SEARCH

SEARCH BY CITATION

Keywords:

  • breast cancer;
  • chemotherapy;
  • metastasis;
  • cancer stem cell;
  • multivariate analysis

Abstract

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

BACKGROUND:

Aldehyde dehydrogenase 1 (ALDH1)-positive cells exhibit stem-like or progenitor ability and have been considered a clinically important diagnostic and therapeutic target in patients with breast cancer. In this study, the authors evaluated responsiveness to chemotherapy of ALDH1-positive cells in primary and metastatic lesions and its relation to prognosis for patients with lymph node-positive breast cancer.

METHODS:

In total, 115 patients who had breast cancer with cytologically confirmed lymph node metastases and who underwent surgery after neoadjuvant chemotherapy (NAC) were evaluated. By using ALDH1 immunohistochemistry in core-needle biopsy specimens of the primary tumor, cytology samples of axillary lymph nodes before NAC, and pathologic samples of each after NAC, the clinical significance of ALDH1-positive cell status was evaluated in primary and metastatic lesions before and after NAC.

RESULTS:

The presence of ALDH1-positive cancer cells, but not ALDH1-negative cancer cells, in primary and metastatic lesions after NAC was associated with a worse prognosis. In multivariate analysis, only ALDH1-positive cells in metastatic lesions after NAC correlated with overall survival. The responsiveness of ALDH1-positive cells to chemotherapy differed between primary and metastatic lesions, and the findings indicated that ALDH1-positive cells in metastatic lesions after NAC may clinically precede those in the primary lesion.

CONCLUSIONS:

The responsiveness of ALDH1-positive cells to chemotherapy in primary and metastatic lesions and its prognostic significance were clarified in patients with breast cancer. The authors concluded that ALDH1-positive status may represent a surrogate marker as a new concept in patients with lymph node-positive breast cancer. Cancer 2012. © 2011 American Cancer Society.


INTRODUCTION

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

The cancer stem cell theory that was established for hematopoietic neoplasms1 has significantly changed the concept of cancer therapy.2-10 In solid tumors, small populations of cancer cells that demonstrate tumorigenic (tumor-initiating) ability were identified first in cluster of differentiation 44 (Indian blood group) (CD44)-positive/heat stable antigen (CD24)-negative breast cancer cells.11 This subpopulation of cells, which demonstrated the probability of stem-like characteristics in cancer tissue, was called cancer stem cells, and cell populations with comparable ability were detected subsequently in various other tumors.3, 6, 10, 12-15 Today, these subpopulations with tumorigenic ability are considered some of the most important diagnostic and therapeutic targets in treating cancer.15-18 Li et al provided clinical evidence of the chemoresistance of CD44-positive/CD24-negative cells in patients with breast cancer.19

Recently, Ginestier et al demonstrated that aldehyde dehydrogenase 1 (ALDH1)-positive cells display stem-like or progenitor abilities in both normal breast epithelium cells and breast cancer cells.20 At the same time, they demonstrated the clinical significance of these cells, because patients who had breast cancer with ALDH1-positive cells had a worse prognosis than patients who had breast cancer without such cells.20 Tanei et al reported that tumors that had ALDH1-positive cells from patients with breast cancer displayed higher rates of resistance to chemotherapy.21 Furthermore, because ALDH1-positive cells exhibit tumorigenic ability and have prognostic significance, ALDH1-positive status may be involved not only in proliferation of the primary tumor but also in the formation of metastases.20

Given these basic and clinical implications, it has been speculated that ALDH1-positive cells in primary tumors and metastases represent some of the most clinically important diagnostic and therapeutic targets in patients with breast cancer.20, 22, 23 However, the responsiveness of ALDH1-positive cells to chemotherapy in primary and metastatic lesions and how that responsiveness is correlated with the prognosis of patients with breast cancer remain unclear. In the current study, we used core-needle biopsies (CNBs) from primary tumors and cytologic samples of axillary lymph nodes before patients received neoadjuvant chemotherapy (NAC), and we also used pathologic samples of both primary tumors and axillary lymph nodes after patients received NAC to evaluate the responsiveness to chemotherapy of ALDH1-positive cells in primary and metastatic lesions and the correlation of that responsiveness to prognosis in patients with lymph node-positive breast cancer.

MATERIALS AND METHODS

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

Patients

We examined paraffin blocks with tumor samples from 115 patients with lymph node-positive breast cancer who had cytologically confirmed axillary lymph node metastases and who had undergone surgery after NAC at Chiba University Hospital between November 2002 and February 2008. We selected patients with histologically proven, clearly invasive breast carcinoma who had undergone axillary lymph node dissection and for whom all resected materials were available for histologic examination. Because the response of lymph node metastases after NAC was not evaluated after level II or III lymph node dissection in patients who had supraclavicular or parasternal metastases, these patients were excluded.

