Using CIC and HoCIC as diagnostic biomarkers for malignant pleural and ascitic effusion

Abstract

Background

Malignant pleural and ascitic effusion (PAE) represents a prevalent complication of advanced cancer with significant diagnostic challenges. Current cytological examination lacks optimal sensitivity, necessitating novel diagnostic targets. Cell-in-cell (CIC) structures, characterized by one cell engulfing another, represent an emerging morphological biomarker with potential diagnostic and therapeutic implications. This study evaluates the diagnostic efficacy of CIC and its homotypic subtype (HoCIC) as independent morphological targets for malignant PAE detection.

Methods

A total of 584 PAE samples were prepared as cell blocks from consecutive patients at a tertiary center (June 2020–May 2024). CIC structures were quantified in hematoxylin and eosin (H&E)-stained sections (per 10 high-power fields, 40× objective). Homotypic CIC (HoCIC) was identified using immunohistochemical (IHC) profiling (CK-pan, CD68) to distinguish tumor cell-tumor cell engulfment. Diagnostic performance was assessed by sensitivity, specificity, accuracy, and receiver operating characteristic (ROC) analysis with Youden’s index optimization.

Results

Among 357 malignant and 227 benign PAE cases, CIC was detected in 87.39% (312/357) versus 8.37% (19/227) of benign cases (p < 0.001). HoCIC demonstrated higher specificity, present in 77.31% (276/357) malignant but only 1.32% (3/227) benign cases (p < 0.001). Diagnostic metrics for CIC: sensitivity 87.4%, specificity 91.6%, accuracy 89.0%, NPV 92.6%; for HoCIC: sensitivity 77.3%, specificity 98.7%, accuracy 85.6%, PPV 98.9%. Optimal cut-off values were 0.5/10 HPF or 1/20 HPF (Youden’s index 0.79 for CIC, 0.76 for HoCIC). Both markers showed significant independent discriminatory value.

Conclusion

CIC and HoCIC represent validated, readily implementable morphological biomarkers with complementary diagnostic utilities—CIC optimized for sensitive screening and HoCIC for specific confirmation. Integration of this tiered morpho-molecular approach into routine cytopathology workflows may enhance diagnostic precision and guide therapeutic decision-making in advanced malignancies.

Introduction

Pleural and ascitic effusion (PAE) can result from over 60 malignant and benign conditions. Malignant PAE represents a critical clinical complication affecting over 50% of patients with advanced metastatic cancer, significantly impacting quality of life and therapeutic decision-making (1, 2). Accurate distinction between benign and malignant effusions is paramount for staging, prognosis, and selection of targeted therapies. Despite advances in imaging and biomarker detection, cytopathological examination of effusion fluid remains the gold standard for diagnosis. However, the sensitivity of conventional cytology ranges from only 43% to 80%, necessitating repeated invasive procedures and delaying definitive treatment (3). Although ancillary techniques—including immunocytochemistry, flow cytometry, and molecular testing—have improved diagnostic yield, these approaches are resource-intensive, require specialized expertise, and remain unavailable in many clinical settings (2). Consequently, there is an urgent unmet need for rapid, cost-effective, readily accessible morphological biomarkers and applicable diagnostic adjunct methods that can enhance diagnostic precision in routine cytopathological practice.

Cell-in-cell (CIC) structures represent a distinctive cellular phenomenon wherein one viable cell is completely internalized within another, first described by H. Stroebe in 1864 (4, 5). This process involves active invasion by the internalized cell (entosis) rather than passive phagocytosis, implicating dynamic cytoskeletal reorganization and altered cell adhesion mechanisms. CIC formation has been increasingly recognized as a morphological hallmark of aggressive tumor behavior, with elevated frequencies documented across diverse malignancies including lung, breast, and gastrointestinal cancers (6, 7). Based on the identity of participating cells, CICs are classified into homotypic CIC (HoCIC)—where tumor cells engulf other tumor cells—and heterotypic CIC (HeCIC), typically involving immune cell internalization (8, 9). Notably, HoCIC predominates in malignant tissues and has been mechanistically linked to tumor immune evasion, aneuploidy induction, and chemoresistance, positioning it as a potential therapeutic target beyond mere diagnostic utility (10, 11).

