Although approximately 70% of breast cancer patients demonstrate a clinical response on neoadjuvant systemic therapy on physical examination or on anatomic radiographic imaging, only 3%–40% achieve a pathologic complete response (pCR). Magnetic resonance imaging (MRI) is superior to physical examination, ultrasound, and mammography in response evaluation during neoadjuvant systemic therapy. The accuracy of breast MRI to predict pCR has a moderate sensitivity, but high specificity. The accuracy of anatomic imaging to assess residual disease and predict pCR depended on anatomic radiographic imaging cancer subtypes. Response monitoring using breast is accurate in triple-negative or HER2-positive tumors. It was inaccurate in estrogen receptor-positive/HER2-negative subtype. Another approach currently under investigation is dynamic contrast-enhanced MRI and diffusion weighted-imaging, (18)F-fluorodeoxyglucose positron emission tomography, fluorodeoxyglucose positron emission tomography/computed tomography.
In 1981, the World Health Organization (WHO) published tumor-response criteria for use in trials where tumor response to chemotherapy was the primary endpoint (1). In response to multiple modifications of WHO criteria, an International Working Party was formed in the mid-1990s to standardize and simplify response criteria. New criteria, known as Response Evaluation Criteria in Solid Tumors (RECIST), were published in 2000 (2).
The specific response categorization (50% for the WHO bidimensional metric, corresponding to a partial response of 30% by RECIST unidimensional measurement) has remained basically unchanged in the evolution of the response assessment. The imaging modalities used to assess tumor size, however, have evolved substantially, which have led to the development of a revised RECIST guideline (version 1.1) (3).
The primary purpose of an imaging endpoint is to serve as an early but accurate indicator of a promising treatment effect. The measure of treatment effect on the phase II endpoint must correlate sufficiently well with the measure of treatment effect on the phase III primary endpoint, which is usually assessed by a difference between two arms on progression-free survival or overall survival.
Both clinical objective response and time to development of disease progression (progression-free survival) are important endpoint in phase II clinical trials in advanced breast cancer. However, many clinical trials (4–9) established that pathologic complete response (pCR) is of greater importance for patient outcome than clinical response. Despite the fact that no standardized definition for pCR exists, many trials (7,9–12) selected pCR as a primary endpoint.
Several imaging techniques are available to predict the effect of preoperative therapy among patients with breast cancer. In many recent trials, physical examination, mammography, and sonographv are still the primary diagnostic tools for the evaluation of therapy (13). According to revised RECIST guideline (version 1.1), ultrasound examinations should not be used in clinical trials to measure tumor regression or progression of lesions because the examination is subjective and operator dependent (3). The RECIST Working Group and same expert (3,14) did not believe that there is at present sufficient standardization and widespread availability to recommend adoption of functional imaging (18F-fluorodeoxyglucose positron emission tomography, dynamic contrast-enhanced magnetic resonance imaging [MRI]) as alternative assessment methods.
However, some studies have shown that MRI can determine response to chemotherapy treatment more accurately than mammography, ultrasound, and clinical examination, and now extensive research has been done on the performance of MRI in evaluating the efficacy of preoperative therapy in patients with breast cancer (11,13,15). Yuan et al. (15) performed a meta-analysis of value MRI in prediction of pathologic complete remission after preoperative therapy. This meta-analysis, which included data from 1212 patients, has shown that contrast-enhanced MRI has high specificity (90.7%) and relative lower sensitivity (63.1%) in predicting pathologic complete remission after preoperative therapy in patients with breast cancer [IIIС]. On the other side, a meta-analysis of preparation MRI for staging the primary breast cancer showed that MRI has high sensitivity (90%) and lower specificity (72%) (16) [IIIС].
In other studies (17–21), it was observed that the accuracy of breast imaging techniques to assess clinical response and predict pCR depended on breast cancer subtypes. It was observed that MRI was able to predict pCR more accurately in patients with HER2-positive tumors when compared with HER2-negative tumors or in patients with triple-negative disease (13,17,20). For example, NeOAdjuvant Herceptin Trial (NOAH)study (22) of locally advanced HER2-positive disease comparing pCR rates with clinical response to chemotherapy plus trasturumab reveals a close correlation between clinical and pathological findings [IIB], so that patients who achieved a clinical CR were more likely to have achieved a pCR than patients who only achieved a clinical partial response (67% vs 19%). It was also observed that the average size discrepancy between MRI and pathology was greater in tumors demonstrating limited Ki-67 staining and high level of estrogen receptor (ER) positivity when compared with tumors demonstrating increased Ki-67 staining and ER-negative status.
In ECTO II trial (12), achievement of pCR following three different anthracycline- and taxane-containing neoadjuvant chemotherapy regimens was more common in ER-negative than ER-positive women (45.3% vs 10.4%). Only in group of patients with ER-negative tumors, high rates of clinical response (cCR) translated into a high rates of pCR (pCR/cCR = 82.6%). These and other findings (17,21–24) suggest that breast cancer intrinsic subtype and the type of neoadjuvant systemic therapy influence on achieving cCR and pCR. These data also suggest that it is inappropriate to use pCR as the primary endpoint in neoadjuvant trials testing cytotoxic chemotherapy in patients with low grade (1 or 2) ER-positive tumors (24).
The goals of neoadjuvant therapy of breast cancer have clinically changed and evolved because it was firstly applied to women with inoperable locally advanced and inflammatory breast cancers in the early 1970s. The extended indication to allow more breast-conserving surgery has widened the application of neoadjuvant treatments and provided evidence for the association between favorable long-term outcome and intermediate endpoints, like pCR after chemotherapy or decreased tumor proliferation measured by Ki-67 after endocrine therapy (25).
A key question is whether pCR and Ki-67 can take the role of qualified and validated surrogate markers of drug efficacy. The recent improvements in understanding the molecular basis of breast cancer heterogeneity has provided conceptual framework for interpreting the role of pCR as potential surrogate marker and that different molecular subtypes will require different intermediate surrogate endpoints (8).
The validation of the intermediate surrogate markers of efficacy would dramatically change the landscape of development of new drugs for a comparative analysis of the intermediate endpoint instead of the final survival endpoint.
- polymerase chain reaction
- positron-emission tomography
- magnetic resonance imaging
- diagnostic radiologic examination
- computed tomography
- physical examination
- neoadjuvant therapy
- diagnostic imaging
- breast cancer
- systemic therapy
- residual tumor
- fluorodeoxyglucose positron emission tomography
- magnetic resonance imaging of breast
- estrogen receptor positive
- complete remission
- her2 positive
- her2 negative
- contrast-enhanced magnetic resonance imaging
- response evaluation criteria in solid tumors