Abstract
Purpose: To evaluate HER2 status in residual tumor identified at the time of surgery in patients not achieving a pathologic complete response (pCR) and to determine the effect of alterations in HER2 status on recurrence-free survival (RFS).
Experimental Design: Clinicopathologic data for patients with HER2-overexpressing breast cancer receiving neoadjuvant therapy with a taxane, anthracycline, and concomitant trastuzumab between 2004 and 2007 were reviewed. Surgical specimens for patients achieving less than a pCR were assessed to determine if there was enough residual tissue to evaluate posttreatment HER2 status. RFS was determined using the Kaplan-Meier method and compared by the log-rank statistic.
Results: A pCR was achieved in 72 of the 142 (50.7%) patients. Residual tumor was sufficient to assess posttreatment HER2 status in 25 patients. Fluorescence in situ hybridization done on pretreatment specimens confirmed HER2 amplification before beginning therapy. Eight (32.0%) posttreatment tumors were found to be HER2-negative by fluorescence in situ hybridization. At a median follow-up of 37 months (range, 8-56 months), the RFS was significantly better for patients with tumors that retained HER2 amplification (87.5% versus 50%, P = 0.04).
Conclusion: High pCR rates are achieved in patients with HER2-positive breast cancer treated with neoadjuvant trastuzumab in combination with anthracyclines and taxanes. One third of patients with significant residual disease loses HER2 amplification, and this change is associated with poor RFS. Residual tumor identified at the time of surgery should be reassessed for HER2 status, and novel adjuvant therapy strategies need to be studied in this population. (Clin Cancer Res 2009;15(23):7381–8)
- breast cancer
- HER2/neu
- neoadjuvant chemotherapy
- trastuzumab
Translational Relevance
This study confirmed that patients with HER2-overexpressing breast cancer treated in the neoadjuvant setting with trastuzumab-based systemic therapy achieve a high rate (∼50%) of pathologic complete response. Importantly, in patients not achieving a pathologic complete response who had significant residual disease, fluorescence in situ hybridization showed that the tumors from one third of these patients no longer had amplification of the HER2 gene. Those patients with tumors that were no longer HER2 amplified had a significantly worse recurrence-free survival than those with tumors that retained HER2 amplification. Taken together, these data suggest that residual tumor identified at the time of surgery in patients receiving trastuzumab-based neoadjuvant therapy should be reassessed for HER2 status and that novel adjuvant therapy strategies need to be studied in this population.
The HER2/neu (HER2) gene is amplified in ∼25% of breast cancers (1). Gene amplification results in overexpression of the HER2 protein, which is associated with an aggressive clinical course to include a shorter disease-free interval after adjuvant therapy and decreased overall survival (OS; refs. 2–4). The natural history of HER2-overexpressing breast cancer has been altered, however, by the routine use of trastuzumab, a monoclonal antibody targeting the extracellular domain of the HER2 protein. Trastuzumab has been shown to improve survival in patients with metastatic HER2-positive breast cancer (5, 6) as well as in patients with earlier stage disease. Several large, multicenter adjuvant therapy trials showed that the addition of trastuzumab to systemic chemotherapy reduces recurrence by ∼50% and improves OS by 30% (7, 8). Trastuzumab has also been shown to be efficacious when given in the neoadjuvant setting with pathologic complete response (pCR) rates ranging from 7% to as high as 65% in patients with both early and locally advanced breast cancer (9–14). Despite these successes with trastuzumab therapy, not all HER2-positive tumors respond and some patients whose tumors do respond will experience disease recurrence. Investigators from our group recently reported a case of a patient with HER2-positive breast cancer who received adjuvant trastuzumab but relapsed with HER2-negative metastatic disease (15). In a study conducted to evaluate changes in HER2 status in metastatic lesions of patients previously treated with trastuzumab, Pectasides et al. showed that 37% of patients no longer had HER2 expression/amplification, and these patients had significantly shorter time to tumor progression than the group who remained HER2-positive (16).
The purpose of the current study was to evaluate HER2 gene amplification status using fluorescence in situ hybridization (FISH) in the residual tumors of patients who received neoadjuvant systemic therapy with paclitaxel and FEC (5-fluorouracil, epirubicin, and cyclophosphamide) with concomitant weekly trastuzumab. We also sought to determine the effect of changes in HER2 status on recurrence-free survival (RFS).
