129

classification tree by the following eight attributes: DE, PS, serum AP, serum LDH. mediastinal spread, sex, WBCC, and liver metastasis. The four groups were distinguished by median survival umes of 59,49,35, and 24 weeks, respectively (P = .OOOl). Interactions among prognostic factors are emphasized in the RECPAM classification model as evi- denced by reassignment of patients across conventional staging barriers into alternate prognostic groups. The advantages of using RECPAM over the more convenhonai Cox regression techniques for a new staging system are discussed.

Survival determinants in extensive-stage non-small-cell lung can- cer: The Southwest Oncology Group experience Albain KS, Crowley II, J_eBlanc M, Livingston RB. Southwest Oncol- ogy Group, Operarions Ofice, 5430 Fredericksburg Rd. San Anmnio.

7X 78229-6197. J Clin Oncol 1991;9:1618-26. We analyzed the 2,531-patient Southwest Oncology Group exten-

sive-stage non-small-cell lung cancer (ENSCLC) data base from 1974 to 1988 to(l)assessthe interactionsofhost-ortumor-relatedprognostic factors and therapy using Cox modeling and recursive partitioning and amalgamation (RPA) to determine whether each independently pre- dicts outcome, and (2) use RPA to define prognostic subsets with different survival potentials. Good performance status (PS), female sex, and age = 70 years were significant independent predictors in a Cox model applied to the entire population. In a second Cox model for patients with good PS enrolled on recent studies, hemoglobin level = 11 .O g/dL, normal lactate dehydrogenase (LDH), normal calcium, and a smgle metastatic site were sigmficant favorable factors. The use of cisplalin was an additional independent predictor of improved outcome in both Cox models after adjustments for year of accrual and all prognostic variables. The favorable effect of cisplatin was observed in each of six RPA-derived subgroups from the entire population. A second RPA of 904 patients from recent trials (nearly all received cisplatin-based therapy) resulted in three distinct prognostic subsets based on PS,age, hemoglobin, and LDH: = I -year survivals were 27%. 16%, and 6% (P < .OOOl). The best survival occurred for patients with a good PS who had a hemoglobin level = 11 g/dL and who were older than 47 years. This analysts suggests that although several factors were Independent variables in the Cox models, three important prognostic subgroups were easily defined through RPA. Together with other analyses, our results suggest the need to modify the stage IV category m NSCLC.

Correlation of modal chromosome number of cultured non-small cell lung carcinomas with DNA index of solid tumor tissue Siegfried JM. Ellison DJ, Resau JH, Miura I, Testa JR. Department of Pharmacology, University of Pirrsbwgh. E 1347 Biomedical Science

Tower, Pittsburgh, PA 15261. Cancer Res 199 I;5 1~3267-73. The modal chromosome number of 13 non-small cell lung carcino-

mas placed into culture was compared to the DNA index of the tumor tissue as measured by flow cytometry in order to determine whether cytogenetic results from such cultures are representative of the original solid tumor. The modal chromosome number observed in culture, which ranged from 45-146, fell within the range of aneuploidy pre- dicted from the DNA content of the original tissue in all 13 cases. In 7 cases, flow cytometry results showed that the aneuploid G,/G, popttla- lion of the tumor tissue (DNA index of 1.5 or higher) represented 1 I- 76% of the cells present, while diploid cells (presumably normal tissue) made up the remainder of the population. In these 7 cases, modal chromosomes numbers of 61-92 were found in tumor cells cultured from the tissue. In 3 cases, only a diploid or neardiploid population was found by flow cytometry, consistent with the near-diploid modal chromosome number of cultured cells observed (45-55). In 3 cases, the aneuploid Cl/G0 population (DNA index of 1.5, 1.6, and 3.2) of the original tissue represented only a small fraction of the solid tumor (I- 5% of cells). Modal chromosome number found in cells cultured from these 3 cases was 64-69, 62-68, and 136-146, which is in close agreement with the aneuploid peak observed in the tissue. Histological

analysis of the tumor tissue in two of the latter cases showed large numbers of infiltrating lymphocytes and/or stromal tissue which could have dominated the measurement by flow cytometry. In the third case, tumor cells made up at least 75% of the spectmen examined, implying that part of the population in the “diploid” peak contained tumor cells in this specimen. Only the aneuploid population was detected in culture of this tumor. Agreement between flow cytometry andcytogenetics was found in cases in which metaphase spreads were obtained within a few days of culture as well as after several months. These results indicate that highly aneuploid populations are found m many, but not all, non- small cell lung tumors. Although in some cases multiple populations may exist in the tumor which do not all proliferate in vitro. tumor cells which are found in culture of solid lung carcinomas are representative of the original tumor. Flow cytometry findings in the solid tumors confirmed the findingsofaneuploidy observed by cytogenetic analysis.

Neurological paraneoplastic syndromes in patients with small cell lung cancer. A prospective survey of 150 patients Ehington GM, Murray NMF, Spiro SG, Newsom-Davis J. Deparnenr

of Clinical Neurology, Radcliffe Infirmary. Woodsrock Road, Oxford

OX2 6HE. J Neural Neurostug. Psychiatry 1991;54:764-7. One hundred and fifty patients presenting with small cell lung cancer

(SCLC) to chest physicians, were assessed neurologically. Neummtts- cular or autonomic deficits were common and occurred in up to 44% of cases. Weakness, dry mouth, and weight loss were not mutually independent and may represent the syndrome formerly described as carcinomalous neuromyopathy. By contrast, undoubted paraneoplaslic syndromes were much less commonly detected. Two patients had the Lambert-Eaton myasthenic syndrome (LEMS) and one had subacute sensory neuropathy (SSN). In these patients, neurological symptoms antedated other manifestations of cancer, by between six and 17 months. The 95% confidence Interval for the prevalence of LEMS or SSN among SCLC patients was O-4%, consistent with the results of previous retrospective or smaller studies: summing these, the overall prevalence of LEMS among SCLC patients is close to 3%, which tmpliesabout250newcasesperannum in Englandand Wales. IfLEMS and SSN are the least uncommon neurological paraneoplastic syn- dromes in SCLC patients, this may reflect the accessibility of motor nerve terminals and dorsal root ganglia to cross-reactive anti-lumour cell antibodies.

Quality of life in lung cancer Geddes DM. Royal Bromplon and National Heart Hospital. Sydney

Street, London SW3 6NP. Respir Med 1991;85:SuppI 8:7-l 1. The prognosis of a patient with lung cancer is poor and the quality is

at least as important as the quantity of remaining life. Quality of life is a useful concept which is almost impossible to define but there are a number of important factors which contribute to it. Culture, religion, previous experience and the point of view of the individual all contri- bute to which of these factors are considered most important. Any quality of life assessment will, therefore, only apply to a defined community. The measurement of quality of life in cancer trials should concentrate on a few important categories such as physical symptoms, psychology and social factors and should be simple rather than conpre- hensive. In addition, a few frequent measures are better than an occasIonal comprehensive survey and ideally, both approaches should be combined and compared. Such measurements are most useful for comparative trials rather than for making an overall quality of life estimate. Finally, for routine clinical use outside clinical trials the quality of life index or the Kamofsky scale is recommended.

HLA antigens and bronchogenic carcinoma in the Greek popula- tion Toumbis M,Z.ervas J, Anagnostopoulou 0, Konstantopoulos K. Krim- beni G, Kotsovoulou V et al. Second Deparrmenr of Chesr Medicine.

Athens Hospiral of Chest Diseases, 152 Mesigion Av.. GR-11527

Athens. Acta Oncol 1991;30:575-8.