The Atypical and Suspicious for Malignancy Categories of the WHO Reporting System for Lymph Node, Spleen, and Thymus Cytopathology: Review of their Diagnostic Utility, Limitations, and Clinical Impact

ABSTRACT

Background

Fine‐needle aspiration biopsy (FNAB) is the preferred first‐line diagnostic tool for evaluating lymphadenopathy due to its minimally invasive nature and cost‐effectiveness. However, cytopathological interpretation of lymph node FNAB remains challenging because of the wide morphological spectrum of lymphoid lesions. The World Health Organization (WHO) Reporting System for Lymph Nodes, Spleen, and Thymus Cytopathology (WHO System) incorporates the “Atypical” and “Suspicious for malignancy” (“SFM”) categories to manage diagnostic uncertainty, yet their reproducibility and clinical implications remain undetermined.

Methods

This review with illustrative case‐based applications of the WHO System discusses the use, diagnostic performance, and limitations of the “Atypical” and “SFM” categories. Comparative analysis of published data, including interobserver concordance and risk of malignancy (ROM) studies, has been performed, with reference to the Sydney System and the multicenter DELYCYOUS study.

Results

Cases categorized as “Atypical” demonstrate highly variable ROM (28.6%–76.9%, mean ≈65%), reflecting heterogeneity in application and ancillary test integration, whereas the “SFM” category consistently exhibits high ROM (82%–100%, mean ≈95%). Interobserver agreement remains poor (κ = 0.075 for “Atypical”; κ = 0.104 for “Suspicious for malignancy”), highlighting interpretive subjectivity. Diagnostic pitfalls include artifactual monomorphism mimicking high‐grade lymphoma and underdiagnosis of paucicellular Hodgkin lymphoma or T‐cell proliferations. Ancillary tests, while essential, may also yield misleading results—such as monoclonal B‐cell populations in reactive conditions—potentially leading to overinterpretation.

Conclusions

The “Atypical” and “SFM” categories are indispensable for diagnostic stratification and clinical management but suffer from limited reproducibility and intrinsic ambiguity. Their optimal application requires standardized diagnostic criteria of specific lesions, which are provided in the WHO System, institutional consistency, and integrated use of cytomorphological, immunophenotypical, and molecular data to minimize diagnostic uncertainty and improve patient outcomes.

1. Introduction

Fine Needle Aspiration Biopsy (FNAB) is widely recognized as an accurate, minimally invasive, and cost‐effective method for diagnosing enlarged lymph nodes in different clinical settings [1, 2, 3, 4]. Despite its well‐established clinical utility, the broad pathological spectrum of lymph node lesions, including benign, infectious, and malignant conditions, continues to pose significant diagnostic challenges. In particular, the cytomorphological overlaps between resting lymph nodes and small cell lymphomas (including chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL), mantle cell lymphoma (MCL), lymphoplasmacytic lymphoma (LPL), and low‐grade follicular lymphomas (FL) where small centrocytes predominate), and between reactive follicular hyperplasia and follicular lymphoma with a more prominent mixed cell type, remain major limitations [5]. Furthermore, these cytomorphological and interpretative difficulties have historically been exacerbated by the lack of standardized reporting systems that ensure diagnostic consistency and clinical reliability for lymph node cytopathology.

To address these challenges and to standardize the performance, the Sydney System for the Performance, Classification and Reporting of Lymph Node FNAB was introduced, providing a structured five‐tier diagnostic framework with standardized terminology, defined risks of malignancy (ROM), and clear recommendations for further diagnostic management [6]. This system was refined and expanded within the World Health Organization (WHO) Reporting System for Lymph Nodes, Spleen, and Thymus Cytopathology (WHO System), establishing an international and globally applicable classification [7].

The WHO System stratifies lymph node FNAB results into five diagnostic categories: “Inadequate/Insufficient/Non‐Diagnostic”, “Benign”, “Atypical”, “Suspicious for malignancy” (“SFM”), and “Malignant”. Among the categories defined by the WHO system, the “Atypical” and “SFM” categories occupy critical positions in the diagnostic spectrum, bridging benign and malignant interpretations. They represent key interpretive “gray zones” and their application reflects both the inherent limitations of lymph nodal cytopathology and the variable access to and integration of ancillary testing such as flow cytometry, immunocytochemistry, and molecular studies.

This review is based on a focused and critical appraisal of the published evidence, integrated with expert experience, concerning the diagnostic performance, reproducibility, and clinical impact of the WHO “Atypical” and “SFM” categories in lymph node cytopathology. Particular emphasis has been placed on the most recent literature, especially studies providing histopathological follow‐up, robust estimates of risk of malignancy, and data on interobserver reproducibility.

2. Comparative Diagnostic Features

The WHO System defines the “Atypical” category as follows: “A specimen classified as ‘Atypical’ shows predominantly benign cytomorphological findings with minimal features that may raise the possibility of a malignant lesion, but is insufficient, either in quantity or quality, for a definite benign or malignant diagnosis” [7].

