600E in PTC would be expected to have a better chance to promote the overall aggressiveness of the tumor, given the well-recognized role of BRAF^V600E in promoting the aggressiveness of PTC (7–10). This result shows the biological relevance of subclonality of BRAF^V600E, hence further supporting its existence. It should be pointed out that although these data are consistent with the notion that BRAF^V600E is mostly a secondary genetic event in PTC, the possibility that it may also be a primary genetic event that initiates these PTC tumors cannot be completely ruled out. This is suggested by the fact that in the studies of Guerra et al. (5, 6), some PTC tumors harbored BRAF^V600E in nearly 50% of alleles, consistent with these tumors harboring heterozygous BRAF^V600E in all the cells—a result that would be expected for a monoclonal tumor initiated by BRAF^V600E. There is also a theoretical possibility that BRAF^V600E may occur as a primary genetic event that initiates the tumorigenesis of PTC in which late powerful secondary genetic events occur and take over as drivers of PTC progression, whereas the original BRAF^V600E may be removed by the DNA repair machinery of the cell. As a consequence, many PTC cells may not harbor the BRAF^V600E mutation, although the tumor may have originated from monoclonal BRAF^V600E in the beginning. This would result in the subclonal genetic pattern for BRAF^V600E found in the Guerra et al. (5, 6) studies. Thus, in this sense, even with these quantitative studies on BRAF^V600E alleles, the chicken–egg puzzle in the relationship between BRAF^V600E and PTC seems to remain. Assuming that BRAF^V600E is mostly a secondary genetic event in PTC, it will then remain to be a challenging and yet important task to identify the real primary initiating genetic event in PTC.

Regardless of whether BRAF^V600E is the chicken or egg in the tumorigenesis of PTC, once present, it is a strong driver of aggressiveness of PTC as documented in many previous studies (7–10) and now further supported by the Guerra et al. study (6). While supporting the generally accepted negative prognostic role of BRAF^V600E in PTC, the Guerra et al. (6) study also suggests that it may be useful to quantitatively test the subclonality of BRAF^V600E in every case of PTC harboring this mutation to help more accurately measure the risk and prognosis of PTC. Use of BRAF^V600E inhibitors has become a prominent treatment for BRAF^V600E-harboring human cancers (29). This is well exemplified by the success of this treatment for melanoma (30–32). Preclinical studies also demonstrate that this is a highly promising treatment for thyroid cancer (33, 34), which is currently under clinical testing. However, resistance to BRAF^V600E inhibitors has recently emerged as a major therapeutic challenge in eradicating melanoma (35); this is likely to be encountered also with thyroid cancer. Several partially responsible molecular mechanisms have been revealed (35). The subclonal nature of BRAF^V600E may represent also an important mechanism involved in this drug resistance because tumor cells harboring the wild-type BRAF may be insensitive to BRAF^V600E inhibitors, giving the tumor the ability to regrow. Testing the subclonality of BRAF^V600E and its extent may help better predict the response of PTC to treatment with BRAF^V600E inhibitors. Moreover, given the possible secondary occurrence nature of BRAF^V600E, it may also be useful to test this mutation in recurrent or metastatic tumors to better guide risk assessment and appropriate treatment, even if the corresponding primary PTC tumor does not harbor the mutation. Complete resolution of the puzzle regarding the causal relationship between BRAF^V600E and PTC and which factors are seminal in disease development will hopefully result in improved treatment approaches. It is hoped that this is not comparable to the real chicken–egg puzzle, which thus far has been beyond the power of humankind to solve.