Dexamethasone desensitizes hepatocellular and colorectal tumours toward cytotoxic therapy

Cancer Letters 242 (2006) 104–111 www.elsevier.com/locate/canlet

Dexamethasone desensitizes hepatocellular and colorectal tumours toward cytotoxic therapy C. Zhang a,1, A. Kolb b,1, J. Mattern c, N. Gassler d, T. Wenger a, K. Herzer e,f, K-M. Debatin g, M. Bu¨chler b, H. Friess b, W. Rittgen h, L. Edler h, I. Herr a,g,* a

Molecular Urooncology, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany b Department of General Surgery, University of Heidelberg, Heidelberg, Germany c Clinical Cooperation Unit Nuclear Medicine, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany d Department of Pathology, University of Heidelberg, Heidelberg, Germany e Tumour Immunology Programme, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany f Department of Internal Medicine, University of Mainz, Mainz, Germany g Department of Pediatrics, University of Ulm, Ulm, Germany h Department of Biostatistics, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany Received 8 August 2005; received in revised form 25 October 2005; accepted 26 October 2005

Abstract The glucocorticoid dexamethasone is frequently used as co-treatment in cytotoxic cancer therapy, e.g. to prevent nausea, to protect normal tissue or for other reasons. While the potent pro-apoptotic properties and the supportive effects of glucocorticoids to tumour therapy in lymphoid cells are well studied, the impact to cytotoxic treatment of colorectal and hepatocellular carcinoma is unknown. We tested apoptosis-induction, viability, tumour growth and protein expression using 8 established cell lines, 18 surgical specimen and a xenograft on nude mice. In the presence of dexamethasone we found strong inhibition of apoptosis in response to 5-FU, cisplatin, gemcitabine or g-irradiation, enhanced viability and tumour growth of colorectal and hepatocellular carcinomas. No correlation with age, gender, histology, TNM, the p53 status and induction of therapy resistance by dexamethasone cotreatment could be detected. These data show that glucocorticoid-induced resistance occurs not occasionally but is common in colorectal and hepatocellular carcinomas implicating that the use of glucocorticoids may be harmful for cancer patients. q 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Cancer therapy; Glucocorticoids; Corticosteroids; Nausea; Apoptosis

1. Introduction Colorectal and hepatocellular carcinomas are a common cause of cancer mortality worldwide. Sixty Abbreviations DEX, dexamethasone; GCs, glucocorticoids; GR, glucocorticoid receptor; 5-FU, 5-fluorouracil; HCC, hepatocellular carcinoma. * Corresponding author. *Tel.: C49 6221 42 3366; fax: C49 6221 42 3362. E-mail address: [email protected] (I. Herr). 1 Both these authors have contributed equally.

percent of patients with colorectal carcinoma develop liver metastases during the treatment of their disease which are a major cause of morbidity and mortality. Traditional systemic chemotherapy regimens, i.e. fluorouracil (5-FU) have yielded response rates of approximately 20% [1]. More recently developed agents, such as gemcitabine or platinum analogues have been found to be useful either alone or in combination with 5-FU. Despite these newer agents and combinations, response rates remain low [2] and alternative treatment strategies have been explored.

