K-ras genotypes and prognosis in non-small-cell lung cancer

Annals of Oncology 6 (SuppI 3): S15-S20, 1995. C 1995 Kluwer Academic Publishers. Printed in the Netherlands.

Symposium article K-ras genotypes and prognosis in non-small-cell lung cancer R. Rosell,1 M. Monzo,2 F. Molina,1 E. Martinez,2 A. Pifarre,21. Moreno,1 J. L. Mate,3 J. M de Ante,2 M. Sanchez2 & A. Font1 1

Medical Oncology Service, 2 Laboratory of Molecular Biology of Cancer, 3 Department of Pathology, University Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain

Background: Despite major advances in the treatment of many kinds of cancer over the past 25 years, the overall 5-year survival of non-small-cell lung cancer patients has scarcely improved. Even in stage I which has the best outcome long-term survival still falls below 70%. Since intriguing data suggest that the identification of genetic markers might allow prognosis to be assessed case by case. We were prompted to evaluate K-ras gene mutations as a putative prognostic marker in this neoplasm. Materials and methods: We used the polymerase chain reaction (PCR) followed by allele specific oligonucleotide (ASO) hybridization or single-strand conformation polymorphism (SSCP) assays, to detect K-ras mutations in DNA from formalin-fixed, paraffin-embedded tumor samples. K-ras mutations were examined in 192 stage I to IV non-small-cell lung cancer patients. Results: K-ras mutations were detected in 51 of 192 of the cases studied (27%). All K-ras mutations detected by PCR/ ASO hybridization were also identified by SSCP. In stage I

disease, the median survival time was 46 months in those patients whose tumors had no K-ras mutations and 21 months in those with aspartic acid and serine mutations at K-ras codon 12; in patients with stage IIIA disease, median survival time was 16 months in the K-ras negative group and 7 months in the aspartic acid and serine mutation group. No significant differences were observed for the remaining amino acid substitutions of K-ras, nor were they observed at all in more advanced disease. Conclusions: K-ras gene status has strong prognostic value in patients with stage IHA non-small-cell lung cancer. The survival curve for patients with stage I and K-ras codon 12 aspartic or serine mutations is close to that of patients with stage IIIA without K-ras mutations. However, a nonsmall-cell lung cancer K-ras genotypic classification should be validated in larger studies.

Key words: K-ras genotypes, allele specific oligonucleotide (ASO) hybridization, single-strand conformation polymorphism (SSCP)

mutations are more oncogenic than other ras mutations such as cysteine 12 and valine 12 [2]. Hitherto, the Members of the ras gene family are highly conserved in presence of a K-ras mutation has been considered a nature, being found in species as divergent as insect, potential prognostic marker. Reports suggest that fungi, and mammals [1]. Ras genes encode proteins that tumors that acquire a K-ras point mutation in vitro may bind guanine nucleotides, exhibit a slow intrinsic become resistant to radiation and to several drugs, inGTPase activity and are localized at the inner face of cluding cisplatin [3]. The poorer survival of those the cytoplasmic membrane. Ras oncogenes are the patients with this genomic perturbation prompt us to most frequently detected oncogenes in human and speculate that this biological marker may independentrodent tumors [1]. The highly mutable sequences in- ly contribute to drug resistance. Despite the high incidence and clinical relevance of clude those at codons 12, 13 and 59-61, and are typinon-small-cell lung cancer, relatively little information cally activated by single point mutation. Diverse amino acid substitutions at codon 12 acti- is available detailing the assortment of genomic pervate the K-ras gene, presumably by reducing the rate of turbations involved in this disease. Currently, determinGTP hydrolysis to GDP and inhibiting the physiologi- ing prognosis and selecting patients for therapy rely cal deactivation of ras proteins. This state of ras mutat- mainly on clinical and pathological staging. Patients ed protein might result in continuous signal transduc- with TNM stage I cancer (T1N0) usually have acceptable life span, on the other hand, patients with stage tion by K-ras and unregulated cell growth. Point mutations that encode any amino acid but pro- 111B and IV have curtailed survival. However, predictline at codon 12 are transforming [1]. Furthermore, in ing outcome in patients with intermediate stages is both in vitro and in vivo transfection and tumorigen- rather difficult. Patients with clinically resectable stage icity assays, it has been demonstrated that aspartic 12 niA non-small-cell lung cancer have a five-year surIntroduction

