Correlation of KAI1/CD82 Gene Expression with Good Prognosis in Patients with Non-Small Cell Lung Cancer1

[CANCERRESEARCH56.1751â€”1755,April15.19961 Advances in Brief

Masashi Adachi, Toshihiko Taki, Yoshiaki leki, Cheng-long Huang, Masahiko HigasMyama, and Masayuki Miyake2

Department of Thoracic Surgery and Department V of Oncology, Kitano Hospita4 Tazuke Kofukai Medical Research institute, 13-3 Kamiya,na-cho, Kita-ku, Osaka 530 fM. A., T. T., Y. 1., C. H., M. M.J and Department ofSurgery. The Centerfor Adult Diseases of Osaka, Osaka 537 [M. HI, Japan

Abstract associated with poor prognosis (5, 6), and antigenic changes of the major blood group ABH antigensand Lewis-related antigensmay As part of our evaluation of members of the transmembrane 4 super reflect the clinical aggressivenessof NSCLC (7, 8). family as possible prognostic predictors, we performed a retrospective Recentinvestigationshavesuggestedthata genedesignatedKAIJ, study on the expressionof the recently Identified KAIJ geneby tumors of the lung This gene,which is identical to CD82,suppressestumor metes which is identical to CD82, is a metastasissuppressorgene for human tasis of prostate cancer,and Its decreasedexpressionmay be involved in prostatecancer(9). KAI1/CD82, a type III integralmembraneprotein malignant progression. We used reverse transcription-PCR to analyze with four transmembrane domains, belongs to the structurally distinct tumor tissues from 151 lung cancer pelients 74 tumors were stage I, 17 family of cell surface membrane glycoproteins, which includes mo were stageII, and 60 were stageifi. Our results indicate that while 35 tility-related protein-i (MRP-i)/CD9 (10), CD37 (1 1), CDS3 (12), patients had tumors in which the KAIJ/CD82 gene was conserved (posi ME49i/CD63 (13), TAPA-1/CD81 (i4), CO-029 (iS), and Sm23 tive), 116 patients had tumors with reduced geneexpression(negative). (16). Members of this so-calledTM4SF are variously expressedon The overall survival rate of patients with KAI1/CD82-positlvetumors was leukocytes and in a variety of mammalian tissues (17, 18), but their significantly higher than that ofpatients with KAJI/CD82-negativetumors actual precise functions remain unknown. In our previous studies on (77.4% versus 383%; P 0.002). Furthermore, the overall survival rate MRP-i/CD9, we showed that it regulates cell motility and that cul of patients with KAJ1/CD82-positiveadenocarcinomawasalso much higher than that of individuals whoseadenocarcinomahadreducedKAIJI tured tumor cells transfected with MRP-l/CD9 cDNA have reduced CD82 expression (73.4% versus 27.1%; P = 0.009). Multivariate analysis motility and metastaticpotentialto the lung (19). In addition, immu with the Cox regression model indicated that KAIJ/CD82 pOSItIVIty cor nohistochemical and immunoblotting analysesof breast cancer tissues related best with the overall survival rate, except for lymph node status. revealedthat MRP-i/CD9 expressiondiminishedas the clinical stage Our data suggest that high KAI1/CD82 gene expression by tumors of the of a given breast cancer advanced, and that MRP-1/CD9 gene and lung may beassociatedwith a goodprognosis.Thesefindings complement protein expression in metastatic lymph nodes was strikingly lower our earlier studieson MRP-1/CD9, anothermemberof the transmem than in the correspondingprimarymammary tumors(20). Moreover, brane 4 superfamily, whose reduced expressionIn non-small cell lung RT-PCR analysis of tumor specimens of patients with lung cancer cancer appearsto be a factor of poor prognosis.This set of observations disclosed that reduced MRP-1/CD9 gene expression was associated suggeststhat assessmentof the expression status of KAI1JCD82 and with poor prognosis (21). These data suggestedthat low MRP-1/CD9 MRP-1/CD9 by tumors may provide prognostic information on the din ical behavior of lung cancer. expression might be associatedwith the metastatic potential of certain human tumors. We have now extended our studies on the TM4SF by Introduction determining whether the levels of expression of the KAIJ/CD82 gene by tumor tissues are of value as a prognostic factor in predicting the As in many other countries, lung cancer has become the most clinical behavior oflung cancer. For this purpose, we applied RT-PCR commoncauseof cancerdeathof men in Japan(1). Althoughsurgical to measureKAIJ/CD82 gene expressionin tumor tissuesfrom 151 resection is considered to be the best treatment for NSCLC3 when the patients with NSCLC. tumor is localized in the pulmonary parenchyma, the survival rates of NSCLC patients have remained stable over the last decade. Even in Materials and Methods resectedstageI lung cancer,recurrencesstill occur in 20 to 30% of patients(2â€”4).Thesedatawould confirmthe notionthatmostpatients Clinical Characteristics of the Patients. We studied all patientswith present to their physician with systemic disease.To have an impact on NSCLCwhounderwentsurgerybetweenJanuary1991andDecember1993at survival, treatment protocols must be improved, and it seems to be theDepartmentofSurgeryoftheCenterfor Adult DiseasesofOsakaandatthe justified to evaluate surgery plus adjuvant chemotherapy or radiother Departmentof ThoracicSurgeryof the Kitano Hospital, Medical Research Institute of Osaka. Tumor-node-metastasis (TNM) staging designations were apy even in early diseasestages(2â€”4).Recentadvancesin molecular madeaccordingtotheInternationalPost-SurgicalPathologicalStagingSystem biology and genetics have created new diagnostic and treatment (22). A total of 169 patients was initially considered for inclusion in the study. possibilitiesfor clinical oncology.Mutationsof the tumor suppressor However,becauseadvancedstagelung cancer(stageIV) involves several gene p53 and of the K-ras oncogene have been reported to be ill-defined factorsandpatientshaddistantmetastases,eightindividualswith stageIV diseasewereexcluded,as were six patientswho had two or more Received12/28/95;accepted3/1/96. formsof cancersandfour patientswhodiedof causesotherthanlung cancer. The costs of publication of this article were defrayed in part by the payment of page Consequently,151patientswith lung cancerupto stage11Thwereincludedin charges. This article must therefore be hereby marked advertisement in accordance with the study.Their clinical recordsand histopathologicaldiagnoseswerefully 18 U.S.C. Section 1734 solely to indicatethis fact. documented. The salient clinical characteristics of the patients are presented in I This work was supported in part by Grants-in-Aid from the Ministry of Education in Japan(toM.M.). Table 1. This report includes follow-up data as of November 1, 1995. The 2 To whom requests for reprints should be addressed at Kitano Hospital, Tazuke medianfollow-up periodfor all patientswas41.2 months,with a rangeof 22 Kofukai Medical Research Institute, 13-3, Kamiyama-cho, Kits-ku, Osaka 530, Japan. to 58 months. 3 The abbreviations used are: NSCLC, non-small cell lung cancer; RT-PCR, reverse transcription-PCR;CD, cluster of differentiation antigen; TM4SF, transmembrane4 Tumor Specimens.Toascertainthepresenceofcancercells,one-halfof superfamily. each fresh tumor tissue specimen was immediately embedded in OCT (Miles, 1751