0304-3835/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2005.10.037

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

One such strategy is hepatic arterial infusion of chemotherapy to deliver high concentrations of cytotoxic agents directly to liver metastases. For hepatocellular carcinoma (HCC) local treatments may be useful in selected patients, but not for many with advanced disease. Treatments trialled for advanced disease include radiotherapy, and systemic as well as intra-arterial chemotherapy. The use of cytotoxic agents in HCC has been disappointing, with few agents showing response rates above 20% including singleagents and combinations of 5-FU, doxorubicine, etoposide, cisplatin, taxanes and gemcitabine [3]. Most chemotherapeutic treatments including hepatic arterial infusion of chemotherapy are mitigated by the addition of dexamethasone (DEX) and similar glucocorticoids (GCs) [4]. GCs are commonly used as comedication in cancer therapy [5]. In the early 1960s, GCs were introduced in remission induction of childhood leukaemia [6]. Subsequently, GCs were used in combination with cytotoxic drugs to treat all haematological tumours, and to manage treatment- and diseaseassociated symptoms in solid cancer. GCs were used on the basis of their recent well-understood pro-death effects in lymphoid cells, and on their effectiveness in treating tumour or treatment-related oedema, inflammation, pain, electrolyte imbalance, to stimulate appetite, to prevent nausea and emesis, or toxic reactions caused by cytotoxic treatment in cancer therapy [5,7]. While GCs generally support therapy of lymphoid tumour cells, some studies describe inhibition of cancer therapy in some cell lines of solid tumours [5,8–11]. Concomitantly, a growing body of evidence from single cell lines strongly suggest a GCconferred cellular resistance to cancer therapy, e.g. by enhancing DNA repair capacity, blocking apoptosis, suppressing host anti-tumour immune responses and causing biophysical resistance to drug access by triggering diabetes [5,7,12,13]. A recent report demonstrates an increased risk of skin cancer and lymphomas in patients treated with GCs [14]. Remarkably, prospective randomized clinical trials evaluating any potential impact of GCs on tumour control have never been performed. Most importantly, it is not even known, whether GC-conferred resistance in solid tumours is a common or an only occasionally problem. To analyze whether DEX might affect the outcome of cytotoxic therapy of colorectal and hepatocellular carcinomas, 8 cell lines, 18 primary cell lines freshly isolated from resected tumour specimens and one xenograft line growing on nude mice were examined. Cells were treated with 5-FU, cisplatin, gemcitabine or g-irradiation in the presence or absence of DEX. DEX

105

inhibited apoptosis, promoted viability and tumour growth of all examined cancer cells independent of the p53 status. These results suggest that induction of therapy resistance by co-medication of GCs occurs not only occasional but is a common problem in colorectal and hepatocellular carcinoma. 2. Material and methods 2.1. Cell lines and culture The following established human tumour cell lines were used: Caco-2, CX-1, SW707, SW948 (colon), SW707 (rectal), HepG2, Huh7, Hep3B (hepatocellular). HepG2 cells express small amounts of wild type p53, Huh7 cells express mutant p53 with increased halflife as a result of a point mutation at codon 220 and Hep3B cells are deficient of p53 [15]. Cells were grown at 37 8C in DMEM. DMEM was obtained from Life Technologies Gibco BRL (Karlsruhe, Germany) supplemented with 10% heat-inactivated fetal bovine serum (Sigma, Deisenhofen, Germany), 25 mM HEPES and 2 mM L -glutamine (all from Gibco/Life Technologies, Paisley, Scotland). The cell lines were cultured as recently described [9]. 2.2. Tumours Solid tumours were resected, transported in Liforlabw medium (Oncoscience, Wedel, Germany) to the laboratory where the fresh tissues were minced in RPMI medium supplemented with 20% heat-inactivated fetal bovine serum (Sigma, Deisenhofen, Germany), 25 mM HEPES, 2 mM L-glutamine and Pen/Strep (all from Gibco/Life Technologies, Paisley, Scotland) under sterile conditions, counted by trypan blue exclusion and immediately analyzed by the MTTassay. Patient material was obtained under the approval of the ethic committee of the University of Heidelberg. Diagnoses were established by conventional clinical and histological criteria according to the World Health Organization (WHO). All surgical resections were indicated by principles and practice of oncological therapy. Neither GCs nor neoadjuvant chemotherapy were applied prior to surgery. 2.3. Nude mice and xenografts CX-1 colon carcinoma cells were injected subcutaneously into the right anterior flank of 6–10 weeks old NMRI (nu/nu) female mice. After the tumours had reached a mean diameter of about 8–10 mm, mice were randomized divided into groups of 5–6 animals each and treatment was started. The mice were given 0.28 mg/l DEX in the drinking water, and the amount water consumed daily was approximately 30 ng/g body weight. Cisplatin (5 mg/kg) was injected i.o. at two consecutive days followed by daily measuring tumour growth with calipers and tumour volumes (v) were calculated using

106

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

the formula vZ1/2 (length!width2). Mice were humanely euthanized at tumour sizes O3000 mm3. 2.4. Cytotoxic treatment

receiving cytotoxic drugs. On this basis, scores were calculated per therapeutic dose (values 0, 1, 2, 3) as well as per patient in total (values ranging from 0 to 9). A sample was declared as being significantly resistant to cytotoxic treatment

