J Cancer Res Clin Oncol DOI 10.1007/s00432-015-1939-9
ORIGINAL ARTICLE – CANCER RESEARCH
Follicular variant of papillary thyroid carcinoma (FVPTC): histological features, BRAF V600E mutation, and lymph node status Ann E. Walts · James M. Mirocha · Shikha Bose
Received: 6 February 2015 / Accepted: 12 February 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose Follicular variant of papillary thyroid carcinoma (FVPTC) is currently treated like conventional papillary thyroid carcinoma (cPTC). Recent reports indicate that encapsulated FVPTC behaves like follicular adenomas, while infiltrative FVPTC behaves like cPTC. This raises the possibility that histology and/or mutation status might help personalize management of FVPTC regarding extent of surgery, intensity of follow-up, and targeted therapy. This study correlates histological features, immunoreactivity for CK19, HBME, and Gal, and BRAF V600E mutation with lymph node (LN) metastasis and follow-up in FVPTC. Methods Forty-eight FVPTC (21 with regional lymph node metastasis [LN+] and 27 with negative lymph nodes [LN−]) were reviewed. Demographics, tumor focality, size, circumscription, follicular architecture, lymphovascular invasion, extrathyroidal extension (ETE), and margin status were charted. Macrodissected formalin-fixed paraffin-embedded sections from 47 (21 LN+ and 26 LN−) cases were analyzed for BRAF V600E (1799T>A) mutation using real-time PCR. Correlations between the variables and LN status were calculated.
Presented in part at the United States & Canadian Academy of Pathology Annual Meeting, Baltimore, MD, March 2013. A. E. Walts (*) · S. Bose Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, CA 90048, USA e-mail: [email protected] J. M. Mirocha Biostatistics Core, Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA 90048, USA
Results Sixty-two percent of cases with ETE demonstrated LN metastasis, while 59 % of cases with circumscribed tumors were LN−. In multivariable analysis, ETE and tumor size ≥1 cm were the best predictors of LN+ status, whereas in cases without ETE, the infiltrative pattern and tumor size provided the “best fit.” Immunostains and BRAF mutation status were not helpful. All four tumors that recurred were LN+, with infiltrative borders, and lacked the BRAF mutation. Conclusions Tumor circumscription, extrathyroidal extension, and tumor size ≥ 1.0 cm are predictors of lymph node status in FVPTC. Keywords Follicular variant · Papillary thyroid carcinoma · BRAF V600E mutation · Extrathyroidal extension · Circumscription · Histological features
Introduction The follicular variant of papillary thyroid carcinoma (FVPTC) has presented challenges to pathologists and clinicians since the term was introduced (Hazard et al. 1955; Lindsay 1960, 1964) to describe tumors that “resemble papillary carcinoma in biologic behavior and all morphologic features with the exception that papillae are not present” (Chem and Rosai 1977). As subsequently defined by Livolsi (2003) and presented in the World Health Organization Classification of Tumors (DeLellis et al. 2004), FVPTC is composed “entirely or almost entirely of follicles lined by cells exhibiting nuclear features of papillary thyroid carcinoma.” Not surprisingly, the diagnosis and management of FVPTC have been confounded by the lack of evidence-based consensus guidelines regarding the extent of papillary growth “allowable” in FVPTC
13
J Cancer Res Clin Oncol
and the subjective nature of “unequivocal” when used to describe the nuclear features of conventional PTC (cPTC) (i.e., nuclear clearing, overlapping, grooves, and pseudoinclusions) required to distinguish FVPTC from follicular adenoma particularly in cases where the lesion is encapsulated or sharply circumscribed rather than infiltrative into stroma or vascular channels and not associated with metastases (Daniels 2011). The roles of BRAF and RAS mutation analysis and of cytokeratin-19 (CK19), HBME-1 (HBME), and galectin-3 (Gal) immunostains as adjuncts in the diagnosis and management of follicular-patterned thyroid lesions also remain controversial. Currently, at our institution, FVPTC is usually managed like cPTC. However, recent studies indicating that encapsulated FVPTC (1) behaves like and therefore might be treated like follicular adenomas, while infiltrative FVPTC behaves like and should be treated like infiltrative cPTC and (2) exhibits lower BRAF mutation and higher RAS mutation rates than cPTC have raised the possibility that histopathology and/or mutation status might help personalize management regarding extent of surgery, intensity of follow-up, and targeted therapy in cases of FVPTC. This study was designed to correlate histological features, immunoreactivity for CK19, HBME, and Gal, and BRAF V600E mutation status with regional lymph node (LN) metastasis and follow-up in a histomorphologically welldefined cohort of FVPTC.