In line with this definition, the “Atypical” category is used for specimens that are largely consistent with benign or reactive processes but exhibit some abnormalities that prevent a confident exclusion of malignancy. These may include: a monotonous lymphoid population with mild nuclear enlargement or subtle nuclear irregularities where the differential diagnosis often lies between a lymph node in a resting phase and a small cell lymphoma; or a mixed lymphoid population with both small and large cells including centrocytes and centroblasts where the differential diagnosis includes a reactive follicular hyperplasia and follicular lymphoma. In these situations, ancillary techniques, including flow cytometry, immunocytochemistry and molecular assays, are typically required to clarify the interpretation [8, 9].

The “Atypical” category also encompasses cases in which the cytopathological material is impacted by technical limitations such as low cellularity, poor smearing, inadequate preparation, or poor‐quality staining. These factors often result in an indeterminate interpretation. Similarly, the presence of a few atypical, possibly non‐lymphoid, cells within a lymphoid population or in a paucicellular background may be difficult to interpret and can pose significant challenges for clinical management of possible metastatic disease, in the absence of proper clinical context such as prior history of malignancy.

In contrast, the WHO System defines the “SFM” category as follows: “A specimen that demonstrates some cytopathological features suggestive of malignancy but insufficient—either in quantity or quality—to make an unequivocal diagnosis of malignancy” [7]. A specimen is therefore categorized as “SFM” when cytopathological features are highly suggestive of malignancy, yet the findings remain inconclusive. A definitive diagnosis is often precluded by limited cellularity, suboptimal sample quality, or non‐diagnostic ancillary studies. Cytopathological features in this category may include pronounced cellular atypia or other findings strongly concerning for lymphoma—such as single markedly atypical cells resembling Hodgkin cells—or metastatic disease, but without ancillary evidence sufficient to confirm malignancy.

Both the “Atypical” and “SFM” categories play critical but distinct roles in clinical management of lymphadenopathy assessed by FNAB, and this is demonstrated by the clear hierarchical gradient in the ROM that separates these two categories, confirming their usefulness to clinicians in establishing appropriate further diagnostic management of patients. However, reported ROM values for the “Atypical” category vary widely, ranging from 28.6% to 76.9% [10, 11, 12, 13, 14, 15, 16, 17, 18] (mean ≈65%) [19, 20], reflecting variability in the application of the category across institutions, patient cohorts, and availability and use of ancillary testing [19]. “SFM” typically demonstrates ROMs between 82% and 100% [10, 11, 12, 13, 14, 15, 16, 17, 18] (mean ≈95%) [19, 20] (Table 1).

TABLE 1.

Risk of malignancy (ROM) according to WHO category.

WHO category	Level of suspicion	ROM range (%)	Mean ROM (%)	Notes
Atypical	Indeterminate; features overlap with reactive or benign processes	28.6–76.9	65	Shows wide inter‐institutional variability; ROM may be influenced by patient cohorts, rapid on‐site evaluation (ROSE) and ancillary studies
Suspicious for malignancy	Strongly suggestive of malignancy; warrants urgent diagnostic confirmation	82–100	95	Very high ROM; typically, prompt immediate tissue biopsy or surgical management

These two categories are distinguished by the extent of cytopathological atypia and presence or absence of conclusive ancillary studies. “Atypical” denotes uncertainty as to whether the lesion represents a lymphoproliferative lesion or metastatic disease, while “SFM” implies a higher probability of malignancy, with a definitive diagnosis limited mainly by insufficient or indeterminate sampling and inconclusive ancillary testing. In this context, it is important to bear in mind that the “Atypical” category should be used when the cytopathologist is inclined toward a benign diagnosis but hesitates for some reason. The use of this category will reflect the experience of the cytopathologist, the local practice of cytopathology, the availability of ancillary testing to clarify a specific diagnosis, and the support of clinicians within an active multidisciplinary setting to correlate history, clinical findings and imaging. The WHO System emphasizes that this category should be used sparingly and not as a waste basket, and the reasons a case is categorized as “Atypical” should always be clearly stated in a report. The use of the “SFM” category similarly should include a statement as to which malignancy is suspected but inherently will have a higher ROM based on the cytopathological features present.

3. Clinical Utility

Despite interpretive challenges, the “Atypical” and “SFM” categories play a pivotal role in clinical decision‐making by prompting additional diagnostic investigations. Their primary value lies in guiding, firstly the decision to obtain further tissue samples by FNAB, core needle biopsy (CNB) or surgical biopsy when diagnostic uncertainty persists, and secondly the selection of appropriate ancillary tests. FNAB can yield material suitable for the full range of ancillary techniques applicable to CNB and excision biopsy, including flow cytometry, immunocytochemistry, and molecular testing, each offering distinct advantages depending on the clinical context and cytomorphological features. Flow cytometry allows assessment to demonstrate monomorphic lymphoid populations and B‐cell clonality through light chain restriction and surface marker profiling, thereby facilitating the distinction between reactive proliferations and B‐cell lymphomas [21, 22, 23]. The choice and extent of ancillary testing should be tailored to the specific diagnostic context and to the availability and results of flow cytometry. When high‐quality flow cytometric data are available and demonstrate a clearly reactive or on the other hand a clonally defined lymphoid population, the need for extensive immunocytochemical panels may be reduced and limited to confirmatory or subtype‐specific markers, for example cyclinD1 when cytopathology and flow cytometry have suggested mantle cell lymphoma. In contrast, when flow cytometry is inadequate, inconclusive, or unavailable—particularly when a cell block represents the main diagnostic substrate—a broader immunocytochemical panel is warranted. Marker selection should remain morphology‐driven and clinically oriented.