Table I Relationshipof KAI1/CD82 gene expressionfactorsKA1I/CD82and various prognostic Statistical Analysis. Overallcancer-specificsurvivalwasdefinedfrom the statusCharacteristicsTotal date of the operationto the date of death due to cancer.The statistical significanceof the differencebetweenthe incidenceof KAIJ/CD82-positive (â€”)PAge (n)(+) expressionandseveralclinical andpathologicalparameterswasassessedby @ at surgery(yr) the testor by Mann-Whitney'sU test.The Kaplan-Meiermethodwasused 43NSâ€•>609522605613 to estimatetheprobabilityof theoverallsurvivalasafunctionof time(27)and 73SexFemale401 was compared using the log-rank test (28). Multivariate analyses were per formed by using the Cox regressionmodel (29) using the SAS statistical 28NSL@MaleIll23 2 88Tumor package(SASInstitute,Cary,NC) andthefive factors(KAIJ/CD82statusof the tumor, sex,ageat surgery,T status,and N status)studied;scoreswere statusT14414 300.048'@T26515 assigned to each variable for the regression analysis. All Ps are based on 50T3305 two-tailed statistical analysis; P < 0.05 was considered to indicate statistical 25T4121 significance. 11Nodal statusN08827 610.008cN1264 Results 22N2304 26N370 KAIJ/CD82 Expression in Lung Tumor Tissues Analyzed by 7Pathological Quantitative RT-PCR. The KAIJ/CD82 expression rate ranged stageStage from 0 to 2.5 with a mean of 0.695. Of the iSi primary lung 490.003'StageI7425 15StageII172 cancers studied, 35 carcinomas (23.2%) were evaluated as KAIJ/ 35StageLIla427 CD82-positive and 116 tumors (76.8%) as negative (Fig. 1). 17Histology IHb181 Among the 116 KAJJ/CD82-negative tumors, there were 73 whose gene conservation rate ranged from 0. 1 to 1.2, and 43 whose Adenocarcinoma942371NSâ€•Squamous 38Large cell cancer5012 conservation rate was SO.1. 7Total cell cancer70 Relationship between KAIJ/CD82 Gene Expression and Known numberofpatients15135 116 Prognostic Factors. Analysisof the iS 1patientswhoselung cancers a NS, not significant. were testedfor KAIJ/CD82 geneexpressionrevealedthat there were [email protected]. no statisticallysignificantrelationships(f test)betweengeneexpres C Mann-Whitney's U test. sion and the patients' age at surgery, sex, or histology of the tumor (Table 1). By contrast, KAIJ/CD82 expression was significantly as Kankakee,IL).Frozensectionswerecuton thecryostatto athicknessof6 @m sociated (Mann-Whitney's U test) with lymph node status and immediately stained with H&E. One-half of a given tumor specimen (P = 0.008), tumor status (P 0.048), and pathological stage containing only cancer cells was used for RT-PCR. (P = 0.003). Of the 88 patients with N0 stage, 30.7% had positive RT-PCR Analysis. Total cellular RNA waspurified from freshor frozen KAJJ/CD82 gene expression, compared with 15.4% and 13.3% of tumortissuesbytheacidguanidiniumthiocyanateprocedure(23).First-strand those with N1 and N2 stage, respectively. None of the tumors of cDNA synthesiswas performedwith 5 ,.tg of total RNA using a cDNA patients with N3 stage was KAIJ/CD82 positive. On the other hand, synthesis kit (Pharmacia, Piscataway, NJ) and following the manufacturer's the tumors of 31.8% of the patients with T1 stage had conserved protocol. For PCR amplification, we used a 1 @laliquot of the reaction KAIJ/CD82 gene expression, as did 23.1% of the tumors of pa mixture.To obtainreproduciblequantitativeperformanceoftheRT-PCRassay tients with T2 stage, 16.7% of those with T3 