|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Alteration of p53 Pathway in Squamous Cell Carcinoma of the Head and Neck: Impact on Treatment Outcome in Patients Treated With Larynx Preservation IntentByFrom the Departments of Medicine, Radiation Oncology, Surgery, Pathology, and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY. Supported in part by the Overman Fund and the Garban Intercapital Fund.Address reprint requests to: David G. Pfister, MD, Memorial Sloan-Kettering Cancer Center, Box 188, 1275 York Ave, New York, NY 10021; email: pfisterd{at}mskcc.org
PURPOSE: To identify the role of p53 pathway alteration(s) as predictors of treatment outcome in patients with advanced, resectable, squamous cell carcinoma (SCC) of the larynx and pharynx treated with larynx preservation (LP) intent.
PATIENTS AND METHODS: Seventy-one patients treated on two consecutive LP protocols were studied based on availability of representative tissues. We analyzed the expression pattern of p53, its upstream regulator mdm2, and downstream transcriptional target p21/WAF1 by immunohistochemistry. Positive phenotype was defined as RESULTS: Positive phenotype was observed in 35 (49%) of 71 cases for p53, in 52 (74%) of 70 for mdm2, and in 37 (54%) of 68 for p21. There was no correlation between p53 phenotype and p21 nuclear accumulation. The mdm2-negative phenotype was most predictive of major response at the primary tumor site (P = .088). p53-positive phenotype was associated with worse local control with LP (LCLP; 49% v 23%, P = .053) and inferior overall survival (OS; 51% v 29%, P = .017) at 5 years. On Cox regression analysis, p53-positive phenotype predicted inferior OS (P = .033) and showed a trend for worse LCLP (P = .102). When analyzed in a multivariate model as continuous variables, p53 showed a stronger correlation with inferior OS (P < .01), and mdm2 was associated with worse OS (P < .01). CONCLUSION: Among the three markers studied, our data support p53 phenotype as the most informative predictor of unfavorable outcomes in the LP setting, and suggest a role for mdm2 phenotype that requires further exploration. Our analysis does not support a p53-dependent mechanism for p21 expression. Prospective and larger studies are necessary before integration of these molecular markers as part of molecular staging and predictors for organ preservation or other outcomes.
CONVENTIONAL TREATMENTS of surgery and radiation therapy for advanced squamous cell carcinoma (SCC) of the larynx and pharynx are associated with significant morbidity affecting speech, swallowing, and overall quality of life. Combination chemoradiotherapy yields survival results comparable with total laryngectomy and radiotherapy, with a possibility of larynx preservation (LP). However, the LP approach is not successful in all cases, and more than 30% of surviving patients require salvage laryngectomy.1-3 Reliable predictors of treatment outcome are needed to minimize the morbidity and cost of unsuccessful LP because many patients may be best served by early referral for surgical intervention. In this study, we tested the hypothesis that inactivation of the p53 pathway might confer tumor resistance to apoptosis-inducing chemoradiotherapy, resulting in a decrease in local control and overall survival (OS). Alteration of p53 pathway is of interest in this setting because TP53 mutations and/or p53 protein overexpression are common events in SCC of the head and neck,4,5 and several studies have reported a correlation between TP53 mutation and response to cisplatin and fluorouracil, the chemotherapeutic agents historically most frequently used in this setting.6,7 Moreover, several clinical studies have demonstrated a correlation between TP53 mutation and radioresistance.8,9 A recognized limitation of most studies is that they are confined to the analysis of TP53 mutation or p53 expression patterns but do not examine