Despite their established association with malignancy, CIC structures have not been systematically integrated into clinical cytopathology workflows. A major limitation has been the technical constraints of conventional preparations: in cytological smears or liquid-based monolayers, cellular overlap and three-dimensional clustering obscure CIC morphology, precluding reliable quantification (12). The advent of cell block technology—wherein effusion samples are processed into paraffin-embedded tissue blocks—has revolutionized diagnostic cytopathology by enabling high-quality histological sections, comprehensive immunohistochemical (IHC) profiling, and standardized morphological assessment (13–15). Importantly, cell block preparations preserve cellular architecture and allow clear visualization of CIC structures in hematoxylin and eosin (H&E)-stained sections, overcoming the limitations of traditional cytology.

From a translational oncology perspective, CIC and HoCIC represent targetable morphological phenotypes that bridge diagnostic pathology and precision medicine (16). The quantification of these structures requires no specialized molecular infrastructure—merely standard microscopy—and thus offers immediate applicability across diverse clinical settings, including resource-limited environments. Furthermore, emerging evidence suggests that CIC formation involves E-cadherin-dependent cell adhesion and Rho-ROCK signaling pathways, providing mechanistic insights that could inform future therapeutic interventions (17). While the complete molecular triggers of CIC formation remain under investigation, their robust association with malignant transformation establishes HoCIC as a high-specificity diagnostic target and CIC as a sensitive screening biomarker for malignant effusions.

In this context, we conducted a large-scale diagnostic accuracy study to rigorously evaluate CIC and HoCIC as independent morphological markers for distinguishing benign from malignant PAE. By leveraging cell block technology and standardized IHC validation, we aimed to establish clinically applicable cut-off values and diagnostic algorithms that could be directly integrated into routine cytopathological practice, ultimately improving patient stratification and therapeutic outcomes in advanced cancer management.

Materials and methods

Study design and patient population

This prospective diagnostic accuracy study was conducted at the Second Hospital of Xi’an Jiaotong University, China, between June 2020 and May 2024. Consecutive patients presenting with newly diagnosed pleural or ascitic effusion requiring diagnostic evaluation were enrolled. Inclusion criteria comprised: (i) age ≥18 years; (ii) available clinical and imaging data; (iii) sufficient effusion volume for cell block preparation (≥50 mL); (iv) definitive benign or malignant diagnosis established within 3 months of initial sampling. Exclusion criteria included: (i) inadequate cell block material for assessment; (ii) loss to follow-up before definitive diagnosis.

A total of 584 PAE samples from 294 male and 290 female patients (median age 62 years, range 23–93 years) met inclusion criteria. The final cohort comprised 357 malignant cases (61.1%) and 227 benign cases (38.9%). Malignant effusions were confirmed by histopathological biopsy, cytological findings consistent with clinical–radiological evidence of malignancy, or disease progression during follow-up. Benign effusions were established by negative cytology/histology with ≥6 months clinical follow-up without malignant evolution. Primary tumor sites, systemic treatment histories, and laboratory parameters were systematically recorded.

Cell block preparation and morphological assessment

All effusion samples were processed using standardized cell block protocols to ensure reproducible diagnostic material. Fresh effusion specimens (50–200 mL) were centrifuged at 1,500 × g for 10 minutes. Cell pellets were mixed using X-Gel standardized cell block protocols and fixed in 10% neutral buffered formalin for 30 minutes, embedded in paraffin, and sectioned at 4 μm thickness. This methodology preserves cellular architecture and enables high-quality IHC profiling.

Hematoxylin and eosin (H&E) staining was performed on all sections. Two board-certified pathologists (Rui Guo. and Dongxuan Li, with 15 and 12 years of cytopathology experience, respectively) independently reviewed cases in a blinded fashion. Discrepancies were resolved by consensus with a third senior pathologist. Clinical, laboratory, and imaging data were integrated to establish final diagnoses according to international cytopathological guidelines.