Materials and Methods
Cell lines and treatments
The BT-474 cell line was purchased from the American Type Culture Collection. Cells were maintained in DMEM/Ham F12 1:1 (DMEM/F12) supplemented with 10% fetal bovine serum and 2 mmol/L l-glutamine (Life Technologies, Inc. Ltd.) at 37°C in 5% CO2. Trastuzumab (Herceptin; kindly provided by F. Hoffmann, La Roche) was dissolved in sterile apyrogen water and stored at 4°C. Trastuzumab-resistant BT-474 (BT-474R) cells were obtained by culturing the parental BT-474 cells in the presence of increasing concentrations of trastuzumab (up to 500 nmol/L) for >18 mo. Genetic analysis was done using single nucleotide polymorphism (SNP) arrays on the clones and parental cell lines. Protein extraction, Western blot, and immunohistochemistry were done as previously described (17).
Patient selection
The Department of Breast Medical Oncology database was queried to identify patients with histologically confirmed, HER2-overexpressing (defined as immunohistochemical 3+) or amplified (FISH positive), nonmetastatic, invasive breast cancer who received the neoadjuvant systemic chemotherapy-based regimen with concomitant trastuzumab described below. Patient and tumor characteristics including age at diagnosis, presenting clinical stage, histology, nuclear grade, estrogen receptor (ER) and progesterone receptor (PR) status, presence or absence of lymphovascular invasion, type of surgery, and pathologic response in the breast and axilla were recorded. Follow-up data was updated through January 2009. The University of Texas M.D. Anderson Cancer Center Institutional Review Board approved this study.
Pathology
The breast cancer diagnosis was confirmed by a review of core biopsy material by dedicated breast pathologists. The histologic subtype of all tumors was defined according to the WHO classification system (18), and the modified Black's nuclear grading system was used (19). Immunohistochemical analysis was done to determine ER and PR status. Nuclear staining of ≥10% was considered positive. HER2 status was evaluated by immunohistochemistry and further confirmed by FISH in tissue obtained before initiation of neoadjuvant chemotherapy. Interpretations of these assays were based on the most recent American Society of Clinical Oncology/College of American Pathologists guidelines (20).
FISH analysis of breast carcinoma was done using the PathVysion HER-2 DNA probe kit (Vysis, Inc.). Briefly, this assay uses two directly labeled fluorescent DNA probes that specifically target the HER2 locus and CEP17, the α-satellite DNA sequence at the centromeric region of the chromosome. For the pretreatment biopsy specimens, all areas of invasive tumor were screened under a fluorescent microscope to evaluate the possibility of heterogeneity among tumor cells. No heterogeneity was identified. Sixty tumor cells (versus 20 cells as per the manufacturer's recommendation) in each case were then scored for HER2 and CEP17 signals. Among the posttreatment specimens, we scored all tumors cells identified up to 60 when present. For cases with reduced residual tumor cell density due to treatment response, we scored a minimum of 20 tumor cells for HER2 and CEP 17 signals. A FISH ratio (HER2 gene signals to chromosome 17 signals) was determined and if >2.2 was considered positive.
A pCR was defined as no residual invasive disease in the breast and axilla on final pathologic assessment. For patients achieving less than a pCR who had enough residual tumor tissue, a dedicated breast pathologist (Y.W.) reassessed HER2 status in the pretreatment biopsy specimen and in the posttreatment residual tumor using FISH (described above) to determine if HER2 gene amplification was present.
Treatment
Paclitaxel was given weekly for 12 wk at a dose of 80 mg/m2/wk i.v. This was followed by four cycles of FEC75 (fluorouracil 500 mg/m2 epirubicin 75 mg/m2, cyclophosphamide 500 mg/m2 i.v., given the first day of each cycle) given every 3 wk. Trastuzumab was given as a loading dose of 4 mg/kg i.v. on the first day and then subsequently given weekly at a dose of 2 mg/kg concomitantly with both the anthracycline and taxane chemotherapy. After completion of neoadjuvant systemic therapy, patients underwent appropriate surgery with either a segmental or total mastectomy. The axillary lymph nodes were assessed with sentinel lymph node biopsy for patients who presented initially with node-negative disease and with axillary lymph node dissection for patients who were documented to have axillary lymph node metastasis before beginning neoadjuvant systemic therapy. Surgery was followed by radiation therapy when indicated and appropriate endocrine therapy for patients with hormone receptor–positive disease. Trastuzumab was continued to complete 1 y of therapy.
