Profile of BRAFV600E, BRAFK601E, NRAS, HRAS, and KRAS Mutational Status, and Clinicopathological Characteristics of Papillary Thyroid Carcinoma in Indonesian National Referral Hospital

Introduction

In recent decades, there has been a dramatic increase in the incidence of thyroid cancer.^1,2 In terms of worldwide and Indonesian cancer incidence, it is currently ranked 7th and 12th, respectively.³ Papillary thyroid carcinoma (PTC) is a follicular cell-derived tumor attributed to 80–85% of thyroid cancer.¹ Based on its distinctive histopathological characteristics, striking variants including tall and columnar cells, oncocytic, solid/trabecular, and those frequently display extrathyroidal extension (ETE), lymph node metastasis (LNM), and distant organ metastasis, are particularly aggressive.⁴

The presence of gene mutation is associated with the behavior and prognosis of the disease. B-Rapidly accelerating fibrosarcoma (BRAFV600E) and RAT Sarcoma (RAS) mutations are well-published driver mutations in the development of PTC. BRAFV600E mutation is typically present in the classic and tall-cell variants, commonly related to a higher level of aggression.⁵ Patients with BRAF mutation are twice as likely to experience a relapse of their illness and possess greater mortality rates than those without the mutation.^6,7 The human RAS gene is divided into Kirsten RAT sarcoma (KRAS), neuroblastoma (NRAS), and Harvey (HRAS), as opposed to BRAFV600E mutation, are more prevalent in the follicular variant of PTC.⁵ Better prognosis and more indolent disease behavior have been associated with RAS mutation.^8,9

The prevalence of BRAFV600E and RAS mutations varied between Western and Asian countries, which was believed to be caused by geographic heterogeneity, race, and other risk factors. Americans and Europeans carried BRAFV600E mutation in around 35–60% of the patients.^10,11 Meanwhile, the prevalence of BRAFV600E mutations was quite high among Asian countries, though the numbers varied.¹² On the other hand, a relatively similar prevalence of RAS-positive PTC was found in Europe, the USA, and Asian countries.¹³

Hitherto, studies about PTC and its mutational status are still limited in Indonesia. Yet, most of the studies used immunohistochemistry modality and/or cytology specimens with limited samples.^14–18 Given the dearth of research describing the prevalence of BRAF and RAS mutations in PTC in Indonesia, this study aimed to assess the prevalence of BRAFV600E, BRAFK601E, and RAS (NRAS, HRAS, and KRAS) mutations and their association with the clinicopathological profiles of PTC that were related to the tumor behavior and prognosis in Indonesian population.

Subject and Methods

Subject Selection and Evaluation of Histologic Parameters

Every patient who underwent a complete thyroidectomy and was diagnosed with PTC at the Cipto Mangunkusumo Hospital-Faculty of Medicine Universitas Indonesia between 2019 and August 2021 was retrospectively collected. A total of 172 patients were included after eliminating patients with insufficient samples, inaccessible medical records, inappropriate hematoxylin and eosin (H&E) stained slides, and formalin-fixed paraffin-embedded (FFPE) tumor specimens. Medical records were used to acquire clinical information, such as age, gender, and clinical stage. Two certified pathologists from our institution blindly examined the pathological information microscopically, including the tumor size, histological variant, multifocality, nuclear score, LNM, ETE, and lymphovascular invasion (LVI). We excluded PTC with high-grade features (high mitosis index and necrosis) and all follicular variants were invasive. Interobserver agreement was analyzed by using Kappa analysis with nearly perfect agreement results.

BRAFV600E, BRAFK601E, HRAS, NRAS, and KRAS Mutational Analysis

DNA Isolation and Purification

Using the QIAamp DNA FFPE Tissue Kit (Qiagen, Valencia, CA) in the integrated Laboratory of FMUI-CHM, we extracted genomic material from 5-μm thick sections of FFPE tumor tissues. Following the manufacturer’s directions, we carried out the subsequent steps of melting the paraffin with xylene, lysing the tissue with Proteinase K, heating, DNA binding, and washing. After 8000 rpm centrifugation with a QIAamp MinElute Column, pure DNA products were obtained (Qiagen, Valencia, CA). Spectrophotometers called NanoDrop TM 2000/2000c were used to measure the final DNA product’s quantity (Thermo Fisher Scientific, Waltham, MA). Absorbance measurement was used to evaluate quality. An A260/A280 ratio of 1.8 to 2.0 indicates a high-quality DNA sample.

