|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Locoregional Failure 10 Years After Mastectomy and Adjuvant Chemotherapy With or Without Tamoxifen Without Irradiation: Experience of the Eastern Cooperative Oncology GroupFrom the Joint Center for Radiation Therapy, Harvard Medical School, and Beth Israel Deaconess Medical Center; Harvard School of Public Health and Dana-Farber Cancer Institute, Boston, MA; Johns Hopkins University School of Medicine and Johns Hopkins Oncology Center, Baltimore, MD; Fox Chase Cancer Center; University of Pennsylvania School of Medicine and Hospital of the University of Pennsylvania, Philadelphia, PA; Brown University School of Medicine and Roger Williams Medical Center, Providence, RI; University of Pretoria School of Medicine, Pretoria, Republic of South Africa; Rush-Presbyterian Medical School and Illinois Masonic Medical Center, Chicago IL; and AMC Cancer Research Center, Denver, CO. Address reprint requests to Abram Recht, MD, Department of Radiation Oncology, Beth Israel Deaconess Medical Center, East Campus, Finard Building B25, 330 Brookline Ave, Boston, MA 02215; email arecht{at}caregroup.harvard.edu
PURPOSE: To assess patterns of failure and how selected prognostic and treatment factors affect the risks of locoregional failure (LRF) after mastectomy in breast cancer patients with histologically involved axillary nodes treated with chemotherapy with or without tamoxifen without irradiation. PATIENTS AND METHODS: The study population consisted of 2,016 patients entered onto four randomized trials conducted by the Eastern Cooperative Oncology Group. The median follow-up time for patients without recurrence was 12.1 years (range, 0.07 to 19.1 years). RESULTS: A total of 1,099 patients (55%) experienced disease recurrence. The first sites of failure were as follows: isolated LRF, 254 (13%); LRF with simultaneous distant failure (DF), 166 (8%); and distant only, 679 (34%). The risk of LRF with or without simultaneous DF at 10 years was 12.9% in patients with one to three positive nodes and 28.7% for patients with four or more positive nodes. Multivariate analysis showed that increasing tumor size, increasing numbers of involved nodes, negative estrogen receptor protein status, and decreasing number of nodes examined were significant for increasing the rate of LRF with or without simultaneous DF. CONCLUSION: LRF after mastectomy is a substantial clinical problem, despite the use of chemotherapy with or without tamoxifen. Prospective randomized trials will be necessary to estimate accurately the potential disease-free and overall survival benefits of postmastectomy radiotherapy for patients in particular prognostic subgroups treated with presently used and future systemic therapy regimens.
IRRADIATION CLEARLY reduces the risk of locoregional failure (LRF) for patients with invasive breast cancer treated with mastectomy. However, for many years it has been controversial whether this reduction also results in decreased risks of distant failure (DF) and, ultimately, death due to cancer for patients who receive systemic therapy.1-6 The two largest trials conducted to answer this question among node-positive patients treated with chemotherapy showed that radiation therapy after modified radical mastectomy not only reduced LRF rates but also significantly improved disease-free and overall7 or cause-specific8 survival rates. Nonetheless, the applicability of these two trials to making treatment decisions for patients treated today has been questioned. There is only limited information available from either of these trials or other studies on how commonly measured prognostic factors affect the risk of LRF and, hence, the potential benefits of irradiation in particular patient subgroups. Therefore, we examined the long-term risk of LRF after mastectomy in node-positive patients entered onto randomized trials conducted by the Eastern Cooperative Oncology Group (ECOG). Patients received chemotherapy or chemohormonotherapy but not irradiation in relation to a number of prognostic and treatment variables. A previous report analyzed LRF at 3 years in two of these trials (E5177 and E6177).9 Here we present updated results for the patients included in this prior study as well as patients treated on two additional trials (E4181 and E5181).
