This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schiller, J. H. Articles by Johnson, D. H. Search for Related Content PubMed PubMed Citation Articles by Schiller, J. H. Articles by Johnson, D. H.

Lack of Prognostic Significance of p53 and K-ras Mutations in Primary Resected Non–Small-Cell Lung Cancer on E4592: A Laboratory Ancillary Study on an Eastern Cooperative Oncology Group Prospective Randomized Trial of Postoperative Adjuvant Therapy

From the William S. Middleton Veterans Administration Hospital and University of Wisconsin, Madison; University of Wisconsin, Milwaukee, WI; Eastern Cooperative Oncology Group, Dana-Farber Cancer Institute, Boston, MA; University of Rochester Cancer Center, Rochester; Beth Israel Medical Center, New York; Albany Medical Center, Albany; State University of New York, Syracuse, NY; Evanston Hospital, Evanston, IL; University of Washington, Seattle, WA; Latter-Day Saints Hospital and University of Utah School of Medicine, Salt Lake City, UT; Mayo Clinic, Rochester, MN; and Vanderbilt University Medical School, Nashville, TN.

Address reprint requests to Joan H. Schiller, MD, University of Wisconsin, Department of Medicine, Medical Oncology Section, K4/666 Clinical Science Center, 600 Highland Ave, Madison, WI 53792; email jhschill{at}facstaff.wisc.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To determine the prognostic and predictive significance of p53 and K-ras mutations in patients with completely resected non–small-cell lung cancer (NSCLC).

PATIENTS AND METHODS: Patients were randomized preoperatively to receive adjuvant postoperative radiotherapy (Arm A) or radiotherapy plus concurrent chemotherapy (Arm B). p53 protein expression was studied by immunohistochemistry (IHC) and p53 mutations in exons 5 to 8 were evaluated by single-strand conformational analysis. K-ras mutations in codons 12, 13, and 61 were determined using engineered restriction fragment length polymorphisms.

RESULTS: Four hundred eighty-eight patients were entered onto E3590; 197 tumors were assessable for analysis. Neither presence nor absence of p53 mutations, p53 protein expression, or K-ras mutations correlated with survival or progression-free survival. There was a trend toward improved survival for patients with wildtype K-ras (median, 42 months) compared with survival of patients with mutant K-ras who were randomized to chemotherapy plus radiotherapy (median, 25 months; P = .09). Multivariate analysis revealed only age and tumor stage to be significant prognostic factors, although there was a trend bordering on statistical significance for K-ras (P = .066). Analysis of survival difference by p53 by single-stranded conformational polymorphism and IHC, interaction of p53 and K-ras, interaction of p53 and treatment arm, nodal station, extent of surgery, weight loss, and histology did not reach statistical significance.

CONCLUSION: p53 mutations and protein overexpression are not significant prognostic or predictive factors in resected stage II or IIIA NSCLC. K-ras mutations may be a weak prognostic marker. p53 or K-ras should not be routinely used in the clinical management of these patients.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
LUNG CANCER IS THE leading cause of cancer-related deaths in the United States, responsible for more deaths than breast cancer, colon cancer, and prostate cancer combined. Although surgery can be curative in 70% to 80% of patients who present with node-negative disease, individuals who present with nodal involvement have a significantly poorer prognosis. Cure rates with surgery alone for stage II (N1) or resectable stage IIIA (N2) disease range from 30% to 50%. Identification of resectable patients who are at risk for recurrent disease (prognostic factors), or who are likely to benefit from additional treatment (predictive factors), would be helpful in identifying patients who may benefit from additional therapy.

Classic prognostic factors have included the tumor-node-metastasis staging system, performance status, sex, and weight loss. Recently a number of molecular markers have been identified which are implicated in the causation of cancer. These markers, including overexpression of certain oncogenes or inactivation of tumor suppressor genes, have been found in significant numbers of patients with lung cancer and are therefore logical candidates for prognostic or predictive factors.

Ras oncogenes have been implicated in the pathogenesis of a wide variety of human and animal tumors.¹ These genes (K-, H-, and N-) encode membrane-bound 21-kd proteins with GTP-binding activity, which play an important role in cellular signal transduction pathways. The ras proteins acquire transforming potential when point mutations in codons 12, 13, or 61 lead to amino acid substitutions, presumably because alterations at these sites are associated with loss of the intrinsic GTPase activity.^1,2 The frequency at which ras mutations in human cancer are found depends on the organ of origin, being rare in squamous cell carcinomas and adenocarcinomas of the breast, stomach, and ovary and occurring in 30% to 50% of adenocarcinomas of the lung and colon^3-8 and almost all pancreatic cancers.⁹

The incidence and prognostic significance of K-ras mutations in lung cancer have been studied in a number of retrospective series. These mutations occur primarily in codon 12 and are found more often in adenocarcinomas obtained from smokers than nonsmokers.¹⁰ In some retrospective series, the presence of the K-ras mutation has been found to be a negative prognostic factor, associated with early relapse and shortened survival.^7,11-15

The p53 gene, located on the short arm of human chromosome 17, encodes for a nuclear phosphoprotein involved in the regulation of cell proliferation.¹⁶ The mutant gene product, which tends to accumulate to high levels in cancer cells, is believed to exert a dominant negative effect over coexpressed normal p53. Alterations of either the gene or protein product have turned out to be one of the most common changes identified in human malignancies. In resected lung cancers, point mutations of the p53 gene have been found in all histologic types, including approximately 45% of resected non–small-cell lung cancer (NSCLC) and even more frequently in SCLC.^17-22