Downloaded from http://annonc.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on June 15, 2015

Summary

16 minutes to inactivate proteinase K, and used for DNA amplification. Integrity of extracted DNA was analyzed in 0.8% agarose TBE gel. PCR/allele specific oligodeoxvnucleotide (ASO) hybridization One to three uJ of DNA sample were used as a template and amplified in 10 mM Tris HC1 (pH 9.0 at 25 "C), 50 mM KCI, 0.1% Triton X-100,1.5 mM Mg Cl2, 200 uM of each dNTP and 0.5nmol of each amplimer in a 50 |il volume. Two pairs of primers were used to amplify K-ras exon 1 (KR1U 5' GGCCTGCTGAAAATGACTGA 3' and KR1D-1 5' TGATTCTGAATTAGCTGTAT 3') to detect K-ras mutations at codon 12 and 13, and K-ras exon 2 (KR2U 5' ACCTGTCTCTTGGATATTCT 3' and KR2D 5' TGATTTTGATTATTATATGC 3') to detect K-ras mutations at codon 61. DNA in the PCR mix was denatured at 94 *C for 1 minute, amplified for 35 cycles at the following temperatures: 94 "C for 20 seconds, 56 "C for 20 seconds and 72 "C for 20 seconds in a 9600 Perkin-Elmer Cetus DNA thermal cycler (Perkin-Elmer Cetus, U.S.A.). After completion of PCR 5 |il of the reaction was analyzed in 2% agarose-TBE gel. For ASO hybridization, PCR samples were denatured at 95 "C for 5 minutes, and 10 ul were dotted onto a 5 x SSC (1 x SSC is 150 mM NaCl, 15 mM Na citrate, pH 7.6) pretreated dried nylon membrane (Hybond, Amersham, U.K.). Membranes were dried and the DNA was cross-linked by exposure to UV radiation for 2 minutes (UV cross-linking, Stratagene, CA, U.S.A.). Each membrane was hybridized with a radiolabelled oligodeoxynucleotide specific for each K-ras point mutation in 5 ml of hybridization solution (5 x SSPE, 5 x Denhardt's solution, 0.5% SDS, 100 mM sodium pyrophosphate) at 42 'C for 5 hours. K-ras 12 oligodeoxynucleotide set was purchased from Clontech (U.S.A.). Four pmol of each oligodeoxynucleotide was 5' terminally labelled at 37 *C for 30 minutes with 10 nCi (3.3 pmol) of alfa-32P-dATP, 70 mM Tris HCI pH 7.6, 10 mM MgCl2, 5 mM DTT and 1 u of T4 polynucleotide kinase (Promega, U.S.A.) in a 25 u.1 volume. Hybridization was performed with 0.1 pmol of labelled oligodeoxynucleotide. Membranes were then washed with 5 x SSC at room temperaTable 1. Patient characteristics.