CIC quantification protocol and CIC counting

Cell-in-cell (CIC) structures were defined as complete internalization of one or more viable cells within the cytoplasm of another intact cell, with preserved nuclear morphology of both participants. Quality control measures: (i) only intact CIC structures with discernible membranes were counted; (ii) apoptotic bodies and phagocytosed cellular debris were excluded; (iii) fields with <50 tumor cells were excluded from counting.

To ensure clinical applicability, CIC quantification was conducted by specialized cytopathologists using Mackay’s validated criteria: systematic screening of tumor-enriched areas under 40× objective lens with standardized reporting as CICs per 10 high-power fields (HPFs) (18).

Immunohistochemical profiling and HoCIC identification

IHC staining was performed on 4-μm cell block sections using the Ventana BenchMark XT automated staining platform (Roche Diagnostics, Rotkreuz, Switzerland). Antibody selection—CK-pan (AE1/AE3), CK7 (OV-TL 12/30), CD68 (KP-1), and additional lineage-specific markers—was guided by morphological features observed in corresponding H&E sections. All reagents, including prediluted primary antibodies and UltraView Universal DAB Detection Kits, were sourced from Roche Diagnostics (Shanghai, China). Automated staining followed manufacturer-standardized protocols with rigorous positive and negative controls run concurrently.

CIC structures were classified as homotypic (HoCIC) and HeCIC. HoCIC was defined as cytokeratin-positive tumor cell engulfing another CK+ cell. Heterotypic CIC (HeCIC) was defined by cytokeratin-negative engulfing cells (typically CD68-positive macrophages). Only HoCICs were quantified in 10 HPFs (40×) of tumor-rich areas by blinded cytopathologists for diagnostic analysis. Cases with ≥1 HoCIC per 10 HPFs were classified as positive.

Statistical analysis

The diagnostic performance of CIC and HoCIC as morphological biomarkers was comprehensively evaluated by calculating sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy with 95% confidence intervals (95% CIs). Receiver operating characteristic (ROC) curve analysis was employed to determine optimal diagnostic cut-off values by maximizing Youden’s index (J = sensitivity + specificity − 1), with the area under the ROC curve (AUC) and corresponding 95% CIs computed to quantify discriminative ability. Analyses used The PASW Statistics 18. p < 0.05 was considered statistically significant.

Results

Patient characteristics and clinical data

A total of 584 PAE samples from consecutive patients were analyzed, comprising 431 pleural effusions (73.8%) and 153 ascitic effusions (26.2%). The final cohort included 357 malignant cases (61.1%) and 227 benign cases (38.9%), with a male-to-female ratio of 1.05:1 (294 males, 290 females) and median age of 62 years (range 23–93 years).

Malignant PAE originated from diverse primary malignancies (Table 1): lung cancer predominated (60.2%, 215/357), followed by ovarian cancer (21.8%, 78/357), gastrointestinal cancers (14.6%, 52/357), and breast cancer (3.4%, 12/357). Other rare primary sites (n = 12) included pancreatic, renal cell, and unknown primary carcinomas.

Table 1.

The sensitivity, specificity and accuracy of CIC and HoCIC in pleural and ascitic effusion.

Markers	Diagnosis	Malignancy (n)	Benign (n)	Cases (n)	P	SEN (%)	SPE (%)	PPV (%)	NPV (%)	ACC (%)
CIC (n)	Pos (+)	312	19	331	0.000	87.4	91.6	94.3	82.2	89.0
	Neg (–)	45	208	253
HoCIC (n)	Pos (+)	276	3	279	0.000	77.3	98.7	98.9	74.3	85.6
	Neg (–)	81	224	305
Total(N)		357	227	584

Benign PAE encompassed multiple etiologies (Table 1): non-specific/other benign effusions (28.6%, 65/227) constituted the largest subgroup, followed by cardiogenic effusion (17.2%, 39/227), hepatic cirrhosis (16.7%, 38/227), inflammatory effusion (15.9%, 36/227), renal origin (14.1%, 32/227), and tuberculous pleuritis/peritonitis (7.5%, 17/227).

CIC and HoCIC prevalence in PAE samples

CIC structures were identified in 87.4% (312/357) of malignant PAE samples versus 8.4% (19/227) of benign cases (p < 0.001), yielding a 10.4-fold higher detection rate in malignancy (Figure 1). Quantitative analysis revealed markedly elevated CIC densities in malignant effusions: median 6.0 CICs/10 HPF (range 0–126; IQR 2.0–12.0) compared to median 0.0/10 HPF (range 0–5; IQR 0.0–0.0) in benign cases (Mann–Whitney U test, p < 0.001). The mean CIC count in malignant samples was (9.10 ± 10.42)/10 HPF versus (0.17 ± 0.67)/10 HPF in benign controls (Figure 1).

Figure 1.

Cell-in-cell (CIC) structure (red arrow) in HE-stained sections from cell blocks of malignant pleural effusions samples (Lung adenocarcinoma). The red arrow points to CIC, 40×.

HoCIC structures was detected in 77.3% (276/357) of malignant cases versus 1.3% (3/227) of benign cases (p < 0.001), representing a 59.5-fold enrichment in malignant effusions (Figures 2–4). Quantitative distribution showed median 2.0 HoCICs/10 HPF (range 0–113; IQR 0.0–5.0) in malignant samples versus median 0.0/10 HPF (range 0–2; IQR 0.0–0.0) in benign cases (p < 0.001); mean counts were (3.37 ± 9.20)/10 HPF versus (0.02 ± 0.16)/10 HPF, respectively (Figures 2–4).

Figure 2.

Cell-in-cell (CIC) structures in CK7 immunostaining sections from cell blocks of malignant PAE samples (Lung adenocarcinoma). The red arrow points to heterotypic CIC (HeCIC), CK7 negative cell engulfing CK7 positive cell; The blue arrow points to homotypic CIC (HoCIC), CK7positive cell engulfing CK7 positive cell. 40×.

Figure 4.

Cell-in-cell (CIC) structures in CD68 immunostaining sections from cell blocks of malignant PAE samples (Lung adenocarcinoma). The red arrow points to heterotypic CIC (HeCIC), CD68 positive cell engulfing CD68 negative cell; The blue arrow points to homotypic CIC (HoCIC), CD68 negative cell engulfing CD68 negative cell. 40×.

Figure 3.

Cell-in-cell (CIC) structures in TTF-1 immunostaining sections from cell blocks of malignant pleural effusions samples (Lung adenocarcinoma). The red arrow points to heterotypic CIC (HeCIC), TTF-1 negative cell engulfing TTF-1 positive cell; The blue arrow points to homotypic CIC (HoCIC), TTF-1 positive cell engulfing TTF-1 positive cell. 40×.

Comparative diagnostic discrimination: While CIC exhibited higher sensitivity (87.4% vs. 77.3%), HoCIC demonstrated markedly superior specificity (98.7% vs. 91.6%) and positive likelihood ratio (59.5 vs. 10.4), positioning HoCIC as a high-confidence confirmatory biomarker and CIC as a sensitive screening target.

Diagnostic performance of CIC and HoCIC

CIC demonstrated robust diagnostic discrimination for malignant PAE with sensitivity 87.4%, specificity 91.6%, and overall accuracy 89.0%. NPV was 82.2%, indicating reliable exclusion of malignancy when CIC was absent. HoCIC exhibited a distinct diagnostic profile: sensitivity 77.3%, specificity 98.7%, and accuracy 85.6%. PPV reached 98.9%, establishing HoCIC as a high-confidence confirmatory biomarker.

Comparative analysis revealed complementary strengths: CIC achieved significantly higher sensitivity (87.4% vs.77.3%) and NPV (82.2% vs. 74.3%), whereas HoCIC demonstrated markedly superior specificity (98.7% vs.91.6%). Both markers showed statistically significant independent discriminatory value (χ² test, p < 0.001 for each). Notably, overall accuracy did not differ significantly between CIC and HoCIC (89.0% vs. 85.6%, p = 0.127), indicating equivalent diagnostic utility despite divergent sensitivity-specificity trade-offs.

