|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Prognostic Value of p53 for Local Failure in Mastectomy-Treated Breast Cancer PatientsFrom the Departments of Radiology (Division of Radiation Oncology), Medical Oncology, and Pathology, University of Texas Health Science Center at San Antonio, San Antonio, and Baylor Methodist Breast Center at Baylor College of Medicine, Houston, TX. Address reprint requests to Richard C. Zellars, MD, Division of Radiation Oncology, Department of Radiology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr, San Antonio, TX 78284-7800.
PURPOSE: The loss of p53 function is a recognized adverse prognostic factor in invasive breast cancer. Several studies have shown a relationship between the nuclear accumulation of p53 protein (a surrogate marker of p53 inactivation) and poor disease-free and overall survival. In general, however, these studies did not report the prognostic value of p53 for local failure, which we have therefore assessed retrospectively here. MATERIALS AND METHODS: Accumulation of p53 protein was evaluated by immunohistochemistry in 1,530 mastectomy-treated breast cancer patients (259 radiation therapy [RT]– and 1,271 mastectomy only [No RT]–treated patients). Statistical comparisons were made between p53 protein accumulation, estrogen/progesterone receptors, nodal status, tumor size, and local failure rate (LFR). Local failure was defined as tumor recurrence involving the chest wall and/or the ipsilateral supraclavicular/axillary lymph nodes. The median follow-up period was 62 months. RESULTS: In the No RT group, the LFR was 9.1% and 16.5% in p53-negative and p53-positive patients, respectively (P < .001). Multivariate analysis revealed that p53 protein accumulation was significantly associated with an increased risk of local relapse (relative risk [RR], 1.7; 95% confidence interval [CI], 1.2 to 2.4). Nodal status and tumor size were also significant factors. In the RT group, the LFR was 9.3% and 21.5% in p53-negative and p53-positive patients, respectively (P = .009). Multivariate analysis revealed that p53 protein accumulation was significantly associated with an increased risk of local relapse (RR, 2.5; 95% CI, 1.1 to 5.7), as was nodal status. CONCLUSION: Nuclear accumulation of p53 protein is independently associated with a significantly increased local failure rate in breast cancer patients treated with mastectomy, with or without radiation.
TWENTY PERCENT TO 30% of node-positive breast cancer patients will develop a local-regional recurrence after being treated by mastectomy with or without chemotherapy.1-3 Ultimately, 50% of these recurrences will be controlled with local modalities; however, the remaining 50% will die with uncontrolled local-regional disease.4,5 In the hope of preventing such events, early, randomized clinical trials evaluated the role of postmastectomy radiation therapy (RT). Although these trials consistently demonstrated a significant improvement in local control, evidence of survival benefit was lacking.6 As a result, many considered RT ineffectual in improving survival, and some suggested that postmastectomy RT may be detrimental.6 The debate over this topic continues7-9 with the recent publication of two randomized prospective trials in premenopausal node-positive women.1,2 In these trials, the two arms of each study differed only with respect to the addition of postmastectomy radiation. The results showed that there was not only an improvement in local control but also a significant improvement in survival in the patients who received adjuvant RT. Although confirmatory trials would provide greater certainty on this issue, it is reasonable to expect that a population exists with sufficient risk of local-regional failure that adjuvant RT will not only improve local-regional control but also possibly improve survival. However, because the personal, social, and financial cost of postmastectomy RT can be substantial, it would be helpful to identify breast cancer patients at greatest risk of local-regional recurrence and therefore most likely to benefit from postmastectomy RT. Known predictors of an increased risk of local-regional failure are lymph node positivity, tumor size, grade of anaplasia, extent of surgery, and possibly lymph-vascular invasion.1,7,10-13 Because some of these prognostic factors parallel those of distant failure, we hypothesized that some other factors prognostic of distant failure may be associated with local-regional recurrence. Some well-recognized prognostic factors of distant failure are estrogen/progesterone receptor status, S-phase fraction, DNA ploidy, and, more recently, p53 accumulation. Although many of these prognostic factors have been studied for some time, the tumor suppressor gene p53 is relatively new and is currently the focus of much attention in breast cancer research. Research has shown that altered function of this gene is associated with decreased disease-free and overall survival.14-17 Interestingly, considering the extent to which p53 has been evaluated for prognostic purposes, there is little in the literature concerning its prognostic value for local-regional failure in mastectomy-treated breast cancer patients. Accordingly, to identify breast cancer patients who have the highest risk of local relapse and would most likely benefit from adjuvant RT, we evaluated the local failure prognostic value of p53.
Patient Data The following patient information is from the San Antonio breast cancer database. A team of experienced data managers regularly visits each physician’s office or clinic to obtain follow-up information. The primary source of information is the patient chart or clinic record. Hospital tumor registries are a second source of information. Once the updated information is entered into the data files, a computer-generated report is produced and sent back to the medical oncologist, surgeon, or tumor registry. This report lists each patient and selected demographic and patient status variables and requests that errors or omissions be corrected and returned. For the purposes of quality assurance, audits are performed. Patients’ charts are reviewed and compared with the database. No significant differences have been found between randomly chosen patient charts and the corresponding database printouts. Information in the database has been independently verified on a sampling basis on multiple occasions over time The records of 1,530 mastectomy-treated breast cancer patients from the San Antonio database were reviewed. The patient population was subdivided into two groups: those who received mastectomy followed by adjuvant RT and those who received mastectomy only (No RT). Data concerning the following prognostic factors were collected and analyzed: tumor size, axillary lymph node status, estrogen/progesterone receptor status (estrogen receptor–negative, < 3 fmol/mg; progesterone receptor–negative, < 5 fmol/mg),18,19 DNA ploidy and S-phase fraction (determined by flow cytometry),20 type of systemic therapy, local failure rate, and p53 protein accumulation. Local failure was defined as a recurrence on the chest wall or in the axillary or supraclavicular nodal regions. Local recurrences as a first site of recurrence were included in the analysis. In all local-regional failure analyses, simultaneous local-regional and distant failure was counted as an event (ie, a local-regional failure). In the competing-risks analyses, distant failure alone as a first failure was a competing risk, preventing further observation of local-regional failure. In the standard survival analysis, distant failure alone as a first failure was treated as a censoring event. Median follow-up for local recurrence was 62 months, with 80% of the patients followed up for more than 2 years (range, 1 to 228 months).
Assessment of p53 Function
Statistical Analysis
Patient and Tumor Characteristics A total of 1,530 patients were included in these analyses, 259 (17%) of whom received adjuvant RT and 1,271 (83%) of whom did not (Table 1). Because the patient population was not randomized, there were a number of differences between the study groups. Patients who received RT had more positive axillary lymph nodes (P < .001), larger tumors (P < .001), and higher S-phase fractions (P = .04) and were more often estrogen receptor–negative (P < .001) and progesterone receptor–negative (P < .001). These patients more frequently received adjuvant chemotherapy (P < .001) and adjuvant endocrine therapy (P = .04). However, there were no significant differences in age at diagnosis, DNA ploidy status, or p53 status between patients who did or did not receive adjuvant RT.
All local failure rates were determined by the cumulative incidence method (unless otherwise stated) to account for competing risks. However, for purposes of comparison, the local failure rate, as determined by the Kaplan-Meier method, is also presented in Tables 2 and 3. The 5-year actuarial local failure rates were 15.9% and 13.0%, respectively, for patients who did or did not receive adjuvant RT after surgery. Five-year actuarial local failure rates are shown in Tables 2 and 3 for each of the factors included in these analyses. The higher local failure rate for patients who received radiation was probably due to biologically more aggressive tumors in this group, as manifested by a greater preponderance of poor prognostic factors, because this was not a randomized study. Table 4 further illustrates this point. It shows the distribution of patients and their local control status by tumor stage and the number of involved nodes.
For the 1,271 patients who did not receive adjuvant RT, the following factors were associated with higher local failure rates in univariate analyses: positive axillary lymph nodes (P < .001), large (> 2 cm) tumors (P < .001), ER negativity (P = .04), high S-phase fractions (P < .001), administration of adjuvant chemotherapy (P = .003), and positive p53 status (P < .001) (Table 2). The apparent negative effect of adjuvant chemotherapy on local control was due to the greater likelihood of poor-risk patients (eg, those with more involved nodes) being given adjuvant treatment, so that any benefit of this treatment was more than offset by the initially higher risk. No significant relationships were observed with regard to age at diagnosis, DNA ploidy, progesterone receptor status, or adjuvant endocrine therapy (Table 2). Risk of local relapse in the mastectomy-only patients was nearly twice as high for p53-positive patients compared with p53-negative patients (16.5% v 9.1%, respectively; P < .001) (Fig 1A).