stage, and 8.3% of for KAI1/CD82, we titrated the amount of starting cDNA and the number of amplification cycles. All subsequentassayswere carried out by using the those with T4 stage. With respect to pathological stage, the pro parametersthatyielded amplificationof both KAI1/CD82 and(3-actinDNA portion of patients whose tumors had conservedKAIJ/CD82 gene (the internal control) within a linear range.On the basisof the nucleotide expression decreased from 33.8% of those with stage I to 5.6% of sequence of KAI1/CD82 (9), 5'-AGTCCTCCCTGCTGCTGTGTG-3' was patients with stage Ilib. used as the senseprimer and 5'-TCAGTCAGCK3TGGGCAAGAGG-3' as the Association of KAJJ/CD82 Gene Expression with Overall Sur antisenseprimer.This primer pair amplifiesa 1031-bpfragment(nucleotides vival of Lung Cancer Patients. Comparing the survival among all 65-1095).Thereactionmixturewassubjectedto30 PCRamplificationcycles 151 patients according to the expression status of the KAJJ/CD82 of 40 s at 94Â°C, 40 5 at 60Â°C, and 90 s at 72Â°C. @-Actin DNA amplification gene, the overall survival rate for patients with positive tumors was was used as the internal PCR control (24); the senseprimer was 5'-CTGTCF significantly better than that of individuals whose tumors had reduced GOCGGCACCACCAT-3',andtheantisenseprimerwas5'-GCAACTAAGT KAJJ/CD82 gene expression (Table 2; Fig. 2). The overall survival CATAGTCCGC-3'.The samePCRconditionswereusedto amplify @3-actin DNA. Tubescontainingall ingredientsexcepttemplateswereincludedin all rate of the 35 patientswith KAJJ/CD82-positive tumorswas 77.4%, runs and served as negative reaction controls. Preparations of the KAI1/CD82- and that of the 1i6 patients with tumors that had reduced expression positivehumanendothelialcell line ECV3O4wereusedas positivecontrols was 38.5% (P = 0.002). Patientswith KAII/CD82-positive adenocar (25). The amplified DNA samples were run on a 1% agarose gel, and bands cinomas had progressively increased overall survival (Fig. 3A) in were visualizedwith ethidium bromide and photographedwith a Polaroid comparison with those whose adenocarcinomas had reduced gene camera. Densitometric analysis of the photographic negatives was used for expression (73.4% versus 27.1%; P = 0.009). The patients with bandquantification(26). squamous cell carcinoma (Fig. 3B) only showed such a tendency Specimen Classification Based on RT-PCR Results. The value obtained (87.5% versus 62.8%; P 0.064). for KAI1/CD82 by densitometryof the bandof a given tumor tissuesample Prognostic Value of KAIJ/CD82 Status. The variables used in wasdividedby thatof (3-actinandwasreferredto asthe KAIJ/CD82expres Cox regressionanalysisare shown in Table 3. The estimatedprog sionratio.Theexpressionratioof thetumorwasdividedby thatof thehuman endothelialcell line ECV3O4toobtaintheKAII/CD82 geneconservationrate. nostic value of each variable in relation to overall survival among the When the conservation rate value of a given specimen was > 1.2. it was 151 patients studied is expressedas a P value. Two variables (N stage considered to indicate conserved (positive) KAJJ/CD82 gene expression, and if andKAIJ/CD82 status)werefoundto be significantprognosticfactors the value was 1.2, it was consideredasdenotingnot conserved(reduced/ of survival; KAIJ/CD82 had a significant P of overall survival negative)expression. (P = 0.046). Other variables (T stage, sex, age at surgery, body 1752

Fig. 1. A, agarosegel electrophoresisof RT PCR-amplified 103l-bp KA1J/CD82 cDNA. Lane 1. size marker; Lane 2, human endothelial cell line KAIâ€”1/CD82 ECV3O4(positive control);Lanes3 and5, primary adenocarcinomasof the lung with conserved KA1I/CD82 gene expression (positive); Lanes 4 and6, primary adenocarcinomasofthe lung with reducedKA1I/CD82 geneexpression(negative); 1 2345678910 Lanes 7 and 9, primary squamous cell carcinomas ofthe lungevaluatedasKAJJ/CD82genepositive; Lanes 8 and 10, primary squamous cell carcinomas of the lung evaluatedas KAIJ/CDS2genenega B tive. B, agarosegel electrophoresisof amplified @-actinDNA (internal PCR control) of each specimen.

@@ â€” . â€” /3 â€”actin

1 2345678910

Table2 Overall survival rate of 151patientswith NSCLCaccordingto activated human T cells (30). CD82 was subsequently cloned as 1A4 clinicopathologica! characteristics and KAJI/CD82 status andC33 by otherlaboratories(31, 32). IA4 expressioncorrelateswith KAIJ/CD82status(no. conjugate formation, susceptibility to lysis by lymphokine-activated (%)KAII/CD82of patients)Overall survivalrate killer effectors, and reduced expression of HLA class I surface antigen KAIJ/CD82KAJJ/CD82KA1I/CD82Characteristics(+) (31).C33is associatedwithinhibitionofhumanTcell leukemiavirus type 1-induced syncytium formation (32). Expression of CD82 is not (â€”)(+)(â€”)pAge restricted to leukocytes; the protein is also expressedat high levels by at surgery(yr)6013 4372.727.50.023>6022 most other types of cells, including epithelial cells, endothelial cells, 7385.656.20.022SexFemale12 and fibroblasts (33). However, as with other members of the TM4SF, theprecisefunctionof CD82 remainsunknown,but its involvementin 2878.630.60.131Male23 signal transduction hasbeen suggested.A monoclonal antibody to 1A4 144.00.008Tumor 8880. induces calcium mobilization in U937 cells and tyrosine phosphoryl statusT114 3010037.70.040T215 ation of various proteins (34). These effects depend on the dual 5066.252.90.218T35 participation of both CD82 and Fc receptors, the main transducing 2580.023.10.036â€˜F4I receptor of monocytic cells. 11016.40.001Nodal The patterns of recurrencein lung cancer indicate that distant statusN027 6183.369.90.096N14 metastasesremain the primary cause of death of patients treated by 2275.023.10.433N24 surgery alone or surgery plus chemotherapy and that they are different 2650.018.50.174N30 for the various histological tumor types (2, 35). Martini a al. (35) 70.0Pathological reported that in stage II NSCLC, local recurrence occurred more stageStage 4981.879.30.436StageI25 frequently in patients with squamous carcinoma than those with 1550.024.00.861StageII2 adenocarcinoma. By contrast, distant metastasesare by far the most 3585.727.40.013Stageffib1lila7 common type of recurrence in patients treated for adenocarcinoma of 17010.30.122HistologyAdenocarcinoma23

