The current prospective study was conducted at the Kitasato University Hospital (Sagamihara, Japan). The recruitment criteria were as follows: i) Histologically or cytologically confirmed SCLC and confirmation of the clinical stage based on the results of examination by chest X-ray, computed tomography (CT) of the chest and abdomen. In addition to other procedures as indicated, including brain magnetic resonance imaging (MRI) and positron emission tomography (PET) scanning or radionuclide bone scanning; ii) chemotherapy-naïve; iii) evaluable or measurable disease; iv) adequate bone marrow, hepatic and renal function; v) no active concomitant malignancy; and vi) written informed consent. Tumor response was classified in accordance with the Response Evaluation Criteria for Solid Tumors. CT, brain MRI and PET scanning or radionuclide bone scanning were routinely conducted to evaluate tumor progression. The institutional review board at the Kitasato University Hospital approved the study protocol and all patients provided written informed consent.

A peripheral blood specimen was collected to analyze the presence of CTCs at each of the following five time periods: Within seven days prior to commencing treatment (baseline); chemotherapy cycles two and three; following chemotherapy cycle four; and the time point at which progressive disease was confirmed.

The OBP-401 virus was used to detect CTCs as described previously (15). Briefly, a 7.5-ml peripheral-vein blood sample was drawn into tubes containing citric acid, phosphoric acid and dextrose (Nacalai Tesque, Inc., Kyoto, Japan). The himac CF12RX (Hitachi, Ltd., Tokyo, Japan) was used for centrifugation at 540 × g for 5 min. Following washing with phosphate-buffered saline (PBS) and RPMI-1640 medium (Sigma-Aldrich, St. Louis, MO, USA) by centrifugation, the samples were infected with a 4×10⁸ plaque-forming unit of OBP-401 virus for 24 h at 37°C. Each sample was fixed with 4% paraformaldehyde (Wako Pure Chemical Industries, Osaka, Japan) and treated with a surface-active agent (Emalgen 2025 G; Kao Chemicals, Tokyo, Japan) to degrade the red blood cells. The remaining cells were scraped from the slides, applied to glass slides and examined using a fluorescence microscope (IX71; Olympus Corporation, Tokyo, Japan).

The glass slides were soaked in PBS to remove the coverglass and the cells were scraped from the slides, suspended in PBS and applied to glass slides by cytocentrifugation in a Cytospin 4 cytocentrifuge (Thermo Fisher Scientific, Waltham, MA, USA). The cells were subsequently blocked with blocking buffer [1% normal goat serum (Millipore, Bedford, MA, USA) in PBS] for 1 h at room temperature and incubated for 1 h at room temperature with mouse monoclonal anti-pan cytokeratin antibody (ab961; Abcam, Cambridge, UK). Next, the cells were washed with PBS and treated with Alexa Fluor 405-conjugated secondary antibody (Invitrogen Life Technologies, Carlsbad, CA, USA) for 1 h at room temperature. Following washing with PBS, the cells were mounted and examined using a fluorescence microscope (IX71; Olympus, Tokyo, Japan) and the number of CTCs in the 7.5 ml peripheral venous blood was counted.

Previously, GFP-positive cells with relatively small diameters have been observed in blood samples obtained from cancer patients and healthy controls (16). The small GFP-positive cells were double-stained with a variety of anti-CD antibodies (CD2/3/13/14/15/16/19/45/203c; Biolegend, Inc., San Diego, CA, USA) to characterize the cell attributes and the results showed that the majority (~70%) of the small GFP-positive cells were monocytes (16). Accordingly, to define CTCs in the blood samples of cancer patients, the cut-off value of cell diameter for CTCs was determined as >8.4 μm. This value was deduced from the average value plus two standard deviations, by analyzing the diameter distribution of monocytes in the blood samples of healthy controls, which statistically indicated that >95% of monocytes may be excluded from the GFP-positive cells in the blood samples of cancer patients. Consequently, in the present study, GFP-positive cells >8.4 μm in diameter were counted as CTCs (17). A GFP-positive cell that was isolated from the peripheral blood of a SCLC patient is shown in Fig. 1.

The primary analysis was a comparison between overall survival (OS) in the unfavorable and favorable groups stratified according to the selected threshold of baseline CTC count. OS and progression-free survival (PFS) were measured from the date of when the baseline blood sample was collected to the date when clinical progression was confirmed by mortality or censoring at the last follow-up examination. PFS and OS curves were plotted using the Kaplan-Meier method and differences in survival time were analyzed for statistical significance with the log-rank test. Student’s t-test was performed to evaluate absolute change in the mean CTC count. Cox proportional hazards regression was used to determine the hazard ratios (HRs) for OS, which were adjusted for age, gender, pretreatment stage (extensive disease versus limited disease), Eastern Cooperative Oncology Group performance status, serum lactate dehydrogenase (LDH) levels, serum Na levels and the baseline CTC count prior to the initiation of chemotherapy. P<0.05 was considered to indicate a statistically significant difference and the statistical analysis was performed using the SPSS for Windows software program, version 17 (SPSS, Inc., Chicago, IL, USA).

The 30 patients with histologically or cytologically confirmed SCLC (28 males and two females; mean age, 69 years; age range, 51–85 years; limited disease patients, n=8; and extensive disease patients, n=22) who met the inclusion criteria between April 2009 and December 2011 were the subjects of the current study. The patient characteristics are summarized in Table I. At the time of analysis, 28 of the 30 (93%) evaluable patients had experienced disease progression, and 25 of the 30 (83%) evaluable patients had succumbed to their diseases, resulting in a median PFS of 5.7 months (95% CI, 4.8–6.5 months) and median survival time of 11.5 months (95% CI, 9.2–13.7 months). The median follow-up period for determining the survival time was 12.0 months from the baseline blood sample collection.

CTCs were detected in 29 of the 30 (96%) patients. The blood samples following two cycles of chemotherapy (prior to chemotherapy cycle three) for CTC analysis were obtained from 29 patients. The comparison between the mean CTC count at baseline and following two cycles of chemotherapy, according to treatment response, are shown in Table II. Among the patients who exhibited a partial response (PR) following two cycles of chemotherapy, the mean CTC count tended to increase (2.32 cells/7.5 ml) compared with the mean CTC count at baseline (0.84 cells/7.5 ml) regardless of a reduction in tumor volume (P=0.05).

The group of 21 patients with CTC counts of <2 cells/7.5 ml at the baseline exhibited a significantly longer median survival time (14.8 months; 95% CI, 11.5–18.2) than the group of nine patients with a CTC count of ≥2 cells/7.5 ml (3.9 months; 95% CI, 3.3–4.6) (P=0.007; Fig. 2). The group of patients that exhibited a PR following two cycles of chemotherapy who had a CTC count of <2 cells/7.5 ml prior to chemotherapy cycle three tended to have a longer median PFS (8.3 months; 95% CI, 5.3–11.3) compared with the group with a CTC count of ≥2 cells/7.5 ml (3.8 months; 95% CI, 2.6–5.0) (P=0.07; Fig. 3).

The clinical factors that were identified as significant in relation to survival time in the univariate analysis were stage at diagnosis, serum LDH levels and baseline CTC count (Table III). Multivariate analysis with adjustment for these factors identified all of the factors (HR, 3.91; 95% CI, 1.19–12.87; P=0.026) to be independent prognostic markers for OS (Table IV).