Tumor-associated macrophages and individual chemo-susceptibility are influenced by iron chelation in human slice cultures of gastric cancer.

Purpose: Presence of tumor-associated macrophages (TAM) and high levels of ferritin and lipocalin 2 (Lcn2) in the tumor microenvironment are associated with poor prognosis in many types of cancer. Here we investigate whether iron deprivation influences TAM phenotype and chemotherapy resistance in tumor slice cultures (TSC) of gastric cancer. Results: TAM remained morphologically and functionally stable for four DIV. DFO treatment for 72 h decreased ferritin expression in TAM and in the tumor stroma but did not alter Lcn2 expression. TAM phenotype was altered after 72 h of cisplatin or DFO treatment compared with control conditions. Single DFO treatment and combined treatment with cytotoxic drugs significantly increased tumor cell apoptosis in TSC of gastric cancer. Methods: TSC were manufactured by cutting tissue of gastric cancer resection specimens in 350 µm thick slices and cultivating them under standard conditions on a filter membrane, at an air-liquid interface. After 24 h ex vivo, TSC were treated with irinotecan (100 nM) or cisplatin (10 µM) alone and in combination with deferoxamine (DFO; 10 µM, 100 µM), respectively, for 72 h. After four days in vitro (DIV) the TSC were fixated with paraformaldehyde, paraffin embedded and analyzed by immunohistochemistry for apoptosis (cPARP), proliferation (Ki67), TAM (CD68, CD163), ferritin, and Lcn2 expression. Conclusions: TAM are well preserved and can be studied in TSC of gastric cancer. Iron deprivation significantly increased tumor cell apoptosis.

Individual Susceptibility Analysis Using Patient-derived Slice Cultures of Colorectal Carcinoma.

BACKGROUND: Nonresponse to chemotherapy in colorectal carcinoma (CRC) is still a clinical problem. For most established treatment regimens, no predictive biomarkers are available. Patient-derived tumor slice culture may be a promising ex vivo technology to assess the drug susceptibility in individual tumors. METHODS: Patient-derived slice cultures of CRC specimens were prepared according to a standardized protocol and treated with different concentrations of 5-fluorouracil (5-FU) and an adapted FOLFOX regimen (5-FU and oxaliplatin) to investigate histologic response. Additionally, a semi-automatized readout using fluorescent stain-specific segmentation algorithms for Image J was established to quantify changes in tumor proliferation. Nonresponse to chemotherapy was defined as persisting tumor cell proliferation. RESULTS: Slices treated with 5-FU showed lower tumor cell fractions and dose-dependent alterations of proliferating tumor cells compared with controls (1 µM, Δ +3%; 10 µM, Δ -9%; 100 µM, Δ -15%). Individual tumor samples were examined and differences in chemotherapy susceptibility could be observed. Untreated slice cultures contained an average tumor cell fraction of 31% ± 7%. For all samples, the histopathologic characteristics exhibited some degree of intratumoral heterogeneity with regard to tumor cell morphology and distribution. The original tumor matched the features found in slices at baseline and after 3 days of cultivation. CONCLUSIONS: Patient-derived slice cultures may help to predict response to clinical treatment in individual patients with CRC. Future studies need to address the problem of tumor heterogeneity and evolution. Prospective correlation of ex vivo results with the clinical course of treated patients is warranted.