This Article Full Text (PDF) From JCO 1983 Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Google Scholar Articles by Andreopoulou, E. Articles by Hortobagyi, G. N. PubMed PubMed Citation Articles by Andreopoulou, E. Articles by Hortobagyi, G. N.

Prognostic Factors in Metastatic Breast Cancer: Successes and Challenges Toward Individualized Therapy

Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, TX

A pivotal article by Hortobagyi et al¹ in the 1983 Volume 1 issue of Journal of Clinical Oncology addressed the impact of predictive and prognostic models in the setting of metastatic breast cancer. At that time, clinical variables were used in a forward stepwise logistic regression procedure to determine the patient characteristics important in predicting response to doxorubicin-containing chemotherapeutic regimens. Three models for assessing prognosis were generated by this study, which pioneered the use of mathematical algorithmic probability equations and biologic stratification as predictors for individual, rather than random, outcomes.

Within the past two decades, the management of metastatic breast cancer has evolved, with the introduction of new chemotherapeutic agents (taxanes, vinorelbine, gemcitabine, and capecitabine), newer hormonal agents, including third-generation aromatase inhibitors, and biologic therapies, such as trastuzumab, bevacizumab, and lapatinib. However, metastatic breast cancer remains incurable, though therapeutic innovation has resulted in modest improvements in survival rates. The original article proposed a model of prognostication with a purely demographic and anatomic base. Since then, thinking has evolved to a biologic and genetic platform. The introduction of novel technology has enabled scientists to further classify breast tumors using methods that can measure differences in thousands of biologic pathways simultaneously. The purpose of this editorial is to highlight the article's original perspective in light of the rapidly changing field of predictors of response and individualized therapy for the treatment of patients with breast cancer.

The original models proposed are still useful in the estimation of risk in patients with breast cancer. Even today, anatomic staging continues to play a major role in guiding treatment decisions. The extent of the disease, disease-free intervals, prior adjuvant therapy, and performance status remain of prognostic value. Although the level of risk may be useful in deciding between certain types of regimens (chemotherapy v hormonal therapy) as an initial treatment option, it does not offer guidance as to actual drug selection for individual patients.

It has become clear that the unpredictable clinical behavior of metastatic breast cancer reflects the biologic heterogeneity of the disease. This heterogeneity introduces subgroups that benefit from therapeutic innovations and likely account for the modest improvements in survival. Current technology provides a powerful method to help identify and validate new drug targets, as well as potential diagnostic and prognostic markers for use as predictors of response to new molecular therapeutics. In light of this, the focus has moved increasingly away from dose escalation and intensification of chemotherapy in otherwise unselected patients to the premise of targeting specific biologic pathways in selected patients using pharmacologic or immunologic methods.

Although the prototype target in breast cancer has been known for approximately 40 years, we only came to accept, after the 1995 Oxford overview, that essentially all the benefit derived from tamoxifen is limited to patients with estrogen receptor (ER) –positive breast cancer, regardless of stage and age.² However, not every patient with ER-positive breast cancer benefits from tamoxifen. For this reason, several endocrine approaches have been developed that interrupt estrogen stimulation, either by directly modulating the ER-signaling pathway or by lowering serum or tumor concentrations of estrogen. Beyond the classic pathway, new evidence suggests that cross-talk exists between the ER and growth factor receptor pathways, contributing to de novo and acquired resistance to endocrine therapy.³ Inappropriate activation of growth factor signaling can readily promote endocrine therapy failure in breast cancer cells, by either overriding the growth-inhibitory properties of antiestrogenic drugs or by the establishment of a new self-propagating autocrine loop that efficiently drives resistant cell growth.