Similar to observations with K-ras mutations, p53 mutations have been retrospectively correlated with clinical features. These studies have shown that p53 mutations are associated with younger age and squamous histology, but not sex, tumor stage, nodal status, neuroendocrine differentiation, or prior chemotherapy.^20,23 However, the relationship between p53 alterations and survival in NSCLC has been controversial. p53 overexpression has been variously correlated with worse outcome, better prognosis, or no influence on patients’ survival.^24-30

Although the interaction and cooperation of mutant p53 and K-ras has been postulated to be an important determinant of tumor progression, few studies have analyzed both p53 and K-ras mutational status and their relationship to patient survival or determined the importance of their interaction as prognostic factors.^26,27,31 p53 and K-ras genes have been found to be frequently, but not always, mutated in the same lung carcinoma cell lines.³² However, the prevalence of p53 mutations in NSCLC cell lines with ras mutations did not differ from that in cell lines without ras mutations.²³ Therefore, prospective studies that assess the prognostic utility of these genetic abnormalities are required.

E3590 was a prospective randomized intergroup trial of postoperative adjuvant therapy in patients with completely resected stages II and IIIA NSCLC. After stratification for nodal status, weight loss, histology, and lymph node dissection, patients were randomized to receive either thoracic radiation therapy or four cycles of cisplatin and etoposide plus concurrent thoracic radiation therapy. The objectives of this laboratory study were to determine the incidence and prognostic significance of p53 and K-ras mutations in patients with stages II and IIIA NSCLC undergoing postoperative thoracic radiation therapy plus or minus chemotherapy.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patient Study
EST3590 (Intergroup 0115) was a randomized, prospective trial of adjuvant radiotherapy plus or minus chemotherapy in patients with resected stages II and IIIA NSCLC. The objectives of the study were to determine if concomitant chemoradiotherapy was superior to radiotherapy alone in prolonging survival and preventing local tumor recurrence after resection. The clinical results of the study are reported elsewhere.³³

To be eligible for E3590, the clinical study, patients were required to have complete resection of the primary tumor and thorough mediastinal lymph node sampling or dissection. They were then randomized to receive either four cycles of cisplatin (60 mg/m² intravenously on day 1) and etoposide (120 mg/m² intravenously days 1 to 3) administered concurrently with thoracic radiotherapy (RT) (50.4 Gy in 28 daily Gy fractions) or RT alone. Patients were stratified by weight loss (< 5% v > 5%), histology (squamous v other histologies), nodal biopsy technique (lymph node sampling v complete lymph node dissection), and lymph node station (N₁ v N₂).

All patients registered onto E3590 were eligible for entry to the laboratory study, E4592, provided that at least one paraffin block of primary tumor or a portion of tumor tissue removed from the block was available from the patient. All patients gave informed consent for both the clinical study (E3590) and the laboratory correlative study (E4592).

DNA Extraction
With the aid of a hematoxylin and eosin stained slide, 15-micron thick paraffin sections were dissected (microdissected) free of normal tissue. Paraffin was removed by two exchanges of xylene and the xylene removed by two exchanges of ethanol. The pellet was dried and digested overnight at 37°C with 100 µL of 0.2% proteinase K (Boehringer, Indianapolis, IN). Samples were then heated to 95°C for 10 minutes to kill the proteinase K, and 10 µL of mussel glycogen (Boehringer) was added as a carrier. Samples were extracted once with 100 µL phenol:chloroform and once with chloroform. DNA was precipitated with 3 µL of 5M NaCl and two volumes of ethanol and frozen 30 minutes at -70°C. The DNA was pelleted and washed once with 70% ethanol at -20°C. The pellet was dried and resuspended in 100 µL of TE buffer, pH 8 (10 mM Tris, 0.1 mmol/L EDTA) and allowed to dissolve for 2 days at 4°C before use. DNA concentrations were determined on 20 samples and ranged from 200 to 500 µg/mL.

p53 Mutational Analysis
p53 mutations in exons 5 to 8 were evaluated by single-strand confirmation polymorphism (SSCP) analysis. Genomic DNA was subjected to polymerase chain reaction (PCR) using oligonucleotide primers; primer sequences were as follows:

Exon 5a, sense strand: TGA CTT TCA ACT CTG TCT CCT T

Exon 5a, antisense strand: CAT GTG CTG ACT GCT TGT

Exon 5b, sense strand: GTT GAT TCCACA CCC CCG

Exon 5b, antisense strand: GTC GTC TCT CCA GCC CCA

Exon 6, sense strand: AGG CTT CTG ATT CCT CAC TGA

Exon 6, antisense strand: CCA GAG ACC CCA GTT GCA AAC

Exon 7, sense strand: TGG GCC TGT GTT ATC TCC T

Exon 7, antisense strand: GCA CAG CAG GCC AGT GTG CAG

Exon 8, sense strand: CTT CTC TTT TCC TAT CTT G

Exon 8, antisense strand: CTT CTT GTC CTG CTT GCT T

PCR reactions were designed to generate fragments from p53 exons 5 to 8, ranging from 148 to 182 bp in length. PCR conditions were optimized according to polymerase, primer sequences, amounts of primers, fragment sizes, magnesium concentrations, the use of cosolvents (DMSO, formamide, betaine, glycerol, PEG), and cycling parameters to obtain efficient amplification of DNA from paraffin-embedded tumor samples with little or no background. Primers selected are listed above. Optimized reaction conditions were as follows: all reactions contained 1 µL of DNA (0.2 to 0.5 µg) 40 mmol/L of Tricine-KOH, pH 9.2, 15 mmol/L of KOAc, 3.5 mmol/L Mg(OAc)₂, 3.75 µg/mL of BSA, 200 µm of dNTPs, 1 µL of KlenTaq-1 DNA polymerase (Advantage cDNA Polymerase Mix, CLONTECH Laboratories, Palo Alto, CA), and 100 ng of DNA, in a 50-µL reaction volume. In addition, exon 5a contained 0.2 µmol/L of each primer and 0.2 x SB (PCR Sample Buffer, containing betaine, Epicenter Technologies, Madison, WI). Optimized amplification conditions were as follows: 94°C, 5 minutes; (94°C, 10 seconds; 61°C, 10 seconds; 72°C, 10 seconds) x 33 cycles; 72°C, 5 minutes; and 4°C, hold. Exon 5b contained 0.2 µmol/L of each primer and 5% DMSO. Optimized amplification conditions were as follows: 94°C, 5 minutes; (94°C, 10 seconds; 62°C, 10 seconds; 72°C, 10 seconds) x 33 cycles; 72°C, 5 minutes; and 4°C, hold. Exon 6 contained 0.16 µmol/L of each primer and 0.2 x SB. Optimized amplification conditions were as follows: 94°C, 5 minutes; (94°C, 30 seconds; 59°C, 8 seconds; 72°C, 8 seconds) x 33 cycles; 72°C, 5 minutes; 4°C, hold. Exon 7 contained 0.4 µmol/L of each primer and 0.2 x SB. Optimized amplification conditions were as follows: 94°C, 5 minutes; (94°C, 20 seconds; 64°C, 10 seconds; 72°C, 20 seconds) x 33 cycles; 72°C, 5 minutes; 4°C, hold. Exon 8 contained 0.2 µmol/L of each primer and 0.2 x SB. Optimized amplification conditions were as follows: 94°C, 5 minutes; (94°C, 10 seconds; 58°C, 12 seconds; 72°C, 10 seconds) x 33 cycles; 72°C, 5 minutes; 4°C, hold.

Amplified fragments were treated with exonuclease I (USB, Cleveland, OH) to remove excess primers. PCR reactions were heated to 70°C for 30 minutes, then cooled to 37°C. Fifteen units of Exo I were added, and the samples were incubated for 30 to 40 minutes at 37°C. Samples were heated again to 70°C for 30 minutes to inactivate the Exo I and stored at 4°C until analysis.

Exo I-digested PCR fragments were subjected to SSCP analysis.¹⁹ A cold SSCP technique (NOVEX, San Diego, CA) was used. Twenty percent precast TBE-polyacrylamide gels were run in the NOVEX ThermoFlow Mini-Cell Unit and stained with silver nitrate (SilverXpress, NOVEX). Gels were run at 320 V; optimum run temperature and time were determined for each amplicon: exon 5a (167 bp), 2 hours at 25°C; exon 5b (148 bp), 2 hours at 19°C; exon 6 (160 bp), 4 hours at 4°C; exon 7 (172 bp), 4 hours at 24°C; and exon 8 (182 bp), 8 hours at 14°C. Silver staining was performed using the manufacturer’s protocols.

After SSCP, the gels were stained using the Novex SilverXpress staining kit and photographed using a digital imaging system (Alpha Innotech, San Leandro, CA). Seventy seven exons were analyzed by direct sequence analysis and compared with SSCP. Fifteen of 19 exons that were mutant by SSCP were also mutant on sequencing; 57 of 58 exons that were wildtype by SSCP were also wildtype on sequencing (concordance rate, 96%).

p53 Immunohistochemistry (IHC)
Five-micron thick paraffin sections were deparaffinized with xylene and rehydrated through a series of graded alcohol to water. Antigen retrieval consisted of boiling for 10 minutes in acetate buffer. Staining with monoclonal antip53 clone DO-7 (Dako, Carpenteria, CA) was performed on a Ventana gen II or Ventana ES auto-stainer and detected using Ventana antibody detection reagents (Ventana, Tucson, AZ.) Blinded analysis of slides was performed by two observers. Slides were scored for intensity (0 to 4+) and percentage positivity: 0 (0%), 1+ (< 25%), 2+ (25% to 50%), 3+ (50% to 75%), 4+ (> 75%).

K-ras Mutational Analysis
K-ras mutations were analyzed using primer-engineered RFLP. Primer sequences and PCR conditions have been previously described.^4,32 Primers were synthesized by the University of Wisconsin Biotechnology Center (Madison, WI).

PCR was performed on 1 µL of DNA extract (approximately 0.2 to 0.5 µg) per 50-µL reaction using AmpliTaq DNA polymerase (Perkin Elmer, Branchburg, NJ) in a 96-well thermocycler (M.J. Research, Watertown, MA). After PCR amplification, K-ras fragments were digested as follows (enzymes from New England Biolabs, Beverly, MA): codon 12, Ban I, 37°C overnight; codon 13, Hae III, 37°C overnight; codon 61, positions 1 and 2, Bcl I, 50°C 5-7 hours; and codon 61, position 3, Ear I, 37°C overnight.

The digested products were run next to an undigested portion of the same sample on a 4% agarose gel consisting of a 1:1 mix of NuSieve GTG (FMC, Rockland, ME) and Analytic Grade Agarose (BioRad, Hercules, CA). Gels were stained with ethidium bromide, and bands were imaged on a UV lightbox (IBI, New Haven, CT) and photographed with a polaroid camera.

