Early Diagnosis of Lung Cancer by Detection of Tumor Liberated Protein

GIULIO TARRO,* ANTONIO PERNA, ANO CIRO ESPOSITO Department of Diagnostics, Unit of Mo/ecu/ar Bi%gy, "D. Cotugno" Hospita/, Naples, and Unit of Path%gy "M. Scarlato" Hospita/, Scafati (SA), /ta/y

Tumor liberated protein (TLP) is a protein that can be used to reveal the early development of a tumor. Besides being formed in thl tumor, TLP is released in the blood when a patient starts producing cancer cells, which in tum enables the physician to intervene at, stage when the cancer is operable. To date, the available studies of tumor markers in lung cancer patients are CEA, NSE, TPA Chromogranine, CA 125, CA 19-9, and Cyfra 21-1. The sensitivity and specificity for serum markers ranges between 50 and 90% depending on the study and the cl inical samples analyzed. Most of these markers show an increased rate of positivity as the stagl advances. There are very limited data on TLP to draw any firm conclusion regardingthe diagnostic value ofthis marker. TLP has beer detected in 53.1 % of non-small celllung cancer (NSCLC) patients (N = 534) with 75% being positive in the early stage (stage I) an< dropping to 45% in the late stage (stage IV). However, 7.6% blood donar sera and 17.4% chronic lung disease sera have also teste< positive. In a confirmation study, the specificity was 89.94% and the sensibility was 63.63% from stage 111 to IV NSCLC patients.ln al initial study ofTLP as a marker for early detection in stage I, NSCLC patients showed a sensitivity of 66.7% and a specificity of 800;' for TLP compared to a sensitivity of 33.3% far CA 19-9, 11.1 % for Cyfra 21-1 and CA 125, and 0% for CEA; the specificity for ali fou of the latter markers was 100%. Using immunohistochemical analysis with peroxidase anti-peroxidase (PAP), we observed tha NSCLC cells were positive; we used the specific rabbit antiserum to TLP, which turned out negative in the presence of 1 mg/mlofthl synthetized peptide. The pre-serum was a/so negative. The same reactivity was found early in the modified epithelial cel/s o interstitial lung fibrosis and might be a predictive marker of celi transformation. The site of the peroxidase positivity wa~ cytoplasmic, of diffuse and/or granular type. J. CelI. Physiol. 203: 1-5,2005. © 2004 Wiley-Liss, Inc.



Lung cancer is the main cause of mortality in the world for men between 35 and 70 years of age. The highest incidence rates are found in Europe and North America. It is estimated that there were approximately 375,000 cases oflung cancer in Europe in 2000,303,000 in men and 72,000 in women; the resulting deaths totalled about 347,000 (280,000 men and 67,000 women) (Tyczynski et al., 2003). It is clear that lung cancer is a very important health problem for men, considering the mortality rate associated with the disease and the fact that the choices oftherapy for this tumor are limited. Air pollution, polynuclear aromatic hydrocarbons mostly derived from car traffic, radon and tobacco smoking surely play a fundamental role in causing this disease and therefore the first approach must be prevention measures. However, even though tobacco smoking and environmental factors' are certain causes of risk, they alone do not soom sufficient to justify the differences found in mortality and incidence for lung cancer within a country and between various countries. One must consider other important risk factors that act as independent factors or as modifiers or modulators of the effects of smoking. For this reason, new approaches are required to counteract this particular epidemie.

The discovery of biomarkers, that is markers of the biological process associated with cell growth, repre≠sents a basic target in tumor diagnosis, sincethey can be used for early diagnosis, tumor relapse and metastasis, as well as for prognosis, site of tumor, and rnonitoring of therapy.

Several studies have shown that the expression of different biornarkers is frequent in eelliines oflung cancer. and many authors have oonsidered the presence of these markers in blood as an index of the extension of disease, prognosis, and l'espanse to therapy. Some turnor markers

© 2004 WILEY-LISS, INe.

