Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Strategies to Translate Preclinical Information to Breast Cancer Patient Benefit

Strategies to Translate Preclinical Information to Breast Cancer Patient Benefit Abstract Despite the progress in understanding breast cancer biology, translation of basic findings into clinical applications still appears to be a complex process, and few molecular markers/signatures are in routine clinical use or currently challenged for their clinical utility. Disease complexity, certainly, represents an obstacle to successful translation, but methodological pitfalls in development and validation steps also contribute. Translational research should be planned as a round-trip from the bench to the bedside and back. The preoperative/neoadjuvant setting represents an ideal model because it allows identification and validation of treatment response predictors and of pharmacodynamic markers associated with clinical downstaging, investigations on in vivo action mechanism of drugs, and indirect validation of findings from preclinical models. Availability of well-annotated, high-quality biospecimens; standardized, reproducible, and robust assays to detect molecular markers/signatures even on few cells; prospective planning of study design; and regulatory issues adequately fitting preclinical and clinical needs represent fundamental assets for translational studies. After decades of research aimed at identifying cellular and molecular alterations correlated with, or causative of, breast cancer progression, the translation of basic findings into clinical applications as well as our understanding of breast cancer biology are still modest. Despite major achievements such as the development of powerful drugs specifically targeting molecular alterations, the development of genomic classifiers associated with metastatic potential, as well as the possibility to predict response and toxicity for several drugs, the outcome of the first decade of the genome era has not yet substantially affected the care of breast cancer patients. However, the identification of molecular intrinsic subtypes, which are associated to distinct patient clinical outcomes (1), corroborated biologically driven treatment options. In fact, hormonal therapy is the first choice treatment for luminal A, slowly proliferating estrogen receptor (ER)–positive tumors, trastuzumab is the standard of care for HER2-positive cancers, and chemotherapy with dose-dense regimens is recommended for basal-like ER-negative rapidly proliferating tumors. Several obstacles, including the complexity of disease biology itself, but also of analytical approaches used to develop molecular predictors in some ways limited a successful translation. However, a lack of preclinical focus on the intended clinical application, and some misunderstanding about the entire process of development and validation, coupled with gaps in effective and efficient collaboration among the key players involved in the different phases of translating candidate markers/signatures into routine practice, played a major role. The Round Trip of Translational Research The pathway toward personalized medicine is currently a work-in-progress, which takes advantage of continuous technology developments and of an open access to omics data for the recognition of genetic and environmental risk factors, the definition of prognostic signatures (even specific for the different metastatic sites), and the identification of new drug targets. Optimization of opportunities to generate new hypotheses definitely relies on a “round-trip” model, with a translation of information direct from preclinical to clinical development and back to preclinical studies passing through, and gathering information from, clinical validation. Examples of translational research in breast cancer account for success stories, with a biunivocal relation between target and drug (ie, besides ER, the “first” druggable target targeted by selective estrogen receptor modulators, also HER2 targeted by trastuzumab, HER2 and HER1/epidermal growth factor receptor targeted by lapatinib, BRCAness, and poly (ADP-ribose) polymerase inhibitors). However, also some failures have been recently observed (ie, antiangiogenic therapy in advanced cancers). Obstacles in translating preclinical data into a high level of evidence of clinical utility have been recently discussed by Richard Simon (2). Even when the clinical needs have been clearly focused, each of the three main components of the translational research pathway appears to be characterized by problems and pitfalls (Figure 1). During the preclinical phase, use of the classical hypothesis-driven approach may be characterized by some reductionism and by limitations due to inadequacy/lack of preclinical models. Conversely, the novel data-driven approach, which takes advantage of the enormous amount of information obtained by omics approaches, may instead suffer from poor understanding of biology and inadequate validation of data. Figure 1 View largeDownload slide The full circle of translational research in cancer. Figure 1 View largeDownload slide The full circle of translational research in cancer. Clinical translation of preclinical findings represents a very complex phase, where the bench-derived hypothesis has to be challenged in the context of the patient. Such phase is often retrospective and requires availability of well-annotated clinical biospecimens coupled with a strong interaction between preclinical and clinical investigators (3). Its main pitfalls include the lack of a proper validation; the use of “convenience samples” (instead of samples derived from case series of patients homogeneous in terms of clinicopathologic features, treatments, and tumor biologic features); and poor consideration of assay reproducibility, robustness, sensitivity, and specificity (4). Inadequate performance of the clinical developmental phase is often responsible for the failure of translating biological information into clinical practice. In fact, despite the huge number of articles published in the last 30 years on prognosticators or predictors of treatment response, only ER and progesterone receptor, HER2 amplification, the biochemical markers uPA and PAI-1, and the OncoType Dx (Genomic Health, Redwood City, CA) recurrence score have been recommended by the American Society of Clinical Oncology consensus panel for their clinical utility as risk selection