Key Words. Breast cancer Pathological complete response Prognosis Clinical stage

Transcription

1 The Oncologist Breast Cancer Prognostic Value of Initial Clinical Disease Stage After Achieving Pathological Complete Response SHAHEENAH DAWOOD, a,f KRISTINE BROGLIO, b SHU-WAN KAU, a RABIUL ISLAM, a W. FRASER SYMMANS, c THOMAS A. BUCHHOLZ, d SEAN E. MCGUIRE, d FUNDA MERIC-BERNSTAM, e MASSIMO CRISTOFANILLI, a GABRIEL N. HORTOBÁGYI, a ANA M. GONZALEZ-ANGULO a Departments of a Breast Medical Oncology, b Quantitative Sciences, c Pathology, d Radiation Oncology, and e Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas, USA; f Department of Medical Oncology, Dubai Hospital, Dubai, United Arab Emirates Key Words. Breast cancer Pathological complete response Prognosis Clinical stage Disclosure: No potential conflicts of interest were reported by the authors, planners, reviewers, or staff managers of this article. LEARNING OBJECTIVES After completing this course, the reader will be able to: 1. Define survival outcomes in women with early-stage breast cancer who achieve pathological complete response following primary systemic chemotherapy. 2. Define the prognostic value of initial clinical stage in women with breast cancer who achieve pathological complete response following primary systemic chemotherapy. 3. Define survival outcomes in women with inflammatory breast cancer who achieve pathological complete response following primary systemic chemotherapy. CME Access and take the CME test online and receive 1 AMA PRA Category 1 Credit at CME.TheOncologist.com ABSTRACT The aim of this retrospective study was to determine the prognostic impact of initial clinical stage in patients who achieved a pathological complete response (pcr) after receiving primary systemic chemotherapy (PST). Between 1977 and 2006, 489 patients who had achieved a pcr after receiving an anthracyclinebased PST regimen were identified. Recurrence-free survival (RFS) and overall survival (OS) were estimated with the Kaplan Meier product limit method and the differences between groups were compared using the log-rank statistic. Cox proportional hazards models were fit to determine the association of initial clinical stage with survival outcomes after adjusting for patient and tumor characteristics. The median age was 47 years. Twenty (4.1%) patients had stage I disease, 243 (49.7%) had stage II disease, 189 (38.7%) had stage III disease, and 37 (7.5%) had inflammatory breast cancer (IBC). At a median follow-up of 45 months, 59 (12%) patients had experienced disease recurrence. The 5-year RFS and OS rates for the whole cohort were 87.8% and 89.3%, respectively. Lower clinical stage at diagnosis was associated with Correspondence: Ana M. Gonzalez-Angulo, M.D., Department of Breast Medical Oncology, Unit 1354, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas , USA. Telephone: ; Fax: ; agonzalez@mdanderson.org Received June 21, 2007; accepted for publication November 12, AlphaMed Press /2008/$30.00/0 doi: /theoncologist The Oncologist 2008;13:6 15

2 Dawood, Broglio, Kau et al. 7 statistically significant higher RFS and OS rates. In a multivariate model, patients with clinical stage IIIB/C disease and those with IBC had lower RFS rates than patients with clinical stage I/II/IIIA disease. In addition, patients with clinical stage IIIB/C disease and those with IBC had a greater hazard of death than patients with clinical stage I/II/IIIA disease. Overall, patients who achieved a pcr had a low rate of recurrence. However, higher clinical stage and IBC were associated with worse outcomes in breast cancer patients who achieved a pcr after PST. The Oncologist 2008;13: INTRODUCTION It is estimated that 180,510 new cases of breast cancer will have been diagnosed in the year 2007, with approximately 40,910 women expected to die from this disease [1]. With the introduction of anthracycline-based polychemotherapy regimens, 5-year survival rates have significantly improved for patients with early-stage breast cancer [2]. With the introduction of trastuzumab, a monoclonal antibody targeting the human epidermal growth factor receptor (HER)-2/neu receptor, further improvements in survival have been observed among patients with HER-2/neu positive disease [3, 4]. Further improvements in survival will depend on methods targeted at individualizing therapy in order to attain maximum therapeutic efficacy. Primary systemic chemotherapy (PST), a standard approach to treating women with locally advanced and inflammatory breast cancer, is increasingly being used in patients with earlier stage disease. It is associated with a number of advantages, including downstaging of tumors, thereby improving available surgical options and allowing for in vivo assessment of chemosensitivity and thus providing the option to modify chemotherapeutic regimens when necessary. Furthermore, attaining a pathological complete response (pcr) following PST has been shown by a number of investigators [5 9] to be a surrogate marker for superior long-term outcome, thought to result from the eradication of distant mircometastatic invasive residual disease. Taking this phenomenon one step forward, a number of studies have focused on developing regimens to improve the rates of pcr, such as with the addition of taxanes [6] and trastuzumab [10]. In addition, several groups are currently using pcr as a surrogate marker for survival in the design of clinical trials. Investigators have focused on identifying pathological [11] and clinical [12] factors to predict long-term outcomes in patients with residual disease following PST. In an earlier report based exclusively on patients who achieved a pcr after PST, we showed that premenopausal women with stage IIIB disease who had 10 lymph nodes removed for pathological examination had a higher risk for distant recurrence [13]. The aim of the current retrospective study was to confirm, using an updated and larger cohort, our initial findings, and to primarily determine the prognostic impact of initial clinical stage in patients who achieve a pcr. METHODS Patient Population We searched a database maintained at the Breast Medical Oncology Department of The University of Texas M. D. Anderson Cancer Center (MDACC) to retrospectively identify patients with histologically confirmed invasive breast cancer who had received PST and achieved a pcr. Medical charts for all patients were reviewed to confirm the accuracy of clinical variables recorded within the database. All patients had to have received an anthracycline-based chemotherapy regimen either pre- or postoperatively. Exclusion criteria included patients who were male, had stage IV disease, did not receive anthracycline-based chemotherapy, had bilateral disease, or did not undergo definitive surgery. Variables recorded included patient demographics, tumor characteristics, initial clinical stage, administration of taxanes, surgical details, and recurrence information. This retrospective study was approved by the institutional review board of the MDACC. Staging and Pathology Review Clinical stage at diagnosis was based on the staging criteria proposed by the American Joint Committee on Cancer (Sixth Edition) in 2003 [14]. Initial biopsy specimens of all patients were used to confirm invasive carcinoma, categorize histological type according to the World Health Organization classification system [15], and define histological grade according to the modified Black s nuclear grading system [16]. pcr was defined as the complete absence of invasive disease in both the breast and axillary lymph nodes. Operative specimens were reviewed to confirm the absence of invasive disease and to record the presence or absence of residual ductal carcinoma in situ (DCIS). Receptor studies were performed on pretreatment needle biopsy samples. Specimens obtained before 1993 were assessed for hormone receptors (HRs) by the dextran-coated charcoal ligand-binding method to determine estrogen receptor

3 8 Value of Clinical Stage After Pathological Complete Response (ER) and progesterone receptor (PgR) status. Thereafter, receptor status was determined using immunohistochemistry (IHC) on 4- m paraffin-embedded tissue sections stained with the following monoclonal antibodies: (a) 6F11 (Novacastra Laboratories Ltd., Burlingame, CA) for ERs and (b) 1A6 (Novacastra Laboratories Ltd.) for PgRs. HER- 2/neu status was determined, using initial biopsy specimens, by IHC analysis and/or fluorescence in situ hybridization (FISH). A patient whose tumor stained 3 by IHC or was gene amplified by FISH was considered to have a HER-2/neu positive tumor. Patients with HER-2/neu negative tumors had no demonstrable staining by IHC and/or no gene amplification by FISH. Statistical Analysis and Outcome Measures Patient characteristics were tabulated by stage and compared across stage groups with 2, Fisher s