Statistical Design
This correlative study to E3590 was based on anticipated accrual rates, assuming that 203 cases would be available for analysis from E4592. The design also assumed a 25% K-ras mutation rate. With these assumptions, the study design for E4592 gave 90% power, using a two-sided level 0.05 log-rank test for detecting a 9-month difference in median survival (median survival of 21 months for patients with wildtype K-ras and median survival of 12 months for patients with mutant K-ras).

Statistical Analysis
Fisher’s exact test³⁴ was used for comparisons with respect to categorical end points (eg, incidence of mutation). Survival time was calculated as the time to death from the date of randomization on E3590; patients who were alive were censored at the last known date alive. Progression-free survival was calculated as the time to progression or death without progression from the date of randomization on E3590; patients who were alive and progression-free were censored as of date of last known follow-up.

Survival distributions for survival time and progression-free survival were estimated with the Kaplan-Meier method³⁵ and compared with the log-rank test.³⁶ Statistical significance was set at a level of 0.05, and all log-rank test P values reported are two-tailed.

Cox proportional hazards model were used to estimate the joint effect of prognostic factors on survival. In the Cox model, stepwise selection was used to determine parsimonious models. Statistical significance was set at a level of 0.10 and all P values reported are two-tailed. Possible factors for inclusion in the model consisted of treatment arm, p53 (SSCP) mutation status, p53 (IHC) mutation status, K-ras mutation status, interaction of p53 (SSCP) and K-ras mutation status, the stratification factors (nodal stage, histology, weight loss, and lymph node dissection), and selected baseline patient characteristics (age group, sex, race, Eastern Cooperative Oncology Group performance status, T stage, and tumor-node-metastasis stage.

Four hundred eighty-eight patients were entered onto the parent clinical study of E3590 between April 1991 and February 1997. Two hundred seventeen patients were registered onto the laboratory correlative study, 111 in arm A and 106 in arm B (Table 1). All patients were considered assessable for survival in an intent-to-treat analysis. Twenty samples were never submitted. Of the 197 samples, 14 were unassessable for p53 analysis by SSCP (12 cases: too many infiltrating lymphocytes to allow dissection of tumor, two failed PCR); and 13 samples were unassessable for K-ras analysis (12 cases: unable to adequately dissect tumor; one sample failed PCR amplification).

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

Possible explanations for this difference in the survival include the fact that E4592 became active in August 1993, whereas E3590 was activated in April 1991. Though patients who were already registered onto E3590 were eligible for E4592, informed consent was required, which meant that patients who died early could not have been registered onto E4592. In addition, patients were often not registered onto both studies at the same time even after 1993. There was an average difference of approximately 3 months (range, 0 to 26 months) from the time of registration onto E3590 and registration onto E4592.

p53 (SSCP)
Of 183 assessable tumors, 83 (45%) had p53 mutations by SSCP. One of 78 putatively normal regional lymph nodes was also found to have a p53 mutation. Distribution of the p53 by exons is listed in Table 2. Sixty percent of patients with squamous cell carcinoma had a p53 mutation by SSCP, compared with 37.5% of patients with nonsquamous histology (P = .005). The median survival of the 83 patients with mutant p53 by SSCP was 38 months (95% CI, 31 to 61 months); for the 100 patients with wildtype p53, it was 52 months (95% CI, 28 to 66 months; P > .83 by log-rank analysis) (Fig 1A).

View larger version (25K): [in this window] [in a new window]   Fig 1. Overall survival of patients with mutant and wildtype p53 by (A) SSCP or (B) immunohistochemistry.

There was no significant difference in progression-free or overall survival by p53 mutational status—wildtype versus mutant—when analyzed by treatment arm. The median survival for patients randomized to radiation therapy alone was 38 months for mutant p53 and 52 months for wildtype p53; for patients randomized to combined radiation and chemotherapy, the median survival for mutant and wildtype p53 was nearly identical at 40 months. There was no difference in progression-free or overall survival between wildtype and mutant p53 when analyzed by performance status, weight loss, nodal station, histology, type of lymph node dissection, age, sex, or race. (Data not shown.)

p53 (IHC)
Overexpression of p53 was also determined by IHC. Ninety-nine of 180 tumors (55%) overexpressed the p53 protein (staining 2 to 4+). No difference in median survival was observed between p53 overexpressors and wildtype p53 (40 months [95% CI, 33 to 61 months] v 42 months [95% CI, 27 to 86 months], respectively (P = .93 by log-rank analysis) (Fig 1B).

Among the 81 patients with wildtype p53 by IHC, 42 patients have died, with a 1-year survival rate of 76.5% (95% CI, 67.3% to 85.8%). Fifty of the 99 patients with mutant p53 by IHC have died with a 1-year survival rate 84.8% (95% CI, 77.8% to 91.9%). The difference in survival distributions by p53 (IHC) mutational status was not statistically significant (risk ratio of wildtype v mutant is 0.975; 95% CI, 0.65 to 1.45; log-rank P value = .93).

Similar to p53 mutational analysis by SSCP, no difference in overall survival was observed when p53 expression was broken down by subgroup. The median survival of patients on arm A with both p53 overexpression and wildtype p53 by IHC was 45 months; for patients on arm B, it was 40 months versus 42 months (P = .65 by log-rank analysis).

The IHC data were analyzed by both percentage of cells staining positive and intensity of staining (Table 3). No differences in qualitative results were identified when the data were analyzed by intensity of staining or percentage of cells staining positive (data not shown). An IHC index was also derived by multiplying the percentage of cells positive with the intensity of staining, but again no qualitative difference in results was observed.

K-ras
Forty-four of 184 assessable tumors were positive for K-ras for an overall incidence of 24% (Table 2). One of the 85 putatively normal lymph nodes had detectable K-ras mutations. The majority of these mutations in the tumors were in codon 12. Only three of 63 squamous cell carcinoma patients had K-ras mutations, compared with a 33% incidence in patients with nonsquamous histology (P < .05) (Table 2). Thirty-one percent of patients younger than 60 years had mutant K-ras, compared with 17% of patients older than 60 years who had K-ras mutations (P = .025 by univariate analysis).

The median survival of the 44 patients with K-ras mutation was 30 months (95% CI, 34 to 64 months): the median survival for the 140 patients with wildtype K-ras was 42 months (95% CI, 34 to 64 months) (Fig 2). This was not statistically significant by log-rank analysis (P = .38). Seventy-nine of the 140 patients with wildtype p53 K-ras have died with a 1-year survival rate of 82.9% (95% CI, 76.6% to 89.1%). Among the 44 patients with mutant K-ras, 26 have died with a 1-year survival rate of 70.4% (95% CI, 56.9% to 83.9%). The difference in survival distributions by K-ras mutation status was not statistically significant (risk ratio of wildtype v mutant is 0.82; 95% CI for risk ratio, 0.52 to 1.27; log-rank P value = .38). No difference in progression-free survival was observed.

View larger version (20K): [in this window] [in a new window]   Fig 2. Overall survival of patients with mutant and wildtype K-ras.

a) Patients on the chemotherapy arm of the study: 70 patients who had wildtype K-ras had median survival of 41.8 months, whereas 20 patients with mutant K-ras had median survival of 24.7 months. (Risk ratio of wildtype v mutant K-ras was 0.59; 95% CI, 0.32 to 1.075; and log-rank P value = .09) (Fig 3).

b) Patients with Eastern Cooperative Oncology Group performance status 1: 82 patients who had wildtype K-ras had median survival of 34.4 months, compared with 29 patients with mutant K-ras who had median survival of 35.9 months, (risk ratio of wildtype v mutant K-ras was 1.08; 95% CI, 0.625 to 1.87; and log-rank P value = .08).

c) Patients with weight loss 5%: 34 patients who had wildtype K-ras had median survival of 38.3 months, compared with six patients with mutant K-ras who had median survival of 17.8 months (risk ratio of wildtype v mutant K-ras was 0.39; 95% CI, 0.14 to 1.06; and log-rank P value = .05).

d) Patients with nodal stage of N1: 71 patients who had wildtype K-ras had median survival of 45.2 months, and 20 patients with mutant K-ras had median survival of 23.6 months (risk ratio of wildtype v mutant K-ras was 0.47; 95% CI, 0.25 to 0.90; and log-rank P value = .02).

View larger version (27K): [in this window] [in a new window]      Fig 3. Survival of patients randomized to (A) RT or (B) RT plus chemotherapy by K-ras status.

View larger version (25K): [in this window] [in a new window]   Fig 4. Survival of patients with (A) wildtype or (B) mutant K-ras by treatment arm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

We observed a 45% to 55% incidence of p53 mutations by SSCP and IHC, respectively; similar to what has been reported in other studies.^17,18 However, we were unable to find any correlation between p53 mutations and overall survival or between p53 mutations and any of the patient characteristics we examined. p53 abnormalities by either SSCP or IHC were neither prognostic nor predictive factors. We attempted to correlate survival with IHC using either the percentage of cells positive or the intensity of staining and were unable to find a correlation. We also derived a p53 IHC index by multiplying the percentage of positive cells with the intensity of staining, similar to an analysis occasionally used to determine the estrogen receptor positivity in breast cancer. No correlation was observed.

Three other studies found an improved survival with p53 overexpression. Lee et al²⁵ studied 156 patients and found that those whose tumors stained with 50% or more cells staining had a significantly superior survival (65 months) than those with low or negative staining (median survivals, 26 months v 33 months, respectively; P = .002, comparing high v low or negative). Conversely, another series studied 85 NSCLC tumors and identified p53 overexpression in 55% and mutations in 45%; a significant reduction in survival was noted with p53 overexpression (P = .01), but not gene mutation.²⁴ Other studies have shown no correlation between p53 mutational status and survival.²⁹

The reason for such conflicting information on the prognostic significance of p53 protein expression is not known. Differences in the monoclonal antibody used, in the methods of antigen retrieval, in the types of specimens (eg, frozen v fixed), or in the criteria for positive staining may explain some of the differences, although one would not expect such a wide divergence of results. However, studies using SSCP to assess mutational status have generally shown no association between p53 mutation and survival.