Materials and methods

No K-ras mutations

Patients One hundred and ninety-six patients with biopsy-proven non-smallcell lung cancer entered this study (stages I to IV, according to the international TNM criteria [18]). All patients included were studied between January 1988 and October 1992. Tumor specimens were obtained by means of either fiberoptic bronchoscopy or surgery. A portion of each tumor was obtained in the Pathology Department and appropriate adjacent samples were reviewed histologically. The patients' main characteristics studied are summarized in Table 1. Seventy-two patients stage I1IB and IV received carboplatin plus etoposide as chemotherapeutic regimen, complete details have been published elsewhere [15]. DNA isolation Three to five 5 nm paraffin embedded tumor sections were cut depending on the amount of tissue available and collected in an eppendorf tube. One ml of xylene was added to remove the paraffin. Samples were then centrifuged at 10.000 g for 5 minutes and pellets rinsed with ethanol. After centrifugation 400 u.1 of lysis buffer (10 mM Tris HC1 pH 8.5, 2.5 mM Mg Cl2, 50 mM KC1, 0.5% Tween 20 containing 250 \ig/m\ of proteinase K) was added, the tubes were incubated at 56 'C for 24 hours with regular shaking to extract the DNA and centrifuged at 12.000 g for 20 minutes to clean up the samples. Superaatants were collected, incubated at 95 "C for 8

No. of patients 141 (73%) Age (years) Below 50 11 (8%) 50 to 59 37 (26%) 60 to 69 67 (48%) 70 or more 26 (18%) Sex Male 137 (97%) Female 4 (3%) Performance status (KPS) 60% 3 (2%) 70% 23 (16%) 80% 40 (29%) 90% 75 (53%) Histological type Squamous cell 81 (75%) Adenocarcinoma 41 (71%) 19(73%) Large cell Stage I 29(21%) 11 (8%) II IHA 54 (38%) IIIB and IV 47 (33%)

Certain K-ras mutations

K-ras codon 12 Asp and Ser mutations

36 (19%)

15(8%)

9 6 16 5

(25%) (16%) (44%) (14%)

2(13%) 5 (38%) 6 (46%) 2(13%)

34 (94%) 2 (6%)

14 (93%) 1 (8%)

0 4(11%) 15 (42%) 17(47%)

0 3 (23%) 6 (46%) 6 (46%)

20 (19%) 13(22%) 3 (12%)

7 (6%) 4 (7%) 4(15%)

5 (14%) 2(6%) 13(36%) 16(44%)

2(13%) 0 5 (33%) 8 (54%)

Downloaded from http://annonc.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on June 15, 2015

vival rate of about 15% or higher when preoperative chemotherapy is given [4, 5]. Better means of formulating the prognosis in patients with non-small-cell lung cancer would improve patient selection for chemoradiotherapy approaches in the stage ULA. setting. Well-established features of ras mutations in nonsmall-cell lung cancer include the fact that: 1) approximately 30% of lung adenocarcinomas contain ras mutations, primarily in K-ras [6]; 2) ras mutations are found in 36% of non-small-cell lung cancer cell lines, including adenocarcinomas (28%) and other nonsmall-cell lung cancer histologic types (45%) [7]; 3) ras mutations portend a poor prognosis in lung adenocarcinoma [8]. Notwithstanding, it has been confusingly postulated that: 1) K-ras mutations have a poor prognosis irrespectively of the histologic type [9]; 2) K-ras mutations have prognostic value only in resected stage I (NO) lung adenocarcinoma 10]; 3) in advanced nonsmall-cell lung cancer only K-ras 12 mutants have prognostic value [11]; 4) K-ras mutations occur almost exclusively in a third of adenocarcinomas [8, 12,13]; 5) K-ras mutations occur independently of the histologic type [7, 9, 14, 15]; and 6) in squamous cell lung carcinomas a higher incidence of k-ras mutations has been found than in adenocarcinoma biopsies [16]. The present study comprises 192 non-small-cell lung cancer patients who were studied to develop a practical molecular genetic test for assessing prognosis by means of formalin-fixed paraffin-embedded sections as a source of DNA. Ras mutations were analyzed by polymerase chain reaction (PCR) followed by allele specific oligonucleotide (ASO) hybridization [17] or by single-strand conformation polymorphism (SSCP) [14].