Materials and methods After IRB approval, our departmental files were searched for thyroid resections diagnosed as FVPTC with accompanying lymph node excisions. H&E-stained sections were reviewed by two pathologists (AW and SB). After all cases with >5 % papillary architecture and all cases with multiple tumors that had cPTC had been excluded, 48 FVPTC were available for study. The study set included 21 FVPTC with metastases to regional lymph nodes (LN+) and 27 FVPTC with negative regional lymph nodes (LN−) at thyroidectomy. Demographics, operative procedure, tumor focality (uni- versus multifocal), size (maximum diameter), circumscription (well circumscribed versus infiltrative), the presence of follicular architectural distortion, solid foci, lymphovascular invasion (LVI), extrathyroidal extension on microscopic evaluation, and resection margin status were recorded. Well-circumscribed tumors showed a sharply demarcated broad pushing border with or without an identifiable capsule, whereas infiltrative tumors showed tentacular growth into adjacent benign thyroidal parenchyma (Fig. 1a, b). Follicular architectural distortion was defined as elongated/oval follicles with branching, infoldings, and/or outpouchings (Fig. 1c). In cases with
13
Fig. 1 Representative hematoxylin- and eosin-stained sections of FVPTC showing a circumscribed border, b infiltrative border, and c follicular architecture distortion
multifocal tumor, the largest tumor nodule was selected for study. All lymph nodes excised had been entirely submitted for microscopic evaluation. Follow-up information was culled from our thyroid tumor registry and electronic medical records.
J Cancer Res Clin Oncol Table 1 Immunohistochemical staining Antibody
Antibody vendor
Antibody clone
Antibody dilution
Pretreatment
Instrument
Detection
HBME-1 CK19
Dako Cell Marque
HBME-1 RCK108
1:200 Predilute
Dako PT link-high pH On board pretreatment-high pH
Dako Autostainer Ventana Benchmark Ultra
Dako Envision+ Ventana Ultraview DAB
Galectin-3
Cell Marque
9C4
Predilute
On board pretreatment-high pH
Ventana Benchmark Ultra
Ventana Optiview DAB
Immunohistochemical staining Serial sections of a representative block from each formalin-fixed paraffin-embedded FVPTC were immunostained for CK19, HBME, and galectin-3. Immunohistochemical detection was performed on 4-μm tissue sections in accordance with the manufacturers’ recommendations using the antibodies, pretreatment, instrument, and detection shown in Table 1. All slides were run with positive and negative control tissues. Hepatocellular carcinoma and benign liver, mesothelioma, and papillary thyroid carcinoma and benign tonsil served as positive and negative controls for CK19, HBME, and Gal, respectively. The slides were subsequently counterstained with Mayer’s hematoxylin. The percentage of tumor cells exhibiting cytoplasmic and/or membrane staining with each of the three antibodies were semiquantitatively jointly evaluated by two pathologists (AW and SB) using a double-headed microscope. Interobserver differences were resolved by discussion until a consensus was reached.
BRAF V600E mutation detection Macrodissected formalin-fixed paraffin-embedded sections from 47 of the FVPTC were analyzed for BRAF V600E (1799T>A) mutation using real-time PCR. The DNA extracted from the remaining FVPTC was insufficient for mutation analysis. Briefly, the macrodissected tissues were deparaffinized, and DNA was extracted using the QIamp DNA Mini Kit (Qiagen, Valencia CA). DNA concentration was assessed by the Nanodrop spectrophotometer. PCR to detect BRAF V600E mutation was performed using Taqman Minor Groove Binders as follows: BRAF-F: 5′-CTACTGTTTTCCTTTACTTACTACACCT CAGA-3′, BRAF-R: 5′-ATCCAGACAACTGTTCAAACTGATG-3′, BRAF wild-type probe: 5′-VIC-CTAGCTACAGtGAAA TC-BHQ-3, BRAF mutant probe: 5′-FAM-TAGCTACAGaGAAATCBHQ-3′. Sample DNA and controls were adjusted to 5 ng/ul. BRAF V600E-positive DNA was purchased from Coriell
Cell Repository (HBT -38, Cat#4-088-2560, Coriell Institute, Camden, NJ). Mutation-negative DNA from Invivoscribe Technologies (IVS0000, Cat#4-092-0010, Invivoscribe Technologies, Inc., San Diego, CA) served as a negative control. Mutation-positive DNA was diluted in negative DNA to obtain the 50 % positive and 5 % positive (LOD) controls. Samples were cycled as follows: stage 1—50 °C for 10 min; stage 2—95 °C for 15 min; stage 3—50 cycles of 94 °C for 15 s and 60 °C for 1 min with data collection at the end of each cycle of stage 3. The threshold was set below the LOD and above the negative control in the BRAF mutation detector. Samples with BRAF mutant amplification Ct above the 5 % LOD were detected for BRAF mutation (BRAF DET). Samples with BRAF mutant amplification Ct
ORIGINAL ARTICLE – CANCER RESEARCH
Follicular variant of papillary thyroid carcinoma (FVPTC): histological features, BRAF V600E mutation, and lymph node status Ann E. Walts · James M. Mirocha · Shikha Bose
Received: 6 February 2015 / Accepted: 12 February 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose Follicular variant of papillary thyroid carcinoma (FVPTC) is currently treated like conventional papillary thyroid carcinoma (cPTC). Recent reports indicate that encapsulated FVPTC behaves like follicular adenomas, while infiltrative FVPTC behaves like cPTC. This raises the possibility that histology and/or mutation status might help personalize management of FVPTC regarding extent of surgery, intensity of follow-up, and targeted therapy. This study correlates histological features, immunoreactivity for CK19, HBME, and Gal, and BRAF V600E mutation with lymph node (LN) metastasis and follow-up in FVPTC. Methods Forty-eight FVPTC (21 with regional lymph node metastasis [LN+] and 27 with negative lymph nodes [LN−]) were reviewed. Demographics, tumor focality, size, circumscription, follicular architecture, lymphovascular invasion, extrathyroidal extension (ETE), and margin status were charted. Macrodissected formalin-fixed paraffin-embedded sections from 47 (21 LN+ and 26 LN−) cases were analyzed for BRAF V600E (1799T>A) mutation using real-time PCR. Correlations between the variables and LN status were calculated.
Presented in part at the United States & Canadian Academy of Pathology Annual Meeting, Baltimore, MD, March 2013. A. E. Walts (*) · S. Bose Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, CA 90048, USA e-mail: [email protected] J. M. Mirocha Biostatistics Core, Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA 90048, USA
Results Sixty-two percent of cases with ETE demonstrated LN metastasis, while 59 % of cases with circumscribed tumors were LN−. In multivariable analysis, ETE and tumor size ≥1 cm were the best predictors of LN+ status, whereas in cases without ETE, the infiltrative pattern and tumor size provided the “best fit.” Immunostains and BRAF mutation status were not helpful. All four tumors that recurred were LN+, with infiltrative borders, and lacked the BRAF mutation. Conclusions Tumor circumscription, extrathyroidal extension, and tumor size ≥ 1.0 cm are predictors of lymph node status in FVPTC. Keywords Follicular variant · Papillary thyroid carcinoma · BRAF V600E mutation · Extrathyroidal extension · Circumscription · Histological features
Introduction The follicular variant of papillary thyroid carcinoma (FVPTC) has presented challenges to pathologists and clinicians since the term was introduced (Hazard et al. 1955; Lindsay 1960, 1964) to describe tumors that “resemble papillary carcinoma in biologic behavior and all morphologic features with the exception that papillae are not present” (Chem and Rosai 1977). As subsequently defined by Livolsi (2003) and presented in the World Health Organization Classification of Tumors (DeLellis et al. 2004), FVPTC is composed “entirely or almost entirely of follicles lined by cells exhibiting nuclear features of papillary thyroid carcinoma.” Not surprisingly, the diagnosis and management of FVPTC have been confounded by the lack of evidence-based consensus guidelines regarding the extent of papillary growth “allowable” in FVPTC
13
J Cancer Res Clin Oncol
and the subjective nature of “unequivocal” when used to describe the nuclear features of conventional PTC (cPTC) (i.e., nuclear clearing, overlapping, grooves, and pseudoinclusions) required to distinguish FVPTC from follicular adenoma particularly in cases where the lesion is encapsulated or sharply circumscribed rather than infiltrative into stroma or vascular channels and not associated with metastases (Daniels 2011). The roles of BRAF and RAS mutation analysis and of cytokeratin-19 (CK19), HBME-1 (HBME), and galectin-3 (Gal) immunostains as adjuncts in the diagnosis and management of follicular-patterned thyroid lesions also remain controversial. Currently, at our institution, FVPTC is usually managed like cPTC. However, recent studies indicating that encapsulated FVPTC (1) behaves like and therefore might be treated like follicular adenomas, while infiltrative FVPTC behaves like and should be treated like infiltrative cPTC and (2) exhibits lower BRAF mutation and higher RAS mutation rates than cPTC have raised the possibility that histopathology and/or mutation status might help personalize management regarding extent of surgery, intensity of follow-up, and targeted therapy in cases of FVPTC. This study was designed to correlate histological features, immunoreactivity for CK19, HBME, and Gal, and BRAF V600E mutation status with regional lymph node (LN) metastasis and follow-up in a histomorphologically welldefined cohort of FVPTC.
