160 Abstracts/Lung Cancer I4 (1996) 149-179

~nwntion~ of IGF-I. Given reports of elevated IGFBP secretion in SCLC and inhibitionof IGF-I bioactivity by IGFBPs, these Endings may indicate that increased serum IGFBPs disrupt IGF-I regulation of GH secretion and glucose homeostasis.

Survival in early-stage non-small cell lung cancer NesbittJC,PutnamJB Jr, WalshGL,RothJA,MoumainCF. Deparrmenr of Thoracic. Cardiovascular Surgery, M.D. Anderson Cancer Center, Box 109.1515 Holcombe Blvd., Houston, lX 77030. AnnThorac Surg 1995;60:46&72.

The duration of survival in early-stage lung cancer (stages I and II) varies between reports in the literature. Several reasons account for this: patient population heterogeneity, inconsistent staging, anatomic variability, dissimilar tumor morphology, and unpredictile tumor biology. Tbis report addresses some of the issues in early-stage non- small cell lung caucer that relate to variability between estimates of survival in end stage reporting. We review several large series since the introduction of the IntemationsJ Staging System in 1986 and other selected, contemporary reports that address end results in patients with pathologic stage I or stage II lung cancer. Overall survival for patients withpathologicstageIdiseaseis64.6%(range,55%to72%)aud41.2% for patients with stage II disease (range, 29% to 51%). Reducing morphologic differences by placing patients in groups based on the TNMsubsetandrelInementincategorixationbymatchingTNMsubsets based on histology and other factors can improve considerably homogeneity and enhance prognostic predictability. The development of more accurate measures for predicting prognosis may serve to clarify the roles of primary and adjuvant treatment, particularly in those patientswithearly-stagediseaseassociatedwithpoorprognosticfactors in whom the potential for long-term survival is reduced.

The tumour marker ‘NeuronSpecific enolase’ as prognosis indicator in small-cell bronchial carcinoma Hoster M, Kirchheiner T, Ruble K-H. Klinikfir Pneumologie, Klinik Ambrock, Ambrocker Weg 61. D-58091 Hagen. Pneumologie 1995;49: 470-4.

To assess the predictive v&e of the tumour marker NSE in respect ofsllrvivatimep~~inpatientswithsmall-cellbronchial carcinoma we performed serial measurements of the NSE concentration in 67 patients in whom the small-ceII bronchial carcinoma had been newly diagnosed, before and during chemotherapy or radiotherapy. Pretherapeutic NSE determination proved au important predictive parameter with regard to survival time prognosis. In patients with an initial NSE concentration of over 60 @ml the survival time was significantly reduced (horn 12.1 months to 8.4 months, p < 0.01). The pretherapeuticNSEconcentrationsareassociatedwiththeinitialtumour stage. Sorial NSE determinations reflect the course of the disease. A drop in NSE concentrations during the early phase of therapy may be important as a secondary prognosis indicator.

P~mostic factors obtahed by a pathdo& examikio~ in completely resected ~owsmall-cell lung cancer: An BBfdYSiS in each pathologic stage 1chhse Y, Yano T, Asoh H, Yokoyama H, Yoshino I, Katauda Y. Department of Chest Surgery National Kyushu Cancer Center. 3-I-I. Notame, Minami-ku, Fukuoka 815. J Thorac Cardiovasc SUrg 1995;110:601-5.

We attempted to clarify what factors predominantly influence the mvival of patients with non-small-cell lung cancer in each pathologic stage on the basis of information generaBy obtained by a pathologic examination of completely resected non-small-cell lung cancer. The

subjectsinchtded243patientswithstageI,63withstageII,and108with stage IIIA disease. Pathologic features used in the analysis were as follows: tbegreatesttumorsize(%~.Ocmversus~3.O cm),tbehistologic cell type (squamous versus nonquamous celI carcinoma), the grade of differentiation, and tumor invasion of pleuraand vessels. In stage IIIA. the extent of the metastasis to the lymph nodes was also included in the analysis. The significant prognostic factors @ < 0.05) in stage I demonstratedbyatmivariateanaIysisofthesurvivaIcurvesincmdedt& tumor size, the grade of differentiation (well differentiated versus moderately and poorly differentiated tumor), pleural involvement, and invasion of the artery and vein. In addition, the histologic cell type and thepleuralinvolvementiustageIIandmvssionofthevemandtheextent of metastasis to the lymph nodes (NO and Nl versus N2) in stage IIIA were also found to be significant prognostic factors. A muhivariate prognostic factor analysis showed that the grade of ditTeremiation, pleural involvement, and venous invasion in stage I; the histologic cell type and pleural involvement in stage II; and venous invasion and mediastinai lymph node metastasis in stage IIIA were all predominant pmgnosticfactors.The~observationsthereforesuggestthata~olo~gthologic examination can identify the patients with a poor prognosis, which is different among the stages.

Small cell lung cancer with and without superior veer cava syndrome: A multivariate analysis ofprogno8tic factors in 408 axses Wurschmidt F, Bunemann H, Heilmann H-P. H-H-It&fur fur Strahlentherapie, AKSt. Georg. Lahmuhlenstr. 5, D-20099 Hamburg. Int J Radiat Gncol Biol Phys 1995;33:77-82.

Purpose: Patients with small cell lung cancer (SCLC) and superior vena cava syndrome (SVCS) are widely believed to have a grave prognosis. The purpose of this study was to detemtine the prognosis of patients with SCLC and SVCS as compared to SCLC without SVCS. Methodr and Materials: A retrospective analysis of 408 cases of SCLC f SVCS was performed. Three- hundred and sixty showed no clinical signs of SVCS and 43 (11%) had SVCS; in 5 patients no adequate information was available about clinical signs of SVCS. AU patients were classified as limited disease cases. About 98% received chemotherapy usually as the tirst treatment followed by radiotherapy. A median total dose of 46 Gy (range 30 to 70 Gy) was given at 2.0 Gy per fraction five times weekly. A prophylactic cranial hmdiation was applied if a complete remission was achieved alter chemotherapy or atIer 30 Gy of brad&ion. Kaplan-Meier survival curves are shown and comparisons were made by the log-rank and the Gehan/Wilcoxon test. To adjust for prognostic factors, a proportional hazards analysis was done. Results: Patients without SVCS had 5-year survival rates (* SE) and a median survival time (MST, 95% confidence intervals) of 1 I % l

2% and 13.7 months (12.7- 14.5) in UICC Stage I to III; in Stage III the figures were 9% f 2% and 12.6 months (11.2-13.7). In comparison, SCLC with SVCS had 5-year survival rates of 15% f 7% and MST of 16.1 months (13.8-20.5). The difference was significant in univariate analysis (Stage III disease: p = 0.008 by the log- rank test). In a muhivariate analysisofaIlpatients, Stage (Stage I+11 >III; p=O.O003), SVCS (yes > no; p = O.OOS), and Karnofsky performance status (% 70 -z SO-100%; p = 0.008) were of significant importsnce. Conclusions: SVCS is a favorable proguostic sign in SCLC. The treatment should bc curatively intended.

Long-term survival after removal of a single brain metastasis from lung carcinoma Giamundo A, Maiuri F, Iaconetta G, Signorelli F. Institute of Neurosurgery, School of Medicine, Universily ‘Federico II’, Naples. Cancer J 1995;8:215-7.

Abstracts/Lung Cancer 14 (19%) 149- I79 161

An unusual case of a IO-year survival after removal of lung adenc- carcinoma and brain metastasis is described and the pertinent literature is briefly reviewed. Although the prognosis ofpatients with lung tumors and brain metastases has significantly improved, cases with survival of more than 10 years remain rare. Patient candidates for long survivals are those with lung tumors different from the oat-cell carcinoma and with single supratentorial brain metastases, in whom complete removal of both primary and metastatic tumor is possible. In this selected group of patients an aggressive surgical treatment is justified.

Molecular and biological factors in the prognosis of non-small cell lung cancer Kanters SDJhI, Lammers SJ, Voest EE. Deportment Internal Medicine. University Hospital Utrecht. PO Box 85500. 3508 GA Utrecht. Eur Respir J 1995;8: 1389-97.

in 93.7% of cases. More specific information was obtained by express cytologic examination in 97.9% of cases of malignant growth and in 64.4% with benign tumors. On the basis ofthe test results, preoperative intratumor chemotherapy technique suggested by the authors may be recommended for clinical use.

p53 and disease progression in patients with non-small cell lung cancer Sauter ER. Gwin JL, Mandel J, Keller SM. Department ofSurgery, Beth Israel Medical Center. New York, NY 10029. Surg Oncol 1995;4: 157- 61.

For patients with non-small cell lung Caftcer the tumour/node/ metastasis (TNM) staging system and other conventional prognostic factors fail to predict the outcome oftreatment and survival accurately. New prognostic factors are urgently needed to improve understanding ofthe biologicalbebaviourofthediedifferentsubtypesofnon-small cell lungcancerandtorecognizepatients withagoodorpoorprognosis. This review will focus on molecular and biological factors published in the English language literature between 1988 and 1994. To be included in this survey, the predictive value ofa specific prognostic factor had to be contirmed by multivariate analysis in at least two different studies. Blood group antigen expression, ras oncogenes, microvessel density, and factors reflecting the proliferative state of the tumour may be important determinants of outcome of treatment. The search for new detemtinants of prognosis has provided insight in the complex tumour biology ofnon-small cell lung cancer and indicated possible targets for rumow therapy. Several promising prognostic factors have now been recognized. To validate these factors, prospective studies of a large patient population are needed. This ultimately serves the recognition of subsets of patients who may benefit from adjuvant therapy.

The peculiarities of metastatic spreading of lung cancer in cases of hereditary-related incidence ShutkinVA, WagnerRI, Barchuk AS. PetrovRes. InstituteofOncologv. Ministery of Health of Russian Fed.. St. Petersburg. Vopr Onkol 1994;40:25-9.

Specific features of metastatic spreading of lung cancer have been compared in 258 surgical patients from families with a history of lung cancer incidence (group I) and in 861 controls (group II). Lesions in lymph nodes at various stages were more frequent in group I, metastatic spreading incidence increasing in step with stage. In is noteworthy that more frequent lesions in mediastinal lymph nodes (stage IV) were observed in group I (54.2%). as compared with 11.8% in controls (P<O.OOl). Multiple lesions of lymph nodes at all stages of metastatic spreading featured prominently in 48.5% of group I, as compared with 17.0% in group II (P<O.OO I ). Metastatic spreading was studied versus such characteristics of primary tumor as size, histologic pattern and pattern of tumor growth.

The effectiveness of transthoracic biopsy in diagnosis and treatment of neoplasms of chest organs Rodzaevsky SA, Babiy YaS. Yugrinov GG, Pozmogov AI, KIapchuk AG, Tuganova T.N et al. UT. Res. Inst. Oncol. andRadio1.. Ministry of Health of L&aine. Kiev. Vopr Onkol 1994;40:29-35.

A complex examination of 1400 patients with pathologies of the lungsandmediastinum wascaniedoutusingroentgenologic,endoscopic, biopticsndcytoIogicmethods.DiagnosiswascontinnedmorphologicalIy

Present methods of predicting nodal progression preoperatively in patients with non-small cell hmg cancer (NSCLC) are inadequate. Our hypothesis was that ~53 expression in primary NSCLC would predict disease progression, making it a useful marker of adverse outcome. From 1987 to 1992,sixty-eight consecutive NSCLC patientsunderwent potentiallycurative lungresectionandmediastinal lymphnodedissection byonesurgeon.Primarytumoumwereanalysedusingthep53monoclonal antibody 180 1. p53 overexpression was found in 53% of tumours. ~53 expressiondidnot correlate withage,gender, histology or stage. A trend toward a higher incidence of ~53 expression was seen in tumours with nodal spread (P = 0.06), and ~53 expression correlated significantly (P = 0.03) with improved disease-free survival in patients with squamous cell carcinoma (SCC). p53 was the fourth most important independent predictor of survival, behind histology, gender and nodal disease. As a weakindependentpredictorofsurvival, thecorrelationofp53 expression with survival in patients with SCC must be evaluated with caution. If borne out in a larger patient population, ~53 expression may be a marker of nodal disease progression in patients with NSCLC.

Prognostic significance ofCD44 expression in adenocarcinoma of the lung Clarke MR. Landreneau RI, Resnick NM, Crowley R. Dougherty GJ. Cooper DL et al. Department of Pathologv, (Jniversity Pittsburgh Sch. Medicine, ZOOLothropStreet. Pittsburgh, PA 15232-2582. JClin Path01 Clin Mol Path01 1995;48:M200-4.

Aims - To determine whether expression of CD44 in neoplasia is associated with tumour grade, stage and prognosis. Methods - The immunohistochemical expression of CD44 was evaluated using the mouse antihuman monoclonal antibody 3G I2 which recognises regions shared by all CD44 isoforms to determine whether expression in formalin fixed, paraffin wax embedded tissue correlates with tumour grade, stage or survival in adenocarcinoma of the lung. Thirty one adenocarcinomas of the lung. 16 TZNO and I5 T2Nl . and their nodal metastases were studied. Results - Of the 3 1 tumours, 25 were positive for the CD44 antigen. CD44 expression correlated with tumour grade. in that intense staining was seen only in moderately and/or poorly differentiated tumours. CD44 did not correlate with nodal status. tumow size, pleural invasion, angiolymphatic invasion, or host inflammatory response, but did correlate with survival. A median survival of 46 monthswasobservedinpatientswithmoderatetostrongCD44expression compared with 24 months for those with no or weak expression. Nine patients were alive without evidence of disease at a median follow up of 61 months. Six (66%) of these nine patients had strong CD44 expression. This contrasts with strong expression in only three ( 17%) of the 17 patients dying with a median survival of 28 months. Conclusion - In primary adenocarcinoma of the lung loss of CD44 expression IS associated with less favorable outcome and may indicate a more aggressive neoplasm. CD44 may be a useful prognostic marker m lung carcinoma.