The discriminative performance of CIC and HoCIC was further quantified by receiver operating characteristic (ROC) curve analysis. CIC demonstrated excellent discriminative ability with an AUC of 0.912, significantly exceeding the 0.90 threshold for high-accuracy biomarkers. HoCIC achieved comparable performance with an AUC of 0.884. The AUC difference of 0.028 was not statistically significant (DeLong test, p = 0.187), confirming equivalent overall diagnostic validity (Table 2, Figure 5).

Table 2.

The AUC of using CIC, HoCIC as morphological markers for diagnosing malignant PAE.

Test result variable(s)	Area	Std. errora	Asymptotic Sig.b	Asymptotic 95% confidence interval
				Lower bound	Upper bound
CIC	0.912	0.013	0.000	0.888	0.937
HoCIC	0.884	0.014	0.000	0.856	0.911

The test result variable(s): CIC, HoCIC has at least one tie between the positive actual state group and the negative actual state group. Statistics may be biased. AUC, Area Under the Curve; CIC, Cell-in-cell; HoCIC, homotypic CIC; PAE, pleural and ascitic effusions;

^a Under the nonparametric assumption.

^b Null hypothesis: true area = 0.5.

Figure 5.

The ROC Curve of using CIC and HoCIC as morphological markers for diagnosing malignant PAE. ROC curve, receiver operating characteristic curve; CIC, Cell-in-cell; HoCIC, homotypic CIC; PAE, pleural and ascitic effusions.

Optimal diagnostic cut-offs were established by maximizing Youden’s index. Youden’s index maximization identified 0.5/10 HPF (1/20 HPF) as the optimal threshold for both (J = 0.790 and 0.760, respectively), yielding 87.4%/91.6% sensitivity/specificity for CIC and 77.3%/98.7% for HoCIC, underscoring their complementary clinical utilities (Table 2, Figure 5).

Discussion

Historical context and nomenclature evolution

Cell-in-cell (CIC) structures were first described over a century ago, yet only recent advancements in high-resolution microscopy and molecular characterization have reignited scientific interest in this phenomenon (19, 20). CIC typically depicts one viable cell completely internalized within another, characterized by a crescent-shaped nucleus in the host cell and a distinct vacuole containing the inner cell. Initially termed “cell cannibalism” by Leyden in 1925 (21), this phenomenon has been variably described as “cell phagocytosis,” “bird’s eye cells,” or “signet-ring cells” in historical literature (22). The terminology has evolved substantially, with “cannibalism,” “emperipolesis,” and “entosis” now representing distinct mechanistic pathways of cell engulfment (23).

In 2007, Overholtzer et al. established entosis as a non-apoptotic form of programmed cell death, wherein a living cell actively invades a neighboring host cell to form an internalized structure, ultimately undergoing lysosomal degradation or release (24–26). Conversely, emperipolesis—first defined by Humbel in 1956—describes lymphocyte invasion into tumor cells with preservation of both cellular identities (27). From a molecular classification perspective, cannibalism and entosis are categorized as homotypic CIC (HoCIC), whereas emperipolesis represents heterotypic CIC (HeCIC). HeCIC typically occurs in physiological and inflammatory states, involving histiocytes, megakaryocytes, monocytes, and fibroblasts. Although CIC structures—including both HoCIC and HeCIC—have been documented in normal tissues and non-neoplastic conditions (28, 29), CIC and HoCIC demonstrate marked enrichment in malignant contexts, with robust association to tumor progression, invasion, metastasis, and adverse prognosis. Due to modern cell visualization techniques, these characteristics establish CIC and HoCIC as independent prognostic and diagnostic biomarkers with significant translational potential.

Technical advancements: from cytological smears to cell block technology

A critical barrier to clinical implementation of CIC-based diagnosis has been technical limitations of conventional preparations. CIC structures are poorly visualized in traditional cytological smears or liquid-based monolayers due to cellular overlapping, three-dimensional clustering, and loss of architectural context. The advent of cell block technology has revolutionized diagnostic cytopathology by enabling high-quality histological sections, comprehensive immunohistochemical profiling, and standardized morphological assessment. Our study leverages this technical platform to demonstrate that CIC can be reliably quantified in H&E-stained cell block sections, unlocking its diagnostic potential for routine clinical practice.