Modern therapies also target deregulated critical biochemical pathways present in cancer cells that result in changes in normal cellular processes such as apoptosis, proliferation, angiogenesis, DNA repair, cell cycle progression, invasion, and metastases.⁴ Like ER, some of these molecular defects have prognostic or predictive value, whereas others are implicated in resistance to chemotherapy or hormone therapy. Drugs that target these pathways have been demonstrated to improve response and/or survival in patients with metastatic breast cancer. The adverse prognosis of the human epidermal growth factor receptor 2 (HER2) –positive breast cancer has been recognized since the seminal publication by Slamon et al⁵ 20 years ago. Ten years ago, randomized trials of anti-HER2 therapy added to first-line chemotherapy in patients with HER2-overexpressing metastatic disease showed increased clinical benefit and survival.⁶ For certain agents, such as lapatinib and trastuzumab, a preclinical knowledge of the drugs’ target led to appropriate patient selection, which improved response to therapy and enhanced clinical development. In these clinical trials, patient selection was performed using immunohistochemistry or fluorescent in situ hybridization to detect the target receptor. Although beneficial for the development of these agents, both technologies have limitations in reproducibility, interpretation, and tissue requirements needed to measure multiple biologic pathways.

The application of modern genomic technology is revealing a new genetic taxonomy that reclassifies cancers based on a molecular signature. The availability of the human genome and subsequent sequencing of cancer genomes are having an impact on the prognostic classification of tumors and discovery of targets. Several investigators have shown that gene expression profiles differ across tumor types defined by ER/progesterone receptor and HER2. Sorlie and Perou⁷ initially identified four tumor groups (basal like, HER2/neu+, normal breast, and ER-positive luminal type) based on gene expression patterns. Of particular interest was the finding that ER-positive tumors may be divided into at least two distinctive groups, luminal type A and B, with different outcomes. Another important implication is that ER-negative breast cancer includes at least two biologically distinct subtypes: The HER2-positive type and the basal-like type, which represents a different clinical entity that is associated with shorter survival and high frequency of P53 mutations.

Several lessons were learned regarding heterogeneity in well-studied pathways such as HER-2. For example, PTEN deficiency has been reported as a powerful predictor for trastuzumab resistance. Additionally, studies suggest that PI3K-targeting therapies could overcome this resistance.⁸ Furthermore, investigators have also recently reported that the HER-2/neu status of tumor cells detected in metastatic sites, bone marrow, or peripheral blood of patients with advanced disease differ from that of the original primary tumor, suggesting a clonal selection or genetic instability.⁹ Moreover, the presence of concomitant alterations of other oncogenes like topo-IIa seems to better define biologic subsets of the disease and may contribute to treatment decisions.

Researchers from our group have observed that tau expression modulates response to paclitaxel, suggesting that low microtubule-associated protein tau mRNA may be a promising single gene marker.¹⁰ An association between low tau expression and ER-negative status was also reported. This may partly explain why ER-negative disease is more sensitive to paclitaxel. Many tumors are not fully sensitive despite low tau expression, suggesting additional pathways of resistance. This is consistent with the belief that response to chemotherapy is a multifactorial process and that no single marker will be informative. Thus multigene predictors that use information from several distinct molecular pathways of resistance will likely be more powerful than any single gene.

The integration of genomic and functional data allows for better understanding of how the transcriptome responds to drug treatment, how these changes may be predictive of sensitivity, and the identification of molecular pharmacodynamic markers of drug response.

It is likely that tumors are driven by combinatorial effects of several molecular abnormalities. Simultaneously tackling some or all of these molecular drivers is likely to provide most therapeutic benefit and reduce the risk of resistance. As an example, inhibition of poly(ADP-ribose) polymerase might potentiate the activity of DNA-damaging agents in BRCA1 mutation carriers. The majority of breast cancers that develop in BRCA1 mutation carriers are basal like. Preclinical and clinical studies of BRCA1-mutated breast cancers suggest sensitivity to DNA damaging chemotherapeutics. Inhibition of the PI3K/AKT signal that regulates translocation of MDM-2 to the nucleus results indirectly in persistent p53 activation, enhancing chemosensitivity. Of particular interest are the heat shock protein 90 molecular chaperone, which has the ability to affect several cancer drugs simultaneously.