Overexpression of p53 by immunostaining has not been shown to correlate well with p53 mutations detected by SSCP. In one study, the concordance rate was only 67%, and although there was a negative survival advantage with positive p53 immunostaining by univariate analysis, no survival correlation was identified with the presence of gene mutations.²⁴ In another series, less than one half of those that stained strongly positive had p53 mutations.³⁷ Some of the discrepancy may be a result of technical difficulties associated with microdissection and contamination of tumor extraction with normal cells. Furthermore, positive p53 immunostaining does not invariably correlate with gene mutation. Differences in overexpression may be dependent on the type of mutation, with p53 missense mutation resulting in high expression, and deletions, splicing mutations, and nonsense mutants resulting in low levels of p53.^38,39

In our study, no correlation between the presence or absence of K-ras mutation and overall survival or progression-free survival was observed, although there was a trend toward improved survival for patients with wildtype K-ras compared with mutated K-ras in patients receiving chemotherapy plus radiation therapy. This had originally been reported as statistically significant; however, with further follow-up, the P value dropped from .012 to .09.⁴⁰

The mutational status of K-ras has been extensively studied as a prognostic marker in NSCLC. The presence of the K-ras mutation has been reported to be a negative prognostic factor, associated with early relapse and shortened survival.^7,11-15 However, these studies generally are retrospective, involve small numbers of tumors (typically less than 50),¹⁰ or use frozen tumor specimens or cell lines obtained over a period of many years^7,11,12,41 or from heterogeneous groups of patients who received a number of different, ill-defined treatments.^13,15,42 In other studies, the presence of K-ras mutations has not been found to be of prognostic significance by multivariate analysis in patients with resected early-stage NSCLC⁴³ or found to be only significant in stage II NSCLC patients in subgroup analyses.¹⁵ A recent prospective study reported an association between K-ras mutations and K-ras survival that was statistically significant for stage I tumors only.⁴⁴

K-ras and p53 status were analyzed to determine whether the addition of chemotherapy to radiotherapy impacted on survival in either the wildtype or mutational group. No differences in survival were noted between K-ras–mutated patients receiving RT or chemotherapy plus RT. Similarly, no differences in survival were noted in patients with mutated p53 between treatment arms, suggesting that these mutations are not factors predictive for response to chemotherapy.

Few studies have looked at the prognostic interaction between genetic mutations. Abnormalities in p53 and erbB-2 confirmed a worse prognosis than either one alone.⁴⁵ Patients who were bcl-2 positive and p53 negative survived significantly longer than those who were bcl-2 negative or p53 positive.⁴⁶ In our study, we failed to detect any interaction between p53 and K-ras in multivariable analysis.

This study is one of the few multi-institutional studies examining K-ras and p53 as potential prognostic and predictive markers in patients entered a prospective randomized trial. We conclude that p53 protein overexpression or mutational status does not have independent prognostic significance. The number of K-ras mutations detected in this study was relatively small, thus decreasing the power of the study to detect statistically significant differences. Although a trend toward differences in survival was observed in arm B between patients with mutant or wildtype K-ras, K-ras remains at best a weak prognostic marker. Thus, we conclude that the clinical usefulness of these two molecular markers in predicting prognosis or response to treatment for stages II or IIIA NSCLC is limited and should not become part of standard clinical care.

	ACKNOWLEDGMENTS

 
This study was conducted by the Eastern Cooperative Oncology Group (Robert L. Comis, MD, Chair) and supported in part by Public Health Service grants no. CA21076, CA23318, CA49957, CA66636, and CA21115 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD. J.H.S. is also partially supported by the William S. Middleton Veterans Administration Hospital.

We thank David Carbone, MD, PhD, and Gerard Bittner for their expertise and help with the development of the p53 and K-ras assays, Tim Toonen, MD, for his help with p53 IHC, and Terry Oberly, MD, for his advice and direction with the p53 IHC.

	NOTES

 
The contents of this study are solely the responsibility of the authors and do not represent the official views of the National Cancer Institute.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Barbacid M: Ras genes. Annu Rev Biochem 56: 779-827, 1987[Medline]

2. Bos J: Ras oncogene in human cancer: A review. Cancer Res 49: 4682-4689, 1989[Abstract/Free Full Text]

3. Forrester K, Almoguera C, Han K, et al: Detection of high incidence of K-ras oncogenes during human colon tumorigenesis. Nature 327: 298-303, 1987[Medline]

4. Mitsudomi T, Viallet J, Mulshine JL, 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 6: 1353-1362, 1991[Medline]

5. 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 5: 1037-1043, 1990[Medline]

6. Rodenhuis S, van de Wetering 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 317: 929-935, 1987[Abstract]

7. 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 323: 561-565, 1990[Abstract]

8. Tortola S, Marcuello E, Gonzalez I, et al: p53 and K-ras gene mutations correlate with tumor aggressiveness but are not of routine prognostic value in colorectal cancer. J Clin Oncol 17: 1375-1381, 1999[Abstract/Free Full Text]

9. Almoguera C, Shibata D, Forrester K, et al: Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 53: 549-554, 1988[Medline]

10. Slebos RJC, Hruban RH, Dalesio O, et al: Relationship between K-ras oncogene activation and smoking in adenocarcinoma of the human lung. J Natl Cancer Inst 83: 1024-1027, 1991[Abstract/Free Full Text]

11. Rodenhuis S, Slebos RJC: Clinical significance of ras oncogene activation in human lung cancer. Cancer Res 52: 2665s-2669s, 1992 (suppl)

12. Mitsudomi T, Steinberg S, Oie H, et al: Ras gene mutations in non-small cell lung cancers are associated with shortened survival irrespective of treatment intent. Cancer Res 51: 4999-5002, 1991[Abstract/Free Full Text]

13. Kobayashi T, Tsuda H, Noguchi M, et al: Association of point mutation in c-Ki-ras oncogene in lung adenocarcinoma with particular reference to cytologic subtypes. Cancer 66: 289-294, 1990[Medline]

14. 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 52: 2903-2906, 1992[Abstract/Free Full Text]