exact, or Kruskal-Wallis tests as appropriate. Recurrence-free survival (RFS) was calculated from the date of surgery to the date of local or distant disease recurrence or the date of last follow-up. Patients who died before experiencing disease recurrence were considered censored at their date of death for the analysis of RFS. Overall survival (OS) was calculated from the date of surgery to the date of death from any cause or last follow-up. The median follow-up time was calculated as the median observation time among all patients and among patients still alive at their last follow-up. Survival outcomes were estimated with the Kaplan Meier product limit method and compared between groups with the log-rank statistic. Cox proportional hazard models were fit to determine the association between initial clinical stage and survival outcomes after adjustment for potential confounding patient and tumor characteristics. In the model, clinical stage was considered as three separate groups: (a) stage I/II/ IIIA (operable), (b) stage IIIB/C (nonoperable), and (c) inflammatory. Inflammatory breast carcinoma was considered separately because of its known inherently aggressive biology and poor prognostic outcome. Terms of HR status (ER/PgR positive versus negative), nuclear grade, and menopausal status were included in the model regardless of statistical significance because of their clinical significance. Other variables were considered for inclusion in the model if statistical significance was attained by either the Wald test or the likelihood ratio test (LRT). Analyses were performed with SAS 9.1 (SAS Institute, Inc., Cary, NC) and S-Plus 7.0 (Insightful Corp., Seattle, WA). p-values.05 were considered statistically significant. RESULTS Patient Characteristics Between 1977 and 2006, 489 patients in total who fit the eligibility criteria were identified and included in the analysis. One hundred seventy-two (35.2%) patients were diagnosed before 2000, 227 (46.4%) patients were diagnosed between 2000 and 2004, and 90 (18.4%) patients were diagnosed between 2005 and Patient characteristics crosstabulated by initial clinical stage are summarized in Table 1. All patients received an anthracycline-based chemotherapy regimen, of whom 476 (97.3%) received anthracylines preoperatively and 13 (2.7%) received anthracylines in the postoperative period. Anthracylinebased chemotherapy was administered using one of the following regimens: (a) FEC (500 mg/m 2 5-fluorouracil, 100 mg/m 2 epirubicin, 500 mg/m 2 cyclophosphamide); (b) FAC (500 mg/m 2 5-fluorouracil on days 1 and 4, 50 mg/m 2 doxorubicin i.v. by continuous infusion over 72 hours on days 1 3, 500 mg/m 2 cyclophosphamide on day 1); (c) AC (60 mg/m 2 doxorubicin, 600 mg/m 2 cyclophosphamide); (d) VACP (500 mg/m 2 5-fluorouracil i.v. on day 1, 50 mg/m 2 doxorubicin i.v. by continuous infusion over 72 hours on days 1 3, 500 mg/m 2 cyclophosphamide i.v. on day 1, 1 mg/m 2 vincristine i.v. on day 1, 100 mg/m 2 prednisone orally on days 1 to 5). Three hundred twenty-seven (67%) patients received taxanes (paclitaxel or docetaxel) in either the pre- or postoperative period. Two hundred ninetysix (60.5%) patients received all chemotherapy preoperatively and 193 (39.5%) patients received part of their chemotherapy regimen in the preoperative period and part in the postoperative period. Sixty (12%) patients had trastuzumab as part of their PST. Twenty (4.1%) patients had stage I disease, 243 (49.7%) had stage II disease, 86 (17.6%) had stage IIIA disease, 60 (12.3%) had stage IIIB disease, 43 (8.8%) had stage IIIC disease, and 37 (7.5%) had inflammatory breast cancer. The median age for the whole cohort was 47 years (range, years). The predominant histological subtype was infiltrating ductal carcinoma (96%). Seventy-seven percent and 85% of patients had a negative ER assay and grade III disease, respectively. Information on HER-2/neu status was available for 318 (65%) patients, of whom almost half (44%) had tumors that were HER-2/neu positive, which tended to occur more frequently in patients with higher stage disease. Fifty-four percent of patients underwent a mastectomy, while 46% of patients had breast-conserving surgery. Four hundred one (82%) patients had an axillary lymph node dissection, 84 (17%) patients had a sentinel lymph node biopsy, and four (1%) refused axillary lymph node sampling. The median number of lymph nodes removed was 13 (range, 0 39).