criteria in early-stage breast cancer patients (5). The round trip for translational research leads to a final phase of clinical validation/utility, a step that should be performed in the clinical context in which the molecular marker/classifier has been identified, and indeed requires standardized and reproducible assays, a predefined statistical plan, and consecutive biological specimens from an appropriate patient population. To speed-up clinical validation and translation of biomarkers, a “prospective–retrospective” design has been recently proposed, which employs the use of archived specimens from patients entered a previous prospective clinical trial and an a priori written definition of the study protocol (6). The Preoperative Therapy Model To speed-up the identification of predictors of treatment response within the context of clinical trials, an ideal translational model can be identified in the preoperative (or neoadjuvant) setting (7) for the following reasons: 1) breast tumors are clinically detectable, measurable, and accessible for sequential tissue sampling; 2) preoperative treatment and serial sampling offer great opportunities to study tumor biology; 3) preoperative treatment is relatively short, and results are quickly available, thus allowing a rapid assessment of treatment efficacy; 4) available techniques permit to perform comprehensive molecular analyses even on few cells. Such a clinical approach allows pre- and posttreatment sampling to evaluate mRNA and microRNA expression, gene mutation, amplification and methylation, protein expression, single-nucleotide polymorphisms, and to quantify and qualify circulating tumor cells. Pretreatment biologic findings could be challenged as predictors of treatment response as a function of tumor volume reduction and/or of pathological response. Posttreatment changes of biological profiles could instead provide information on treatment activity in terms of pharmacodynamic markers and action mechanisms for new drugs. The use of primary systemic therapy as a test bench for the identification of molecular predictors of treatment response was classified among the highest priority topics by an international Web-based consultation of breast cancer professionals to identify the issues most widely considered to be of the utmost importance (8). Overview of Translational Studies Performed in the Context of Primary Systemic Treatments About 500 articles on predictive factors have been published in the last 25 years on more than 4000 women with stage I–III breast cancers subjected to preoperative cytotoxic and/or endocrine systemic treatments. In the pregenomic era, besides ER and tumor grade, pretreatment proliferation indices and apoptosis-related markers proved to moderately predict treatment efficacy, whereas posttreatment proliferation was predictive of long-term clinical outcome (9–16), regardless of the type of primary treatment (Table 1). Over time, collection of biospecimens and biological information from all the women entering clinical trials became the rule because majority of the patients’ consent to donate leftover material for research. The possibility of performing comprehensive molecular analyses on limited amount of tumor cells or on formalin-fixed paraffin-embedded material and on blood samples allowed for identification gene expression signatures predictive of likelihood of tumor reduction and/or pathologic complete response following endocrine (17) or cytotoxic (18,19,20) treatments, as well as for detection of changes in serum protein levels associated with clinical outcome (21). Table 1 Overview of biomarker results from preoperative systemic treatment: association with clinical endpoints* Clinical endpoint  Primary chemotherapy  Primary endocrine therapy  Tumor shrinkage  High pretreatment PI  Changes in PI    Absence of ER and/or PgR  ER and/or PgR positivity    High histological grade  EGFR expression    Increased apoptotic index  HER2 overexpression    wt-p53      mdr-1, P-gp changes    DFS and/or OS  Low-posttreatment PI  Changes in PI    Low histological grade      wt-p53      Low P-gp      Low HER2 expression    Clinical endpoint  Primary chemotherapy  Primary endocrine therapy  Tumor shrinkage  High pretreatment PI  Changes in PI    Absence of ER and/or PgR  ER and/or PgR positivity    High histological grade  EGFR expression    Increased apoptotic index  HER2 overexpression    wt-p53      mdr-1, P-gp changes    DFS and/or OS  Low-posttreatment PI  Changes in PI    Low histological grade      wt-p53      Low P-gp      Low HER2 expression    * References (9–16). DFS = disease-free survival; EGFR = epidermal growth factor receptor; ER = estrogen receptor; OS = overall survival; PgR = progestrone receptor; PI = proliferation indices; P-gp = P-glycoprotein. View Large Overall, none of the reported gene signatures is robust enough to be clinically useful mainly because of the lack of appropriate validation, and translational studies often suffer from small case series and statistical faults. Such studies, however, represent a proof-of-principle that gene and protein expression profiling may predict response to preoperative systemic treatments. Novel Approaches to Turn Preclinical Information into Clinical Benefit Genomic Classifiers Derived From Cell Lines and Small Interfering RNA Screening Modern approaches to discovering treatment response predictors rely on mechanism-driven (or biology-driven) hypotheses (Figure 2). At present, such approaches resort to the use of 1) cell lines and/or xenograft models to define gene expression signatures associated with drug response and 2) functional genomics to predict chemosensitivity. The former approach has been used to better understand the temporal response to tamoxifen by identifying treatment-induced changes in gene expression in an ER-positive breast cancer xenograft model (22), and the resulting early/transient proliferation and continuous/late/estrogen response genes proved to be associated with clinical outcome in 404 women entered four tamoxifen-treated subsets. Figure 2 View largeDownload slide Functional genomics to accelerate the discovery of drug-specific response predictors. Figure 2 View largeDownload slide Functional genomics to accelerate the discovery of drug-specific response predictors. The functional genomic approach is based on silencing target genes in cultured cells to systematically study the ability of individual genes to alter tumor