SEARCH

SEARCH BY CITATION

Keywords:

  • inflammatory breast cancer;
  • combined therapy;
  • race;
  • African American;
  • treatment adherence

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

BACKGROUND:

The authors compared treatment adherence rates and outcome in Caucasian and African American patients with inflammatory breast cancer (IBC).

METHODS:

The records of 55 (25 Caucasian and 30 African American) IBC patients treated with curative intent from 1995 to 2009 were reviewed. All patients received neoadjuvant doxorubicin (Adriamycin) and/or taxane-based chemotherapy, and mastectomy with or without radiotherapy. The median follow-up period for Caucasian and African American patients was similar (39.5 months and 36.1 months, respectively).

RESULTS:

There was no difference between races in median age, tumor size, grade, and receptor status at diagnosis. The number of patients who completed neoadjuvant chemotherapy, surgery, and radiotherapy did not differ by race (84% of Caucasians vs 86.7% of African Americans) nor did the median length of time to complete trimodality treatment (263 [range, 207-422] days for Caucasians vs 262 [range, 165-371] days for African Americans). There was a trend toward slightly higher pathological complete response rates in Caucasian than African American women (20% in Caucasians vs 6.7% in African Americans, P = .23). Despite slightly better response rates to neoadjuvant chemotherapy, Caucasian patients did not have higher 3-year local control rates (70% in Caucasians vs 64% in African Americans, P = .73). However, there was a trend toward higher 3-year overall survival in Caucasian versus African American patients (73% in Caucasians vs 55% in African Americans, P = .09) and higher distant metastasis-free survival (60% in Caucasians vs 40% in African Americans, P = .19).

CONCLUSIONS:

This study is among the largest to examine patients with IBC by race. Being Caucasian or African American did not appear to impact treatment adherence. However, African American patients tended to have poorer response to standard treatment and worse outcome than Caucasian patients. Cancer 2011;. © 2011 American Cancer Society.

Inflammatory breast cancer (IBC) is an aggressive, rare subtype of locally advanced breast cancer with rapid disease progression and clinical findings of skin erythema, edema (peau d'orange), breast induration, warmth, and asymmetric enlargement. Typically, extensive lymphovascular invasion by tumor emboli is present and involves the superficial dermal plexus of vessels in the papillary and high reticular dermis.1 The National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database shows that IBC accounts for <1% to approximately 5% of all breast malignancies in the United States.2, 3

Patients with IBC are typically younger at diagnosis than patients diagnosed with all other types of breast cancer (median age, 57 vs 64 years).4 IBC patients also have a worse clinical outcome than do other patients with T4 disease.3 Median survival for a patient with IBC is approximately 30 months, and the 10-year survival rate is <30%.3 The advent of multimodality therapy, consisting of neoadjuvant chemotherapy, particularly anthracyclines and taxane-based treatment, surgery, radiotherapy (XRT), and hormonal therapy, has markedly improved outcomes over the past 2 decades.5, 6 Currently, local control rates for patients treated with chemotherapy, mastectomy, and postmastectomy radiation now approach 70% to 80%, and 5-year survival rates are 30% to 40%.5, 6

The proportion of African American women diagnosed with IBC is higher than the proportion of Caucasian women diagnosed with IBC.7 African American breast cancer patients have worse survival rates compared with Caucasian patients in all stages of breast cancer.7, 8 Socioeconomic disadvantages and decreased access to healthcare may contribute to lower survival rates in African Americans, leading to delayed diagnosis and treatment, in addition to treatment noncompliance.9 Later stage breast cancer presentation appears to be the major contributor to the lower survival rates described in African American women.10, 11 However, even after adjusting for stage and age, the biologically aggressiveness of the disease may have a substantial impact on outcome.9 African American women with breast cancer are at increased risk for developing high-grade, estrogen and progesterone receptor-negative disease and having lymph node positivity.12-16 Triple negative tumors have been known to be most prevalent among younger African American women17 and more frequent in premenopausal African American women (39%) compared with postmenopausal African American women (14%).18 African American women also face an increased risk of being diagnosed at a younger age with breast carcinoma.17, 19 However, there is a lack of evidence confirming that African American women with IBC have a poorer prognosis than Caucasian women with IBC because of the overall low incidence of IBC and markedly aggressive disease course. Previous studies of outcome differences between African American and Caucasian IBC patients have analyzed heterogeneous groups of IBC patients (both nonmetastatic and metastatic) treated with different antiquated treatment modalities.2, 8 The goal of our study was to review the outcomes of African American and Caucasian patients with nonmetastatic IBC treated with modern neoadjuvant chemotherapy agents, surgery, and radiation treatment at Emory University in inner city Atlanta.

