Lung Cancer (2008) 59, 240—245

avai lab le at www.sc iencedi rec t .com

journa l homepage: www.e lsev ier .com/ locate / lungcan

Serum levels of apoptosis biomarkers, survivinand TNF-alpha in nonsmall cell lung cancer

Duygu Derin ∗, Hilal Oguz Soydinc, Nese Guney, Faruk Tas, Hakan Camlıca,Derya Duranyıldız, Vildan Yasasever, Erkan Topuz

Institute of Oncology, Istanbul University, Istanbul, Turkey

Received 12 June 2007; received in revised form 26 July 2007; accepted 6 August 2007

KEYWORDSSurvivin;TNF-alpha;Apoptosis;Nonsmall cell lungcancer

SummaryContext: This study was conducted to investigate the prognostic role and the effects ofchemotherapy on serum apoptosis biomarkers consisting of survivin and tumor necrosis factoralpha (TNF-alpha) in patients with advanced stage nonsmall cell lung cancer (NSCLC).Materials and methods: Fifty-seven patients with newly diagnosed NSCLC were enrolled intostudy. Performance status was 0 or 1 in 47 patients and 2 in 10 patients. Thirty-two of themwere no or less than 10% weight loss. Patients were treated with platinum-based chemotherapy.Serum levels of TNF-alpha and survivin were determined by enzyme-linked immunosorbent assay(ELISA) technique.Results: While serum survivin levels in patients were not significantly different from controls(p = 0.321), serum TNF-alpha levels in patients were found significantly higher than in controls(p = 0.029). We found that serum TNF-alpha levels were increased (p < 0.001), whereas serumlevels of survivin (p = 0.025) were decreased by the chemotherapy effects. The changes of theTNF-alpha and survivin serum levels due to chemotherapy effect showed a significant negativecorrelation (r = −0.36 p = 0.007). The increase of serum TNF-alpha levels was independent fromchemotherapy response; however, the reduction of serum survivin levels was found only signif-icant in the chemoresponsive group (p = 0.039). While older age, weight loss and performancestatus yielded prognostic value, neither TNF-alpha nor survivin levels proved to be significantfor survival.Conclusion: Our findings suggest that the reduction in the serum survivin levels of advanced

NSCLC patients after chemotherapy can be used as a predictor of response to the chemotherapybut not that of survival.© 2007 Elsevier Ireland Ltd. All rights reserved.

∗ Corresponding author at: Kayseri Egitim ve Arastırma Hastanesi, Medikal Onkoloji Klinigi, Kayseri, Turkey.Tel.: +90 352 336 88 84x17 80; fax: +90 352 336 96 90.

E-mail address: dygderin@yahoo.com (D. Derin).

0169-5002/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.lungcan.2007.08.005

cancer 241

Table 1 Patient characteristics

Parameter n

No. of patients 57Age, yr median (range) 56 (33—77)Gender, male/female 50/7Weight loss (10%), yes/no 25/32Performance status, 0—1/2 47/10Histology, squamous cell/other 29/28Stage, IIIB/IV 20/37No. of metastasis, single/multiple 24/13Haemoglobin (12 g/dl), low/normal 27/28Albumin (3.5 g/dl), low/normal 5/41LDH (450 U/L), normal/elevated 37/13

wtopcdsWortil(a

2

Bcrbaf

(MdctsoTAahcs

Serum levels of survivin and TNF-alpha in nonsmall cell lung

1. Introduction

Nonsmall cell lung cancer (NSCLC) is the leading causeof cancer death in the world. More than three quar-ters of patients with NSCLC present with locally advanced(stage IIIB) or metastatic disease (stage IV). Although plat-inum containing combination chemotherapy is the mainstayof first-line therapy, overall survival of patients follow-ing treatment is limited [1]. The benefit of combinationchemotherapy in advanced disease is poor [2], therefore, itis important to identify biomarkers able to predict responseto chemotherapy and survival of patients undergoing treat-ment.