CNSE, Chromogranine, CEA, TP A, and CYFRA 21-1) havE been evaluated in many patients with SUlall celI lunli cancer (SCLC) as well as with NSCLC .• ;However, tht diagnosis and follow-up of this patho'ogy are not ye1 satisfactory, although a relationsmp \Vitti. clinical rele≠vance has been found for the NSE and chromograninE markers in SCLC; the CYFRA 21-1 marker, on the othel hand, has been useful only in the follow-up of NSCLC fOI the squamous subtype CSoichiro et al., 2001). Recently, there has been renewed interest in screening as spiraJ computerized tomography can detect smalI asymptomatic lesions more etfectively than conventional radiography. Although cure rates for such lesions are very good, there il; to date no evidence for the etfectiveness of mass screenin{i strategies (Tyczynski et al., 2003).


In 1983 a tumor associated antigen (TAA) was isolated from non small cell lung (NSCL) cancer (Tarro et al.: 1983) and named tumor liberated protein CTLP) (Tarro 2000). TLP is composed of soluble lipoglycoproteins; itf: size measures between 48.1 and 61 A and it has aIl isoelectric point less than 7.0 and a molecular weight 01 214,000 Da. It was first analyzed for amino acid composition and then some partial sequences were

Contract grant sponsor: Foundation T. and L. de Beaumont Bonelli (for Cancer Research).

*Correspondence 00: Giulio Tarro, Division of Virology, D. Cotugno Hospital, Via Posillipo n. 286,80123 Naples, Italy. E-mail: gitarro@tin.žt

Received 6 February 2004; Accepted 29 June 2004 Published online in Wiley InterScience ( 25 August 2004. DO!: 1O.lO02,jcp.20195




obtained. A 100 kDa (CSH-275) fragment was identified in TLP and one epitope was used as an antigen to produce antibodies directed towards specific sites ofTLP by inoculation in rabbits (Tarro et al., 1993). This antiserum was named CSH-419 and used by some authors for research on TAA in lung and colorectal cancer patients (Garaci et al., 1996; Guadagni et al., 2000). A reasonably high level of purification of TLP enabled us to obtain an initial amino acid sequence which in turn led to the synthesis of the foliowing three specific peptides from lung and colon cancer (Tarro, 1999): ArgThrAsnLysGluAlaSerlle; GlySerAlaXphe≠ThrAsn; AsnGlnArgAsnArgAsp. A fourth peptide, Gly≠ProProGlu ValGlnAsnAlaAsn, was obtained from TLP which was purified from an urogenital tumor.

Recent studies with rabbit antibody raised against a peptide from lung and colon cancer, made up of eight amino acids, revealed that TLP is produced in the cytoplasm ofthe malignant cells (Table 1, Fig. lA). Since it was not detected in normal tissues, we concluded that it was a tumor-specific antigen (Tarro et al., 1998). Our work on TLP indicates that this protein appears to be a new tumor marker with potential clinical applications. Using Western blot analysis, we were able to show that TLP was expressed in alilung cancer-derived celilines, in breast cancer-derived celi lines, and in colorectal≠derived celilines that were studied (Tarro et al., 1998). Since lung and intestinal tissues have a common embryonic origin, it is not unreasonable to suggest that TLP might be tumor-specific antigen for at least some of the malignancies that arise from tissues of epithelial origino Since we were not able to detect TLP in normal lung tissues, nor in the osteosarcoma celiline SAOS-2 (Tarro et al., 1998), the notion that TLP is indeed a tumor-specific antigen is supported. On the basis of the immunohistochemical studies which revealed a large amount of TLP in the columnar epithelial celis and lumen (secreted TLP), as illustrated in Figure lA, it appears that TLP is a cytoplasmic antigen which is specific for epithelial celi-derived neoplasms. Preincu≠bation ofthe antibody with the corresponding immuniz≠ing antigenblocked the immunocytochemical reaction, thus confirming the specificity of the antibody (Tarro et al., 1998).

