Three-dimensional nanostructured substrates enable dynamic detection of ALK-rearrangement in circulating tumor cells from treatment-naive patients with stage III/IV lung adenocarcinoma

Abstract

Background: Circulating tumor cells (CTC) shows great prospect to realize precision medicine in cancer patients.

Methods: We developed the NanoVelcro Chip integrating three functional mechanisms. NanoVelcro CTC capture efficiency was tested in stage III or IV lung adenocarcinoma. Further, ALK-rearrangement status was examined through fluorescent in situ hybridization in CTCs enriched by NanoVelcro.

Results: NanoVelcro system showed higher CTC-capture efficiency than CellSearch in stage III or IV lung adenocarcinoma. CTC counts obtained by both methods were positively correlated (r = 0.45, p < 0.05). Further, Correlation between CTC counts and pTNM stage determined by NanoVelcro was more significant than that determined by CellSearch (p < 0.001 VS p = 0.029). All ALK-positive patients had 3 or more ALK-rearranged CTC per ml of blood. Less than 3 ALK-rearranged CTC was detected in ALK-negative patients. NanoVelcro can detect the ALK-rearranged status with consistent sensitivity and specificity compared to biopsy test. Furthermore, the ALK-rearranged CTC ratio correlated to the pTNM stage in ALK-positive patients. Following up showed that CTCs counting by NanoVelcro was more stable and reliable in evaluating the efficacy of Clozotinib both in the short and long run compared with CellSearch. Changing of NanoVlecro CTC counts could accurately reflect disease progression.

Conclusion: NanoVelcro provides a sensitive method for CTC counts and characterization in advanced NSCLC. ALK-rearrangement can be detected in CTCs collected from advanced NSCLC patients by NanoVelcro, facilitating diagnostic test and prognosis analysis, most importantly offering one noninvasive method for real-time monitoring of treatment reaction.

Keywords: Circulating tumor cell; Diagnostics; Lung cancer; Microfluidics; Nanomaterials.

Fig. 1

a 1.0 mL of blood is collected from NSCLC patients. b CTCs are captured in Nano Velcro chip. c Immunostaining assay is carried out to identify CTCs. d FISH assay is used to detect the ALK status in CTCs

Fig. 2

NanoVelcro device configuration and work mechanism. a NanoVelcro CTC Chip is composed of an overlaid PDMS-based chaotic mixer, a patterned silicon nanowire (SiNW) substrate, and a multilayer chip holder to assemble both functional components together. b Herringbone features introduce a helical flow in the microchannel facilitating the immunological recognition between CTC and anti-EpCAM coated SiNW substrate. c Schematic representation of an immunofluorescence protocol developed for identification of CTC (CK+/CD45−/DAPI+) from non-specifically captured WBCs (CK−/CD45+/DAPI+) and cell debris. d Silanation reaction were employed to covalently link streptavidin onto the SiNW substrate, allowing conjugation of biotinylated anti-EpCAM prior to CTC detection studies

Fig. 3

Comparison of NanoVelcro and CellSearch in CTC capture ability. a Immunofluorescence staining of CTC immobilized on NanoVelcro substrates. Typical micrographs of CTC and WBCs were immobilized on a NanoVelcro substrate. Scale: white bars correspond to 10 μm. b The enumerated CTC by NanoVelcro from 1.0 ml blood and by CellSearch from 7.5 ml in NSCLC adenocarcinoma patients at their first visit (Normalized to 1.0 ml scale for comparison). Red represents the enumerated CK+/CD45− CTC by NanoVelcro in patient blood; Green represents CellSearch counts CTC. c Linear regression plots were computed for CTC counts obtained by CellSearch and NanoVelcro (Normalized to 7.5 ml scale for comparison). d Correlation analysis between pTNM stage with NanoVelcro and CellSearch CTC counts respectively in all NSCLC adenocarcinoma patients (Normalized to 1.0 ml scale for comparison)

Fig. 4

Detection of anaplastic lymphoma kinase (ALK) gene abnormalities in circulating tumor cells (CTC) and tumor specimens of ALK-positive patients. a Examples of isolated or clusters of ALK-rearranged CTCs detected by filter-adapted fluorescent in situ hybridization (FISH) and of ALK-rearranged tumor cells in tumor specimens detected by FISH. Green arrows show an ALK rearrangement with a split 3′ and 5′ (red/green) signal. Red arrows show an ALK rearrangement with only the 3′ signal. b Examples of isolated CTCs with a gain of native ALK copies. In cells with a native ALK status, the overlapping of probes results in a fused (3′ 5′, yellow) signal. Scale: white bars correspond to 10 μm

Fig. 5

Determination of the anaplastic lymphoma kinase (ALK)–rearranged circulating tumor cells (CTC) optimal cutoff value in ALK-positive and ALK negative patients. a Prevalence of ALK-rearranged CTC in ALK-positive and ALK-negative patients. b Determination of the cutoff value for 3 ALK-rearranged CTC per ml blood by using receiver operating characteristic curve analysis. NPV negative predictive value, PPV positive predictive value

Fig. 6

Receiver operating characteristic (ROC) curve in differentiating ALK-positive and ALK-negative patients, the corresponding AUC was approximately 0.989

Fig. 7

Clinical significance of circulating tumor cells (CTC) harboring anaplastic lymphoma kinase (ALK) in ALK-positive patients. a CTC counts by NanoVelcro positively correlated to CellSearch technique in ALK-positive patients. b In ALK-positive patients, CTC counts by NanoVelcro related to pTNM stage, but CellSearch

Fig. 8

Dynamic change in circulating tumor cell (CTC) counts before and after initiation of crizotinib treatment for 8 months in one ALK-positive patient. a Tumor tissues from this patient were positive for TTF-1, CEA and CK-7. b Serial lung CT scan changes are presented. Lung tumor showed gradual shrinkage after crizotinib treatment. c Serial CTC counts by NanoVelcro and CellSearch are plotted along with radiological changes. CTC counts by NanoVelcro are always higher than CellSearch before lung tumor disappearing on CT scan after crizotinib treatment for 6 months. NanoVelcro accurately reflects the disease progressions with the disappearing of tumor, while CellSearch rebounded abnormally