記住我
Abstract
Background: The term noninvasive follicular tumor with papillary-like nuclear features (NIFTP) was introduced as a surrogate for noninvasive encapsulated follicular variant of papillary thyroid carcinoma with a defined set of histopathologic criteria. There are very few studies depicting the cytological cues for the diagnosis of NIFTP. The objective of the study was to determine the spectrum of cytological features in fine needle aspiration cytology (FNAC) smears of cases histopathologically diagnosed as NIFTP. Methods: This was a retrospective cross-sectional study conducted over a duration of four years between January 2017 and December 2020. All surgically resected cases (n-21), who met the diagnostic criteria of NIFTP on histopathology and who underwent preoperative FNAC were included and reviewed in the study. Results: Out of a total of 21 cases, at FNAC, diagnosis of benign, suspicious for malignancy, follicular variant of papillary thyroid carcinoma, and classic papillary thyroid carcinoma (PTC) was rendered in 14 (66.6%), 2 (9.5%), 2 (9.5%), and 3 (14.28%), respectively. Scanty cellularity was noted in 12 (57.1%) cases. Papillae, sheets, and microfollicles were seen in 1 (4.7%), 10 (47.6%), and 13 (61.9%) cases, respectively. Nucleomegaly, nuclear membrane irregularities, nuclear crowding, and overlapping were seen in 7 (33.3%), 9 (42.8%), and 9 (42.8%), respectively. Nucleoli, nuclear grooving, and inclusions were seen in 3 (14.2%) 10 (47.6%), and 5 (23.8%) cases, respectively. Conclusion: At FNAC, NIFTP can be found in every category of The Bethesda System for Reporting Thyroid cytopathology (TBSRTC). Nuclear membrane irregularities, nuclear grooving, mild nuclear crowding, and overlapping were noted in a modest number of cases. However, the absence or rare occurrence of features such as papillae, inclusions, nucleoli, and metaplastic cytoplasm may help prevent an overdiagnosis of malignancy.
Keywords: Fine needle aspiration, noninvasive follicular tumor with papillary-like nuclear features, papillary thyroid carcinoma, thyroid
How to cite this article: Introduction 
Fine needle aspiration cytology (FNAC) with or without ultrasound guidance is an important tool for triaging patients with thyroid lesions.[1] The greatest advantage of thyroid FNAC is its ability to identify benign proliferations with good accuracy and its definitive role in the diagnosis of malignancies such as classic papillary thyroid carcinoma (PTC), medullary and anaplastic carcinoma. However, the major caveat of FNAC in thyroid lesions is the lesions in the intermediate category, especially follicular patterned lesions, for which it serves as a screening test.[2] In the last decade, the incidence of follicular variant papillary thyroid carcinoma (FVPTC) is increasing, most of which are diagnosed as indeterminate on cytology.[3] Follicular variant of papillary thyroid carcinoma comprises two subtypes, encapsulated FVPTC (E FVPTC) and invasive variant (FVPTC I), based on the absence or presence of capsular and/or vascular invasion, respectively. In a large series of 109 cases with a follow-up of 10 to 26 years, the E FVPTC showed an indolent course without recurrence or metastasis.[4] In view of these findings, an international expert panel proposed renaming this tumor as noninvasive follicular tumor with papillary-like nuclear features (NIFTP). Although, NIFTP was considered as an immediate precursor of FVPTC, it is now considered to be at the benign end of the spectrum of FVPTC and is characterized by complete encapsulation, the presence of nuclear features of PTC, the absence of psammoma bodies, true papillae, and solid areas on histopathology. The World Health Organization 2018 book on thyroid tumors has included NIFTP as a separate entity.
The change in the terminology of E FVPTC to NIFTP and the ensuing shift from the malignant category has raised several issues, the most significant being the incidence of NIFTP, the cytomorphological criteria for diagnosing NIFTP at FNAC, and the resulting change in Risks of malignancy (ROM) in TBSRTC categories.[5],[6] Since TBSRTC came before NIFTP, the category in which NIFTP cases should be classified remains contentious. Some studies state that NIFTP is most commonly diagnosed as category II, III, and IV.[7] In contrast, Hahn et al.[8] reported that 75% of NIFTP cases fall into the suspicious malignancy category. In terms of preoperative diagnosis, few studies advocate making the notion of rendering a diagnosis of NIFTP based on a few combinations of cytomorphonuclear features. However, Hirokawa et al.[9] pragmatically stated the difficulty of distinguishing NIFTP from FVPTC I on cytology with caution for pathologist with a zest of preoperative diagnosis of NIFTP. Few studies have attempted to determine the cytomorphologic features that best describe NIFTP at FNAC.