other critical components of the p53 pathway (Fig 1). The product encoded by the MDM2 gene binds to the transcriptional activation domain of p53, blocking its ability to regulate gene expression. The MDM2 gene itself is a transcriptional target of p53, thus resulting in a feedback loop that controls both the activity of p53 and the expression of mdm2. Moreover, mdm2 is involved in p53 degradation because it binds to its amino-terminal region and, functioning as an ubiquitin E3-ligase, presents p53 for proteosome-mediated enzymatic degradation.10-12 Thus, mdm2 is recognized as a critical p53-negative regulator. Both in vitro and in vivo data support the oncogenic role of MDM2 in several tumor models. In vitro, the overexpression of mdm2 in murine cells increases their growth rate.13 Transgenic mice overexpressing mdm2 were tumor prone, with 50% developing lymphomas and sarcomas. Moreover, these mice showed an increased incidence of sarcomas in comparison to p53-null mice. This increased incidence was maintained in crosses onto the p53-null background, demonstrating a p53-independent role for MDM2 in tumorigenesis.14
The lack of data regarding the functional status of p53 downstream products in analyzed tumors represents another drawback in the published studies. The described yeast-based assay can distinguish between silent TP53 mutations acquired by the malignant cells and those that truly contribute to the malignant phenotype, but this assay is difficult to perform as part of routine clinical investigations.15 Accordingly, we chose to study the expression of p21/WAF1 as an indirect evidence of p53 function. It has been reported that p21/WAF1 gene expression may serve as an indicator of p53 activity because p21/WAF1 is under the direct transcription control of p53.16 However, serum or individual growth factors, such as epidermal growth factor and fibroblast growth factor, which are known to be upregulated in SCC of the larynx, can also induce p21 expression in p53-deficient cells.17 Thus, there are at least two separate pathways that could account for the induction of p21 in SCC of the head and neck, one linked to DNA damage recognition (p53 dependent) and the other produced by signaling mechanisms caused by certain cellular mitogens (p53 independent). In the present study, we analyzed the expression pattern of p53 and its upstream regulator mdm2 and downstream transcriptional target p21/WAF1 in 71 patients with advanced SCC of the larynx and pharynx treated on one of two consecutive LP protocols at Memorial Sloan-Kettering Cancer Center (MSKCC) between 1988 and 1995. The association between these markers and response to induction chemotherapy, local control with LP (LCLP), time to distant failure, and OS was examined to determine their potential prognostic impact.
Patient Selection One hundred seventy patients with advanced SCC of the larynx and pharynx treated on one of two consecutive LP protocols at MSKCC between 1988 and 1995 were the subjects of our investigation. Entry criteria were similar among the two studies and have been previously reviewed.18,19 All patients had the following: histologically confirmed SCC; advanced-stage SCC of the larynx, hypopharynx, or oropharynx that would, if treated surgically, jeopardize the larynx; assessable/measurable disease present; Karnofsky performance status (KPS)  60%; and adequate hematologic, renal, hepatic, cardiac, and auditory functions to tolerate therapy with high-dose cisplatin-based chemotherapy. All patients gave written informed consent for participation in these protocols approved by the MSKCC Institution Review Board. Archival representative blocks from the primary tumors were available for 71 of 170 patients’ tumors, which formed the basis of this study. These patients were entered onto a separate Institution Review Board–approved tissue-acquisition protocol to conduct the correlative immunohistochemical study.
Initial Evaluation
Treatment
Evaluation of Tumor Response
Assessment of Treatment Outcome
Immunohistochemistry Methods
Statistical Methods
Correlations of biologic marker expression and response (major and complete) to chemotherapy (both at the primary site and overall) were performed using the
Multivariate analysis was felt to be necessary to correct for potential baseline characteristic confounders as well as imbalances in the treatment regimens between groups. Given the available sample size and related outcome events, only a limited number of covariates could be entered into each model. Accordingly, the following strategy was used. Each potential pretreatment prognostic factor (KPS, stage, age, treatment regimen, weight loss, symptoms, comorbidity [Charlson Comorbidity Index], alcohol use, cigarette use, albumin, ALT, hemoglobin, platelets, mean corpuscular volume, lactic acid dehydrogenase, tracheostomy, and feeding tube status) were also analyzed with univariate analysis by log-rank tests for significance in predicting OS and LCLP. Each potential prognostic pretreatment factor that was determined to be significant in predicting OS (ie, log-rank test < 0.05) was then included in a multivariate analysis using Cox’s proportional-hazards model,31 doing the regression in a forward stepwise manner with a threshold significant value of 0.05. Then, while only including the variables that remained significant in the multivariate model, each biologic marker was added separately into the Cox’s proportional-hazards model and reanalyzed in a forward stepwise manner with a threshold significant value of 0.05. The biologic markers were analyzed in a similar manner for their significance in predicting LCLP. A multivariate assessment focusing on time to distant failure was not performed because the number of distant failure events (n = 10) was judged insufficient to yield a stable analysis. Median follow-up data was computed using the inverse Kaplan-Meier method32 and was obtained through December 1, 2000. Nonparametric inferences were performed using the
Table 1 compares the study population with available archival tissues (n = 71) with unavailable cases (n = 99). No significant differences in baseline variables were observed, with the exception of the location of primary disease, with a greater proportion of larynx cancer (49%, tissue available v 62%, tissue unavailable) in the group for which tissue was unavailable. The reason for unavailability of specimens was frequently that patients had their diagnostic biopsies performed outside MSKCC, with no available tissues in the archival bank. Immunophenotypic study of p53 expression was performed in all 71 cases, whereas representative slides for mdm2 and p21 analysis were available for 70 (99%) of 71 and 68 (96%) of 71 cases, respectively.
p53-positive phenotype, defined as 20% of tumor cell showing nuclear immunoreactivity, was observed in 35 (49%) of 71 cases (Fig 2 upper panel). There were no statistically significant differences in baseline variables between p53-negative and p53-positive cases (Table 2), although the power of such comparison is limited.
Fifty-two (74%) of 70 tumors were positive for mdm2 (Fig 2 middle panel), and 37 (54%) of 68 tumors were positive for p21 overexpression (Fig 2 lower panel). We examined all possible correlations between the three markers, and we did not observe any statistically significant association. Because of the role p53 plays in the induction of p21 transcription, this analysis was of special interest. There was no correlation found between p53 wild phenotype and p21 expression (P = .64). As noted in Table 2, for the most part, baseline characteristics were comparable among negative and positive cases. However, we observed a higher proportion of patients whose tumor was negative for p21 (P = .089) or mdm2 (P = .154) overexpression who participated in the earlier protocol (1988-90). To determine whether there might be significant clinical or prognostic differences between patients on the older versus newer protocols, exploratory analyses were performed. Although the numbers compared limit the power of the significance testing, no significant imbalance in age, KPS, primary site, T stage, N stage, or overall stage was found.
Correlation With Response to Induction Chemotherapy No molecular marker, analyzed individually or in combination, significantly predicted tumor response to induction chemotherapy (P values ranged between .88 to .98). However, lack of mdm2 expression showed a trend with major response at the primary site (P = .088) and complete response at the primary site and neck (P = .10). Because there were no major differences in the induction regimen or response rates of the different protocols (except for recycling time), these comparisons were not controlled by treatment regimen.
Correlation With Local Control (LCLP) and Time to Distant Failure
Only 10 patients had documented distant failure, approximately one fourth the rate of local failure observed, and in only four patients was it the sole site of failure. Accordingly, given the limited number of distant failure events, the related analyses should be interpreted cautiously. None of studied markers significantly predicted time to distant failure, although the p53-negative group had a somewhat higher incidence of distant failure (P = .10). A multivariate assessment was not performed because the relatively small number of distant events was judged insufficient to yield a stable analysis.