4 Dawood, Broglio, Kau et al. 9 Table 1. Patient characteristics by clinical stage I II III Inflammatory p-value n % n % n % n % n Age at diagnosis (yrs) Minimum Median Maximum Menopausal status Premenopausal 8 40% % % 19 51% Postmenopausal 12 60% % 81 43% 18 49%.20 Race/ethnicity Black 4 20% 35 14% 30 16% 2 5% Hispanic 2 10% 46 19% 29 15% 5 14% White 13 65% % % 27 73% Other 1 5% 14 6% 12 6% 3 8%.78 Histology Other 1 5% 10 4% 9 5% 0 0% Ductal 19 95% % % %.63 Grade I 0 0% 2 1% 2 1% 0 0% II 3 15% 26 11% 26 15% 10 32% III 17 85% % % 21 68%.08 ER Negative 12 67% % % 23 82% Positive 6 33% 50 22% 41 25% 5 18%.55 PgR Negative 10 56% % % 19 79% Positive 8 44% 56 25% 31 21% 5 21%.15 HER-2 Negative 10 67% % 46 46% 8 44% Positive 5 33% 71 38% 53 54% 10 56%.048 Lymphovascular invasion Negative 17 85% % % 9 31% Positive 3 15% 29 12% 30 16% 20 69%.0001 Taxanes No 6 30% 62 26% 80 42% 14 38% Yes 14 70% % % 23 62%.003 Surgery Mastectomy 3 15% % % % Segmental 17 85% % 67 35% 0 0%.0001 n nodes removed Minimum Median Maximum (continued)

5 10 Value of Clinical Stage After Pathological Complete Response Table 1. (Continued) Postsurgery, 23% (n 114) of breast specimens had evidence of residual DCIS. Survival Outcomes The median follow-up times among all patients and among patients still alive were 45 months and 46 months, respectively. At the time of analyses, 59 (12%) patients had experienced disease recurrence and 53 (11%) patients had died. No recurrences were observed among the 20 patients with stage I disease. Six percent (n 16) of patients with stage II disease at diagnosis, 26% (n 26) of patients with stage III disease, and 46% (n 17) of patients with inflammatory breast cancer experienced a recurrence. The Kaplan Meier 5-year RFS and OS rate estimates are summarized in Table 2. Five- and 10-year RFS rate estimates for the whole cohort were 87.8% (95% confidence interval [CI], 84.5% 91.3%) and 82.8% (95% CI, 77.8% 88%), respectively. Five- and 10-year OS rate estimates for the whole cohort were 89.3% (95% CI, 86% 92.8%) and 83.3% (95% CI, 77.9% 89%), respectively. Lower tumor (T), node, and overall clinical stage (Fig. 1) was associated with statistically significant higher RFS and OS rates. The number of lymph nodes removed at the time of definitive surgery was not significantly associated with either RFS (hazard ratio [HR], 0.97; 95% CI, ; p.15) or OS (HR, 0.98; 95% CI, ; p.32). Other factors, such as HR status, HER-2/neu status, and the presence of residual DCIS were not significantly associated with either RFS or OS. A nonsignificant trend toward a longer RFS time was observed among patients who received adjuvant radiation therapy than among those who did not (p.08). Patients who received taxanes also tended to have a longer OS duration (p.04). I II III Inflammatory p-value n % n % n % n % Residual DCIS No 16 80% % % 35 95% Yes 4 20% 64 26% 44 23% 2 5%.046 Adjuvant hormone No 11 55% % % 32 86% Yes 9 45% 69 28% 51 27% 5 14%.08 XRT No 3 15% 61 25% 25 13% 1 3% Yes 17 85% % % 36 97%.001 Abbreviations: DCIS, ductal carcinoma in situ; ER, estrogen receptor; HER-2, human epidermal growth factor receptor 2; PgR, progesterone receptor; XRT, radiotherapy. Table 3 summarizes the results of the multivariate models for RFS and OS. Variables considered in the models included HR status, grade, menopausal status, clinical stage, and radiotherapy. HER-2/neu status was not considered in the model because of the large amount of missing information. In the model for RFS, after adjusting for the included variables, clinical disease stage remained significantly associated with RFS (LRT p.0001). Patients with clinical stage IIIB/C disease (HR, 2.73; 95% CI, ; p.01) and inflammatory breast cancer (HR, 8.83; 95% CI, ; p.0001) had a shorter RFS duration than patients with clinical stage I/II/IIIA disease. In the model for OS, after adjusting for the included variables, clinical disease stage remained significantly associated with OS (LRT p.009). Patients with clinical stage IIIB/C disease (HR, 2.63; 95% CI, ; p.03) and inflammatory breast cancer (HR, 4.68; 95% CI, ; p.001) had