Diagnostic Efficacy of BRAFV600E Immunocytochemistry in Thyroid Aspirates in Bethesda Category IV and Papillary Thyroid Carcinoma

Abstract

Background:

In papillary thyroid carcinoma (PTC), BRAFV600E is a common mutation and is associated with aggressive clinical behaviour. Immunocytochemistry (ICC) and molecular testing are recommended in the Bethesda System for Reporting Thyroid Cytopathology 2017 (TBSRTC) category III, IV and V.

Aims:

The current study aimed to evaluate the diagnostic efficacy of conventional FNAC versus FNAC with BRAFV600E immunostaining in cases of TBSRTC category IV, cases of suspicious for PTC and cases of PTC.

Methods and Material:

The study included a prospective case series of 45 patients with clinically palpable thyroid nodules with TBSRTC category IV, category V (suspicious for PTC) and PTC. The corresponding histology specimens of all the 45 cases were also analyzed. Immunostaining for BRAFV600E was performed on FNAC cell blocks and their corresponding histology sections using anti-BRAF (VE1) clone (Ventana). The diagnostic efficacy of the BRAFV600E immunostaining was compared on cytological specimens with histological specimens

Results:

BRAFV600E immunostaining helped to improve the sensitivity of the cytology to confirm the PTC as a diagnostic aid for thyroid FNAs. Cytology alone had a sensitivity of 62.96% and a lower specificity of 60.70%. The combination of both the tests together provided 84.62% sensitivity and much higher specificity of 100%. PPV was also increased to 100% and NPV was raised 94.12%.

Conclusions:

The performance of BRAFV600E immunostaining on the cytological specimen is a rapid, simple and cost-effective test and could be considered in TBSRTC category IV and suspicious and malignant cases of PTC.

INTRODUCTION

Approximately 70%–80% of cases of clinically palpable thyroid nodules can be accurately diagnosed on fine-needle aspiration cytology (FNAC).[1,2] The remaining 20%–30% cases fall in indeterminate categories of the Bethesda System for Reporting Thyroid Cytopathology 2017 (TBSRTC), including category III, that is, atypia of undetermined significance/follicular lesion of undetermined significance (AUS/FLUS), category IV, that is, follicular neoplasm/suspicious for follicular neoplasm (FN/SFN) and category V, that is, suspicious for malignancy (SFM). These lesions are the most challenging to diagnose accurately for further management. Currently, only the surgical excision with histopathological analysis is the only way to provide the final diagnosis.[3] Detection of molecular biomarkers is recommended to improve the cytological diagnostic accuracy of these lesions.[4,5,6] The most common mutation that occurs in PTC is a threonine-to alanine nucleotide transversion at position 1799(T1799A) in exon 15, which results in a valine-to-glutamate substitution at position 600 (V600E).[7] BRAFV600E mutation is a specific biomarker for PTC with a mutation rate of 53.0% to 80.6%.[8,9] In PTC, BRAFV600E mutation is an important prognostic molecular marker, and can improve the diagnostic accuracy of FNAC, and is significantly associated with aggressive clinicopathologic characteristics.[10] In follicular thyroid carcinoma (FTC), BRAFV600E mutation detection has a limited role, but in the case of FN on cytology, its detection is associated with the risk of extra-thyroid extension and metastasis after surgery.[11]

Different DNA-based methodologies have been successfully applied for identifying BRAF mutations, such as direct sequencing, allele-specific polymerase chain reaction (PCR), or real-time PCR. But despite their excellent sensitivity, these methods are time-demanding, expensive and they require strict quality control.[12,13,14,15] Therefore, as an alternative to these limitations, the development of BRAFV600E mutation-specific antibody (clone VE1) is now rapidly gaining popularity as an easier detection technique for processed formalin-fixed, paraffin-embedded (FFPE) tissues from various tumors.[14] Although there is a high concordance rate of VE1 immunostaining with molecular methods in surgical specimens of PTC, its evaluation is rarely conducted in thyroid cytology specimens, and its diagnostic efficacy in cytology has not been well established.[16,17,18,19,20] The role of BRAFV600E mutation to differentiate benign from malignant lesions among indeterminate cytological categories of TBSRTC remains controversial, to date. A recent meta-analysis including 32 eligible studies proposed that BRAFV600E mutation played a limited role in the diagnosis of indeterminate cytological categories due to its low sensitivity, despite a specificity of nearly 100%.[21,22] In contrast, one study from China of 314 thyroid nodules including 52 cases of TBSRTC category III/IV and 13 cases of TBSRTC category V concluded that BRAFV600E mutation could improve the prediction of malignancy in indeterminate nodules.[21] The diagnostic efficacy of BRAFV600E mutation to differentiate between TBSRTC categories III, IV, and V has not been well established. Molecular testing is usually not recommended in the cases reported as PTC on cytology. In the present study, we aimed to evaluate BRAFV600E immunostaining with the VE1 antibody for a diagnosis of papillary thyroid carcinoma (PTC) in cell blocks of FNA thyroid and its utility in category IV and V (suspicious for PTC) cases.