Despite the accessibility of CIC detection in H&E preparations, distinguishing HoCIC from HeCIC remains challenging without molecular validation. To address this, we employed pan-cytokeratin (CK-pan), CD68, and CD45 IHC staining to definitively phenotype CIC structures—establishing HoCIC as cytokeratin-positive tumor cell-tumor cell interactions and HeCIC as cytokeratin-negative host cell-tumor cell engagements.

In general, a standard panel of epithelial markers (CK-pan, CK7, CEA, claudin-4) and macrophage markers (CD68) and mesothelial markers (calretinin, WT-1, D2-40) were routinely applied to cases with suspicious or inconclusive cytomorphology. Additionally, for cases with confirmed metastatic carcinoma, site-specific IHC markers (such as TTF-1 for lung, CDX2 for gastrointestinal tract, GATA3 for breast/urothelial, and PAX8 for renal/gynecological primaries) were applied when clinically indicated to identify the primary tumor origin. Conway AB et al. employed cell-in-cell morphology in the diagnostic evaluation of metastatic small cell carcinoma involving the thoracic cavity, which was further supported by immunoperoxidase staining showing tumor cell positivity for CD56, synaptophysin and TTF-1 (30).

This morpho-molecular approach enhances diagnostic specificity and aligns with precision medicine paradigms. We emphasized that our diagnostic approach aligns with standard cytopathological practice and that CIC detection serves as an adjunctive diagnostic marker rather than replacement IHC marker in the diagnostic workup.

Diagnostic performance and clinical implications

Our study reveals significantly elevated CIC and HoCIC densities in malignant PAE compared to benign effusions (9.10 ± 10.42 vs. 0.17 ± 0.67/10 HPF for CIC; 3.37 ± 9.20 vs. 0.02 ± 0.16/10 HPF for HoCIC; both p < 0.001). Notably, these counts exceed those reported in primary solid tumor tissues, likely reflecting unique biological pressures in the serous cavity microenvironment (31).

Metastasis represents a key biological hallmark of aggressive malignancy, driven by nutrient deprivation, hypoxia, and detachment-induced stress (32, 33). The serous cavity provides a specialized niche where tumor cells must adapt to anchorage-independent survival and limited nutrient availability (34). We hypothesize that CIC formation serves as an adaptive survival mechanism under these stressful conditions, enabling tumor cells to internalize neighbors for nutrient acquisition—analogous to “cellular cannibalism”. This biological imperative explains the elevated CIC prevalence in malignant effusions versus primary tumor tissues, and positions CIC detection as a functional biomarker of metastatic competence.

Meanwhile, we also observed a significant positive correlation between the number of CICs detected in pleural and peritoneal effusion samples and the total tumor cell count in these effusions—that is, cases with a greater number of tumor cells tended to have a higher number of CICs. Since the aim of this study was to assess the qualitative diagnostic value of CIC counts in distinguishing benign from malignant pleural and peritoneal effusions, we did not conduct an in−depth analysis of the correlation between CIC and tumor cell counts. In future studies, we plan to further investigate the clinical value of the ratio of CIC count to tumor cell count in pleural and peritoneal effusion samples, as well as its potential biological significance.

Complementary diagnostic strategy: CIC as screening, HoCIC as confirmation

Huang CC et al. have noted that although cell-in-cell structures and other cytological characteristics are diagnostic indicators for squamous cell carcinoma, poorly differentiated cases may show overlapping features with reactive mesothelial cells, mesothelioma, metastatic adenocarcinoma, or urothelial carcinoma in serous cavity effusions. Consequently, correlation with clinical history, a well-selected immunohistochemical marker panel, and cytomorphological features is imperative to secure an accurate diagnosis (35). Kazuki Nabeshima And Matsumoto S employed virtual microscopy with FISH to analyze and identify cell-in-cell engulfment as useful cytomorphological criteria for differentiating benign reactive mesothelial hyperplasia from malignant mesothelioma (36–38).