Designs of Included Trials The current analysis included patients with histologically involved axillary lymph nodes treated with modified radical mastectomy or radical mastectomy who were entered onto four randomized prospective trials conducted by ECOG. Adjuvant irradiation was not permitted in any of these trials. Patients with tumors of any size were entered onto the two previous trials; patients with tumors larger than 5 cm were excluded from the two later ones. Patients with clinical signs or certain pathologic features of locally advanced disease (skin ulceration or satellite nodules, skin infiltration of > 2 cm in diameter, muscle infiltration or fixation, peau d'orange of more than one third of the breast, dermal lymphatic involvement, axillary nodal fixation, arm edema, or supraclavicular or internal mammary adenopathy) or metastatic disease were excluded. For entry onto each trial, patients underwent measurement of the estrogen-receptor protein (ERP), which could be either positive (defined as  10 fmol/mg protein) or negative. Postmenopausal status (for patients who had not undergone hysterectomy or oophorectomy) was defined as having no menstrual periods for at least 12 months before diagnosis. Brief summaries of the included trials are listed in Table 1. From 1978 to 1982, trial E5177 recruited 553 assessable premenopausal patients to one of three treatment arms: cyclophosphamide 100 mg/m2 orally on days 1 to 14, methotrexate 40 mg/m2 intravenously on days 1 and 8, and fluorouracil 600 mg/m2 intravenously on days 1 and 8 administered for 12 cycles at 28-day intervals (CMF12); CMF12 with the addition of prednisone 40 mg/m2 orally on days 1 through 14 for 12 cycles (CMFP12); or CMFP12 with the addition of tamoxifen 10 mg orally twice daily continuously during the entire treatment period for 12 cycles (CMFPT12).10
Trial E6177, also conducted from 1978 to 1982, included 223 assessable postmenopausal patients (aged Trial E4181, conducted from 1982 to 1986, included 802 assessable postmenopausal patients (without age limit) in three arms: CMFPT for four cycles (CMFPT4), CMFPT for 12 cycles, or CMFPT for 12 cycles followed by a further 4 years of tamoxifen (CMFPT60). One patient included in a previous report of this trial12 was later found to be ineligible and was therefore excluded from our study. After 1986, disease-free patients on this third arm were eligible to be rerandomized to stop tamoxifen treatment after a total duration of 5 years or to continue tamoxifen treatment indefinitely. Trial E5181, conducted from 1982 to 1987, included 533 assessable premenopausal patients in two arms: CMFPT for 12 cycles (CMFPT12/60) or 12 total cycles of two different regimens delivered in alternation ("Alternating-T12/60". For the odd-numbered cycles, CMFPT was administered with the omission of prednisone after cycle 3 and the addition of fluoxymesterone (Halotestin; Pharmacia & Upjohn, Bridgewater, NJ) 10 mg orally twice daily; for the even-numbered cycles, patients were administered 21-day cycles of vinblastine 4.5 mg/m2, doxorubicin 45 mg/m2, and thiotepa 12 mg/m2, all given intravenously on day 1, with Halotestin and tamoxifen also given as described above [VATHT]).13 This trial also incorporated a second randomization after chemotherapy was completed (step 2) to continue tamoxifen for a total of 5 years (TAM60) or to stop it at the end of chemotherapy (TAM12); patients who continued tamoxifen treatment for 5 years were also subsequently eligible (if disease-free) for another randomization at 5 years to continue tamoxifen treatment indefinitely beyond 5 years or stop it. Further details for each trial separately—including definitions of menopausal status for patients undergoing reproductive-organ surgery, prerandomization evaluation, other exclusion criteria, total accrual and reasons patients were not assessable, stratification factors, patient and tumor characteristics, chemotherapy and hormonal regimens, dose-modification rules, and follow-up techniques—may be found in the publications referenced above.
Characteristics of the Current Study Population Patient and tumor characteristics for the current study group are described in Table 2. Pathologic tumor size and stage were determined from the gross examination, as obtained from the original pathology report. The number of examined and involved nodes was also obtained from the original pathology report. The simultaneous distribution of tumor sizes (using 1-cm intervals) and the number of involved nodes are listed in Table 3.
A number of pathologic characteristics (in addition to gross tumor size and axillary nodal involvement) were recorded from the original report or central slide review for patients entered onto trials E5177 and E6177 only. Given the limited number of patients for whom such pathologic data were available (only 28% of the entire study population), the lack of standardized criteria for processing specimens and reporting them by the original institutional pathologists, and potential changes in pathologic definition and assessment that may have taken place since these materials were initially reviewed, we have elected not to analyze the impact of these pathologic factors for this report.
End Points
Follow-Up and Statistical Analysis
Five- and 10-year actuarial rates (and the SEs of these rates) were calculated by the cumulative incidence method.14 Comparison of cumulative incidences was performed using the method of Gray.15 Comparisons of the patterns of failure in different groups and correlations between prognostic factors were examined with the Because the goal of the current study was to assess the patterns of failure and incidences of LRF in specific subgroups in this population, overall survival results are not reported here.
Results in the Study Population as a Whole The patterns of failure (ie, the proportion of failures in each of the three aforementioned categories) did not vary in a statistically significant manner among the four trials or between different arms within individual trials (data not shown). Hence, patients from these trials were combined for further analysis. Of the 2,016 study patients, 1,099 (55%) experienced disease recurrence. The first sites of failure (as a proportion of the entire population) were as follows: isolated RF, 254 (13%); LRF with simultaneous DF, 166 (8%); and DF only, 679 (34%; this group included six cases without specific failure sites reported). The 5-year cumulative incidences of isolated LRF, LRF with simultaneous DF, LRF with or without DF, and DF only were 10.1% (SE, 0.7%), 6.3% (SE, 0.6%), 16.4% (SE, 0.8%), and 24.5% (SE, 1.0%), respectively. The 10-year rates were 12.6% (SE, 0.8%), 8.0% (SE, 0.6%), 20.7% (SE, 0.9%), and 32.3% (SE, 1.1%), respectively. The median time to development of an isolated LRF was 2.6 years, compared with 2.8 years for development of LRF with simultaneous DF and 3.0 years for development of DF only. Approximately 80% of the 10-year incidences of isolated LRF and LRF with or without DF occurred by 5 years, similar to the proportion of patients who developed DF only by 5 years.
