GYNECOLOGIC ONCOLOGY ARTICLE NO.
71, 177–184 (1998)
GO985170
Establishment of a New Cell Line, OKT1, from Small Cell Carcinoma Secreting Ectopic ACTH of the Uterine Cervix Hideyuki Ohtake, M.D., Hidetaka Katabuchi, M.D.,1 Hironori Tashiro, M.D., Salma Khan, M.D., Yukitoshi Fukumatsu, M.D., and Hitoshi Okamura, M.D. Department of Obstetrics and Gynecology, Kumamoto University School of Medicine, Kumamoto City, 860-8556, Japan Received April 7, 1998
Objective. Small cell carcinoma of the uterine cervix is rare and represents a unique entity among gynecological tumors. It sometimes demonstrates neuroendocrine differentiation, including adrenocorticotropin (ACTH) secretion. In this study, we established a new cell line, OKT1, from a case of carcinoma secreting ectopic ACTH without Cushing’s syndrome and determined the character of the cell line. Methods. OKT1 was established from OKT tumor cells, derived from a biopsy specimen of small cell cervical carcinoma, and serially heterotransplanted into nude mice. To characterize OKT1, the cell morphology, growth properties, immunohistochemical properties, hormone- and tumor-associated antigen secretion, tumorigenic potential, DNA profile, and chromosomal alteration were studied. Results. The population doubling time of OKT1 was approximately 27 h. The cytological properties of OKT1, including DNA ploidy pattern, were similar to those of the primary tumor. Neuroendocrine differentiation was shown in the OKT1 cells by the positive immunocytochemical staining of neuron-specific enolase (NSE) and the presence of NSE and ACTH in the culture media. The xenograft of 1 3 108 OKT1 cells into nude mice yielded tumor mass. Furthermore, OKT1 demonstrated HPV type 18 and absence of a p53 gene mutation from exons 5 through 8. Conclusion. To our knowledge, OKT1 is the first cell line established from small cell cervical carcinoma with ACTH secretion. © 1998 Academic Press
INTRODUCTION Small cell cervical carcinoma is a rare tumor comprising 2–5% of all cervical neoplasms [1]. It usually occurs in women younger than those in whom ordinary forms of cervical carcinoma develop. The tumor widely disseminates early in its development and mostly results in a poor prognosis. In such a tumor, neuroendocrine differentiation is sometimes shown, but carcinoma secreting ectopic adrenocorticotropin (ACTH) is 1
To whom correspondence and reprint requests should be addressed at Department of Obstetrics and Gynecology, Kumamoto University School of Medicine, Honjo 1-1-1, Kumamoto City, Kumamoto 860-8556, Japan. Fax: 181-96-363-5164. E-mail:
[email protected].
extremely rare [2]. Although the disease has gained much attention because of its aggressive clinical behavior and neuroendocrine character, the pathogenesis remains a subject of controversy. For its biological characterization, establishment of a tumor cell line is essential. To the best of our knowledge, only three cell lines of small cell carcinoma originating from the uterine cervix have been established to date [3–5], and whether these are capable of producing ectopic ACTH is unclear. The novel cell line we established, OKT1, is the first small cell cervical carcinoma producing ectopic ACTH. MATERIALS AND METHODS Cell Culture The patient, a 28-year-old Japanese woman, had an International Federation of Gynecology and Obstetrics stage IIB small cell carcinoma of the uterine cervix [6]. Histological diagnosis was made according to the new terminology of endocrine tumors of the uterine cervix [7]. She underwent radical surgery and intravenous combination chemotherapy. Twenty months later, she died of multiple metastatic tumor spread throughout the body. Her laboratory data showed an elevation of serum ACTH. Although serum ACTH reached 779 pg/ml, Cushing’s syndrome did not develop. Tumor tissue was excised from the recurrent vaginal tumor of a small cell cervical carcinoma. The tumor was serially heterotransplanted into nude mice (BALB/c nu/nu) up to the 10th generation [8]. The tumor of the last generation, designated OKT, was cut into small pieces and tissue debris was removed in Hank’s balanced salt solution (Gibco, Grand Island, NY). After filtration of the tumor pieces with 70-mm nylon mesh (Yaoki, Tokyo, Japan), 7 3 106 viable cells were aseptically collected and suspended in RPMI 1640 (Gibco) and supplemented with 15 mM/L of Hepes (Nacalai Tesque, Kyoto, Japan), 10% fetal calf serum (FCS; HyClone Laboratories Inc., Logan, UT), 100 U/mL penicillin, and 100 mg/mL streptomycin (Gibco). The cells were incubated at 37°C in a humidified atmosphere of 5% CO2 and 95% air. When confluent, the cells were subcultured after detachment with trypsin– ethylenediaminetetraacetic acid. They were ini-
177
0090-8258/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
FIG. 1. Phase contrast microscopy of OKT1 cells. Four different characteristics are recognized: Type 1 (a); Type 2 (b); Type 3 (c); and Type 4 (d). Original magnification 3400.