Immunohistochemical evaluation

of the role of TLP

lmmunohistochemically, all the lung cancer speci≠mens with high or medium levels of difIerentiation

showed TLP cytoplasmic immunostaining (Table 1). The evaluated colorectal carcinomas also were positive for TLP expression (Tarro et al., 1999). No TLP protein was detected in the surrounding normal tissues or in inflammatory celis. In some specimens tms specific staining was also recognized, as well as in the lumen of atypical glands and in bronchial secretion. This last observation, taken together with TLP expression in all the adenocarcinoma-derived celllines (including MCF-7 and HT-29) and tumor specimens, could suggest that TLP is a secretory product ofthe neoplastic cells. On the other hand, the lower differentiated squamous carcino≠mas, the small celliung carcinomas (well known to be an aggressive type oflung cancer) and the two sarcomas did not exhibit any TLP immunostaining (Tarro et al., 1999).

Combining the immunohistochemical data with the results of the Western blot analyses, we conclude that TLP is a cytoplasmatic antigen specific for epithelial celi-derived neoplasms only, because it was not detect≠ed in non-neoplastic tissues celis, in neuroendocrine tumors, or in mesenchymal cell-derived tumors and normal cells (Rasi et al., 2000). In addition, finding that TLP is detectable in neoplasms and tumor celilines with common embryological features seems to suggest that TLP might be an "oncofetal antigen."



Immunoreactivity for TLP was found in neoplastic tissues by anti NQRNRD serum prepared after rabbit immunization obtained by using the sequence length of six amino acida (Tarro, 1999) according to the pattern already perforrned in the Cold Spnngs Hs.rbor Labora-

                                                                               tory (Tarro et al., 1993).      y

The immunostaining was always ~wasmic, with a low- to-absent background. AlI adenocarcmomas showed a high expression level ofTLP (Fig. lC,D). Tbe protein was distributed either in small granules or in larger accumulations (Fig. 2A.B); in some specimens, TLP was detected in the lumen of atypical glands and in the bronchial secretions (Fig. 2C,D), suggesting that TLP could be considered a secretory product ofthe neoplastic cells. We found cytoplasmic-specific TLP staining of various expression levels in alI medium- and well≠differentiated specimens of squamous carcinomas (Fig. lE). However, TLP was noi detected in the infiltrating lobular carcinoma of the breast (Fig. 3e), urothelial carcinoma (Fig. 30), or in the myocardium


TABLE 1. Reactivity of rabbit antiserum to the TLP-derived peptide (RTNKEASIC)




2. Lung

3.  Lung

4.  Lung

5.  Lung

6.Lung 7. Lung 8.Lung

7.  Lung

8.    Lung

9.     Lung

10. Skin

11.  Breast

12.  Breast

13.  Thymus

14.  Endometrium

15. Bladder

16.  Colon

Origin of tissue

Normal adult

Fetus 18 weeks

Interstitial inftammation Bronchiolo-alveolar adenocarcinoma Acinar adenocarcinoma

Papillary adenocarcinoma

Epidennoid carcinoma (differentiated) Epidermoid carcinoma (non-keratinizing) Small celi carcinoma


Malignant mesothelioma Malignant melanoma Infiltrating duct carcinoma Intraductal adenocarcinoma Malignant thymoma Adenocarcinoma

Papillary urothelial corcinoma Adenocarcinoma

Peroxidase positivity



O 3+ 3+ 3+ 2+ o o o O O


O o o

O 2+

!,:,\RLY llETE<.'TIOt': ClF lXt':(; C,\:\('I-:H IJY TU'





Fig. I. Histological seclion or n lung udennc(lI'cillomu incubated wilh Uw rabbit antiscrum to thc tumor liberated prlliein ('l'LP) pcptidc I{TNKEi\SI (A) und or CI squumous lung "arcinomu IDJ "" or u lung ndenoclll'cinoll1u \e anel DI incubull'd wii h the rahbit ant.iserulllln t.hc 'l'LP pcpt ide NC-tHNRD. The bruwn slaining n'p,.('s('nL~ the dii't.ribll≠tinn nfTLP in t.hcsc malignanlli~SlWi' lA. B and D: ,∑180: C: . R~()I.


lTable 21. A rnild oxpression level 01' l'LP was a]so dctccted in thc colorccLal adenocarcinoma examined cFig. 3A). Some outhors detected TLP in 53.117, 01' NSCL cancer patients, with 07r positivity in paticnts with tumors other than NSCL cancer;;, 7.6'ž; positi\'ily in unknown blood donors, alld 17.417, positž"ity in patients with chronic lung discascs h:wing ;)11 elevated risk far lung cancer (Garaci et al, 1996). 'l'ho antibody to NQRNRD sequence stains the alveolnr epithelium modified in interstitial lung fibrosis (Fig. 3B), 'l'his epithel ium changes from Oai io cubical: ihis alteration is tel'll1cd cubica] allomol'ph iSI11, indicating a retu rn of the alveolar epitheliul11 to the look of thc retallung berorc cxhibiting respil'aiory functions. Lung parcnchyma with diffuse inten;titial fibrosis does noi clisplay uny respiratory funciion due to thc compleie vanishing 01' septal capillaz'y circulation in this pathoJogy, as in t.h fetal lung {Perna et al., 1999L The pm'enchYl11a with

uch Inl)(JjfJcatiuIlS is Illore exposecl io neoplastic irans≠fOl'll1ation (Garaci et al.. 1996), Thc study of thc c1istribution 01' anti-TLP anlibodies in heallhy pcople a Il cl patients wilh neoplasia, or in other patients afTcctcd by Jung inflammator)' paiholog-ies who could l'alI L1nder thc category or Iling ncopla:o:i'l, is very interestin,g IEsposiio et al .. 1997), 1/1 ract, lhe rcuctivil)' Si),,'11UI in genel'8 I pat.hologie:; i!'l n uflcfiJ J tcst to USSCSi:ž thc spcci≠lžcity anel posit.ivc prcdict.ive v\llua 01' Lh\;! atil;iay. Finally,


rigo 2. TLP iIllIl111nohisluehclllll',,1 cxprcssiull b," thc alllibud~' lu NqR:--lHD ~eqtll'nrl'. A l'rpl'Psl'nIlllil'1' ~taining in a lungo ndplloral'ei≠IwnHl \A: . :i20: B: . 4S0: c: . :320: D: . -ISO I.


we believe that scnsltlvlty can increase scnsibility. with il11provement. o[ thc lst generaiion test (Tarru unel Esposito. 200~1.




Previou:> siudies have shown thc l'cadivity or the 1'I.lbbit anLiserlllll ubtainccl againsi lhc pcptide seqllencc RTNKEASl ('l'alTo el al .. ] 993. ] 998. :2001: Garaci eL al.. ] 996: 'l'ano, 1999, :2000. 2002: Rasi cl al.. 2000),


inl;e this scquencc is analogous to that present in cOl'in (RTQKEASIl, :J. proteic companent or h1ll1lan myocarclilll11. polyclana] antibodies were procluced b,v PRIMM (Milan. lLaly) in rabbiis againsi boih varianis l and 2 or I;orin w,l.. l'or lbc scqllcnl;c~ of juncižon or delction, respe<.:tively, to c1eten11ine whether the 'l'LP \Vas homologous or onl." 311alogolls lo lh!' COl'in.

An ELlSA sandwich was preparcd using t.hc rabbit aniiserum allii-RTNKEASI-KLH as a capillring anii≠body. purifžcd for immune affžnit.v (l;.\l with pcpticle. whel'c:ls c1irrerent antisera were usecl as t.he tracing' anLibody: poly Ab anti val' l delclion-BSA pl1l'i/ied wit.h proiein A: pol.v anti val' ] dcletion-KLH pllrifžccl l'or lA wilh pcpLide; p01v anlž \'tlr l juncl.ion-BSA plll'ificci wiih pl'olein A: pul." anii valI jllnction-KLll purifll.:d l'or lA wit.h pt'pticle: poi} 'lnti V\.1l' 2 deletion-BSA punllcd wlth proicin A; poJy anli val' 2 dclciion-KLH





sera were firsL purified against. pepLide + BSA and then after gel filtration on a NAP 25 column tpharmacia, Uppsala, Sweden l, as required by the conjugation proLocol (NaHCO;l 0,1 M -l- 0.9'H NaCI). Artel' gel filtra≠tion, lyophilization, and dissolution, the purified sera had higher antibody titcrs. The serum + EZ Iink was tested at diITerent dilutions (1:10,1:50, 1:100, 1:1,000) and a significantlossofactivity was measul'ed, When w Lested the sera at lower di lutions (1:10, 1:50, l: 100l, the results confirmed the loss or adivity.