The ability to detect the presence of minimal residual disease in the bone marrow, lymph nodes, and peripheral blood also provides new opportunities to direct patient therapy. In 2004, a study by Cristofanilli et al¹¹ demonstrated the significance of circulating tumor cells in patients with measurable metastatic breast cancer and the prognostic implications before initiating therapy. Subsequently, important questions about the biologic characteristics of circulating cancer cells and the reasons for the reduced capacity of systemic treatments to arrest or eradicate the cancer were raised. Comprehensive analysis of circulating tumor cells is likely to provide new insights into the biology of breast cancer and will contribute to defining novel treatments and better predictive scope.

Efforts being made to genotype and phenotype these cells should lead to further monitoring of these rare events. Stem-like progenitor cells may also be at play and will need to be targeted in future approaches.

Furthermore, personalized treatment selection studies in metastatic breast cancer based on pharmacogenomic predictors aims to determine whether selection of patients by molecular features will increase the clinical benefit rate. Genotyping of CYP2D6 and other enzymes involved in tamoxifen metabolism genotype can best guide selection of endocrine therapy for breast cancer.

The ultimate biomedical application of basic research is the translation of fundamental knowledge into cost-effective tailored treatment. Future emphasis should be on the contemporaneous development of molecular diagnostics and therapeutics, because one without the other precludes optimal implementation of personalized therapy.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

Conception and design: Eleni Andreopoulou, Gabriel N. Hortobagyi

Manuscript writing: Eleni Andreopoulou, Gabriel N. Hortobagyi

Final approval of manuscript: Eleni Andreopoulou, Gabriel N. Hortobagyi

REFERENCES

1. Hortobagyi GN, Smith TL, Legha SS, et al: Multivariate analysis of prognostic factors in metastatic breast cancer. J Clin Oncol 1:776-786, 1983[Abstract]

2. Tamoxifen for early breast cancer: An overview of the randomized trials—Early Breast Cancer Trialists’ Collaborative Group. Lancet 351:1451-1467, 1998[CrossRef][Medline]

3. Shou J, Massarweh S, Osborne CK, et al: Mechanisms of tamoxifen resistance: Increased estrogen receptor-HER2/neu cross-talk in ER/HER2-positive breast cancer. J Natl Cancer Inst 96:926-935, 2004[Abstract/Free Full Text]

4. Hanahan D, Weinberg RA: The hallmarks of cancer. Cell 100:57-70, 2000[CrossRef][Medline]

5. Slamon DJ, Clark GM, Wong SG, et al: Human breast cancer: Correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235:177-182, 1987[Abstract/Free Full Text]

6. Slamon DJ, Leyland-Jones B, Shak S, et al: Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344:783-792, 2001[Abstract/Free Full Text]

7. Perou CM, Sørlie T, Eisen MB, et al: Molecular portraits of human breast tumours. Nature 406:747-752, 2000[CrossRef][Medline]

8. Nagata Y, Lan KH, Zhou X, et al: PTEN activation contributes to tumor inhibition by trastuzumab and loss of PTEN predicts trastuzumab resistance in patients. Cancer Cell 6:117-127, 2004[CrossRef][Medline]

9. Meng S, Tripathy D, Shete S, et al: HER-2 gene amplification can be acquired as breast cancer progresses. Proc Natl Acad Sci U S A 101:9393-9398, 2004[Abstract/Free Full Text]

10. Rouzier R, Rajan R, Wagner P, et al: Microtubule-associated protein tau: A marker of paclitaxel sensitivity in breast cancer. Proc Natl Acad Sci U S A 102:8315-8320, 2005[Abstract/Free Full Text]

11. Cristofanilli M, Budd GT, Ellis MJ, et al: Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 351:781-791, 2004[Abstract/Free Full Text]

This Article Full Text (PDF) From JCO 1983 Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Google Scholar Articles by Andreopoulou, E. Articles by Hortobagyi, G. N. PubMed PubMed Citation Articles by Andreopoulou, E. Articles by Hortobagyi, G. N.