178 OHTAKE ET AL.
NEW CELL LINE FROM CARCINOMA SECRETING ECTOPIC ACTH
FIG. 2. 3400.
179
Immunocytochemistry of OKT1 cells. NSE is immunocytochemically positive in the cytoplasm of cells. Hematoxylin stain. Original magnification
tially adhered to fibroblasts derived from nude mice and then grown to develop into remarkable piles. After 20 passages, the cells were completely free of the fibroblasts and were suspended in the medium. Nonadherent culture was passaged weekly by transfer of floating-cell aggregates (1:10 split), and aliquots of each passage were kept frozen in liquid nitrogen in RPMI 1640 containing 10% dimethyl sulfoxide (Nacalai Tesque) and 10% FCS. In Vitro Morphology Cultured cells were observed by phase contrast microscopy and were plated onto Lab-tek tissue culture chamber slides (Nunc, Naperville, IL). The cells were fixed in 95% ethanol and stained by an immunocytochemical technique. Mouse antihuman epithelial membrane antigen (EMA) monoclonal antibody (Dakopatts, Glostrup, Denmark), mouse anti-human cytokeratin monoclonal antibody gp 56 kDa (Immnunotech, Marseilles, France), mouse anti-swine vimentin monoclonal antibody (Dakopatts), mouse anti-human carcinoembryonic antigen (CEA) monoclonal antibody (Dakopatts), rabbit antihuman neuron-specific enolase (NSE) polyclonal antibody (Dakopatts), rabbit anti-cow S-100 polyclonal antibody (Dakopatts), and mouse anti-ACTH monoclonal antibody (Dakopatts) were used. Immunoperoxidase staining of EMA, cytokeratin, vimentin, CEA, and ACTH was performed by the avidin– biotin complex (ABC) method with a Vectastain ABC kit (Vector, Burlingame, CA). For the immunoperoxidase detection of NSE and S-100, the peroxidase anti-peroxidase
method was performed with PAP kit (Dako, Carpinteria, CA). Peroxidase activity was visualized with 3,39-diaminobenzidine (Sigma, St. Louis, MO) as a substrate in Tris–HCl buffer (0.5 mg/ml, pH 7.6) containing 0.01% H2O2. Nuclear staining was performed with hematoxylin in water. Appropriate positive controls were used for each immunoprotein. Negative controls were prepared by omitting the primary antibody. For transmission electron microscopy, the cultured cells were collected at each passage and fixed in chilled 0.1 M cacodylate-buffered 2.5% glutaraldehyde solution for 1 h, postfixed with 1% osmium tetroxide dissolved in 0.1 M cacodylate buffer, dehydrated through a graded ethanol series, and then embedded in Epon 812 (Ohken, Tokyo, Japan). Ultrathin sections were made using an Ultratome MT 6000-XL (RMC, Tucson, AZ) with a diamond knife (Diatome; Bienne, Switzerland). They were stained with lead citrate and uranyl acetate and observed under a 12-A electron microscope (Hitachi, Tokyo, Japan). Hormone and Tumor Marker Assays ACTH in cultured medium was analyzed by two-site immunoradiometric assay (IRMA). NSE, CEA, and squamous cell carcinoma related antigen (SCC-Ag) were analyzed in cultured medium by radioimmuno assay (RIA). Each assay was performed using an Allegro ACTH Kit (Nichols, CA), Eiken “NSE” (Eiken, Tokyo, Japan), CEA z RIABEAD (Dinabotto, Tokyo, Japan), and SCC z RIABEAD (Dinabotto), respectively. The level of these substances in culture medium was
180
OHTAKE ET AL.
