Lymph node (LN) metastases from nonhematopoietic malignancies are rare. Their distinction from hematopoietic malignancies typically requires integration of morphology, immunohistochemistry (IHC), and genetic analyses. Among these, neuroendocrine carcinoma (NEC) is immunophenotypically characterized by the expression of CD56 (neural cell adhesion molecule, NCAM), synaptophysin, and chromogranin, as detected by IHC 1. More recently, CD200 expression detected by IHC has been reported in NEC where it may serve as an additional marker for differentiating nonhematopoietic malignancies 2. In contrast, CD200 expression assessed by flow cytometry (FCM) is well documented in nonneoplastic B cells and in hematopoietic malignancies such as chronic lymphocytic leukemia/small lymphocytic lymphoma and multiple myeloma 3-5. Moreover, FCM is routinely used for immunological profiling of hematopoietic malignancies—commonly on LN suspensions and fine-needle aspiration samples—but its use in nonhematopoietic malignancies remains underexplored. Nevertheless, CD56 expression has been reported in NEC using FCM 6. The present study evaluates CD200 and CD56 expression in LN metastases of nonhematopoietic malignancies by FCM, with a particular focus on their diagnostic value in NEC. The FCM data were generated from a retrospective, consecutive series (January 2022—May 2025) of LN samples diagnosed with metastases from nonhematopoietic malignancies. The cohort included cases of NEC (n = 6), squamous cell carcinoma (n = 10), adenocarcinoma (n = 10), and melanoma (n = 11). All diagnoses were established according to the WHO Classification of Tumors series 1, 7. In all cases, LN imprints stained by May–Grünwald–Giemsa were evaluated by two expert cytologists, and all diagnoses were confirmed by histology. FCM was performed using 12-color analysis on a BD FACSLyric cytometer and analyzed with BD FACSuite software version 6.1 (Becton Dickinson, San Jose, CA, USA), within the routine diagnostic workflow of our laboratory. Daily instrument quality control was conducted using cytometer setup and tracking (CST) beads. Fresh LN suspensions, obtained after mechanical dissociation of biopsy fragments, were stained using a single-tube 12-color screening panel routinely used in our laboratory for lymphoid immunophenotyping. The panel included the following antibody combinations: CD8 and surface κ light chain (FITC; clone SK1), CD56 and surface λ light chain (PE; clone NCAM16.2), CD3 (PerCP-Cy5.5; clone SK7), CD19 (PE-Cy7; clone SJ25C1), CD5 (APC; clone L17F12), CD10 (APC-R700; clone HI10a), CD2 (APC-H7; clone RPA-2-10), CD20 and CD4 (V450; clones L27 and RPA-T4), CD200 (BV510; clone MRC OX-104), HLA-DR (BV605; clone G46-6), CD7 and CD23 (BV711; clones M-T701 and M-L233), and CD45 (BV786; clone HI30). All antibodies were obtained from BD Biosciences (San Jose, CA, USA). Samples were initially gated based on forward scatter (FSC) properties to exclude doublets (FSC-H vs. FSC-A), followed by FSC/side scatter (SSC) gating to exclude nonviable cells. The SSC/CD45 dot plot was then used to distinguish CD45− cells (nonhematopoietic malignancies) from CD45+ cells (leukocytes). CD56 and/or CD200 expression was considered positive if detected in 20% or more of CD45− cells. In all samples, mean fluorescence intensity (MFI) values of CD56 and CD200 expression were measured within the CD45− gated population. These MFI values, expressed in relative linear units on an arbitrary bi-exponential scale (ranging approximately from 10−2 to 105), were extracted for each subset. The ratio of CD56 MFI in CD45− cells to that in NK cells (defined as CD56+ cells), and the ratio of CD200 MFI in CD45− cells to that in B cells (defined as CD200+ cells), were then calculated. To reduce inter-sample variability related to staining intensity and cytometer performance, CD56 and CD200 expression levels were expressed as normalized MFI ratios relative to these internal reference populations, allowing more reliable quantitative comparison across samples. Statistical analyses were performed using the Kruskal–Wallis test. When the Kruskal–Wallis test indicated significance, post hoc pairwise comparisons were performed using Dunn's test with Bonferroni correction to adjust for multiple testing. A p-value < 0.05 was considered statistically significant after correction. All cases of NEC (6/6 100%) expressed CD56, confirming prior observations 1. In squamous cell carcinoma, adenocarcinoma, and melanoma, CD56 expression was observed less frequently, in 3/10 (30%), 2/10 (20%), and 6/11 (54%) of cases, respectively (Table 1). Furthermore, the ratio of CD56 MFI in CD45− cells to NK cells was significantly higher in NEC cases compared to other cases (Figures 1 and 2). Additionally, CD200 expression was more frequent in NEC than in adenocarcinoma and melanoma (5/6 83%, 2/10 20%, and 2/11 18%, respectively). No cases of squamous cell carcinoma expressed CD200. Notably, the ratio of CD200 MFI in CD45− cells to B cells was significantly higher in NEC cases than in other cases. These