3 The abbreviations used are: LOH. loss of heterozygosity: SCLC. small cell lung cancer: NSCLC, non-small cell lung cancer. Vol. 2, 1185-1189, July /996 Clinical Cancer Research 1185 Loss of Heterozygosity at 3p in Non-Small Cell Lung Cancer and Its Prognostic Implication1 Tetsuya Mitsudomi,2 Tsunehiro Oyama, Kinue Nishida, Akira Ogami, Toshihiro Osaki, Kenji Sugio, Kosei Yasumoto, Keizo Sugimachi, and Adi F. Gazdar Department of Surgery II, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan [T. M.. K. Sugio, K. Sugim.l; Department of Surgery II, University of Occupational and Environmental Health, School of Medicine, Kitakyushu 807, Japan fT. M., T. Oy., K. N., A. 0.. T. Os., K. Sugio. K. Y.1; and Hammon Cancer Center and Department of Pathology. University of Texas, Southwestern Medical Center. Dallas, Texas 75235-8593 [A. F. G.] ABSTRACT We examined 1 10 patients with non-small cell lung cancer who underwent consecutive pulmonary resection for loss of heterozygosity (LOH) at the short arm of chromo- some 3 (3p). We performed a PCR-based microsatellite polymorphism analysis for detection of LOH. The microsat- ellite markers used were D3S966 (3p2l.3), D3S1007 (3p2l.3- 22), and D3S1228 (3pl4.l-l’L3). Of 98 informative cases, 3p LOH was found in 45 (46%). 3p LOH was more prevalent in squamous cell carcinoma (24/35, 69%) than in adenocarci- noma (18/52, 35%; P 0.0019). There was no significant association between 3p LOH and sex, disease stage, or grade of differentiation. However, patients with 3p LOH tended to survive for a shorter period of time (P = 0.0631, log rank test). There was no such tendency in squamous cell carci- noma (P = 0.7513), but in adenocarcinoma, the difference of survival was significant (P = 0.0015). Cox’s proportional hazards model also predicted that 3p LOH was an independ- ent poor prognostic marker in adenocarcinoma (P = 0.0502) but not in squamous cell carcinoma or in the entire cohort (P = 0.7866 and 0.1371, respectively). LOH at 3p may help to identify non-small cell lung cancer patients with a poor prognosis, who thus need an intensive postoperative fol- low-up protocol or who are suitable for novel investigational therapeutic approaches. It is also suggested that the putative tumor suppressor gene at 3p may have a different role in squamous cell carcinoma and adenocarcinoma of the lung. Received 1 1/20/95; revised 1/3 1/96; accepted 3/27/96. I This work was supported in part by Grants-in-Aid 0445435 1 and 07457300 from the Ministry of Education, Science and Culture in Japan and by grants from the Fukuoka Cancer Society and the Kaibara Morikazu Medical Science Promotion Foundation. 2 To whom requests for reprints should be addressed, at Department of Thoracic Surgery. Aichi Cancer Center Hospital. 1- I Kanokoden. Chikusa-ku, Nagoya 464. Japan. Phone: 81 (52) 762-61 1 1: Fax: 81 (52) 764-2963. INTRODUCTION Loss of genetic material at a certain chromosomal locus resulting in LOH3 is frequently seen in various types of human malignancies, and it is considered as a hallmark of a tumor suppressor gene (1, 2). In lung cancer, LOH is frequently present at many chromosornal arms including 1p, 2q, 3p, Sq. 8p, 9p, 1 lp, l3q, l7p, and l8q (3-5). Among these abnormalities, LOH at multiple loci on the short arm of chromosome 3 (3p) is the most frequent genetic lesion in lung cancer, which was first identified by Whang-Peng et a!. (6) by means of cytogenetics. Of interest, it is the earliest appearing molecular abnormality described to date in the pathogenesis of lung cancer (7, 8). Subsequent numerous studies using RFLP revealed that (a) allelic loss of 3p occurs in a significant fraction of NSCLC and in almost all SCLC (4, 5, 9-15), and (b) there are at least three distinct regions located at 3p25, 3p2l.3, and 3p14-cen corn- monly deleted in lung cancer ( 14). However, corresponding tumor suppressor genes have not been cloned. Previous studies suggested that the 3p LOH is associated with less differentiated histology or with the advanced stage in adenocarcinoma of the lung ( 15) and with a trend toward a poor prognosis (16). However, clinical implication or a prognostic impact of the 3p LOH has not been established. In this study, as a part of systematic search for molecular prognostic markers in NSCLC (17-20), we examined 1 10 patients with NSCLC for 3p LOH using PCR-based microsatellite polymorphism analysis. We correlated the finding with various clinical features includ- ing patient survival. MATERIALS AND METHODS Patients and DNA. During a 21-month period from July 1991 to April 1993, 143 consecutive patients underwent pulmo- nary resection for treatment of NSCLC at the Department of Surgery II, University of Occupational and Environmental Health as a routine clinical practice. Of them, 1 10 patients (77%) were studied. The criterion of the patients for inclusion in this study was solely availability of paired normal lung tissue and tumor materials. There were 61 adenocarcinornas, 38 squa- mous cell carcinomas, 7 large cell carcinomas, and 4 other types. Thirty-eight patients had stage I disease, 8 had stage II. 38 had stage lila, I5 had stage IlIb. and I 1 had stage IV disease. For postoperative follow-up, patients were asked to visit our clinic for examinations including chest X-ray and tumor marker determination every month for the first year, every other month for the second year, and then every 3 months for the third year and so on. Computed tomographic scanning and bone scinti- gram were performed at least once every year after the opera- Research. on November 28, 2021. © 1996 American Association for Cancer clincancerres.aacrjournals.org Downloaded from