Abnormalities in the control of programmed cell deathor apoptosis, play an important role in tumorigenesis [3].Apoptosis is a multistep cascade regulated by proteins thatpromote or counteract cell death. The induction of apoptosiswas found to be a common event for different classes of anti-cancer agents and it is believed to be one of the main cellularmechanism by which chemotherapy and radiation therapykill cancer cells [4]. Antineoplastic agents induce apoptosisbecause DNA damage lead to inhibition of the antiapoptoticmolecules such as survivin and to product cytokines such astumor necrosis factor alpha (TNF-alpha) [5]. Inhibition ofsurvivin and increase of the TNF-alpha, activate caspase-3and consequently apoptosis occurs [6].

Survivin is a recently identified protein of the inhibitorof apoptosis protein (IAP) family which supresses apopto-sis [7]. It is undetectable in normal differentiated tissues[8,9] but becomes notably expressed in most common humancancer in vivo [9—11], especially in lung and breast can-cer cells [9,12]. Many reports have indicated correlationbetween survivin expression and either unfavorable prog-nosis [11,13—15], lack of response to chemotherapy [14,16]in various tumor tissues. In NSCLC, survivin expression havebeen reported to be indicators of poor prognosis, associatedwith a worse overall patient survival [17—19].

TNF-alpha is a cytokine that triggers apoptosis throughthe activation of caspase in a wide variety of cells [20].An antiproliferative effect of TNF-alpha has been demon-strated in various malignancies, such as colon [21,22], andrenal carcinoma [23], as well as malignant melanoma [24]. Invitro studies have shown an antiproliferative effect of TNF-alpha against various NSCLC cell lines [25,26]. A positivecorrelation between high TNF-alpha expression and favor-able prognosis in NSCLC was also demonstrated [27].

Based on these data, some mediators of apoptosis, suchas survivin and TNF-alpha may be predictor of responseto chemotherapy and prognostic factor on survival. In thisstudy, we aimed to investigate the prognostic roles and theeffects of chemotherapy of serum TNF-alpha and survivinlevels in patients with advanced stage NSCLC.

2. Materials and methods

2.1. Patients and methods

From June 2005 to November 2006, 57 patients recentlycytologically or histologically diagnosed with NSCLC andadvanced nonsurgical stage, were enrolled in the study.The patient characteristics are listed in Table 1. Patients

rbAts

Response to chemotherapy, yes/no 38/18Survival, alive/dead 20/37

ere not included if poor performance status (PS), morehan Eastern Cooperative Oncology Group (ECOG) PS 2,r any other chronic disease producing weight loss wasresent. The patients were treated with platinum-basedombination chemotherapy. Response to chemotherapy wasetermined after two cycles of chemotherapy according totandard World Health Organization (WHO) criteria [28].HO complete response is defined as the disappearance

f all evidence of disease, partial response is defined as aeduction by at least 50% in the sum of the product of thewo longest diameters of all lesions and progressive diseases defined as a 25% increase in the sum of the products of allesions or in the product of any one lesion. Normal controlsn = 24) were recruited from among the Institute personnel,nd all were in excellent health at the time of the study.

.2. Serum levels of survivin and TNF-alpha

lood samples were obtained from patients with lung can-er and healthy controls by venipuncture and clotted atoom temperature in fasting conditions on first admissionefore chemotherapy and after two cycles of chemother-py. The sera were collected following centrifugation androzen immediately at −20 ◦C until analysis.

The human total survivin enzyme immunometric assayEIA) kit (TiterZyme EIA, Assay Design, Inc., Ann Arbor,I, USA; Catalog # 900-111) was used for the quantitativeetermination of survivin in serum. The kit uses a mono-lonal antibody to survivin immobilized on a microtiter plateo bind the survivin in the standards or samples. After ahort incubation, the excess sample or standard is washedut and a rabbit polyclonal antibody to survivin is added.his antibody binds to the survivin captured on the plate.fter a short incubation, the excess antibody is washed outnd goat anti-rabbit immunoglobulin G (IgG) conjugated toorseradish peroxidase is added, which binds to the poly-lonal survivin antibody. Excess conjugate is washed out andubstrate is added. After a short incubation, the enzyme

eaction is stopped and the color generated is read at 450 nmy an automated ELISA reader (Rayto, RT-1904C Chemistrynalyzer). The measured optical density is directly propor-ional to the concentration of survivin in either standards oramples. A standard curve is prepared from survivin standard

