IntroductionNon-small cell lung cancer (NSLSC) stem cells (CSCs) have been shown to be responsible for bone metastasis by interacting with osteoclasts (OCs) and creating an immunosuppressive environment in the bone pre-metastatic niche. Now, we investigated the interaction among OCs, IL-15-stimulated NK cells and CSCs to understand whether NK cells can interfere with the pro-metastatic crosstalk between OCs and CSCs.MethodsIn vitro co-cultures of autologous OCs, NK cells and spheres enriched for CSC from NSCLC A549 cell line (A549/s) were set up both on plastic and bone slices, and OC activity was evaluated through quantification of tartrate-resistant acid phosphatase and of resorption area. The expression of NK cell receptors and ligands was studied through flow cytometry on NK cells, OCs and CSCs. The NK cell degranulation activity was investigated as CD107a expression.ResultsThe number of multinucleated/TRAP+ OCs and TRAP activity decreased when OCs were cultured with NK cells, and with NK cells+A549s. In presence of NK cells, A549s adhered to the bone slice surface and grew, suggesting that NK cells promote the growth of cancer cells rather than block it. Next, we studied whether NK cell phenotype and activity could be modulated by OCs and A549s. When NK cells were co-cultured with A549 cells, OCs, or with the combination of OCs and A549s, we observed a significant increase in the cytotoxic subset of CD56+CD16+ cells, a reduced expression of activating receptors (DNAM-1, NKp44) and an increased expression of inhibitory receptors (TIGIT, TIM3) on NK cells. The NK cell degranulation activity was inhibited by the presence of OCs. The addition of an anti-TIGIT antibody only partially reactivated NK cells, as indicated by a modest control of A549 cell growth, suggesting that NK cell exhaustion was induced by OCs.DiscussionAll together these data show that OCs negatively affect the NK cell cytotoxic activity, allowing the growth of NSCLC CSCs. Our findings reveal a previously unrecognized role of OCs in modulating the immune microenvironment by dampening NK cell function.

Circulating tumor cells (CTCs) sensing the chemokine CXCL12 and invading hyaluronic acid hydrogel (CXCL12 loaded hydrogel, CLG) might display metastatic attitude. CLG recovered-human lung (H460, A549), and ovarian cancer cells (IGROV-1) overexpressed the CXCL12 receptor, CXCR4, developed larger spheres and activate transcriptional programs of invasion and stemness. Interestingly, CLG-U87 glioblastoma cells highly expressed EpCAM and significantly upregulated the very specific NGFR, NTS, AQP1 and CMKLR1 genes, associated with cell proliferation, migration/invasion and metastasis. In a syngeneic model, subcutaneous CLG significantly attracted GFP-Lewis lung carcinoma (LLC) cells impairing lung colonization within four hours from cell injection and up to twenty-one days. In lung, M1 macrophages rapidly increased post cancer cells injection while M2 prevailed after 10 days (T10). Neutrophils (Ly6Ghigh and Ly6Glow) infiltrated the lungs after ten days from cells injection with late expansion of immature Ly6Glow. According to lung niche, Empty gel (EG) and CLG were early infiltrated by macrophages and later by neutrophils. CLG generated an immunosuppressive environment defined by M1/M2/Ly6Glow able to capture and divert metastatic CTCs from the lung colonization.

Background Metastatic recurrence represents the major clinical challenge in early-stage lung cancer after curative surgery. Here, we investigated the role of circulating extracellular vesicles and particles (EVPs) in promoting formation of pre-metastatic niches (PMNs).Methods Plasma-derived EVPs were obtained by ultracentrifugation from pre-surgery blood samples of patients with poor prognosis. Heavy-smokers cancer free individuals were used as control. EVP were characterized following MISEV guidelines. Functional experiments were carried out in vitro in 2D and 3D-bioprinted models as well as in vivo.Results EVPs from patients with early relapse show distinct molecular profiles, characterized by elevated levels of miR-29a and complement protein C4a. These EVPs preferentially target endothelial cells inducing a pro-inflammatory condition with upregulation of VCAM1 and CXCL1. In turn, endothelial modulation stimulated fibroblast activation and promoted neutrophils recruitment supporting PMNs formation. Mechanistically, we demonstrate that miR-29a and C4A act synergistically through SPARC down-modulation promoting cancer cell colonization. Preconditioning of mouse lungs using EVPs from patients with poor prognosis increased metastatic growth of human tumor cells, which was inhibited by miR-29a blockade.Conclusions Circulating EVPs could be novel prognostic biomarkers and key players in PMN formation offering new targets to reduce relapses in lung cancer.