The rcsuspension of Iyophilized sel'um in water doe not make it possible to obtain a homogeneous solution, but l'ather one conLaining precipitates, Therefore, the conjugation on 0,3 mg in 50 ~d or total and non-specilic IgG, conducted according to the proLocol, might have limitations l'or thc final yield, FurLhermore, we evalu≠ated the immunohistochemical expl'ession ol'TLP using the polyclonal antibodies to corin w,t, val'iants 1 and 2 l'or thc sequence ofjunction ancl deleLion, TLP was not dctccted in the clilferenLiatccl adenocarcinomns by ali the rabbit antisel'a (Fig. 4A,BJ, whereas the immunos≠taining was low to absent in the myocardium (Fig, 4C,Dl,


AIMS AND PERSPECTIVES FOR TLP Ongoing studies have attempted Lo achieve the com≠plete seguence 01' TLP (Tano, 2002), which in turn


Fig. 3. 'l'LP imlllulluhistochcmicai cxpression hy t1w antibÚdy lo NQRNRD sequence, A reprcscnl:ui\'c staining in a colo-l'celaI car≠cinomn tA:'480), Thc anlibod.\' LO NQR~RD scqucncý slaim; thc alvenlar L'pithcliull1 Illodificd in thc inlerstitiallung fibrosis (B: ,4801, NCI!Hlivc l'n!' TI.P illll1lllnostainiog nn adcnocal'cinomfl or lhc hrcast le: ,480. am} an urulclial curcinol1llJ IO: . :l201,


purifžcd for lA with pcptide; poJy anti val' 2 junction≠BSA purified with protein A: poly anti var 2 junction≠KLH pl.lrifžed for IA with peptide.

Ali antibodies were pUl"i{-jed by the poDI or antiscra of two l'abbits illlmunizeci and conjugated with biotine or with horseradish peroxidase (HRPl. Amigen samples were human myocardiulll anel norl1131 lung, The cun≠cenLl'atinn 01' HeLa cells and NSCLC extracts were lanci 511g/ml. respcctively, As Lhe control, an ELlSA standarel plaLe wa:; pl'cpal'ed to check thc con'cct procedure for the use uftbe Lestccl antibodies (gel fžltl'ation. Iypbilization, and dissolution in HzO, conjugaLion); thc coating wa pCl'fol'mecl with tbc antigens by which tbe antibodie against thc corin variants were induced in rabbits, The


TABLE 2, Hcactivil.v or ,'abhit antiscl'lIm LO thc TLP-dcrived peplidc fxQRNHlJl

PCl'Oxidnsc posi! h'il,\"



I nle!'slili,,1 libl'O;is ,\dcJJuciI n.:illlJrna Epidermoid clrdnuma M\'OCflrciiuITI

1.;fillrHling dUCI cnrcinurrw .-\dl'nOCOI'cinOIlI!t

C ,..,theltul Cill.,.inoll1o

;1. ~ . n l)

fig. 'I, A Iling ~.dcnOC1II'cllloma nl'gnti\'(; liJI' 'l'LP iml11unOSlllining- ily thL' :lIllibud)' lo tlH.' cOl'in v:t •. iaol I ,,l' jllnctioll lA: .4801 and by lhe anlibodv 111 lh" co!"in \'<1,';al1l 2 or juoclion (B:, ∑ISO), A human 1I1.\∑uulrcliulll nq;ut hT l'or illlmUIlOt'i1 ni,iillg by ih~ ;-'nt ihn,.1~∑ fo t hp {'m'in \'ill'innll ofjullctiunlC: ,4801 <lncl ~lighUy pll-"ili\'c fili' imlllunnSloill≠illg 1;.1 thc> anlihod,Y lo thc cOl'in variflnt 2 ofjuncliu" IU: .411111.




should enable in vitro preparations oflarge quantities of TLP by genetic engineering. The RT-PCR of the TLP fragment (300 bp), from the A549 celi line, was subcloned into the PGEM-T easy plasmid (vector) (Promega, Madison, Wl), resulting in the T-easy TLP construct (3,300 bp). The fragment (300 bp) could be eleaved with EcoRI. A putative open reading frame was deduced (Tarro, 2002).

The availability of such antigen preparations will facilitate future studies on the role of TLP in human malignancy. It could also enable the preparation of an assay for early diagnosis of the corresponding tumors (Soichiro et aL, 2001; Tarro and Esposito, 2002) and might even be useful in the generation of a specific anticancer vaccine to prevent neoplastic disease in sub≠jects at risk of developing cancer, through the stimula≠tion of the immune system by preparing on attack against celis that express this protein.

With the aim of identifying with the potential role of TLP in early lung cancer detection, anti- TLP antibodies were studied in lung cancer patients (Tarro and Esposito, 2002). Low sensitivity could result from the appearance of specific immunocomplexes able 10 block the detection ofTLP, when anti-TLP antibodies appear in the serum ofthe patient. In view ofthis possibility, we performed a study to determine the presence of anti- TLP antibodies in sera from NSCL cancer patients and also 10 determine whether these antibodies represent an immune response to TLP related to human lung cancer (Esposito et al., 1997). It was also important to show the correlation of the antibody with the time of appearance of the TLP in the blood of the patients (Tarro and Esposito, 2002). Furthermore, in 1997 it was shown for the first time that this antigen obtained from human lung cancer is capable of producing a humoral immune response (Esposito et a!., 1997).

Hybridomas producing anti-TLP monoclonal antibo≠dies (MAbs) were produced using the somatic fusion method with spleens of mice immunized with TLP≠derived peptide RTNKEASIC (paapep), conjugated with KLH and BSA (Tarro et al., 2001). Among the MAbs selected, eight belonged to the IgM elass and two to the IgG class. One of these MAbs, lF2/11, was selected for the setting up of an ELISA for titration of TLP in biologica! fluids. The test is based on the competition of the MAbs for antigen coated on microtiter plates and antigen in solution. The ELISA is able to recognize not only pep9aa but, more importantly, a crude preparation ofTLP, TLP in tumor extracts and TLP in human serum (Tarro et al., 2001).

For only six lung adenocarcinomas (Perna et al., 2001), we used the monoclonal antibody (MAb) 10 TLP diluted 1:100 on a single section with the goal oflooking for a possible relationship between intracellular TLP concentration, degree ofneoangiogenesis, and extension of angioinvasion (Ray and Stettler-Stevenson, 1994; Rak et al., 1995; Fontanini €t al., 1997; Pema et al., 1999). In two cases, we observed a high level of intra≠cellular TLP, a moderate degree of neoangiogenesis and lack of angioinvasion; in four cases, a high level ofintra≠cellular TLP corresponded with extended neoangiogen≠esis and elear-cut angioinvasion (Pema et al., 2001).

In conclusion, the results of our study emphasize the importance ofneoangiogenesis and angioinvasion in the

evolution of malignancies capable of strongly influen≠cing survival, as well as the antigenic proteic products associated with neoplastic tissue (TLP among them) and their possible role in the process of metastasis.

N ew studies presented in this review artiele show that a rabbit antibody raised against a second peptide of six amino acids (NQRNRD) from NSCLC belonging to the same protein previously detected and characterized (TLP) is produced in the cytoplasm of malignant celis. Our study indicates that this protein appears early in the modified epithelial cells of interstitiallung fibrosis and might be a predictive marker of celi transformation. These findings might allow for the preparation of an ELISA with a first sandwich between the polyclonal antibody against RTNKEASI and the serum sample capturing the antigen, and then a second inununological reaction between the second rabbit antibody (anti≠NQRNRD) and the previous complex for revealing the reactivity. A specificity >95% and an increased sensi≠tivity would yield a good method for the detection ofTLP antigen in the early stages oflung cancer.


Esposito c, Tarro G, Cuomo N, Morelli F. 1997. Anti-TLP antibodies in lung cancer patients. Int Med 5:191-194.

Fontanini G, Lucchž M, Vignanti S, Mussi A, Ciardiello F, De Laurentis M, De Placido S, Basolo F, Angeletti CA, Bevilacqua G~1997. Angiogenesis as a prognostic indicator of survival in non-small celIlung carcinoma: A prospective study. J Nati Cancer Inst 89:881-886.

Garaci E, Sinibaldi P, Rasi G. 1996. A new tumor-associated-antigen ofnon-small celI lung eancer: Tumor liberated protein (TLP) a possible turnor marker. Antieancer Res 16:2253-2256.

Guadagni F, Graziano P, Roselli M, Mariotti S, Bemard P, Sinibalsi-Vallebona P, Rasi G, Garaci E. 2000. Differential expression of a new tumor-associated≠antigen l'LP during human colorects1 cancer tumorigenesis. Am J Pathol 154:993-999.

Perna A, Tarro G, Esposito C, Baldi C, Annunziata S, Morellf'F, Cuomo N, Di Spirito A 1999. Neoangiogenesis and prognosis in lung can""'t- Int Med 7: 111≠113.

Pema A, Tarro G, Pema F, Esposito C. 2001. Irnrnunocl\elJUcaI evaluation of endothelal damage by angioinvasivity. Int J Clin Invest 9:3'l-33.

Ra.k JW, St Croix BD, Kerbel RS. 1995. Consequences of angiogenesis for tumor progression, metastasis and cancer therapy. Anticancer Drugs 6:3-18.

Rasi G, Sinibaldi-Vallebona p. Serafino A, Bernard P, Pierimarchi P, Guorino E, Faticanti-Scucchi L, Graziano P, Guadagni F, Garaci E. 2000. A new human tumor-associated antigen (TLP) is naturally espressed in rat DHD-K12 colorecta1 tumor cells. Int J Cancer 85:540-544.

Ray JM, Stettler-Stevenron WG. 1994. The role of matrix metalloproteases and their inhibitors in tumor invasion, metastatis and angiogenesis. Eur Resp J 7: 2062-2072.

Soichiro A, Hidek:i K, Neomichi I, Yasushi N, Masayuk:i S, Michiko A, Takayuk:i K. 2001. Optimal combination of seven tumor markers in prediction of advanced stage at first examination of patients with non-small celI lung caneer. Antieancer Res 21:3085-3092.

Tarro G. 1999. Tumor liberated protein (TLP). 1ta potential for diagnosis and therapy. Anticancer Res 19:1755-1758.

Tarro G. 2000. An overvŐew of the lung turnor liberated protein (TLP).

Characterization or the genetic immunologic profile. Int Med 8:5-8.

Tarro G. 2002. Characterization of a fragment containing a putative TLP eDNA sequence. Anticancer Res 22:2693-2696.

Tarro G, Esposito C. 2002. Progress and new hope in the fight against cancer:

Novel developments in early detection oflung cancer. Int Med 10:7-11.

Tarro G, Pederzini A, Flaminio G. Maturo S. 1983. Human tumor WltigeDS inducing in vivo delayed hypersensitivity and io vitro mitogenic activity. Oncology 40:248-254.

Tarro G. Marshak DR, Peroa A, Esposito C. 1993. Antigenic regioOB of tumor liberated protein complexes and antibodies agaiost the same. In: Carpi A, Sagripanti A, Grassi B. editors. Third Internation Congress. Advances in Management of Malignancies. Pisa, ltaly: 6/10 Decemher 1993. Biomed & Pharmacother 47:237.

Tarro G, Giordano A, Esposito C, Pema A, Claudio PP. 1998. Immunoistochem≠ieaI eharaeterizatioo of turnor liberated particles (TLPl expression pattern in lung cancer. Anticancer Res 18:2365-2370.

Tarro G, Perna A, Esposito C. Morelli F, Cuomo N, Di Spirito A 1999. Airns and perspectives for tumor liberated protein (TLP). Riv It Rie Med Chir 7:43-46.

Tarro G, Esposito C. Perna A, Cuomo N, Di Spirito A, Morelli F. Perna F, Maio A, Nolli ML. 2001. Production and characterization of anti∑TLP derived peptide (RTNKEASIC) monoclonal ant.ibodies. Int Med 9:55-61.

Tyczynski JE, Bra)' F, Parkin DM. 2003. Lung canoer in Europe in 2000:

Epidemiology, prevention, and early detection. Lancet QnCQI 4:45-55.