Combined fine needle aspiration cytology and core needle biopsy in the same setting: A two‐years’ experience

Fine needle aspiration cytology (FNAC) combined with rapid on‐site evaluation (ROSE) and ancillary techniques is an accurate diagnostic tool for many pathologies. However, in some cases, it may not be sufficient for actionable diagnoses or molecular testing, especially for cases that require large immunohistochemical panels or cases in which histological features are mandatory for the diagnosis. Core needle biopsy (CNB), on the contrary, provides samples that are suitable for histological features and sufficient for all ancillary studies. However, CNB is often performed by radiologists or clinicians without the direct participation of cytopathologists, which can lead to missed or delayed diagnoses. This study reports on the experience of combining FNAC and CNB performed in one setting by cytopathologists. The aim was to evaluate the impact of CNB on FNAC and the diagnostic efficiency of the combined procedures.


| INTRODUC TI ON
0][11][12] The incorporation of rapid onsite evaluation (ROSE) further enhances its utility by gauging sample adequacy, leading to a notable decrease of approximately 20% in inadequate rates. 4,13[20][21][22][23][24] Nonetheless, despite the documented high sensitivity and specificity values, 2,4,25 FNAC may prove insufficient to reach actionable diagnoses in certain pathologies, particularly those necessitating extensive immunohistochemical panels and molecular testing, or intricate cases wherein histological architectural features are pivotal for diagnosis. 5,20Core needle biopsy (CNB) emerges as a potent sampling technique offering the advantage of revealing tissue architecture and obtaining samples more substantial than those garnered via FNAC and CB.7][28][29][30] Like FNAC, CNB can be conveniently performed in outpatient clinics with minimal patient discomfort and low risk of complications, making it especially pertinent for individuals unable or unwilling to undergo surgery.
The utility of CNB has been underscored by numerous studies; however, it is frequently conducted by radiologists and clinicians in the absence of direct pathologist involvement. 25,31This approach entails certain limitations, including the lack of information derived from FNAC and ROSE and challenges in assessing sample adequacy in CNB specimens.To overcome these limitations, a potential solution is performing FNAC, ROSE and CNB in a single setting.Yet, only a limited number of studies exist in which interventional cytopathologists have conducted these procedures. 5,32,33 our institution, the integration of combined FNAC and CNB within a single setting has been implemented over the past 4 years.
This study seeks to evaluate the impact of this approach on the diagnostic efficacy of FNAC as well as both procedures collectively.

| Ethics and study design
Approval from a local ethics committee was not required as this study is retrospective.All patients provided preliminary informed consent for the diagnostic procedures and usage of anonymized data as outlined below.Data were retrieved from electronic records for all consecutive combined FNAC and CNB cases performed between 1 January 2020 and 31 December 2021.A total of 142 cases were selected from our institute's archival files.

| Fine needle aspiration cytology (FNAC)
Fine needle aspiration cytology and ROSE were conducted by experienced cytopathologists (VC, AC, PZ).ROSE was employed to assess adequacy, offer preliminary diagnoses and guide the decision to proceed with CNB.Briefly, the FNAC procedures were performed using 25-or 30-mm 23-gauge needles or spinal 22G needles (in case of deep-seated targets) attached to disposable 20 mL syringes.One smear was air-dried and immediately stained with Diff-Quik stain for ROSE.Further smears were allocated to Papanicolaou or Diff-Quik according to ROSE results, and fixed accordingly.In cases suspicious in actionable diagnoses in 91.5% of all cases (n = 130/142).A complete molecular assessment was successfully performed in 14.7% of cases (n = 21/142) utilizing either FNAC or CNB material.

Conclusions:
The combined use of FNAC and CNB in one setting improves the diagnostic accuracy of both procedures.This approach exploits the advantages of each procedure, enhancing the accuracy of the final diagnosis.