17 ture and subsequently with 5 x SSPE (1 x SSPE is 180 mM NaH2PO4, 1 mM EDTA and 0.1% SDS for 20 minutes at 1-2 *C below the melting point (Tin) of the oligodeoxynucleotide. Specific hybridization was detected by exposing the membranes to autoradiographic film (Kodak OAR, U.SA.) using intensifying screens at -80'C. PCR/single-strand conformation polymorphism (SSCP)

Fig. 1. Detection of K-ras point mutations in non-small cell lung cancer by PCR-SSCP analysis. Genomic DNAs from surgical specimens that show mobility shifts correspond to GTT (valine) mutations at codon 12 were also detected by ASO hybridization.

Statistical analysis Dichotomous variables were analyzed with Pearson's chi square test. Mann-Whitney U test was used to compare interval variables. Survival curves were calculated by the Kaplan and Meier method [19], comparison between patients with a mutation and patients without a mutation, was done by the exact log-rank statistic. P-values for the comparison between two proportions were calculated with the percentiles of Fisher's exact test. Multivariate analysis for survival was performed by using the Cox model [20].

Results K-ras mutations were detected in 51 of 192 non-smallcell lung cancer specimens (27%) by means of PCR of DNA segments containing exons 1 and 2 followed by either ASO hybridization or SSCP assays. K-ras mutations were more commonly found at codon 12 with wild-type glycine (GGT) changing to aspartic acid (GAT) in 10 individual tumors, serine (AGT) in 5, valine (GTT) in 14, cysteine (TGT) in 9, arginine (CGT) in 2 and alanine (GCT) in 1. Ras mutations at codon 61 were: leucine (CTA) in 1 case, glutamine (GAA) in 1 case, proline (CCA) in 1 case, arginine (CGA) in 5 cases and lysine (AAA) in two. Examples of dot-blot and SSCP analysis are illustrated in Figs. 1 and 2. Mutations were present in adenocarcinomas (17 of 58) (29%) as well as other histological types, 27 of 108 (25%) squamous cell carcinomas and in 7 of 26 (27%) large cell undifferentiated carcinomas. The baseline patient characteristics are depicted in Table 1. There was a trend towards there being K-ras negative tumors in lower stages for instance, in 29 of 36 (81%)

Fig. 2. Detection of K-ras mutations by allelle-specitic oligonucleotide (ASO) hybridization, corresponding to valine.

stage I patients, 11 of 13 stage II (84%), 54 of 72 stage IHA (75%) and 24 of 31 stage IIIB (78%) as opposed to 23 of 40 stage IV patients (58%). The overall survival of the 192 patients analysed according to TNM classification was as follows: 43 months median survival time in stage I, 18 months in stage II, 14 months in stage IHA, 10 months in stage ITTR and 4 months in stage IV (figure not shown). When we split survival according to K-ras status a notable 46.4 months median survival time was observed in the stage I ras negative group, 42 months in certain ras mutations (except aspartic and serine codon 12 mutations) and 21 months in aspartic acid and serine codon 12 K-ras mutation group. However, the differences did not reach statistical significance. In the 72 stage UIA patients, median survival time was 16 months for the negative ras mutation group, 14 months for certain ras mutations and only 7 months for aspartic and serine codon 12 mutation group (p - 0.01). Table 2 depicts characteristics of the 72 stage IHA patients broken down by K-ras genotypes. We observed that those patients whose tumors contained K-ras codon 12 aspartic acid or serine mutations were more prone to

Downloaded from http://annonc.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on June 15, 2015