10.009Squamouscell12 7173.427. (%) 3887.562.80.064cancerLarge @ 100 @@@1_@,@ --, KAIâ€”1/CD82 positive (n=35) 90 747.6Totalnumberof35cell cancer0 a, @@80 11677.438.50.002patients @ 70 @ 6) >60 @ 50 weight, serumalbumin, and carcinoembryonicantigen)had no rela C,) 40 c@ 30 tion with the KAIJ/CD82 status or its prognostic value. a, > 20 0 Discussion P=0.002

A gene designatedKAIJ was recently isolatedfrom human chro 0 10 20 30 40 50 60(M) Months after Surgery mosomellpii.2 (9). It hasbeensuggestedthatdecreasedexpression of this gene is involved in the malignantprogressionof prostateand Fig. 2. Overall survival of 151 patientswith lung cancer in relation to tumor KAIJI CD82 status. Thirty-five patients with KAIJ/CD82-positive tumors did not reach median othercancers(9). Â£411isidenticalto CD82, first identifiedby cDNA survival, whereasmediansurvival time of I 16 patientswith tumors that had reduced cloning as the R2 antigen that is strongly up-regulatedin mitogen expressionwas 1123 days. P was determinedwith the log rank test. 1753

A KAIJ/CD82 gene expression is associated with good prognosis of (%) NSCLC patients,especiallythose with adenocarcinoma.This set of 100 results would suggest that analysis of the status of the MRP-1/CD9 90 KAI 1/CD82 positive (nâ€”23) and KAJJ/CD82 genes in tumor tissues may provide prognostic in a, i@8o formation for patients with adenocarcinoma of the lung. However, as @ 70 reportedpreviously,severalmembersof TM4SF associatewithMHC @ > 60 ) class II glycoproteins in B cells or CD4 and CD8 in T cells, which @ 50 would suggest that there remain hitherto unknown multicomponent membrane signaling complexes in which members of the TM4SF are Ca30 a, important constituents (36, 37). To date it remains unknown whether > 20 0 there are interactions between MRP-1/CD9 and KAIJ/CD82, and to P=0.009 @@ 0 ,, I solvesuchproblems,muchmore in vitro investigationscombined @ 10 20 30 40 50 60 (M) with other molecular and biological components are needed, which Months after Surgery lead to useful clinical studies to disclose whether MRP-1/CD9 and B KAIJ/CD82 genes may improve prognostic evaluation in patients with lung cancer. (%) KAI-1 /CD82 positive (n=12) Our data show that high expression of the KAIJ/CD82 gene might 100 be a feature of some lung cancers with low metastatic potential. 90 @ a, L .@ i@80 However, it must be noted that certain patients with lymph node @ 70 metastaseshad primary tumors that were KAII/CD82 positive, even in an early stage.These differences with respect to the expression of this .@> 60 KAI-1 /CD82 reduced (n=38) gene point to several possible explanations: (a) as the product encoded (I) 40 by the KAIJ/CD82 gene has three N-linked glycosylation sites, con Ca 30 sistent with their presumed role in metastasis suppression, an aberrant a, > 20 glycosylation could result in the loss of the protein's function (38); 0 P=0.064 and (b) gene mutations could give rise to several types of altered 0 10 20 30 40 50 60 (M) proteins with diminished normal function. These and other questions

Months after Surgery about the role of KAJI/CD82 in lung cancer remain to be answered. Lung cancer remains one of the overwhelming social problems Fig. 3. A, overall survival of 94 patients with adenocarcinoma of the lung in relation because of its high incidence, poor surgical curability, and low effects to tumor KAJI/CD82 status. Median survival time of 23 patients with KAIJ/CD82-positive tumors was not reached and that of 71 patients with tumors that had reduced expression of combined systemic chemotherapy (2â€”4). In addition, screening was1059days.B,overallsurvivalofSOpatientswithsquamouscellcarcinomaofthelung programs have proven ineffective in increasing the overall survival in relation to tumor KAIJ/CD82 status. Median survival times of both of KAIJ/CD82 statuswere not reached. and reducing cancer-related mortality (39). Considering these circum stances,postoperative adjuvant therapy could become very important for improving the prognosis of lung cancer patients. It is conceivable the lung with the brain asthe mostfrequentsite(2, 35). The resultsof that the role of perioperativeadjuvant therapy may be of value for our studieson membersof the TM4SF are of particular interestin patientswhoseadenocarcinomahasreducedKAIJ/CD82 expression, relation to these clinical observations. As previously shown, reduced even if the tumor is localized to the pulmonary parenchyma. In MRP-1/CD9 gene expression correlates with poor prognosis in pa- contrast, patients with KAIJ/CD82-positive tumors may not require tients with lung cancer,and patientswith MRP-1/CD9-positive ade- adjuvant therapy after surgery. The determinationof the statusof nocarcinoma have a significantly better overall survival rate than KAIJ/CD82 expression, in combination with other molecular and adenocarcinomapatientswhose tumor has reducedMRP-1/CD9 cx- biochemicalassays,may improve the prognosticevaluation of lung pression (21). On the other hand, as documented here, conserved cancer patients and enhance the ability to prospectively identify

NSCLCVariableAssignedTable 3 Multivariate regressionanalysis offive variables in predicting overall survival of 151patientswith CI)PKAII/CD82Negative0â€”0.87810.44073.96980.416score(3SEWald g@Hazard ratio (95%

(0.175â€”0.986)0.046PositiveISexMale00.42370.31631.79451.528

(0.822â€”2.839)0.180FemaleIAge

(yr)6000.03100.27440.01281.031 (0.602â€”1.766)0.910>601T

statusT,10.25730.14743.05021.293(0.969-1.727)0.081T221,3T44N

statusN000.68640.136925.14541.987 (1.519-2.598)<0.001N1IN,2N,3

1754

individuals who are destined for early disease recurrence and who 18. Horejsi, V.. and Vlcek, C. Novel structurallydistinct family of leukocytesurface requireadjuvanttherapy.However, large-scaleprospectiveandretro glycoproteinsincludingCD9, CD37, CD53, and CD63. FEBS Lett., 288: 1â€”4,1991. 19. Ikeyama, S., Koyama, M., Yamaoko, M., Sasada,R., and Miyake, M. Suppressionof spective studies on various types of cancers are needed to establish cell motility and metastasisby transfection with human motility-related protein whether KAII/CD82 is indeed of practical utility as a prognostic (MRP-l/CD9) DNA. J. Exp. Med., 177: 1231â€”1237,1993. predictor. 20. Miyake, M., Nakano, K., leki, Y., Adachi, M., Huang, C., Itoi, S., Koh, T., andTaki, T. Motility related protein-l(MRP-l/CD9) expression:inversecorrelationwith me tastasesin breast cancer. Cancer Res., 55: 4127â€”4131,1995. Acknowledgments 21. Higashiyama, M., Taki, T., leki, Y., Adachi, M., Huang, C., Koh, T., Kodama, K., Doi. 0., and Miyake, M. Reducedmotility relatedprotein-I (MRP-l/CD9) gene We thankDr. MasatoYagitafor critical editorialhelpwith themanuscriptand expressionasa factor of poor prognosisin non-smallcell lung cancer.Cancer Res., Dr. Syun-itsuHatanakafor statisticalanalysis. 55: 6040â€”6044,1995. 22. Hermanek, P., and Sobin, L. H. (eds.). TNM classification of malignant tumors, Ed. References 4, pp. 75â€”80.Berlin:Springer-Verlag, 1987. 23. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acid 1. Statisticsand Information DepartmentVital Statistics1993 Japan,vol. 1, pp. 274â€” guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156â€” 279. Tokyo: Ministry of Health and Welfare, 1995. 159, 1987. 2. Bains, M. S. Surgical treatment of lung cancer. Chest, 100: 826â€”837,1991. 24. Nakajima-lijima, S., Hamada, H., Reddy, P., and Kakunaga, T. Molecular structure of 3. Bunn, P. A., Zandwijk, N. V., Pastorino,U., Aisner, J., Alberto, P., Arriagada, R., thehumancytoplasmic@-actingene:interspecieshomologyofsequencesinthe Carney,D., Comis, R., Diurich, C., Gatzemeier,U., Ginsberg,R., Greco, F. A., introns.Proc.NatI. Aced.Sci. USA, 82: 6133â€”6137,1985. Hansen, H. H., Harper, P., Henriksson, R., Huber, H., Klener, P., LeChevalier, T., 25. Takahashi, K., and Sawasaki, Y. Rare spontaneously transformed human endothelial Lewensohn, R., Murray, N., Niederle, N., Postmus, P., Rosell, R., Scaglioni, G., cell line providesusefulresearchtool.In Vitro Cell. Dev. Biol., 28A: 380â€”382,1992. Sculier, J. P., Splinter, T., Stahel, R., Symann, M., Thatcher, N., Tonato, M., and 26. Scheck, A. C., Mehta, B. M., Beikman, M. K., and Shapiro, J. R. BCNU-resistant Turrisi, A. First Euro-American forum on lung cancer treatment. Eur. J. Cancer, 30A: human glioma cells with overepresentationof chromosomes7 and 22 demonstrate 1â€”4,1994. increasedcopynumberandexpressionofplatelet-derivedgrowthfactorgenes.Genes 4. Mountain, C. F. A new international staging system for lung cancer. Chest, 89: Chromosomes & Cancer, 8: 137â€”148,1993. 2255â€”2335,1986. 27. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations. 5. Mitsudomi, T., Oyama, T., Kusano, T., Osaki, T., Nakanishi, R., and Shirakusa. T. J. Am. Stat. Assoc.,53: 457â€”481,1958. Mutations of the p53 gene as a predictor of poor prognosisin patientswith non 28. Mantel, N. Evaluation of survival data and two new rank order statistics arising in its small-cell lung cancer. J. Natl. Cancer Inst., 85: 2018â€”2023,1993. consideration. Cancer Chemother. Rep., 50: 163â€”170,1966. 6. Rosell,R., Li, S., Skacel,Z., Mate, J. L, Macsue,J., Canela,M., Tolosa,E, Armengol, 29. Cox, D. R. Regressionmodelsand life-tables.J. R. Stat. Soc. B, 34: 187â€”220,1972. P., Bamadas, A., and Ariza, A. Prognostic impact of mutated K-ms gene in surgically 30. Gaugitsch.H. W., Hofer, E., Huber, N. E., Schnabal,E., and Baumruker,T. A new resectednon-small cell lung cancer patients.Oncogene,8: 2407-2412, 1993. superfamilyof lymphoid and melanomacell proteinswith extensivehomologyto 7. Miyake, M., Taki, T., Hitomi, S., and Hakomori, S. Correlation of expressionof Schistosoma mansoni antigen Sm23. Eur. J. Immunol., 2!: 377â€”383,1991. @ antigens with survival in patients with carcinoma of the lung. N. Engl. J. 31. Gil, M. L., Vita, N., Lebel-Binay, S., Miloux, B., Chalon, P., Kaghad,M., Marchiol Med., 327: 14â€”18,1992. Fournigault, C., Conjeaud, H., Caput, D., Ferrara, P., Fradelizi, D., and Quillet-Mary, 8. Ogawa, J., Sano,A., Inoue, H., and Koide, S. ExpressionofLewis-relatedantigenand A. A memberof the tetraspanstransmembraneproteinsuperfamilyis recognizedby prognosis in stage I non-small cell lung cancer. Ann. Thorac. Surg., 59: 412â€”415, a monoclonalantibodyraisedagainstan HLA classI-deficient, lymphokine-activated 1995. killer-susceptible, B lymphocyte line. J. Immunol., 148: 2826â€”2833,1992. 9. Dong, J-T., Lamb, P. W., Rinker-Schaeffer, C. W., Vukanovic, J., Ichikawa, T., 32. Imai, T., Fukudome, K., Takagi, S., Nagira, M., Furuse, M., Fukuhara, N., Nishimura, Isaacs,J. T., and Barrett, J. C. KAI-l. a metastasissuppressorgene for prostate cancer M., Hinuma, Y., and Yoshie, 0. C33 antigen recognizedby monoclonalantibodies on humanchromosomellpl 1.2.Science(WashingtonDC), 268: 884â€”886,1995. inhibitory to human T cell leukemia virus type 1-induced syncytium formation is a 10. Miyake, M., Koyama, M., Seno, M., and lkeyama, S. Identification of the motility member of a new family of transmembraneproteinsincluding CD9, CD37, CD53, relatedprotein(MRP-l), recognizedbymonoclonalantibodyM3l-l5, whichinhibits and CD63. J. Immunol., 149: 2879â€”2886,1992. cell motility. J. Exp. Med., 174: 1347â€”1354,1991. 33. Engel, P., Wagner, N., and Tedder, T. F. CD82 workshop report. In: S. F. 11. Classon, B. J., Williams, A. F., Willis, A. C., Seed, B., and Stamenkovic, 1. The primary structureof thehumanleukocyteantigenCD37,a specieshomologueofthe Schlossman,L. Boumsell, W. Gilks, J. M. Harlan, T. Kishimoto, C. Morimoto, J. Ritz, S. Shaw, R. Silverstein, T. Springer, T. F. Tedder, and R. F. Todd (eds.), rat MRC OX-44 antigen.J. Exp. Med., 169: 1497â€”1502,1989. 12. Amiot, M. Identification and analysisof cDNA clones encodingCD53: a pan Leukocyte Typing V, pp. 691â€”693.Oxford: Oxford University Press, 1995. leukocyteantigenrelatedto membranetransportproteins.J. Immunol., /45: 4322â€” 34. Lebel-Binay, S., Lagaudriere, C., Fradelizi, D., and Conjeaud, H. CD82, member of 4325, 1990. the tetra-span-transmembraneproteinfamily, is a costimulatoryprotein for T cell 13. Houa, H.. Ross, A. H., Huebner,K., Isobe, M., Wendeborn,S., Chao, M. V., activation. J. lmmunol., 155: 101â€”110,1995. Ricciardi,R. P.,Tsujimoto,Y., Croce,C. M., andKoprowski,H. Molecularcloning 35. Martini, N., Burt, M. E., Bains, M. S., McCormack, P. M., Rusch, V. W., and and characterizationof an antigen associatedwith early stagesof melanomatumor Ginsberg,R. J. Survival after resectionof stageII non-smallcell lung cancer.Ann. progression.CancerRes., 48: 2955â€”2962,1988. Thorac.Surg.,54: 460â€”466,1992. 14. Oren,R., Takahashi,S.,Doss,C., Levy, R., andLevy, S. TAPA-l, the targetof an 36. Angelisova, P., Hilgert, I., and Horejsi, V. Association of four antigens of the antiproliferativeantibody,definesanewfamily of transmembraneproteins.Mol. Cell tetraspansfamily(CD37, CD53, TAPA-l, and R2/C33)with MHC classII glyco Biol., 10: 4007â€”4015,1990. proteins.Immunogenetics.39:249â€”256,1994. 15. Szala, S., Kasai, Y., Steplewski, Z., Rodeck, U., Koprowski, H., and Linnenbach, 37. Imai, T., and Yoshie, 0. C33 antigen and M38 antigen recognized by monoclonal A. J. Molecular cloning of eDNA for the human tumor-associatedantigenCO-029 antibodiesinhibitoryto syncytiumformationby humanT cell leukemiavirus type 1 andidentificationof relatedtransmembraneantigens.Proc.Natl. Acad.Sci.USA,87: are both members of the transmembrane 4 superfamily and associatewith each other 6833â€”6837,1990. andwith CD4 or CD8 in T cells.J. Immunol.,151: 6470â€”6481,1993. 16. Wright, M. D., Henkle, K. J., andMitchell, G. F. An immunogenicM@23,000integral 38. Hakomori, S. Aberrant glycosylationin tumors and tumor-associatedcarbohydrate membraneproteinof Schistosomamansoniwormsthat closely resemblesahuman antigens. In: G. F. Vande-Woude and G. Klein (eds.), Advances in Cancer Research, tumor-associated antigen. J. Immunol.. 144: 3195â€”3200,1990. pp. 257â€”331.San Diego: Academic Press, 1989. 17. Wright, M. D., and Tomlinson,M. G. The ins and outs of the transmembrane4 39. Fontana, R. S., Sanderson,D. R., and Woolner, L. B. Lung cancer screening:the superfamily. Immunol. Today, 15: 588â€”594,1994. Mayo program.J. Occup.Med., 28: 746â€”750,1986.

1755

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1996 American Association for Cancer Research. Correlation of KAI1/CD82 Gene Expression with Good Prognosis in Patients with Non-Small Cell Lung Cancer

Masashi Adachi, Toshihiko Taki, Yoshiaki Ieki, et al.

Cancer Res 1996;56:1751-1755.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/8/1751

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/56/8/1751. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.