Correlation With OS It should be emphasized that these last two hazard ratios, obtained when the representative markers were used as continuous variables, reflect the increased risk associated with a 1% increase in staining. The scale of change used will affect the hazard ratio. For example, if the staining was to increase from 20% to 40%, the related hazard risk becomes 1.49. Because of the disparity in mdm2 staining characteristics between the 1988 to 1990 and 1991 to 1995 groups, we evaluated the impact of mdm2 expression on survival in these two groups separately to see if its predictive value might be affected. On univariate analysis, mdm2 overexpression (as a continuous variable) delineated a strong trend (P = .086) in the older group and was also significant in the newer group (P = .014). Because the specific groups were now 33 and 37 patients, respectively, we felt a multivariate assessment involving more than two variables would be too unstable. Accordingly, a Cox model was performed using KPS, the most predictive variable in a previous model, with mdm2 as a continuous variable. In this analysis, mdm2 overexpression was significant in both groups (P = .021 and .030, respectively).
Avoidance of total laryngectomy is the goal of organ/function preservation in SCC of the larynx and pharynx. Several groups have demonstrated that combination chemoradiotherapy is feasible in patients with advanced disease whose surgical management would have required total laryngectomy. Survival results were similar to those obtained historically using standard management, and a substantial proportion of patients avoided total laryngectomy. However, in more than one third of patients selected for chemoradiotherapy, the treatment will fail.1-3 Better markers of treatment outcome are needed. This study supports p53 phenotype, among the three markers studied, as the most informative predictor of unfavorable outcome in patients undergoing LP treatment. We could not demonstrate a role for p21 expression as an indirect evidence of functioning p53. Nonetheless, the analysis suggests a role for mdm2 phenotype that requires further study because of the observed trend with chemotherapy response as well as the correlation with OS when analyzed as a continuous variable. The preliminary results, however, need to be interpreted cautiously and require validation. Most of the published studies examining the role of p53 as a prognostic marker in SCC of the head and neck have focused on correlation with pathologic features, mostly in the setting of initial presentation. Few studies have examined the correlation of p53 and response to specific treatment intervention and outcome. Our results are in concordance with studies that revealed a strong correlation between p53 nuclear accumulation and worse treatment outcome in patients with advanced SCC.33,34 Of special interest is a study conducted by the Department of Veterans Affairs Laryngeal Cancer Cooperative Study Group, which examined the correlation between p53 nuclear accumulation and response to chemotherapy in SCC of the larynx. This study showed that p53 nuclear accumulation significantly correlated with successful organ preservation,35 a finding that is divergent from our analysis. There are several possible explanations for the observed dichotomy in the results, including differences in methodology and cutoff points used to define p53 overexpression. The same group has subsequently published an article on a subset of patients that showed a correlation between p53 mutation and worse outcome, and there was no correlation between p53 mutation and p53 overexpression.36 Other differences between the studies were that our study was not limited to laryngeal carcinoma as was the Veterans Affairs study, and had a higher proportion of patients with T4 (59% v 26%) and N2/N3 (60% v 28%) disease, affecting the anticipated incidence of treatment failure and related ability to detect prognostic differences. There are two studies of mdm2 in SCC of the larynx, which suggest that mdm2 overexpression may be an early event in pathogenesis37,38 because it occurs in preneoplastic lesions that progress to invasive SCC. No studies have directly correlated mdm2 overexpression with clinical outcome in SCC. The mdm2 protein product could be considered a novel target of treatment for patients with advanced SCC of the head and neck. Recently, enhancement of drug-induced apoptosis by antisense oligodeoxynucleotides and synthetic peptides targeted against mdm2 has been described.39,40 These novel antitumor agents, which have proven sufficient in blocking the degradation of p53 in vitro and have shown antitumor activity using xenografts models, should be considered as potential candidates for the treatment of tumors overexpressing mdm2. The relatively high frequency of mdm2 overexpression in our study population (74%) and the observed trend with chemotherapy response as well as correlation with worse outcome merit further consideration of mdm2 as a biomarker and target of treatment in SCC of the larynx and pharynx. Our observation that p21/WAF1 accumulates in invasive tumor cells (Fig 2 lower panel) and the lack of correlation with wild-type p53 phenotype suggest that p21 overexpression in advanced SCC may reflect a p53-independent mechanism. In support of this postulate, overexpression of epidermal growth factor receptor and/or basic fibroblast growth factor are reported in epithelial hyperplastic laryngeal lesions and observed in both early and advanced head and neck SCC.41,42 The postulate that p21 expression, in this setting, is not p53-dependent has been previously suggested.43,44 A similar observation has been reported for both prostate and breast cancers.45,46
There were several limitations in this study that need to be considered. First, tissue collection was not inclusive. This problem is common to many studies that examine potential biomarkers and can potentially bias the final results. Second, the observation that mdm2- and p21-positive staining was less frequent in the 1988 to 1990 group, cannot be entirely explained. Statistical chance is possible but should not be assumed.47 It is reassuring in this regard, however, that mdm2 remained predictive of survival in both groups. Third, by convention, overexpression of p53, p21, and mdm2 is defined as In conclusion, our study supports p53 phenotype as the most informative predictor of unfavorable outcome, as well a potential role for mdm2, which will require further studies and exploration. Future prospective research, in which all tissue in a treatment group is analyzed, will be required to truly understand the impact these molecular markers may have on outcome.
1. The Department of Veterans Affairs Laryngeal Cancer Study Group: Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 324: 1685-1690, 1991[Abstract] 2. Lefebvre JL, Chevalier D, Luboinski B, et al: Preliminary results of a European Organization for Research and Treatment of Cancer phase III trial: EORTC Head and Neck Cancer Cooperative Group. Cancer 88: 890-899, 1996 3. Pfister DG, Shaha AR, Harrison LB: The role of chemotherapy in the curative treatment of head and neck cancer. Surg Oncol Clin N Am 6: 749-768, 1997[Medline] 4. Gottschlich S, Maune S, Preugschat J, et al: p53 analysis of laryngeal cancer in exon 4 to 9. Anticancer Res 20: 2613-2616, 2000[Medline] 5. Morawski K, Gabriel A, Namyslowski G, et al: Clinical application of proliferating cell nuclear antigen, oncoprotein p53 and tumor front grading analysis in patients operated on for laryngeal cancer. Eur Arch Otorhinolaryngol 256: 378-383, 1999[CrossRef][Medline] 6. Nakashima S, Natsugoe S, Matsumoto M, et al: Expression of p53 and p21 is useful for the prediction of preoperative chemotherapeutic effects in esophageal carcinoma. Anticancer Res 20: 1933-1937, 2000[Medline]
7.
Thor AD, Berry DA, Budman DR, et al: erbB-2, p53, and efficacy of adjuvant therapy in lymph node-positive breast cancer. J Natl Cancer Inst 90: 1346-1360, 1998 8. Concin N, Zeillinger C, Stimpfel M, et al: p53-dependent radioresistance in ovarian carcinoma cell lines. Cancer Lett 150: 191-199, 2000[CrossRef][Medline] 9. Raybaud H, Fortin A, Bairati I, et al: Nuclear DNA content, an adjunct to p53 and Ki-67 as a marker of resistance to radiation therapy in oral cavity and pharyngeal squamous cell carcinoma. Int J Oral Maxillofac Surg 29: 36-41, 2000[CrossRef][Medline] 10. Prives C: Signaling to p53: Breaking the MDM2-p53 circuit. Cell 95: 5-8, 1998[CrossRef][Medline] 11. Freedman DA, Wu L, Levine AJ: Functions of the MDM2 oncoprotein. Cell Mol Life Sci 55: 96-107, 1999[CrossRef][Medline]
12.
Oren M: Regulation of the p53 tumor suppressor protein. J Biol Chem 274: 36031-36034, 1999 13. Fakharzadeh SS, Trusko SP, George DL: Tumorigenic potential associated with enhanced expression of gene that is amplified in a mouse tumor cell line. EMBO J 10: 1565-1595, 1991[Medline]
14.
Jones SN, Hancock AR, Vogel H, et al: Overexpression of Mdm2 in mice reveals a p53-independent role for Mdm2 in tumorigenesis. Proc Natl Acad Sci USA 95: 15608-15612, 1998 15. Bureik M, Jungbluth A, Drescher R, et al: Human p53 restores DNA synthesis control in fission yeast. Biol Chem 378: 1361-1371, 1997[Medline] 16. Ozdemir E, Kakehi Y, Yoshida O: p21 (WAF-1/CIP-1), a downstream regulator of functional p53 loss, in transitional cell carcinoma of urothelium. Eur Urol 38: 230-234, 2000[CrossRef][Medline] 17. Cao L, Yao Y, Lee V, et al: Epidermal growth factor induces cell cycle arrest and apoptosis of squamous carcinoma cells through reduction of cell adhesion. J Cell Biochem 77: 569-583, 2000[CrossRef][Medline]
18.
Pfister DG, Harrison LB, Strong EW, et al: Organ-function preservation in advanced oropharynx cancer: Results with induction chemotherapy and radiation. J Clin Oncol 13: 671-680, 1995
19.
Pfister DG, Bajorin D, Motzer R, et al: Cisplatin, fluorouracil, and leucovorin: Increased toxicity without improved response in squamous cell head and neck cancer. Arch Otolaryngol Head Neck Surg 120: 89-95, 1994 20. American Joint Committee on Cancer: TNM staging, in Beahrs O, Henson D, Hutter R, et al (eds): Manual for Staging of Cancer, ed 3. Philadelphia, PA, JB Lippincott, 1988, pp 33-45 21. Harrison LB, Pfister DG, Fass DE, et al: Concomitant chemotherapy-radiation therapy followed by hyperfractionated radiation therapy for advanced unresectable head and neck cancer. Int J Radiat Oncol Biol Phys 21: 703-708, 1991[Medline] 22. Miller AB, Hoogstraten B, Staquet M, et al: Reporting results of cancer treatment. Cancer 47: 207-214, 1981[CrossRef][Medline]
23.
Kris MG, Gralla RJ, Kelsen DP, et al: Trial of vindesine plus mitomycin in stage-3 non-small cell lung cancer, an active regimen for outpatient treatment. Chest 87: 368-372, 1985 24. Cordon-Cardo C, Dalbagni G, Saez GT, et al: p53 mutations in human bladder cancer: Genotypic versus phenotypic patterns. Cancer 56: 347-353, 1994 25. Dalbagni G, Presti JC Jr, Reuter VE, et al: Molecular genetic alterations of chromosome 17 and p53 nuclear overexpression in human bladder cancer. Diagn Mol Pathol 2: 4-13, 1993[Medline]
26.
Osman I, Drobnjak M, Fazzari M, et al: Inactivation of the p53 pathway in prostate cancer: Impact on tumor progression. Clin Cancer Res 5: 2082-2088, 1999 27. Osman I, Scher HI, Zhang ZF, et al: Alterations affecting the p53 control pathway in bilharzial-related bladder cancer. Clin Cancer Res 3: 531-536, 1997[Abstract]
28.
Cordon-Cardo C, Latres E, Drobnjak M, et al: Molecular abnormalities of mdm2 and p53 genes in adult soft tissue sarcomas. Cancer Res 54: 794-799, 1994 29. Agresti A: Categorical Data Analysis. New York, NY, Wiley, 1966, pp 39-44 30. Kaplan E, Meyer P: Nonparametric estimation for incomplete observation. J Am Stat Assoc 53: 457-481, 1958[CrossRef] 31. Cox DR: Regression models and life-tables. J Royal Stat Soc 34: 187-220, 1972 32. Schemper M, Smith TL: A note on quantifying follow-up in studies of failure time. Control Clin Trials 17: 343-346, 1996[CrossRef][Medline]
33.
Temam S, Flahault A, Perie S, et al: p53 gene status as a predictor of tumor response to induction chemotherapy of patients with locoregionally advanced squamous cell carcinomas of the head and neck. J Clin Oncol 18: 385-394, 2000 34. Warnakulasuriya S, Jia C, Johnson N, et al: p53 and P-glycoprotein expression are significant prognostic markers in advanced head and neck cancer treated with chemo/radiotherapy. J Pathol 191: 33-38, 2000[CrossRef][Medline] 35. Bradford CR, Zhu S, Wolf GT, et al: Overexpression of p53 predicts organ preservation using induction chemotherapy and radiation in patients with advanced laryngeal cancer: Department of Veterans Affairs Laryngeal Cancer Study Group. Otolaryngol Head Neck Surg 113: 408-412, 1995[CrossRef][Medline]
36.
Bradford CR, Zhu S, Poore J, et al: p53 mutation as a prognostic marker in advanced laryngeal carcinoma: Department of Veterans Affairs Laryngeal Cancer Cooperative Study Group. Arch Otolaryngol Head Neck Surg 123: 605-609, 1997 37. Capaccio P, Carboni N, Pignataro L, et al: Cyclin D1, p53, mdm2, and Ki67 protein expression in preneoplastic lesions of the larynx. J Chemother 9: 113-114, 1997[Medline] 38. Pignataro L, Capaccio P, Pruneri G, et al: The predictive value of p53, MDM-2, cyclin D1 and Ki67 in the progression from low-grade dysplasia towards carcinoma of the larynx. J Laryngol Otol 112: 455-459, 1998[Medline] 39. Sato N, Mizumoto K, Maehara N, et al: Enhancement of drug-induced apoptosis by antisense oligodeoxynucleotides targeted against Mdm2 and p21WAF1/CIP1. Anticancer Res 20: 837-842, 2000[Medline] 40. Wang H, Zeng X, Oliver P, et al: MDM2 oncogene as a target for cancer therapy: An antisense approach. Int J Oncol 15: 653-660, 1999[Medline] 41. Maiorano E, Botticella MA, Marzullo A, et al: Expression of ER-D5 and EGFr in laryngeal carcinoma and pre-malignant epithelium. Acta Otolaryngol Suppl 527: 95-99, 1997[Medline] 42. Dellacono FR, Spiro J, Eisma R, et al: Expression of basic fibroblast growth factor and its receptors by head and neck squamous carcinoma tumor and vascular endothelial cells. Am J Surg 174: 540-544, 1997[CrossRef][Medline] 43. Erber R, Klein W, Andl T, et al: Aberrant p21 (CIP1/WAF1) protein accumulation in head-and-neck cancer. Int J Cancer 74: 383-389, 1997[CrossRef][Medline] 44. Van Oijen MG, Tilanus MG, Medema RH, et al: Expression of p21 (Waf1/Cip1) in head and neck cancer in relation to proliferation, differentiation, p53 status and cyclin D1 expression. J Oral Pathol Med 27: 367-375, 1998[Medline]
45.
Gohring UJ, Bersch A, Becker M, et al: p21 (waf) correlates with DNA replication but not with prognosis in invasive breast cancer. J Clin Pathol 54: 866-870, 2001 46. Sarkar FH, Li Y, Sakr WA, et al: Relationship of p21 (WAF1) expression with disease-free survival and biochemical recurrence in prostate adenocarcinomas (Pca). Prostate 40: 256-260, 1999[CrossRef][Medline]
47.
Jacobs TW, Prioleau JE, Stillman IE, et al: Loss of tumor marker-immunostaining intensity on stored paraffin slides of breast cancer. J Natl Cancer Inst 88: 1054-1059, 1996 Submitted June 27, 2001; accepted March 21, 2002.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test unless there was concern regarding an inadequate number of observations, in which case a Fisher’s exact test was used.