242 D. Derin et al.

Table 2 Distribution of serum survivin and TNF-alpha values in patients with NSCLC and healthy controls

Patients (n = 57) Controls (n = 24) p

Median Range Median Range

.96

dt

IaicmksibbibfiStstiacrocAta

2

FuwapCataO

sls

3

TwTicw(

ll

pw(atnfwTsTep

rmnf

4

Survivin (pg/ml) 25.02 2.8—199TNF-alpha (pg/ml) 8.8 0.3—33.

ilutions and the amount of the product was then quanti-ated as picogram per milliliter (pg/mL).

The BioSource hTNF-alpha kit (BioSource International,nc. 542 Flynn Road Camarillo, California 93012, USA; Cat-log # KHC3011) is a solid phase sandwich enzyme-linkedmmunosorbent assay (ELISA). A monoclonal antibody spe-ific for hTNF-alpha has been coated onto the wells of theicrotiter strips provided. Samples, including standards of

nown hTNF-alpha content, control specimens, and patientpecimens, are pipetted into these wells. During the firstncubation, the hTNF-alpha antigen binds to the immo-ilized (capture) antibody on one site. After washing, aiotinylated monoclonal antibody specific for hTNF-alphas added. During the second incubation, this antibodyinds to the immobilized hTNF-alpha captured during therst incubation. After removal of excess second antibody,treptavidin—Peroxidase (enzyme) is added. This binds tohe biotinylated antibody to complete the four-memberandwich. After a third incubation and washing to remove allhe unbound enzyme, a substrate solution is added, whichs acted upon by the bound enzyme to produce color. Aftershort incubation, the enzyme reaction is stopped and the

olor generated is read at 450 nm by an automated ELISAeader (Rayto, RT-1904C Chemistry Analyzer). The intensityf this colored product is directly proportional to the con-entration of hTNF-alpha present in the original specimen.standard curve is prepared from hTNF-alpha standard dilu-

ions and the amount of the product was then quantitateds picogram per milliliter (pg/mL).

.3. Statistical analysis

requency tables and statistical analyses were performedsing by SPSS 11 for Windows. Normality was checkedith Kolmogorov—Smirnoff test. Comparison between serumpoptotic substances and prognostic parameters were com-ared with Student’s t-test and Mann—Whitney U-test.

omparison of survivin and TNF-alpha levels before andfter chemotherapy was calculated with paired samples t-est. Receiver-operating characteristics (ROC) curves werepplied to find the cutoff level of TNF-alpha and survivin.verall survival was estimated with Kaplan Meier analy-

Itmc

Table 3 Distribution of serum values of survivin and TNF-alpha in

Before chemotherapy

Median Range

Survivin (pg/ml) 25 2.8—199.9TNF-alpha (pg/ml) 8.8 0.3—33.6

36.2 11.6—106 0.3214.77 0.3—15.3 0.029

es and differences between groups were calculated withog-rank test. p value smaller than 0.05 was considered asignificant.

. Results

he levels of serum TNF-alpha and survivin in patientsith advanced NSCLC and healthy controls are shown in

able 2. The baseline serum TNF-alpha levels were signif-cantly higher in patients with advanced NSCLC than in theontrol group (p = 0.029). However, serum survivin levelsere not significantly different from healthy control group

p = 0.321).None of the prognostic parameters analyzed was corre-

ated significantly with the serum TNF-alpha and survivinevels (Table 1).

When we analyzed the effect of chemotherapy on serumarameters, we found that TNF-alpha levels (p = 0.001)ere significantly increased and conversely, survivin levels

p = 0.025) were significantly decreased due to chemother-py effect (Table 3). In the subgroup analyses, we foundhat TNF-alpha (p = 0.002 for responsive group, p = 0.018 foronresponsive group) levels were increased independentlyrom chemotherapy response while survivin levels (p = 0.039)ere significantly decreased only in the responsive group.he survivin levels were not changed in the nonrespon-ive group (p = 0.114) (Tables 4 and 5). The changes of theNF-alpha and survivin serum levels due to chemotherapyffect showed a significant negative correlation. (r = −0.36,= 0.007) (Fig. 1).