DNA samples were amplified as above and 1 (il of a 1/100 dilution of the PCR was used as a template for a 20 (il nested PCR reaction. DNA amplification was performed in 10 mM Tris HC1 (pH 9.0 at 25 *C), 50 mM KC1, 0.1% Triton X-100, 1.5 mM MgCl2, 200 uM dATP, 200 uM dTTP, 200 uM dGTP, 150 uM dCTP, 0.5 uCi (50 uM) of a-32P-dCTP (Amersham, U.K.), 0.5 uM of KR1U and 0.5 uM of an internal downstream K-ras primer (KR1D-2 5' GGTTGATGGTGTTCAAATAT 3') in a 9600 Perkin-Elmer Cerus DNA thermal cycler. The thermal profile used was 94 *C for 1 minute, 15 x (94 *C for 20 seconds, 56 *C for 20 seconds, and 72 #C for 20 seconds), and 74 *C for 4 minutes. Two to four |il of the nested PCR reaction were mixed with an equal volume of denaturating solution (98% formamide, 10 mM EDTA pH 8.0, 0.02% xylene cyanol, 0.02% bromophenol blue). The DNA mixture was denatured at 95 'C for 5 minutes, chilled on ice and loaded onto a nondenaturing 6% polyacrilamide - 1 xTBE gel containing 10% glycerol. Electrophoresis was performed at 4 W for 16 hours at 4 *C. The gel was dried and finally exposed to an autoradiographic film (Kodak OAR, U.S.A.) using intensifying screens for several hours at -80'C.

18 Table 2. Characteristics of 72 stage IIIA NSCLC patients broken down by K-ras mutation status. Certain K-ras mutations

K-ras codon 12 Asp and Ser mutations

54 (75%)

13 (18%)

5 (9%)

32 (74%) 15 (71%) 7 (88%)

8 (19%) 4 (19%) 1 (12%)

3 (7%) 2 (9%)

15 (28%) 21 (39%) 18 (33%)

1 (8%) 5 (38%) 7 (54%)

1 (20%) 4 (80%)"

16

14

7b

• p-0.05. b p-0.01.

develop distant metastases. K-ras related median survival time in stage INK was 10 months in both the negative ras mutation group and in the certain ras mutation group, whereas it was 2 months in the aspartic acid and serine codon 12 mutation group. No differences surfaced in stage IV group as median survival time was undistinguishable among the three established subsets (4 months). Using multivariate analyses, stage and K-ras mutation were found to be associated with outcome. Discussion Our study illustrates that K-ras mutations may well be a relevant prognostic marker mainly in early stages of non-small-cell lung cancer. The subgroup of stage HIA patients whose tumors had K-ras codon 12 aspartic or serine mutations (9%) had a worse outcome (7 months median survival time) whereas survival in patients whose tumors had no ras mutations was substantially better (26% five-year survival). An intermediate group of patients was made up of those whose tumors had ras mutations other than aspartic acid and serine (18%). Mounting evidence indicates that K-ras gene mutations could be a notable prognostic marker in nonsmall-cell lung cancer. K-ras mutations were detected in nineteen of 69 lung adenocarcinomas in a Dutch study, mostly in stage I, and those patients whose tumors had mutated K-ras had poorer survival [8]. This finding was confirmed in a series of 66 non-small-cell lung cancer cell lines in the stage I-HIA setting but not in more advanced disease. When survival in advanced disease was broken down into those tumors with K-ras codon 12 mutations as compared with the remainder (ras negative plus mutations at codon 61), the difference turned out to be significant [7]. Sugio et al. [10] also found K-ras status-linked differences among resected lung adenocarcinomas but only in stage I. hi our previous experience we detected 13 tumors with K-ras

Downloaded from http://annonc.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on June 15, 2015

No. of patients Histological type Squamous cell Adenocarcinoma Large cell Relapse No Local Distant Median survival time (months)

No ras mutation

mutations among 66 resected non-small-cell lung cancers of any histological type with significant differences in survival [9]. More recently, Kern et al. [21] observed 16 of 44 stage I-IIIA lung adenocarcinoma patients harboring K-ras mutations, the difference in survival approached significance (p =° 0.06), although in the subset of 6 patients (14%) whose tumors contained K-ras mutations and c-erbB-2 overexpression there was a much worse outcome. In addition to K-ras status, other genomic perturbations have been singled out as having prognostic value, such as p53 mutations [22, 23] and c-erbB-2 overexpression [24]. However in a recent study, c-erbB-2 was expressed in 93% of lung adenocarcinomas [25] as compared with 34% in the above mentioned study [24]. These differences are likely to be related to immunohistochemistry-related conditions. Formalin fixation and paraffin embedding of tissue may alter c-erbB-2 antigens, whereas the high frequency which was detected using frozen tissue sections could be linked to the type of monoclonal antibody. This high frequency of expression is not likely to confer prognosis to this protein overexpression. Overexpression of p53 protein detected immunohistochemically, (a surrogate for missense-type p53 gene mutations) has been reported to be associated with decreased survival of non-small-cell lung cancer patients but not with p53 gene mutations [26]. On the other hand, a mutation both in p53 and K-ras genes has been detected very uncommonly in the same lung tumor (4%-7%) [25, 27]. This contradicts data obtained from cell lines in which p53 gene mutations in exon 8 had a significantly higher incidence of ras mutations (54%) than those in cell lines with p53 gene mutations at other exons (16%) [28]. It seems, however, that the presence of mutations in one gene did not correlate with the presence of mutations in the other gene, a fact that has also been observed in colorectal and endometrial carcinomas [2, 29]. Our experience (data in preparation) does not support the theory of meaningful cooperation between these genes in vivo. The presence of K-ras mutation and p53 gene mutations probably defines separate groups of patients. Poor tumor differentiation has not attracted much attention to date, although that could be a point of criticism when addressing the role of K-ras mutations as prognostic marker. Mutations in K-ras were found to be equally common in the different grades of tumor differentiation in one study [27] and similarly, accumulation of K-ras mRNA in tumor tissue has been reported to be independent of cell differentiation. Microsatellite instability in non-small-cell lung cancer has recently been documented. This manifestation of the replication error (RER) phenotype was described in sporadic colorectal carcinomas. When microsatellite markers from chromosome 3p were used, genetic instability was recorded in 34% of cases and, more importantly, some of them harbored K-ras mutations [30] a phenomenon discovered in endo-

19

Acknowledgements

We thank Dr. Jose Javier Sanchez for his contribution in the statistical analysis and Maura O'Sullivan for her editing of the manuscript. Supported in part by a grant from Bristol-Myers Squibb Company, Spain and by a Spanish Ministry of Health grant, Fondo de Investigaciones Sanitarias, FIS (95/0177). We thank the following investigators who, in addition to the authors, contributed patient tumor samples to this trial: Dr. Antonio Anton, Hospital Miguel Servet, Zaragoza, Dr. Catalina Vadell, Hospital del Mar, Barcelona.

References 1. Barbacid M. Ras genes. Annu Rev Biochem 1987; 56: 779827.

2. Peinado MA, Fernandez-Renart M, Capella G et al. Mutations in the p53 suppressor gene do not correlate with c-K-ras oncogene mutations in colorectal cancer. Jnt J Oncol 1993; 2: 123-34. 3. Sklar MD. The ras oncogenes increase the intrinsic resistance of NIH 3T3 cells to ionizing radiation. Science 1988; 239: 645-7. 4. Roth JA, Fossella FV, Komaki R et al. A randomized trial comparing perioperative chemotherapy and surgery with surgery alone in resectable stage IIIA non-small cell lung cancer. J Natl Cancer Inst 1994; 86:673-80. 5. Rosell R, Gomez-Codina J, Camps C et al. A randomized trial comparing preoperative chemotherapy plus surgery with surgery alone in patients with non-small cell lung cancer. N Engl J Med 1994; 330: 153-8. 6. Rodenhuis S, Slebos RJC, Boot AJM et al. Incidence and possible clinical significance of K-ras oncogene activation in adenocarcinoma of the human lung. Cancer res 1988; 48: 5738-41. 7. Mitsudomi T, Viallet J, Mulshine R et al. Mutations of ras genes distinguish a subset of non-small-cell lung cancer cell lines from small-cell lung cancer cell lines. Oncogene 1991; 6: 1353-62. 8. Slebos RJC, Kibbelaar RE, Dalesio O et al. K-ras oncogene activation as a prognostic marker in adenocarcinoma of the lung. N Engl J Med 1990; 323: 561-5. 9. Rosell R, Li S, Skacel Z et al. Prognostic impact of mutated K-ras gene in surgically resected non-small cell lung cancer patients. Oncogene 1993; 8: 2407-12. 10. Sugio K, Ishida T, Yokoyama H et al. Ras gene mutations as a prognostic marker in adenocarcinoma of the human lung without lymph node metastasis. Cancer Res 1992; 52: 2903-6. 11. Mitsudomi T, Steinberg SM, Oie HK et al. Ras gene mutations in non-small cell lung cancers are associated with shortened survival irrespective of treatment intent. Cancer Res 1991; 51: 4999-5002. 12. Rodenhuis S, Van de Watering ML, Mooi WJ et al. Mutational activation of the K-ras oncogene. A possible pathogenetic factor in adenocarcinoma of the lung. N Engl J Med 1987; 317: 929-35. 13. Capella G, Cronauer-Mitra S, Peinado MA, Perucho M. Frequency and spectrum of mutations at codons 12 and 13 of the c-K-ras gene in human rumors. Environ Health Perspect 1991; 93:125-31. 14. Suzuki Y, Orita M, Shiraishi M et al. Detection of ras gene mutations in human lung cancers by single-strand conformation polymorphism analysis of polymerase chain reaction products. Oncogene 1990; 5:1037-43. 15. Rosell R, Li S, Anton A et al. Prognostic value of K-ras 12 genotypes in patients with advanced non-small cell lung cancer receiving carboplatin with either intravenous or chronic oral dose etoposide. Int J Oncol 1994; 5:169-76. 16. Nelson MA, Garewal HS, Rosenberg R et al. Improved detection of K-ras mutations in non-small cell lung cancer by a sensitive PCR method: A potential biomarker. Proc Am Soc Clin Oncol 1993; 12:469 (Abstr). 17. Slebos RJC, Boerrigher L, Evers SG. A rapid and simple procedure for the routine detection of ras point mutations in formalin-fixed paraffin-embedded tissues. Diagn Mol Pathol 1992; 1:136-41. 18. Mountain CE A new international Staging system for lung cancer. Chest 1986; 89 (Suppl 4> 225S-32S. 19. Kaplan EL, Meier P. Non parametric estimation from incomplete observations. Am Stat Assoc J 1958; 53: 457-81. 20. Cox DR. Regression models and life-tables. J R Stat Soc 1972; 34: 187-220. 21. Kern JA, Slebos RJC, Top B et al. C-erb-2 expression and codon 12 K-ras mutations both predict shortened survival for patients with pulmonary adenocarcinomas. J Clin Invest 1994; 93:516-20. 22. Horio Y, Takahashi T, Kuroishi T et al. Prognostic significance

Downloaded from http://annonc.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on June 15, 2015

metrial carcinoma [31] where K-ras mutations were significantly more frequent in RER positive (56%) phenotype than in RER negative (14%) tumors. In experimental models it seems that mismatch repair deficiency in human chromosome 3 contributes to microsatellite instability and enhances the level of tolerance to the cytotoxic action of DNA-alkylating agents [32]. Furthermore as progression from a treatment-sensitive to a treatment-resistant tumor state is a virtually universal phenomenon in patients with small-cell lung cancer, transitions from small-cell lung cancer to non-smallcell lung cancer phenotype can be detected. This has been proven in cell culture models when the insertion of the viral Harvey ras genes into small-cell lung cancer cell lines induces the appearance of many non-smallcell lung cancer phenotypic features [33]. It is important to make two caveats to this analysis: there may be geographical differences (in the frequency, type and prognostic significance of K-ras mutations) and a larger study is desirable to substantiate the hypothesis that the presence of specific K-ras genotypes may be an independent prognostic factor. A genotypic classification has been established in colorectal adenocarcinomas analyzed by K-ras status. Those patients whose tumors had codon 12 aspartic acid mutations were more inclined to develop distant hematogenous metastases, whereas codon 13 aspartic had a very indolent clinical course as well as mutations at codon 12 valine substitution. The remaining mutations were more aggressive than those tumors with wild-type K-ras [34]. In summary, some K-ras codon 12 genotypes may help to define tumor aggressiveness and could serve to select resected non-small-cell lung cancer patients as candidates for adjuvant chemotherapy, whereas in the more advanced disease setting we hypothesize that K-ras mutations could help to identify patients who exhibit chemoresistance.

20

23.

24. 25. 26.

28.

29.

30. 31. 32.

33. 34.

lesions of the human uterine endometrium. Cancer Res 1993; 53: 1883-8. Shridar V, Siegfried J, Hunt J et al. Genetic instability of microsatellite sequences in many non-small cell lung carcinomas. Cancer Res 1994; 54: 2084-7. Duggan BD, Felix JC, Muderspach LI et al. Microsatellite instability in sporadic endometrial carcinoma. J Natl Cancer Inst 1994; 86:1216-21. Koi M, Umar A, Chauhan DP et al. Human chromosome 3 corrects mismatch repair deficiency and microsatellite instability and reduces N-methyl-N'-nitro-N-nitrosoguanidine tolerance in colon tumor cells with homozygous hMLHl mutation. Cancer Res 1994; 54:4308-12. Falco JP, Baylin SB, Lupu R et al. v-rasH induces non-small cell phenotype, with associated growth factors and receptors, in a small cell lung cancer cell line. J Clin Invest 1990; 85:1740-5. Finkelstein SD, Sayegh R, Bakker A, Swalsky P. Determination of tumor aggressiveness in colorectal cancer by K-ras-2 analysis. Arch Surg 1993; 128: 526-32.

Correspondence to: Rafael Rosell.M.D. University Hospital Germans Trias i Pujol Medical Oncology Service Box 72 08916 Badalona, Barcelona Spain

Downloaded from http://annonc.oxfordjournals.org/ at University of Iowa Libraries/Serials Acquisitions on June 15, 2015

27.

of p53 mutations and 3p deletions in primary resected nonsmall cell lung cancer. Cancer Res 1993; 53:1-4. Mitsudomi T, Oyama T, Kusano T et al. Mutations of the p53 gene as a predictor of poor prognosis in patients with non-small cell lung cancer. J Natl Cancer Inst 1993; 85: 201823. Kern JA, Schwart DA, Nordberg JE et al. pl85™™ expression in human lung adenocarcinomas predicts shortened survival. Cancer Res 1990; 50: 5184-91. Bongjorno PF, Whyte RI, Lesser EJ et al. Alteration of K-ras, p53 and erbB-2/neu in human lung adenocarcinomas. J Thorac Cardiovasc Surg 1994; 107: 590-5. Carbone DP, Mitsudomi T, Rusch V et al. p53 protein overexpression, but not gene mutation, is predictive of significantly shortened survival in resected non-small cell lung cancer patients. Proc Am Soc Clin Oncol 1993; 12: 334 (Abstr). Ridanpaa M, Karjalainen A, Anttila S et al. Genetic alterations in p53 and K-ras in lung cancer in relation to histopathology of the tumor and smoking history of the patient. Lnt J Oncol 1994; 5:1109-17. Mitsudomi T, Steinberg SM, Nau MM et al. p53 gene mutations in non-small-cell lung cancer cell lines and their correlation with the presence of ras mutations and clinical features. Oncogene 1992; 7:171-8fj' Enomoto T, Fujita M, Jnoue M et al. Alterations of the P53 tumor suppressor gene and its association with activation of the c-K-ras-2 protooncogene in premalignant and malignant