a greater hazard of death than patients with clinical stage I/II/ IIIA disease. In the multivariate model, lower grade was significantly associated with longer RFS and OS times. Taxanes did not remain significantly associated with OS in the multivariate setting. The multivariate models for OS and RFS were additionally evaluated after removing the patients who had inflammatory breast cancer, with similar results obtained (data not shown). DISCUSSION As expected, the results of this retrospective study show that the incidence of recurrence in patients with breast cancer who achieve a pcr after PST is small (12%), consistent with previous studies that have reported recurrence rates in the range of 13% 25% [5, 7 9, 13]. The highest rate of recurrence was observed among patients with inflammatory

6 Dawood, Broglio, Kau et al. 11 Table 2. Five-year recurrence-free survival and overall survival estimates Recurrence-free survival Overall survival 5-Year 95% CI p-value 5-Year 95% CI p-value All 87.8% (84.5% 91.3%) 89.3% (86% 92.8%) Age at diagnosis % (78.5% 88.8%) 88.3% (83.9% 92.9%) % (90.4% 97.5%) % (85.3% 96.1%).81 Menopausal status Premenopausal 84.4% (79.4% 89.8%) 87.7% (83% 92.7%) Postmenopausal 91.5% (87.4% 95.8%) % (86.1% 96.1%).45 Race/Ethnicity Black 94.1% (87.6% 100%) 92.8% (86.1% 100%) Hispanic 84.9% (76.2% 94.6%) 78.9% (67.3% 92.6%) Other 84.5% (64.4% 100%) 96.4% (89.8% 100%) White 87.1% (83% 91.5%) % (86.8% 94.6%).61 T category T1 98.6% (95.9% 100%) 98.6% (95.8% 100%) T2 90.5% (86% 95.2%) 94.3% (90.3% 98.4%) T3 96.4% (92.6% 100%) 93.9% (88.2% 100%) T4 67.8% (57.9% 79.4%) % (61.9% 82.8%).0001 N category N0 88.6% (83% 94.6%) 90.8% (85.4% 96.6%) N1 91.3% (86.7% 96.1%) 93.7% (89.6% 98%) N2 89.1% (80.4% 98.6%) 82.9% (72.8% 94.5%) N3 72.9% (60.2% 88.5%) % (67.1% 94.7%).03 Clinical stage I/II 94.4% (91.1% 97.8%) 95.7% (92.4% 99.1%) IIIA 94.0% (88.5% 99.9%) 93.1% (87.4% 99.2%) IIIB 70.3% (58.1% 85.1%) 82.6% (72.9% 93.7%) IIIC 85.4% (69.5% 100%) 85.3% (75.0% 96.9%) Inflammatory 50.8% (35.7% 72.2%) % (60.4% 91.8%).002 Histology Other 93.3% (81.5% 100%) 100.0% (100% 100%) Ductal 88.4% (85% 91.9%) % (85.9% 93.1%).08 Grade I/II 88.5% (79.2% 98.8%) 95.8% (90% 100%) III 89.2% (85.8% 92.8%) % (85.8% 93.6%).25 HR Negative 88.5% (84.2% 93.1%) 89.1% (84.1% 94.4%) Positive 91.7% (86.4% 97.3%) % (88.5% 98.4%).70 HER-2 Negative 90.4% (85.5% 95.6%) 94.7% (90.7% 98.8%) Positive 88.3% (80.7% 96.7%) % (93% 100%).47 Lymphovascular invasion Negative 91.4% (88.3% 94.6%) 90.9% (87.4% 94.5%) Positive 76.1% (65.7% 88.1%) % (75.6% 95.3%).36 (continued)

7 12 Value of Clinical Stage After Pathological Complete Response Table 2. (Continued) Recurrence-free survival breast cancer (46%), while patients with stage I disease in our cohort did not experience relapse. In addition, we showed that higher clinical stage was predictive of shorter RFS and OS times. As shown in the recently updated results from the Early Breast Cancer Trialists Collaborative Group [2], adjuvant anthracycline-based polychemotherapy has substantially decreased the risk for recurrence and death, presumably by eradicating distant micrometastatic disease. The recent addition of trastuzumab to adjuvant regimens for patients with HER-2/neu positive tumors has substantially decreased the risk for recurrence and subsequently improved OS in this cohort [3, 4]. Despite these major advances, a substantial portion of patients will recur, a fact that impacts survival outcomes. Traditional factors predicting for recurrence have included node-positive disease, higher T stage, high nuclear grade, ER-negative disease, and HER-2/neu positive disease [17]. Recurrences are presumably attributed to the inability to assess which patients are resistant to treatment when administered in the adjuvant setting. Hypothetically, the ability to achieve a pcr, indicating high tumor chemosensitivity, would also indicate eradication of all distant invasive micrometastatic disease, with resulting low recurrence rates. Our study demonstrates this phenomenon Overall survival 5-Year 95% CI p-value 5-Year 95% CI p-value