Primary tumors were pathologically confirmed based on CNB, and positive axillary lymph node metastases were confirmed by fine-needle aspiration biopsy. No patients had received preoperative radiation or hormone therapy. Follow-up was available for all 115 patients; 29 patients developed distant metastasis, and 22 patients died of breast cancer. Survival data were collected up to May 1, 2011 (median follow-up, 71 months; range, 39-102 months). Patient outcomes were defined as overall survival (OS). The duration of OS was determined as the time between the start of NAC and the date of death.

We histologically examined all CNB specimens and resected primary tumors and axillary lymph node metastases. All carcinomas were classified according to the criteria of the World Health Organization. The combined histologic grade (1, 2, or 3) of infiltrating ductal carcinoma was obtained according to a modified Scarff-Bloom-Richardson histologic grading system with guidelines as suggested by Nottingham City Hospital pathologists. Nuclear grading was based on nuclear polymorphism. Staging at the time of diagnosis was based on the tumor-lymph node-metastasis (TNM) classification. Tumor size and lymph node status were evaluated separately. The study protocol was approved by the ethics committee of our institute, and written informed consent was obtained from all patients before participation.

Local and Systemic Therapy

All axillary lymph node metastases, including infraclavicular lymph node metastases, were diagnosed by ultrasound-guided cytology at the initial visit. All patients received neoadjuvant doxorubicin (60 mg/m2) plus cyclophosphamide (600 mg/m2) at 3-week intervals for 4 cycles; then, they received paclitaxel (80 mg/m2) at weekly intervals for 12 cycles. Within 4 weeks after NAC, all patients underwent breast surgery. Patients with infraclavicular metastases underwent level III lymph node dissection, and patients without infraclavicular metastases underwent level II lymph node dissection. All patients with residual axillary metastases received radiotherapy to the chest wall and supraclavicular lymph nodes. For all estrogen receptor (ER)-positive patients, tamoxifen (tamoxifen citrate) or an aromatase inhibitor was prescribed as adjuvant treatment regardless of age or any other prognostic factor.

Pathologic Examination and Assessment of Dissected Primary Tumors and Lymph Nodes

Pathologic response of primary tumors and lymph nodes to NAC was assessed for all patients. All lymph nodes were sliced into multiple serial sections of approximately 2 mm thickness. These sections were stained with hematoxylin and eosin (H&E) and evaluated by pathologists. In each primary tumor and axillary lymph node, residual cancer cells were evaluated. A pathologic complete response (pCR) was defined as the absence of residual cancer cells in both the primary lesion and axillary lymph nodes.

Immunohistochemical Staining of ALDH1

For ALDH1 immunostaining, the paraffin-embedded sections were deparaffinized in xylene and rehydrated in graded alcohol. Antigen enhancement was achieved by incubating the sections in citrate buffer, pH 6.0 (DAKO Japan, Kyoto, Japan), as recommended. ALDH1 antibody (monoclonal, immunoglobulin [Ig]G isotype; BD Biosciences, Piscataway, NJ) was used at 1:200 dilution. Staining was performed using the Envision Kit (DAKO Japan) according to the manufacturer's protocols. Paraffin sections of normal liver tissue were used as a positive control for ALDH1 antibody staining (Fig. 1a). Immunohistochemical staining was evaluated independently by 3 investigators (T.F., M.S., and M.N.) who were blinded to the clinical outcomes. At the same time as the ALDH1 immunostaining, other serial sections were examined for cytokeratin (AE1/AE3; Dako Cytomation, Kyoto, Japan) using immunohistochemistry, and we excluded staining of stroma and lymphocytes in primary tumors and lymph nodes. The cytoplasmic staining of cancer cells was considered ALDH1-positive. By using the criteria described by Ginestier et al,20 immunohistochemical staining of ALDH1 was classified as 3+ (≥50% positive tumor cells), 2+ (<50% and ≥10% positive tumor cells), 1+ (<10% and ≥5% positive tumor cells), or negative (<5% positive tumor cells). For subsequent analyses, tumors that had 1+, 2+, or 3+ expression of ALDH1 were considered ALDH1-positive. ALDH1 immunostaining was performed for all CNB samples before NAC and in all pathologic samples of primary tumors and axillary lymph nodes after NAC (Fig. 1b-e).

thumbnail image

Figure 1. These photomicrographs reveal immunohistochemical and immunocytochemical staining for aldehyde dehydrogenase 1 (ALDH1) in specimens from patients with breast cancer. (a) The liver exhibited strong ALDH1 cytoplasmic immunoreactivity as a positive control. Immunohistochemical staining was performed in residual primary tumor after neoadjuvant chemotherapy (NAC) and produced (b) ALDH1 negative results and (c) ALDH1-positive (3+) results, and immunostaining of residual axillary metastases after NAC produced (d) ALDH1 negative results and (e) ALDH1-positive (3+) results. (f) Liver cells collected by fine-needle aspiration exhibited strong ALDH1 immunoreactivity as a positive control for immunocytochemical staining. Immunocytochemical staining was performed for cell-transferred smears before NAC and produced (g) ALDH1 negative results and (h) ALDH1-positive results (original magnification, ×400 in a-h).