The median survival of all patient group was 10.2 months,ange 6.6—13.8 months. Older age, weight loss and perfor-ance status yielded prognostic value (Table 6). Conversely,

either TNF-alpha nor survivin levels proved to be significantor survival.

. Discussion

n this study, we investigated the serum levels of two impor-ant apoptotic substances, TNF-alpha and survivin, usingonoclonal antibodies in patients with advanced nonsmall

ell lung carcinoma. To our knowledge, this is the first

patients with NSCLC before and after chemotherapy

After chemotherapy p

Median Range

19.2 4.4—178.9 0.02511.8 0.3—35 <0.001

Serum levels of survivin and TNF-alpha in nonsmall cell lung cancer 243

Table 4 Distribution of serum values of survivin and TNF-alpha before and after chemotherapy, in patients with partial responseor stable disease (subgroup analysis)

Before chemotherapy After chemotherapy p

Median Range Median Range

Survivin (pg/ml) 25.4 2.8—199.9 19.1 6.38—178.9 0.039TNF-alpha (pg/ml) 8.63 0.9—33.6 11.8 0.3—26.5 0.002

Table 5 Distribution of serum values of survivin and TNF-alpha before and after chemotherapy, in patients with progressivedisease (subgroup analysis)

Before chemotherapy After chemotherapy p

Median Range Median Range

Survivin (pg/ml) 24.8 3.5—178.1TNF-alpha (pg/ml) 9.1 0.2—25.5

mvs

rshwcwcrtwrcsnasgsa

Fig. 1 Effect of chemotherapy on serum TNF-alpha and sur-vivin levels in patients with advanced NSCLC (p < 0.001 andp = 0.025, respectively).

report on serum survivin levels in NSCLC patients detectedusing monoclonal antibodies. Previous studies on apoptosishave been evaluated survivin expression in frozen tissue byreverse transcriptase polymerase chain reaction (RT-PCR)and/or in paraffin-embedded materials by immunohisto-chemistry (IHC). Therefore, we discuss our survivin resultsin comparison with the findings provided by these studies.

Survivin overexpression was found in lung cancer, but it

was not observed in other lung tissue lesions [9,17,29,30]. Incontrast to previous reports, we found that serum survivinlevels were not differents from healthy controls in patientswith NSCLC. This finding may be explained by the different

Table 6 Statistically significant prognostic factors on sur-vival in advanced NSCLC patients and the value of serumparameters on survival

Prognostic factor p value

Age 0.02Weight loss 0.005Performance status 0.002Serum surviving 0.434Serum TNF-alpha 0.725

scetsNcswsi

itsdi

18.6 4.4—132.6 0.11412.5 1.5—35 0.018

aterial in which we assesed survivin. We investigated sur-ivin levels in serum whereas all the other studies assessedurvivin in tumor tissues.

The relationship between survivin expression andesponse to the chemotherapy was assessed on tumor tis-ue of patients with NSCLC. Vischioni and colleagues [31]ave shown that the expression of survivin in patientsith advanced NSCLC did not correlated with response tohemotherapy. In another study [32], survivin expressionas assessed in patients with NSCLC after two cycles ofhemotherapy. They found a statistically significant cor-elation between survivin gene expression and responseo chemotherapy and additionally, survivin-negative casesere responsive to the chemotherapy. Neither of them

eported the variation in the survivin expression due tohemotherapy effects. Our study is the first report in thisetting. We found that serum survivin levels were sig-ificantly decreased after chemotherapy. In the subgroupnalyses, this reduction was significant only in chemore-ponsive group and was not significant in the non-responsiveroup. According to our findings, the decrease in the serumurvivin levels of advanced NSCLC patients after chemother-py, can be used as a predictor of response to chemotherapy.