Circulating tumor cells (CTCs) are cells that shed from primary tumors into the bloodstream, leading to the formation of metastases. Their presence and numbers reflect the tumor burden, making them valuable biomarkers. Consequently, a simple blood draw can be used to non-invasively monitor tumor progression and treatment efficacy, a rapidly growing approach known as liquid biopsy [1]. However, the extreme rarity and heterogeneity of CTCs pose significant challenges for their identification, requiring methods that are often time-consuming, complex, and expensive.

Lung cancer is the primary cause of cancer-related deaths, with a five-year overall survival rate of 15 %, mainly related to late diagnosis at advanced stages. Despite recent improvements in cancer medicine, metastatic disease still, in fact, represents the main challenge for effective therapies. Targeting disseminated cells requires a better understanding of the underlying biological mechanisms and a paradigm shift to devise innovative treatments. Nanomedicine could open up new perspectives in cancer management, as nanoparticles offer many advantages in drug delivery due to their intrinsic properties, such as better bioavailability and the potential for selective targeting. A strategy to enhance delivery at the site of action is to use nanomedicines targeting specific receptors overexpressed in cancer cells or the tumor microenvironment. We recently identified TAS1R3 as an ideal target, enriched in metastatic lung cancer cells and highly expressed in the CD133+ cancer stem cell (CSC) subset. We developed and evaluated a targeted formulation of sphingomyelin nanoemulsions functionalized with a TAS1R3-specific aptamer (SN-5F). SN-5F exhibited high stability in both aqueous media and biofluids and was efficiently internalized by cell models overexpressing the targeted receptor, including the usually hard-to-target CD133+ CSC population. SN-5F was evaluated in 2D and 3D cell cultures, as well as in vivo using patient-derived xenografts, where it successfully reached disseminated cancer cells and delivered hydrophobic drugs to CSCs, resulting in reduced tumor-forming potential. These results highlight the promise of rationally designed, receptor-targeted nanomedicines as a novel approach for anti-metastatic lung cancer therapy.

Background Due to their enhanced responsiveness and persistence, cytokine-induced memory-like (CIML)-natural killer (NK) cells have emerged as new immunotherapeutic tools against malignancies. However, their effects on tumor-cell spread and metastases in solid tumors remain poorly investigated. Moreover, a clear identification of the most effective CIML-NK subsets, especially in controlling cancer stem cells (CSC), is still lacking.Methods We performed combined phenotypical and functional analyses of CIML-NK cell subsets, either selected by flow-cytometry gating, or generated from sorted CD56bright/CD56dim NK cells.By co-culture experiments, we analyzed the effect of CIML-NK cells on non-small cell lung cancer (NSCLC) cell spheroids, or patient-derived xenografts (PDX), assessing changes in their CSC content, tumorigenicity, and/or tumor disseminating capability in vivo. CIML-NK cells were also infused in PDX-bearing mice to validate their effect on the CSC dissemination from the PDX to the lungs.Finally, we generated and functionally analyzed CIML-NK cells from patients with stages I/III NSCLC (n=6).Results We show that CIML-NK cells exert antitumor activity mostly through their CD56bright cell subset, which greatly expands during CIML differentiation. Compared with NK cells conventionally activated with interleukin-2, CIML-NK cells express lower levels of check-point receptors, TIGIT and TIM3, and higher effector functions against NSCLC cells from PDX, and against in vitro-generated tumor spheroids. Remarkably, CIML-NK cells also significantly reduce the CSC-containing CD133+ cell subpopulation within spheroids and PDX, and limit tumor cell tumorigenicity and ability to disseminate CSCs from primary tumors to distant sites. Sorting experiments on CIML or tumor cell subsets reveal that CD56bright cells drive most of this anti-CSC activity, and suggest that such functional advantage could be related to increased expression of LFA-1 and ICAM-1 on CD56bright cells and CSCs, respectively. We also show that the tri-specific killer cell engager (TriKE) 1615133 significantly enhances CIML-NK cell activity against CSCs. Finally, we demonstrate that CIML-NK cells, capable of killing autologous tumor cells and responding to the 1615133 TriKE, could be induced from patients with NSCLC.Conclusions Our study discloses for the first time the therapeutic potential of CIML-NK cells in controlling CSCs and metastatic spread, highlighting the role of the CD56bright subset expansion and 1615133 TriKE for optimizing CIML-NK-based therapies against metastatic tumors.

and Purpose: Biomarkers for immunotherapy in NSCLC remain inadequate, with PD-L1 continuing to underperform despite its ongoing use. "Ghost biomarkers", such as circulating immune profiling, hold potential to reveal host immunity dynamics and identify ways to enhance immune fitness for improved therapeutic outcomes. This study uses single-cell technologies to analyse circulating immune cells in stage IV NSCLC patients (pts) receiving first line mono-immunotherapy (IO) or chemo (CT)-IO, aiming to identify subpopulation who might benefit or not from treatment. Blood samples from 33 stage IV NSCLC pts were collected at the baseline of first-line IO or CT-IO, as part of the Horizon Europe-funded I3LUNG study (NCT05537922) conducted at IRCCS Istituto Nazionale dei Tumori of Milan. Single-cell RNA sequencing was performed by Chromium Single Cell 5’ v2 kit (10X Genomics) on PBMCs, enriched for neutrophils sorted by FACS (CD66b+CD15+). After quality control, 30 samples were included in the analysis. Among 30 pts, 27 were treated with CT-IO and 3 with mono-IO, 25 were adenocarcinoma, 4 squamous and 1 NOS carcinoma. UMAP analysis revealed distinct "swan neck" clusters of immature neutrophils, predominantly expressing LCN2 gene, in pts with disease progression (p-value <0.05, logFC -3, 67). These cells were notably absent in pts with complete response or partial response. Conversely, responders showed the enrichment of a mature subset of IGHM TCL1A, IGHD expressing B cells ("smart” B cells) (all genes with p-value <0.05 and logFC >2.5). Of note, TCL1A gene is involved in the formation of tertiary lymphoid structures, which are positively correlated with IO responses. Although based on a very small cohort in which no progressive disease occurred, we observed T and NK cells enrichment in mono-IO treated pts with high PD-L1 expression. To our knowledge, this is the largest prospective cohort that analyses circulating single-cell PBMCs in stage IV NSCLC pts treated with a first-line IO based. We identified novel circulating immune subsets, such as "swan neck" neutrophils and "smart” B cells, which emerged as promising host biomarkers. These subsets consistently acted as negative and positive predictors, respectively to first-line IO-based therapy. Arsela Prelaj, Monica Ganzinelli, Alessandro Guidi, Chiara Cavalli, Cecilia Silvestri, Antonino Belfiore, Laura Mazzeo, Claudia Proto, Mario Occhipinti, Marta Brambilla, Teresa Beninato, Leonardo Provenzano, Luca Inverizzi, Giulia Bertolini, Daniele Lorenzini, Filippo De Braud, Luca Agnelli, Andrea Vingiani, Sabina Sangaletti, Giuseppe Lo Russo. Unmasking host immune complexity: Single-cell transcriptomic analysis of circulating PBMCs in non-small cell lung cancer treated with immunotherapy based first line - I3LUNG study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7146.

Circulating tumor cells are a key biomarker in liquid biopsy, offering a noninvasive approach to monitor and guide therapeutic decision in cancer patients. The extreme rarity and the heterogeneity of CTCs make their identification extremely challenging. To overcome these limitations, CTC can be isolated exploiting an unbiased workflow based on Parsortix® strategy, which enriches for CTC according to their size and deformability, in combination with DEPArray ${ }^{\text {TM }}$ technology that allows visualization and selection of single CTC. Then, expert physicians manually analyze the digital images of cells acquired with DEPArray technology. The introduction of an automatic system for CTC identification into clinical workflows could facilitate the early detection of metastases, optimize treatment decision-making, and improve patient outcomes. In this paper, we present a Deep Learning-based classification pipeline to distinguish CTCs from leukocytes in a liquid biopsy, focusing on improving diagnostic procedure and accuracy. The proposed method automatically classifies images acquired with DEParray technology, obtained at “Fondazione IRCCS-Istituto Nazionale dei Tumori” of Milan. It is based on ResNet architecture, a Convolutional Neural Network for image analysis and very popular in the medical field. The introduction of a carefully designed augmentation process allows incrementing data variability, obtaining improving in the model performance. The proposed method achieves an F1 score of $0,777 \pm 0,012$. The obtained promising results encourage us to do future research in this field.

Colorectal cancer is the third leading cause of death in developed country, especially in the presence of metastases. Metastatic cancer cells depend on antioxidants like glutathione (GSH) to tackle oxidative stress as they adapt to distinctive environments different from those of the primary tumor. The mechanisms that cause oxidative stress in cancer cells are poorly understood, but it was recently discovered that the Chloride Intracellular Channel 1 (CLIC1) protein supports tumor cell Reactive Species of Oxigen (ROS) overproduction. CLIC1 is a ubiquitous protein from the CLICs family, which is unique in its ability to be metamorphic. In normal cells, CLIC1 protein is expressed as a cytosolic form and binds antioxidant glutathione (GSH). Under the chronic stress typical of cancer cells, CLIC1 undergoes a conformational change, translocating to the plasma membrane (tmCLIC1). This transition releases GSH into the cytoplasm while tmCLIC1 facilitates reactive oxygen species (ROS) overproduction, fostering an aggressive cancer phenotype. Studies reveal a contradiction where cancer cells simultaneously exhibit high levels of both ROS and GSH, which are usually inversely related. In this work, we demonstrate that this paradox is linked to tmCLIC1 activity, where GSH release acts as a defense mechanism, while tmCLIC1 overexpression amplifies ROS production, driving cancer cell proliferation and metastasis. Blocking tmCLIC1 activity has shown a reduction of proliferation, migration, and invasion of metastatic colorectal cancer cells (CRC). The study also demonstrated that forcing non-metastatic CRC cells (SW480) to release GSH using the antioxidant N-Acetyl Cysteine (NAC) enriched cytoplasmic GSH and increased tmCLIC1 expression. This effect, however, was absent in cells where CLIC1 was silenced, proving that GSH release depends on tmCLIC1 activity. These cells were converted from non-aggressive cell lines to metastatic ones in vitro and in vivo. These findings challenge the traditional view of antioxidants in cancer treatment, suggesting that tmCLIC1 inhibition could serve as a novel therapeutic strategy against metastatic colorectal cancer, addressing a pressing societal health concern. Francesca Cianci, Guido Rey, Carlotta Tacconi, Luca Palloni, Matteo Ranucci, Giulia Bertolini, Luca Roz, Alessandro Fantin, Michele Mazzanti. Oxidative stress increases the aggressiveness of colorectal cancer by activating the antioxidant transmembrane chloride intracellular channel 1 (tmCLIC1). [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1507.

Circulating Tumor Cells (CTCs) are crucial biomarkers in liquid biopsy, offering a noninvasive tool for cancer patient management. However, their identification remains particularly challenging due to their limited number and heterogeneity. Labeling samples for contrast limits the generalization of fluorescence-based methods across different hospital datasets. Analyzing single-cell images enables detailed assessment of cell morphology, subcellular structures, and phenotypic variations, often hidden in clustered images. Developing a method based on bright-field single-cell analysis could overcome these limitations. CTCs can be isolated using an unbiased workflow combining Parsortix® technology, which selects cells based on size and deformability, with DEPArray™ technology, enabling precise visualization and selection of single cells. Traditionally, DEPArray-acquired digital images are manually analyzed, making the process time-consuming and prone to variability. In this study, we present a Deep Learning-based (DL) classification pipeline designed to distinguish CTCs from leukocytes in blood samples, aimed to enhance diagnostic accuracy and optimize clinical workflows. Our approach employs images from the bright-field channel acquired through DEPArray technology leveraging a ResNet-based Convolutional Neural Network. To improve model generalization, we applied three types of data augmentation techniques and incorporated fluorescence (DAPI) channel images into the training phase, allowing the network to learn additional CTC-specific features. Notably, only bright-field images have been used for testing, ensuring the model's ability to identify CTCs without relying on fluorescence markers. The proposed model achieved an F1-score of 0.798, demonstrating its capability to distinguish CTCs from leukocytes. These findings highlight the potential of DL in refining CTC analysis and advancing liquid biopsy applications.

Abstract Epithelial-mesenchymal transition (EMT) is a crucial biological process during development which also appears to play a fundamental role in the dissemination of cancer cells. It is known that hybrid states (H), in which cells have mixed epithelial (E) and mesenchymal (M) properties, are particularly conductive to metastasis but the possibility of seeding secondary lesions is also closely linked to the intrinsic tumor initiation potential (stemness). An overall vision taking into account both the EMT spectrum and stemness properties could therefore be the key to fully understanding the metastatic process. In this study we aim to build a mathematical model capable of describing the transitions in the EMT spectrum distinguishing stem from non-stem cell subpopulations in lung cancer cells. Using E-cadherin and N-cadherin staining we evaluated experimentally by flow-cytometry the modulation of epithelial (E), hybrid (H) and mesenchymal (M) phenotypes in A549 lung cancer cells after the induction of EMT by TGFβ treatment and subsequent return to steady state after TGFβ withdrawal. Stem and non-stem populations were identified with the previously validated CD133 marker. To mathematically describe the cellular dynamics we used a system of differential equations assuming transitions E-H, E-M and H-M, for stem and non-stem compartments initially treated as decoupled. Parameters used in the equations were the three switching rates corresponding to the three transitions allowed. To test the robustness of the model we solved the equation multiple times (500) by first stochastically changing the initial conditions and then the initial parameters, obtaining the respective parameters optimized by minimizing the residual between experimental data and model. To decouple the parameters, equations were then solved stochastically 1000 times changing two of three initial parameters, obtaining the optimized third parameter and the respective minimal residual. Parameters fitting closely the experimental data during TGFβ-mediated EMT induction were obtained, indicating that the transition from H to M appears to be absent in the non-stem population but present in the stem population, while the transition from E to M is always present. We also observed that after TGFβ withdrawal the inclusion of an epigenetic decay factor is necessary to model the latency period before MET insurgence when EMT is previously induced in the system. We can conclude that during EMTinduction, the dynamics in the non-stem compartment can be described by a bifurcation model in which population E becomes either M or H and remains permanently in that state, stressing the stability of the H state and also that E cells might have a predisposition to transform into M or H. Furthermore, a portion of H stem cells is able transform into M during EMT and therefore appears more plastic which could underscore a greater metastatic potential. Citation Format: Arianna Di Bernardo, Sarthak Sahoo, Melissa Balsamo, Elsa Quaranta, Mohit Kumar Jolly, Giulia Bertolini, Luca Roz. Modeling the dynamics of lung cancer cells along the epithelial−mesenchymal spectrum in the stem and non-stem cell compartments [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7376.

Abstract PURPOSE and INTRODUCTION: In vivo a dermal filler hyaluronic gel-based loaded with CXCL12 (CLG) was able to divert B16-hCXCR4 cells from lung metastases. Herein, CLG was assessed for the capability to isolate human circulating cancer cells (CTCs). Moreover, in vivo CLGs and lungs were characterized to dissect natural (lung) and artificial (gel) microenvironment composition. EXPERIMENTAL DESIGN: “TRAP4MET” clinical trial was conducted in 48 advanced cancer patients characterized at diagnosis for CLG-dependent CTCs-isolation as compared to ScreenCell™ filters. C57/B6 mice were s.c. injected with Empty Gel (EG) or CLG and five days later i.v. injected with GFP-LLC (Lewis lung) cells. Lungs and gels were collected at time 0 (before tumour cells injection), after 4 hours and 10 days post tumour cells inoculation. Lungs and gels were analysed through flow-cytometry for GFP-LLC cells, innate and adaptive immunity. RESULTS: In TRAP4MET clinical trial CLG-CTCs were isolated in 8/8 patients with ovarian (OC), 6/8 with lung (LC), 6/8 with colorectal (CRC), 8/8 with endometrial (EC), 8/8 with renal (RCC) cancer and 5/8 with glioblastoma (GBM). In OC, LC and GBM, CLG isolated more CTCs than the conventional ScreenCell™ (CLG/SC ratio=1.88 for OC, 2.47 for LC and 11.89 for GBM). To dissect the in vivo efficacy of CLG, GFP-LLC were i.v. injected in C57/B6 mice five days later the s.c CLG or EG injection. Five days after CLG/EG gels and lungs were recovered. In CLG a lower % of total Macrophages (MΦ), inactive/precursor Tregs and higher % of M2-MΦ was observed compared to EG while no major differences were revealed in lungs from CLG/EG/CTRL groups. 4 hours post injection revealed in CLG a lower % of MΦ, lower inactive/precursor Tregs and a higher % of M2- MΦ and CXCR4+ MΦ versus EG. In correspondent CLG-lungs, lower % of mature neutrophils and inactive/progenitor Tregs as compared to EG-lung and CTRL-lung, respectively. 10 days post cells inoculation, CLG gels revealed again low total MΦ, low inactive/precursor Tregs and high M2-MΦ, CXCR4+ MΦ as to EG. The corresponding CLG-lungs displayed higher non-aged neutrophils and and NK cells, lower CXCR4+ MΦ, lower total neutrophils, lower aged (CXCR4+) neutrophils and lower inactive Tregs as compared to EG and CTRL, respectively. Consistently, GFP-LLC cells were higher in CLG compared to EG at either 4 hours and 10 days post cell injection while reduced in lungs of CLG-mice compared to EG- and CTRL-lungs mice at 4 hours and 10 days post cell inoculation, respectively. CONCLUSION: CLG may support OC, LC and GBM- CTC counting in cancers at today orphan of CTCs reliable methods. In vivo, CLG attracted GFL-LLC cells while reducing lung GFP-LLC cells as early as after 4 hours post cell inoculation. CLG/EG and correspondent lung analysis revealed an immunosuppressive microenvironment within CLG compared to EG reduced in the corresponding lungs. Citation Format: Giuseppe Guardascione, Luigi Portella, Dario Guido Di Febbraro, Giulia Bertolini, Giuseppina Rea, Caterina Ieranò, Crescenzo D'Alterio, Maria Napolitano, Sara Santagata, Anna Maria Trotta, Emilia Scarpa, Sabrina Chiara Cecere, Alessandro Ottaiano, Giuliano Palumbo, Alessandro Morabito, Teresa Somma, Roberto Pacelli, Sandro Pignata, Stefania Scala. CXCL12-loaded-hydrogel (CLG) in vivo modifies lung metastatic niche toward an immunoactive microenvironment reducing lung metastasis development [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6815.

Circulating tumor cells (CTC), released by primary tumors into the bloodstream, represent a valuable source to inform on cancer heterogeneity, cancer progression, metastatic disease and therapy efficacy without the need of invasive tumor biopsies. However, the extreme rarity and heterogeneity of CTCs, occurring at genotypic, phenotypic and functional levels, poses a major challenge for the study of this population and explains the lack of standardized strategies of CTC isolation. Lung cancer, the leading causes of cancer-related death worldwide, is a paradigmatic example of how CTC heterogeneity can undermine the clinical utility of this biomarker, since contrasting data have been reported using different isolation technologies. Some evidences suggest that only a fraction of CTC, characterized by stem-like feature and partial epithelial-mesenchymal transition (EMT) phenotype, can sustain metastasis initiation. Cancer stem cells (CSCs) have the potential to maintain primary tumors, initiate metastasis and escape both chemotherapy and immunotherapy treatments. Moreover, a close connection has been reported in several tumor types among hybrid phenotype, characterized by retention of epithelial and mesenchymal traits, acquisition of CSC feature and increased metastatic potential. This review focuses on the phenotypic and functional heterogeneity of CTCs and the resulting implications for their isolation and clinical validation, especially in the setting of non-small cell lung cancer (NSCLC). In particular, we discuss the most relevant studies providing evidence for the presence and prognostic/predictive value of CTC subsets characterized by stem-like and hybrid EMT phenotype. Despite technical and conceptual issues, tracking circulating CSCs has the potential to improve the prognostic/predictive value of CTCs in NSCLC setting and could provide novel insights into the comprehension of the metastatic process and identification of novel therapeutic targets.

Background Epithelial to mesenchymal transition (EMT) endows cancer cells with pro-metastatic properties, which appear most effective when cells enter an intermediate hybrid (H) state, characterized by integrated mesenchymal (M) and epithelial (E) traits. The reasons for this advantage are poorly known and, especially, it is totally unexplored whether the interplay between H-cells and NK cells could have a role. Here we characterize the pro-metastatic mechanics of non-small cell lung cancer (NSCLC) H-cells and their subset of cancer-initiating cells (CICs), dissecting crucial interactions with NK cells.Methods Human lung cancer cell lines and sublines representative of E, M, or H states, assessed by proteomics, were analyzed in vivo for their tumor-forming and disseminating capabilities. Interactions with NK cells were investigated in vitro using migration assays, cytotoxic degranulation assays, and evaluation of CD133+ CICs modulation after coculture, and validated in vivo through NK cell neutralization assays. Correlation between EMT status, NK cell infiltration, and survival data, was evaluated in a cohort of surgically resected NSCLC cases (n=79).Results We demonstrated that H-cells, have limited dissemination capability but show the highest potential to initiate metastases in vivo. This property was related to their ability to escape NK cell surveillance. Mechanistically, H-cells expressed low levels of NK-attracting chemokines (CXCL1 and CXCL8), generating poorly infiltrated metastases. Accordingly, proteomics and GO enrichment analysis of E, H, M cell lines showed that the related secretory processes could change during EMT.Furthermore, H-CICs uniquely expressed high levels of the inhibitory ligand B7-H3, which protected H-CIC from NK cell-mediated clearance. In vivo neutralization assays confirmed that, indeed, the pro-metastatic properties of H-cells are poorly controlled by NK cells.Finally, the analysis of patients revealed that detection of hybrid phenotypes associated with low NK infiltration in NSCLC clinical specimens could identify a subset of patients with poor prognosis.Conclusions Our study demonstrates that H-cells play a central role in the metastatic spread in NSCLC. Such pro-metastatic advantage of H-cells is supported by their altered interaction with NK cells and by the critical role of B7-H3 in preserving their H-CIC component, indicating B7-H3 as a potential target in combined NK-based therapies.

Background:Circulating Tumor Cells (CTCs) represent a small, heterogeneous population that comprise the minority of cells able to develop metastasis. To trap and characterize CTCs with metastatic attitude, a CXCL12-loaded hyaluronic-gel (CLG) was developed. CXCR4+cells with invasive capability would infiltrate CLG. Methods:Human colon, renal, lung and ovarian cancer cells (HT29, A498, H460 and OVCAR8 respectively) were seeded on 150 μl Empty Gels (EG) or 300 ng/ml CXCL12 loaded gel (CLG) and allowed to infiltrate for 16 h. Gels were then digested and fixed with 2 % FA-HAse for human cancer cell enumeration or digested with HAse and cancer cells recovered. CLG-recovered cells migrated toward CXCL12 and were tested for colonies/spheres formation. Moreover, CXCR4, E-Cadherin and Vimentin expression was assessed through flow cytometry and RT-PCR. The clinical trial "TRAP4MET" recruited 48 metastatic/advanced cancer patients (8 OC, 8 LC, 8 GBM, 8 EC, 8 RCC and 8 EC). 10 cc whole blood were devoted to PBMCs extraction (7 cc) and ScreenCell™ filters (3 cc) CTCs evaluation. Ficoll-isolated patient's PBMCs were seeded over CLG and allowed to infiltrate for 16 h; gels were digested and fixed with 2 % FA-HAse, cells stained and DAPI+/CD45-/pan-CK + cells enumerated as CTCs. Results:Human cancer cells infiltrate CLG more efficiently than EG (CLG/EG ratio 1.25 for HT29/1.58 for A498/1.71 for H460 and 2.83 for OVCAR8). CLG-recovered HT29 cells display hybrid-mesenchymal features [low E-cadherin (40 %) and high vimentin (235 %) as compared to HT29], CXCR4 two-fold higher than HT29, efficiently migrate toward CXCL12 (two-fold higher than HT29) and developed higher number of colonies (171 ± 21 for HT29-CLG vs 131 ± 8 colonies for HT29)/larger spheres (spheroid area: 26561 ± 6142 μm2 for HT29-CLG vs 20297 ± 7238 for HT29). In TRAP4MET clinical trial, CLG-CTCs were isolated in 8/8 patients with OC, 6/8 with LC, 6/8 with CRC, 8/8 with EC, 8/8 with RCC cancer and 5/8 with GBM. Interestingly, in OC, LC and GBM, CLG isolated higher number of CTCs as compared to the conventional ScreenCell™ (CLG/SC ratio = 1.88 for OC, 2.47 for LC and 11.89 for GBM). Bland and Altman blot analysis and Passing and Bablok regression analysis showed concordance between the methodological approaches but indicate that SC and CLG are not superimposable suggesting that the two systems select cells with different features. Conclusion:CLG might represent a new and easy tool to isolate invasive CTCs in multiple cancers such as OC, LC and GBM at today orphan of reliable methods to consistently detect CTCs.