For the 259 patients who did receive adjuvant RT, high S-phase fraction (P < .001) and positive p53 status (P = .009) were associated with higher local failure rates (Table 3). No significant relationships were observed with positive axillary lymph nodes, tumor size, age at diagnosis, estrogen or progesterone receptor status, or adjuvant endocrine therapy or chemotherapy (Table 3). However, the numbers of patients and local failures were relatively small in some subsets, resulting in decreased statistical power to detect significant differences. Nevertheless, the risk of local failure was substantially higher for p53-positive patients than for p53-negative patients (21.5% v 9.3%, respectively; P = .04) (Fig 1B). Although accounting for the competing risk of distant failure results in lower estimates of local failure rates compared with the Kaplan-Meier estimate, the relative differences among groups were virtually identical, regardless of the method of estimation. That is, accounting for the competing risk of distant failure does not change the direction or relative magnitude of the effect in the univariate setting. Therefore, we used Cox proportional hazards regression, which censors competing risks, to determine which factors provide independent prognostic information for predicting local failure. Both forward stepwise variable selection and backward elimination were used to construct the most parsimonious models. Separate multivariate analyses in patients who did or did not receive adjuvant RT confirmed that p53 status was a strong independent predictor of local failure (Table 5). The factors that were also significant independent predictors of local failure in the No RT group were involved nodes (P < .001), tumor size (P = .005), lack of adjuvant endocrine therapy (P = .003), and high S-phase fraction (P = .02). The sample size in this analysis was decreased to the 889 patients with complete data for all factors. Nevertheless, the multivariate relative risk of local failure in this subgroup of p53-positive tumors was 1.7 (P = .006).
The increased local failure risk of larger tumors and involved nodes was expected. Adjuvant endocrine therapy was shown to be beneficial in this analysis, whereas it did not seem to be beneficial in the univariate analysis. This finding is probably because the benefit of endocrine therapy was counterbalanced by the greater likelihood of its being used in poor-risk patients. In patients who received adjuvant RT, involved axillary lymph nodes (P = .02), p53 accumulation (P = .02), and lack of adjuvant chemotherapy (P = .01) were independent predictors of local failure. In the RT group, the multivariate relative risk for local failure in p53-positive tumors was 2.5 (P = .02). Lack of chemotherapy was associated with an increased risk of local failure in the multivariate analysis (relative risk, 2.5), although no similar relationship was identified in the univariate analysis. As with endocrine therapy in the No RT group, the patients most likely to receive chemotherapy were also those at highest risk. Therefore, the potential benefit of chemotherapy was counterbalanced by more biologically aggressive tumors with intrinsically higher relapse rates. Once confounding variables were adjusted for, the benefit of chemotherapy on local control became apparent.
Given the percentage of women who will have a local-regional recurrence after mastectomy and the local control and survival benefit of adjuvant RT in select patients, it is important to identify novel molecular markers predictive of local failure. These new predictors will aid in the identification of women most likely to benefit from additional local therapy. In this study, we assessed traditional indicators of distant disease-free survival for local failure prognostic value. We found that a promising new predictor of distant disease-free survival also has local failure prognostic value. That putative predictor is the tumor suppressor gene p53. The p53 tumor suppressor gene is an important negative regulator of cellular proliferation. The p53 gene product is induced in response to DNA damage. Evidence shows that the expression of p53 protein leads to cell cycle arrest in G1 and in some cases to programmed cell death, or apoptosis.24-26 Arrest of the cell cycle, presumably to allow for DNA repair, as well as apoptosis, prevents the replication of damaged DNA and the subsequent propagation of genetic defects. This DNA replication control mechanism maintains fidelity in chromosomal transmission and has earned p53 the title "guardian of the genome."26 Alterations in p53 lead to loss of its cell growth–regulatory function, resulting in accelerated cell growth and increased DNA mutation frequency. The unchecked propagation of these mutations is thought to contribute to the development of human cancers. The hypothesis that altered p53 is intimately involved in the genesis of tumors is supported by the fact that it is the most commonly found genetic alteration in human cancers. With regard to breast cancer, altered p53 has been identified in 15% of in situ and 50% of invasive disease. Because loss of p53 function leads to higher proliferative and lower apoptosis rates, altered p53 should therefore be associated with a worse clinical outcome. Indeed, many studies in breast cancer have shown a correlation between p53 protein accumulation and a poor clinical outcome.14-17 In general, however, these studies did not report the prognostic value of p53 for local failure, which we have retrospectively assessed here. This is the largest study of its kind in the literature, and it demonstrates that the nuclear accumulation of p53 is a significant adverse prognostic factor