Polymerase Chain Reaction and DNA Sequencing

BRAF Exon 15 Mutation Analysis

KOD One Polymerase Chain Reaction (PCR) Master Mix was used to perform PCR (Toyobo KMM-201). As shown in Table 1, specific primers were employed to amplify exon 15 of the BRAF gene. In the Agilent Surecycler 8800 thermal cycler, PCRs were produced under the following conditions: (i) 94°C denaturation for 2 min, (ii) 40 cycles of 98°C denaturation for 15 sec, 53°C and 55°C annealing for 5 sec, and 68°C elongation for 1 sec, (iii) 68°C elongation for 10 sec, and then (iv) 4°C hold. To verify the quality of the PCR results, the presence of a band using 1% TBE agarose electrophoresis was assessed. DNA sequencing was done by using BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Waltham, MA) and ABI PRISM 3730xl Genetic Analyzer (Applied Biosystems, Carlsbad, CA). By comparing sequences with the Basic Local Alignment Search Tool (BLAST) and manual reading confirmation, tumors harboring the BRAFV600E and BRAFK601E mutations were verified.

Table 1 Primer Pairs for Mutational Analysis

RAS Mutation Analysis

We used the MyTaq HS Red Mix Kit to perform PCR (Bioline). Exon 2 and exon 3 of the HRAS, NRAS, and KRAS genes were amplified using certain primers, as shown in Table 1. The following settings were used to achieve PCRs in the Veriti 96-Well Fast Thermal Cycler from Applied Biosystem in Carlsbad, California: (i) 95°C PCR initial activation step for 5 min (ii) 40 cycles of 95°C denaturation for 15 sec, 64°C annealing for 30 sec for exon 2 and 3 HRAS, 55°C annealing for 30 sec for exon 2 NRAS, 60°C annealing for 30 sec for exon 3 NRAS, 54°C annealing for 30 sec for exon 2 KRAS and 55°C annealing for 30 sec for exon 3 KRAS, and 72°C extension for 30 sec, (iii) 72°C final extension for 3 min, and (iv) 4°C hold before taken out of the machine. The presence of a band was assessed by using 1% TBE agarose electrophoresis to verify the quality of PCR results. DNA sequencing was done by using BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Waltham, MA) and ABI PRISM 3730xl Genetic Analyzer (Applied Biosystems, Carlsbad, CA). By comparing sequences with the Basic Local Alignment Search Tool (BLAST) and manual confirmation, tumors harboring the NRAS, HRAS, and KRAS mutation were verified.

Statistical Analysis

All of the research data were processed using Statistical Program for Social Science (SPSS) ver.20. BRAF and RAS mutational status, patient gender, clinical stage, histological variation, and other categorical data were given as frequencies and percentages. Age and tumor size were presented as median values based on the distribution abnormality of the numerical data. The bivariate analysis employed the Chi-square test or Fisher’s exact test to look at the relationship between variables in categorical data. If the p-value for each test was less than 0.05, the analysis was regarded as significant.

Result

We summarized the patients’ baseline characteristics in Table 2. This study mostly consisted of patients <55 years old, female, low clinical stage, tumor size <4 cm, multifocal, unencapsulated, without LVI, and without ETE. Distant metastasis was found in the liver, lung, vertebrae, costae, and other bones. The most common histologic variant was the classic variant followed by follicular, tall-cell, solid, and oncocytic variants (Figure 1).

Table 2 Baseline Characteristics of the Patients

Figure 1 Histopathology variants of papillary thyroid carcinoma (PTC) using H&E staining. (A) Classic variant of PTC showed marked papillary structures with fibrovascular core. (B) Follicular variant of PTC showed tumor with predominantly follicular pattern lined by typical PTC nuclei. (C) Tall-cell variant of PTC is defined by more than 30% of the tumor consisted of tall-cell morphology. (D) Solid/trabecular variant of PTC with predominantly solid pattern.

We examined exon 15 of the BRAF gene and exons 2 and 3 of all NRAS, HRAS, and KRAS genes (Table 3). This study revealed that 65 patients were BRAFV600E mutants and 37 patients were RAS mutants. Almost all BRAFV600E mutations were heterozygous and only one homozygous mutant was found. Only one patient harbored BRAFK601E mutation. BRAF and RAS mutations were mutually exclusive. Our result showed that substitution of glutamine-to-arginine at exon 3 NRAS mutation (CAA (Gln) > CGA (Arg); Q61R) was the most common type of mutation, followed by glutamine-to-arginine substitution at exon 3 HRAS (CAG (Gln) > CGG (Arg)) and glycine-to-arginine substitution at exon 2 NRAS (GGT (Gly) > CGT (Arg)) (Figure 2).

Table 3 BRAF and RAS Mutations in PTC

Figure 2 (A) BRAFV600E mutation in exon 15 showed T to A substitution (GTG/Valine to GAG/Glutamate). (B) NRAS mutation in exon 3 showed A to G substitution (CAA/Glutamine to CGA/Arginine). (C) HRAS mutation in exon 3 showed A to G substitution (CAG/Glutamine to CGG/Arginine). (D) KRAS mutation in exon 3 showed A to G substitution (CAA/Glutamine to CGA/Arginine).

The association between clinicopathological characteristics of PTC and mutational status is displayed in Table 4. There was a significant association of BRAFV600E mutation with high clinical stage, tall-cell variant, non-encapsulated tumor, LVI, ETE, and LNM. Whereas only follicular variants, encapsulated tumors, and no ETE were associated with RAS mutation.

Table 4 Differences Between Clinicopathological Characteristics of PTC and BRAFV600E and RAS Mutation

Furthermore, we performed a subgroup analysis to investigate if there are any differences between clinicopathological characteristics and NRAS-HRAS mutations after excluding KRAS-mutated patients. Except for tumor multifocality, there were no differences in the majority of clinicopathological characteristics between NRAS and HRAS mutational status (Table 5).

Table 5 Differences of Clinicopathological Characteristics of PTC Between NRAS and HRAS Mutation

Discussion

RAS/RAF/MEK/ERK (MAPK, mitogen-activated protein kinase) signaling is a crucial route that controls cellular proliferation, differentiation, and survival. BRAF and RAS genes are well-known proto-oncogenes that contribute to the development of PTC. Missense mutations in codon 600 of exon 15 (V600E) account for most of the activating mutations in the BRAF gene. Their association with the behavior of the disease remains controversial despite prior publications concerning the issue.^19,20

BRAFV600E had been known as the most prominent genetic mutation for thyroid carcinoma for decades. However, its worldwide prevalence in PTC ranges widely from 27.3% to 73.4%.²¹ A large meta-analysis study from the Asian population also showed a similar prevalence rate for BRAFV600E mutation, ranging from 23% to 83% in overall PTC cases.¹² Different sample sizes, multiple tissue sources, various sequence reading techniques, and other geographic considerations including genetic and environmental status may contribute to the significant discrepancies among studies. In the present study, BRAFV600E mutation was detected in 37.8% of PTC patients. This result is similar to the previous finding from Southeast Asia by Navarro-Locsin et al,²² which reported the BRAFV600E mutation rate was 38.5%. Similar to our result, there were studies in Indonesia conducted by Heriyanto et al²³ and Perdana AB et al¹⁸ that used molecular methods and reported BRAFV600E mutation rates were 40.3% and 31%, respectively. A single-center cohort study in Singapore reported BRAF mutation in 56% of the PTC patients.²⁴ An even higher percentage of BRAF mutation was also found in Japanese (82.1%), Vietnamese (83%), and South Korean population (81.3%).²⁵ Among other regions in Asian continents, BRAFV600E mutation was particularly more saturated in Southeast and East Asia.¹² We hypothesize the difference between Indonesia and other Southeast Asian countries is because of geographic conditions including environmental factors. Although still controversial, a high iodine intake has been associated with a higher risk for the occurrence of BRAF mutation in PTC.²⁶ A relatively poor iodine intake in certain provinces in Indonesia may be a plausible explanation for a lower percentage of BRAFV600E mutation in this country,^27,28 which needs to be elucidated.

Beside BRAFV600E, a mutation that can occur in exon 15 of the BRAF gene is BRAFK601E. This rare variant showed mutation at nucleotide 1801 in which an A>G transition occurred, producing lysine to glutamic acid substitution.²⁹ In contrast with BRAFV600E, BRAFK601E-mutated PTC is classified as a relatively low-grade tumor with no aggressive behavior such as ETE and metastases. In this study, we found only one case that harbored the BRAFK601E mutation. This particular case featured an encapsulated tumor, a follicular variant, no ETE, low stage, with neither distant nor LNM.

RAS was the second-most frequent genetic mutation discovered in thyroid carcinoma, primarily in follicular-cell-derived malignancies. The prevalence of RAS mutation in PTC was varied, ranging from 6.7% to 68.8%.^30–32 Our study identified RAS mutation in 21.5% of the PTC patients. This result seems to be similar to a study conducted in India, which reported RAS mutation in 20% of the patients.³³ Lower rate of RAS mutation was found in Saudi Arabia, in which the rate of RAS mutation was 6.1%.³⁴ Among the three homologous isoforms, NRAS was the most frequent mutation displayed in thyroid nodules.^30,34,35 This is similar to our finding where NRAS and HRAS mutation rates were 59.5% and 37.8%, respectively. Our study also supports evidence from Patel et al³⁵ where the majority of both NRAS and HRAS mutations were detected in codon 61. Furthermore, one case of KRAS mutation was observed in codon 61.³⁶ The KRAS mutation was higher in follicular adenoma than the NRAS mutation, which had a stronger oncogenic potential.^30,35

