Biomarkers

Introduction

Breast cancer is a heterogeneous disease process. The World Health Organization recognizes many morphologic subtypes. By gene expression, invasive breast cancer is classified into at least four major subtypes: luminal A, luminal B, human epidermal growth factor receptor 2 (HER2)-positive, and triple-negative (ER/PR- and HER2-negative). The biology and the prognosis of these subtypes are vastly different. Genetic profiles of these cancer subtypes are also interlinked with the traditional breast biomarkers.

According to the US National Institutes of Health’s (NIH) Working Group and Biomarkers Consortium, a biomarker objectively measures normal biologic processes, pathogenic processes, or a pharmacologic response to a therapeutic intervention. For more than 40 years, estrogen receptor (ER) has been known to play a crucial central role in the development and progression of breast cancer. Researchers and clinicians have used treatments that target ER in order to block the progression and recurrence of breast cancer. With the passage of time, additional biomarkers were discovered. Today ER, progesterone receptor (PR), and HER2 are important biomarkers. They can predict not only the response to treatment but also the prognosis and disease recurrence. These biomarkers can be evaluated in breast core biopsies or excision specimens. Determination of the biomarker status on core biopsy is especially important when neoadjuvant therapy is considered. Evaluation and reporting of these biomarkers are now almost completely standardized throughout the United States. The newest (8th) edition of the American Joint Committee on Cancer (AJCC) staging system has added a clinical Prognostic Staging Group that incorporates the results of these ancillary tests to the traditional (TNM) system, tumor size, lymph node status, and metastasis.

Other commonly used but less defined biomarkers are Ki67 (a proliferation marker), programmed death ligand 1 (PD-L1), basal markers such as CK5/6, epithelial growth factor receptor (EGFR), androgen receptor (AR), and p53. PD-L1 has gained significant importance in the recent years, primarily in triple-negative (ER/PR/HER2-negative) breast cancers. Currently, there is no universal protocol for reporting most of these markers. The significance of p53 expression and its impact on treatment is even less defined; therefore p53 will not be discussed in this chapter.

Epithelial cadherin (E-cadherin) and p120 confirm the diagnosis of lobular differentiation. These immunostains are not traditional breast biomarkers; however, invasive lobular carcinoma (ILC) has a different prognosis independent of the stage. Therefore we are considering them as breast biomarkers.

Estrogen Receptor: Historical Perspective

ER status has long been recognized as an important predictive and prognostic biomarker in breast cancer. Approximately 70% to 75% of breast cancers are ER-positive. A majority of these ER-positive tumors respond to adjuvant hormonal therapy. This treatment is beneficial in reducing disease recurrence and improving 15-year mortality from breast cancer in women of all ages. ER assessment by immunohistochemistry (IHC) is now the standard in the management of breast cancer.

Like other steroid receptors, ER belongs to the nuclear receptor superfamily. It consists of 553 amino acids forming an N-terminal domain with transcription activation functions, a central DNA-binding domain, and a ligand-binding domain at the carboxy terminal. Two isoforms of ER exist: ERα and ERβ. They are encoded by two different genes ( ESR1 on chromosome 6 and ESR2 on chromosome 16, respectively) with no homology in their amino acid sequences. While the specific roles of ERβ are still being elucidated, ERα is the most studied and clinically measured isoform. The ligand for ER is the female sex hormone 17β-estradiol (E2) that in its physiologic state mediates growth and differentiation of the breast ducts. The estrogen signaling pathway is activated when the hormone diffuses through the cell membrane and binds to nuclear ER, inducing its dimerization. The ligand-receptor complex then promotes downstream gene transcription.

The association between estrogen hormone and breast cancer pathogenesis has been known since the late 1800s. Tumor regression with oophorectomy, adrenalectomy, or hypophysectomy has been documented for metastatic and inoperable breast tumors. Jensen and colleagues first reported on the characterization and measurements of ER in the mid-1960s. A few years later, McGuire and colleagues observed that there was variability in the concentration of ER in primary and metastatic breast cancers and emphasized that an assay for ER must be quantitative. Their studies provided the early insights into this hormone receptor’s utility as a biomarker for breast cancer.

Before the era of IHC, ligand-binding assays were used primarily for receptor quantitation. These biochemical procedures were often cumbersome and involved extraction of receptor proteins by homogenization of fresh-frozen tumor tissue and incubation of the homogenates with radioactive and nonradioactive E2, followed by separation of the bound and unbound hormone. Scatchard plots and standard curves were subsequently used to quantify ER, which was expressed in femtomoles of ER protein per milligram of cytosol protein. Some of the earlier receptor analytic methods included Sephadex gel filtration, protamine sulfate precipitation, sucrose density gradient ultracentrifugation, and dextran-coated charcoal assay. The latter two methods were commonly used, with dextran-coated charcoal assay being the preferred because of its ease of use and accuracy. Although newer methodologies such as enzyme immunoassays and IHC have made ligand-binding assays obsolete, they were the first assays that broadened our understanding of ER and the response to hormonal therapy, helped determine positive and negative cutoff levels for ER, and were even used to validate the newer assays.

The application of IHC for ER quantitation gained widespread acceptance in the 1990s with the development of new monoclonal antibodies to ER and different antigen retrieval techniques. IHC offers many advantages over the traditional biochemical assays; most importantly, it can be performed on very small amounts of tumor in formalin-fixed, paraffin-embedded (FFPE) tissue as well as frozen tissue. The immunohistochemical stain is applied to a microscopic slide, which permits direct visualization of anti-ER antibodies binding to tumor cells and helps differentiate ER staining of stromal cells, necrotic tumor, and benign parenchyma from the tumor cells. Furthermore, in situ and invasive components and different morphologic subtypes can be selectively evaluated with IHC. Worldwide, the current practice is to use IHC exclusively on paraffin sections.

Progesterone Receptor: Historical Perspective

Like ER, PR status is an independent predictive factor for benefit from adjuvant endocrine therapy and a prognostic indicator for recurrence in breast cancer. The discovery was preceded by observations that a subset of ER-positive breast cancers failed to respond to hormonal treatment, suggesting that ER was not a sufficient indicator of hormone dependence in breast cancer by itself. ER-positive/PR-positive breast cancers fare better compared with ER-positive/PR-negative tumors to adjuvant treatment. Studies by Horwitz and colleagues, the same group that extensively studied ER, suggested that the presence of PR might serve as an indicator of the functionality of the estrogen signaling pathway in the breast.

PR, too, is a member of the steroid receptor subgroup of ligand-activated transcription factors within the large nuclear receptor superfamily. It contains 946 amino acids, a DNA-binding domain sandwiched between an N-terminal domain with transcription activation and inhibitory functions and a C-terminal ligand-binding domain. The central DNA-binding domain of PR shows considerable sequence homology to that of ER. The two isoforms, namely PR-A and PR-B, are encoded by the same gene (unlike the isoforms of ER) and are identical except that the truncated PR-A is short of 164 amino acids that are seen at the N-terminal end of PR-B. In the normal breast, PR-A and PR-B are coexpressed at equal levels in luminal epithelial cells, suggesting that both proteins are required to mediate physiologically relevant signaling (i.e., formation of lobular-alveolar structures, modulation of milk synthesis, and duct development). In breast cancer, predominance of one isoform is common, suggesting that the resultant unbalanced expression of PR-A and PR-B may induce aberrant targeting of genes. PRs are under the control of E2 or related estrogens. PR is synthesized by tumor cells that are stimulated by estrogens through an interaction with ER. Thus PR is a surrogate marker of ER activity, and it is rare that PR-positive cells do not also express ER.

Ligand-binding assays (sucrose gradient ultracentrifugation and dextran-coated charcoal assays) were the gold standard for early characterization and measurement of PR. With the advent of monoclonal antibodies to PR, IHC largely replaced the biochemical assays for PR measurement in the mid-1990s.

Several studies demonstrated that IHC was superior to ligand-binding assays and enzyme immunoassays for assessing ER and PR status in primary breast cancer and had equivalent or better ability to predict response to adjuvant endocrine therapy. Reproducible and reliable IHC assays with proper standardization are essential for meaningful clinical application.

Receptor Status Assessment: Why is it Important?

The assessment of hormone receptors in primary invasive breast cancer is now mandatory. Both ER and PR are strong predictive factors and relatively weak prognostic factors for response to adjuvant and therapeutic hormonal therapy. Historically, tamoxifen has been the standard adjuvant endocrine treatment for premenopausal women who have ER-positive breast cancer, usually for 5 years. It is now recommended that higher-risk patients receive adjuvant ovarian function suppression (with gonadotropin-releasing hormone, GnRH agonist therapy), in addition to either tamoxifen or aromatase inhibitor (AI) therapy to further lower the risk of breast cancer recurrence. While either tamoxifen or AIs can be used in the adjuvant treatment of postmenopausal patients, AIs are preferred as initial therapy as they may reduce the recurrence risk and improve survival compared with tamoxifen. Over the past decade, multiple randomized control trials have studied the role of extended adjuvant endocrine therapy beyond 5 years. The benefits of extended therapy include reductions in risk of locoregional and distant recurrence and in contralateral breast cancer. According to the current recommendation, women with stage II or III breast cancer should extend endocrine therapy for up to 10 years. Findings from several recent phase III randomized control trials have prompted a shift in these practice guidelines. Based on recent data 10 years of tamoxifen treatment can further reduce recurrence and approximately halve breast cancer mortality during the second decade after diagnosis.

Therapeutic benefit was directly proportional to the level of ER, with patients with higher expression of ER in the breast cancers demonstrating the most benefit. Also, low ER or PR is associated with a high risk of recurrence after hormonal therapy. ER-negative cancers respond favorably to chemotherapy and are more likely to achieve a complete pathologic response after neoadjuvant chemotherapy than ER-positive tumors. Interestingly, single hormone receptor–positive tumors (ER-positive/PR-negative and ER-negative/PR-positive) show less response to tamoxifen, although some reports claim that patients with ER-negative/PR-positive tumors may derive benefit from tamoxifen. Such knowledge underscores the importance of identifying tumors that truly express ER and PR.

Hormone Receptor Assessment

It was expected that IHC, like ligand-binding assays, was an intrinsically quantitative method that demonstrated a direct linear relationship between the amount of ER protein present in tumor cell nuclei and the amount of ER antigen detected by IHC. However, a number of studies have shown that this is not the case. Preanalytic and analytic factors such as tissue fixation time, antigen retrieval method, and antigen detection method influence ER IHC. Controversies in hormone receptor IHC were due to lack of standardization of the test, poor reproducibility among laboratories, and lack of proficiency testing. Standardization of IHC testing and establishing cutoff values for positivity are critical to avoid false-negative results. In 2001 an NIH consensus development panel recommended that patients with any expression of hormone receptor in their tumor cells may benefit from hormonal therapy. They implied that a mere positive or negative ER status suffices in therapeutic decision-making. Later studies, however, emphasized that quantification of hormonal receptors by IHC helps better identify patients who may benefit from adjuvant chemotherapy and also clarify why some patients do not respond to hormonal therapy.