for local failure in mastectomy-treated invasive breast cancer. This was also true when the study population was divided according to the presence or absence of adjuvant RT. We present the results of two statistical methodologies for analyzing local failure. The first, the cause-specific cumulative incidence method, was used because it accounts for competing risks in this population. The second, the more common but less accurate Kaplan-Meier method, was used to make comparisons of our results with other published literature easier. Multivariate analysis revealed that p53 remains a significant independent prognostic factor. The sample size of this study strengthens the reliability of these findings. We analyzed our patients who received adjuvant RT separately for two reasons. First, the RT group was significantly different from the No RT group in several characteristics. Second, there is substantial controversy in the literature about the effect of p53 on radiation sensitivity.27-29 For instance, in a transplanted fibrosarcoma model using tumors that differed only with respect to functional p53 status, mutant p53 tumors were more radioresistant than their wild-type counterparts.30 Conversely, other laboratory studies have found no correlation between p53 status and radiation sensitivity.31 However, in a recent clinical publication, p53 protein accumulation was associated with increased resistance to RT in head and neck cancer patients.32 This finding is consistent with other clinical studies that have shown an increased resistance to chemotherapeutic agents associated with altered p53 function.33 It is not possible with the present data to definitively assess the influence of p53 on radiation sensitivity. However, the association of p53 protein accumulation with a similar magnitude of change in local failure (approximately double) in both the No RT and RT groups does not support the hypothesis of differential radiation sensitivity. Interestingly, a comparable yet relatively small study of conservatively treated breast cancer patients by Silvestrini et al34 supports the hypothesis of differential radiation sensitivity. In that study, p53 had local failure–predictive value in patients treated with lumpectomy alone but did not have local failure–predictive value in patients treated with lumpectomy and radiation. A limitation to this analysis may be the retrospective design of this study. Because therapy was not standardized, treatment preferences could have biased the results. In particular, details of the systemic adjuvant therapies selected were not available. Systemic therapy has been demonstrated to have an effect on local control in some studies.35 In the multivariate analysis, however, p53 status remained an independent prognostic factor after the effect of systemic adjuvant therapy had been considered. In summary, there is a significant direct correlation between local failure and altered p53 function in mastectomy-treated breast cancer patients. p53 has prognostic significance independent of traditional prognostic factors. Perhaps as more prognostic indicators become available, this information will help to identify those patients who are at greater risk of local failure and therefore may warrant more aggressive local management.
Supported in part by Specialized Program of Research Excellence grant no. P50-CA-58183, Medical Oncology Program Project grant no. CA-30195, and San Antonio Cancer Institute grant no. P30-CA-54174 from the National Institutes of Health, Bethesda, MD. We thank J. Dignam and J. Bryant for kindly providing the cumulative incidence SAS macros and Carole M. Elledge for editorial assistance in the preparation of the manuscript.
1. Overgaard M, Hansen P, Overgaard J, et al: Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med 337:949-955, 1998
2.
Ragaz J, Stewart JM, Nhu L, et al: Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 337:956-962, 1998
3.
Early Breast Cancer Trialists’ Collaborative Group: Effects of radiotherapy and surgery in early breast cancer: An overview of the randomized trials. N Engl J Med 333:1444-1455, 1995 4. Schwaibold F, Fowble B, Lawrence J, et al: The results of radiation therapy for isolated local regional recurrence after mastectomy. Int J Radiat Oncol Biol Phys 21:299-310, 1991[Medline] 5. Aberizk W, Silver B, Henderson C, et al: The use of radiotherapy for treatment of isolated locoregional recurrence of breast carcinoma after mastectomy. Cancer 58:1214-1218, 1986[Medline] 6. Cuzick J, Stewart H, Peto R, et al: Overview of randomized trials of postoperative adjuvant radiotherapy in breast cancer. Cancer Treat Rep 71:15-25, 1987[Medline] 7. Haagensen CD, Bodian C: A personal experience with Halsted’s radical mastectomy. Surg 199:143-150, 1984 8. Pierce LJ, Lichter AS: Defining the role of postmastectomy radiotherapy: The new evidence. Oncology 10:991-1001, 1996[Medline] 9. Pierce L, Glatstein E: Postmastectomy radiotherapy in the management of operable breast cancer. Cancer 74:477-485, 1994 (suppl)[Medline] 10. Valagussa P, Bonadonna G, Veronesi U, et al: Patterns of relapse and survival following radical mastectomy. Cancer 41:1170-1178, 1978[Medline] 11. Donegan WL, Perz-Mesa CM, Watson F: Biostatistical study of locally recurrent breast carcinoma. Gynecol Obstet 122:529-540, 1996 12. Mentzer SJ, Osteen RT, Wilson RE: Local recurrence and the deep resection margin in carcinoma of the breast. Surg Gynecol Obstet 163:513-517, 1986[Medline] 13. Houghton J, Baum M, Haybittle JL: Role of radiotherapy following total mastectomy in patients with early breast cancer: The Closed Trials Working Party of the CRC Breast Cancer Trials Group. World J Surg 18:117-122, 1994[Medline]
14.
Allred DC, Clark GM, Elledge R, et al: Association of p53 protein expression with tumor cell proliferation rate and clinical outcome in node-negative breast cancer. J Natl Cancer Inst 85:200-206, 1993 15. Elledge RM, Fuqua SAW, Clark GM, et al: The role and prognostic significance of p53 gene alterations in breast cancer. Cancer Res Treat 27:95-102, 1993 16. Friedrichs K, Gluba S, Eidtmann H, et al: Over-expression of p53 and prognosis in breast cancer. Cancer 72:3641-3647, 1993[Medline]
17.
Kovach JS, Hartmann A, Blaszyk H, et al: Mutation detection by highly sensitive methods indicates that p53 genetic mutations in breast cancer can have important prognostic value. U S A 93:1093-1096, 1996
18.
Powell B, Garola RE, Chamness GC, et al: Measurement of progesterone receptor in human breast cancer. Cancer Res 39:1678-1682, 1979
19.
McGuire WL, De La Garza M, Chamness GC: Evaluation of estrogen receptor assays in human breast cancer tissue. Cancer Res 37:637-639, 1977 20. Dressler LG, Seamer LC, Owens MA, et al: DNA flow cytometry and prognostic factors in 1331 frozen breast cancer specimens. Cancer 61:420-427, 1988[Medline] 21. Gelman R, Gelber R, Henderson IC, et al: Improved methodology for analyzing local and distant recurrence. J Clin Oncol 8:548-555, 1990[Abstract] 22. Gaynor JJ, Feuer EJ, Tan CC, et al: On the use of cause-specific failure and conditional failure probabilities: Example from clinical oncology data. JASA 88:400-409, 1993 23. Pepe MS, Mori M: Marginal or conditional probability curves in summarizing competing risks failure time data? Stat Med 12:737-751, 1993[Medline] 24. Ruley HE: p53 and response to chemotherapy and radiotherapy. Oncol Update 11:7-19, 1997 25. Kastan MB: The p53 tumor suppressor gene: A multifaceted cancer threat. Adv Oncol 12:3-7, 1996 26. Lane DP: p53, guardian of the genome. Nature 358:15-16, 1992[Medline]
27.
McIlwrath AJ, Vasey PA, Ross GM, et al: Cell cycle arrests and radiosensitivity of human tumor cell lines: Dependence on wild-type p53 for radiosensitivity. Cancer Res 54:3718-3722, 1994 28. Kubiczek EK, Yin J, Lin K, et al: p53 mutational status and survival of human breast cancer MCF-7 cell variants after exposure to x-rays of fission neutrons. Radiat Res 142:256-262, 1995[Medline]
29.
Lee JM, Bernstein A: p53 mutations increase resistance to ionizing radiation. Acad Sci U S A 90:5742-5746, 1993
30.
Lowe SW, Bodis S, McClatchey A, et al: p53 status and the efficacy of cancer therapy in vivo. Science 266:807-810, 1994
31.
Brachman DG, Beckett M, Graves D, et al: p53 mutation does not correlate with radiosensitivity in 24 head and neck cancer cell lines. Cancer Res 53:3667-3669, 1993
32.
Raybaud-Diogene H, Fortin A, Morency R, et al: Markers of radioresistance in squamous cell carcinomas of the head and neck: A clinicopathologic and immunohistochemical study. J Clin Oncol 15:1030-1038, 1997 33. Lowe SW, Ruley HE, Jacks T, et al: p53 dependent apoptosis modulates the cytotoxicity of anti-cancer agents. Cell 74:957-967, 1993[Medline]
34.
Silvestrini R, Veronisi S, Benini E, et al: Expression of p53, glutathione S-transferase-pi, and Bcl-2 proteins and benefit from adjuvant radiotherapy in breast cancer. J Natl Cancer Inst 89:639-645, 1997 35. Fisher B, Redmond C, Poisson R, et al: Eight-year results of a randomized clinical trial comparing total mastectomy and lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 320:822-828, 1989[Abstract] Submitted August 26, 1999; accepted January 10, 2000.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 tests. Distant failure before local failure is a competing risk and a potential source of bias if ignored.