15. Graziano SL, Gamble GP, Newman NB, et al: Prognostic significance of K-ras codon 12 mutations in patients with resected stage I and II non-small-cell lung cancer. J Clin Oncol 17: 668-675, 1999[Abstract/Free Full Text]

16. Chang F, Syrjanen S, Syrjanen K: Implications of the p53 tumor-suppressor gene in clinical oncology. J Clin Oncol 13: 1009-1022, 1995[Abstract]

17. Greenblatt MS, Bennett WP, Hollstein M, et al: Mutations in the p53 tumor suppressor gene: Clues to cancer etiology and molecular pathogenesis. Cancer Res 54: 4855-4878, 1994[Free Full Text]

18. Bennett W, Hussain S, Vahakangangas K, et al: Molecular epidemiology of human cancer risk: Gene-environment interactions and p53 mutation spectrum in human lung cancer. J Pathol 187: 8-18, 1999[Medline]

19. Orita M, Iwahana H, Kanazawa H, et al: Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc Natl Acad Sci U S A 86: 2766-2770, 1989[Abstract/Free Full Text]

20. Chiba I, Carbone D, Minna J, et al: Mutations in the p53 gene are frequent in primary, resected non-small cell lung cancer. Oncogene 5: 1603-1610, 1990[Medline]

21. Takahashi T, Nau M, Chiba I, et al: p53: A frequent target for genetic abnormalities in lung cancer. Science 246: 491-494, 1989[Abstract/Free Full Text]

22. Iggo R, Gatter K, Bartek J, et al: Increased expression of mutant forms of p53 oncogene in primary lung cancer. Lancet 335: 675-679, 1990[Medline]

23. 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 7: 171-180, 1992[Medline]

24. Carbone DP, Mitsudomi T, Chiba I, et al: p53 immunostaining positivity is associated with reduced survival and is imperfectly correlated with gene mutations in resected non-small cell lung cancer. A preliminary report of LCSG 871. Chest 106: 377S-381S, 1994 (suppl)[Medline]

25. Lee J, Yoon A, Kalapurakal S, et al: Expression of p53 oncoprotein in non-small-cell lung cancer: A favorable prognostic factor. J Clin Oncol 13: 1893-1903, 1995[Abstract/Free Full Text]

26. Scagliotti G, Ferrari G, Leonardo E, et al: Preliminary report on the expression of p53, I-ras, Ki67 in radically resected non-small cell lung cancers randomized in a clinical trial of adjuvant chemotherapy (ALPI trial). Proc Am Soc Clin Oncol 16: 457a, 1997 (abstr 1643)

27. Kwiatkowski D, Harpole D, Godleski J, et al: Prognostic factor analysis of 250 stage I NSCLC patients: Pathologic features are more important than molecular analyses. Proc Am Soc Clin Oncol 16: 457a, 1997 (abstr 1644)

28. Quinlan D, Davidson A, Summers C, et al: Accumulation of p53 protein correlates with a poor prognosis in human lung cancer. Cancer Res 52: 4828-4831, 1992[Abstract/Free Full Text]

29. McLaren R, Kuzu I, Dunnill M, et al: The relationship of p53 immunostaining to survival in carcinoma of the lung. Br J Cancer 66: 735-738, 1992[Medline]

30. Pappot H, Francis D, Brunner N, et al: p53 protein in non-small cell lung cancer as quantitated by enzyme-linked immunosorbent assay: Relation to prognosis. Clin Cancer Res 2: 155-160, 1996[Abstract]

31. Kim YC, Park KO, Kern JA, et al: The interactive effect of Ras, HER2, P53 and Bcl-2 expression in predicting the survival of non-small cell lung cancer patients. Lung Cancer; 22: 181-190, 1998[Medline]

32. Lehman TA, Bennet WP, Metcalf RA, et al: C. p53 mutations, ras mutations, and p53-heat shock 70 protein complexes in human lung carcinoma cell lines. Cancer Res 51: 4090-4096, 1991[Abstract/Free Full Text]

33. Keller S, Adak S, Wagner H, et al: Prospective randomized trial of postoperative adjuvant therapy in patients with completely resected stages II and IIIa non-small cell lung cancer: An intergroup trial (E3590). Proc Am Soc Clin Oncol 18: 465a, 1999 (abstr 1793)

34. Mehta C, Patel N: A network algorithm for performing Fisher’s exact test in RXC contingency tables. J Am Stat Assoc 78: 427-434, 1983

35. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958

36. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50: 163-170, 1966[Medline]

37. Marchetti A, Buttitta C, Merlo G, et al: p53 alterations in non-small cell lung cancers correlate with metastatic involvement of hilar and mediastinal lymph nodes. Cancer Res 53: 2846-2851, 1993[Abstract/Free Full Text]

38. Bodner SM, Minna JD, Jensen SM, et al: Expression of mutant p53 proteins in lung cancer correlates with the class of p53 gene mutation. Oncogene 7: 743-749, 1992[Medline]

39. Kastan MB, Zhan Q, el-Deiry WS, et al: A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 71: 587-597, 1992[Medline]

40. Schiller J, Adak S, Feins R, et al: Prognostic significance of p53 and K-RAS mutations in primary resected non-small cell lung cancer: Preliminary results from a prospective randomized trial of postoperative adjuvant therapy (Intergroup Trial 0115) Proc Am Soc Clin Oncol 18: 464a, 1999 (abstr 1789)