LRF Rates in Relation to Prognostic and Treatment Variables
On univariate analysis, ERP status had a statistically significant effect on LRF with or without DF (Table 4) (rates not shown). However, this effect might be related to both prognostic and treatment effects, ie, not all trial arms contained tamoxifen, and tamoxifen was used for varying durations in different arms. Hence, this effect could have occurred because patients with ERP-positive tumors have lower intrinsic risks of LRF after chemotherapy than do patients with ERP-negative tumors, because some ERP-positive patients received tamoxifen, or because both influences were at work. Therefore, we examined failure rates for each individual arm of each trial according to ERP status (Table 7). Testing differences between ERP-positive and -negative patients for LRF with or without DF showed ERP status had a statistically significant impact within the CMFPT12 arm of E5177, the CMFPT arm of E5181 (ignoring duration of tamoxifen), and the alternating-regimen arm of E5181 (again, ignoring tamoxifen duration), as well as for patients on E5181 (regardless of chemotherapy regimen) randomized to receive only 12 months of tamoxifen.
We also analyzed the impact of both systemic and surgical treatment factors on the risk of LRF. Because of differing time periods of accrual, population characteristics, and selection criteria for these trials, results in one arm of one trial cannot be compared directly with or combined with those of an arm in a different trial in analyzing treatment effects without introducing potentially severe biases. Therefore, we did not analyze the impact of delivered chemotherapy doses, protocol compliance, or tamoxifen duration on the risk of LRF in the study population as a whole. However, because of their potential clinical importance, we did examine two particular aspects of systemic treatment in more detail for individual trials for LRF with or without DF. There were no statistically significant differences between the rates of LRF with or without DF between different chemotherapy regimens within each trial when ERP status was taken into account (Table 7). There was a trend for the use of 12 months of tamoxifen to reduce LRF, compared with its nonuse, in E6177 but not E5177; however, even in E6177, this effect was modest and not statistically significant. There were no statistically significant differences in the risk of LRF with or without DF in patients who were randomized to receive 60 months of tamoxifen in E4181 and E5181 compared with those who received shorter durations of treatment, with the exception of patients with ERP-negative tumors in E5181 (P = .02).
We also examined the risk of LRF in the entire study population in relation to the number of axillary nodes recovered from surgery. The 10-year risk of LRF with or without DF in the 61 patients who had two to five examined nodes was 27.2% (SE, 5.9%) compared with 22.2% (SE, 2.3%) for the 353 patients with six to 10 examined nodes and 20.0% (SE, 1.0%) for the 1,598 patients with 11 or more examined nodes. Trends for isolated LRF in the three subgroups were similar (data not shown). On univariate analysis, the number of examined nodes was not a statistically significant factor for rates of either isolated LRF or LRF with or without DF (Table 4). For the population as a whole, the 10-year rates for LRF with or without DF in the 69 patients treated with radical mastectomy compared with those for the 1,947 patients treated with modified radical mastectomy were 31.1% (SE, 5.7%) and 20.3% (SE, 0.9%), respectively (P = .04).
Proportional subdistribution hazards models that directly assess the effect of covariates on the cumulative incidence of a particular type of treatment failure were used to analyze the impact of age, menopausal status, tumor size (divided by T stage), number of involved nodes (divided into the groups of one to three, four to seven, and
LRF Rates for Patients With One to Three Involved Axillary Nodes
LRF Rates for Patients With Four or More Involved Axillary Nodes
Results for Specific LRF Sites
The 10-year cumulative incidences of specific sites of LRF are listed in Table 5 for patient subgroups defined by combinations of T stage and number of involved axillary nodes. All sites of first failure were scored for this analysis when patients developed more than one simultaneous local and/or regional site of failure (with or without simultaneous DF). Because of its rarity, internal mammary node failure was not analyzed. The tumor sizes and numbers of involved nodes for these four patients were 1.0 cm with one involved node, 1.9 cm with one node, 4.2 cm with four nodes, and 2.5 cm with five nodes. The 10-year cumulative incidences of specific sites of LRF are listed in Table 11 for patient subgroups defined by combinations of the number of involved axillary nodes and the number of examined axillary nodes. Patients with T1, T2, or T3 tumors were included. For patients with one to three involved axillary nodes, the number of examined nodes had a statistically significant impact on univariate analysis only for the risk of axillary failure. For patients with four or more involved axillary nodes, the number of examined nodes had a statistically significant impact only for the risk of supraclavicular-infraclavicular failure.