Download figure to PowerPoint

Histologic Grade and Estrogen Receptor, Progesterone Receptor, Human Epidermal Growth Factor Receptor 2, and Ki-67 Status

Routine histologic examination was performed with H&E staining. Histologic grade was determined using the Scarff-Bloom-Richardson grading system. ER and progesterone receptor (PR) status was defined as positive when ≥1% of tumor cells showed had immunohistochemical staining (ER, clone 1D5; PR, clone PgR636; DAKO Japan). Staining for human epidermal receptor 2 (HER2) was scored into 4 grades (0, 1, 2, or 3), and tumors with scores of 3+ were considered HER2-positive. Tumors with scores of 2+ were determined by fluorescence in situ hybridization using PathVision HER2 DNA Probe kits (SRL Inc., Scottsdale, Ariz). Tumors that contained >2 genes per cell also were considered HER2-positive. Ki-67 was defined as positive when ≥15% of tumor cells had positive immunohistochemical staining (clone MIB-1, IgG isotype; DAKO Japan).

Cell Transfer Technique

Coverslips from Papanicolaou-stained smears were removed by soaking in xylene, and the smear was then covered with Mount-Quick liquid medium (Newcomer Supply, Middleton, Wis). After microwaving for 5 minutes to polymerize the liquid plastic, the slides were soaked in warm water, and the plastic film containing smear-medium complex was peeled and lifted from the original glass slide. After marking by the cytopathologist, the plastic film was cut with scissors and transferred onto several positively charged glass slides with the original orientation of cells to the slide still intact. Those glass slides, with the divided smear sections, were placed in the microwave for 5 minutes to secure adhesion. The Mount-Quick medium was then removed in 3 changes of xylene for up to 30 minutes and rehydrated through a series of ethanol to distilled water. Slides were then placed in phosphate-buffered saline.

Immunocytochemical Staining of ALDH1

Smear sections were incubated at room temperature for 1 hour with ALDH1 antibody at 1:300 dilution. Staining was achieved using the Envision Kit according to the protocols of the manufacturer. Fine-needle aspiration biopsy of liver tissue was used as a positive control for ALDH1 antibody staining (Fig. 1f). ALDH1 immunostaining was performed for these smear sections (Fig. 1g,h).

Statistical Analysis

Statistical analysis was performed using software. The OS duration was calculated using the Kaplan-Meier method and was compared using log-rank tests. Associations between the prevalence of ALDH1-positive cells and clinical parameters were assessed using the chi-square test. A Cox proportional hazards regression model was used for univariate and multivariate analyses. P values < .05 were considered statistically significant.

RESULTS

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

ALDH1-Positive Cancer Cells, Not ALDH1-Negative Cancer Cells, in Primary Tumor After Chemotherapy Were a Prognostic Factor in Patients With Lymph Node-Positive Breast Cancer

In total, 115 patients with breast cancer who were diagnosed preoperatively as lymph node positive by cytology underwent surgery after NAC. Of these, 16 patients (13.9%) had no residual primary tumor, and 99 patients (86.1%) had residual primary tumor after NAC. No significant difference in OS was observed between patients with and without residual tumor after NAC (P = .056) (Fig. 2a). To evaluate ALDH1-positive cells in primary tumor before NAC, ALDH1 immunohistochemistry was performed in CNB specimens from all patients. Of the 115 patients, the ALDH1 staining grade in primary tumors before NAC was negative in 70 patients (60.9%), 1+ in 32 patients (27.8%), 2+ in 9 patients (7.8%), and 3+ in 4 patients (3.5%). Forty-five patients who had grade 1+ or greater ALDH1 staining were considered ALDH1-positive. Of all 115 patients, 45 patients (39.1%) had ALDH1-positive cells in their primary tumor, and 70 patients (60.9%) did not. Furthermore, to evaluate the prognostic significance of ALDH1-positive cells in primary tumor after NAC, ALDH1 immunohistochemistry was performed on surgical specimens from all patients. Of 70 patients who had ALDH1-positive cells (−) in CNB specimens before NAC, only 1 patient had ALDH1-positive cells in residual primary tumor after NAC. The other patients had ALDH1-positive cells (−) or had no residual tumor. After NAC, 35 patients (30.4%) had ALDH1-positive cells in residual primary tumor, and 80 patients (69.6%) did not. OS was significantly shorter in patients who had ALDH1-positive cells (+) in residual primary tumor after NAC than in those with ALDH1-positive cells (−) (P < .0001) (Fig. 2b). Finally, of all 115 patients, 15 patients (13.1%) had no residual tumor, 65 patients (56.5%) had residual tumor with ALDH1-positive cells (−), and 35 patients (30.4%) had residual tumor with ALDH1-positive cells (+) (Table 1). OS was significantly shorter in patients who had residual tumor with ALDH1-positive cells (+) than in patients who had no residual tumor (P = .0036) (Fig. 2c). It is noteworthy that the OS of patients who had residual tumor with ALDH1-positive cells (−) was significantly longer than that in patients who had ALDH1-positive cells (+) (P < .0001), and their OS did not differ significantly compared with the OS of patients who had no residual tumor (P = .314) (Fig. 2c). These data indicate that, in residual primary tumor after NAC, ALDH1-positive cancer cells, but not ALDH1-negative cancer cells, represent a prognostic factor in patients with lymph node-positive breast cancer.