In this study, we retrospectively analyzed our 15-year single-institution experience in treating nonmetastatic IBC patients. The primary objective of this study was to compare treatment adherence rates and outcome in Caucasian and African American nonmetastatic IBC patients treated with definitive intent. All received standard neoadjuvant doxorubicin (Adriamycin) and/or taxane-based chemotherapy, and mastectomy with or without XRT and endocrine therapy if indicated.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

Study Population

The records of 55 (25 Caucasian and 30 African American) nonmetastatic IBC patients who were treated with curative intent from 1995 to 2009 at Emory University including the Emory Clinic, Emory Midtown Hospital, and Grady Memorial Hospital were analyzed. Patients were staged according to the sixth edition of the American Joint Committee on Cancer Classification and Staging of Tumors.20 Fifty-one patients had stage IIIB (T4d N0/N1/N2a M0) disease, and 4 patients had stage IIIC (T4d N3b/c M0) disease: inflammatory carcinoma (T4d), no regional lymph node metastasis (N0), metastasis in movable ipsilateral axillary lymph node(s) (N1), metastasis in ipsilateral axillary lymph nodes fixed to each other (matted) or to other structures (N2a), metastasis in ipsilateral internal mammary lymph node(s) and axillary lymph node(s) (N3b), metastasis in ipsilateral supraclavicular lymph node(s) (N3c), no distant metastasis (M0). This study was approved by the Emory University Institutional Review Board.

All patients had complete history, physical examination, mammography, and ultrasound for the evaluation of regional nodes and breast tissue. Routine blood cell counts, serum chemistry tests, chest x-ray, computed tomography (CT), and bone scan were performed to detect distant metastasis. Positron emission tomography/CT scan and breast magnetic resonance imaging were performed in some patients as a part of their workup.

The diagnosis of IBC was based on physical examination in all patients. Dermal lymphatic invasion on pathology was seen in 32 patients (58.2%), unknown in 12 patients (21.8%), and absent in 11 patients (20%).

All patients received neoadjuvant doxorubicin (Adriamycin) and/or taxane-based chemotherapy and mastectomy. Because of poor response to neoadjuvant chemotherapy, adjuvant chemotherapy (mostly consisting of Adriamycin- or Taxol-based regimens) was administered to 14 Caucasian and 12 African American women. Clinical response to neoadjuvant chemotherapy was determined via physical examination and radiological studies and defined as either complete in cases where all clinical signs of IBC disappeared or partial response in cases with any improvement in clinical signs.

After neoadjuvant chemotherapy, modified radical mastectomy (MRM) was performed in 23 Caucasian and 28 African American patients. Axillary lymph node dissection was performed in all MRM patients except 2 Caucasian patients with negative sentinel lymph node. Simple mastectomy was performed in 2 Caucasian and 2 African American patients with clinically persistent T4d tumor and/or massive residual cancer in breast after neoadjuvant chemotherapy. A resection margin of ≥2 mm was accepted as negative margin and <2 mm as close/positive margin. A pathologic complete response was defined as no residual invasive carcinoma in the pathology report.

Neoadjuvant chemotherapy was followed by mastectomy in the majority of patients; however, 2 patients with poor response to chemotherapy were treated with preoperative XRT.

Although all patients were prescribed XRT, 5 patients did not receive XRT because of progressive disease and patient refusal (3 Caucasians and 2 African Americans). However, 3 patients with progressive disease received salvage chest wall XRT in their course of treatment. Ten patients in this cohort received their adjuvant XRT at a facility outside of the Emory system. Among patients who received XRT at Emory, 50 gray (Gy) was delivered to the chest wall and draining lymphatics (including internal mammary, axillary, and supraclavicular lymph nodes), followed by a 10- to 16-Gy boost to the scar and/or chest wall in 2-Gy fractions per day. A 10-Gy boost to the supraclavicular fossa was delivered in 1 patient with supraclavicular involvement. Tissue equivalent bolus was placed on the chest wall beginning with the first day of XRT until achieving skin erythema.

