The Prognostic Role of Circulating Tumor Cells (CTCs) in Lung Cancer

Lung cancer affects over 1. 8 million people worldwide and is the leading cause of cancer related mortality globally. Currently, diagnosis of lung cancer involves a combination of imaging and invasive biopsies to confirm histopathology. Non-invasive diagnostic techniques under investigation include “liquid biopsies” through a simple blood draw to develop predictive and prognostic biomarkers. A better understanding of circulating tumor cell (CTC) dissemination mechanisms offers promising potential for the development of techniques to assist in the diagnosis of lung cancer. Enumeration and characterization of CTCs has the potential to act as a prognostic biomarker and to identify novel drug targets for a precision medicine approach to lung cancer care. This review will focus on the current status of CTCs and their potential diagnostic and prognostic utility in this setting.


INTRODUCTION
Lung cancer is the leading cause of cancer-related mortality among men and women worldwide (1). In 2012, the incidence was estimated at 1.8 million new cases, accounting for 12.9% of all new cancers diagnosed globally (2). There is an estimated 18% survival rate beyond 5 years for all stages combined, with poor outcomes largely due to late diagnosis (1,3). The majority of patients present with locally advanced or metastatic disease, with ∼20-30% of patients presenting with early stage disease (3,4). Late diagnosis is a major underlying cause for this advanced disease presentation (5). The annual mortality rate for lung cancer is higher than for colon, breast, and prostate cancers combined (6). The majority of patients presenting with advanced stage at diagnosis contributes to this poor outcome (4).
There are two main types of lung cancers, small cell lung carcinoma (SCLC) and non-small cell carcinoma (NSCLC). NSCLC is the most common, accounting for 80% of cases (7). NSCLC has three main histological subtypes: adenocarcinoma, squamous cell (epidermoid) carcinoma, and large cell undifferentiated carcinoma. Adenocarcinoma accounts for ∼40% of cases although is increasing in relative incidence, and usually starts in mucus secreting epithelial cells (167). The prognosis of NSCLC subtypes depends on the stage of the tumor and the treatment availability.
Small cell lung cancer (SCLC) accounts for about 15% of all lung cancers diagnosed annually and up to 25% of lung cancer deaths. SCLC is characterized by a more aggressive clinical phenotype than NSCLC with progression to metastatic disease earlier in the disease course (8).
SCLC and NSCLC arise from different cell types and demonstrate varying clinical features as shown in Table 1.
Lung cancer may be initiated through exposure to carcinogens. The main risk factor for lung cancer is the use of tobacco. Tobacco is known to initiate and promote carcinogenesis and accounts for 85% of lung cancer cases (9). Additional known risks include exposure to pollutants such as asbestos, tar and metals including arsenic, and chromium. Common symptoms include persistent cough, worsening breathing, pneumonia that fails to resolve, chest discomfort, wheezing, blood in the sputum, and hoarseness (3,10). A minority are asymptomatic, detected by chance through investigation of other illnesses or in screening programs (11).
Treatment options depend on the intent of treatment and may include loco-regional treatment such as surgery, image guided ablation including radical chemo-radiotherapy, stereotactic ablative radiation treatment, thermal ablation or cryotherapy, or systemic treatment such as chemotherapy, targeted agents, and immunotherapy, alongside novel agents under current investigation in clinical trials (11). An example of the power of targeted therapies in a precision medicine approach was demonstrated in 2004 by Lynch et al. (12) and Paez et al. (13) who demonstrated that patients with EGFR mutations present in the tumors of patients with non-small cell lung cancer exhibited a dramatic response to getfitinib, the epidermal growth factor (EGFR) tyrosine kinase inhibitor (TKI), bringing personalized medicine to reality for a subset of NSCLC patients (12,13). Utilization of expensive systemic targeted therapies, however, has traditionally required invasive biopsies in order to assess for targetable tumoral aberrations. This presents a challenge for the monitoring of lung cancers due to the requirement for longitudinal sampling of tumors (14).

METASTASIS AND EPITHELIAL-MESENCHYMAL TRANSITION
Metastasis is an extremely complex, multistep process. Cells must gain the ability to intravasate into the blood from the bulk tumor, travel through the blood undergoing sheer stressors and immune evasion, and extravasate to favorable metastatic sites such as bone, brain and liver (15)(16)(17). In order to detach from the primary tumor and disseminate into the blood, cells must undergo a cellular process known as epithelialmesenchymal transition (EMT) (18). EMT enables tumor cells to become motile and enhances migratory capabilities which in effect allows cells to penetrate into the lymph vasculature and circulate as single or clusters of circulating tumor cells (CTCs) (19). Whilst in blood, CTCs exist in a dynamic EMT state (20). CTCs extravasate having undergone the reverse process known as mesenchymal to epithelial transition (MET) and colonize at distant organs, (21). EMT is thought to support cell invasiveness but restrict proliferation, thereby maintaining cancer cell survival in metastatic sites whereas MET re-activates proliferative potential (22). The famous "seed and soil" hypothesis proposed by Stephen Pagent in the Nineteenth century suggesting that tumor cells (the "seed") have a preference to metastasize in certain organs (the 'soil) (23). This hypothesis has since been revisited by Fidler and Langly, still holding significance in cancer research today (24,25).

CIRCULATING TUMOR CELLS IN LUNG CANCER
CTCs were first described by an Australian physician, Thomas Ashworth in 1869, where cancer cells in the blood were observed which resembled the cells of the primary tumor (26). CTCs play a central role in the metastatic spread of lung cancer, that is ultimately responsible for patient morbidity and mortality from the disease (27). While the concept of CTCs were described over one hundred years ago, it is only recently that they have been utilized in cancer diagnosis and prognosis (28).
Evidence has shown that the presence of CTCs in the blood correlates with poor overall survival in patients with metastatic prostate, breast and colon cancers (29)(30)(31). Patients with SCLC have on average 10 times more CTCs than patients with any other tumor type (32)(33)(34).
Molecular targeted therapies such as tyrosine kinase inhibitors (TKIs) in epidermal growth factor receptor (EGFR) mutants and anaplastic lymphoma kinase (ALK) inhibitors in ALK rearranged NSCLC patients have recently advanced the management of lung cancer for a limited proportion of patients (35)(36)(37)(38)(39). To determine eligibility for such targeted therapies, tumor biopsies have traditionally been necessary, increasing the likelihood of biopsyrelated complications (40). Even in patients developing resistance to first line EGFR TKIs, liquid biopsies using circulating tumor DNA plasma only detect T790M mutations in ∼80% of cases, particularly in low volume disease, making a repeat biopsy necessary. Tumor heterogeneity within the primary site or between primary and metastatic sites, can also create potential sampling bias, which may mask the true genetic profile of the cancer. The prospect of longitudinal sampling in order to monitor for the development of therapeutic resistance to treatments is likewise limited if invasive biopsies are essential (41,42).
Use of CTCs as a liquid biopsy is promising for serial assessment of tumor evolution during the course of the disease and during systemic treatment in a less invasive, real-time manner, by a simple blood draw (19,43). This liquid biopsy also provides potential for the early diagnosis of cancer and valuable insights into tumor heterogeneity and genomic diversity for the early diagnosis of cancer and guidance of clinical treatment (44,45). A sensitive and unbiased isolation method to capture CTCs is therefore essential to provide tumoral material for analysis and potentially drive treatment decisions (46,47).

CIRCULATING TUMOR CELL DETECTION METHODS IN LUNG CANCER
CTCs have the potential to accompany standard screening tests and be used for molecular characterization of a tumor (48). Detection of CTCs in NSCLC has been challenging due to the rarity in circulation (a few CTCs per billion normal blood cells) and the presence of non-epithelial characteristics (49). It is therefore imperative that sensitive and specific CTC detection methods are developed and optimized to assist in better patient monitoring and management (50)(51)(52)(53)(54). The advantages and disadvantages of the isolation methods in lung cancer are discussed and summarized in the Table 2. A summary of the CTC lung cancer studies are highlighted in Table 3.

EX-VIVO EXPANSION OF CIRCULATING TUMOR CELLS
Despite limitations of current CTC isolation techniques, these cells have been detected in a number of cancers, including breast, head, and neck cancer, lung, prostate, colon and gastric cancer (21,50,53,109,(132)(133)(134)(135). Successful ex-vivo culture of CTCs represents a "Holy Grail" in the study of cancer metastasis as it allows for in depth characterization of metastasis initiating cells as well as the testing of functional assays (136).
Short-term CTC culture (3-14 days) has been achieved in a number of cancer types, even from early stage cancers (137)(138)(139). This allows for the recapitulation of the disease in an ex vivo/in vivo setting for the testing of therapies and functional analysis (140). A summary of this is in Table 4. In comparison, long-term cultures have only been established in advanced metastatic cases where a large number of CTCs have been isolated (111,142,143) ( Table 5). Long-term culture studies have shown that some CTCs in patient blood are immortalized and can be cultured ex vivo into stable cell lines (Figure 1) (139). There are only a few reports of successful long-term culture, notably, in patients with advanced stages of disease (136,145,146). CTC-expansion has been limited due to the influence of CTC enrichment. Certain cancers also require specific culture conditions for primary and metastatic samples (136). The successful culture of CTCs long-term holds great promise in developing personalized cancer treatment for testing of therapeutic efficacy using drug screening (140). This approach could assist in determining the choice of therapeutic regimen beneficial for patients and hence holds significance in advancement of precision medicine and personalized oncology (139).
Three main strategies are used for the propagation of CTCs in culture; two-dimensional (2D) culture, very commonly used for expansion of CTCs short-term, three-dimensional (3D) culture used for long-term expansion and xenotransplantation and four dimensional (4D) shown to mimic the process of metastasis (137,(147)(148)(149)(150).
The expansion of CTCs in-vivo to generate patient derived xenografts (PDXs) may also be used to comprehensively analyse advanced disease biology and present a valuable model to understand cancer metastasis. The use of PDX's have been shown to mimic patient's disease and mirror response to chemotherapy (e.g., Platinum agents) (142,151). However, PDXs have been challenging due to CTC heterogeneity causing unreliability of these models to translate clinically. PDX model development also takes 4-8 months and therefore are not optimal for rapid studies necessary for patients with advanced disease (151). In an ideal world cancer cell lines would be routinely generated from each cancer patient but this is not realistic at present (136,139,152).

CLINICAL SIGNIFICANCE
The immediate need for early detection of lung cancer recurrence and monitoring treatment response is essential to facilitate Positive Immunoselection EpCAM positive based selection Ability to process larger volumes of blood for the capture of higher numbers of CTCs.
As above GILUPI CellCollector (GILUPI Nanomedizin) (55) Ephesia CTC-chip (56) Negative Immunoselection Depletion of Leukocytes by CD45 Antibodies Has the ability to avoid false-negative results or loss of CTCs due to phenotypic heterogeneity.
CTCs are often contaminated with remaining blood cells resulting in low purity. improved survival of patients. Previous studies have shown computerized tomography (CT) screening has helped to reduce mortality, however CT has risks such as radiation exposure, leading to an increased risk of long-term cancer (153). This signifies the need for less invasive techniques for the early detection of metastasis and aid the personalized treatment of lung cancer. The use of CTCs as a liquid biopsy has the potential to accompany standard screening tests and also allow for molecular and genetic characterization of the tumor (48). Enumeration of CTCs could provide a biomarker for cancer surveillance following treatment of early, locally advanced and advanced lung cancer and provided a better understanding on the mechanisms of metastasis (33). Although chemotherapy, targeted small molecules and immune checkpoint inhibitor

(Continued)
Frontiers in Oncology | www.frontiersin.org  therapies have shown significant benefits, the occurrence of acquired drug resistance and disease relapse are very common. Through serial sampling a longitudinal analysis of CTCs for identification of tumor evolution could provide valuable insights into mechanisms underlying resistance (154). Detection of CTCs in lung cancer has been challenging, as CTCs usually present with non-epithelial characteristics (49). This emphasizes the need for more sensitive technologies to better capture CTCs for in-depth characterization and functional studies using cell culture and xenograft models. This will then ultimately assist in optimizing personalized therapies for lung cancer patients, with CTCs potentially being a prognostic biomarker.

CONCLUSION
The clinical significance of CTCs is yet to be established, however, advances in CTC detection and single-cell profiling have significantly improved our knowledge of underlying mechanisms of the evolution and dissemination of cancer and is progressively being translated to clinical studies. With lung cancer being the largest cause of cancer mortality worldwide, one of the biggest challenges for managing and treating patients is the lack of early screening/diagnostic methods (4). The isolation of CTCs from cerebrospinal fluid (CSF), may represent a unique subpopulation CTCs with ability to survive the journey in blood circulation and subsequent invasion of the CNS (105,155). CTCs hold great promise as biomarkers for the early diagnosis and treatment  selection of patients as well as broadening the current knowledge of metastasis (154). Recurrence and progress of the disease, severity of symptoms and side-effects dramatically decrease patient's quality of life (QoL) (156). Therefore there is a vital need to monitor tumor evolution and understand mechanisms underlying development of therapeutic resistance.
Challenges for the field to address include the low sensitivity and specificity of current technologies prohibiting their use in current clinical settings, the large number of CTCs required for the development of CTC lines and patient xenografts for downstream functional analyses and the limited number of CTCs frequently found in patients with early stage disease (157). CTCs have demonstrated prognostic clinical utility is breast, lung and prostate cancers using the CellSearch technology (158,159). Recent studies have demonstrated renewed interest in the FDA-approved Cellsearch platform for CTC PD-L1 analysis (160)(161)(162). These studies demonstrate how CTCs could be used to identify patients for anti PD-1/PD-L1 therapy (immunotherapy). Cellsearch relies on CTC enrichment using EpCAM (when CTCs undergo EMT, EpCAM is downregulated). As such the field is moving toward labelfree technologies for CTC isolation. Currently, there are a number of technologies to enrich CTCs (i.e., Rarecyte, iChip, ISET, DEPArray, EPISPOT etc). The current label-free technologies are being validated for a number of cancers in larger clinical trials (163,164). This is highlighted by the Cancer-ID network consortium in standardizing CTC/ctDNA and exosome isolation, analysis and reporting (165). The current gold standard in isolating CTCs from patient blood relies on the EpCAM status of these cells, thereby excluding a large majority of CTCs present in the blood of metastatic patients. Furthermore, Cellsearch does not allow for subsequent culture as the cells are fixed (166). CTCs as a liquid biopsy have valuable potential to improve early diagnosis, monitoring of disease, and direct treatment of lung cancer, however a better understanding of CTC biology is crucial for the field to move forward.

AUTHOR CONTRIBUTIONS
JK, AK, KO, CP: Idea. JK, AK, MW: Preparation of figures and tables. All authors were involved in the preparation, review and editing of the manuscript.