LRF after mastectomy is an important clinical problem for several reasons. LRFs may be difficult to control and cause substantial morbidity.19 In addition, such failures may also reduce patients' chance of cure. The Danish and British Columbia trials,7,8 as well as others,20-23 furnish strong support for the general principle that attaining maximal initial locoregional tumor control is necessary to achieve the best possible outcome in patients with positive axillary nodes treated with mastectomy and chemotherapy. In the Danish trial, radiotherapy reduced the odds of any recurrence or death by 41% on multivariate analysis7; in the British Columbia trial,8 the reduction in the relative risk of any recurrence at 15 years was 33%. The relative reductions in the risk of death from any cause in these two trials were 29% and 26%, respectively. Data on the risk of LRF as a first failure site in patient subgroups defined by the number of involved nodes are limited and somewhat conflicting. For example, LRF rates in patients with one to three involved nodes in the Danish trial7 (crude rate, 30%) and British Columbia trial8 (10-year actuarial rate, 16%; 15-year actuarial rate, 33%) were substantially higher than those in the few other published series with more than 5 years of follow-up that reported results according to the number of involved nodes (6% to 13%).24-26 Similar discrepancies may be noted between results in these two trials for patients with four or more positive nodes (crude rate, 42%; 10- and 15-year actuarial rates, 41% and 46%, respectively) and those of other investigators (14% to 24%).22,24-27 The results in the current study (which contains a substantially larger number of patients than any published series known to the authors) concur with the majority of reported series; that is, the risk of LRF with or without DF at 10 years was 12.9% in patients with one to three positive nodes and 28.7% for patients with four or more positive nodes. One reason for these discrepancies is the use of differing definitions of LRF. For example, patients with simultaneous DF were scored as having LRF in the analysis of the Danish trial but not in the British Columbia trial. Supraclavicular nodal failures may be scored as DF rather than regional failure. (Note that a substantial proportion of all LRFs in our series was in the supraclavicular or infraclavicular nodes, eg, approximately 30% of LRFs in patients with one to three involved axillary nodes.) Another reason for these discrepancies may be different distributions of tumor sizes and other prognostic factors among the reported populations. Patients may also be monitored at different intervals and assessed for recurrence differently from study to study. Finally, another reason may be statistical, ie, the use of unadjusted Kaplan-Meier estimation rather than cumulative incidence statistics or crude rates of recurrence in populations subject to competing risks of failure.28-30 In our study, risk factors usually considered associated with increasing risk of systemic relapse (increasing tumor size, increasing number of involved nodes, and negative ERP status) were found to be associated with an increased risk of LRF. Three studies have suggested that patient subgroups defined by combinations of prognostic factors (in particular, tumor size and number of involved axillary lymph nodes) may be more prognostic of the risk of LRF than single factors alone.9,31,32 However, these three studies scored only isolated LRF. Follow-up time was also short—only recurrences within the first 3 years of follow-up were reported. In addition, no study before the current one has examined LRF or DF-only rates in patients treated with systemic therapy according to narrow intervals of tumor size (eg, < 1 cm, 1 to 2 cm, 2 to 3 cm, etc) and nodal status (eg, one positive node, two nodes, etc). In addition, few other investigators have performed multivariate analyses of the risk of LRF. A recent study examined the correlates of LRF for patients with 10 or more positive axillary nodes.33 Both the number of involved nodes and tumor size exerted statistically significant effects on this risk, whereas patient age and ERP level did not. (However, the actual rates of failure in patient subgroups were not described.) The impact of prognostic factors besides tumor size, nodal status, and ERP status on the risk of LRF is uncertain. Such factors include the presence of vascular or lymphatic invasion,34 tumor grade,32,34,35 HER2 expression,36 p53 expression,37 and the distance of tumor from the pectoral fascia.38-40 We were unable to examine the impact of these pathologic factors on the risk of LRF in our population because of inadequate information. The impact of different systemic therapies on LRF has not been well studied. Some regimens may be inferior to others. For example, adding doxorubicin to fluorouracil and melphalan (L-PAM) resulted in a substantial decrease in LRF in one randomized trial.41 In another study, LRF was less frequent in patients treated with CMF than in those treated with L-PAM.42 However, regimens that are commonly used today seem to vary little from one another in preventing LRF. For example, CMF- and doxorubicin-based regimens were approximately equally effective in this regard in the National Surgical Adjuvant Breast and Bowel Project B-15 and Southwest Oncology Group 8313 trials.43,44 A randomized trial conducted in Canada in node-positive premenopausal women compared standard oral CMF to a regimen using a different anthracycline (epirubicin).45 The incidence of isolated chest wall failure (as the first failure site) was approximately the same in the two arms (8% in the CMF-treated patients and 10% in the cyclophosphamide/epirubicin/fluorouracil [CEF]-treated patients), whereas the risk of DF was significantly reduced in the CEF arm. The administration of doxorubicin after CMF did not decrease the risk of LRF in a study conducted in Milan (although data were not separately presented for patients treated with mastectomy or breast-conserving therapy).46 In our study, there was also little difference in LRF rates between premenopausal patients treated with or without doxorubicin in trial E5181. Therefore, we believe that the results reflected in our study (in which patients were treated predominantly with variants of classic oral CMF) will likely be similar to those achieved with other currently used regimens. Whether more recent innovations in chemotherapy, such as adding taxanes or using high-dose programs, will reduce LRF rates compared with those of standard regimens remains to be determined. However, increasing the dose-intensity of cyclophosphamide did not decrease LRF rates in the National Surgical Adjuvant Breast and Bowel Project B-22 trial when compared with standard doses.47 In addition, the risk of LRF was still substantial (three of eight patients) in a pilot program of high-dose chemotherapy in patients with 10 or more involved nodes who were not irradiated.48 Routine addition of tamoxifen to chemotherapy for patients with ERP-positive tumors might reduce the risk of LRF compared with that when chemotherapy alone is used. Nonetheless, few data are available to evaluate this conjecture. Oophorectomy did not reduce LRF rates for premenopausal patients who received CMFP in the International Breast Cancer Study Group trial II.49 The Danish Breast Cancer Group trial 82b also contained a third arm in which patients received CMF plus tamoxifen for 1 year. The 7-year rate of LRF for patients who received CMF plus tamoxifen was 43%, compared with 36% for patients who received CMF and 12% for those who received CMF plus radiotherapy.50 (A recent report of this study did not update the results in the CMF plus tamoxifen arm.51) However, these two trials included patients with both ERP-positive and -negative tumors, and the Danish trial used only 1 year of tamoxifen; both factors dilute the potential impact of hormonal therapy. The Italian Breast Cancer Adjuvant Chemo-Hormone Therapy Cooperative Group trial I enrolled only node-positive patients (both premenopausal and postmenopausal) with ERP-positive tumors. The incidences of isolated LRF in patients who received chemotherapy or both chemotherapy and tamoxifen (given for 5 years) were 13% and 5%, respectively, at a median follow-up time of 5 years.52 However, these results included LRF in patients treated with breast-conserving therapy and those treated with mastectomy; in addition, an unstated number of patients received postmastectomy radiotherapy.53 Hence, the implications of their findings for patients treated with mastectomy without radiotherapy cannot be assessed easily. In our study, trials E5177 and E6177 showed a trend for reduced LRF rates in patients who received both chemotherapy and tamoxifen compared with those in patients who received chemotherapy alone, which nonetheless was not statistically significant. However, the duration of tamoxifen treatment in these two trials (only 1 year) would be considered inadequate today.54 None of the trials included in this study compared chemotherapy alone to chemotherapy plus 5 years of tamoxifen. The data from trial E4181 suggest that longer durations of tamoxifen treatment might have a modestly greater impact on LRF rates than shorter durations, but such an effect was very limited, if real, in E5181. Therefore, the ultimate value of tamoxifen in the reduction of LRF rates in patients who receive chemotherapy after mastectomy remains uncertain. Another factor that may account for some differences in LRF rates between series is variability in surgical technique. For example, in the Danish trial, the LRF rate was 40% for 133 patients who had an axillary dissection specimen from which only zero to three nodes were recovered, compared with 32% for 511 patients in whom four to nine nodes were recovered and 27% for 211 patients in whom 10 or more nodes were examined.7 Our own study (Table 8) showed a similar trend, which was statistically significant for total LRF rates on multivariate analysis (Table 4). The rate for LRF with or without DF was particularly high for patients with four or more involved nodes with only four to five examined nodes. In retrospect, it is not surprising that the number of nodes examined was more significant when adjusting for the effect of number of positive nodes in the multivariate model than it was on univariate analysis. This is because the cases with a smaller number of nodes examined must (on average) have a smaller number of positive nodes as well. In our study, the main impact of the number of examined nodes for patients with one to three involved nodes seemed to be on the risk of axillary recurrence rather than local or supraclavicular-infraclavicular failure. In patients with four or more involved nodes, the effect of the number of examined nodes was greatest with regard to supraclavicular-infraclavicular failure. However, the numbers of patients in the subgroups with five or fewer examined nodes were small, particularly for patients with four or more involved nodes; hence, these results should be considered tentative. Two randomized trials compared radical mastectomy with modified radical mastectomy. In a trial performed in Alabama, LRF rates were higher in patients treated with modified radical mastectomy,55 whereas in a trial conducted in England, LRF rates were the same in both arms.56 However, neither report explicitly stated whether radiotherapy was given to patients, and neither used systemic therapy. We suspect that the increased rate of LRF with or without DF observed in the small group of patients treated with radical mastectomy in our study may reflect selection bias, ie, more extensive surgery was used by the surgeon for patients with more extensive disease. Caution is needed in the interpretation, generalization, and clinical use of our results for several reasons. With regard to issues of internal validity, as is usual when multiple hypotheses are tested, a large number of P values were generated, and hence some were likely to be significant as a result of chance. Because we did not adjust our analysis for this problem, our results should be interpreted carefully. In addition, the numbers of patients in some of the subgroups identified were limited, resulting in fairly large SEs of the estimated risks of LRF and low power to detect clinically relevant differences. With regard to the external validity or generalizability of our findings, menopausal status was a protocol eligibility factor; therefore, it could have led to differential sampling of the population of all breast cancer patients. In addition, our population could have been relatively enriched with patients at lower risk of LRF than the "average patient" if those considered at high risk of LRF were preferentially not entered onto our trials because institutional policies directed that they should be irradiated. The management implications of our findings are not straightforward. The limitations of the available study material did not allow us to examine the effect of pathologic factors, such as lymphatic vessel invasion, tumor grade, or margin status, that might substantially affect the risk of LRF. It is also possible that newer chemotherapy regimens and changes in surgical practices might result in lower LRF rates than those found here, although evidence for this possibility is lacking at present. Data from other trials and institutions on these points are urgently needed. In addition, such data need to be given separately for patients treated with mastectomy and breast-conserving therapy, as the lower axilla is included in standard tangential fields that are used to irradiate the breast.57 Finally, both the benefits and costs of treatment options must be considered when making clinical decisions. It seems likely that, as with systemic therapy,54,58 adjuvant radiation therapy will yield comparable proportional reductions in LRF for all patient subgroups, but the absolute size of differences in outcome will be substantially greater for patients at higher risk than for those at lower risk. Certain subgroups of patients may therefore have such low failure rates that the absolute benefits of such treatment will be insufficient to outweigh the potential side effects. We will not debate here as to what LRF rate is sufficiently high to serve as a threshold for recommending postmastectomy radiotherapy. There seems to be substantial consensus among experts regarding the value of routine irradiation for patients with four or more involved nodes or locally advanced cancers and considerable controversy (even among radiation oncologists) regarding its value for patients with one to three involved nodes.59 We believe our results are valuable for a different reason in that they show the complexity of assessing the risks of LRF in patient subgroups, especially when differing systemic treatments are used. In such a situation, prospective randomized trials will be necessary to estimate accurately the potential impacts of postmastectomy radiotherapy on disease-free and overall survival benefits for patients in particular prognostic subgroups treated with currently used and future systemic therapy regimens. In conclusion, our study showed that LRF was a substantial problem for node-positive breast cancer patients treated with mastectomy, despite the use of chemotherapy with or without tamoxifen. Among the factors that could be examined in this study population, tumor size and number of involved axillary nodes were the variables most closely related to the risk of developing an isolated LRF. Increasing tumor size, increasing number of histologically involved axillary nodes, negative ERP status, and decreasing number of examined axillary nodes significantly increased the risk of LRF with or without simultaneous DF. Further retrospective and prospective research on the correlates of such failure is needed to confirm and extend these findings.
Conducted by the Eastern Cooperative Study Group (Robert L. Comis, MD, Chair) and supported in part by United States Public Health Services grants no. CA23318, CA16116, CA18281, CA21692, CA66636, and CA21115 from the National Cancer Institute, National Institutes of Health and the Department of Health and Human Services. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.
1. Cuzick J, Stewart H, Rutqvist L, et al: Cause-specific mortality in long-term survivors of breast cancer who participated in trials of radiotherapy. J Clin Oncol 12:447-453, 1994[Abstract] 2. Recht A: The return (?) of postmastectomy radiotherapy. J Clin Oncol 13:2861-2864, 1995 (editorial) [Medline] 3. Harris JR, Morrow M: Local management of invasive breast cancer, in Harris JR, Lippman ME, Morrow M, et al (eds): Diseases of the Breast. Philadelphia, PA, Lippincott, 1996, pp 485-545