Materials and methods After IRB approval, our departmental files were searched for thyroid resections diagnosed as FVPTC with accompanying lymph node excisions. H&E-stained sections were reviewed by two pathologists (AW and SB). After all cases with >5 % papillary architecture and all cases with multiple tumors that had cPTC had been excluded, 48 FVPTC were available for study. The study set included 21 FVPTC with metastases to regional lymph nodes (LN+) and 27 FVPTC with negative regional lymph nodes (LN−) at thyroidectomy. Demographics, operative procedure, tumor focality (uni- versus multifocal), size (maximum diameter), circumscription (well circumscribed versus infiltrative), the presence of follicular architectural distortion, solid foci, lymphovascular invasion (LVI), extrathyroidal extension on microscopic evaluation, and resection margin status were recorded. Well-circumscribed tumors showed a sharply demarcated broad pushing border with or without an identifiable capsule, whereas infiltrative tumors showed tentacular growth into adjacent benign thyroidal parenchyma (Fig. 1a, b). Follicular architectural distortion was defined as elongated/oval follicles with branching, infoldings, and/or outpouchings (Fig. 1c). In cases with
13
Fig. 1 Representative hematoxylin- and eosin-stained sections of FVPTC showing a circumscribed border, b infiltrative border, and c follicular architecture distortion
multifocal tumor, the largest tumor nodule was selected for study. All lymph nodes excised had been entirely submitted for microscopic evaluation. Follow-up information was culled from our thyroid tumor registry and electronic medical records.
J Cancer Res Clin Oncol Table 1 Immunohistochemical staining Antibody
Antibody vendor
Antibody clone
Antibody dilution
Pretreatment
Instrument
Detection
HBME-1 CK19
Dako Cell Marque
HBME-1 RCK108
1:200 Predilute
Dako PT link-high pH On board pretreatment-high pH
Dako Autostainer Ventana Benchmark Ultra
Dako Envision+ Ventana Ultraview DAB
Galectin-3
Cell Marque
9C4
Predilute
On board pretreatment-high pH
Ventana Benchmark Ultra
Ventana Optiview DAB
Immunohistochemical staining Serial sections of a representative block from each formalin-fixed paraffin-embedded FVPTC were immunostained for CK19, HBME, and galectin-3. Immunohistochemical detection was performed on 4-μm tissue sections in accordance with the manufacturers’ recommendations using the antibodies, pretreatment, instrument, and detection shown in Table 1. All slides were run with positive and negative control tissues. Hepatocellular carcinoma and benign liver, mesothelioma, and papillary thyroid carcinoma and benign tonsil served as positive and negative controls for CK19, HBME, and Gal, respectively. The slides were subsequently counterstained with Mayer’s hematoxylin. The percentage of tumor cells exhibiting cytoplasmic and/or membrane staining with each of the three antibodies were semiquantitatively jointly evaluated by two pathologists (AW and SB) using a double-headed microscope. Interobserver differences were resolved by discussion until a consensus was reached.
BRAF V600E mutation detection Macrodissected formalin-fixed paraffin-embedded sections from 47 of the FVPTC were analyzed for BRAF V600E (1799T>A) mutation using real-time PCR. The DNA extracted from the remaining FVPTC was insufficient for mutation analysis. Briefly, the macrodissected tissues were deparaffinized, and DNA was extracted using the QIamp DNA Mini Kit (Qiagen, Valencia CA). DNA concentration was assessed by the Nanodrop spectrophotometer. PCR to detect BRAF V600E mutation was performed using Taqman Minor Groove Binders as follows: BRAF-F: 5′-CTACTGTTTTCCTTTACTTACTACACCT CAGA-3′, BRAF-R: 5′-ATCCAGACAACTGTTCAAACTGATG-3′, BRAF wild-type probe: 5′-VIC-CTAGCTACAGtGAAA TC-BHQ-3, BRAF mutant probe: 5′-FAM-TAGCTACAGaGAAATCBHQ-3′. Sample DNA and controls were adjusted to 5 ng/ul. BRAF V600E-positive DNA was purchased from Coriell
Cell Repository (HBT -38, Cat#4-088-2560, Coriell Institute, Camden, NJ). Mutation-negative DNA from Invivoscribe Technologies (IVS0000, Cat#4-092-0010, Invivoscribe Technologies, Inc., San Diego, CA) served as a negative control. Mutation-positive DNA was diluted in negative DNA to obtain the 50 % positive and 5 % positive (LOD) controls. Samples were cycled as follows: stage 1—50 °C for 10 min; stage 2—95 °C for 15 min; stage 3—50 cycles of 94 °C for 15 s and 60 °C for 1 min with data collection at the end of each cycle of stage 3. The threshold was set below the LOD and above the negative control in the BRAF mutation detector. Samples with BRAF mutant amplification Ct above the 5 % LOD were detected for BRAF mutation (BRAF DET). Samples with BRAF mutant amplification Ct