For B‐cell processes in the absence of flow cytometry, an extended immunocytochemistry panel is recommended to include CD20, CD3, CD5, CD10, CD23, kappa/lambda, cyclin D1, and Bcl2. For T‐cell processes, panels can include CD3, CD5, CD7, CD2, CD4, CD8, and markers for T‐follicular helper cells (TFH) (CD10, Bcl6, PD1, CXCL13, ICOS) if TFH origin is suspected [24, 25]. Additional markers such as CD30, ALK, Bcl2, Bcl6, cyclin D1, TIA‐1, granzyme B, and Ki67 are used based on cytomorphological features which raise specific suspicions [26]. Immunocytochemistry is also particularly helpful for identification of metastatic disease [27].

Molecular studies including fluorescence in situ hybridization (FISH), polymerase chain reaction (PCR) for gene rearrangement, and next‐generation sequencing (NGS) enable detection of specific genetic abnormalities, including immunoglobulin heavy chain (IgH) or T‐cell receptor (TCR) rearrangements, which assist in diagnosing lymphoproliferative disorders when immunophenotypic data are inconclusive [28, 29, 30]. Detection of specific translocations such as t(14;18) for follicular lymphoma, t(11;14) for mantle cell lymphoma, and t(2;5) for ALK‐positive anaplastic large cell lymphoma by FISH or PCR can further refine diagnosis and subtype classification. NGS may be considered if standard techniques are inconclusive and suspicion for a clonal process remains [31]. These ancillary studies can also identify molecular alterations characteristic of metastatic disease, such as EGFR mutations for lung and other cancers or other specific mutations that can be helpful for targeted therapy [32].

When ancillary studies remain inconclusive or the sample proves inadequate, current recommendations for these WHO System diagnostic categories support repeating a FNAB with rapid on site evaluation (ROSE), CNB, or excisional biopsy to achieve diagnostic resolution.

4. Diagnostic Challenges

The main limitation of these categories is their poor diagnostic reproducibility. The DELYCYOUS study on the Sydney System revealed low interobserver concordance: Fleiss' κ = 0.075 for ALUS/AUS category and κ = 0.104 for the Suspicious category. These contrasted sharply with the high agreement for the definitive categories of Non‐diagnostic, Benign and Malignant [33]. This reflects the inherent subjectivity in evaluating borderline cases, particularly when ROSE or ancillary data are unavailable.

Too much reliance on cytomorphology can introduce interpretive pitfalls. Reactive lymphoid hyperplasia may raise the suspicion of a high‐grade lesion when large immunoblasts are prominent and cluster due to spreading artifacts (“concentric oval” effect) [34] simulating lymphoma and leading to inappropriate “Atypical” or “SFM categorization”. The monotonous appearance of small lymphoid cells may also raise the possibility of a small cell lymphoma, such as CLL/SLL or MCL. Conversely, classical Hodgkin lymphoma, especially the nodular sclerosis subtype, may be underdiagnosed because of low cellularity or paucity of diagnostic Hodgkin or Reed–Sternberg cells, occasionally resulting in categorization as “Benign” or “Atypical” [35].

False‐negative results may also arise in T‐cell lymphomas, such as peripheral T‐cell lymphoma (PTCL) or nodal T follicular helper cell lymphoma (TFHCL), angioimmunoblastic type, which sometimes exhibit cytomorphological overlap with reactive lymphadenopathy. Flow cytometry has historically shown lower specificity in the evaluation of T‐cell lymphoproliferative disorders compared with B‐cell lymphomas, but with the recent introduction of TRBC1 and TRBC2 antibodies, its specificity in assessing T‐cell proliferations has significantly improved [36, 37]. Atypical T‐cell populations can be seen in reactive conditions prompting the application of ancillary studies such as flow cytometry or PCR to exclude monotypic T‐cell populations or monoclonal T‐cell receptor gene rearrangements, respectively [24]. Among B cell lymphoproliferative disorders, potential flow cytometry pitfalls include finding small monoclonal B‐cell populations detected by flow cytometry in reactive conditions, such as Hashimoto thyroiditis and AIDS‐related lymphadenopathy, which may lead to overinterpretation as “SFM” or “Malignant” [38].

5. Illustrative Clinical Cases

Cases categorized as “Atypical” and “SFM” are generally characterized by suboptimal sampling, poor smear quality including issues with preparation, distribution, and fixation, and limited cellular material, often precluding the use of ancillary studies.

However, in clinical practice, the range of cases that may fall into these two categories is considerably broader, even when direct smears are of good quality and sufficient cytopathological material is available for ancillary techniques. The following examples illustrate this wide diagnostic spectrum:

5.1. Case 1—Atypical Category (Possible Reactive Process)

A 58‐year‐old man with a prior history of melanoma of the back presented with unilateral swelling and edema in the left inguinal region of uncertain etiology. Ultrasound examination demonstrated an oval‐shaped lymph node, approximately 20 mm, with cortical thickening but preserved fatty hilum, findings that raised the possibility of a reactive or inflammatory process.