thumbnail image

Figure 2. The clinical significance of residual pathologic tumor and aldehyde dehydrogenase 1 (ALDH1)-positive cells was examined in primary lesions from patients with breast cancer after they received neoadjuvant chemotherapy (NAC). (a) In pathologic analyses of primary lesions after NAC, overall survival (OS) did not differ significantly between patients who had residual primary tumor (orange line) and patients who had no residual primary tumor (blue line) (P = .056). (b) In an ALDH1 immunohistochemical analysis of primary lesions after NAC, the OS of patients who had ALDH1-positive cells (+) (red line) was significantly shorter than the OS of patients who had ALDH1-positive cells (−) (light blue line) (P < .0001). (c) In evaluations that considered both pathologic residual tumor and ALDH1-positive cells in primary lesions after NAC, the OS of patients who had residual primary tumor with ALDH1-positive cells (+) (red line) was significantly shorter than the OS of patients who had no residual primary tumor (blue line) (P = .0036). The OS of patients who had residual primary tumor with ALDH1-positive cells (−) (purple line) was significantly longer than the OS of patients who had ALDH1-positive cells (+) (red line) (P < .0001), but there was no significant difference compared with the OS of patients who had no residual primary tumor (blue line) (P = .314).

Download figure to PowerPoint

Table 1. Aldehyde Dehydrogenase 1 Status in Primary Tumor and Axillary Metastases Before and After Neoadjuvant Chemotherapy (n = 115)
AssessmentNo. of Patients
  1. Abbreviations: ALDH1, aldehyde dehydrogenase 1; CNB, core needle biopsy; NAC, neoadjuvant chemotherapy.

Before NAC 
  Primary tumor (CNB sample) 
  ALDH1-positive cells (+)45
  ALDH1-positive cells (−)70
After NAC (surgical specimen) 
 Primary tumor 
  Residual cancer cells with ALDH1-positive cells (+)35
  Residual cancer cells with ALDH1-positive cells (−)65
  No residual cancer cells15
 Axillary metastases 
  Residual cancer cells with ALDH1-positive cells (+)19
  Residual cancer cells with ALDH1-positive cells (−)53
  No residual cancer cells43

ALDH1-Positive Cancer Cells, Not ALDH1-Negative Cancer Cells, in Axillary Metastases After Chemotherapy Were a Prognostic Factor in Patients With Lymph Node-Positive Breast Cancer

All 115 patients were diagnosed preoperatively with lymph node-positive disease by cytology. In these patients, we evaluated residual metastases after NAC. There were 43 patients (37.4%) who had no residual metastases after NAC and 72 patients (62.6%) who exhibited residual metastases. OS was significantly shorter in patients with residual metastases after NAC than in those with no residual metastases (P = .0019) (Fig. 3a). To evaluate the prognostic significance of ALDH1-positive cells in axillary lymph node metastases after NAC, ALDH1 immunohistochemistry was performed in surgical specimens of all axillary lymph nodes. Nineteen patients (16.5%) had ALDH1-positive cells in residual axillary metastases, and 96 patients (83.5%) did not. The OS of patients who had ALDH1-positive cells (+) in residual axillary metastases after NAC was significantly shorter than the OS of patients who had ALDH1-positive cells (−) (P < .0001) (Fig. 3b). Finally, of all 115 patients, 43 patients (37.4%) had no residual axillary metastases, 53 patients (46.1%) had residual axillary metastases with ALDH1-positive cells (−), and 19 patients (16.5%) had residual axillary metastases with ALDH1-positive cells (+) (Table 1). The OS of patients who had residual axillary metastases with ALDH1-positive cells (+) was significantly shorter than the OS of patients who had no residual axillary metastases (P < .0001) (Fig. 3c). It is noteworthy that the OS of patients who had residual axillary metastases with ALDH1-positive cells (−) was significantly longer than the OS of patients who had ALDH1-positive cells (+) (P < .0001), and their OS did not differ significantly compared with the OS of patients who had no residual axillary metastases (P = .188) (Fig. 3c). These data indicate that, in residual axillary metastases after NAC, ALDH1-positive cancer cells, and not ALDH1-negative cancer cells, are a prognostic factor for patients with lymph node-positive breast cancer.