The present study was planned with the aim to determine the spectrum of cytological features in FNAC smears of cases diagnosed as NIFTP on histopathology.
Materials and Methods This was a retrospective cross-sectional study. Institutional ethical committee clearance was obtained before conducting the study. All cases strictly following the inclusion criteria and exclusion criteria for NIFTP and in whom preoperative FNAC was done were included in the study. The study duration was four years, from January 2017 to December 2020. The sampling method adopted was purposive sampling.
For NIFTP, diagnosis following inclusion and exclusion criteria were applied as stated in the literature as follow: an encapsulated tumor, showing follicular patterned growth with no well-formed papillae, no psammoma bodies, <30% solid, trabecular, or insular growth pattern. The cells should have nuclear features of papillary carcinoma of moderate grade (nuclear score 2). There should be no capsular or vascular invasion, tumor necrosis or high mitotic activity.
Cases in which smears demonstrated faded staining were excluded from the study.
Selection of cases. There were a total of 55 cases diagnosed as PTC on histopathology during the four-year study period. These included classic PTC, follicular variants of PTC, and well-differentiated tumor with uncertain malignant potential. Among these, after reviewing, those cases in which the diagnosis strictly met the inclusion and exclusion criteria for NIFTP and in whom preoperative FNAC was available accounted for 21 cases, which formed the study population.
Demographic details along with ultrasound findings were obtained from the case records. Details of local examination findings like the size, side, solitary, or diffuse involvement, consistency, signs of inflammation, and presence or absence of lymph nodes were noted from the records. The routine protocol followed for FNAC in our institute, which is applicable for cases in the present study is as follows; FNAC of thyroid was performed using a 24-gauge needle without aspiration technique. If the initial aspirate was colloidal or fluid, the entire fluid was aspirated and FNAC was done from the residual solid area. The aspirated fluid was centrifuged and sediment smears were prepared and processed. Normally, multiple site aspirations were performed for large nodules. In case of incidentally detected thyroid swellings on ultrasound, complex swellings and lesions where initial FNAC was negative due to low cellularity or cyst fluid-only specimens, FNAC was performed under ultrasound guidance. The FNAC procedure was performed by a cytopathologist. Fifty percent of the smears were fixed in 95% ethyl alcohol for Haematoxylin and Eosin and Pap staining and 50% were air dried for Giemsa staining.
The cytology slides were collected and the diagnosis was reviewed by two experienced pathologists (Dr AK and Dr SM) with more than ten years of experience in cytopathology. The cellularity of the smears, the amount of colloid and cytological details, both architectural and nuclear features, were noted in a predetermined format. Features in fine needle aspiration cytology smears of all 21 cases were classified into different diagnostic categories of TBSRTC after review.
The type of surgery performed was noted. The gross details like the size of the nodule and focality were noted from the case records. The histopathology of all the 21 cases was reviewed by two experienced pathologists, who were blinded for the cytological diagnosis, and the architectural and nuclear features were noted in the given format.
Before reviewing the cases, all the cytology smears and histopathology sections were coded by a pathologist not involved in the study.
For FNAC, cellularity; sparse or moderate/abundant, architectural features like pseudo papillae, crowded clusters/sheets and/or microfollicles, nuclear features such as nucleomegaly, nuclear membrane irregularity, nuclear grooving, intranuclear cytoplasmic inclusions, nucleoli, nuclear crowding and overlapping, cyst macrophages, metaplastic cytoplasm, and psammoma bodies were recorded for each case.
Statistical analysis was performed using Microsoft Excel 2011. Standard descriptive analysis was performed. Quantitative data was expressed as mean, standard deviation, percentages, and proportions. Comparison of the different cytodiagnostic categories of the present study was compared with the data available in the existing literature. Similarly, the cytological details were also compared with that reported in the existing literature.
Results All cases strictly followed the inclusion and exclusion criteria for NIFTP and had an appropriate FNAC over a period of four years from January 2017 to December 2020 were included in the study. Of the 55 cases histopathologically diagnosed as PTC during the study period, 21 cases fulfilled the criteria of NIFTP and these cases also had FNAC at our Institute. Therefore, the total sample size of the study was 21 cases. The demographic details are shown in [Table 1]. The mean age at presentation was 41 ± 3.2 years. The female-to-male ratio was 1:0. All the 21 cases (100%) presented had a solitary thyroid nodule on clinical examination. Out of 21 cases, the maximum number of cases (13/21,61.9%) underwent total thyroidectomy, four (19%) cases underwent hemithyroidectomy, two cases (9.5%) underwent lobectomy, while two cases (9.5%) underwent radical neck dissection as part of surgical treatment. Most of the lesions were unifocal (18/21,85.7%) and the mean size of the nodule was 2.5 ± 0.2 cm. The distribution of cases into TBSRTC is shown in [Table 2].
Table 2: Distribution of all NIFTP cases in the TBSRTC categories of NIFTPThe distribution of NIFTP cases in TBSRTC categories is depicted in [Table 2]. Out of total 21 cases, at FNAC, diagnosis of benign, suspicious for malignancy, follicular variant of papillary thyroid carcinoma (FV-PTC), and classic PTC was rendered in 14 (66.6%), 2 (9.5%), 2 (9.5%), and 3 (14.28%) cases, respectively.
There were no cases diagnosed in atypical category.
The cytomorphological features of NIFTP cases is summarized in [Table 3]. Scanty cellularity was noted in 12 (57.1%) cases. Most of the cases with scanty cellularity were diagnosed as benign (n-9), followed by suspicious for follicular neoplasm (n-2), and suspicious for malignancy (n-1). Papillae, sheets, and microfollicles were seen in 1 (4.7%), 10 (47.6%), and 13 (61.9%) cases, respectively. Nucleomegaly, nuclear membrane irregularity, nuclear crowding and overlapping, nucleoli, nuclear grooving, and inclusions were seen in 7 (33.3%), 9 (42.8%), 9 (42.8%), 3 (14.2%), 10 (47.6%), and 5 (23.8%), respectively. The representative Bethesda category-wise cytology images of histopathologically confirmed cases of NIFTP have been depicted in [Figure 1], [Figure 2], [Figure 3], [Figure 4].
Figure 1: a and b: Shows cytology smears of histopathology-proven NIFTP which was diagnosed as benign at cytology. (May Grunwald Giemsa, ×100). Figure C shows the histopathology with features satisfying the criteria of NIFTP (Haematoxylin and Eosin, ×100)
Figure 2: a and b: Shows cytology smears of histopathology-proven NIFTP which was diagnosed as follicular neoplasm/suspicious for follicular neoplasm at cytology. (Haematoxylin and Eosin, ×100)
Figure 3: a and b: Shows cytology smears of histopathology-proven NIFTP which was diagnosed as suspicious of malignancy at cytology. [(Haematoxylin and Eosin, ×200 (a) (Haematoxylin and Eosin, ×100) (b) ]
Figure 4: a: Shows cytology smears of histopathology-proven NIFTP which was diagnosed as papillary thyroid carcinoma. Figure b shows the histopathology with features satisfying the criteria of NIFTP (Haematoxylin and Eosin, ×400) Discussion Distinguishing NIFTP from classic PTC is of paramount importance as the latter putatively needs total thyroidectomy. The preoperative diagnosis of NIFTP warrants excision to rule out invasion and infiltration, albeit requiring a conservative approach. Furthermore, radioactive iodine or further correction surgeries are not required.[10]
Few papers have been published to answer the queries related to NIFTP and to identify the best combination of features that can accurately predict the diagnosis of NIFTP in cytology. The most significant and earliest study in this regard was that of Nikiforov et al.[4] who proposed scoring based on nuclear and cytological features.
Maletta et al.[11] retrospectively analyzed the FNAC findings of 96 histologically proven cases of NIFTP, focusing on nuclear and cytoplasmic features that allow accurate prediction of NIFTP at cytology. Among all the nuclear features studied, authors observed the best concordance for nuclear membrane irregularity while stating nuclear size enlargement as the least reliable criterion. Similar findings were observed in the present study wherein nuclear membrane irregularity was seen in 42% of cases, whereas increase of nuclear size was observed in only 7% cases. However, Yan le et al.[12] found that nucleomegaly was present in all 14 cases studied. In contrast to the above observations, Bizzarro et al.[13] reported that the nuclear size was significantly smaller in NIFTP as compared to follicular variant of papillary thyroid carcinoma - invasive (FVPTC I).
In the present study, NIFTP was the most frequently clustered in Bethesda category II (66.6%). Our findings corroborated with the findings of Ibrahim et al.[14] Nonetheless, there are studies which state that a significant number of NIFTP cases are reported to fall under atypia of undetermined significance, suspicious of follicular neoplasm and non-diagnostic category.[11],[15],[16],[17],[18] Pressure from the clinician to provide a definite diagnosis and thereby deliberate attempt by the cytopathologist to place the lesion in a specific category might have resulted in no cases being diagnosed in the atypical category in the present study. Surprisingly, even after the review of slides, no cases were placed in the Atypia of undetermined significance (AUS) category. Though studies from Western countries have reported higher percentage of NIFTP that are preoperatively diagnosed as malignant (7 to 28%), it is interesting and imperative to note that the percentage in Asian countries is quite low ranging from 0 to 4.7%.[5],[9],[16] In the present study, NIFTP was diagnosed preoperatively as PTC in 14.28% cases, which is slightly higher than the cases reported in the previous studies. In a recent study by Hirokawa et al.[9] from Japan, 75% of NIFTP cases were placed in suspicion for malignancy or malignant category. In the present study, 19% of the cases were reported suspicious of malignancy. Comparison of NIFTP cases in various diagnostic categories with that in the literature is shown in [Table 4].
Table 4: Comparison of NIFTP cases in various diagnostic categories with that in the literatureThe threshold for nuclear features to segregate FVPTC from NIFTP is a matter of contention. Literature reveals considerable geographical variations among the expert pathologist for the reproducibility of the cytological features.[19],[20] Due to enormous disputation, a simplified scoring system has been introduced that takes nuclear size and shape, nuclear membrane irregularity, and nuclear chromatin into consideration, with each parameter acquiring a score of one, if present distinctively. A score of two or three entails a lesion as NIFTP.[21] The nuclear features when present diffusely would leverage the diagnosis toward NIFTP over hyperplastic nodule or goiter.
Studies have compared cytological features of NIFTP with conventional PTC as well as with E FVPTC and FVPTC I.[11] In the present study, lesions other than NIFTP diagnosis at histopathology were not included in the study. The dominant cytological pattern emerging out of these studies in NIFTP cases included a microfollicular pattern of arrangement, nuclear crowding and overlapping, irregular nuclear membrane, and powdery chromatin.[21] In NIFTP, the nuclear and cytological features are not as well defined as in classical PTC, while it is practically impossible to distinguish E FVPTC and FVPTC I in cytology. However, in a study by Ibrahim et al.,[14] the authors reported that high cellularity and obvious nuclear features favored a diagnosis of FVPTC I in contrast to E FVPTC. The presence of true papillae, intranuclear inclusions, and psammoma bodies slant the diagnosis toward the classic PTC. In the present study, microfollicular pattern was seen in (61.9%) of the cases, in accordance with that reported in the literature.[7],[12] Pseudo papillae were found in 4.7% (1/21) of cases in the present study. Yan et al.[12] and Brandler et al.[7] also noted papillae in two and three cases, respectively. The macro follicles when aspirated with the needle and spread on the slide without being fragmented may resemble a large sheet with an anatomical margin that can mimic a papilla. Cytopathologists must be cautious of this finding and the presence of occasional papillae should be considered in the context of the overall smear morphology. In the present study, nuclear membrane irregularity and nuclear crowding and overlapping were noted in (42.8%) of the cases, in concordance with the studies reported in literature.[22] Among the various cytoarchitectural, nuclear, and cytological features analyzed by Strickland et al.,[17] microfollicular pattern and nuclear membrane irregularity were the most reliable to distinguish classic PTC from NIFTP. In another study by Strickland et al.,[21] the authors proposed that nuclear features could not differentiate NIFTP from FVPTC, while the nuclear score varied significantly between NIFTP and benign nodules, confirming its role in identifying NIFTP with confidence at FNAC.
Nuclear grooving was noted in 47.6% of cases in the present study. Nuclear grooving, stand-alone is not a reliable criterion for the diagnosis of PTC, as other lesions like Hashimoto's thyroiditis and goiter show the presence of nuclear grooves. Several authors have reported the insignificant role of nuclear grooves in the diagnosis of NIFTP at cytology. However, Hirokawa reported nuclear grooves to be present in 96.4% of cases of NIFTP.[9]
Intranuclear pseudo inclusions (INPI) have always been considered as an important finding in the diagnosis of PTC with DeMay Richard et al.[23] suggesting that the presence of even one INPI would suggest a diagnosis of PTC. Chandler et al. proposed scoring of INPI and noted that more than INPI would suggest NIFTP. Intranuclear pseudo inclusions are distinctively absent in NIFTP.[17],[24]
It is important that the nuclear features deemed important for recognizing NIFTP should be viewed in the context of the surrounding thyroid since the processing artifact, slide preparation and staining, and the underlying thyroid condition may alter the nuclear features. Comparison of various cytological features of NIFTP cases with that in the literature is shown in [Table 5].
Table 5: Comparison of various cytological features of NIFTP cases with that in the literatureThe incidence of NIFTP shows wide regional variation. Hahn et al.[8] reported an incidence of 15–28% in western countries[11],[25],[26], while studies from countries of Asian origin show a low incidence of NIFTP of 1.5%.[16] In the present study, the incidence of NIFTP was 38.1%. Disagreement over the nuclear and other features of FVPTC among the expert pathologists across different regions have been reported in the literature. Likewise, even the methodology of sample collection, the retrospective nature of the study, and the inclusion and exclusion criteria deployed all account for the discrepancies in the incidence rates of NIFTP reported worldwide.[9],[10]
In the present study, 57.1% cases of NIFTP had low cellularity. Similar results were reported by Maletta et al.[11] Possible explanation for scant cellularity in the smears could be the inherent nature of the lesions. Yang GCH et al.[27] reported low cellularity in the majority of the NIFTP cases studied while reporting hypercellularity in only 5.6% of cases. In contrast to these findings, they demonstrated high cellularity in smears of E FVPTC and FVPTC I.
In the present study, 13/21 cases (61.9%) underwent total thyroidectomy despite the fact that most of the lesions (14/21, 66.6%) were diagnosed as benign at FNAC. Among the 13 cases which underwent total thyroidectomy, seven cases (7/13, 53.85%) were placed in suspicious or malignant category at FNAC, for which total thyroidectomy was justifiable. In the remaining six cases (6/13,46.1%), total thyroidectomy was performed where the initial FNAC diagnosis was benign. The decision to perform total thyroidectomy in the cases diagnosed as benign on cytology was based on the clinicians' decision, taking into consideration the disease characteristics, ultrasound features (presence of nodules in both the lobes), patients' socioeconomic background as well as the patients' informed opinion. One of the reasons was the need to avoid reoperation in cases of hyperthyroidism to eliminate the risk of malignant transformation due to radioiodine therapy administered to these patients and also in cases of large nodular goiters. Similarly, the fact that the institute where this study was conducted, is a rural setup, wherein patients follow up after surgery is a challenge, a total thyroidectomy was considered in cases diagnosed as benign at FNAC, based on informed decision by the patient.
With the introduction of NIFTP, a significant change in ROM was observed in the different TBSRTC categories. Several individual studies and metanalysis have shown that, when NIFTP is placed in benign category, there is reduction in ROM from category II to V.[28] In a multiinstitutional study in Asia, the only category which showed a relative reduction in ROM was follicular neoplasm/suspicious for follicular neoplasm (24.4%). There was no significant change in the risk of malignancy in the other categories. Similarly, the significant finding that emerged from this study was the absence of a statistically significant difference between the reduction in ROM in categories with or without NIFTP. Kim et al.[15] reported a slight decrease in ROM in all categories but the decrease was not statistically significant. In a systematic metanalysis by Bongiovanni et al.,[29] the introduction of NIFTP lead to a decrease in ROM in category V and VI by 14% and 3%, respectively. Most significantly, studies demonstrated a reduction in ROM in suspicious malignancy category from 23.4% to 41.5%.[5],[28] In the present study, the ROM was decreased in category V.
Conclusion At cytology, NIFTP can be found in any category of TSBRTC. Most NIFTPs were not diagnosed as PTC on thyroid FNAC. In the present study, NIFTPs were most frequently categorized as benign. Although nuclear membrane irregularity, nuclear grooving, and mild nuclear crowding and overlapping were noted in a modest number of cases, the absence or rare occurrence of features like pseudo papillae, inclusions, nucleoli, and metaplastic cytoplasm could help prevent overdiagnosis of malignancy.
Acknowledgments
Authors acknowledge the support provided by Dr MG Shivaramu, Principal AIMS, Dean, Health Sciences (Medical), Adichunchanagiri Institute of Medical Sciences, Adichunchanagiri University, BG Nagara for the support provided.
Statement of ethics
Informed consent was obtained from the patients. The study was approved by the Institutional Ethical Committee, AIMS, ref no IEC/AIMS/232.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References 
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/joc.joc_192_21
留言 (0)