The significance of survivin expression in the progno-is of NSCLC patients is not well characterized. Monzo andolleagues [17] found that the patients without survivinxpression had better overall survival. In agreement withhis study, two other studies reported that survivin expres-ion in tumor tissue is a predictor of shortened survival inSCLC [19,33]. In contrast to these studies, Vischioni andolleagues [31] and Falleni and colleagues [30] reported thaturvivin expression in tumor tissues did not correlate withorse prognosis in NSCLC. In agreement with the mentioned

tudies, we found that serum survivin levels do not have anympact on the survival of advanced NSCLC patients.

We found significantly elevated serum levels of TNF-alpha

n patients with advanced NSCLC compared with normal con-rols. Most of the published data demonstrated elevatederum TNF-alpha levels in patients with NSCLC [34—37]. Theata of these studies confirmed our results. The other find-ng in our study is the serum level increase of TNF-alpha

2

drspeleTwwtltltt

prIsiai

Tscsaatatstcsrsgaao

C

N

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

44

ue to chemotherapy effect. This increase was regardless ofesponse to chemotherapy but it is greater in the chemore-ponsive group than in the non-responsive group (p = 0.002,= 0.018, respectively). There are scanty data about theffects of chemotherapy on TNF-alpha levels. Niiya and col-eagues [38] reported that doxorubicin induced TNF-alphaxpression in human lung carcinoma cells. On the other side,as and colleagues [35] showed that serum TNF-alpha levelsere not changed by the chemotherapy effect in 28 patientsith NSCLC. Similarly, O’Brien and colleagues [39] reported

hat there were no detectable change in serum TNF-alphaevel after chemotherapy. The most probable explanation ofhese results may be small sample size of the studies. In aarger study group, elevation of serum TNF-alpha levels dueo chemotherapy effects may probably be detected similaro our study.

Studies which investigated the role of TNF-alpha levels onrognosis, found that serum levels of this cytokine was noteliable indicator of survival in patients with NSCLC [34—37].n agreement with the results of these studies, we found thaterum TNF-alpha level did not have any impact on survivaln this study. As expected, performance status, weight lossnd age older than 60 yielded prognostic value on survivaln our study.

In conclusion, the present study showed higher serumNF-alpha levels in advanced NSCLC patients, while serumurvivin levels were not significantly different from healthyontrol groups. Survival was influenced by neither serumurvivin nor TNF-alpha levels. Application of chemother-py provided significant decrease in serum levels of survivinnd increase in serum levels of TNF-alpha. The changes ofhe TNF-alpha and survivin serum levels due to chemother-py effect were negatively correlated. This finding suggestshat increased serum TNF-alpha levels and decreased serumurvivin levels induced apoptosis. Although the increase inhe serum TNF-alpha levels was regardless of response tohemotherapy, it was more significant in the chemorespon-ive group than in the nonresponsive group. Additionally,eduction of serum survivin levels after chemotherapy wasignificant only in chemoresponsive group. Our findings sug-est that the decrease in the serum survivin levels ofdvanced NSCLC patients after chemotherapy, can be useds a predictor of response to the chemotherapy but not thatf survival.

onflict of interest

one.

eferences

[1] Belani CP, Langer C. First-line chemotherapy for NSCLC:an overview of relevant trials. Lung Cancer 2002;38(Suppl4):13—9.

[2] Gridelli C, Rossi A, Maione P. Treatment of non-small-cell lungcancer: state of the art and development of new biologicagents. Oncogene 2003;22:6629—38.

[3] Cross M, Dexter TM. Growth factors in development, transfor-mation and tumorigenesis. Cell 1991;64:271—80.

[4] Kerr JF, Winterford CM, Harmon BV. Apoptosis: its significancein cancer and cancer therapy. Cancer 1994;73:2013—26.

[5] Raff M. Cell suicide for beginners. Nature 1998;396:119—22.[

D. Derin et al.

[6] Reed JC. The survivin saga goes in vivo. J Clin Invest2001;108:965—9.

[7] LaCasse EC, Baird S, Komeluk RG, Mackenzie AE. The inhibitorsof apoptosis (IAPs) and their emerging role in cancer. Oncogene1998;17:3247—59.