In general, PTC has predominantly occurred in women, between the fourth to fifth decades of life.^37–39 This is in line with our observation that 76.2% of PTC patients were female with the median age of diagnosis was 44 years. Despite the lack of association observed, BRAFV600E and RAS mutation were more prevalent in male and older patients. This pattern is similar to the previous studies.^39–43 One phenomenon of this event could be the neoplastic transition that may begin in older individuals.^44,45

In this cohort study, most of the patients were diagnosed in stage I and II of the disease. Several studies have explained the association between the clinical stage of PTC and BRAFV600E mutation.^21,46 Our finding is consistent with the previous studies in which we found a significant correlation between BRAFV600E mutation and high clinical stage (p = 0.033). RAS mutation was slightly prevalent (28.6%) in the clinical stage III and IV (p = 0.502).

Consistent with the preceding literature,^43,47 we observed the two most prevalent PTC variants were classic (34.9%) and follicular (30.2%). Interestingly, the number of tall-cell variants found in this study was quite high (21.5%). Our institution is a national referral hospital that is classified as a quaternary-level hospital. Mostly, we received advanced cases that cannot be solved at primary, secondary, and tertiary levels of care. Hence, the samples obtained were usually complicated cases with aggressive behavior.

BRAFV600E mutation has long been associated with aggressive variants of PTC.^21,48,49 In the present study, BRAFV600E mutation was predominantly found in tall-cell (86.5%), and classic variants (36.7%). We also discovered a significant positive correlation between BRAFV600E mutation and histologic variants (p = <0.001), especially classic (OR: 3.722; 95% CI: 1.434–9.606) and tall-cell variants (OR: 41.143; 95% CI: 11.979–141.308). Compared to the BRAFV600E mutation, the RAS mutation is unlikely to be displayed in the aggressive types of thyroid tumors. It is commonly associated with follicular adenoma, follicular thyroid carcinoma, and follicular variant of PTC.^34,50 This present study supports the previous evidence in which we observed that RAS mutation is more prevalent and showed a significant association with the follicular variant of PTC (p = <0.001).

The existence of tumor multifocality and capsules in PTC has been considered as the predictive marker for disease recurrence.^41,51,52 Previous studies concluded that BRAFV600E mutation significantly affects tumor multifocality development.^46,53,54 Those findings were incongruities to our present study in which we found no association between multifocality in both BRAFV600E and RAS mutation. However, HRAS-mutated patients were significantly inclined to show multifocality compared to NRAS-mutated patients (p = 0.016). We also found a significant association between the presence of tumor capsules in BRAFV600E and RAS mutation. BRAFV600E mutation was significantly more saturated in non-capsulated tumors (p = 0.001), whereas RAS mutation was significantly more prevalent in encapsulated tumors (p = 0.017). These findings also corroborate several existing studies.^21,55

LVI, ETE, and LNM are important prognostic factors in cancer, reflecting the aggressiveness of disease behavior. Supporting previous evidences,^21,46,48,54 a significant association between BRAFV600E mutation and the presence of LVI (p = 0.043), ETE (p = <0.001), and LNM (p = 0.009) were displayed in this study. Compared to the BRAFV600E mutation, we only found that RAS mutation is significantly associated with the absence of LNM (p = 0.033). The distribution of RAS mutation, which is more prevalent in encapsulated tumors, might be able to explain this finding.

Several limitations of this study may include a small sample size and limited data related to therapy and survival rate of the patients which may better represent disease aggressiveness. We also did not investigate TERT promoter mutation that frequently exists concurrently with BRAF or RAS mutation. Compared to a single mutation, some studies have reported that co-occurrence between TERT promoter mutation with either BRAFV600E or RAS mutation is associated with an increased PTC aggressiveness highlighted by a higher recurrence rate and decreased disease-free survival rate. It is known that the coexistence of TERT and BRAFV600E creates a special mechanism that increased the expression of the TERT mRNA in PTC.^56,57

In conclusion, this cohort study consisted of 172 PTC patients with the majority of female, being diagnosed under 55 years, and low clinical stage. BRAFV600E mutation was found in 37.8% of the patients. RAS mutations were found in 21.5% of the patients. One patient harbored BRAFK601E mutation and displayed an indolent morphology. Among RAS genes, 59.5% of the patients harbored NRAS mutation, 37.8% of the patients harbored HRAS mutation, and 2.7% of the patients harbored KRAS mutation. There was a significant association between BRAFV600E mutation and high clinical stage, tall-cell variants, non-encapsulated morphology, LVI, ETE, and LNM. Only follicular variant, encapsulated morphology, and absence of ETE were associated with RAS mutations. HRAS-mutated patients tend to show multifocality.