A US‐guided FNAB was performed, ROSE confirmed sample adequacy, and two additional passes were obtained. The Giemsa‐stained smears revealed a dispersed lymphoid population with scattered cellular fragments composed of blood vessels in lymphoid tangles. The lymphoid population was polymorphic, consisting predominantly of small mature lymphocytes, immunoblasts, and several histiocytes with cytoplasmic tingible bodies (Figure 1A). Numerous eosinophils were also identified throughout the background (Figure 1B).

FIGURE 1.

Case 1—Atypical Category (possible reactive process). (A, B) Smear preparations show a polymorphic lymphoid population with scattered tingible body macrophages and numerous eosinophils (orange arrows). The background is rich in small mature lymphocytes with intermixed larger activated forms (May–Grünwald–Giemsa stain, 40X). (C) Flow cytometric analysis was performed by gating the CD45‐positive leukocyte population (P3). Within this gate, CD19‐ and CD3‐expressing cells were evaluated (data not shown). The CD19‐positive B‐cell population accounts for 52% of total cellularity. Two distinct CD19‐positive subpopulations with different fluorescence intensities are identified. In the CD19 versus SSC‐A dot plot, the population gated as P7 demonstrates lambda light chain restriction, consistent with a clonal B‐cell expansion. In contrast, the CD19‐positive cells within the P4 gate show a polyclonal immunophenotype, expressing both kappa and lambda light chains. [Color figure can be viewed at wileyonlinelibrary.com]

The Papanicolaou‐stained smears showed similar features. Cell‐block material was non‐contributory. No cytopathological features suggestive of metastatic melanoma were observed. Although the presence of eosinophils could raise concern for Hodgkin lymphoma, no diagnostic Reed–Sternberg or Hodgkin cells were identified. The overall cytomorphological pattern was more consistent with dermatopathic lymphadenitis or reactive lymphoid hyperplasia with eosinophilia.

One of the FNAB was suspended in phosphate‐buffered saline (PBS) for flow cytometric analysis, which revealed a predominantly polyclonal background with a small lambda‐restricted CD19‐positive B‐cell population, accounting for approximately 10% of all cells (Figure 1C). This was interpreted as a non‐diagnostic finding, possibly reactive, insufficient on its own to warrant categorization as “SFM” but falling within the interpretive spectrum of the “Atypical” category of the WHO System.

The integrated interpretation, considering ultrasound, cytomorphology, and flow cytometry, favored a reactive process, and the case was therefore classified as “Atypical” as a probable reactive process. The report recommended clinical and imaging correlation, close follow‐up, and, if lymph node enlargement persisted or progressed, repeat FNAB after 2 months, or if clinical suspicion was high, CNB/excisional biopsy for diagnostic confirmation.

Follow‐up after 2 months was negative, with resolution of lymphadenopathy. Upon further inquiry, the patient reported having sustained an insect bite on the left leg while gardening shortly before the onset of swelling—providing a plausible reactive etiology for the lymph node changes.

5.1.1. Key Diagnostic Issue

Lymphoid polymorphism with prominent eosinophilia and a small B‐cell lambda‐restricted population raises the challenge of distinguishing a reactive dermatopathic lymphadenitis from early or subtle lymphoproliferative neoplasm.

5.1.2. How the WHO Categorization Was Applied

The predominantly reactive background was coupled with a small, non‐diagnostic B‐cell clone which made the cytopathologist reluctant to classify the case as “Benign,” and the findings were insufficient for “SFM”. The case was therefore assigned to the “Atypical” category of the WHO System. An alternative would have been to categorize the case as “Benign” with close follow‐up.

5.1.3. Teaching Point

Minor B‐cell clones may appear in reactive conditions and can suggest a diagnosis of monoclonal B‐cell lymphocytosis, rather than malignancy. Clinical correlation and follow‐up are critical to avoid overdiagnosis of lymphoma.

5.2. Case 2 – Atypical Category (Possible Non‐Lymphoid Cell)

A FNAB of a right axillary lymph node was performed in a 45‐year‐old woman during the evaluation of a clinically enlarged node detected on routine imaging. The FNAB smears demonstrated a predominantly polymorphous lymphoid population, composed of small mature lymphocytes, histiocytes, and scattered plasma cells. In addition, occasional small cohesive tissue fragments of non‐lymphoid cells were identified. These cells exhibited round to oval nuclei with smooth nuclear contours, small nucleoli, and abundant cytoplasm (Figure 2). No significant cytopathological atypia, mitotic figures, or pigment were observed.

FIGURE 2.

Case 2—Atypical Category (possible non‐Lymphoid Cell). (A) A small tissue fragment of large cells with abundant cytoplasm and round, nuclei with nucleoli is seen within a polymorphous lymphoid background (May–Grünwald–Giemsa stain, 40×). (B) A melanocytic nevus within the lymph node capsule identified after excisional biopsy (Hematoxylin and Eosin stain, 40X). [Color figure can be viewed at wileyonlinelibrary.com]

Based on the FNAB features, the case was categorized as “Atypical” according to the WHO System, due to the presence of non‐lymphoid elements that could not be definitively characterized as benign or malignant on FNAB alone. No material was available to prepare a cell block for immunocytochemical studies.

Subsequent excision of the lymph node revealed preserved nodal architecture with a small, well‐circumscribed focus of benign melanocytic cells confined to the capsule and subcapsular region (Figure 2B). These cells were arranged in nests and demonstrated strong immunoreactivity for S‐100, Melan‐A, and PRAME, confirming the diagnosis of a small intracapsular nodal nevus.

The cytopathological finding of rare non‐lymphoid cell tissue fragments on FNAB thus correlated histopathologically with benign nevoid cell inclusions within the lymph node that can mimic atypical or metastatic processes on cytopathology and illustrate the diagnostic challenges such lesions can pose.

5.2.1. Key Diagnostic Issue

Presence of rare small tissue fragments of non‐lymphoid cells that could not be definitively characterized on FNAB, raising the differential diagnosis between benign inclusions of nevoid cells and potential metastatic cells.

5.2.2. How the WHO Categorization Was Applied

Because these cells appeared cytomorphologically bland, but their nature could not be confirmed in the absence of immunocytochemistry, a benign interpretation was not appropriate. The case was placed in the “Atypical” category to reflect this uncertainty.

5.2.3. Teaching Point

Benign melanocytic inclusions in lymph nodes may mimic metastatic or atypical cells on cytopathology. When immunophenotypic characterization is not possible, the “Atypical” category communicates diagnostic uncertainty without overcalling malignancy.

5.3. Case 3 – Atypical (Possible Lymphoid, Non‐Primary Lesion)

A 58‐year‐old man had been affected by T1 mycosis fungoides (MF) for 3 years. The patient presented with scattered cutaneous patches on the trunk, arms, and legs, without clinically evident lymph nodes or organ involvement (stage T1 according to EORTC). He was therefore treated with topical corticosteroids and phototherapy. The patient subsequently developed right axillary lymphadenopathy. The lymph node measured 18 mm, was oval and hypoechoic on ultrasound (US), and displayed a preserved, though reduced, hilum. Because the patient's compliance and clinical condition were not optimal for surgical excision, a US‐guided FNAB was performed. Smears at ROSE showed mature small lymphocytes, follicular center cells, and dendritic reticulin cells. Small lymphocytes were identifiable by their small size, round shape, and dense compact chromatin. Follicular center cells were medium‐sized and irregularly shaped centrocytes or larger and round centroblasts, with a bluish rim of cytoplasm and nuclei with granular chromatin and two or more nucleoli (Figure 3A). Scattered medium‐sized lymphoid cells with dense chromatin were also detected. These latter cells were neither mature small lymphocytes nor clearly identifiable as follicular center cells (Figure 3A,B). Flow cytometry showed that 40% of the gated cells were CD19‐positive B cells with balanced light‐chain expression. T cells displayed a balanced CD4/CD8 ratio and co‐expression of CD2/CD3/CD7. However, the flow profile could not exclude the possibility of MF cell infiltration in the lymph node. Therefore, the FNAB diagnosis was considered “Atypical” according to the WHO System, and strict clinical and US follow‐up was recommended. The patient subsequently added anti‐inflammatory therapy; after 1 month, clinical and US re‐evaluation showed that the lymph node had decreased in size and later almost completely disappeared.

FIGURE 3.

Case 3—Atypical (possible lymphoid, non‐primary lesion). (A, B) Smear preparations show a densely cellular field composed of a heterogeneous population of lymphoid cells with variable size and nuclear morphology (May–Grünwald–Giemsa stain (A) and Papanicolaou stain (B), 40×). [Color figure can be viewed at wileyonlinelibrary.com]

5.3.1. Key Diagnostic Issue

The polymorphous smear and FC results with absence of an aberrant T‐cell population made T‐cell lymphoma less likely. However, in this specific clinical context, the possibility of MF cell infiltration could not be ruled out.

5.3.2. How the WHO Categorization Was Applied

Based on the clinical, cytopathological, and flow cytometry findings, the WHO System categorization was “Atypical”, and a close clinical and US follow‐up was advised.

5.3.3. Teaching Point

Patients with MF may develop lymphadenopathy for several reasons as with the general population, including reactive lymphadenopathy and dermatopathic lymphadenopathy. Nevertheless, lymph node involvement by MF has to be considered which would indicate advanced‐stage MF (IIB), requiring systemic therapy; therefore, a pathological evaluation of the lymph node was necessary. When a lymph node does not show diffuse structural replacement on histopathology, scattered malignant cells may be microscopically detectable if they are clearly distinct from the surrounding lymphoid cells. When cytopathological differences are subtle, flow cytometry may be useful to identify clonal populations, particularly in cases of B‐cell clonality. In T‐cell lymphomas, such as in this case, flow cytometry is less effective unless TRBC1 and/or TRBC2 antibodies are applied. PCR analysis of TCR gamma or TCR beta gene rearrangements can also be used in similar cases, demonstrating multiple peaks with a Gaussian distribution, confirming polyclonality. Therefore, in the absence of clonality assessment, it was advisable to categorize this case as “Atypical”.