41. Harada M, Dosaka-Akita H, Miyamoto H, et al: Prognostic significance of the expression of ras oncogene product in non-small cell lung cancer. Cancer 69: 72-77, 1992[Medline]

42. Rodenhuis S, Slebos R, Boot A, et al: Incidence and possible clinical significance of K-ras oncogene activation in adeno-carcinoma of the human lung. Cancer Res 48: 5738-5741, 1988[Abstract/Free Full Text]

43. Mehdi S, Tatum A, Newman N, et al: Prognostic markers in resected stage I and II non-small-cell lung cancer: An analysis of 260 patients with 5 year follow-up. Clin Lung Cancer 1: 59-67, 1999[Medline]

44. Nelson H, Christiani D, Mark J, et al: Implications and prognostic value of K-ras mutation for early stage lung cancer in women. J Natl Cancer Inst 91: 2032-2038, 1999[Abstract/Free Full Text]

45. Harpole D, Herndon J, Wolf W, et al: A prognostic model of recurrence and death in stage I non-small cell lung cancer utilizing presentation histopathology and oncoprotein expression. Cancer Res 55: 51-56, 1995[Abstract/Free Full Text]

46. Ohsaki Y, Toyoshima E, Fujiuchi S, et al: bcl-2 and p53 protein expression in non-small cell lung cancers: Correlation with survival time. Clin Cancer Res 2: 915-920, 1996[Abstract]

Submitted February 23, 2000; accepted September 8, 2000.

This article has been cited by other articles:

				M.-S. Tsao, S. Aviel-Ronen, K. Ding, D. Lau, N. Liu, A. Sakurada, M. Whitehead, C.-Q. Zhu, R. Livingston, D. H. Johnson, et al. Prognostic and Predictive Importance of p53 and RAS for Adjuvant Chemotherapy in Non Small-Cell Lung Cancer J. Clin. Oncol., November 20, 2007; 25(33): 5240 - 5247. [Abstract] [Full Text] [PDF]

				L. A. Robinson, J. C. Ruckdeschel, H. Wagner Jr, and C. W. Stevens Treatment of Non-small Cell Lung Cancer-Stage IIIA: ACCP Evidence-Based Clinical Practice Guidelines (2nd Edition) Chest, September 1, 2007; 132(3_suppl): 243S - 265S. [Abstract] [Full Text] [PDF]

				H. H. Nelson, M. Wilkojmen, C. J. Marsit, and K. T. Kelsey TP53 mutation, allelism and survival in non-small cell lung cancer Carcinogenesis, October 1, 2005; 26(10): 1770 - 1773. [Abstract] [Full Text] [PDF]

				D. A. Eberhard, B. E. Johnson, L. C. Amler, A. D. Goddard, S. L. Heldens, R. S. Herbst, W. L. Ince, P. A. Janne, T. Januario, D. H. Johnson, et al. Mutations in the Epidermal Growth Factor Receptor and in KRAS Are Predictive and Prognostic Indicators in Patients With Non-Small-Cell Lung Cancer Treated With Chemotherapy Alone and in Combination With Erlotinib J. Clin. Oncol., September 1, 2005; 23(25): 5900 - 5909. [Abstract] [Full Text] [PDF]

				L. Thomas, L. A. Doyle, and M. J. Edelman Lung Cancer in Women: Emerging Differences in Epidemiology, Biology, and Therapy Chest, July 1, 2005; 128(1): 370 - 381. [Abstract] [Full Text] [PDF]

				S. Singhal, A. Vachani, D. Antin-Ozerkis, L. R. Kaiser, and S. M. Albelda Prognostic Implications of Cell Cycle, Apoptosis, and Angiogenesis Biomarkers in Non-Small Cell Lung Cancer: A Review Clin. Cancer Res., June 1, 2005; 11(11): 3974 - 3986. [Abstract] [Full Text] [PDF]

				K. M.W. Pisters and T. Le Chevalier Adjuvant Chemotherapy in Completely Resected Non-Small-Cell Lung Cancer J. Clin. Oncol., May 10, 2005; 23(14): 3270 - 3278. [Abstract] [Full Text] [PDF]

				G. R. Simon, S. Sharma, A. Cantor, P. Smith, and G. Bepler ERCC1 Expression Is a Predictor of Survival in Resected Patients With Non-small Cell Lung Cancer Chest, March 1, 2005; 127(3): 978 - 983. [Abstract] [Full Text] [PDF]

				K M Fong, Y Sekido, A F Gazdar, and J D Minna Lung cancer * 9: Molecular biology of lung cancer: clinical implications Thorax, October 1, 2003; 58(10): 892 - 900. [Abstract] [Full Text] [PDF]

				S. A. Ahrendt, Y. Hu, M. Buta, M. P. McDermott, N. Benoit, S. C. Yang, L. Wu, and D. Sidransky p53 Mutations and Survival in Stage I Non-Small-Cell Lung Cancer: Results of a Prospective Study J Natl Cancer Inst, July 2, 2003; 95(13): 961 - 970. [Abstract] [Full Text] [PDF]

				M. D. Brundage, D. Davies, and W. J. Mackillop Prognostic Factors in Non-small Cell Lung Cancer* : A Decade of Progress Chest, September 1, 2002; 122(3): 1037 - 1057. [Abstract] [Full Text] [PDF]

G. Bepler, A. Gautam, L. M. McIntyre, A. F. Beck, D. S. Chervinsky, Y.-C. Kim, D. M. Pitterle, and A. Hyland Prognostic Significance of Molecular Genetic Aberratio