K E Y W O R D S
ancillary techniques, core needle biopsy, fine-needle aspiration cytology, rapid on-site evaluation of non-Hodgkin lymphoma (NHL), FC immunophenotyping was performed on material suspended in phosphate-buffered saline.Cell blocks were prepared for non-lymphomatous pathologies requiring ancillary studies.FNAC diagnoses were classified based on specific classification systems for lymph nodes 33,34 and thyroid, 6 while other targets were classified as negative, suspicious or positive for malignancy.Positive cases, when possible, were further specified by histotype, characterized using specific cytological features and ICC and FC techniques. 22,35,36All cytologic smears were examined and signed out by the same performers of the procedure (VC, AC, PZ).

| Core needle biopsy (CNB)
Patients eligible for CNB were selected based on requests from referring doctors or the cytopathologist's impression after FNAC and ROSE, following detailed explanations and informed consent.
Haemostasis alterations with an increased risk of haemorrhage were exclusion criteria, and anticoagulant therapy was suspended (Figure 1).Local anaesthesia was administered before CNB using an 18G needle, with one to two biopsies taken from each lesion using a new needle for each biopsy.Ultrasound and Colour Doppler US evaluation were performed in all the cases before FNAC and CNB to assess the specific features of the targets (signs of colliquation, presence of large vessels, relationship with nearby organs).This information was particularly useful in large, superficial targets in which CNB was performed without US guide.The remaining cases were performed with US guide.This approach allowed to perform CNB safely in an outpatient setting.CNB samples were immediately fixed in 10% neutral buffered formalin, embedded in paraffin and processed according to standard histology protocols, as previously described. 37Immunohistochemistry was performed on a Ventana Benchmark ultra-automated immunostainer (Ventana Medical Systems) using standard protocols and manufacturer-provided reagents.The workflow for combined FNAC-CNB in one setting is reported in Figure 11.

| Core needle biopsy impact and statistical analysis
The comparison between FNAC and CNB was consecutive because they were consecutively performed by the same operators.
Moreover, at the moment of CNB evaluation, we were aware of FNAC results including ROSE and ancillary techniques (mainly flow cytometry).At the same time, each procedure was evaluated, diagnosed and reported as two distinct procedures in which the sole data obtained from each of them (for instance: FC for FNAC and ICC for CNB) were utilized for the diagnoses of each sample.(FC and ICC for FNAC, immunohistochemistry for CNB).Specific entities for NHL and primary tumour metastases were reported when feasible, otherwise, malignant cases were designated as not otherwise specified (NOS).Surgical samples, clinical follow-up and CNB served as diagnostic gold standards when available.

| Sample characteristics
The series encompassed a majority of lymph node samples (68%, have been compared with the corresponding CNB (see Table 3).
Regarding the two cases diagnosed as 'lymphoma NOS' on both FNAC and CNB, they were DLBCL with extensive fibrosis which hampered both FNAC and CNB.

| Impact of CNB
The impact of CNB on the diagnostic process was evaluated.CNB confirmed FNAC diagnoses in 40.1% (n = 57/142) of cases, improved diagnoses in 47.2% (n = 67/142), allowed diagnoses not feasible through FNAC alone in 2.1% (n = 3/142) of cases and changed prior FNAC diagnoses in 2.1% (n = 3/142).We considered the allowed and changed cases of particular interest because they represent a substantial contribution of CNB to the diagnosis of difficult or complex cases (Table 4).
Namely, allowed cases concerned two lymphadenopathies and a soft tissue mass.Of the lymphadenopathies, one was diagnosed as atypical by FNAC and as Rosai-Dorfman disease at CNB (Figure 6), and the second was diagnosed ALUS by FNAC and MCL at CNB.The soft tissue mass was diagnosed as atypical at FNAC and leiomyosarcoma at CNB. Changed cases concerned two lymphadenopathies and a parotid nodule.One lymphadenopathy was diagnosed as DLBCL by FNAC and Anaplastic ALK+ lymphoma by CNB (Figure 7); the other lymphadenopathy was diagnosed as a metastasis from a poorly differentiated carcinoma considered secondary to a previous carcinoma of the penis but was found out to be a diagnosed poorly differentiated adenocarcinoma of the lung by CNB.Finally, the parotid nodule was diagnosed as a pleomorphic adenoma at FNAC and malignant myoepithelioma by CNB.These cases are reported and summarized in
This involved sequencing EGFR, BRAF, KRAS, ALK, ROS1, RET and MET for lung adenocarcinoma metastases using NGS in 18 cases.
Additionally, BRAF mutation was detected in a case of lymph node

(A) (B) (C) (D) (E) (F)
thyroid and salivary glands.However, FNAC's limitations in providing detailed histological information and its potential lack of diagnostic material, particularly in cases of soft tissue tumours, have led to its inadequacy for certain malignancies.This is especially true for lymphomas, where primary FNAC diagnoses have been met with criticism due to the need for surgical excision for histological control and subtype classification. 42,43In contrast, CNB offers a viable alternative to surgical biopsy, providing tissue architecture insights and ample material for ancillary testing, all while avoiding the risks and costs associated with surgery.
While CNB boasts benefits, especially for lesions that are difficult to excise surgically or where patients refuse surgical intervention, the procedure has been questioned when performed by non-cytopathologists. 44,45 CNB lacks the advantage of ROSE and the obtained material may not represent the entire lesion, leading to potential diagnostic inaccuracies, particularly for lymphomas.
However, the combined approach of performing both FNAC and CNB in the same setting mitigates the shortcomings of each method (Figure 11).FNAC samples various parts of large tumours, accounting for tumour heterogeneity, while CNB offers architectural details

(A) (B) (E) (D) (C) (F)
33]46 This approach also aids in identifying patients who genuinely require surgical excision and those for whom surgery would be excessive, delaying diagnosis and appropriate therapy.Notably, the comparison between the two procedures was consecutive because they were consecutively performed by the same operators.Moreover, at the moment of CNB evaluation, we were aware of FNAC results including ancillary techniques (mainly FC).However, FNAC and CNB were diagnosed and reported as two distinct procedures in which only the data obtained from each of them were utilized to produce two distinct diagnostic reports.This approach highlights both the advantages and the limits of CNB, which in some cases performed worse than FNAC (see Table 4).Despite that the combined FNAC/CNB approach has demonstrated significant effectiveness, particularly for cases requiring histological confirmation where surgical excision is challenging or refused by patients.Notably, this approach accelerates diagnosis, as CNB is quicker than the surgical route in both sampling and processing.Moreover, it streamlines the diagnostic process, benefiting cases that necessitate surgical excision by aiding surgeons and clinicians in timing surgery assessments and therapeutic decisions.This approach has been explored with combined FNAC and CNB performed by clinicians or radiologists, revealing heightened diagnostic accuracy for both techniques. 10,20,47However, the approach is further optimized when conducted by interventional cytopathologists, as skill requirements are largely similar. 5This synergy improves the adequacy of FNAC, identifies cases warranting CNB and minimizes time and cost related to performing procedures in separate settings.
In their study, Domanski et al.Nonetheless, it is crucial to consider clinical guidelines.For instance, the Sydney System recommends histological control for the initial diagnosis of both Hodgkin and non-Hodgkin lymphoma.In our study, CNB procedures were conducted, processed and interpreted in a manner consistent with histological samples, and they were accepted by clinicians as valid histological controls.This approach allowed CNB to serve as a substitute for surgical biopsies, which became particularly valuable in cases where surgical biopsies were hindered by specific clinical contexts or declined by patients.
However, the combined FNAC/CNB approach can serve as an alternative to surgical excision for actionable diagnoses.Jelloul et al. 47 showed that with this approach, the lymphoma classification rate reached 81%, as opposed to the 40% subclassification rate achieved with FNAC alone.This method proves effective for both initial diagnosis and surveillance cases with a previous lymphoma history.In 4 years (range: 18-91 years).The cases involved lymph nodes (n = 96/142), soft tissue nodules or masses (n = 31/142), liver (n = 7/142), thyroid (n = 3/142), parotid (n = 3/142) and bone (n = 2/142) nodules or masses.Clinical suspicions and requests for both FNAC and CNB were communicated by referring doctors in 52.4% of cases (n = 74/142), while in the remaining cases, CNB was performed based on ROSE results and the decision of interventional cytopathologists.In all cases, patients were informed about the procedures, including the possible performance of CNB, based on clinical indications, ROSE results, feasibility and provided signed informed consent.
24 hours before the procedure because normal coagulation indices and platelet count are necessary to reduce bleeding risk.Despite the absence of a consensus regarding threshold values that preclude this kind of procedure, platelet count <50.000/μL and Quick Time <50% are commonly used indices.Moreover, the International Normalized Ratio should be <2.0 prior to low-risk procedures and <1.5 in medium−/high-risk procedures.These criteria were used as additional exclusion criteria.Selection criteria included target diameter ≥20 mm, accessibility by palpation or US guidance, adequate clearance from large vessels (≥10 mm), and absence of haemorrhagic or necrotic FNAC material at ROSE.The needle used was a 18-gauge Histocore Biopsy Instrument, HC18110, 18G X 100 -Notch 19 mm (BIP, Biomed Instrumente & Produkte, Germany) Molecular testing was performed on tumours when clinically indicated to detect targetable molecular alterations.Depending on the availability of material, scraped smears or CNB sections were utilized.Massively parallel sequencing was used for wide gene panel assessments (Cancer Panel, MiSeq, Illumina) while Idylla cartridges (Biocartis) were employed for single or limited mutations when appropriate. 15,16,38-40F I G U R E 1 (A) Topical local anaesthesia administered at the site of the previous FNAC before performing CNB.(B) Schematic representation of the CNB procedure: (1) Needle tip positioned on the target surface; (2) First shot inserts the needle into the target; (3) Second shot advances the covering to cut and trap a tissue fragment; (C) Needle insertion, possibly using the same track as FNAC; (D) Ultrasound-guided CNB.

F
I G U R E 2 (A) Lymph node FNAC displaying monomorphic lymphoid cells with small, irregular nuclei, granular chromatin and absent or small nucleoli (Diff-Quik 270X); FC showed light chain restriction.FNAC diagnosis: small-cell NHL (NOS).(B) CNB revealing diffuse small-cell NHL (H&E 270X).(C) IHC CD20 positivity.(D) IHC CD10 positivity.(E) Ki67 positivity (5%-10% of cells).Final actionable diagnosis: low-grade follicular lymphoma.U R E 3 (A) Lymph node FNAC showing monomorphic lymphoid cells with medium-sized, irregular nuclei, dense chromatin and small nucleoli (Diff-Quik 270X); FC revealed CD20+, CD19+, FMC7+, CD5+, CD23-, CD10-, lambda light chain+, kappa-.FNAC+FC diagnosis: mantle cell lymphoma.(B) CNB as histological control displaying diffuse NHL (H&E 270X).(C) IHC Cyclin D1 positivity.D: Ki67 positivity (>50% of cells).Final actionable diagnosis: mantle cell lymphoma.metastasis from melanoma using Idylla.For lymph node metastases from gastric adenocarcinoma, KRAS, NRAS and BRAF status was assessed by NGS and microsatellite instability was assessed by RT-PCR in two cases.4| DISCUSS ION The adoption of FNAC as an alternative to open surgical biopsy has proven effective for many solid tumours, such as those in the F I G U R E 4 (A) Lymph node FNAC showing polymorphic large lymphoid cells with dispersed chromatin, evident nucleoli and small lymphocytes in the background (Diff-Quik 430X); FC ineffective.FNAC diagnosis: large cell NHL, NOS.(B) IHC CD20 positivity on CNB.(C) Bcl-6 positivity.(D) MUM1 positivity.Final actionable diagnosis: DLBCL, ABC type.U R E 5 (A) Lymph node FNAC displaying atypical large cells with one or two evident nucleoli scattered among small lymphoid cells (Diff-Quik 270X); ICC on additional smears ineffective.(B) CNB showing lymphoid cell groups surrounded by fibrotic tissue (H&E 106X).(C) IHC CD30 positivity in scattered cells.(D) IHC CD30 positivity in a Reed-Sternberg cell.Final actionable diagnosis: Hodgkin lymphoma, nodular sclerosis.U R E 6 (A) lymph node FNAC showing a polymorphous smear with cells having large, dense cytoplasm in an inflammatory background; few lymphoid cells are detectable (Diff-Quik stain 270X), (B) epithelioid cells, isolated or loosely cohesive with eccentric, ovoid, atypical, nucleolated nuclei (Diff-Quik stain 430X), this case was described and reported as atypical.(C) CNB showing a nodal architecture subverted by epithelioid cells; follicles and lymphocytes are absent (H&E 270X), (D) histiocytes with epithelioid appearance and intermingled inflammatory cells (H&E 430X), (E) diffuse positivity for CD68, (F) positivity for S100; CD1a and Langerin were negative.CNB diagnosis was Rosai-Dorfman disease.7 (A, B) lymph node FNAC showing a polymorphous smear with large, dispersed, atypical nucleolated cells and small lymphocytes in the background (Diff-Quik and Papanicolaou stains 430X); Flow cytometry showed few CD19 positive events.FNAC diagnosis was DLBCL.(C) CNB showing a subverted nodal architecture by large cells; residual lymph node is visible at the top of the fragment (H&E, 270X).(D) CD45 positivity in the residual lymph node and negativity in the neoplastic area (E) positivity for CD30 in the neoplastic cells and negativity in the residual lymph node cells.(F) positivity for ALK in the neoplastic cells and negativity in residual lymph node.Final diagnosis was anaplastic, ALK+ lymphoma.
It would have been interesting to reverse the comparison (CNB versus FNAC as reference) but as CNB were reported by the same operators, blinding was impossible, and the reverse comparison would have been influenced by the awareness of the first diagnosis; therefore, it has not been performed in the present study.

F I G U R E 1 0 1 46
Abbreviations: CNB, core needle biopsy; FNAC, fine needle aspiration cytology; NOS, not otherwise specified.Actionable diagnoses: accurate diagnoses with indication of specific pathological entities or primary tumour in cases of metastases.
To evaluate the impact of CNB on the diagnostic process, FNAC diagnoses were compared with subsequent CNB diagnoses.The impact of CNB was categorized as follows: non-contributory, in cases of inadequate samples; confirmed, when the CNB and FNAC diagnoses were the same; 33proved, when the CNB diagnosis was consistent with the FNAC diagnosis and further specified the corresponding entity; allowed, when CNB produced a diagnosis that could not be reached by FNAC; changed, when the CNB changed the previous FNAC diagnosis.Statistical analysis was conducted using R v4.0.141andMicrosoftExcel (Microsoft, Redmond, WA), including descriptive statistics for continuous and categorical variables.3|RESULTS3.1 | Lymph nodes and extra-nodal sites analysisMost FNAC and CNB procedures (78.6%, n = 112/142) were USguided, while others were performed free-handedly after ultrasound examination for palpable and superficial targets.Results obtained from FNAC and CNB of lymph nodes and extra-nodal sites were analysed separately due to distinct diagnostic and classification approaches.For lymph nodes, diagnoses included inadequate, negative, atypical, suspicious and malignant.33InadequateFNAC results were identified by ROSE and promptly repeated, thus not appearing in the corresponding results.Conversely, 12 out of 142 CNB samples (8.45%) were deemed inadequate due to reasons such as excessive necrosis, dominance of capsule or perilesional tissues, tissue fibrotic distortion or technical bias.After initial diagnoses, specific entities were reported based on clinical data, morphological features and ancillary technique results 32/96), and benign cases represented 5.2% (n = 5/96).Combined lymph node FNAC and CNB diagnoses yielded actionable results in 91% (87/96) of cases (Table1).The analysis of extra-nodal sites re- 96/142; Figures 2-7) and extra-nodal lesions (32%, n = 46/142; Figures 8-10).Among lymph nodes, lymphoma accounted for 61.5% of cases (n = 59/96), solid neoplasm metastases constituted 33.3% (n = diagnosis of occult primary neoplasms.As the diagnosis of specific subtypes of lymphomas may be challenging both on FNAC and CNB samples, lymphomas difficult or not identifiable on FNAC have been analysed.For this purpose, entities diagnosed as NHL-NOS by FNAC

Table 4
. CNB was non-contributory in 8.4% (n = 12/142) of cases.Among non-diagnostic CNB, only two were performed without US guidance, because the targets were localized in the subcutaneous soft tissue and were palpable.Notably, non-contributory CNB were caused by determined by extensive fibrosis with distortion of the tissue (4 cases), necrosis (2 cases) and sampling of non-representative areas of the lesions (6 cases).Overall, CNB exerted a positive impact on the diagnostic process in 51.4% (n = 73/142) of total cases.The combined FNAC and CNB procedure led to actionable diagnoses in 91.5% (n = 130/142) of all cases.

Table 3 )
. As expected, other than some of the most frequent entities, MZL and PTLD were significantly represented (2 cases each).The first column reports the final diagnoses of 43 cases where FNAC and CNB were concordant; the second column reports cases diagnosed as 'atypical' or 'lymphoma, not otherwise specified' by FNAC and for which a diagnosis was reached with CNB; finally, the third column lists the final diagnoses of cases where FNAC and CNB were discordant.
33In fact, it is important to note that while FNAC diagnoses are actionable in many cases, such as in thyroid cytology, the same level of certainty may not apply to other organs, particularly in lymph node FNAC and lymphomas.In the case of lymphomas, primary diagnoses made through FNAC, although Note:supported by ancillary techniques and indications of specific nosologic entities, are often requested to be histologically confirmed by clinicians.However, FNAC does indeed demonstrate similar capabilities, as evidenced by numerous studies, including the present one.