5.4. Case 4—Suspicious For Malignancy (Possible Lymphoid Neoplasm)

A 59‐year‐old man presented with generalized lymphadenopathy. FNAB of a cervical lymph node revealed a monomorphic population of intermediate to large lymphoid cells exhibiting moderate to high N:C ratio, irregular nuclear contours and inconspicuous nucleoli (Figure 4A). Flow cytometric analysis was limited by suboptimal cell viability, and the corresponding cell block displayed only moderate cellularity (Figure 4B). Immunocytochemistry showed strong CD20 expression (Figure 4B, inset) with only scattered CD3‐positive T cells, while evaluation of additional markers was not feasible due to insufficient material. Based on the cytomorphologic features and the incomplete immunophenotypic profile, the case was categorized as “SFM” according to the WHO System. A subsequent CNB confirmed the diagnosis of diffuse large B‐cell lymphoma (DLBCL).

FIGURE 4.

Case 4—Suspicious for malignancy (possible lymphoid neoplasm). (A) Monomorphic population of intermediate‐to‐large lymphoid cells with a moderate‐to‐high N:C ratio, irregular nuclear contours, and inconspicuous nucleoli. Small T‐cell lymphocytes are present in the background (May–Grünwald–Giemsa stain, 40X). (B) Cell‐block section showing similar cytomorphologic features (Hematoxylin–eosin stain, 40X). The medium‐to‐large cells are CD20‐positive (inset, ABC, 40X). Further immunohistochemical evaluation on the cell block was not possible due to specimen exhaustion. Flow‐cytometric analysis was limited by suboptimal cell viability. [Color figure can be viewed at wileyonlinelibrary.com]

5.4.1. Key Diagnostic Issue

Monomorphic B‐cell populations with nuclear atypia and limited evaluable ancillary studies pose difficulty in definitively diagnosing a lymphoid neoplasm. The generalized lymphadenopathy, cytopathological findings, and the predominance of B‐cells on immunocytochemistry establish the “SFM” categorization.

5.4.2. How the WHO Categorization Was Applied

The cytomorphology was strongly suggestive of a large cell, B‐cell lymphoma supported by the limited immunocytochemistry showing strong CD20 expression and scattered CD3‐positive cells, but the immunophenotypic data were incomplete on limited cell block material. Since a definitive diagnosis could not be reached, the case was categorized as SFM.

5.4.3. Teaching Point

When cytomorphology is highly suggestive of malignancy but ancillary studies are insufficient, the “SFM” category appropriately signals the need for prompt tissue biopsy to avoid diagnostic delay.

5.5. Case 5 – Suspicious for Malignancy (Possible Non‐Lymphoid Lesion)

A 67‐year‐old man with a history of cutaneous melanoma presented with an enlarged left axillary 30 mm lymph node. Ultrasound‐guided FNAB was performed, and brownish material was collected. Direct smears showed a proteinaceous background containing debris, with scattered medium cells with eccentric abundant micro‐vacuolated or granular cytoplasm containing melanin pigment and small, irregular, often indented nuclei, suggestive of melanophages (Figure 5A). A few lymphoid cells were present in the background. Similar morphological features were observed in the cell block, with occasional small cellular aggregates (Figure 5B). Immunocytochemical analysis for SOX10 and MART1 was negative in the probable macrophages.

FIGURE 5.

Case 5—Suspicious for malignancy (possible non‐lymphoid lesion). (A) Direct smears showed a necrotic background containing debris, with scattered medium cells with eccentric abundant microvacuolated or granular cytoplasm containing melanin pigment and small nuclei, suggestive of melanophages. A few lymphoid cells were present in the background (Papanicolaou stain, 40X). (B) Similar morphological features were observed in the cell block showing foamy macrophages and some small lymphoid cells entrapped within the necrotic tissue (Hematoxylin–eosin stain, 40X). [Color figure can be viewed at wileyonlinelibrary.com]

Based on cytopathological features, the case was categorized as “SFM” according to the WHO System because the background material and probable melanophages were suspicious of necrotic metastatic melanoma but no viable melanoma cells were present.

The patient's clinical management included repeat FNAB, which showed several tissue fragments of atypical melanocytic cells, and a diagnosis of metastatic melanoma was made.

5.5.1. Key Diagnostic Issue

Scant cellularity with the presence of pigmented probable melanophages and necrosis created uncertainty between benign melanophages and metastatic melanoma cells in a patient with a prior history of melanoma.

5.5.2. How the WHO Categorization Was Applied

The cytomorphological features of pigmented cells with nuclear indentations and coarse brown material were concerning but not definitive for melanoma metastasis, and immunocytochemistry was non‐contributory. Because of this, but even more importantly, the clinical setting, with a history of melanoma and a significantly enlarged 30 mm lymph node, the case was categorized as “SFM”.

5.5.3. Teaching Point

In suspected melanoma metastasis, distinguishing melanophages from malignant melanocytes may be challenging in low‐cellularity FNAB samples. The “SFM” category supports recommending additional sampling, which is often required for definitive diagnosis.

5.6. Case 6 – Suspicious for Malignancy (False‐Positive)

A 22‐year‐old woman presented with fever, sore throat, and bilateral cervical lymphadenopathy of 2 weeks' duration. Physical examination revealed multiple tender, enlarged cervical lymph nodes measuring up to 25 mm. Laboratory tests showed mild leukocytosis with relative lymphocytosis. Because of persistent lymph node enlargement and concern for possible lymphoma, a US‐guided FNAB of a right cervical lymph node was performed.