MATERIALS AND METHODS

This prospective study was approved by the Institutional Ethical Committee (IEC 3/16) and informed written consent was obtained from all the patients before performing FNAC and BRAFV600E evaluation. A total of 334 patients with clinically palpable thyroid nodules, from April 2017 to March 2019 were selected as likely study subjects. In the current study, the number of cases was restricted to 45 as it was a pilot study conducted as an institutional intramural project with limited funding and cost constraints. Inclusion criteria were nodules in TBSRTC categories IV, suspicious for PTC or PTC on cytology with adequate cell yield on cell blocks. The cell blocks were considered adequate if a total of 100 or more viable follicular epithelial cells were present. Exclusion criteria were: 1) nodules in TBSRTC category I, II, or Medullary thyroid carcinoma; 2) cases of TBSRTC categories IV, V, or PTC, where subsequent histological specimens were not received. A total of 45 patients with clinically palpable thyroid nodules were finally enrolled in the current study. FNAC of the thyroid nodules was performed with 23/24-gauge needles. Both air-dried and wet fixed (in 95% ethanol) smears were prepared. The smears were stained with May-Grunwald-Giemsa (MGG), Hematoxylin and Eosin (H and E), and Papanicolaou (Pap) stains.

Cell block preparation: Cell blocks were prepared from the remaining aspirates using Nathan's protocol.[23] The remaining aspirates were rinsed in 50% ethanol and centrifuged in a 10-mL disposable centrifuge tube at 4,000 rpm for 6 minutes. The supernatant fluid was decanted and the deposit (pellet) was fixed in freshly prepared Nathan alcohol formalin substitute (NAFS) consisting of 9 parts of 100% ethanol and 1 part of 40% formaldehyde. The pellet was then processed for sectioning as per the routine processing schedule of the laboratory.

H and E stained sections were examined for each cell block. Slides of FNA smears as well as cell blocks were evaluated. All cases were categorized according to TBSRTC 2017 on cytology. The cases, that were categorized as FN/SFN (category IV), suspicious for PTC (category V), and PTC were included for the study. The subsequent histological specimens of the cases included in the study, were received in our laboratory (e.g., Hemithyroidectomy, subtotal thyroidectomy and total thyroidectomy with or without lymph nodes). The evaluation of histopathology specimen was done as per the current CAP (College of American) guidelines for grossing of resection/thyroidectomy specimens. It included size, laterality, consistency, appearance of the specimen, presence or absence of capsule and capsular or extrathyroidal extension of tumor in cases of suspicious malignancy. Presence or absence of colloid, calcification and areas of necrosis were also noted. Submitted lymph nodes were also evaluated for nodal metastasis.

Sections of FFPE tissues were stained with H and E and evaluated. The cytological diagnosis and the histological diagnosis were compared.

Automated immunohistochemistry: Immunostaining of BRAFV600E was performed with anti-BRAF (VE1) clone (Ventana) on 3-5 μm sections from the cell blocks and FFPE tissue blocks. Immunostaining was performed on a Ventana BenchMark XT automated staining instrument according to the manufacturer's instructions, with onboard deparaffinization, retrieval, and staining (Ventana Medical Systems, Tucson, AZ, USA). The antibody was validated and standardized in our laboratory as per the manufacturer's instructions using BRAFV600E positive cell lines. All immunostained stained sections were independently analyzed by two of the authors (N.A. and N.H.) for cytoplasmic BRAFV600E protein expression and intensity of staining. This inter pathologist correlation was checked by applying kappa statistics, and the inter-observer variation was 0.93 (excellent agreement), with a significance level of <0.05.