Statistical analysis
Patient characteristics were tabulated or described by their median and range overall, by pCR group, and by postneoadjuvant chemotherapy HER2 status group. The χ2 test or Wilcoxon rank sum test was used as appropriate to determine associations between patient characteristics. Median follow-up time was calculated as the median observation time among all patients. Recurrence was defined as recurrence of disease in either local, regional, or distant sites. RFS was defined as the time from diagnosis to the time of first recurrence or last follow-up. Survival distributions were estimated with the Kaplan-Meier method, and the log-rank statistic was used to compare the differences between groups.
Results
Between June 2003 and May 2007, 142 HER2-positive patients were treated with the concomitant trastuzumab and neoadjuvant systemic therapy regimen. Table 1 lists patient characteristics overall and by whether they experienced a pCR. Seventy-two (50.7%) patients achieved a pCR. From the 70 patients with residual disease, 61 (43.0%) had a partial response to neoadjuvant chemotherapy, 6 (4.2%) had stable disease, and 3 (2.1%) had progression of disease. Compared with patients with residual disease, patients who had a pCR were more likely to have ductal histology (versus lobular or mixed ductal/lobular; P < 0.0001), absence of lymphovascular invasion (P = 0.005), and hormone receptor–negative tumors (P = 0.045 for ER; P = 0.046 for PR).
Patient characteristics overall and by pCR
The majority of patients who did not achieve a pCR had a near complete response with only minimal residual disease, such as scattered tumor cells in the primary tumor site or lymph node or minimal cellularity in the surgical specimens. In these patients, HER2 status could not be reassessed. However, in 25 patients achieving less than a pCR, enough residual tissue was available at the time of surgery to reassess HER2 status by FISH. Eight (32.0%) of these patients had tumors that lost HER2 amplification. To confirm that these patients had HER2 gene amplified tumors before receiving the concomitant trastuzumab and neoadjuvant chemotherapy regimen, FISH was repeated on their pretreatment biopsy specimens and homogeneous HER2 amplification was confirmed in all cases (Table 2; Fig. 1). Twenty patients had enough residual disease to reassess ER status to compare with pretreatment ER status. Four (20%) patients had tumors that converted from ER-negative to ER-positive disease. When comparing patients with tumors that lost HER2 gene amplification (n = 8) with those with tumors that remained HER2-amplified (n = 17), there were no significant differences in clinicopathologic features associated with conversion of HER2 status (Table 3).
HER2 gene amplification and hormone receptor status following trastuzumab containing neoadjuvant chemotherapy in patients with enough residual disease identified at the time of surgery to reassess HER2 status
FISH was done to assess HER2 status. A, FISH done on biopsy specimen before treatment with a trastuzumab containing neoadjuvant chemotherapy regimen. Red, HER2 gene; green, CEP17 (centromere of chromosome 17). HER2/CEP17 = 6.22. Due to the intensity of HER2 staining, merged images were not obtained. B, FISH done on residual disease identified at the time of surgery from the same patient after completion of neoadjuvant chemotherapy. Image shown is a merged image of staining for HER2 and CEP17. HER2/CEP17 = 1.1.
Patient characteristics by HER2 status following trastuzumab containing primary chemotherapy
The median follow-up for the entire population was 33.5 months (range, 8-65 months). Patients achieving a pCR had significantly better RFS compared with patients who did not achieve a pCR (P = 0.0175; Fig. 2A). The 3- and 5-year RFS estimate for all patients and the 3-year RFS estimate for those who achieved a pCR versus those who did not achieve a pCR are listed in Table 4. The median follow-up for the patients who achieved less than a pCR and had enough residual tumor tissue to reassess HER2 status was 37 months (range, 8-56 months). Analysis of these patients showed that patients who retained HER2 gene amplification had significantly better RFS compared with patients whose tumors lost HER2 gene amplification (P = 0.041; Fig. 2B). The 3-year RFS estimates for patients whose tumors retained HER2 amplification was 87.5% [95% confidence interval (95% CI), 72.7-100%] versus 50.0% (95% CI, 25.0-100%) for those that did not (Table 4).
Kaplan-Meier plots of RFS by (A) pCR and (B) status of HER2 gene amplification in patients with residual tissue identified at the time of surgery.
Kaplan-Meier estimates of RFS among all patients by pCR and by HER2 status in patients with residual tissue identified at the time of surgery
There have been eight deaths in the entire cohort, two in the group of patients who achieved a pCR versus six in the group achieving less than a pCR (P = 0.137). In the group of 25 patients that had enough residual disease to reassess HER2 status, there has been one death, which occurred in a patient whose tumor had lost HER2 amplification.