Stock solutions of cytotoxic drugs (all obtained from Sigma, Deisenhofen, Germany) were prepared in DMSO (cisplatin, 5-fluorouracil) or cell culture medium (gemcitabine). A stock solutions of DEX (Sigma) was prepared in ethanol. Final concentrations of the solvents in medium were 0.01% or less. Cells were g-irradiated in their flasks using a cesium g-ray source. 2.5. Measurement of apoptosis Cells were stained with fluoresceinthiocyanate (FITC)conjugated annexin V (BD Biosciences, Heidelberg, Germany) and externalization of phosphatidylserine as well as the forward side scatter profile were identified by flow cytometry (FACScan, BD Biosciences) as described [9]. For detection of DNA fragmentation the Nicoletti method was used [16]. 2.6. MTT-assay Primary tumour cells were resuspended at 5!105/ml in 96-well microplates, 100 ml per well. After treatment the MTT-assay was performed as described [17]. 2.7. Statistical analysis 2.7.1. Biopsies The tumor specimen of each patient was investigated in a 2-factorial design consisting of three doses of DEX (0.1Z DEX1, 1ZDEX2, 10 mMZ DEX3) and a control and three doses of gemcitabine (25, 1; 50, 2; 200, 3) or cisplatin (7, 1; 17, 2; 34, 3) and a control (Z0). This results in a total of 16 experimental conditions. Viability of the cells under each condition was determined as mean of 3–8 replicates together with its standard deviation and then standardized on the result of the double control (no cytotoxic agent and no DEX applied); i.e. the viabilities of all conditions were divided by the mean of the double control. For each patient the standardized means were compared separately for each therapeutic dose and for the control by comparing the three DEX doses with its respective control (cytotoxic treatment alone). Notice, that the four means (three DEX doses and control) under the condition of no therapy describe the effect of DEX alone while the other three sets of four means under the three doses of the therapeutic agent describe the resistance of the cells under treatment depending on DEX. We declared a DEX dose group resistant when its mean minus one standard deviation (X j KSDj ) was still larger than the mean plus one standard deviation (X 0 C SD0 ) of the respective control group of that specimen, jZ1, 2, 3 denoting the three dose groups

Fig. 1. DEX inhibits apoptosis and promotes proliferation in colorectal and hepatocellular cancer cells independent of p53 in vitro. (A) CACO-2 and CX-1 cells were left either untreated (CO) or were treated with 5-FU in concentrations indicated in the absence (white bars) or presence (black bars) of DEX (1 mM) which was added 48 h prior to cytotoxic treatment. 72 h following addition of 5-FU apoptosis was analyzed by staining of the cells with annexin-FITC and FACS-analysis. Likewise, viability was detected by the MTT-assay. (B) The human HCC cell lines HepG2 with wild type p53 (p53wt), Huh7 with mutated p53 (p53mut) and HEP3B with deleted p53 (p53K/K) were treated with DEX (1 mM) followed by cisplatin (7 or 13 mM CIS) 48 h later. 72 h post-incubation of cisplatin, apoptosis was analyzed by staining of the cells with annexin–FITC, nicoletti buffer, and measuring the FSC/SSC profile by FACS-analysis. Viability was examined by the MTT-assay. Experiments were performed three times with identical outcome and standard deviations are shown.

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

when it showed a score of 2 for at least one drug dose or when he/she reached the total score of 5 out of a maximum of 9 per time point (i.e. more than 50%). The outcomes of a group of biopsies were summarized in respective ratios obtained by adding the individual scores for each condition as well as in total over the number of samples and by dividing through the number of samples. The results of the evaluation of each single specimen are available upon request. The tumour population was considered as suffering from resistance as a whole, when for one therapeutic dose all ratios were higher than 50% or when 5 of the total of 9 combinations were higher than 50%. The statistical method described was applied independently to all three time points. 2.7.2. Xenografts on nude mice A distribution free test for tumour growth curve analyses was used for therapy experiments with xenografted cancer cells as described [18].

3. Results 3.1. DEX prevents cisplatin-induced apoptosis and promotes proliferation independent of the p53 status in vitro To investigate whether DEX might interfere with apoptosis we treated five established colorectal carcinoma cell lines in the presence or absence of DEX. 72 h later, apoptosis was examined by staining of the cells with annexin-FITC and FACS-analysis (Fig. 1A). While 5-FU alone strongly induced