thumbnail image

Figure 3. These charts illustrate the clinical significance of residual pathologic metastases and aldehyde dehydrogenase 1 (ALDH1)-positive cells in axillary lesions after neoadjuvant chemotherapy (NAC) in patients with breast cancer. (a) In pathologic analyses of axillary lesions after NAC, the overall survival (OS) of patients who had residual axillary metastases (orange line) was significantly shorter than that of patients who had no residual axillary metastases (blue line) (P = .0019). (b) In an ALDH1 immunohistochemical analysis of axillary lesions after NAC, the OS of patients who had ALDH1-positive cells (+) (red line) was significantly shorter than the OS of patients who had ALDH1-positive cells (−) (light blue line) (P < .0001). (c) In evaluations that considered both pathologic residual metastases and ALDH1-positive cells in axillary lesions after NAC, the OS of patients who had residual axillary metastases with ALDH1-positive cells (+) (red line) was significantly shorter than the OS of patients who had no residual axillary metastases (blue line) (P < .0001). The OS of patients who had residual axillary metastases with ALDH1-positive cells (−) (purple line) was significantly longer than that in patients with ALDH1-positive cells (+) (red line) (P < .0001), and was not significantly different from that in patients with no residual axillary metastases (blue line) (P = .188).

Download figure to PowerPoint

ALDH1-Positive Cells in Axillary Metastases After Chemotherapy Were an Independent Prognostic Factor in Multivariate Analysis

Next, we evaluated the difference in prognostic significance between ALDH1-positive cancer cells in residual primary tumor after NAC and that in residual axillary metastases. Of all 115 patients, 35 patients (30.4%) had ALDH1-positive cells in residual primary tumor after NAC, and 80 patients (69.6%) did not. The clinicopathologic status of these 2 groups is provided in Table 2. Patients who had ALDH1-positive cells (+) in residual primary tumor after NAC had characteristics of younger age (P = .0412), pCR (P = .0319), and ALDH1-positive cells (+) in residual axillary metastases (P < .0001) compared with patients who had ALDH1-positive cells (−) (Table 2). Of all 115 patients, 19 patients (26.4%) had ALDH1-positive cells in residual axillary metastases after NAC, and 96 patients (73.6%) did not. The clinicopathologic status of these 2 groups is provided in Table 3. Patients who had ALDH1-positive cells (+) in residual axillary metastases after NAC had characteristics of younger age (P = .0319), lymphovascular space invasion (LVSI) (P = .0032), ALDH1-positive cells (+) in the primary tumor before NAC (P < .0001), and ALDH1-positive cells (+) in residual primary tumor (P < .0001) compared with patients who had ALDH1-positive cells (−) (Table 3). To evaluate ALDH1-positive cells in residual primary tumor and residual axillary metastases as a potential independent prognostic factor after NAC in patients with lymph node-positive breast cancer, univariate and multivariate analyses were performed for correlations between OS and various clinical characteristics, including age, tumor (T) classification, lymph node metastasis before NAC (level III vs level I-II), the number of lymph node metastasis after NAC, LVSI, histologic grade, ER status, PR status, HER status, Ki-67 status, pCR, ALDH1-positive cells in primary tumor before NAC, ALDH1-positive cells in residual primary tumor, and ALDH1-positive cells in residual axillary metastases. In univariate analyses, LVSI (P = .006), pCR (P = .0006), ALDH1-positive cells in primary tumor before NAC (P < .0001), ALDH1-positive cells in residual primary tumor (P < .0001) and ALDH1-positive cells in residual axillary metastases (P < .0001) were correlated significantly with OS (Table 4). Furthermore, multivariate analyses indicated only ALDH1-positive cells in residual axillary metastases (P = .006) correlated with OS (Table 4). These data indicate that ALDH1-positive cells in residual axillary metastases represent an independent prognostic factor after NAC in patients with lymph node-positive breast cancer.

Table 2. Characteristics of Patients With Aldehyde Dehydrogenase 1 (ALDH1)-Positive Cells (+) and ALDH1-Positive Cells (−) in Residual Primary Tumor
 No. of Patients 
CharacteristicALDH1- Positive Cells (−), n = 80ALDH1- Positive Cells (+), n = 35P
  • Abbreviations: ER, estrogen receptor; Her2, human epidermal growth factor 2 receptor; LVSI, lymphovascular space invasion; NAC, neoadjuvant chemotherapy; pCR, pathologic complete response; PgR, progesterone receptor.

  • a

    P < .05.

Age, y  .0412a
 <503910 
 >504125 
Tumor classification  .0829
 T1102 
 T2-T36023 
 T41010 
Lymph node metastases  .8054
 Level I-II6027 
 Level III208 
No. of lymph node metastases after NAC  .4863
 03813 
 1-32813 
 ≥4149 
LVSI  .3618
 Present2514 
 Absent5521 
Histologic grade  .4055
 1-24623 
 33412 
ER status  .7419
 Positive5023 
 Negative3012 
PgR status  .8185
 Positive4318 
 Negative3717 
Her2 status  .1077
 Positive287 
 Negative5228 
Ki-67 status  .5024
 Positive5120 
 Negative2915 
pCR  .0319a
 Yes141 
 No66 34
ALDH1-positive cells in primary tumor before NAC  .0787
 Positive111 
 Negative6934 
ALDH1-positive cells in residual axillary metastases  <00001a
 Positive316 
 Negative7719 
Table 3. Characteristics of Patients With Aldehyde Dehydrogenase 1 (ALDH1)-Positive Cells (+) and ALDH1-Positive Cells (−) in Residual Axillary Metastases
 No. of Patients 
CharacteristicALDH1- Positive Cells (−), n = 96ALDH1- Positive Cells (+), n = 19P
  • Abbreviations: ER, estrogen receptor; Her2, human epidermal growth factor 2 receptor; LVSI, lymphovascular space invasion; NAC, neoadjuvant chemotherapy; pCR, pathologic complete response; PgR, progesterone receptor.

  • a

    P < .05.

Age, y  .0313a
 <50454 
 >505115 
Tumor classification  .1805
 T1120 
 T2-T36914 
 T4155 
Lymph node metastases  .7141
 Level I-II7215 
 Level III244 
No. of lymph node metastases after NAC  .0789
 0474 
 1-33110 
 ≥4185 
LVSI  .0032a
 Present2712 
 Absent697 
Histologic grade  .7575
 1-25712 
 3397 
ER status  .6243
 Positive6013 
 Negative366 
PgR  .6429
 Positive5011 
 Negative468 
Her2 status  .3307
 Positive314 
 Negative6515 
Ki-67 status  .7059
 Positive6011 
 Negative368 
pCR  .0646
 Yes150 
 No8119 
ALDH1 (+) cells in primary tumor before NAC  <.0001a
 Positive2817 
 Negative682 
ALDH1(+) cells in residual primary tumor after NAC  <.0001a
 Positive1916 
 Negative773 
Table 4. Univariate and Multivariate Analyses for Correlations Between Overall Survival and Clinical Characteristics
 Univariate AnalysisMultivariate Analysis
CharacteristicHRPHRP
  • Abbreviations: ALDH1, aldehyde dehydrogenase 1; ER, estrogen receptor; Her2, human epidermal growth factor 2 receptor; HR, hazard ratio; LVSI, lymphovascular space invasion; NAC, neoadjuvant chemotherapy; pCR, pathologic complete response; PgR, progesterone receptor.

  • a

    P < .05.

Age2.214.082  
Tumor classification    
  T2-T4 (vs T1)2.878.222  
  T4 (vs T1-T3)2.368.077  
Lymph node metastasis    
 Level III (vs level I-II)1.014.979  
 ≥4 Lymph node metastases after NAC (vs 0-3)1.485.494  
LVSI3.321.006a1.646.276
Histologic grade 3 (vs 1-2)1.302.559  
ER status2.246.067  
PgR status2.139.086  
Her2 status1.134.785  
Ki-67 status1.207.662  
pCR7.158.0006a2.195.211
ALDH1(+) cells in primary tumor before NAC8.878<.0001a2.049.221
ALDH1(+) cells in residual primary tumor10.044<.0001a1.924.349
ALDH1(+) cells in residual metastases11.575<.0001a3.876.006a

Even if ALDH1-Positive Cells Remain in the Primary Lesion, Patients With Disappearance of Metastatic ALDH1-Positive Cells After Chemotherapy Have a Good Prognosis

It is noteworthy that, in multivariate analyses, ALDH1-positive cells in residual axillary metastases after NAC were correlated significantly with OS, whereas ALDH1-positive cells in residual primary tumor did not (Table 4). To examine why the dynamics of ALDH1-positive cells are more important in metastases than in primary lesions, we evaluated the difference in drug responsiveness between ALDH1-positive cells from the primary lesion and metastatic ALDH1-positive cells after NAC. First, we confined the groups analyzed to 45 patients who were diagnosed preoperatively as ALDH1-positive by immunohistochemistry in CNB specimens. Of these 45 patients, there have been 19 deaths and 21 relapses; these 45 patients included 11 patients (24.4%) with ALDH1-positive cells (−) in primary lesions after NAC and 34 patients (75.6%) with ALDH1-positive cells (+). In the 11 patients who had ALDH1-positive cells (−) in the primary lesion after NAC, only 1 patient (9.1%) had ALDH1-positive cells in an axillary lesion, and the other 10 patients (90.9%) had ALDH1-positive cells (+) in axillary lesions. In comparison, in 34 patients with ALDH1-positive cells (+) in the primary lesion after NAC, 16 (43.1%) had ALDH1-positive cells (+) in axillary lesions, and the other 18 patients (56.9%) had ALDH1-positive cells (−) in axillary lesions. Next, to evaluate metastatic ALDH1-positive cells in axillary lesions before NAC, we attempted reimmunohistochemistry of ALDH1 in preoperative cytologic samples from axillary metastases using the cell-transfer technique. Of the 45 patients analyzed, we succeeded in cell transfer of 38 samples, but we could not transfer sufficient amounts of cells from 7 patients. Of the 38 patients for whom we could perform reimmunohistochemistry of ALDH1, 30 patients had ALDH1-positive cells confirmed in axillary lesions before NAC. Of these 30 patients, there have been 11 deaths and 12 relapses. Because we performed reimmunohistochemistry of ALDH1 only for patients who were diagnosed preoperatively as ALDH1-positive by immunohistochemistry in CNB specimens, our data did not address whether all patients with ALDH1-positive lymph nodes also had ALDH1-positive primary tumors before chemotherapy. Of these 30 patients who had both ALDH1-positive cells in the primary lesion and metastatic ALDH1-positive cells in axillary lesions confirmed before NAC, 5 patients had ALDH1-positive cells (−) in the primary lesion after NAC, and 25 patients had ALDH1-positive cells (+). In the 5 patients who had ALDH1-positive cells (−) in primary lesions after NAC, 4 patients also had ALDH1-positive cells (−) in axillary lesions after NAC. In comparison, among the 25 patients who had ALDH1-positive cells (+) in the primary lesion after NAC, 9 patients (36%) had ALDH1-positive cells (−) in axillary lesions after NAC, and 16 patients (64%) had ALDH1-positive cells (+). In these 25 patients with residual ALDH1-positive cells in the primary lesion, the OS of patients who experienced the disappearance of ALDH1-positive cells in axillary metastases was significantly longer than the OS of patients without such disappearance (P = .0015) (Fig. 4). These data suggest that, even if ALDH1-positive cells remain in the primary lesion, patients who experience the disappearance of ALDH1-positive cells in axillary metastases after chemotherapy can achieve good prognosis.

thumbnail image

Figure 4. Even if aldehyde dehydrogenase 1 (ALDH1)-positive cells remained in the primary lesion, patients in whom metastatic, ALDH1-positive cells disappeared after neoadjuvant chemotherapy (NAC) had a good prognosis. Of 115 patients, 25 patients displayed both ALDH1-positive cells in the primary lesion and metastatic, ALDH1-positive cells in axillary lesions before NAC, and they had residual ALDH1-positive cells in the primary lesion after NAC. Among these patients, the overall survival of those who had ALDH1-positive cells (−) in axillary lesions after NAC was significantly longer than the overall survival of those who had ALDH1-positive cells (+) (P = .0015).

Download figure to PowerPoint

DISCUSSION

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

To the best of our knowledge, this is the first report demonstrating the prognostic significance of residual ALDH1-positive cells in primary lesions, and particularly in metastatic lesions, after NAC in patients with lymph node-positive breast cancer. A subpopulation of cancer cells with tumorigenic ability has been revealed in solid tumors.6, 10-15 This subpopulation, which has stem-like characteristics in cancer tissue, is called cancer stem cells or tumor-initiating cells.6, 10-15 Traditionally, in the diagnosis and therapy of cancer, we treated whole cancer cells in the same manner. Because of the discovery of this subpopulation, we may revise our understanding of the handling of cancer cells in solid tumors. To apply this subpopulation in clinical settings, clarification of relations between this subpopulation and chemotherapy is important.

Recently, it was reported that ALDH1-positive cells had stem-like or progenitor ability in either normal breast epithelium cells or breast cancer cells.20 ALDH1-positive cells can be considered as a subpopulation with high proliferative potency in breast cancer tissue and are regarded as a new diagnostic and therapeutic target in breast cancer patients.20, 22 This study used pre-NAC and post-NAC samples from primary and metastatic lesions to evaluate responsiveness to chemotherapy of the ALDH1-positive subpopulation and the influence on prognosis in breast cancer patients.

In recent years, pathologic complete remission (pCR) of primary tumors and axillary lymph node metastases after NAC has been considered 1 of the goals of treatment for patients with breast cancer.24-29 By comparison, patients who do not achieve pCR have been treated prognostically in the same manner, and no suitable clinical goals have been identified for these patients. In the current study, we evaluated residual ALDH1-positive cells in addition to pathologic cancer cells after NAC in both primary and metastatic lesions. In primary lesions after NAC, separating patient status after NAC into 3 groups of “no residual tumor,” “residual tumor with ALDH1-positive cells (−),” and “residual tumor with ALDH1-positive cells (+),” only patients who had residual tumor with ALDH1-positive cells (+) after NAC had a worse prognosis compared with the other 2 groups. Furthermore, using the same evaluation for metastatic lesions, only patients who had ALDH1-positive cells (+) in residual axillary metastases after NAC had a worse prognosis compared with the other 2 groups. These data indicate that only residual ALDH1-positive cells after NAC represent a prognostic factor in patients with lymph node-positive breast cancer. Our data were derived from detailed analyses of primary and metastatic lesions before and after NAC in a single institution. One of the limitations in this study was the small sample size. Patients with ALDH1-positive cells (+) in primary and metastatic lesions after NAC clearly had a worse prognosis. In contrast, according to our data, no prognostic difference was apparent between patients who had no residual tumor and those who had residual tumor with ALDH1-positive cells (−). This suggests the interesting notion that, if pathologic tumor and metastases remain after NAC, then patients with ALDH1-positive cells (−) are prognostically similar to patients who attain pCR. However, this result may have been attributable to our small sample size, and evaluation in a larger sample is required. The rate of ALDH1-positve cells in the primary tumor before NAC (39.1%) in this study was higher than that in a previous report (19%).20 In our current study, we evaluated patients with advanced breast cancer who had cytologically confirmed axillary lymph node metastases. The reason for the high rate of ALDH1-positve cells in primary tumor may be because of differences in patient selection for this study. In recent therapy for breast cancer, trastuzumab for Her2-positive patients and antihormone therapy for ER-positive and PR-positive patients are clinically important.30-32 Combined therapy with chemotherapy and trastuzumab for patients with Her2-positive breast cancer reportedly raises the rate of pCR after NAC.33 No patients in our study received trastuzumab; thus, we could not evaluate the influence of trastuzumab on ALDH1-positive cells. In addition, the length of observation in this study was insufficient to evaluate the efficiency of hormone therapy. Therefore, our data mainly reflect the influence of chemotherapy, and the therapeutic efficacy of trastuzumab and hormone therapy on ALDH1-positive cells remains to be clarified.

This study also demonstrated that only ALDH1-positive cells in residual axillary metastases were correlated with OS in multivariate analysis. Residual ALDH1-positive cells in metastatic lesions had superior prognostic significance compared with ALDH1-positive cells in primary lesions. Furthermore, our analyses in selected patients who had ALDH1-positive cells in both primary and metastatic lesions before NAC produced 3 interesting results. First, in some patients, ALDH1-positive cells disappeared after NAC in primary or axillary metastatic lesions. Second, the rate of disappearance for ALDH1-positive cells after NAC was higher in axillary metastatic lesions than in primary lesions. Third, in patients who had residual ALDH1-positive cells after NAC in primary lesions, the OS for patients who experienced the disappearance of ALDH1-positive cells from axillary metastatic lesions after NAC was longer than the OS for patients who had residual ALDH1-positive cells. The finding that ALDH1-positive cells, particularly in metastatic lesions, had a complete response to chemotherapy is clinically important. In patients with ALDH1-positive cells, the pCR rate was low, as noted by Tanei et al.21 However, in these patients who did not achieve pCR, those who experienced the disappearance of ALDH1-positive cells after NAC had a good prognosis. These findings suggest that ALDH1-positive cells in axillary lesions offer a surrogate marker for patients with lymph node-positive breast cancer.

In addition, ALDH1-positive cells in residual metastases, but not residual primary tumor, were correlated with OS in multivariate analysis. However, the hazard ratio of ALDH1-positive cells in residual metastases (11.575) and of ALDH1-positive cells in residual primary tumor (10.044) was almost the same in univariate analysis. The presence of ALDH1-positive cells in residual primary tumor also was an important prognostic factor in patients with lymph node-positive breast cancer. These data may be results from a small sample; and, as mentioned above, evaluation and confirmation in a larger sample are needed. Moreover, in general, the drug responsiveness of lymph node metastases and distant organ metastases are not similar in breast cancer patients.34, 35 We should not equate lymph node metastases with distant organ metastases as targets of chemotherapy. However, given the substantially different environment from the primary lesion, the finding that the prognostic significance of ALDH1-positive cells in axillary lesions was superior to that in primary lesions is clinically important.

In conclusion, the responsiveness of ALDH1-positive cells to chemotherapy in primary and metastatic lesions and its associated prognostic significance were revealed in patients with breast cancer. ALDH1-positive cells may represent a surrogate marker as a new concept in patients with lymph node-positive breast cancer. In the future, ALDH1-positive cells may be used for diagnosis and treatment, and these cell populations could be targeted in the development of new chemotherapies. We hope that our results will provide basic data for therapies targeting cell subpopulations in tumors, rather than the whole body, in patients with noncurable, metastatic breast cancer.

FUNDING SOURCES

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

No specific funding was disclosed.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

REFERENCES

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