A novel human ex vivo model for the analysis of molecular events during lung cancer chemotherapy

Abstract

Background: Non-small cell lung cancer (NSCLC) causes most of cancer related deaths in humans and is characterized by poor prognosis regarding efficiency of chemotherapeutical treatment and long-term survival of the patients. The purpose of the present study was the development of a human ex vivo tissue culture model and the analysis of the effects of conventional chemotherapy, which then can serve as a tool to test new chemotherapeutical regimens in NSCLC.

Methods: In a short-term tissue culture model designated STST (Short-Term Stimulation of Tissues) in combination with the novel *HOPE-fixation and paraffin embedding method we examined the responsiveness of 41 human NSCLC tissue specimens to the individual cytotoxic drugs carboplatin, vinorelbine or gemcitabine. Viability was analyzed by LIFE/DEAD assay, TUNEL-staining and colorimetric MTT assay. Expression of Ki-67 protein and of BrdU (bromodeoxyuridine) uptake as markers for proliferation and of cleaved (activated) effector caspase-3 as indicator of late phase apoptosis were assessed by immunohistochemistry. Transcription of caspase-3 was analyzed by RT-PCR. Flow cytometry was utilized to determine caspase-3 in human cancer cell lines.

Results: Viability, proliferation and apoptosis of the tissues were moderately affected by cultivation. In human breast cancer, small-cell lung cancer (SCLC) and human cell lines (CPC-N, HEK) proliferative capacity was clearly reduced by all 3 chemotherapeutic agents in a very similar manner. Cleavage of caspase-3 was induced in the chemo-sensitive types of cancer (breast cancer, SCLC). Drug-induced effects in human NSCLC tissues were less evident than in the chemo-sensitive tumors with more pronounced effects in adenocarcinomas as compared to squamous cell carcinomas.

Conclusion: Although there was high heterogeneity among the individual tumor tissue responses as expected, we clearly demonstrate specific multiple drug-induced effects simultaneously. Thus, STST provides a useful human model to study numerous aspects of mechanisms underlying tumor responsiveness towards improved anticancer treatment. The results presented here shall serve as a base for multiple functional tests of novel chemotherapeutic approaches to NSCLC in the future.

Figure 1

Percentages of dead cells in human NSCLC tissue specimens following 16 h culture period ex vivo. Viability was also assessed without cultivation (native tumor; NAT TU) to compare for culture effects. Data are expressed as mean percentage of dead cells + STD (standard deviation) (n = 6) with Triton X values as 100%.

Figure 2

Caspase-3 induction determined by flow cytometry in human cancer cell lines in the absence (RPMI) or presence of carboplatin (CARB), vinorelbine (VIN) and gemcitabine (GEM), respectively. The cell lines were plated 24 h before addition of the single drug. Percentage of caspase-3 positive cells are shown as mean values + SEM (n = 4–5).

Figure 3

A-D: Induction of cleaved caspase-3 as determined by IHC in CPC-N cells (small-cell lung cancer) in the absence (A) or presence of carboplatin (B), vinorelbine (C) and gemcitabine (D), respectively (all 400×).

Figure 4

Transcription of caspase-3 in four different human NSCLC tissue specimens (n = 3 SCC; n = 1 AC) and one human SCLC (small-cell lung cancer) tissue sample in the absence (RPMI) or presence of carboplatin (CARB), vinorelbine (VIN) and gemcitabine (GEM), respectively. Signals of caspase-3 were normalized to the respective GAPDH signals and the resulting values of the untreated medium control tissues (RPMI) were set 100%. The data are shown for each individual specimen separately and are expressed as mean values ± SEM (n = 2). Lines were drawn through all data points derived from a single specimen to better reveal the individual trend of response.