107

apoptosis, the presence of DEX inhibited death in all cell lines examined. Data are presented for CACO-2 and CX-1 cells. Similar results have been found in SW707, SW403 and SW948 cells (data not shown). Corresponding results were obtained when viability was detected by the MTT-assay. Remarkably, even DEX alone induced faster proliferation and reduced basal apoptosis. To get information about DEX might induce resistance toward other cytotoxic drugs, cells were treated with cisplatin in the presence or absence of DEX. Also under these conditions, apoptosis was inhibited and viability was promoted by DEX (data not shown). To examine apoptosis induction in hepatocellular (HCC) carcinomas together with a putative involvement of the p53 status, the HCC cell lines HepG2 with wild type p53, Huh7 with mutant p53 and HEP3B with deleted p53 were treated with cisplatin in the presence or absence of DEX and 72 h later death was measured by detection of the FSC/SSC profile, apoptosis by staining with annexin-FITC, DNAfragmentation by staining with nicoletti buffer followed by FACS-analysis. Viability was examined by the MTT-assay under the same conditions (Fig. 1B). DEX inhibited all features of death and promoted viability irrespective of the p53 status. However, and as expected, cells with defective p53 are less sensitive to cytotoxic treatment compared to cells with functional p53 as described [15].

Table 1 Characterization of tissue from patients and cytotoxic treatment of isolated cells Tumour

Gender

Age (years)

Histological typing (WHO)

pTNM

RC3 RC4 RC5 RC6 RC7 RC8 RC9 COLO3 COLO4 COLO6 COLO7 COLO9 COLO5 COLO8 HCC1 HCC2 HCC3 HCC4

F M M M F M F F M F M M M M M M M M

54 67 57 67 61 59 75 79 71 55 65 69 65 80 68 68 77 73

Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma Hepatic metastase adenocarcinoma Hepatic metastase adenocarcinoma Hepatocellular carcinoma Hepatocellular carcinoma Hepatocellular carcinoma Hepatocellular carcinoma

pT3, pN1, pT1, pN0, pT3, pN0, pT2, pN0, pT3, pN2, pT3, pN0,

Treatment C/K DEX M0, G3 M0, G2 M0, G3 M0, G2 M0, G3 M0, G3

pT3, pN1 (3/34), M0, G2 pT2, pN0 (0/11), M0, G3 pT4, pN0 (0/66), pM1, G3 pT3, pN0 (0/13), pM1, G2 pT3, pN1, (3/16), pM1, G2

pT1, pN0 (nZ1), pMx, G2 pT2, pN0 (0/1), pMx, G1 pT1, pN0 (0/1), pMx, G3 pT2, pNx, pMx, G2, R0

M, male, F, female, RC, rectal carcinoma, COLO, colon carcinoma, HCC, hepatocellular carcinoma.

CIS CIS GEM GEM GEM 5-FU GAMMA 5-FU GAMMA CIS CIS GEM 5-FU GAMMA 5-FU GAMMA CIS 5-FU GAMMA GAMMA CIS GAMMA CIS 5-FU CIS GAMMA CIS

108

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

lines which may have undergone a selection process, we examined freshly isolated primary colorectal carcinoma cells. The tissue used contained 90% tumour cells morphologically identified by a pathologist. DEX was used in concentrations of 0.1, 1 and 10 mM from which the median concentration resembles peak plasma levels in the clinical setting [19–21]. Cells derived from 7 rectal and 7 colon carcinomas including 2 liver metastases of colon carcinomas and 4 hepatocelluar carcinomas (for tumour characteristics see Table 1) were treated with 5-FU, cisplatin, gemcitabine or g-irradiation in the presence or absence of DEX. 24, 48 and 72 h later, viability was measured (Figs. 2 and 3 and data not shown). DEX strongly enhanced viability in all examined cells and totally neutralized in several cases the cytotoxic effect of cancer therapy. Using statistical analysis induction of resistance by DEX cotreatment of patient material was confirmed to be significant—the method is described in Section 2. Thus, DEX inhibits apoptosis and promotes proliferation of carcinoma cells treated ex vivo.

Fig. 2. DEX induces resistance in resected colorectal carcinomas and derived metastases ex vivo. Tumour cells from patients with rectal (RC8, RC9, RC7, RC3) or colon cancer (COLO6, COLO7) and derived liver metastases (COLO8, COLO5) were freshly isolated from surgical specimens and cultivated in a concentration of 5! 105/ml in the absence (white bars) or presence of DEX (0.1, 1 or 10 mM as indicated) for 24 h. Cells were treated with 5-FU, girradiation, gemcitabine or cisplatin while the controls remained untreated (CO). After incubation for 48 or 72 h with cytotoxic drugs, viability was measured by the MTT-assay. Experiments were performed three times with identical outcome and standard deviations are shown. Statistical analysis of resistance was performed as described in Section 2.

3.2. DEX induces therapy resistance in surgical specimens of colorectal and hepatocellular carcinomas ex vivo To exclude that induction of therapy resistance by DEX is an artefact occurring only with established cell

Fig. 3. DEX induces resistance in cells of resected hepatocellular carcinomas ex vivo. Tumour cells from patients with hepatocellular carcinoma were freshly isolated, treated with cisplatin in the presence or absence of DEX and analyzed as described in Fig. 2.

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

Fig. 4. DEX inhibits therapy-induced regression of colon carcinoma xenografts in vivo. CX-1 colon carcinoma cells were established as xenograft cell line and injected subcutaneously into nude mice treated as described in Section 2. The volumes of the fast-growing tumours were measured at day 3 and 4 and the results are shown in the diagram. Mice were humanely euthanized after the measurement at day 4. Data are presented as mean of 5–6 animals and the P-value of 0.004 for CIS versus CIS/DEX together with the P-value of 0.0152 for CO versus Cis versus DEX versus CIS/DEX was determined according to the Koziol test for tumour growth curve analyses [18].

3.3. DEX induces therapy resistance in xenografted colon carcinomas in vivo To analyze the in vivo effect of DEX we xenografted CX-1 colon carcinoma cells to nude mice. Tumour volumes were measured daily after start of cisplatin therapy and found to be reduced in mice treated with cisplatin alone (Fig. 4). DEX totally prevented the growth-inhibiting effect of cisplatin since the tumours grew as fast as tumours of untreated control mice. Also, DEX alone led to a faster basal growth of the xenografts. 4. Discussion Our data demonstrate for the first time that in vitro, ex vivo and in vivo treatment of colorectal and hepatocellular cancer cells with GCs generally induces resistance toward cytotoxic therapy. We report here, that DEX inhibits 5-FU, gemcitabine-, cisplatin- and g-irradiation-induced apoptosis and promotes viability in all of eight examined established colorectal and hepatocellular cancer cell lines, in all of 18 primary cell lines freshly isolated from resected colorectal and hepatocellular surgical specimens and in xenografted colon carcinoma tumour cells. These results are surprising and suggest a cell type specific effect of

109

GCs, since these agents are well known to act proapoptotic and anti-proliferative in lymphoid cells as demonstrated by results of our and other groups [5,7,9]. Data obtained with colorectal and hepatocellular carcinomas within the present study may be transferred to other solid tumours, since we found in our actual and ongoing screening project induction of resistance by GCs in 95% of 161 examined fresh surgical specimens, xenografts on mice and established cell lines of tumours including bladder, bone, brain, breast, cervix, lung, kidney, ovary, pancreas, prostate and testis, together with neuroblastomas, and melanomas (these data are summarized in Ref. [22]). Our findings strongly suggest that GCs are highly suspicious to induce resistance to cytotoxic therapy in clinical settings, specifically, in patients with colorectal and hepatocellular cancer. This point was recently addressed in a retrospective clinical study evaluating records of 245 of a total of 763 patients with ovarian carcinoma and no negative outcome of GC treatment on survival of patients was found [23]. However, this study cannot give a definitive answer to the question of whether GC treatment, given as part of antiemetic regimen, prevention of allergic reactions or as immunosuppressive therapy is safe in patients with ovarian carcinoma or other solid tumours—the reasons are discussed elsewhere [24]. Furthermore, there are other clinical examinations which clearly show a negative impact of GCs, e.g. an increased metastatic potential in breast cancer patients and an enhanced risk of skin cancer and lymphomas among users of systemic GCs [14,25,26]. The mechanisms by which GCs induce apoptosis in lymphoid cells are well studied. These include depolarization of the mitochondrial membrane potential, enhanced expression of the death receptor CD95 and its ligand, followed by activation of the caspase cascade [27–29]. The same mechanisms that are induced in lymphoid cells are blocked in several carcinoma cells by GCs thereby inhibiting chemoand radiation therapy-induced apoptosis [9,11]. An open question is, how GCs mediate these cell-type specific effects clearly shown to be related to a functional glucocorticoid receptor (GR) since the effect can be inhibited by the GR antagonist RU486 [9,11]. Nothing is known about a direct link between GR signaling and induction of apoptosis. One putative link is the anti-apoptotic co-chaperone BAG-1 which together with the chaperone HSP-70 is involved in regulation of GR binding activity [30]. However, DEX did not up-regulate expression of BAG-1 neither basal nor in the presence of cisplatin as would have been

110

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

expected in the case of an anti-apoptotic influence of BAG-1 on GR signaling. Another possible explanation for the cell-type specific anti-apoptotic properties of GCs may be the differential expression of GR co-activators and corepressors in diverse cell types, as proposed to explain the opposite effects of tamoxifen on mammary versus endometrial tissue [31]. A recent study compared gene expression of a breast cancer cell line with genes found to be regulated by DEX in lymphocytes [11]. Surprisingly, only a few of the genes regulated by DEX in carcinomas are the same as those identified as GC-regulated in lymphocytes. Among the differential regulated set of sequences are apoptotic genes as well as genes involved in signal transduction, metabolism, transcription, cell cycle, DNA repair and others. These recent data strongly suggest that tissue-specific differences in GC-induced apoptosis versus survival outcomes may be due to cell-type-specific transcriptional regulation. In conclusion, we show that application of DEX renders colorectal and hepatocellular cancer cells generally resistant to cytotoxic therapy. This is in contrast to the effect of GCs in lymphoid cells and may involve cell type specific regulation of survival molecules together with anti-apoptotic molecules. The mechanistic background for this differential effect of GCs is not fully clarified until now. GC-induced modification of cell contacts unique for non-lymphoid and especially epithelial cells may interfere with survival signaling by Akt, MKP-1, SGK-1 and Wnt which seem to be crucially involved in GC-induced therapy resistance. These data are reviewed in detail elsewhere [22]. Thus, while some properties of GCs may be of benefit, induction of resistance in tumour cells may be dangerous for patients. Re-evaluation of patient files and controlled randomized prospective clinical studies are urgently needed. Acknowledgements We thank M. Mildenberger for excellent technical assistance and the Tumorzentrum Heidelberg for financial support. References [1]

Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: evidence in terms of response rate. Advanced Colorectal Cancer Meta-Analysis Project, J. Clin. Oncol. 10 (1992) 896–903.

[2] E. Diaz-Rubio, New chemotherapeutic advances in pancreatic, colorectal, and gastric cancers, Oncologist 9 (2004) 282–294. [3] A.K. Nowak, P.K. Chow, M. Findlay, Systemic therapy for advanced hepatocellular carcinoma: a review, Eur. J. Cancer 40 (2004) 1474–1484. [4] N. Kemeny, W. Jarnagin, M. Gonen, J. Stockman, L. Blumgart, D. Sperber, et al., Phase I/II study of hepatic arterial therapy with floxuridine and dexamethasone in combination with intravenous irinotecan as adjuvant treatment after resection of hepatic metastases from colorectal cancer, J. Clin. Oncol. 21 (2003) 3303–3309. [5] H.P. Rutz, I. Herr, Interference of glucocorticoids with apoptosis signaling and host–tumor interactions, Cancer Biol. Ther. 3 (2004) 715–718. [6] E. Frei III, M. Karon, R.H. Levin, E.J. Freireich, R.J. Taylor, J. Hananian, et al., The effectiveness of combinations of antileukemic agents in inducing and maintaining remission in children with acute leukemia, Blood 26 (1965) 642–656. [7] H.P. Rutz, Effects of corticosteroid use on treatment of solid tumours, Lancet 360 (2002) 1969–1970. [8] M. Mariotta, G. Perewusnyk, O.R. Koechli, J.B. Little, M. von Knebel, M. Doeberitz, et al., Dexamethasone-induced enhancement of resistance to ionizing radiation and chemotherapeutic agents in human tumor cells, Strahlenther. Onkol. 175 (1999) 392–396. [9] I. Herr, E. Ucur, K. Herzer, S. Okouoyo, R. Ridder, P.H. Krammer, et al., Glucocorticoid cotreatment induces apoptosis resistance toward cancer therapy in carcinomas, Cancer Res. 63 (2003) 3112–3120. [10] C.A. Mikosz, D.R. Brickley, M.S. Sharkey, T.W. Moran, S.D. Conzen, Glucocorticoid receptor-mediated protection from apoptosis is associated with induction of the serine/threonine survival kinase gene, sgk-1, J. Biol. Chem. 276 (2001) 16649–16654. [11] W. Wu, S. Chaudhuri, D.R. Brickley, D. Pang, T. Karrison, S.D. Conzen, Microarray analysis reveals glucocorticoidregulated survival genes that are associated with inhibition of apoptosis in breast epithelial cells, Cancer Res. 64 (2004) 1757–1764. [12] M. Karin, H.R. Herschman, Dexamethasone stimulation of metallothionein synthesis in HeLa cell cultures, Science 204 (1979) 176–177. [13] H.P. Rutz, Hydrodynamic consequences of glycolysis: thermodynamic basis and clinical relevance, Cancer Biol. Ther. 3 (2004) 812–815. [14] H.T. Sorensen, L. Mellemkjaer, G.L. Nielsen, J.A. Baron, J.H. Olsen, M.R. Karagas, Skin cancers and non-hodgkin lymphoma among users of systemic glucocorticoids: a population-based cohort study, J. Natl Cancer Inst. 96 (2004) 709–711. [15] M. Mueller, S. Strand, H. Hug, E.-M. Heinemann, H. Walczak, W.J. Hofmann, et al., Drug-induced apoptosis in hepatoma cells is mediated by the CD95 (APO-1/Fas) receptor/ligand system and involves activation of wild-type p53, J. Clin. Invest. 99 (1997) 403–413. [16] I. Nicoletti, G. Migliorati, M.C. Pagliacci, F. Grignani, C. Riccard, A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry, J. Immunol. Methods 139 (1991) 271–279. [17] M. Nakamori, M. Iwahashi, M. Nakamura, H. Yamaue, Clinical benefit of chemosensitivity test for patients with regional lymph

C. Zhang et al. / Cancer Letters 242 (2006) 104–111

[18]

[19]

[20]

[21]

[22] [23]

node-positive esophageal squamous cell carcinoma, J. Surg. Oncol. 84 (2003) 10–16. J.A. Koziol, D.A. Maxwell, M. Fukushima, M.E. Colmerauer, Y.H. Pilch, A distribution-free test for tumor growth curve analyses with application to an animal tumor immunotherapy experiment, Biometrics 37 (1981) 383–390. J.P. Ioannidis, P.J. Hesketh, J. Lau, Contribution of dexamethasone to control of chemotherapy-induced nausea and vomiting: a meta-analysis of randomized evidence, J. Clin. Oncol. 18 (2000) 3409–3422. The Italian Group for Antiemetic Research, Dexamethasone, granisetron, or both for the prevention of nausea and vomiting during chemotherapy for cancer, N. Engl. J. Med. 332 (1995) 1–5. M.E. Brady, G.P. Sartiano, S.L. Rosenblum, N.E. Zaglama, C.T. Bauguess, The pharmacokinetics of single high doses of dexamethasone in cancer patients, Eur. J. Clin. Pharmacol. 32 (1987) 593–596. N. Gassler, I. Herr. Mechanisms of glucocorticoid-induced cancer therapy resistance in solid malignant tumors, CSTT, in press. K. Muenstedt, D. Borces, M.K. Bohlmann, M. Zygmunt, R. von Georgi, Glucocorticoid administration in antiemetic therapy, Cancer 101 (2004) 1696–1702.

111

[24] H.P. Rutz, I. Herr, Glucocorticoid administration in antiemetic therapy: is it safe?, Cancer 103 (2005) 2696. [25] H.G. Iversen, G.H. Hjort, The influence of corticoid steroids on the frequency of spleen metastases in patients with breast cancer, Acta Pathol. Microbiol. Scand. 44 (1958) 205–212. [26] P. Sherlock, W.H. Hartmann, Adrenal steroids and the pattern of mestastases of breast cancer, J Am. Med. Assoc. 181 (1962) 313–317. [27] R. Kofler, The molecular basis of glucocorticoid-induced apoptosis of lymphoblastic leukemia cells, Histochem. Cell Biol. 114 (2000) 1–7. [28] S.L. Planey, G. Litwack, Glucocorticoid-induced apoptosis in lymphocytes, Biochem. Biophys. Res. Commun 20 (2000) 307– 312. [29] C.W. Distelhorst, Recent insights into the mechanism of glucocorticosteroid-induced apoptosis, Cell Death Differ. 9 (2002) 6–19. [30] A.C. Cato, S. Mink, BAG-1 family of cochaperones in the modulation of nuclear receptor action, J. Steroid Biochem. Mol. Biol. 78 (2001) 379–388. [31] Y. Shang, M. Brown, Molecular determinants for the tissue specificity of SERMs, Science 295 (2002) 2465–2468.