FIG. 3. Electron micrograph of OKT1 cells. The tumor cells showed a high nucleo– cytoplasmic ratio and the intracellular junctions were infrequent (a). Round dense-core granules with limiting membrane were dispersed in the cytoplasm (b). N, nucleus; Mit, mitochondria. Lead citrate and uranyl acetate stain. Scale bar, 1 mm (a); 0.5 mm (b).
measured in the growth medium following a 5-day culture of 1 3 105 cells. In Vitro Growth Kinetics The growth curve and population doubling time were studied in the cultured cells. Approximately 1 3 105 cells were plated in quadruplicate in 24 multiwell plates (Flow, McLean, VA) and incubated under standard culture conditions as described above. A cell count was taken initially on the second day after culture began and successively at intervals of 12 to 24 h. The total number of viable cells was counted by the trypan blue exclusion test, the values were plotted on a graph, and population doubling time was calculated from the resulting curve. Hetrotransplantation The cultured cells at the 20th passage were washed twice and suspended in 1.0 ml RPMI 1640. Then, 1 3 108 cells were inoculated subcutaneously into the back of female athymic BALB/c nude (nu/nu) mice, aged 5 weeks. Tumor tissues were aseptically extirpated 8 weeks after transplantation and submitted to routine histopathological examination.
Flow Cytometric DNA Analysis After 20 passages the cells were cultured to subconfluence in 25-cm2 flasks. After trypsinization, cells were washed twice in cold phosphate-buffered saline permeated with Triton X-100 (0.5%) and ribonuclease (50 mg/ml) and stained with propidium iodide (10 mg/ml). Analysis was carried out on an Epics Elite cytofluorimeter (FACSscan; Becton Dickinson, CA). A comparison of flow cytometry was made among the cultured cells, the OKT tumor, and the paraffin-embedded primary tumor. The paraffin-embedded specimens were dewaxed, dehydrated, and treated with pepsin at pH 1.7. Propidium iodine was used for staining. Human peripheral lymphocytes from a healthy donor were used as a calibration standard for DNA diploidy with a coeffication of variation of 1.8. The histograms were evaluated for ploidy. Chromosome Analysis Cells at the 25th passage were submitted for routine karyotypic and cytogenetic analysis of Giemsa-based metaphases. Cells were cultured in complete tissue culture medium for 8 h
NEW CELL LINE FROM CARCINOMA SECRETING ECTOPIC ACTH
181
in Colcemid (0.01 mg/ml) prior to conventional Giemsa staining. Human Papillomavirus (HPV) Typing A pair of consensus primers of L1 resions, L1C1 and L1C2, described by Yoshikawa et al. [9], were used. Total cellular DNA from cultured cells at the 25th passage was extracted according to the standard SDS–proteinase K procedure. The polymerase chain reaction protocol employed in this study was 40 cycles of denaturation (95°C, 1.5 min), annealing (48°C, 1.5 min), and extension (70°C, 2 min). Each reaction mixture (100 ml) contained 50 mM KCl, 10 mM Tris–Cl, pH 8.4, 1.5 mM MgCl2, 200 mM of each dNTP, 1 mM of each primer, and 4 units of Taq polymerase (Biotech International Ltd., Bentley, WA). After the reaction, one-tenth (10 ml) of the reaction mixture was electrophoresed through 4% NuSieve GTG agarose gel (Takara Shuzo Co. Ltd., Kyoto, Japan) containing 1 mg/ml ethidium bromide, the gel was photographed, and furthermore one-fourth (25 ml) was used for digestion with the restriction enzymes. The amplified HPV fragments could be typed on the basis of the restriction fragment length polymorphisms. Restriction maps of the currently available enzymes were surveyed by Mapsort (Sequence analysis software package of the Genetics Computer Group, Madison, WI) and the restriction enzymes RsaI, DdeI, and HaeIII were chosen for typing. The HPV types were confirmed by using at least two additional restriction enzymes. As a positive control, SiHa cells, which demonstrated HPV16 DNA integration, were used [10]. p53 Gene Sequence Analysis The DNA of the cells extracted at the 25th passage was examined for mutations in exons 5 through 8 of the p53 gene using direct DNA sequence analysis as previously described [11]. RESULTS Successful sequential passages of OKT tumor were obtained after the 20th passage through over 100 transfergenerations without mouse fibroblasts. Light microscopic observation of culture cells showed no contamination of other cells. The newly established cell line was designated OKT1. The cell line was subgrouped into four types based on morphologic appearance. First-typed cells (Fig. 1a) grew as tightly packed spherical aggregates of floating cells. Second-typed cells (Fig. 1b) grew as relatively dense floating aggregates, showing an amorphous or irregular outline. Third-typed cells (Fig. 1c) grew as very loosely adherent floating aggregates. Fourth-typed cells (Fig. 1d) attached to the culture dish. Each cell type consisted of small overlapping polygonal cells devoid of an epithelioid appearance. NSE was positive immunocytochemically for nearly all cul-
FIG. 4. Representative in vitro growth curve for OKT1 cells. The initial cell inoculum was 1 3 105 cells in RPMI 1640 with 10% FCS. The medium was not replenished during growth.
tured cells (Fig. 2). Immunostain for S-100, CEA, cytokeratin, EMA, and vimentin were all negative. ACTH-positive cells were not detected. However, ACTH and NSE were detected in the culture medium of OKT1 at a concentration of 7.4 ng/ml (value of control, 2.4 ng/ml) and 30 pg/ml (value of control, ,5 pg/ml), respectively. CEA and SCC-Ag were not identified in the culture medium. Ultrastructurally, the cultured cells were tightly packed and had a maximum diameter of 8 –12 mm (Fig. 3a). Their nuclei contained moderate quantities of heterochromatin and occasional nucleoli. The junctional complex was obscure among the cells. The intracellular organelles were poorly developed, but some of the tumor cells were filled with mitochondria, Golgi apparatus, polyribosomes, or small aggregates of filaments. These cells had small numbers of membrane-bound dense-core granules, 120 to 180 nm in diameter (Fig. 3b). A typical in vitro growth curve for OKT1 shows a population doubling time of 27 h during exponential cell growth (Fig. 4). When subcultured from the plateau phase, a lag phase of 96 h was observed. On the other hand, subcutaneous inoculation of 1 3 108 cells results in tumor with a latent period of approximately 8 weeks until the first appearance. Tumor grew progressively as expansive lesion at the site of inoculation and reached a size of 2 3 1 3 1 cm. The histopathological findings were found to be similar to those of the OKT tumor and the original cervical tumor obtained from the patient (Fig. 5). Flow cytometric DNA analysis revealed diploidy in all three specimens, OKT1 cells, OKT tumor, and primary tumor. OKT1 cells had various chromosomal aberrations. Chromosomal analysis revealed a model chromosomal number of 47 (range 45 to 93). Chromosomal changes, including 47, XX, 22, del(2) (q33), 23, add (3) (p21) or del (3) (p24), 25, add (7) (p13–22), del (7) (q22) or t (7; 7) (p22; q22), 210, der (12) add (12) (p13) add (12) (q22–24), 213, add (15) (q24 –26), 217, 120, 16mar were evident in most of the cells (Fig. 6). PCR-RFLPs showed representative confirmatory analysis of OKT1 cells typed as HPV 18. p53 sequence analysis in OKT1 revealed a wild-type p53 gene in exons 5 through 8.
182
OHTAKE ET AL.
FIG. 5. Histology of tumor hetrotransplantated into nude mice (a) revealed small cell carcinoma and resembled the 10th passage OKT tumor (b) that is a derivation of OKT1. Hematoxylin and eosin stain. Original magnification, 3100.
DISCUSSION Small cell carcinoma is an uncommon but aggressive tumor of the uterine cervix associated with rapid progression of disease and a poor prognosis. While clinicopathological studies of this tumor in several series have been reported in recent years [12–14], little information is available not only on effective clinical treatments for but also on the biological properties
of this carcinoma. Therefore, an experimental model is needed. Until now, only three cell lines have been established from small cell carcinoma of the uterine cervix. These are TC-YIK established by Ichimura et al. [3], and SNU-487 and SNU1245 established by Ku et al. [5]. We established here a new cell line, OKT1, derived from a uterine cervical small cell carcinoma with ACTH secretion [8]. First, it was confirmed that the cell line retained the characteristics of the original
NEW CELL LINE FROM CARCINOMA SECRETING ECTOPIC ACTH
FIG. 6.
183
Karyotype of in vitro passage 25 OKT1 cells.
tumor by immunocytochemical and ultrastructural studies and flowcytometric DNA analysis. In addition, we revealed that OKT1 possessed the character of neuroendocrine cells by an immunocytochemical technique and assay using the culture medium. Thus, OKT1 was shown to be different from the other three cell lines in that it produces ectopic ACTH. It is known that this tumor is particularly analogous to small cell carcinoma of the lung in both morphologic and clinical features. The morphological character of OKT1 also markedly resembled that of the cell lines derived from small cell lung carcinoma (SCLC) [15]. In SCLC, studies investigating the potential role of ACTH as a tumor marker have indicated that 20 –30% of patients have elevated plasma levels of ACTH, whereas Cushing’s syndrome is a relatively unusual clinical manifestation of SCLC, occurring in only 2–3% of cases [16]. As these results could be common in small cell cervical carcinoma, it is not surprising that this case had elevated ACTH without Cushing’s syndrome. Moreover, in this study immunocytochemical staining of the cultured cells was negative for ACTH, although the ACTH in the serum of the patient and the culture medium of the OKT1 cells was detected by IRMA. In the lung or vagina, it has been reported that negative staining for ACTH is characteristic for small cell carcinoma with ACTH secretion [17, 18]. The absence of ACTH immunoreactivity may be related to several factors, including limited intracellular storage of ACTH, posttranslational processing of the proopiomelanocortin precursor to peptides other than ACTH [18], and the relatively low level of ACTH synthesis (per each cell) when compared to pituitary or carcinoid tumors. The association between squamous cell carcinoma and adenocarcinoma of the cervix and HPV is well known. Interestingly, HPV has been demonstrated in small cell carcinoma of
the uterine cervix in as many as 85% of patients [13]. Small cell carcinoma was reported to be strongly associated with HPV-18 [12, 13]. OKT1 in this study, SNU-487, and SNU1245, which all have a similar phenotype of floating cells, contained HPV-18 DNA. HPV-positive cervical carcinoma is less frequently associated with mutation of the p53 gene [19], whereas the mutation is most common in many kinds of carcinoma including SCLC [20]. It is thought that p53 mutation could not be required in carcinogenesis of HPV-positive tumors because the HPV oncoprotein, E6, was shown to bind to stimulate the degradation of the p53 protein [19]. In this study, OKT1 bearing HPV type 18 did not show any p53 mutation in exons 5 through 8, similar to many HPV-positive cervical carcinomas. A HPV-18 integrated cell line derived from small cell carcinoma with ACTH secretion of the uterine cervix has, to our knowledge, never before been reported. OKT1 will add to the model available for the study of the pathogenesis of small cell carcinoma of the uterine cervix and, hopefully, to the establishment of a clinical treatment. ACKNOWLEDGMENTS PCR-RFLPs methods of HPV typing were reliable for SRL Corporation. We gratefully acknowledge Dr. Lora Hedrick Ellenson and Dr. Sigard Lax, Department of Pathology, Johns Hopkins University School of Medicine, for p53 sequence analysis.
REFERENCES 1. Kurman RJ, Norris HJ, Wilkinson EJ: Tumors of the cervix. In Rosai J, Sobin LH (eds): Atlas of Tumor Pathology. Third Series Fascicle 4. Tumor
184
2.
3.
4.
5.
6.
7.
8.
9.
OHTAKE ET AL.
of the Cervix, Vagina, and Vulva. Washington, D.C., Armed Forces Institute of Pathology, 1992, pp 104 –106 Hashi A, Yasumizu T, Yoda I, Kou T, Mizuno K, Hirata S, Kato J, Katoh R, Inoue M, Kawaguchi A, Nakazato M, Onaya T: A case of small cell carcinoma of the uterine cervix presenting Cushing’s syndrome. Gynecol Oncol 61:427– 431, 1996 Ichimura H, Yamasaki M, Tamura I, Katsumoto T, Sawada M, Kurimura O, Furuyama J, Kurimura T: Establishment and characterization of a new cell line TC-YIK originating from argyrophil small cell carcinoma of the uterine cervix integrating HPV16 DNA. Cancer 67:2327–2332, 1991 Ichimura H, Yamasaki M, Yamane T, Shimizu H, Katsumoto T, Kurimura O, Kurimura T: Heterotransplantation of argyrophil small cell carcinoma of the uterine cervix integrating HPV16 DNA into nude mice. Jpn J Cancer Res 79:1255–1258, 1988 Ku JL, Kim WH, Park HS, Kang SB, Park JG: Establishment and characterization of 12 uterine cervical-carcinoma cell lines: common sequence variation in the E7 gene of HPV-16-positive cell lines. Int J Cancer 72:313–320, 1997 Suenaga Y, Katabuchi H, Mizutani H, Tashiro H, Fukumatsu Y, Ito M, Okamura H: A case report of cervical small cell carcinoma with an elevation of the serum adrenocorticotropin hormone. Acta Obstet Gynaecol Jpn 45:589 –592, 1993 Albores-Saavedra J, Gersell D, Gilks CB, Henson DE, Lindberg G, Santiago H, Scully RE, Silva E, Sobin LH, Tavassoli FJ, Travis WD, Woodruff JM: Terminology of endocrine tumors of the uterine cervix: results of a workshop sponsored by the College of American Pathologists and the National Cancer Institute. Arch Pathol Lab Med 121:34 –39, 1997 Katabuchi H, Tashiro H, Suenaga Y, Ohtake H, Ohshige A, Mizutani H, Fukumatsu Y, Okamura H: Small cell carcinoma of the uterine cervix can be differentiated from squamous cell carcinoma; morphologic alterations in xenografted small cell carcinoma. Med Electron Microsc 30:202–208, 1997 Yoshikawa H, Kawana T, Kitagawa K, Mizuno M, Yoshikura H, Iwamoto A: Detection and typing of multiple genital human papillomaviruses by DNA amplification with consensus primers. Jpn J Cancer Res 82:524 – 531, 1991
10. O’Leary JJ, Browne G, Johnson MI, Landers RJ, Crowley M, Healy I, Street JT, Pollock AM, Lewis FA, Andrew A, Cullinane C, Mohamdee O, Kealy WF, Hogan J, Doyle CT: PCR in situ hybridisation detection of HPV 16 in fixed CaSki and fixed SiHa cell lines. J Clin Pathol 47:933– 938, 1994 11. Tashiro H, Isacson C, Levine R, Kurman RJ, Cho KR, Hedrick L: p53 gene mutations are common in uterine serous carcinoma and occur early in their pathogenesis. Am J Pathol 150:177–185, 1997 12. Abeler VM, Holm R, Nesland JM, Kjorstad KE: Small cell carcinoma of the cervix. A clinicopathologic study of 26 patients. Cancer 73:672– 677, 1994 13. Stoler MH, Mills SE, Gersell DJ, Walker AN: Small-cell neuroendocrine carcinoma of the cervix. A human papillomavirus type 18-associated cancer. Am J Surg Pathol 15:28 –32, 1991 14. van Nagell JR, Jr, Powell DE, Gallion HH, Elliott DG, Donaldson ES, Carpenter AE, Higgins RV, Kryscio R, Pavlik EJ: Small cell carcinoma of the uterine cervix. Cancer 62:1586 –1593, 1988 15. Carney DN, Gazdar AF, Bepler G, Guccion JG, Marangos PJ, Moody TW, Zweig MH, Minna JD: Establishment and identification of small cell lung cancer cell lines having classic and variant features. Cancer Res 45:2913– 2923, 1985 16. Ratcliffe JG, Podmore J, Stack BH, Spilg WG, Gropp C: Circulating ACTH and related peptides in lung cancer. Br J Cancer 45:230 –236, 1982 17. Colleran KM, Burge MR, Crooks LA, Dorin RI: Small cell carcinoma of the vagina causing Cushing’s syndrome by ectopic production and secretion of ACTH: a case report. Gynecol Oncol 65:526 –529, 1997 18. Coates PJ, Doniach I, Howlett TA, Rees LH, Besser GM: Immunocytochemical study of 18 tumours causing ectopic Cushing’s syndrome. J Clin Pathol 39:955–960, 1986 19. Paquette RL, Lee YY, Wilczynski SP, Karmakar A, Kizaki M, Miller CW, Koeffler HP: Mutations of p53 and human papillomavirus infection in cervical carcinoma. Cancer 72:1272–1280, 1993 20. Lohmann D, Putz B, Reich U, Bohm J, Prauer H, Hofler H: Mutational spectrum of the p53 gene in human small-cell lung cancer and relationship to clinicopathological data. Am J Pathol 142:907–915, 1993