In all patients with estrogen receptor-positive tumors, hormone therapy was given (9 Caucasians and 9 African Americans). The number of human epidermal growth factor receptor 2 (HER-2/neu)-positive patients by race was 8 Caucasians versus 9 African Americans. However, trastuzumab was given to more Caucasian (6 of 8) than African American (1 of 9) patients (P = .02) due to the majority of HER-2/neu–positive Caucasian patients being diagnosed and treated after 2004 when trastuzumab use became routine at our institution. The median follow-up period from the date of surgery for Caucasian and for African American patients was similar (39.5 [range, 7-167] and 36.1 [range, 4-108] months, P = .31, respectively).

Patient and tumor characteristics by race are listed in Table 1. Treatment characteristics by race are displayed in Table 2.

Table 1. Patient and Disease Characteristics
CharacteristicCaucasian (n = 25)African American (n = 30)P
  • Abbreviations: BMI, body mass index; HER-2/neu human epidermal growth factor receptor 2; NS, not significant.

  • a

    Statistically significant.

Age at diagnosis, y [range]49 [35-65]52.5 [30-77]NS
 <50 years old13 (52%)13 (43.3%)NS
Menopausal status  .06
 Premenopausal12 (48%)7 (23.3%) 
 Postmenopausal13 (52%)23 (76.7%) 
Family history  NS
 Positive8 (32%)8 (26.7%) 
 Unknown2 (8%)4 (13.3%) 
BMI >25  NS
 Yes19 (76%)19 (63.3%) 
 Unknown2 (8%)9 (30%) 
Comorbidities8 (32%)20 (66.7%).01a
Breast  NS
 Left14 (56%)16 (53.3%) 
 Right11 (44%)14 (46.7%) 
Clinical stage at diagnosis  NS
 IIIB24 (96%)27 (90%) 
 IIIC1 (4%)3 (10%) 
Tumor size at diagnosis   
 >5 cm9 (36%)8 (26.7%)NS
 Unknown10 (40%)13 (43.3%) 
Clinical nodal status at diagnosis  .045a
 N07 (28%)4 (13.3%) 
 N113 (42%)8 (26.7%) 
 N23 (12%)8 (26.7%) 
 N31 (4%)3 (10%) 
 Unknown1 (4%)7 (23.3%) 
Histology  NS
 Invasive ductal22 (88%)24 (80%) 
 Invasive lobular3 (12%)3 (10%) 
 Invasive ductal and lobular0 (0%)3 (10%) 
Tumor grade at biopsy  NS
 10 (0%)2 (6.7%) 
 27 (28%)2 (6.7%) 
 313 (42%)19 (63.3%) 
 Unknown5 (20%)7 (23.3%) 
Estrogen receptor  NS
 Positive9 (36%)9 (30%) 
 Unknown3 (12%)4 (13.3%) 
Progesterone receptor  NS
 Positive7 (28%)5 (16.7%) 
 Unknown3 (12%)4 (13.3%) 
HER-2/neu  NS
 Positive8 (32%)9 (30%) 
 Unknown7 (28%)7 (23.3%) 
Triple negative  NS
 Yes5 (20%)7 (23.3%) 
 Unknown4 (16%)5 (16.7%) 
Table 2. Treatment Characteristics and Response
Characteristic/ResponseCaucasian (n = 25)African American (n = 30)P
  • A, adriamycin; C, cyclophosphamide; T, taxol. HER-2/neu human epidermal growth factor receptor 2; MRM, modified radical mastectomy; NS, not significant.

  • a

    Statistically significant.

Neoadjuvant chemotherapy  NS
 Dose dense (AC/T)9 (36%)3 (10%) 
 Regular (AC/T)15 (60%)11 (36.7%) 
 Unknown1 (4%)16 (53.3%) 
Endocrine therapy in estrogen receptor-positive patients9 (100%)9 (100%)NS
Trastuzumab use in HER-2/neu–positive patients6 (75%)1 (11.1%).02a
Clinical imaging response  NS
 Partial15 (60%)16 (53.3%) 
 Complete8 (32%)3 (10%) 
 Unknown2 (8%)11 (36.7%) 
Pathological response  .23
 Partial20 (80%)28 (93.3%) 
 Complete5 (20%)2 (6.7%) 
Surgery type  NS
 Simple mastectomy2 (8%)1 (3.3%) 
 MRM23 (92%)29 (96.7%) 
ypN stage  .22
 N07 (28%)6 (20%) 
 N16 (24%)6 (20%) 
 N27 (28%)8 (26.7%) 
 N33 (12%)8 (26.7%) 
 Unknown2 (8%)2 (6.7%) 
Extracapsular nodal extension  NS
 Positive7 (43.8%)11 (50%) 
 Unknown1 (6.3%)0 (0%) 
yp T stage  .08
 T06 (24%)4 (13.3%) 
 T19 (36%)6 (20%) 
 T21 (4%)3 (10%) 
 T33 (12%)3 (10%) 
 T46 (24%)13 (43.3%) 
 Unknown0 (0%)1 (3.3%) 
Grade at surgery  .03a
 12 (10.5%)1 (4%) 
 26 (31.6%)6 (24%) 
 34 (21.1%)17 (68%) 
 Unknown7 (36.8%)2 (8%) 
Lymphovascular invasion  NS
 Yes13 (52%)14 (46.7%) 
 Unknown1 (4%)0 (0%) 
Dermal invasion at surgery8 (32%)12 (40%)NS
Close margin at surgery  NS
 Yes0 (0%)4 (13.3%) 
 Unknown1 (4%)0 (0%) 
Adjuvant chemotherapy  NS
 Yes14 (56%)12 (40%) 
 Unknown1 (4%)7 (23.3%) 
Radiotherapy  NS
 Yes21 (84%)26 (86.7%) 
 No3 (12%)2 (6.7%) 
 Unknown1 (4%)2 (6.7%) 
Contralateral breast recurrence0 (0%)4 (13.3%)NS

Statistical Analysis

Patient, disease, and treatment characteristics were analyzed using Statistics Package for the Social Sciences (SPSS, Chicago, Ill). Chi-square test was used to assess categorical variables and Mann-Whitney U test for ordinal variables. Differences in means were compared by independent samples t test. Recurrence-free survival, distant metastasis-free survival, and overall survival (OS) were estimated using the Kaplan-Meier method. A P value <.05 was accepted as statistically significant. All statistical analysis was 2-sided.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

There was no difference between races in median age (49 [range, 35-65] years for Caucasians vs 52.5 [range, 30-77] years for African Americans) or in the proportion of women <40 years or <50 years of age. Receptor status, HER-2/neu status, tumor size, and grade at diagnosis were also similar in both groups. However, Caucasian patients had significantly fewer clinically involved nodes (N2/N3) at diagnosis than African American patients (P = .045). More Caucasian than African American patients were premenopausal (48% vs 23.3%, respectively, P = .06).

A larger proportion of African American than of Caucasian women had significant comorbidities at diagnosis (66.7% vs 32%, P = .01). The most prevalent was hypertension (50% in African Americans vs 16% in Caucasians) and diabetes mellitus (26.7% in African Americans vs 12% in Caucasians). Furthermore, 1 African American woman had a history of liver transplant.

Despite the difference in pre-existing comorbidities, being African American or Caucasian did not appear to impact treatment adherence. The number of patients who completed neoadjuvant chemotherapy, surgery, and XRT did not differ by race (84% in Caucasians vs 86.7% in African Americans), nor did the median length of time to complete trimodality therapy (263 [range, 207-422] days for Caucasians vs 262 [range, 165-371] days for African Americans).

Although there was no difference in initial stage or receptor status at diagnosis, there was a trend toward slightly higher pathological complete response rates in Caucasian than in African American women (20% in Caucasians vs 6.7% in African Americans, P = .23). After neoadjuvant chemotherapy, Caucasian patients were less likely than African American patients to have ypT4 disease (24% in Caucasians vs 43.3% in African Americans, P = .08), ypN3 disease (12% in Caucasians vs 26.7% in African Americans, P = .22), and grade 3 tumors (21.1% in Caucasians vs 68% in African Americans, P = .03).