4.
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 5. Pierce LJ, Lichter AS: Defining the role of post-mastectomy radiotherapy: The new evidence. Oncology (Huntingt) 10:991-1002, 1996[Medline] 6. Fowble B: Postmastectomy radiotherapy: Then and now. Oncology (Huntingt) 11:213-239, 1997[Medline]
7.
Overgaard M, Hansen PS, 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, 1997
8.
Ragaz J, Jackson SM, Le N, et al: Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 337:956-962, 1997
9.
Fowble B, Gray R, Gilchrist K, et al: Identification of a subgroup of patients with breast cancer and histologically positive axillary nodes receiving adjuvant chemotherapy who may benefit from postoperative radiotherapy. J Clin Oncol 6:1107-1117, 1988 10. Tormey DC, Gray R, Gilchrist K, et al: Adjuvant chemohormonal therapy with cyclophosphamide, methotrexate, 5- flouorouracil, and prednisone (CMFP) or CMFP plus tamoxifen compared with CMF for premenopausal breast cancer patients: An Eastern Cooperative Oncology Group trial. Cancer 65:200-206, 1990[Medline] 11. Taylor SG, Knuiman MW, Sleeper LA, et al: Six-year results of the Eastern Cooperative Oncology Group trial of observation versus CMFP versus CMFPT in postmenopausal patients with node-positive breast cancer. J Clin Oncol 7:879-889, 1989[Abstract] 12. Falkson HC, Gray R, Wolberg WH, et al: Adjuvant trial of 12 cycles of CMFPT followed by observation or continuous tamoxifen versus four cycles of CMFPT in postmenopausal women with breast cancer: An Eastern Cooperative Oncology Group phase III study. J Clin Oncol 8:599-607, 1990[Abstract] 13. Tormey DC, Gray R, Abeloff MD, et al: Adjuvant therapy with a doxorubicin regimen and long-term tamoxifen in premenopausal breast cancer patients: An Eastern Cooperative Oncology Group trial. J Clin Oncol 10:1848-1856, 1992[Abstract] 14. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, John Wiley and Sons, 1980 15. Gray RJ: A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 16:1141-1154, 1988 16. Snedecor GW, Cochran WG: Statistical Methods (ed 6). Ames, IA, Iowa State University Press, 1967 17. Mehta CR, Patel HR: A network of algorithm for performing Fisher's exact test in R x C contingency tables. J Am Stat Assoc 78:427-434, 1983 18. Fine JP, Gray RJ: A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc (in press) 19. Recht A, Hayes DF, Eberlein TJ, et al: Local-regional recurrence after mastectomy or breast-conserving therapy, in Harris JR, Lippman M, Morrow M, et al (eds): Diseases of the Breast. Philadelphia, PA, Lippincott, 1996, pp 649-667 20. Ahmann DL, O'Fallon JR, Scanlon PW, et al: A preliminary assessment of factors associated with recurrent disease in a surgical adjuvant clinical trial for patients with breast cancer with special emphasis on the aggressiveness of therapy. Am J Clin Oncol 5:371-381, 1982[Medline] 21. McArdle CS, Crawford D, Dykes EH, et al: Adjuvant radiotherapy and chemotherapy in breast cancer. Br J Surg 73:264-266, 1986[Medline] 22. Vélez-García E, Carpenter JT, Moore M, et al: Postsurgical adjuvant chemotherapy with or without radiotherapy in women with breast cancer and positive axillary nodes: A South-Eastern Cancer Study Group (SEG) trial. Eur J Cancer 28A:1833-1837, 1992 23. Tennvall-Nittby L, Tenegrup I, Landberg T: The total incidence of loco-regional recurrence in a randomized trial of breast cancer TNM stage II: The South Sweden Breast Cancer Trial. Acta Oncol 32:641-646, 1993[Medline] 24. Goldhirsch A, Gelber RD, Castiglione M: Relapse of breast cancer after adjuvant treatment in premenopausal and perimenopausal women: Patterns and prognoses. J Clin Oncol 6:89-97, 1988[Abstract] 25. Buzdar A, McNeese MD, Hortobagyi GN, et al: Is chemotherapy effective in reducing the local failure rate in patients with operable breast cancer? Cancer 65:394-399, 1990[Medline]
26.
Kaufmann M, Jonat W, Abel U, et al: Adjuvant randomized trials of doxorubicin/cyclophosphamide versus doxorubicin/cyclophosphamide/tamoxifen and CMF chemotherapy versus tamoxifen in women with node-positive breast cancer. J Clin Oncol 11:454-460, 1993 27. Chu AM, Kiel K: Comparison of adjuvant postoperative radiotherapy and multiple-drug chemotherapy (CMF-VP) in operable breast cancer patients with more than four positive axillary lymph nodes. Cancer 50:212-218, 1982[Medline] 28. Gelman R, Gelber R, Henderson IC, et al: Improved methodology for analyzing local and distant recurrence. J Clin Oncol 8:548-555, 1990[Abstract] 29. Caplan RJ, Pajak TF, Cox JD: Analysis of the probability and risk of cause-specific survival. Int J Radiat Oncol Biol Phys 29:1183-1186, 1994[Medline] 30. Schulgen G, Schmoor C, Sauerbrei W, et al: A note on estimating local recurrence rates in clinical trials on the treatment of breast cancer. Breast Cancer Res Treat 49:87-91, 1998[Medline] 31. Sykes HF, Sim DA, Wong CJ, et al: Local-regional recurrence in breast cancer after mastectomy and Adriamycin-based adjuvant chemotherapy: Evaluation of the role of postoperative radiotherapy. Int J Radiat Oncol Biol Phys 16:641-647, 1989[Medline] 32. Pisansky TM, Ingle JN, Schaid DJ, et al: Patterns of tumor relapse following mastectomy and adjuvant systemic therapy in patients with axillary lymph node-positive breast cancer. Cancer 72:1247-1260, 1993[Medline] 33. Diab SG, Hilsenbeck SG, deMoor CA, et al: Radiation therapy and survival in breast cancer patients with 10 or more positive axillary lymph nodes treated with mastectomy. J Clin Oncol 16:1655-1660, 1998[Abstract] 34. O'Rourke S, Galea MH, Morgan D, et al: Local recurrence after simple mastectomy. Br J Surg 81:386-389, 1994[Medline] 35. Bijker N, Rutgers EJT, Peters JL, et al: Modified radical mastectomy or how to achieve optimal loco-regional control in larger breast cancers. Eur J Surg Oncol 20:258, 1994 (abstr) 36. Stal O, Sullivan S, Wingren S, et al: c-erbB-2 expression and benefit from adjuvant chemotherapy and radiotherapy of breast cancer. Eur J Cancer 31A:2185-2190, 1995 37. Zellars RC, Clark GM, Allred DC, et al: Prognostic value of p53 for local failure in mastectomy treated breast cancer patients. Proc Am Soc Clin Oncol 17:104a, 1998 (abstr 401) 38. Ahlborn TN, Gump FE, Bodian C, et al: Tumor to fascia margin as a factor in local recurrence after modified radical mastectomy. Surg Gynecol Obstet 166:523-526, 1988[Medline] 39. 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] 40. Freedman GM, Fowble BL, Hanlon AL, et al: A close or positive margin after mastectomy is not an indication for chest wall irradiation except in women aged fifty or younger. Int J Radiat Oncol Biol Phys 41:599-605, 1998[Medline] 41. Fisher B, Redmond C, Wicherham DL, et al: Doxorubicin-containing regimens for the treatment of stage II breast cancer: The National Surgical Adjuvant Breast and Bowel Project experience. J Clin Oncol 7:572-582, 1989[Abstract] 42. Muss HB, Cooper MR, Brockschmidt JK, et al: A randomized study of chemotherapy (L-PAM vs CMF) and irradiation for node positive breast cancer: Eleven year follow-up of a Piedmont Oncology Association trial. Breast Cancer Res Treat 19:77-84, 1991[Medline] 43. Fisher B, Brown AM, Dimitrov NV, et al: Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen-nonresponsive tumors: Results from the National Surgical Adjuvant Breast and Bowel Project B-15. J Clin Oncol 8:1483-1496, 1990[Abstract] 44. Budd GT, Green S, O'Bryan RM, et al: Short-course FAC-M versus 1 year of CMFVP in node-positive, hormone receptor-negative breast cancer: An Intergroup study. J Clin Oncol 13:831-839, 1995[Abstract] 45. Levine M, Bramwell VH, Pritchard KI, et al: Randomized trial of intensive cyclophosphamide, epirubicin, and fluorouracil chemotherapy compared with cyclophosphamide, methotrexate, and fluorouracil in premenopausal women with node-positive breast cancer. J Clin Oncol 16:2651-2658, 1998[Abstract] 46. Moliterni A, Bonadonna G, Valagussa P, et al: Cyclophosphamide, methotrexate, and fluorouracil with and without doxorubicin in the adjuvant treatment of resectable breast cancer with one to three positive axillary nodes. J Clin Oncol 9:1124-1130, 1991[Abstract]
47.
Fisher B, Anderson S, Wickerham DL, et al: Increased intensification and total dose of cyclophosphamide in a doxorubicin-cyclophosphamide regimen for the treatment of primary breast cancer: Findings from the National Surgical Adjuvant Breast and Bowel Project B-22. J Clin Oncol 15:1858-1869, 1997
48.
Peters WP, Ross M, Vrendenburgh JJ, et al: High-dose chemotherapy and autologous bone marrow support as consolidation after standard-dose adjuvant therapy for high-risk primary breast cancer. J Clin Oncol 11:1132-1143, 1993
49.
International Breast Cancer Study Group: Late effects of adjuvant oophorectomy and chemotherapy upon premenopausal breast cancer patient. Ann Oncol 1:30-35, 1990 50. Dombernowsky P, Hansen M, Mouridsen HT, et al: Randomized trial of adjuvant CMF + radiotherapy (RT) vs CMF alone vs CMF + tamoxifen (TAM) in pre- and menopausal stage II breast cancer. Eur J Cancer 30A:S28, 1994 (suppl 2) (abstr) 51. Overgaard M, Hansen PS, Overgaard J, et al: Randomized trial evaluating postoperative radiotherapy in high-risk postmenopausal breast cancer patients given adjuvant tamoxifen: Results from the DBCG 82c trial. Radiother Oncol 48:S78, 1998 (suppl 1) (abstr) 52. Boccardo F, Rubagotti A, Amoroso D, et al: Chemotherapy versus tamoxifen versus chemotherapy plus tamoxifen in node-positive, oestrogen-receptor positive breast cancer patients: An update at 7 years of the 1st GROCTA (Breast Cancer Adjuvant Chemo-hormone Therapy Cooperative Group) Trial. Eur J Cancer 28:673-680, 1992 53. Boccardo F, Rubagotti A, Bruzzi P, et al: Chemotherapy versus tamoxifen versus chemotherapy plus tamoxifen in node-positive, estrogen receptor-positive breast cancer patients: Results of a multicentric Italian trial. J Clin Oncol 8:1310-1320, 1990[Abstract] 54. Early Breast Cancer Trialists' Collaborative Group: Tamoxifen for early breast cancer: An overview of the randomized trials. Lancet 351:1451-1467, 1998[Medline] 55. Maddox W, Carpenter J, Laws H: A randomized prospective trial of radical mastectomy versus modified radical mastectomy in 311 breast cancer patients. Ann Surg 198:207-212, 1983[Medline] 56. Turner L, Swindell R, Bell W, et al: Radical versus modified radical mastectomy for breast cancer. Ann R Coll Surg Engl 63:239-243, 1981[Medline] 57. Recht A, Houlihan MJ: Axillary lymph nodes and breast cancer: A review. Cancer 76:1491-1512, 1995[Medline] 58. Early Breast Cancer Trialists' Collaborative Group: Polychemotherapy for early breast cancer: An overview of the randomized trials. Lancet 352:930-942, 1998[Medline] 59. Recht A, Bartelink H, Fourquet A, et al: Postmastectomy radiotherapy: Questions for the twenty-first century. J Clin Oncol 16:2886-2889, 1998[Medline] Submitted September 23, 1998; accepted February 18, 1999.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
65 years) in three arms: no adjuvant therapy, CMFP for 12 cycles, or CMFPT for 12 cycles.
2 test