Cytoplasmic BRAFV600E protein expression was considered and the intensity of staining was recorded on a 0 to 3+ scale where the absence of staining was scored 0, weak cytoplasmic staining was scored 1+, moderate cytoplasmic staining was scored 2+, and intense cytoplasmic staining was scored 3+. The evaluation criteria adopted in some studies of the percentage of tumor cells were also used,[18] and the grade which was higher of the above two were used. The final intensity of 0 and 1+ were considered as negative [Figure 1a–d], while the intensity of 2+ and 3+ was considered as positive staining [Figure 2a–d]. Tissues of BRAFV600E positive colorectal adenocarcinoma or papillary thyroid carcinoma were used as positive controls and unremarkable thyroid tissue was used as negative controls.

Figure 1.

Negative BRAFV600E immunostaining: (a and b) Absence of BRAFV600E immunostaining on the cell block and corresponding histopathology specimen of a case of Follicular adenoma (DAB; x200). (c and d) BRAFV600E immunostaining with intensity of 1+ on cell block and corresponding histopathology specimen of a case of PTC (DAB; x200)

Figure 2.

Positive BRAFV600E immunostaining: (a and b) BRAFV600E immunostaining with intensity of 2+ on cell block and corresponding histopathology specimen of a case of PTC (DAB; x200). (c and d) BRAFV600E immunostaining with intensity of 3+ on cell block and corresponding histopathology specimen of a case of PTC (DAB; x200)

Statistics

The statistical analysis was done using the Statistical Package for the Social Sciences (SPSS) software version 24. The study was based on diagnostic test evaluation wherein the diagnostic efficacy of the BRAFV600E ICC on cytological specimens was compared with BRAFV600E IHC on histological specimens (considered the gold standard in the study). The sensitivity, specificity, positive, and negative predictive values were derived. The association between 2 categorical variables was assessed using the Chi-square test. A P value of <0.05 was considered statistically significant.

RESULTS

Correlation of cytological and histological diagnosis with BRAFV600E immunostaining status:

The study included fine needle aspirates of thyroid nodules from 45 patients, with TBSRTC category IV (FN/SFN), suspicious for PTC (TBSRTC category V), and PTC. Of these, 27 (60%) cases were of category IV, all were negative except two were positive for BRAFV600E ICC whereas, among the 6 (13.3%) cases in category V, 33.3% (n = 2) cases were regarded as positive for BRAFV600E ICC. Out of 12 (26.7%) cases of PTC, 58.3% (n = 7) cases were positive for BRAFV600E ICC as shown in Table 1.

Table 1.

Comparison of BRAFV600E immunostaining on cytological category IV, V, VI and on their corresponding histopathology specimens

Cytological diagnosis	Number of case	Histopathological diagnosis No. of cases	BRAF ICC on cell block	BRAF IHC on histology
Cat. IV	27	PTC-follicular variant 02	02	02
		FTC 04	00	00
		FTC-dedifferentiated	00	00
		02
		FTC-minimally	00	00
		invasive 01
		FTC-encapsulated angioinvasive variant 01	00	00
		FA 10	00	00
		HCA 01	00	00
		HN 02	00	00
		MNG 04	00	00
Cat. V	06	PTC-follicular	00	00
		variant 02	02	02
		PTC-classical	00	00
		variant 02	00	00
		PTC- columnar cell variant 01
		Graves’ disease 01
Cat. VI	12	PTC-classical variant 12	07	09
Total	45		11	13

Cat., category; FN/SFN, follicular neoplasm/suspicious for follicular neoplasm; PTC, papillary thyroid carcinoma; FVPTC, follicular variant of PTC; FTC, follicular thyroid carcinoma

The subsequent histological specimen of 27 cases of TBSRTC category IV (FN/SFN), 11 cases were reported as a follicular adenoma (FA) including one case of hurthle cell adenoma (HCA), 2 cases as a hyperplastic nodule (HN), and 4 cases as multinodular goitre (MNG), and while all were negative for BRAF ICC but 2 cases of TBSRTC category IV were considered positive for BRAF ICC and both the cases were eventually confirmed to be cases of follicular variant of PTC (FVPTC) and were also positive for BRAF IHC on their subsequent histological specimens. The remaining 8 cases of this category were Follicular thyroid carcinoma (FTC) and its variants and none were positive for BRAFV600E immunostaining on the cellblock or on the histology sections. Out of the 6 cases of suspicious for PTC (category V), one case was reported as Graves' disease and was negative for BRAFV600E immunostaining on the cell block and on the histology sections. The remaining 5 cases were regarded as malignant on subsequent histology, (PTC 2, FVPTC 2, a columnar variant of PTC 1) including the BRAFV600E ICC positive cases and were confirmed on BRAFV600E IHC, for their positivity [Figure 3]. None of the cases was reported as Non-invasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP).

Figure 3.

Positive BRAFV600E immunostaining in Columnar Cell Variant: (a) Histopathology of case of Columnar cell variant of PTC (b) BRAFV600E immunostaining with intensity of 3+ on cell block (DAB; x200)

All the 12 cases of PTC on cytology were confirmed for their diagnosis on the subsequent histology. 7 cases of this category were considered positive for BRAFV600E ICC on cell block sections and 5 cases were considered negative. The BRAFV600E IHC on the corresponding histology sections was considered positive in 9 cases including the 2 cases that were considered negative on cytology due to cytoplasmic staining of 1+.

Correlation of clinicopathologic and cyto-histological features of the cases included in the study based on BRAFV600E immunostaining status:

In the study, all 45 cases were histologically confirmed. Among the 27 malignant cases on histology, IHC for BRAFV600E was positive in 28.9% (13/45) cases. A comparison of the clinicopathologic characteristics between BRAF-positive and BRAF-negative groups did not show any significant association by age, laterality, and T-stage of the malignant cases. [Table 2] However, a significant association (P < .05) was seen between sex, capsular invasion, lymph nodal metastasis, cytological categories as well as with histological diagnosis.

Table 2.

Correlation of clinicopathologic and histological features of cases included in the study based on BRAFV600E immunostaining status

Features	Number (%)	BRAFV600E Positive	BRAFV600E Negative	P
Age (years) (n=45)
mean±SD	37.44±11.54	10	25	0.929
<45 years >45 years Sex(n=45)	35 (77.8) 10 (22.2)	03	07	0.050*
Male	12 (26.7)	06	06
Female	33 (73.3)	07	26
Side (n=45)
Right	16 (35.6)	03	13	0.220
Left	15 (33.3)	03	12
Both	12 (26.7)	06	06
Isthmus	02 (4.4)	01	01
T-stage (n=27)
T1	04 (14.8)	01	03	0.597
T2	08 (29.6)	04	04
T3	15 (55.6)	08	07
Capsular invasion (n=27)
No	17 (63.0)	04	13	0.004*
Yes	10 (37.0)	08	02
N-Stage (n=16)
NO	05 (31.25)	01	04	0.049*
N1	11 (68.75)	08	03
Cytological diagnosis (n=45)	27 (60.0)	02	25	0.003*
Category IV	06 (13.3)	02	04
Category V	12 (26.7)	07	05
Category VI
Histological diagnosis
PTC-classical variant	14 (31.1)	09	05	0.001*
PTC-follicular variant	04 (8.9)	04	00
PTC-columnar cell variant	01 (2.2)	00	01
FTC	04 (8.9)	00	04
FTC-dedifferentiated	02 (4.4)	00	02
FTC-minimally invasive	01 (2.2)	00	01
FTC-encapsulated angioinvasive	01 (2.2)	00	01
Non-malignant cases	18 (40.0)	00	18

PTC, papillary thyroid carcinoma; FTC, follicular thyroid carcinoma

DISCUSSION

The current study aimed to determine the diagnostic efficacy of BRAFV600E immunostaining in thyroid aspirates in TBSRTC category IV, and suspicious as well as PTC cases on cytology, prospectively. BRAF mutational analysis is a useful diagnostic marker of PTC and is associated with almost 100% specificity to PTC or ATC (Anaplastic thyroid carcinoma) and is a definite indication of malignancy if found positive between the samples.[24,25] This mutational analysis has also been found to be important as confirmation of malignancy in any of the TBSRTC category, so helped in better surgical management with total thyroidectomies instead of lobectomies.[26] In the current study, out of 27 samples in category IV, only 2 samples (7.4%) were found to be BRAFV600E ICC positive, however, there are other reports where >10% of the BRAF mutated cases were seen in this category.[27] Yet, numerous studies have demonstrated low positivity (3–4.6%) and have argued that BRAFV6000E analysis alone might not provide any advantage over the cytology results in the indeterminate category.[28,29] Ohori et al.[30] tried to explain that this difference is probably due to the variability of application of TBSRTC criteria, especially in the demarcation of the category III and category V and category IV and category V, which can vary across centers. The plausible reason for the low prevalence of BRAFV600E mutations in the indeterminate category is likely accountable to majority lesions being follicular adenomas, FTC, and FVPTC in this category.[25] Few studies have debated that BRAFV600E mutational analysis is truly useful in category V where about 15–20% of them usually appear to carry mutations.[31,32] In the current study, 33.3% of lesions in category V (suspicious for PTC) were positive for BRAFV600E immunostaining which signifies their malignant nature.

However, the highest percentage positivity in the study was, as expected, in the cases of malignant PTC, where nearly 58.3% of the PTCs harbored BRAFV600E protein over-expression. The prognostic implication of BRAFV600E testing pre-surgically basically aids in further patient management and serves as a guide for the type of surgery to be performed. The specificity for BRAFV600E protein over-expression, in the current study, was 100% and none of the FTCs or benign lesions stained up for BRAFV600E immunostaining. BRAFV600E protein overexpression was present in 9 (64.3%) cases of the 14 histologically confirmed PTCs and 2 (50%) cases of FVPTCs. From Table 1, it is evident that the main utility of BRAFV600E immunohistochemistry is in confirming 50% of the histologically confirmed cases of FVPTC (2/4).

In the study conducted by Su et al.[33] an increased sensitivity by only 6% was documented in their meta-analysis. However, Nikiforov et al.[34] who assessed the FNA of 1131 patients have reported lower overall combined sensitivity of 60%.

It has been documented in limited studies published in the literature that BRAFV600E estimation in thyroid aspirates improves the diagnostic accuracy of FNAC of thyroid nodules.

In the Indian subcontinent, the study published by Gupta et al.[35] wherein the authors conducted a pilot study on molecular testing in indeterminate categories of thyroid FNAC aspirates. The authors concluded from their study that BRAFV600E and RAS mutation testing from cytology specimens are useful as a rule in test for indeterminate thyroid nodules.

The prognostic significance of BRAFV600E positive cases was evaluated. BRAFV600E positivity is associated with aggressive behavior.[11] In the current study, the association between BRAFV600E positive versus BRAFV600E negative cases was assessed in terms of the clinicopathological features. The BRAFV600E positive cases were more common in females when compared with males (P = 0.050), with the presence of capsular invasion (P = 0.004), lymph node metastasis (P = 0.049). The majority of BRAFV600E positive cases were of the category VI (P = 0.003). All these parameters were statistically significant in the BRAFV600E positive category when compared with the BRAFV600E negative group. Some studies have reported an association of BRAFV600E positivity with lymph node metastasis. In cases that are clinically negative for nodal metastasis, pre-operative BRAFV600E immunostaining and testing can predict the potential of occult metastasis in the central compartment lymph node.

The current study is unique; however, there are some limitations. In the current study, the sample size was small and the results of immunostaining were not compared with any molecular technique. The main objective of the study was to accurately diagnose the PTC cases on cytology with help of BRAFV600E immunostaining on cytological specimens.

This is a novel study wherein BRAFV600E protein expression on FNA cell blocks and subsequent histopathology specimen has been correlated with clinicopathological and prognostic parameters. Predictive and diagnostic data were derived in the form of sensitivity and specificity. The performance of BRAFV600E immunostaining in FNA cell blocks had predictive implications as it added in deciding the type of initial surgery and further patient management. BRAFV600E immunostaining is a rapid, simple, and cost-effective and does not require any special training for interpretation. When compared with molecular techniques, it is neither dependent on DNA quality nor the number of tumor cells in FNA specimens.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.