differential MFI values between NEC and other nonhematopoietic malignancies align with previous IHC findings, which show strong CD200 expression in NEC but weaker or intermediate expression in other nonneuroendocrine tumors 2. The combined ratio of CD56 MFI and CD200 MFI appears particularly informative in the rare cases of CD56+/CD200+ adenocarcinoma and melanoma. In these cases, the MFI ratios of CD56 and CD200 were significantly lower than those observed in NEC. Additionally, none of the CD56+ squamous cell carcinoma cases expressed CD200. Therefore, the combined assessment of CD56 and CD200 MFI expression enabled discrimination of NEC from squamous cell carcinoma, adenocarcinoma, and melanoma. The diagnostic value of FCM in this context is not merely binary but quantitative. While CD56 and CD200 expression can be observed in isolation across various nonhematopoietic malignancies, the combined immunophenotypic profile—defined by significantly higher normalized MFI ratios for both markers—yielded a distinct signature for NEC. This dual-parameter quantitative approach provided a statistically significant separation (p < 0.05) from melanoma, adenocarcinoma, and squamous cell carcinoma. Even in rare cases where nonneuroendocrine tumors showed partial marker expression, the MFI ratios remained significantly lower than those observed in NEC, demonstrating the superior specificity of this combined profile. NEC is well recognized for its expression of CD56 and CD200 1, 2. In this study, we demonstrate that these markers can be efficiently evaluated using FCM, which offers several distinct advantages over traditional methods. Unlike conventional FCM reporting, which typically provides the percentage of malignant or nonmalignant cells expressing a given marker, we assessed CD56 and CD200 using MFI normalized to control cell populations—B cells for CD200 and NK cells for CD56. Interestingly, CD56 and CD200 expression in our cases sometimes exhibited a bimodal pattern by FCM. In some cases, CD56 expression displayed a bimodal distribution within the CD45− tumor population. Notably, both CD56+ and CD56− events exhibited similar FSC/SSC characteristics and formed coherent clusters within the CD45− gate, supporting the interpretation of intratumoral heterogeneity of CD56 expression, although the presence of a minor fraction of damaged or nonviable cells cannot be completely excluded. Normalized MFI values effectively capture the overall expression intensity on malignant cells, allowing statistically significant discrimination of NEC from squamous cell carcinoma, adenocarcinoma, and melanoma, even when only a small fraction of cells is CD56+ or CD200+. FCM also offers a rapid turnaround time, with results obtainable within 12–24 h. This rapid assessment is valuable for preliminary diagnosis and can complement histopathological evaluation. It is essential to interpret FCM results in the context of cytology and additional antibody panels, as some hematologic malignancies, such as acute leukemias and multiple myeloma, may also show CD45− profiles 8. Although LN involvement is uncommon in multiple myeloma and can be distinguished morphologically, these tumors may express CD200 and CD56; however, they also express markers such as BCAM, CD38, and CD138, which are absent in NEC. CD200 has been implicated in promoting tumor growth and metastasis by facilitating immune evasion 9. Emerging evidence supports the potential of CD200-targeted immunotherapy across several malignancies 10. Detecting CD200 expression by FCM, therefore, has important therapeutic implications and may inform clinical decision-making. While histology and IHC remain the gold standards for diagnosing nonhematopoietic malignancies, FCM provides rapid, actionable insights that can guide the selection of ancillary techniques, such as initial IHC panels, and help identify potential therapeutic targets. Overall, the MFI analysis of CD56 and CD200 in NEC underscores the utility of FCM for diagnostic and therapeutic purposes. Larger, multicenter studies are warranted to validate these preliminary findings and to establish the definitive role of CD56 and CD200 as both diagnostic and therapeutic markers in NEC. R.C. reviewed the cytological data and drafted the manuscript. E.B. was responsible for the collection of flow cytometry data. T.F. and M.D. reviewed and confirmed the histological cases. L.B. conducted the studies, analyzed the flow cytometry data, and co-authored the manuscript. All authors contributed to the final version of the manuscript. The authors have nothing to report. This manuscript respects the ethic policy of CHU Lyon for the treatment of human research participants. All procedures were conducted in accordance with the Helsinki Declaration and the Biological Resource Centre policy of the Hospices Civils de Lyon. Furthermore, the research protocol was approved by the institutional review board of the Hospices Civils de Lyon (25-5321). The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Chiriac et al. (Tue,) studied this question.