In this paper, Despite significant differences in sensitivity, specificity, PPV, and NPV, the overall diagnostic accuracy of CIC and HoCIC was comparable (89.0% vs. 85.6%, p = 0.127), indicating equivalent diagnostic utility despite differing sensitivity–specificity trade-offs. This was further supported by ROC curve analysis, while Youden’s index maximization revealed their complementary clinical utilities. Together, these findings support a tiered diagnostic algorithm in which each biomarker serves a distinct clinical role:

1. Primary screening: H&E-based CIC quantification enables rapid, cost-effective initial assessment, leveraging its high sensitivity (87.4%) to effectively exclude malignancy.

2. Secondary confirmation: IHC-validated HoCIC provides targeted validation for CIC-positive cases, capitalizing on its exceptional specificity (98.7%) to definitively rule in malignancy.

3. Clinical integration: Combined CIC/HoCIC scoring facilitates refined risk stratification and standardized reporting, optimizing diagnostic confidence across diverse clinical scenarios.

This sequential morpho-molecular approach maximizes resource efficiency by reserving IHC validation for cases where it delivers maximal incremental value, while maintaining >98% positive predictive value when HoCIC is detected. Such tiered strategies align with precision diagnostic paradigms in resource-variable settings, optimally balancing diagnostic accuracy with cost-effectiveness and workflow integration.

In contrast, CIC/HoCIC assessment requires only routinely stained cytological preparations (smears or cytospins) and cell block sections, which are prepared during the initial evaluation without additional tissue processing or expensive reagents. Therefore, although CIC/HoCIC detection cannot identify the tissue origin of tumors compared with IHC staining, it can complement IHC and provide important adjunctive diagnostic value in determining malignant effusions at initial diagnosis, particularly in:

1. Hospitals in remote areas or economically underdeveloped regions;

2. Cases with suboptimal cell block material or limited tumor cell numbers;

3. Intraoperative or urgent clinical scenarios requiring rapid preliminary diagnosis;

4. Patients who cannot afford high-cost tests such as IHC, or healthcare institutions with limited resources.

Standardization and future directions

Based on ROC-optimized Youden’s index, we recommend a unified cut-off of ≥1 structure per 20 HPF (0.5/10 HPF) for both CIC and HoCIC positivity. This convergent threshold—derived independently for each biomarker—simplifies laboratory implementation and facilitates multi-center validation.

Conclusion

CIC and HoCIC represent validated, readily implementable morphological biomarkers for distinguishing benign from malignant PAE diagnosis, with ≥1/20 HPF as the optimal threshold. The complementary performance profiles—CIC optimized for sensitive screening, HoCIC for specific confirmation—support a tiered morpho-molecular algorithm with >98% PPV in definitive clinical cytopathological diagnosis. Cell block technology enables this precision diagnostic approach, bridging traditional morphology with molecular phenotyping. Future multi-center prospective studies and digital pathology integration will further establish CIC/HoCIC quantification as standard-of-care biomarkers and potential therapeutic targets in malignant effusion management.

Study limitations

Several limitations warrant consideration. First, this single-center retrospective study requires external validation. Second, inter-observer variability in CIC recognition—though minimized by standardized criteria—may persist across less experienced pathologists. Third, the mechanistic basis for elevated CIC in effusions versus solid tissues remains incompletely characterized. Finally, longitudinal outcomes (survival, treatment response) were not assessed, limiting prognostic conclusions.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by grants from the Youth Fund of the National Natural Science Foundation of China (NSFC: 82102726).

Data availability statement

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

Ethics statement

The studies involving humans were approved by the biomedical Ethics Committee of Faculty of Medicine Xi’an Jiaotong University (No.2021.954). The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin due to the retrospective nature of the study and the use of anonymized residual paraffin-embedded tissue specimens, with no additional interventions or risks to patients.

Author contributions

RG: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Writing – original draft, Validation, Writing – review & editing. XL: Data curation, Formal analysis, Investigation, Resources, Visualization, Writing – original draft, Software. DL: Data curation, Investigation, Resources, Writing – original draft, Formal analysis. HH: Data curation, Investigation, Resources, Writing – original draft. CC: Data curation, Investigation, Resources, Writing – original draft, Formal analysis. XH: Resources, Methodology, Writing – original draft. WS: Methodology, Writing – original draft, Resources. JY: Conceptualization, Project administration, Supervision, Writing – review & editing, Visualization.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.