Direct smears were moderately cellular and showed a polymorphous lymphoid population with numerous medium‐to‐large lymphoid cells dispersed singly and in small loose aggregates. These cells exhibited enlarged nuclei, irregular nuclear contours, finely dispersed chromatin, and occasional prominent nucleoli (Figure 6). Mitotic figures were identified. The background contained abundant small mature lymphocytes, immunoblasts, and scattered tingible body macrophages. Occasional apoptotic bodies and focal necrotic debris were present. The overall cytomorphological appearance raised concern for a high‐grade lymphoid proliferation, particularly diffuse large B‐cell lymphoma.

FIGURE 6.

Case 6—Suspicious for Malignancy (false‐positive). Dense lymphoid proliferation composed of small lymphocytes admixed with numerous larger atypical lymphoid cells displaying vesicular chromatin and moderate amounts of cytoplasm. Marked cellular pleomorphism and an increased nuclear‐to‐cytoplasmic ratio are evident (May–Grünwald–Giemsa stain, 20X). [Color figure can be viewed at wileyonlinelibrary.com]

Flow cytometric analysis demonstrated a polyclonal B‐cell population with balanced kappa and lambda light‐chain expression. T cells showed a predominance of CD8‐positive cells with preserved expression of pan–T‐cell markers (CD2, CD3, CD5, CD7) and no aberrant antigen loss. No monotypic B‐cell population was identified. However, due to the marked cytopathological atypia and the clinical presentation with significant lymphadenopathy, the case was initially categorized as “SFM” according to the WHO System, with recommendation for excisional biopsy.

Subsequent excisional biopsy revealed preserved nodal architecture with florid paracortical hyperplasia. Numerous immunoblasts were present within the paracortex, but no evidence of lymphoma was identified. Immunohistochemistry showed a mixed population of CD20‐positive B immunoblasts and CD3‐positive T cells, without light‐chain restriction. In situ hybridization for Epstein–Barr virus–encoded RNA (EBER) demonstrated numerous positive immunoblasts, confirming the diagnosis of infectious mononucleosis.

5.6.1. Key Diagnostic Issue

Marked immunoblastic proliferation with cytopathological atypia and mitotic activity mimicking high‐grade lymphoma on FNAB, particularly in a young patient with prominent cervical lymphadenopathy.

5.6.2. How the WHO Categorization Was Applied

Despite polyclonal findings on flow cytometry, the striking cytomorphological atypia and clinical concern prompted categorization as “SFM” rather than “Atypical.” The category appropriately conveyed a high level of concern and the need for histopathological confirmation, although the final diagnosis proved to be benign.

5.6.3. Teaching Point

Infectious mononucleosis may exhibit florid paracortical immunoblastic proliferation with marked cytopathological atypia, increased mitotic activity, and scattered apoptotic bodies, closely mimicking high‐grade lymphoma on FNAB. In young patients presenting with generalized lymphadenopathy and systemic symptoms such as fever and pharyngitis, this reactive pattern represents an important diagnostic pitfall.

Careful integration of clinical findings, laboratory data (including lymphocytosis), and ancillary studies is essential. In particular, the absence of a monotypic B‐ or T‐cell population and the lack of phenotypic aberrancies on flow cytometry from the involved lymph node strongly favor a reactive process over lymphoma, even in the presence of striking immunoblastic atypia.

Failure to appropriately weigh these elements may lead to overinterpretation of cytopathological atypia and potential overdiagnosis in the “SFM” category. Awareness of this scenario helps prevent false‐positive classifications and unnecessary aggressive management.

6. Discussion

The assignment of a case to the “Atypical” or “SFM” categories often reflects a combination of technical and interpretative factors. Common causes include suboptimal sampling during the FNAB procedure, poor slide preparation or fixation, insufficient material, and a cytopathologist's limited diagnostic experience. The judicious use of these intermediate categories is important to preserve a high negative predictive value (NPV) for the “Benign” category and a high positive predictive value (PPV) for the “Malignant” category [7].

The interpretation and practical application of the “Atypical” and “SFM” categories remain among the most diagnostic challenging areas in lymph node cytopathology. The “Atypical” category is typically reserved for cases showing predominantly benign features but with focal or subtle atypical cytopathological features that preclude the complete exclusion of malignancy. Contributing factors may include suboptimal cellularity, artifactual distortion, overlapping cytomorphological features between reactive and neoplastic lymphoid proliferations, or incomplete ancillary data. Distinguishing monomorphic resting lymph node material, reactive immunoblastic and follicular hyperplasia from low‐grade B‐cell lymphomas remains a classic and recurrent diagnostic problem, often compounded by sampling limitations, partial lymph node involvement, a significant inflammatory/immunological background, or absent ancillary studies.

The “SFM” category, conversely, encompasses cases where cytopathological and/or clinical findings strongly suggest a neoplastic process, but without sufficient cytopathological features to confirm a malignant diagnosis. Again, the causes of “SFM” include poor sampling, poor quality smear making and staining, low cellularity, inherent overlapping features between some lymphomas and reactive processes, and a lack of ancillary testing. This situation is not infrequently seen in cases of Hodgkin lymphoma containing scant diagnostic Reed–Sternberg cells, or when a lymph node is focally involved by large cell lymphoma, poorly differentiated metastatic carcinoma or melanoma. In these settings, integration of clinical findings, imaging correlation, and, when feasible, repeat FNAB or CNB is indispensable in achieving diagnostic certainty.

While the “SFM” category is associated with a consistently high risk of malignancy, its application is not without potential drawbacks. Overuse of the “SFM” category may occur in settings with low tolerance for diagnostic uncertainty, potentially inflating ROM and leading to unnecessary invasive procedures and patient anxiety. In institutions with limited access to rapid ancillary testing, there may be a lower diagnostic threshold for “SFM” assignment, potentially inflating ROM estimates and increasing downstream invasive procedures. False‐positive “SFM” interpretations may arise in cases of reactive immunoblastic proliferations, therapy‐related changes, or necrotic and paucicellular metastatic lesions lacking viable tumor cells. Explicit recognition of these limitations is essential, and the “SFM” category should be reserved for cases in which cytomorphological and clinical features genuinely support a high suspicion of malignancy, rather than serving as a default category for diagnostically challenging specimens.

A persistent source of variability lies in the interobserver reproducibility of these intermediate categories. Differences in individual diagnostic thresholds, the interpretation of the key diagnostic criteria provided in the WHO System, and availability and use of ancillary testing, particularly flow cytometry, can significantly influence the analyses of FNAB of lymph nodes and the ROM estimates for these categories and, consequently, downstream clinical management. These inconsistencies underscore the need for continued refinement of the cytomorphological key diagnostic criteria and definitions, the establishment of standardized diagnostic algorithms [39], and the wider adoption of ancillary techniques such as flow cytometry, immunocytochemistry, and molecular assays to improve diagnostic precision and reproducibility [5, 40].

Another aspect that significantly influences the performance and real‐world reproducibility of the “Atypical” category is the diagnostic mindset with which cytopathologists approach borderline cases. Within the WHO systems, “Atypical” is intended for cases considered probably benign but containing concerning features. This more conservative approach of “probably benign” raises the threshold for categorizing a case as “Atypical”, thereby reducing the size of this category, but theoretically may increase the proportion of false negatives within the “Benign” category. This will be the focus of ongoing studies analyzing the use of the category. Conversely, in some institutions which adopt a more cautious “possibly malignant” stance, there is a broader use of the “Atypical” category, contributing to its relatively high and widely variable ROM seen in the current literature, which leads to confusion among clinicians, a decrease in trust among clinicians as to the usefulness of FNAB of lymph nodes, and potentially increased unnecessary further testing and patient anxiety. Thus, application of the “Atypical” category is influenced not only by morphology but also by institutional culture, risk tolerance, and availability of ancillary testing.

Recognizing these factors is essential to promote standardized categorization and rigorous application of the WHO System's diagnostic criteria for benign and malignant lesions. The aim is to improve cytopathologists' awareness and use of key diagnostic criteria and so improve interobserver consensus by using the WHO System, increase the application of FNAB in lymph node assessment, and improve patient care.

Future multicenter studies are warranted to further delineate the ROM associated with the “Atypical” and “SFM” categories in diverse clinical settings and population cohorts, and a standardized structure for such studies to allow for more accurate comparisons has been suggested [19]. Correlating this data with histopathological and clinical follow‐up outcomes will be critical for the further validation and refinement of the WHO System. Ultimately, a more evidence‐based and standardized application of these “gray zone” categories will enhance the WHO System's clinical applicability, promote consistent communication among clinicians, and optimize patient management.

7. Conclusion

In conclusion, both the “Atypical” and “SFM” categories are important categories within the WHO System for diagnostic stratification and patient management. Their careful application allows cytopathologists to convey diagnostic uncertainty transparently, guide appropriate patient management including selection of ancillary testing, and maintain the overall sensitivity and specificity of FNAB for lymphoproliferative disorders. FNAB of lymph nodes is firmly established as an accurate and universally applicable method for assessing lymphadenopathy and diagnosing infections, inflammatory processes and metastases, and the application of ancillary testing to FNAB material in the form of microbiological studies, immunocytochemistry, and NGS for metastases, flow cytometry, and more recently, NGS for lymphomas has greatly enhanced these roles. However, there will always be indeterminate findings in FNAB as with most diagnostic tests, due to specimen quality, inherent ambiguous features of lesions and neoplasms, interpretative variability, and limits in reproducibility. The optimal use of the “Atypical” and “SFM” categories depends on the rigorous use of standardized diagnostic cytopathological criteria, consistent institutional application, and an integrated approach using clinical, imaging, cytomorphological, and ancillary testing findings. Ongoing research, multicenter validation studies, and continued refinement of standardized reporting frameworks are critical to refine these categories, reduce interobserver variability, facilitate their clinical utility, and optimize patient management.

Author Contributions

Immacolata Cozzolino: conceptualization, data analysis, literature review, manuscript drafting. Mats Ehinger: conceptualization, data analysis, critical revision. Oscar Lin: data analysis, critical revision. Elisabetta Maffei: data analysis, critical revision. Pio Zeppa: data analysis, critical revision. Andrew S. Field: data analysis, critical revision, supervision.

Funding

The authors have nothing to report.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.