To investigate the hypothesis that trastuzumab treatment could play a causative role in selecting HER2-negative (without gene amplification) cells within a population of HER2-positive (with gene amplification) cells, we cultivated HER2 positive BT-474 breast cancer cells in the presence of increasing concentrations of trastuzumab for >18 months isolating several independent subclones. After this period of time, we found that two independent clones treated continuously with trastuzumab (BT-474R) had lost both HER2 overexpression and HER2 gene amplification (Fig. 3) and had acquired resistance to the antiproliferative activity of trastuzumab in vitro (data not shown).
Loss of HER2 overexpression and amplification in BT-474R cells. A, Western blot showing loss of HER2 overexpression in a representative clone of BT-474R cells. Ponceau staining serves as the loading control. B, loss of HER2 overexpression by immunohistochemistry and loss of HER2 gene amplification by FISH (red, HER2 gene; green, CEP17) of a representative clone of BT-474R cells.
Discussion
Patients with HER2-overexpressing breast cancer treated with trastuzumab-based neoadjuvant systemic therapy achieve a high rate of pCR. In the current study, the pCR rate was 51% after treatment with a neoadjuvant regimen that included taxane and anthracycline-based chemotherapy used concurrently with weekly trastuzumab for 24 weeks. The majority of patients not achieving a pCR had very minimal residual disease (near complete response), with only a third of patients having enough tumor tissue identified at the time of surgery to reassess HER2 status. Importantly, one third of patients who had enough residual disease to repeat HER2 testing had lost amplification of the HER2 gene. Patients who had enough residual disease to reassess HER2 status and had lost HER2 gene amplification had a significantly decreased RFS compared with patients whose tumors remained HER2 amplified.
Other investigators have evaluated HER2 expression in paired samples of pretreatment and posttreatment tissue from patients treated with trastuzumab in the neoadjuvant setting. Burstein et al. reported on HER2 status in patients with residual tumor after treatment with 12 weeks of paclitaxel and trastuzumab (9). Their trial enrolled 40 patients, 23 of whom had residual tissue available for HER2 testing by immunohistochemistry. In six (26.1%) cases, all of whom were immunohistochemical 3+ before treatment, the HER2 status changed to 2+ in two patients and 0 in four patients. In a phase II study of 48 patients treated with 12 weeks of neoadjuvant trastuzumab and vinorelbine, Harris et al. reported a HER2 conversion rate of 12% in 18 patients with enough residual tissue to repeat HER2 testing by immunohistochemistry (12). Although the concordance between HER2 overexpression detected by immunohistochemistry and HER2 gene amplification by FISH has been shown to be statistically significant (21–23), there are issues regarding consistency in immunohistochemistry testing that may affect results, including variable fixation, antigen retrieval methods, and observer analysis (24). In addition, FISH has been shown to be more reproducible than immunohistochemistry between central and peripheral laboratories (22, 25). Because we used FISH to determine HER2 gene amplification status pretreatment and posttreatment, we are confident that the changes in HER2 status are not due to artifact or inconsistent testing. Consistent with our findings, Hurley et al. showed that 43% of tumors that had HER2 gene amplification by FISH before treatment with neoadjuvant trastuzumab, docetaxel, and cisplatin became FISH-negative after therapy (13).
It is unclear whether this change reflects response to therapy or a mechanism of resistance. It is possible that a change in HER2 status could reflect the heterogeneity of HER2 expression within the tumor, suggesting that trastuzumab eliminated HER2-overexpressing clones leaving only HER2-negative tumor cells upon completion of therapy. The results obtained with our preclinical model based on BT-474 cells that acquired resistance to trastuzumab support this possibility. It seems likely that the change in HER2 status reflects treatment of HER2-overexpressing clones, and one could speculate that the trastuzumab therapy was effective in treating the HER2-amplified cells in over 65% of tumors, the 50% that achieved a pCR, and the 15% that became HER2 negative.
Another interesting finding from our analysis is that four patients whose tumors were ER-negative pretreatment were found to be ER positive when residual tumor tissue was examined. Previous reports have described cross talk between the ER and the HER2 pathways, and studies have suggested an association between HER2 signaling and resistance to antiestrogens in human breast cancer (26–28). Whereas we acknowledge that the current study reports a small number of patients, the findings suggest that, in some patients with HER2-overexpressing, ER-negative breast cancer treatment with trastuzumab may facilitate sensitivity to antiestrogen therapy by upregulating ER expression. This finding requires further confirmation in a larger cohort of patients, but given the potential therapeutic implications, we recommend that residual tumor tissue identified in patients treated with concurrent trastuzumab and neoadjuvant chemotherapy be reassessed for HER2 and ER status.
Studies incorporating trastuzumab into neoadjuvant chemotherapy regimens have reported pCR rates ranging from 17% to 65% (9–14). One explanation for the high pCR rates using such regimens is the use of two potentially non–cross-resistant chemotherapy agents given sequentially in combination with trastuzumab. This concept is supported by data from the NOAH (NeOAdjuvant Herceptin) trial, which randomized women with HER2-overexpressing locally advanced breast cancer or inflammatory breast cancer to receive doxorubicin, paclitaxel, and cyclophosphamide, methotrexate, and fluorouracil (CMF)-based neoadjuvant systemic therapy with or without concomitant trastuzumab. This trial enrolled 327 women, and the pCR rates were significantly higher in trastuzumab-treated patients (39% versus 20%, P = 0.002; ref. 14). This lower pCR rate compared with our patient cohort may be due to differences in presenting disease stage. An earlier study also focusing on locally advanced and inflammatory HER2-positive disease given 12 weeks of docetaxel, cisplatin, and trastuzumab in 48 patients and reported a pCR rate of 23% in the breast and 17% in the breast and axilla (13). It is difficult to compare pCR rates between trials due to differences in presenting clinical stage, regimens used, and duration of therapy, as well as differing definitions of pCR. However, because the NOAH trial and the trial reported by Hurley et al. enrolled similar patient populations, the differences in the pCR rates are interesting and suggest that the duration of therapy and the use of an anthracycline may be important determining factors for pCR. Currently, the American College of Surgeons Oncology Group is leading a large, multicenter trial (ACOSOG Z1041) comparing a neoadjuvant regimen of FEC75 followed by paclitaxel plus trastuzumab with a neoadjuvant regimen of paclitaxel plus trastuzumab followed by FEC75 plus trastuzumab in patients with HER2-overexpressing breast cancer. Results from this trial should provide conclusive data regarding the utility of administering trastuzumab concurrently with an anthracycline in the neoadjuvant setting.
Achieving a pCR is an important end point for patients receiving neoadjuvant systemic therapy, as it has been shown to correlate with long-term outcomes (13, 29). In the current study, we again show that achieving a pCR is associated with improved RFS. There was a trend toward improvement in OS, although this did not reach statistical significance, which we attribute to the relative short median follow-up time of 33.5 months. Importantly, a novel finding in the current study is the effect on RFS of loss of HER2 gene amplification in patients with measurable residual disease after administration of trastuzumab. Patients whose tumors lost HER2 gene amplification as determined by FISH analysis had a significantly worse RFS than those whose tumors remained HER2 amplified.
In conclusion, we observed that approximately one third of patients with measurable residual disease after administration of a neoadjuvant systemic therapy regimen that included taxane/anthracycline-based chemotherapy used concurrently with weekly trastuzumab for 24 weeks lost HER2 gene amplification. Our data show that this change affects RFS. Patients who had measurable residual disease and converted to HER2-negative disease had a significantly shorter RFS than patients who had measurable residual tumor but retained HER2 gene amplification. This finding could have implications regarding additional adjuvant therapy. Currently, our practice is to administer trastuzumab postoperatively to complete 1 year of therapy based on data from the multicenter adjuvant trials (7, 8). If conversion of HER2 status reflects response to therapy, such that only HER2-negative clones remain, the need to complete 1 year of trastuzumab in the adjuvant setting comes into question. Furthermore, all patients with early-stage HER2-positive disease who relapse after adjuvant or neoadjuvant trastuzumab therapy should have biopsies of their recurrent disease and reassessment of their marker status, as we have shown that a change in marker status correlates with outcome in patients who develop metastatic disease (30). These data suggest that there may be utility in assessing HER2 status in residual disease identified at the time of surgery and that future clinical trials should be designed to investigate the most appropriate strategy for adjuvant therapy in these patients.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Footnotes
Grant support: American Society of Clinical Oncology Career Development Award and NCI 1K23CA121994-01 (A.M. Gonzalez-Angulo), NCI 1K99CA133244-01 (E.A. Mittendorf), and the Nellie B. Connally Breast Cancer Research Fund.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Received July 8, 2009.
- Revision received September 4, 2009.
- Accepted September 9, 2009.
References
Volume 15, Issue 23