Despite slightly higher response rates to neoadjuvant chemotherapy, Caucasian patients did not have significantly higher rates of local control at 3 years (70% in Caucasians vs 64% in African Americans, P = .73). However, there was a trend toward higher 3-year distant metastasis-free survival (60% in Caucasians vs 40% in African Americans, P = .19) and OS in Caucasian women (73% in Caucasians vs 55% in African Americans, P = .09). Figures 1 to 3 show recurrence-free survival, distant metastasis-free survival, and OS for the Caucasian and African American patients.

thumbnail image

Figure 1. Recurrence-free survival is shown for the Caucasian (C) and African American (AA) patients.

Download figure to PowerPoint

thumbnail image

Figure 2. Distant metastasis-free survival is shown for the Caucasian (C) and African American (AA) patients.

Download figure to PowerPoint

thumbnail image

Figure 3. Overall survival is shown for the Caucasian (C) and African American (AA) patients.

Download figure to PowerPoint

Although more HER-2/neu–positive Caucasian than African American patients received trastuzumab, trastuzumab treatment on univariate analysis did not appear to influence outcome in this small patient cohort and could not account for the higher OS rate in Caucasian women.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

IBC is a clinical diagnosis based on clinical presentation and the rapidity of onset, which often relies on the history provided by the patient. Diagnosis of IBC may show considerable variability, particularly in women with darker skin, where erythema and peau d'orange may be more difficult to assess or confused with lymphatic congestion. The strength of this study is its large number of IBC cases treated within a single institution with the systematic standardization of clinical and pathologic diagnosis criteria. Furthermore, we limited the study to nonmetastatic IBC patients who received neoadjuvant doxorubicin (Adriamycin) and/or taxane-based chemotherapy and mastectomy to evaluate a more uniform population treated with standard therapies.

This study reported a large, single-institution 15-year experience in nonmetastatic IBC patients treated with an initial intent to deliver multimodality curative treatment. We demonstrated that being African American or Caucasian did not appear to impact treatment adherence. However, similar to previous studies, African American patients had higher nodal burden at diagnosis and tended to have poorer response to standard treatment compared with Caucasian patients.8, 12, 13, 16 However, other adverse prognostic factors such as triple-negative receptor status, HER-2/neu status, and tumor size at diagnosis were similar for African American and Caucasian women. Although poorer response rates to neoadjuvant chemotherapy did not translate into significantly lower rates of local control at 3 years, there was a trend toward poorer 3-year OS and distant metastasis-free survival in African American patients compared with Caucasian patients.

Previous studies of IBC by race have supported our findings. A recent study analyzing the SEER database showed that African American women with IBC had poorer survival than Caucasian women.7 However, these patients were treated with several regimens at different institutions.7 In an analysis of Florida cancer center registry data, treatment variabilities were high between African American and Caucasian IBC patients.8 A significantly greater fraction of Caucasian women underwent surgery than did African American women, and significantly more African American women were treated with chemotherapy alone compared with Caucasian women.8 The authors concluded that when matched for all indicators of prognosis and treatment regimens, African American women still had overall worse survival outcomes.8

In our study, African Americans had significantly higher rates of comorbidities (P = .01), with hypertension and diabetes mellitus being most prevalent. Similar to our results, analysis of Florida cancer center registry data showed that African American IBC patients had nearly twice the rate of diabetes (P = .002), hypertension (P = .001), and obesity (P = .002) than Caucasian patients.8

In our small study, the median age at diagnosis was not different between African American and Caucasian women with IBC. There was also no difference among the 2 groups in the proportion of women aged <40-years or <50 years. Furthermore, Caucasian women were more likely to be premenopausal than African American patients (P = .06) in our patient population. However, several other studies3, 8 have reported that African American IBC women are often diagnosed at a younger age than Caucasian women.

Despite advances in multimodal therapies, IBC remains a therapeutic challenge, with rapid disease progression and early distant metastasis. The etiology, biology, and molecular profile of IBC need further study to understand and overcome its aggressiveness and resistance to current treatment modalities, particularly in African American women.

Conclusions

Our single-institution experience is among the largest to examine IBC patients by race. Being African American or Caucasian did not appear to impact treatment adherence. However, with limited follow-up, there was a trend toward decreased response to standard multimodality treatment and worse outcome in African American patients than in Caucasian patients. Longer follow-up and future studies are warranted investigating the impact of race and new targeted agents in women with IBC.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES