D. N. Poller and S. Glaysher

Transcription

1 DOI: /cyt BRAF V600 co-testing is technically feasible in conventional thyroid fine needle aspiration (FNA) cytology smears and can reduce the need for completion thyroidectomy Department of Pathology, Queen Alexandra Hospital, Cosham, Portsmouth, UK Accepted for publication 25 August 2013 BRAF V600 co-testing is technically feasible in conventional thyroid fine needle (FNA) aspiration cytology smears and can reduce the need for completion thyroidectomy Introduction: While fine needle aspiration cytology (FNAC) is the mainstay of diagnosis in thyroid nodules, molecular markers of thyroid cancer have recently been shown to be of value in improving the diagnosis and reducing the rates of unnecessary surgery. Method: A technical method is presented for the assessment of the BRAF V600 gene mutation in thyroid cancer using a simple adaptation of a commercially available kit. After standard preparation and reporting of conventionally stained alcohol-fixed Papanicolaou or air-dried Giemsa-stained slides the coverslip is removed from one slide, the DNA is extracted and submitted for PCR analysis. Results: Assessment of the BRAF V600 mutational status is feasible in very small quantities of DNA, requiring just greater than 5 ng per case from a single pre-stained FNA slide using this method. From the 14 cases examined thus far, one Thy4/Bethesda Class V case (suspicious of malignancy) has been identified with a BRAF V600 mutation and this patient, after multidisciplinary discussion, received a total thyroidectomy. Conclusion: Based on this methodology and other published results for the BRAF mutation, we believe that it is now feasible and cost effective for the UK NHS to BRAF co-test all Thy4/Bethesda Class V thyroid FNAs, as the additional cost of BRAF testing will still be much less than the cost of submitting all Thy4 (Bethesda Class V) patients to a partial and then a later completion thyroidectomy. Keywords: thyroid, cytology, fine needle aspiration, FNA, BRAF, papillary thyroid cancer, diagnosis Technical method Introduction In thyroid cancer somatic point mutations occur within the BRAF gene, resulting in a substitution of valine (V) to glutamic acid (E) at position 600. BRAF V600E mutations in the thyroid have been shown to be almost exclusive to thyroid cancer, therefore the BRAF V600E mutation is a highly specific marker of Correspondence: Dr D. N. Poller MD FRCPath, Department of Pathology, Queen Alexandra Hospital, Cosham, Portsmouth PO6 3LY, UK Tel.: +44 (0) ; Fax: +44 (0) ; david.poller@porthosp.nhs.uk human thyroid carcinoma. 1 Various systems for thyroid molecular cytological diagnosis have been developed including combinations of point mutations of BRAF and RAS with RET/PTC and PAX 8/ PPARc rearrangements, the Asuragen mirinform molecular panel 2 or other multiple gene expression markers such as the Veracyte Afirma gene classifier, 3 or using BRAF V600 point mutation alone or in combination with additional immunohistochemical expression of tumour-associated markers such as HBME-1 or CK19. 4,5 Method A commercially available CE-marked method for assessment of BRAF V600 mutation in formalin-fixed 1

2 2 paraffin-embedded (FFPE) cell blocks has been adapted for use in thyroid fine needle aspiration (FNA) cytology. This method is easy to use on airdried and alcohol-fixed FNA cytology smears. Briefly, a single slide showing representative cells is obtained, the cover slip is removed and the DNA is extracted. The Roche Cobas TM Mutation Test for the detection of the BRAF mutation V600E is a CE-IVD marked test for the detection of mutations in FFPE specimens (Roche Diagnostics Ltd., Burgess Hill, UK). This method has been adjusted here for the use with air-dried and alcohol-fixed FNA cytology smears. Methodology for extraction of DNA from FNA cytology smears DNA extraction is performed as outlined in the Cobas 4800 BRAF V600 Mutation Test Protocol EN 9 Doc Rev. 3.0 (cobas DNA Sample Preparation Kit P/N: , cobas 4800 BRAF V600 Mutation Test P/N: ) for in vitro diagnostic use. The following amendments to the protocol are outlined below. A previously prepared and reported alcohol-fixed Papanicolaou or Giemsa-stained cytology slide is added to a container with sufficient xylene volume to cover the mounted area. Slides are soaked overnight at room temperature, for more than 10 h, or until the coverslips are easily removed. The coverslip is removed slowly and gently, taking care not to disturb the cellular material underneath. After the coverslip has been removed, the slides are soaked in xylene for at least another 5 10 minutes to remove excess mounting medium. Cellular material is scraped from the slide with a sterile scalpel into an Eppendorf tube (Eppendorf UK Ltd, Stevenage, UK). Further cell lysis and DNA isolation methods are as per the Cobas DNA Sample Preparation Kit with DNA finally being eluted in 30 ll of DNA elution buffer. DNA quantitation is performed immediately after DNA isolation using a NanoDrop spectrophotometer (Labtech International Ltd, Uckfield, UK). An average of two readings is performed for each sample with measurements reading within 10% of each other. The average DNA concentration for FNA cytology smears was 17.4 ng/ll (range ng/ll), above the minimum requirements for the assay (minimum assay DNA concentration 5 ng/ ll). Results Validation of formalin-fixed paraffin-embedded samples with the Cobas System against external laboratory and alternate PCR system Initially inter- and intra-system validation was performed in 21 selected samples, 13 from anonymized spare tissue cell blocks from colorectal cancer patients and eight from anonymized spare tissue cell blocks from skin melanoma patients where the BRAF V600E status had previously been determined. Newly extracted DNA from these FFPE samples was directly compared with previously validated results, Life Technologies Cast PCR Assay for KRAS and BRAF mutations, as well as externally tested Cobas â BRAF V600 tests, performed at The Royal Marsden Hospital, London, UK (data not shown). All validation sample results showed concordance with those previously tested. Validation of cytology smear slides against formalin-fixed paraffin-embedded samples in BRAF-positive cell lines As extraction of DNA from cytology FNA slides results in the destruction of the slide we devised a system to validate FFPE-processed cells directly with stained cell smears using cell lines. A BRAF V600Emutant cell line SKMEL28 was grown to 90% confluence in tissue culture flasks before being removed and split into two processing groups. In the first group, each were embedded into paraffin cell blocks and in the second group all were made into cytology smears, both air-dried and alcohol-fixed, using a Cytospin TM technique (Fisher Scientific UK Ltd, Leicestershire, UK). Cell line-based smears were prepared as per The Queen Alexandra Hospital Cytology Department, Portsmouth, standard operating procedure for non-gynaecological FNA samples. Half of the cell preparations were stained with Papanicolaou and the other with Giemsa. Cell blocks were made by placing cell suspensions into a solid scaffold, HistoGel (Fisher Scientific UK Ltd) which could then be processed via normal Queen Alexandra Hospital histological standard operating procedure for preparation of FFPE blocks. Slides and FFPE sections were then processed via the Cobas DNA Sample Preparation Kit as per the manufacturer s guidelines and as outlined above. Results were positive and mutation results were seen for both Papanicolaou and Giemsa-stained slides and for FFPE specimens.

3 BRAF V600 co-testing on thyroid FNA smears 3 Validation of retrospective thyroid FNA cytology slides with corresponding thyroid formalin-fixed paraffinembedded tissue samples After the assessment of the cell line slides, seven anonymized classic-type papillary carcinoma retrospective cases where tested where both a FFPE specimen was available from thyroid surgery and a corresponding FNA cytology specimen with adequate cellularity. DNA was isolated and quantified as previously described. All match-paired papillary thyroid carcinoma cases showed concordant results for the Cobas â 4800 BRAF V600 Mutation Test. Five tumours showed matching BRAF V600E mutation genotypes and two showed matching wild-type (WT) genotypes. All seven cases of papillary thyroid carcinoma showed concordant results of the BRAF V600 E mutation on the paired FNA cytology smear and the subsequently excised papillary carcinoma from the same patient (Table 1). Routine clinical application of BRAF V600 mutational analysis We have now prospectively commenced testing of Thy3F (neoplasm possible, follicular), Thy4 (suspicious for malignancy) and Thy5 (malignant) thyroid FNA (equivalent to class IV, V and VI respectively in the Bethesda system). After the testing of 14 specimens we have identified one case positive for the BRAF V600 mutation (Table 2). This was a Thy4 case, which after MDT review and discussion was treated as a cytological diagnosis of thyroid carcinoma. The excised tumour was a 40-mm UICC stage pt2 tall-cell variant of papillary carcinoma, which is known to show BRAF V600 mutation in 60 80% of cases. 6 This patient underwent a single stage total Table 1. Results of validation phase comparing FNA slide result and BRAF V600E on the formalin-fixed paraffinembedded excised tumour Case FNA slide Excised tumour 1 Mutant Mutant Yes 2 Mutant Mutant Yes 3 WT WT Yes 4 Mutant Mutant Yes 5 Mutant Mutant Yes 6 WT WT Yes 7 Mutant Mutant Yes FNA, fine needle aspirate; WT, wild type. Concordant Table 2. Preliminary results after 3 months of service showing one Thy4 case (Case 10) upgraded to management as a cytological diagnosis of carcinoma and a single-stage total thyroidectomy after a multidisciplinary case review Case number thyroidectomy after MDT review of the case and informed discussion with the patient. Discussion Thy category BRAF V600 result 1 3F WT 2 3F WT 3 3F WT 4 4 WT 5 3F WT 6 4 WT 7 3F WT 8 3F WT 9 4 WT 10 4 Mutated 11 5 WT 12 5 WT 13 3F WT 14 3F WT Thy, thyroid cytology; WT, wild type; F, follicular. We have successfully implemented prospective testing for BRAF V600 as an in-house test for thyroid carcinoma on FNA cytology specimens. Furthermore, we believe that in spite of the cost and time involved in BRAF V600 co-testing of thyroid FNA cytology, for the 3 6% of patients that have Bethesda Class V smears, which are suspicious of malignancy, UK RCPath Thy4 7,8 the estimated the cost of BRAF V600 testing at 300 per case will be easily offset by the cost saved on an unnecessary thyroidectomy which is 2973 based on the NHS Schedule Reference Cost. 9 The rationale of this argument is as follows. If the rate of Thy4 aspirates is 3 6% 9 of the total number of thyroid aspirates then based on published literature it is anticipated that approximately one-third of Thy4 or Bethesda Class V Thyroid FNAC would show BRAF V600 mutations. 1,2,4,10 These patients could be spared a two-stage thyroidectomy procedure and instead receive a one-stage thyroidectomy when a BRAF V600 mutation result was positive. The practice of thyroid surgery appears to be rapidly evolving. In a recent leading article in The Journal of Clinical Endocrinology and Metabolism Sarne comments, For patients with an indetermi-

4 4 nate lesion, a positive test for BRAF (obtained as part of a more comprehensive panel because BRAF alone would be insufficiently sensitive) is almost diagnostic of papillary cancer, and those patients should have a total thyroidectomy. 11 The advantage of the Cobas â system is that apart from the process of DNA extraction all other aspects are controlled by the manufacturer and machine. The test gives a single result, WT or mutant and requires little DNA ( 5 ng/ll). Furthermore, as the DNA is extracted from previously stained smears, it does not require the collection of additional material for molecular diagnosis or for the production of cytoblocks. As the slides are checked prior to DNA extraction it is always certain that representative cells are present in the specimen used for mutation analysis. Koh et al. 12 showed that for lesions suspicious of malignancy after BRAF V600E testing sensitivity rates of thyroid FNAC increased from 67.2% to 78.9%, the number of cases that were malignant increasing from 213 cases to 250 out of a total of 317 cases. Johnson et al. also presented a retrospective audit of the use of BRAF V600 mutations in thyroid FNAC. They identified one BRAF V600E mutation in 37 Thy3F cases, four BRAF mutations (three BRAF V600E and one non-braf V600E) in 15 Thy4 lesions and 11 BRAF V600E mutations in 18 Thy5 lesions. 13 The caveats to this new laboratory technique are as follows. First, it is essential to ensure that there are adequate numbers of diagnostic cells present in the slide submitted for DNA analysis to provide adequate DNA. Second, there is always a very small risk of a false-positive mutational analysis result caused by a metastatic tumour to the thyroid that shows a BRAF mutation such as a metastatic melanoma or colorectal carcinoma 14 or from specimen DNA contamination from another patient or from another laboratory sample. Hence the need to ensure that only one diagnostic slide is scraped for DNA analysis and that the other remaining diagnostic slides that show representative cytological material are available for multidisciplinary review before a decision to treat is made. Third, as the BRAF V600 mutation is a relatively insensitive test for thyroid carcinoma and is only mutated in around one-half of all thyroid cancers 1 the majority of non-papillary primary thyroid carcinomas such as follicular variant of papillary carcinoma, poorly differentiated, and anaplastic carcinoma do not show the BRAF V600 mutation and will be BRAF V600 negative (WT) on cytological testing. In summary, we believe that there is a strong cost-benefit case for testing all Thy4/Bethesda Class V FNA for BRAF V600 mutations. We anticipate this will make one-third of surgery for Thy4/Bethesda Class V lesions unnecessary. Our very preliminary experience after a few months of BRAF V600 cotesting of thyroid FNAC suggests that this is indeed the case with one Thy4 patient now being upgraded to a one-stage total thyroidectomy after discussion and re-review of the slides and the case in the Portsmouth Thyroid Multidisciplinary meeting. Note added in proof Since the date of submission of this article we have had two further Thy4 (Bethesda Class V) patients that were BRAFV600 mutated on cytology who will, after multidisciplinary review, proceed to a one stage total thyroidectomy. We also now use BRAF V600 as a routine diagnostic test for challenging histological specimens where papillary carcinoma is considered in the differential histological diagnosis. Acknowledgments We are grateful to The Portsmouth Rocky Appeal for funding the implementation of this diagnostic test, and also to our other colleagues and thyroid multidisciplinary team members, Mr Avi Agrawal, Dr Daren Gibson, Mr Dae Kim, Dr Anne Spedding, Dr Elizabeth Tilley and Mr Constantinos Yiangou. References 1. Nikiforov YE. Molecular diagnostics of thyroid tumors. Arch Pathol Lab Med 2011;135: Nikiforov YE, Ohori NP, Hodak SP et al. Impact of mutational testing on the diagnosis and management of patients with cytologically indeterminate thyroid nodules: a prospective analysis of 1056 FNA samples. J Clin Endocrinol Metab 2011;96: Alexander EK, Kennedy GC, Baloch ZW et al. Preoperative diagnosis of benign thyroid nodules with indeterminate cytology. N Engl J Med 2012;367: Rossi ED, Martini M, Capodimonti S et al. Diagnostic and prognostic value of immunocytochemistry and BRAF mutation analysis on liquid-based biopsies of thyroid neoplasms suspicious for carcinoma. Eur J Endocrinol 2013;168: Cochand-Priollet B, Dahan H, Laloi-Michelin M et al. Immunocytochemistry with cytokeratin 19 and antihuman mesothelial cell antibody (HBME1) increases the

5 BRAF V600 co-testing on thyroid FNA smears 5 diagnostic accuracy of thyroid fine-needle aspirations: preliminary report of 150 liquid-based fine-needle aspirations with histological control. Thyroid:2011;21: Kovacs CM, Nose V. Papillary thyroid carcinoma, tall cell variant. In: Diagnostic Endocrine Pathology: Nose V (ed). Salt Lake City: Amirsys Publishing; 2012: 1: pp Cross PA, Chandra A, Giles T et al. Guidance on the reporting of thyroid cytology specimens. rcpath.org/publications-media/publications/publications. htm#cyto (accessed ). 8. Kocjan G, Chandra A, Cross PA et al. The interobserver reproducibility of thyroid fine-needle aspiration using the UK Royal College of Pathologists classification system. Am J Clin Pathol 2011;135: Poller DN, Kandaswamy P. A simplified economic approach to thyroid FNA cytology and surgical intervention in thyroid nodules. J Clin Pathol 2013;66: Carr R, Ustun B, Chhieng D et al. Radiologic and clinical predictors of malignancy in the follicular lesion of undetermined significance of the thyroid. Endocr Pathol 2013;24: Sarne DH. A piece of the puzzle: what does BRAF status mean in the management of patients with papillary thyroid carcinoma? J Clin Endocrinol Metab 2012;97: Koh J, Choi JR, Han KH et al. Proper indication of BRAF(V600E) mutation testing in fine-needle aspirates of thyroid nodules. PLoS One 2013;8:e Johnson SJ, Hardy S, Roberts C et al. Pilot of BRAF testing in indeterminate (and malignant) thyroid cytology. ; (in press). 14. Moghaddam PA, Cornejo KM, Khan A. Metastatic carcinoma to the thyroid gland: a single institution 20- year experience and review of the literature. Endocr Pathol 2013;24: