Abstracts /Lung Cancer 12 (1995) 265-329

A phase II study testing weekly platinum derivative combination chemotherapy as second-line treatment in patients with advanced small cell lung cancer Sculier J.P. Thiriaux J Bureau G. Latitte J.J. Recloux P Brohee D et al. Insfifvt Jules Bordel, Service de Medecine. 1 Rue Heger-Border. B-1000 Brussels. Int J Oncol 1995;6:425-9.

A phase IJ trial was conducted to determine the effectiveness of weekly administration of cisplatin (25 mpim’ on day I) and carboplatin (100 mgim’ on day 1) as salvage chemotherapy for patients with small cell lung cancer aller first-line chemotherapy without platinum derivatives. Of40 eligible patients, 38 were evaluablc forresponse. Interval behveen last course of Iirst-line chemotherapy and first course of salvage therapy was less than 3 months in 34 and greater in 4. Five partial responses (13%; confidence interval at 95%:0.01-0.25) were documented (including 4 in patients with a treatment-free interval 0 months) as well as 8 no change. 2 I progressions and 4 early deaths due to malignant dweaw. Toxicity consisted mainly of moderate thrombopcnia and Icuwpenia. Grade I ncphrotoxicity was observed in 6 patients. In conclusion, weekly administration of moderate doses of cisplattn and carboplatin as salvage chemotherapy for small cell lung cancer appeared feasible and was associated with a moderate but definitive. anticancer activity.

Lung cancer pathology in smokers, es-smokers and never smokers Muscat JE, Wynder EL. Division ofEpidemio/ogv. American He&h Foundation. 3.20 East 43rdSbeet. New York, NY 10017. Cancer Letter 1995;88:1-5.

The histologic distribution of lung cancer is markedly different in smokers and non-smokers. It is not known whether the histology among former smokers varies according to the number of years since quitting. Using data from a large

casecontrol study of lung cancer, we found that for both men and women, the proportion of adenoenrcinomas increased with the number of years since quitting smoking. Among long-term quitters (>2S years), the proportion of adcnocarcinomas was similar to that in never smokers. These findings have implications for studies of environmental tobacco smoke and lung cancer.

Immunostaining in the diagnosis of pulmonary neuroendocrine carcinomas: An immunobistochemical study with ultrastructural correlations Lay TS, Darkow GVD, Quesenbeny JT. Depurbnen~ of Path&~, Missouri

Universiry Medical Cen~n: One Hospital Drive. Columbia. MO 65212. Am J Surg Path01 1995;19:173-82.

To determine the speciftcity of immunostaining in the diagnosis of pulmonary ncurwmdocrinc carcinomas, we studied 66 ultrastructurally characterized lung cancers with a panel of markers considered to lx selective for neuroendocrine tumors (neuron-spsific enolasc, chromogranin A, Leo 7, synaptophysin) and B72.3, which is reported to be selective for non-small- cell carcinomas. Ncuroendocrinc tumors studied included 13 small-cell carcinomas, four low- grade ncurocndocrine carcinomas, and two large-cell carcinomas with neuroendocrine differentiation. Non-neurocndocrine tumors included 26

adenocarcinomas, 10 squamous cell carcinomas, and 1 I large-eel1 undifferentiated carcinomas. The following percentages of neuroendocrine carcinomas showing immunoreactivity for the various ‘neurocndocrinc markers’ were found: synaptophysin, 100%; neuron-specific cnolase, 74%; chromogranin A. 37%; and Leo 7, 5%. However, carcinomas without morphologic features of ncuroendocrinc differentiation showed the following immunorcactivity: synaptophysin, 62%; neuron-specific enolase, 60%; chmmogranin A, 17%; and Leu 7, 9%. B72.3 immunostaining WBS seen in 81% of the carcinomas without ncuroendocrine features and in 3 I % of the small-sell carcinomas. We conclude that many of the commercially available antibodies used as ncurocndocrine markers are nonspecific in the diagnosis ofpulmonary ncuroendocrine carcinomas.

Prognostic significance of an interface pattent of central tibmsis and tumor cells in peripheral adetmcarcinoma of the lung Yamashiro K, Yasuda S, Nagax A, Hirata T, Nojima T , Nagashima K. Divasion o/ ClinicalPalolo~, DEparrmenloJClinicolResea~h, Sappcwo National Hospital. 4-2. Ktkusui, Shiroishi-ku. Sapporo 003. Hum Pathol 1995;26:67-73.

Fitly-nine surgically resected pulmonary adenoearcinomas were histologically classified into four types (A, B, C, and D) according to the pattern of invasion,

ic, the extent of invasive tumor cell growth in the interface zone behveen alveoli replaced by cancer cells and the central region of tibrosis (so called ‘scar’). In patterns A cancer cells proliferate along the alveolar walls without forming a frankly invasive lesion. In pattern B invasive lesions occupy less than 30% and in pattern C more than 30% of the Iibrosing area. Pattern D refers to Invasion of the bronchial lumen by cancer cells. The 59 tumors included 14 of pattern A, 12 of pattern B, 16 of pattern C, and I7 of pattern D. The 5-year survival rate for patients with pattern A tumors was loo?/., and the rates for patients with pattern B, C, and D tumors were 64.2%. 30.0%, and ll.8%, respectively. Each difference behveen pairs of survival curves was statistically signilicant (P < .OS). We found a correlation between the pattern of invasion and other prognostic factors. However, even in the cases evaluated as favorable by other prognostic factors (stage I, tumor less than 35 m m in diameter, negative for lymph node m&stases, wclldi&rentiated histology, negative for subpleural invasion, negative for vascular invasion) the survival curves boxme steeper going from pattern A to patterns B, C, and D. We conclude that the pattern of invasion is correlated with the prognosis of surgically treated patients. Our study may provide new histological criteria for the prognostic evaluation of pulmonary adcnocarcinoma.

Ate pathologists biased by clinical information? A blinded cross-over study ofthe histopathological diagnosis of mesothelial turnout-s versus pulmmay adenocarcinoma Skov BG, Braendstrup 0, Hirsch FR. Lauritzen AF, Nielsen HW. Skov T . Bronshoj Kirkevej 49. DK-2700 Bronshoj. Lung Cancer (Ireland) 1994.1 I’ 365- 72.

In a blinded cross-over design. we studied whether three pathologists were biased by clinical information when makmg histopathologtcal diagnoses of adcnocarcinoma of the lung and benign and malignant mesothelial turnours. Furthermore, the interobserver variation of these diagnoses was assessed. Forty- one casts of adenocarcinoma of the lung and mesothelial turnours were assessed by three pathologists in four rounds In the first hvo rounds. slides stained by H and E and clinical information were available. Slides and information were matched so that a specific slide in one round was given crucial information suggesting adcnocaminoma and in the other round, the clintcal information suggested mcsothelial tumour. In the third and fourth rounds, a panel of immunohistochemxal stains was added The clinical Information was matched in the same way as in the first and second rounds. Bias by clinical informatlon was observed when the diagnoses were made on slides stained by H and E. while no bias could be demonstrated when immunohistochemical reactions were included. The reproducibility also improved significantly when these shdes were available.

The value ofneuroendocrine markers in non-small cell lung cancers A comparative immunohistopatbok@c study Kiriekogiani-Psoropoulou P, Malamou-Mitsi V, Martinopoulou U, Lcgaki S, Tamvakis N. Vrettou E et al. Department o/Medicine. School ojMedrcine. Universityojloannina. Ioannina 45 IIO. LungCancer@reland) 1994;11- 353-54

In order to estimate the value of immunohistochemical identification of ncuroendocrine (NE) differentiation markers in non-small cell lung carcinomas (NSCLCs), we investigated the expression of five neuroendocrine and neural differentiation-related antigens in 51 NSCLCs. Adddonally, 20 cpithclial lung honors with NE differentiation [ 15 carcinoids and five small cell lung carcinomas (SCLCs)] and 61 epithelial tumors of various other origin (breast, prostate. colon and head-neck carcinomas) were studied. An indirect two-stage immunoperoxidasc method was performed in formalin-fixed and parstlin- embedded tissue specimens, by using commercially available monoclonal antibodies. These antibodies are directed against neuronspec~fic enolase (NSE), chromogranin-A and Leo-7 which are general markers of NE differentiation, bombesin, which is a specific NE secretory product and neurofilament tnplet protein (NFTP), an intermediate filament protein of neuronal differentiation. All live markers demonstrated a positive lmmunoreactivity m NSCLCs. equally distributed to all three histologic subtypes, ranging from I6 to 47% of the cases (NSE 47%, bomb&n 21.5%, Leo-7 21.5 k. chromogranm-A 18% and NFTP 16%). Most of the carcinoids and SCLCs expressed multiple or all NE markers. The other four epithcllal tumors showed a positwe immunoreactivity for born&m, Leu-7 and NFTP, ranging from 1 I to 409’0 of the cases Chromogranm- A was not expressed in any of these tumors, whereas NSE was demonstrated