Taxanes No 85.7% (80.2% 91.6%) 85.4% (79.7% 91.4%) Yes 88.8% (84.4% 93.4%) % (88.3% 96.5%).04 Surgery Mastectomy 82.0% (76.8% 87.6%) 85.1% (79.9% 90.5%) Segmental 94.8% (91.4% 98.3%) % (90.9% 98.5%).01 Nodes removed % (78.5% 92.8%) 87.4% (80.5% 94.9%) % (86.1% 93.6%) % (87.2% 94.8%).15 Residual DCIS No 86.9% (83% 91%) 88.5% (84.6% 92.5%) Yes 90.8% (84.8% 97.3%) % (85.6% 99.4%).32 Adjuvant hormone No 86.3% (82.2% 90.7%) 88.5% (84.5% 92.8%) Yes 91.5% (86.1% 97.2%) % (85.7% 97.5%).47 XRT No 84.9% (76.9% 93.8%) 89.1% (81.5% 97.3%) Yes 88.4% (84.8% 92.2%) % (85.8% 93.3%).33 Abbreviations: CI, confidence interval; DCIS, ductal carcinoma in situ; HER-2, human epidermal growth factor receptor 2; HR, hormone receptor; N, node; T, tumor; XRT, radiotherapy. to be true, with low recurrence rates seen in the whole cohort. However, recurrence rates showed a progressively increasing trend from 6% to 13% to 46% in patients with stage I/II disease, stage III disease, and inflammatory breast cancer, respectively, which in turn had an impact on survival outcomes. This highlights the fact that initial clinical stage still plays an important prognostic role in this cohort, and that achieving a pcr does not in all instances predict for eradication of micrometastatic disease. One could hypothesize that with advancing stage of disease, the incidence of micrometastatic clones resistant to conventional chemotherapeutic agents would be higher. It could then be further deduced that using additional cytotoxic agents with noncrossresistant properties would be advantageous in this cohort. Interestingly, HR status and HER-2/neu status, both known to be important prognostic and predictive factors, did not predict for survival outcomes once pcr was achieved. Although these results need to be confirmed in a larger prospective cohort, they are hypothesis generating, indicating that, perhaps, in the setting of patients achieving a pcr, these two variables have more of a predictive role rather than a prognostic one. This theory is supported by the results of the retrospective study by Guarneri and colleagues who showed that patients who attained a pcr had a

8 Dawood, Broglio, Kau et al. 13 Figure 1. Overall (A) and recurrence-free (B) Survival curves according to initial clinical stage. better overall outcome than patients who had residual disease after PST regardless of their HR status [18]. It should be noted that, in our cohort of patients, 12% received trastuzumab as part of their PST. Because of the small number of patients involved, there was not sufficient power to detect any differences in survival that may have been contributed by trastuzumab. Conversely, because there was an even distribution of trastuzumab administration among patients with stage I, II, and III disease whose tumors over expressed HER-2/neu, no survival differences could be detected. Inflammatory breast cancer is a rare and aggressive form of locally advanced breast cancer. Before the integration of a multidisciplinary approach into the management schema, inflammatory breast cancer used to be a uniformly fatal disease, with 5% of patients surviving past 5 years [19]. Ueno and colleagues [20] reported on the MDACC experience of inflammatory breast cancer over a 20-year period with a cohort of 178 patients who had received doxorubicin-based PST, of whom 12% had achieved a complete response. In that study, the disease-free survival rate at 15 years was 44% for patients who achieved a complete response to PST, 31% for those who achieved a partial response, and 7% for those who did not respond to PST, thereby highlighting the importance of achieving a complete response even in this very aggressive subset of breast cancer. To our knowledge, we report on the largest cohort of patients with inflammatory breast cancer who have achieved a pcr (n 37), demonstrating higher 10-year RFS and OS rates of 50.8% (95% CI, 35.7% 72.2%) and 64.7% (95% CI, 48.4% 86.4%), respectively (data not shown). Despite these impressive results in our cohort, 46% of those with inflammatory breast cancer relapsed. With recent studies indicating that the addition of trastuzumab to adjuvant chemotherapy regimens improves disease-free survival and OS [3, 4] in patients whose tumors overexpress HER-2/neu, and the known higher incidence of HER-2/neu [21] in patients with IBC, incorporation of trastuzumab in this setting may translate into better outcomes. A factor limiting the comparison of pcr rates across different studies is the varied definition of pcr used. In an attempt to standardize the definition, the International Expert Panel on the Use of Neoadjuvant Systemic Treatment has defined pcr as the absence of all evidence of residual disease, including in situ disease, in either the breast or axillary lymph nodes [22]. Interestingly, results from an earlier study conducted at the MDACC [23] retrospectively looking at 2,000 patients who had received PST found that the presence of only residual DCIS did not affect survival outcomes. Our results confirm these findings and indicate that our definition of pcr that incorporates only the absence of invasive disease may be sufficient. However, these results need to be validated in a larger prospective cohort with longer term follow-up. Furthermore, because the definition of pcr also relies on the exclusion of residual cancer, the accuracy and extent of pathologic sampling are critical. Several recent advances in clinical practice are expected to improve the accuracy of this histopathologic definition by directing pathologists to sites of possible residual disease. These include the communication of disease site and treatment history through electronic medical records, placement of metallic indicators in the tumor bed, targeted surgical procedures (segmental mastectomy, sentinel lymph node biopsy), specimen photography and/or radiography, increased pathologist experience with post-treatment specimens, and more detailed histopathologic sampling protocols [24]. In conclusion, although achieving a pcr improves long-term outcomes overall, resulting in a low event rate, initial clinical stage still plays an important prognostic role, with patients with a higher clinical stage at a greater risk for recurrence. Whether or not patients with a higher clinical stage who achieve a pcr would benefit from additional adjuvant systemic therapy is an important research question for future clinical trials.

9 14 Value of Clinical Stage After Pathological Complete Response Table 3. Multivariate models for overall and recurrence-free survival Hazard ratio 95% CI p-value Overall survival IIB/C versus I/II/IIIA Inflammatory versus I/II/IIIA HR (positive versus negative) Grade (III versus I/II) Menopausal status (post- versus pre-) XRT (yes versus no) Recurrence-free survival IIIB/C versus I/II/IIIA Inflammatory versus I/II/IIIA HR (positive versus negative) Grade (III versus I/II) Menopausal status (post- versus pre-) XRT (yes versus no) Abbreviations: CI, confidence interval; HR, hormone receptor; XRT, radiotherapy. ACKNOWLEDGMENTS This work was supported in part by the Susan G. Komen Foundation, the Nellie B. Connally Fund for Breast Cancer Research, and the Inflammatory Breast Cancer Research Group. A.M.G.-A. is supported by K23CA and ASCO Career Development Award. G.N.H. is supported by DAMD , 2P30 CA (PP-4), and P50 CA REFERENCES 1 Jemal A, Siegel R, Ward E et al. Cancer statistics, CA Cancer J Clin 2007;57: Early Breast Cancer Trialists Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: An overview of the randomised trials. Lancet 2005; 365: Piccart-Gebhart MJ, Procter M, Leyland-Jones B et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005;353: Romond EH, Perez EA, Bryant J et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005;353: Kuerer HM, Newman LA, Smith TL et al. Clinical course of breast cancer patients with complete pathologic primary tumor and axillary lymph node response to doxorubicin-based neoadjuvant chemotherapy. J Clin Oncol 1999;17: AUTHOR CONTRIBUTIONS Conception/design: Shaheenah Dawood, Ana M. Gonzalez-Angulo Provision of study materials or patients: Shaheenah Dawood, Shu-Wan Kau, Ana M. Gonzalez-Angulo Collection/assembly of data: Shaheenah Dawood, Shu-Wan Kau, Rabiul Islam, Sean E. McGuire, Ana M. Gonzalez-Angulo Data analysis and interpretation: Shaheenah Dawood, Kristine Broglio, Ana M. Gonzalez-Angulo Manuscript writing: Shaheenah Dawood, W. Fraser Symmans, Thomas A. Buchholz, Sean E. McGuire, Funda Meric-Bernstam, Massimo Cristofanilli, Gabriel N. Hortobágyi, Ana M. Gonzalez-Angulo Final approval of manuscript: Rabiul Islam, W. Fraser Symmans, Thomas A. Buchholz, Sean E. McGuire, Funda Meric-Bernstam, Massimo Cristofanilli, Gabriel N. Hortobágyi, Ana M. Gonzalez-Angulo 6 Bear HD, Anderson S, Smith RE et al. Sequential preoperative or postoperative docetaxel added to preoperative doxorubicin plus cyclophosphamide for operable breast cancer. National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol 2006;24: Wolmark N, Wang J, Mamounas E et al. Preoperative chemotherapy in patients with operable breast cancer: Nine-year results from the National Surgical Adjuvant Breast and Bowel Project B-18. J Natl Cancer Inst Monogr 2001;(30): Fisher B, Bryant J, Wolmark N et al. Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 1998; 16: Chollet P, Amat S, Cure H et al. Prognostic significance of a complete pathological response after induction chemotherapy in operable breast cancer. Br J Cancer 2002;86: Buzdar AU, Ibrahim NK, Francis D et al. Significantly higher pathologic complete remission rate after neoadjuvant therapy with trastuzumab, paclitaxel, and epirubicin chemotherapy: Results of a randomized trial in human epidermal growth receptor 2-positive operable breast cancer. J Clin Oncol 2005;23: Symmans W, Peintinger F, Hatzis C et al. A new measurement of residual cancer burden to predict survival after neoadjuvant chemotherapy. J Clin Oncol 2006;24(18 suppl):536.

10 Dawood, Broglio, Kau et al Jeruss JS, Mittendorf EA, Tucker SL et al. Should pathologic response to neoadjuvant therapy be incorporated into the AJCC staging system for breast cancer? Breast Cancer Res Treat 2006:100(suppl);S Gonzalez-Angulo AM, McGuire SE, Buchholz TA et al. Factors predictive of distant metastases in patients with breast cancer who have a pathologic complete response after neoadjuvant chemotherapy. J Clin Oncol 2005;23: Singletary SE, Allred C, Ashley P et al. Staging system for breast cancer: Revisions for the 6th edition of the AJCC Cancer Staging Manual. Surg Clin North Am 2003;83: The World Health Organization Histological Typing of Breast Tumors Second Edition. The World Organization. Am J Clin Pathol 1982;78: Black MM, Speer FD. Nuclear structure in cancer tissues. Surg Gynecol Obstet 1957;105: Cianfrocca M, Goldstein LJ. Prognostic and predictive factors in earlystage breast cancer. The Oncologist 2004;9: Guarneri V, Broglio K, Kau SW et al. Prognostic value of pathologic complete response after primary systemic chemotherapy in relation to hormone receptor status and other factors. J Clin Oncol 2006;24: Giordano S, Hortobagyi G. Locally advanced breast cancers: Role of medical oncology. In: Bland KI, Copeland EM, eds. The Breast, Third Edition. Philadelphia: Saunders Publishing, 2004: Ueno NT, Buzdar AU, Singletary SE et al. Combined-modality treatment of inflammatory breast carcinoma: Twenty years of experience at M. D. Anderson Cancer Center. Cancer Chemother Pharmacol 1997;40: Turpin E, Bieche I, Berthaeau P et al. Increased incidence of ERBB2 overexpression and TP53 mutation in inflammatory breast cancer. Oncogene 2002;21: Kaufmann M, Hortobagyi GN, Goldhirsch A et a. Recommendations from an international expert panel on the use of neoadjuvant (primary) systemic treatment of operable breast cancer: An update. J Clin Oncol 2006;24: Mazouni C, Peintinger F, Kau SW et al. Effect on patient outcome of residual DCIS in patients with complete eradication of invasive cancer after neoadjuvant chemotherapy. J Clin Oncol 2007;25: Kaufmann M, von Minckwitz G, Smith R et al. International expert panel on the use of primary (preoperative) systemic treatment of operable breast cancer: Review and recommendations. J Clin Oncol 2003;21: