|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2007.14.4147 on February 4 2008 © 2008 American Society of Clinical Oncology. Response to Neoadjuvant Therapy and Long-Term Survival in Patients With Triple-Negative Breast Cancer
From the Departments of Breast Medical Oncology, Biostatistics and Applied Mathematics, and Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, TX; Department of Gynecology and Obstetrics, University of Münster, Münster, Germany; Department of Obstetrics and Gynecology, Marseille Public Hospital System, Marseille; and Breast Cancer Unit and Translational Research Unit UPRES03535, Institut Gustave Roussy, Villejuif, France Corresponding author: Lajos Pusztai, MD, PhD, Department of Breast Medical Oncology, Unit 1354, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030-1439, USA; e-mail: lpusztai{at}mdanderson.org
Purpose Triple-negative breast cancer (TNBC) is defined by the lack of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER-2) expression. In this study, we compared response to neoadjuvant chemotherapy and survival between patients with TNBC and non-TNBC. Patients and Methods Analysis of a prospectively collected clinical database was performed. We included 1,118 patients who received neoadjuvant chemotherapy at M.D. Anderson Cancer Center for stage I-III breast cancer from 1985 to 2004 and for whom complete receptor information were available. Clinical and pathologic parameters, pathologic complete response rates (pCR), survival measurements, and organ-specific relapse rates were compared between patients with TNBC and non-TNBC. Results Two hundred fifty-five patients (23%) had TNBC. Patients with TNBC compared with non-TNBC had significantly higher pCR rates (22% v 11%; P = .034), but decreased 3-year progression-free survival rates (P < .0001) and 3-year overall survival (OS) rates (P < .0001). TNBC was associated with increased risk for visceral metastases (P = .0005), lower risk for bone recurrence (P = .027), and shorter postrecurrence survival (P < .0001). Recurrence and death rates were higher for TNBC only in the first 3 years. If pCR was achieved, patients with TNBC and non-TNBC had similar survival (P = .24). In contrast, patients with residual disease (RD) had worse OS if they had TNBC compared with non-TNBC (P < .0001). Conclusion Patients with TNBC have increased pCR rates compared with non-TNBC, and those with pCR have excellent survival. However, patients with RD after neoadjuvant chemotherapy have significantly worse survival if they have TNBC compared with non-TNBC, particularly in the first 3 years.
Triple-negative breast cancers (TNBCs) are characterized by the lack of expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER-2).1,2 These cancers occur in approximately 20% to 25% of all patients with breast cancer, and are associated with an unfavorable prognosis.2-4 Patients with TNBC derive no benefit from molecularly targeted treatments such as endocrine therapy or trastuzumab, because they lack the appropriate targets for these drugs. The primary goal of this study was to describe the relation between triple-negative receptor status and major determinants of clinical outcome, such as response to neoadjuvant chemotherapy (rate of pathologic complete response [pCR]), progression-free survival (PFS), site-specific distribution of recurrence, postrecurrence survival (PRS), and overall survival (OS). The secondary goal was to explain the relationship between the increased response rate to chemotherapy associated with this breast cancer subtype in contrast to the unfavorable prognosis.
Study Design Of the 1,782 patients diagnosed with nonmetastatic breast cancer between 1985 and 2004 who received neoadjuvant chemotherapy at M.D. Anderson Cancer Center (Houston, TX), 1,118 were included. Patient characteristics of the excluded patients are provided in the Appendix (online only). Patient cases were selected from the Breast Medical Oncology Database based on the following criteria: (1) receipt of at least one cycle of chemotherapy; (2) availability of complete information on clinical (cTNM at diagnosis) and pathologic stage (pTNM after neoadjuvant chemotherapy); (3) response to treatment; and (4) known ER, PR, and HER-2 status. Patients were excluded from the analysis if they had an additional (metachronous or synchronous) breast cancer or if they had received trastuzumab or bevacizumab. Staging was performed according to American Joint Committee on Cancer guidelines.5 Clinical and histologic characteristics of all patients were obtained from medical records and entered prospectively into an institutional clinical database. All primary tumors were reviewed by a dedicated breast pathologist on first referral to our institution. No central pathology rereview was performed for this analysis. Clinical tumor size was determined on the basis of physical examination and imaging tests, including mammograms. This data analysis was approved by our institutional review board.
Pathology Assessment
Statistical Analyses OS was measured from the date of definitive surgery to the date of last follow-up or death. PFS was measured from the date of definitive surgery to the date of last follow-up or disease relapse. PRS was measured from the date of disease relapse to the date of last follow-up or death. Breast cancer relapse was defined as locoregional or distant recurrence/metastasis. Statistical analyses were performed using S-PLUS 7.0 for Windows (Insightful Corp, Seattle, WA) and more details on the statistical methods are given in the Appendix.
Patient Characteristics Two hundred fifty-five patients (23%) were designated as having TNBC and 863 patients (77%) were designated as non-TNBC. Expression of ER, PR, or HER-2 was observed in 645 (58%), 524 (47%), and 272 (24%) patients, respectively. The mean age of patients with TNBC (48 years) was slightly younger compared with non-TNBC (50 years; P = .002). When age group cohorts were compared, there was no association between patient age group and incidence of TNBC relative to non-TNBC. This may not represent the true incidence pattern of TNBC in the general population but rather reflects a selection bias because the database included only patients who received neoadjuvant chemotherapy. African American (31.8%) and Hispanic women (33.6%) showed an increased incidence of TNBC compared with white patients: 32%, 34%, and 19%, respectively (P = .0003 and P = .018, respectively). This observation is concordant with previous reports.2,8 A small subset of patients had inflammatory breast cancer (n = 59); the incidence of TNBC in this subset was similar (27%) to those with noninflammatory histology (23%; P = .253). Table 1 lists patient characteristics and results of multivariate analysis for patients with TNBC compared with non-TNBC. Table 2 lists the chemotherapy regimens that were administered. Adjuvant therapy after surgery consisted of additional cytotoxic therapy in 379 patients. Greater than 99% of patients with ER-positive disease (n = 641) received adjuvant endocrine therapy, including either tamoxifen (n = 465), anastrozole (n = 168), letrozole (n = 6), or exemestane (n = 1). One patient underwent bilateral salpingo-oophorectomy.
Effect of Triple-Negative Status on Response to Neoadjuvant Chemotherapy Overall, 163 patients (15%) experienced pCR compared with 945 patients (85%) with residual disease (RD). In multivariate analysis, increased pCR rates were observed for patients with TNBC compared with non-TNBC (22% v 11%; odds ratio [OR] = 1.53; 95% CI, 1.03 to 2.26; P = .034) for ductal histology compared with nonductal histology (OR = 2.14; 95% CI, 1.08 to 4.25; P = .030) and high nuclear grade (one grade unit increase, OR = 6.63; 95% CI, 3.49 to 12.6; P < .0001). Decreased pCR rates were observed for larger cT (OR = 0.78; 95% CI, 0.64 to 0.95; P = .013). Individually, all ER-positive/PR-negative/HER-2–negative, ER-positive/PR-positive/HER-2–negative, and ER-positive/PR-positive/HER-2–positive breast cancers exhibited significantly lower rates of pathologic complete response compared with TNBC (P < .0001, P < .0001, and P = .0067, respectively). Table 2 lists pCR rates as a function of triple-negative status and chemotherapy regimens. Increased pCR rates in association with TNBC were observed for patients treated with fluorouracil, doxorubicin, and cyclophosphamide/fluorouracil, epirubicin, and cyclophosphamide, or weekly/once every 3 weeks paclitaxel/docetaxel followed by fluorouracil, doxorubicin, and cyclophosphamide plus weekly/once every 3 weeks paclitaxel/docetaxel followed by fluorouracil, epirubicin, cyclophosphamide chemotherapy (P = .0001 and P = .0072, respectively). There was a trend for higher pCR rate for single-agent taxane or other regimens that did not reach statistical significance (P = .82 and P = .33, respectively), probably because of the small sample size in these categories and scarcity of pCR with single-agent therapy.
Effect of Triple-Negative Status on Survival Parameters Decreased OS was also observed for patients with TNBC compared with non-TNBC (3-year OS rates: 74% v 89%; HR = 2.53; 95% CI, 1.77 to 3.57; P < .0001), ductal compared with nonductal histology (HR = 1.80; 95% CI, 1.11 to 2.92; P = .017), high nuclear grade (HR = 2.20; 95% CI, 1.48 to 3.25; P < .0001), and higher tumor stage (HR = 1.22; 95% CI, 1.03 to 1.45; P = .020). Figures 1A and 1D show the Kaplan-Meier estimates of PFS and OS for patients with TNBC and non-TNBC, respectively. The 1-, 3-, and 5-year estimates for PFS were 81%, 63%, and 61% for TNBC; and 90%, 76%, and 70% for non-TNBC, respectively. The 1-, 3-, and 5-year estimates for OS were 90%, 74%, and 64% for TNBC; and 97%, 89%, and 81% for non-TNBC, respectively. Figures 1B and 1E show kernel estimates of the hazard functions of progression and death, respectively. The 1-, 3-, and 5-year progression HR estimates were 0.22, 0.04, and 0.02 for TNBC; and 0.10, 0.05, and 0.03 for non-TNBC, respectively. The 1-, 3-, and 5-year death HR estimates were 0.14, 0.07, and 0.03 for TNBC; and 0.04, 0.05, and 0.05 for non-TNBC, respectively. The estimated hazard curves cross at about 2.5 and 4.5 years for disease progression or death. Figures 1C and 1F give the estimates of the ratio of the OS and PFS hazard functions including point-wise 95% CIs for TNBC compared with non-TNBC, respectively. The 1-, 3-, and 5-year HRs for disease progression were 1.4, 1.0, and 0.9, respectively. The 1-, 3-, and 5-year HRs for death were 1.8, 1.4, and 1.0, respectively. Hence, recurrence and death rates were higher for TNBC compared with non-TNBC only in the first 3 years.
Patients with TNBC/RD had significantly decreased OS compared with those with non-TNBC/RD (HR = 1.5; 95% CI, 1.3 to 1.8; P < .0001). Importantly, no such significant difference was observed among patients who achieved pCR (HR = 1.7; 95% CI, 0.7 to 4.2; P = .24). Individually, ER and PR expression was associated with increased PFS (HR = 0.66; 95% CI, 0.51 to 0.86; P = .0020 and HR = 0.66; 95% CI, 0.50 to 0.86; P = .0023, respectively) and OS (HR = 0.48; 95% CI, 0.35 to 0.65; P < .001 and HR = 0.68; 95% CI, 0.50 to 0.94; P = .020, respectively). HER-2 expression showed a trend toward decreased PFS (HR = 1.27; 95% CI, 0.98 to 1.64; P = 0.071), while not exhibiting any significant effect on OS (HR = 0.98; 95% CI, 72 to 133; P = .89). We stratified patients into eight groups according to ER, PR, and HER-2 expression. PFS was increased for ER-positive/PR-negative/HER-2–negative (P = .0098), ER-negative/PR-positive/HER-2–negative (P = .037), ER-positive/PR-positive/HER-2–negative (P < .0001), and ER-positive/PR-positive/HER-2–positive (P = .029) tumors compared with TNBC. OS was increased for ER-positive/PR-negative/HER-2–negative (P = .0020), ER-positive/PR-positive/HER-2–negative (P < .0001), and ER-positive/PR-positive/HER-2–positive (P = .0007) tumors compared with TNBC (Appendix Figs A1 and A2; Table A1, online only). Tests for interactions between individual receptor expression and outcome (pCR, PFS, or OS) did not yield significant results. In other words, the effect, which might be seen as a consequence of expression of two receptors in the absence of the third, might be significantly different from the effect observed when the latter is present.
Effect of Receptor Expression on Site of Recurrence and Postrecurrence Survival
We present the results of the largest study to date, to our knowledge, that comprehensively examines the clinical phenotype of TNBC with regard to response to neoadjuvant chemotherapy and survival parameters. Our data indicate that pCR rate is higher in TNBC compared with non-TNBC. More importantly, we also showed that patients who achieved pCR had excellent survival regardless of receptor status, but patients with TNBC and residual disease after neoadjuvant chemotherapy have significantly shorter overall and postrecurrence survival than patients with non-TNBC and residual cancer (Fig 2). This is likely due to the combination of at least two factors. First, TNBC in general has poor prognostic features, including significantly higher nuclear grade, increased incidence of visceral metastases, and shorter recurrence-free interval compared with non-TNBC. Second, chemotherapy is the only systemic treatment option for TNBC, whereas patients with non-TNBC may derive benefit from both chemotherapy and endocrine therapy and, more recently, trastuzumab as well. It is therefore not surprising that patients with non-TNBC and residual cancer after neoadjuvant chemotherapy who also received adjuvant endocrine therapy (if ER or PR positive) in this study showed better survival. As expected, the survival of the entire group of non-TNBC was also better than that in the group with TNBC because the majority of patients in both groups had residual cancer.
It is important to recognize that the non-TNBC group is clinically heterogeneous. In general, fewer ER-positive breast cancers are sensitive to chemotherapy than ER-negative cancers.9,10 However, some subsets of patients in the non-TNBC group may have as high, or higher, pCR rates as those with TNBC (Table A1). For example, HER-2–amplified tumors have significantly higher response rates than HER-2–normal tumors in both hormone receptor subsets.11 HER-2–amplified tumors may be particularly sensitive to anthracycline-based chemotherapy.12 Our results also demonstrate that the risk of recurrence is strongly time-dependent, and the hazard curves are different for TNBC and non-TNBC. The risk of relapse and death is significantly and dramatically higher for TNBC during the first 3 years of follow-up. After 3 years, the hazard curves run close to each other and can even cross, suggesting a possibly higher risk of relapse and death for non-TNBC in years 4 to 6. Our results corroborate previous reports on the clinical history of TNBC. Dent et al1 described the clinical course of 1,601 women with breast cancer who received various therapies but not neoadjuvant chemotherapy. Eleven percent of women had TNBC defined by IHC, and showed an increased likelihood of distant recurrence (P < .0001) and death (P < .001) within 5 years of diagnosis, but not thereafter (similar to our results). In that study, the recurrence risk in the non-TNBC group was constant over the period of follow-up. In another study, Carey et al3 examined response rates to neoadjuvant anthracycline-based chemotherapy in different molecular classes of breast cancer. The investigators assigned basal-like, HER-2–positive/ER-negative, and luminal subtypes to 34, 11, and 62 breast cancers, respectively. Similar to our results, they observed a high pathologic response rate in the basal-like group that paradoxically had significantly decreased distant DFS and OS compared with the luminal subtype. Our results add to the previous literature in that we demonstrate in a large number of patients that the worse overall survival of TNBC is primarily determined by the worse survival of patients with residual cancer. Even though TNBC includes more patients with highly chemotherapy-sensitive disease compared with non-TNBC, these individuals still represent only a minority of all patients with TNBC. None of the previous studies correlated tumor phenotype with site of recurrence; in this article, we show that TNBC has a higher predilection for visceral metastasis and early recurrence within the first 3 years of follow-up. Our study has limitations. Almost one third of eligible patients were excluded because of the lack of complete receptor information. Those who were excluded from the analysis were diagnosed in significantly earlier time intervals (in the mid- and late 1980s) and had significantly lower pT, pN, cT, and cN stages compared with those patients included in our study. There were significant differences in tumor grade, patient age, and ethnicity. How and whether the excluded patients could bias our observations is unclear. In addition, our study covered an almost 20-year period during which multiple different and increasingly effective chemotherapy regimens were used. the frequent use of earlier generation chemotherapy regimens (approximately half of the patients received 3 to 6 months of an anthracycline-based regimen without a taxane) is reflected by the relatively low overall pCR rate (11% and 20% for TNBC and non-TNBC, respectively) compared with what some current third-generation regimens that include taxanes can accomplish.13 However, systemic treatment variables (except endocrine therapy) were equally distributed among the TNBC and the non-TNBC subgroup. The variety of regimens used in the study may even be considered a strength of the study, given that it allows for a conclusion that extends across different cytotoxic treatments. It is important to note that we excluded patients who received neoadjuvant or adjuvant trastuzumab, and more than 75% of ER-positive patients received tamoxifen rather than an aromatase inhibitor. Given that both of these modalities are now used routinely, it is possible that the prognostic gap between TNBC and non-TNBC is even wider today than in our historic cohort of patients. In conclusion, patients with TNBC have increased pCR rates compared with non-TNBC, and those with pCR have excellent survival. However, patients with RD after neoadjuvant chemotherapy have significantly worse survival, particularly in the first 3 years. Accordingly, patients with TNBC may be best treated with third-generation adjuvant or neoadjuvant chemotherapy regimens that achieve the highest possible pCR rates. Given the high risk of visceral metastases, these individuals may require closer surveillance in the initial years of follow-up. However, whether earlier detection and aggressive therapy of metastatic recurrence could improve survival is yet to be demonstrated. Importantly, these and previous results illustrate the need to develop novel therapeutic alternatives for this subgroup of patients to alleviate the significantly worse PFS and OS associated with residual disease with current chemotherapies.
The author(s) indicated no potential conflicts of interest.
Conception and design: Cornelia Liedtke, Chafika Mazouni, W. Fraser Symmans, Lajos Pusztai Administrative support: Gabriel N. Hortobagyi, Lajos Pusztai Collection and assembly of data: Chafika Mazouni, Jaime A. Mejia, Ana M. Gonzalez-Angulo, Bryan Hennessy, Marjorie Green, Massimo Cristofanilli Data analysis and interpretation: Cornelia Liedtke, Kenneth R. Hess, Fabrice André, Attila Tordai, W. Fraser Symmans, Ana M. Gonzalez-Angulo, Massimo Cristofanilli, Gabriel N. Hortobagyi, Lajos Pusztai Manuscript writing: Cornelia Liedtke, Ana M. Gonzalez-Angulo, Gabriel N. Hortobagyi, Lajos Pusztai Final approval of manuscript: Cornelia Liedtke, Chafika Mazouni, Kenneth R. Hess, Fabrice André, Attila Tordai, Jaime A. Mejia, W. Fraser Symmans, Ana M. Gonzalez-Angulo, Bryan Hennessy, Marjorie Green, Massimo Cristofanilli, Gabriel N. Hortobagyi, Lajos Pusztai
Patient characteristics. Of the 1,782 patients represented in the neoadjuvant database (restricted to cases between 1985 and 2004 to allow for at least 2 years of follow-up), 89 patients had a second breast cancer diagnosis, 17 patients presented with proven primary metastatic disease, and 26 and seven patients had unknown tumor or nodal stage, respectively. In 437 patients, estrogen receptor (ER) and/or progesterone receptor (PR) and/or human epidermal growth factor receptor (HER-2) status was missing, and 85 patients had received neoadjuvant trastuzumab. After exclusion of these patients, the final study population included 1,118 patients. Patients excluded from this analysis had significantly smaller cT and cN, as well as pT and pN stages, compared with included patients. Excluded patients were also diagnosed at a significantly earlier time interval. No significant difference was observed for tumor grade or ethnicity.
Statistical analysis.
published online ahead of print at www.jco.org on February 4, 2008. Supported by grants to C.L. from the Deutsche Forschungsgemeinschaft; to L.P. from the National Cancer Institute (NCI; Grant No. RO1-CA106290), the Breast Cancer Research Foundation, and the Goodwin Foundation; and to G.N.H. from the NCI [Grant No. 2P30 CA016672 30(PP-4)] and the Nellie B. Connally Breast Cancer Research Fund. T.A. is a visiting professor supported by the Hungarian American Enterprise Scholarship Fund. Presented in part at the 43rd Annual Meeting of the American Society of Clinical Oncology, June 1-5, 2007, Chicago, IL. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Dent R, Trudeau M, Pritchard KI, et al: Triple-negative breast cancer: Clinical features and patterns of recurrence. Clin Cancer Res 13:4429-4434, 2007 2. Bauer KR, Brown M, Cress RD, et al: Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: A population-based study from the California Cancer Registry. Cancer 109:1721-1728, 2007[CrossRef][Medline] 3. Carey LA, Dees EC, Sawyer L, et al: The triple negative paradox: Primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res 13:2329-2334, 2007 4. Haffty BG, Yang Q, Reiss M, et al: Locoregional relapse and distant metastasis in conservatively managed triple negative early-stage breast cancer. J Clin Oncol 24:5652-5657, 2006 5. American Joint Committee on Cancer: AJCC Cancer Staging Manual, (ed 6). New York, NY, Springer, 2002, pp 221-240 6. Fisher B, Redmond C, Fisher ER, et al: Relative worth of estrogen or progesterone receptor and pathologic characteristics of differentiation as indicators of prognosis in node negative breast cancer patients: Findings from National Surgical Adjuvant Breast and Bowel Project Protocol B-06. J Clin Oncol 6:1076-1087, 1988 7. Mazouni C, Peintinger F, Wan-Kau S, et al: Residual ductal carcinoma in situ in patients with complete eradication of invasive breast cancer after neoadjuvant chemotherapy does not adversely affect patient outcome. J Clin Oncol 25:2650-2655, 2007 8. Carey LA, Perou CM, Livasy CA, et al: Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 295:2492-2502, 2006 9. Clarke M, Collins R, Darby S, et al: Early Breast Cancer Trialists’ Collaborative Group (EBCTCG): Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival—An overview of the randomised trials. Lancet 366:2087-2106, 2005[Medline] 10. Berry DA, Cirrincione C, Henderson IC, et al: Estrogen-receptor status and outcomes of modern chemotherapy for patients with node-positive breast cancer. JAMA 295:1658-1667, 2006 11. Andre F, Mazouni C, Liedtke C, et al: HER2 expression and efficacy of preoperative paclitaxel/FAC chemotherapy in breast cancer. Breast Cancer Res Treat [epub ahead of print on April 28, 2007] 12. Pritchard KI, Shepherd LE, O'Malley FP, et al: National Cancer Institute of Canada Clinical Trials Group: HER2 and responsiveness of breast cancer to adjuvant chemotherapy. N Engl J Med 354:2103-2111, 2006 13. Sachelarie I, Grossbard ML, Chadha M, et al: Primary systemic therapy of breast cancer. Oncologist 11:574-589, 2006 Submitted September 13, 2007; accepted November 19, 2007.
Related Correspondence
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
10% positive tumor cells with nuclear staining. HER-2 positivity was defined as either HER2 gene amplification (fluorescent in situ hybridization) or were scored as 3+ (IHC). Nuclear grade was assessed using the modified Black's nuclear grading system.