[8] Altieri DC. The molecular basis and potential role of survivin incancer diagnosis and therapy. Trends Mol Med 2001;7:542—7.

[9] Ambrosini G, Adida C, Altieri DC. A novel anti-apoptosisgene, survivin, expressed in cancer and lymphoma. Nat Med1997;3:917—21.

10] Tanaka K, Iwamoto S, Gon G, Nohara T, Iwamoto M, Tanigawa N.Expression of survivin and its relationship to loss of apoptosisin breast carcinomas. Clin Cancer Res 2000;6:127—34.

11] Kawasaki H, Altieri DC, Lu CD, Toyoda M, Tenjo T, TanigawaN. Inhibition of apoptosis by survivin predicts shorter survivalrates in colorectal cancer. Cancer Res 1998;58:5071—4.

12] Tamm I, Wang Y, Sausville E, Scudiero DA, Vigna N, Oltersdorf T,et al. IAP-family protein survivin inhibits caspase activity andapoptosis induced by FAS (CD95), Bax, caspases and anticancerdrugs. Cancer Res 1998;58:5315—20.

13] Lu CD, Altieri DC, Tanigawa N. Expression of a novel apop-tosis gene, survivin, correlated with tumor cell apoptosisand p53 accumulation in gastric carcinomas. Cancer Res1998;58:1808—12.

14] Kato J, Kuwabara Y, Mitani M, Shinoda N, Sato A, Toyama T,et al. Expression of survivin in esophageal cancer: correlationwith the prognosis and response to chemotherapy. Int J Cancer2001;95:92—5.

15] Takai N, Miyazaki T, Nishida M, Nasu K, Miyakawa I. Survivinexpression correlates with clinical stages, histologic grade,invasive behavior and survival rates in endometrial carcinoma.Cancer Lett 2002;184:105—16.

16] Zaffaroni N, Pennati M, Colella G, Perego P, Supino R, Gatti L,et al. Expression of the anti-apoptotic gene survivin correlateswith taxol resistance in human ovarian cancer. Cell Mol Life Sci2002;59:1406—12.

17] Monzo M, Rosell R, Felip E, Astudillo J, Sanchez JJ, MaestreJ, et al. A novel anti-apoptosis gene: re-expression of survivinmessenger RNA as a prognosis marker in non-small-cell lungcancer. J Clin Oncol 1999;17:2100—4.

18] Hofmann HS, Simm A, Hammer A, Silber RE, Bartling B.Expression of inhibitors of apoptosis(IAP) proteins in non-small-cell human lung cancer. J Cancer Res Clin Oncol 2002;128:554—60.

19] Ikehara M, Oshita F, Kameda Y, Ito H, Ohgane N, Suzuki R, etal. Expression of survivin correlated with vessel invasion is amarker of poor prognosis in small cell adenocarcinoma of thelung. Oncol Rep 2002;9:835—8.

20] Sidoti-De Fraisse C, Rincheval V, Risler Y, Mignotte B, VayssiereJL. TNF-� activates at least two apoptotic signaling cascades.Oncogene 1998;17:1639—51.

21] Schiller JH, Bittner G. Anti-proliferative effects of tumornecrosis factor, gamma interferon and 5-fluorouracil on humancolorectal carcinoma cell lines. Int J Cancer 1990;46:61—6.

22] Kemeny N, Childs B, Larchian W, Rosado K, Kelsen D. A phaseII trial of recombinant tumor necrosis factor in patients withadvanced colorectal carcinoma. Cancer 1990;66:659—63.

23] Skillings J, Wierzbicki R, Eisenhauer E, Venner P, Letendre F,Stewart D, et al. A phase II study of recombinant tumor necrosisfactor alpha in renal cell carcinoma: a study of the NationalCancer Institute of Canada Clinical Trial Group. J Immunother1992;11:67—70.

24] Lienard D, Lejeune FJ, Ewalenko P. In transit metastases of

malignant melanoma treated by high dose rTNF alpha in com-bination with interferon-gamma and melphalan in isolationperfusion. World J Surg 1992;16:234—40.

25] Hong WS, Jett JR, Sasaki Y, Takahashi H, Nakano H, Naka-gawa K, et al. Colony inhibitory effect of recombinant human

can

[

[

[

[

[

Serum levels of survivin and TNF-alpha in nonsmall cell lung

interferon(rH-IFN)-alpha-beta and -gamma on human lung can-cer cell lines. Jpn J Clin Oncol 1987;17:49—55.

[26] Yang SC, Owen-Schaub LB, Grimm EA, Roth JA. Induction oflymphokines-activated killer cytotoxicity with interleukin-2and tumor necrosis factor-alpha against primary lung cancertargets. Cancer Immunol Immunother 1989;29:193—8.

[27] Boldrini L, Gisfredi S, Ursıno S, Lucchi M, Melfi F, Mussi A, etal. Tumor necrosis factor-alpha: prognostic role and relation-ship with interleukin-8 and endothelin-1 in non-small cell lungcancer. Int J Mol Med 2006;17:887—92.

[28] Miller AB, Hoogstraten B, Staquet M, Winkler A. Reportingresults of cancer treatment. Cancer 1981;47:207—14.

[29] Tanabe H, Yagihashi A, Tsuji N, Shijubo Y, Abes S, Watanabe N.Expression of survivin mRNA and livin mRNA in non-small-celllung cancer. Lung Cancer 2004;46:299—304.

[30] Falleni M, Pellegrini C, Marchetti A, Oprandi B, Buttitta F,Barassi F, et al. Survivin gene expression in early non-small celllung cancer. J Pathol 2003;200:620—6.

[31] Vischioni B, van der Valk P, Span SW, Kruyt FA, RodriguezJA, Giaccone G. Nuclear localization of survivin is a positiveprognostic factor for survival in advanced NSCLC. Ann Oncol2004;15:1654—60.

[32] Karczmarek-Borowska B, Filip A, Wojcierowski J, Smolen A,Pilecka I, Jablonka A. Survivin antiapoptotic gene expression

as a prognostic factor in non-small cell lung cancer: in situhbridization study. Folia Histochem Cytobiol 2005;43:237—42.

[33] Kren L, Brazdil J, Hermanova M, Goncharuk VN, Kallakury BV,Kaur P, et al. Prognostic significance of anti-apoptotisis pro-teins survivin and bcl-2 in non-small cell lung carcinoma: a

[

cer 245

clinicopathologic study of 102 cases. Appl ImmunohistochemMol Morphol 2004;12:44—9.

34] Kaminska J, Kowalska M, Kotowicz B, Fuksiewicz M, GlogowskiM, Wojcik E, et al. Pretreatment serum levels of cytokinesand cytokine receptors in patients with non-small cell lungcancer, and correlations with clinicopathological features andprognosis. M-CSF—–an independent prognostic factor. Oncology2006;70:115—25.

35] Tas F, Duranyıldız D, Argon A, Oguz H, Camlica H, Yasa-sever V, et al. Serum levels of leptin and proinflammatorycytokines in advanced-stage non-small cell lung cancer. MedOncol 2005;22:353—8.

36] Kayacan O, Karnak D, Beder S, Gullu E, Tutkak H, Senler FC,et al. Impact of TNF-alpha and IL-6 levels on development ofcachexia in newly diagnosed NSCLC patients. Am J Clin Oncol2006;29:328—35.

37] De Vita F, Orditura M, Auriemma A, Infusino S, CatalanoG. Serum concentrations of proinflammatory cytokines inadvanced non small cell lung cancer patients. J Exp Clin CancerRes 1998;17:413—7.

38] Niiya M, Niiya K, Kiguchi T, Shibakura M, Asaumi N, ShinagawaK, et al. Induction of TNF-alpha, Upa. IL-8 and MCP-1 by dox-orubicin in human lung carcinoma cells. Cancer ChemotherPharmacol 2003;52:391—8.

39] O’Brien ME, Saini A, Smith IE, Webb A, Gregory K, MendesR, et al. A randomized phase II study of SRL172 combinedwith chemotherapy in patients with advanced inoperablenon-small-cell lung cancer and mesothelioma. Br J Cancer2000;83:853—7.