chemosensitivity. The combining of experimental findings with expression profile data from human tumors allows to identify clinical relevant regulators of drug sensitivity. Using such a preclinical approach, Juul et al. (23) identified a paclitaxel response metagene subsequently validated as a pathologic complete response predictor in a series of triple-negative tumors. Tumor-Initiating Cells (TICs) and Drug Resistance The concept that cancer may be a stem cell disease, arising from tissue stem/progenitor cells or driven by cancer cells with stem cell properties, has important implications for the development of new therapeutic strategies able to target the limited fraction of TICs, which share chemoresistance properties with normal stem cells. Present findings support the hypothesis that the residual posttreatment tumors are enriched of TICs because 1) classical chemotherapy appears to increase the proportion of TICs (CD44+/CD24−//low, ALDH1+) (24); 2) epithelial–mesenchymal transition markers are frequently overexpressed (25); 3) residual cancer cells exhibit a gene expression signature peculiar of breast cancer–initiating cells (26). The therapeutic implications of this model system are valuable because TICs might represent new targets, with distinct possible strategies, such as interference with self-renewal and TIC phenotype reversal through epigenetic modulation/reprogramming. Collection of Human Biological Material in Clinical Trials: Opportunities and Obstacles Access to human biological material for research is considered a major bottleneck hindering successful translation of information from bench to bedside. The potential of this tool is provided by the enormous amount of information contained in biological samples, which, along with the continuing evolution of technologies, represent a great opportunity for new treatment strategies. Biobanks are designed to maximize this research potential, with foreseeable advantages for the community and for the industry, which may be able to transform such knowledge into products, whether diagnostics or therapeutics, for health systems. Due to the complex, multidisciplinary nature of biomaterial collection and translational research, integration of biological material collection into clinical trials warrants careful up-front planning and input from a range of expertise, which take into consideration several aspects also including logistics and standard operating procedures for biological material collection, and test to control for some sources of preanalytical variability, such as those due to timing and tissue handling. In the field of biobanking research, many ethical and legal issues are now open, particularly those affecting relationships with donors, researchers, and sponsors. Today, the main problems are dealing with 1) meaning and implications of the concept of ownership of biological material and related information, and of their “donation” for research; 2) extent of informed consent; 3) withdrawal of consent and its consequences, protection of personal data, availability of information useful for the patients and for their family members; 4) guarantees for the community (ie, availability of research results and technologies derived) and individuals (confidentiality of data). The guarantees should affect: Function of the biobank and its procedures (public interest, independence, transparency, confidentiality and professional secrecy, explicit agreements of transfer, and standards of conservation). Decision-making processes and players involved (institutional review boards, ethics committees, boards of trustees including representatives of associations of patients and citizens). Information and communication to all stakeholders (citizens and their associations, health institutions, public and private research institutes, industry, and political institutes). Rights of donors (personal information, use of the samples, and consent withdrawal). Conclusions The most important message derived from decades of translational research in breast cancer relies on focusing on clinically relevant needs, with a strong, intense, and continuous collaboration among preclinical and clinical investigators. Such an effort should be pursued at each step of the round trip from the bench to the bedside, and back. Instrumental to such a translational model are: 1) availability of well-annotated and high-quality biological specimens from patients entered in clinical trials; 2) standardized, reproducible, and robust assays to detect molecular markers/signatures even on limited tumor material and/or on formalin-fixed paraffin-embedded material; 3) quality assessment and assurance of reagents and test systems, including preanalytical processing, commercial kits, and equipment; 4) prospective planning of the study design, even using a prospective–retrospective approach (5); and 5) regulatory issues adequately fitting the need of translational studies. Moreover, critical issues that cannot be considered independently are analytical performance and clinical utility because the clinical value of a test cannot disregard assay reliability, and assay performance cannot overlook its clinical use (3). The neoadjuvant setting represents an ideal model to directly challenge predictive value and clinical utility of molecular markers/signatures identified on specimens of patients entered clinical trials, as well as to validate genomic classifiers derived from preclinical experimental models and functional genomic studies. Funding This work was supported by grants from the Associazione Italiana per la Ricerca sul Cancro (AIRC, the Italian Association for Cancer Research, grant number 4532 to M.G.D.) and by the Italian Ministry of Health (Special Project on Female Cancers). This work was in part supported also by funds obtained through an Italian law that allows taxpayers to allocate the “5 × 1000” share of their payments to support a research institution of their choice. References 1. Morrow PK,  Hortobagyi GN.  Management of breast cancer in the genome era,  Annu Rev Med. ,  2009, vol.  60  2(pg.  153- 165) 2. Simon R.  Lost in translation: problems and pitfalls in translating laboratory observations to clinical utility,  Eur J Cancer. ,  2008, vol.  44  18(pg.  2707- 2713) 3. Taube SE,  Clark GM,  Dancey JE,  McShane LM,  Sigman CC,  Gutman SI.  A perspective on challenges and issues in biomarker development and drug and biomarker codevelopment,  J Natl Cancer Inst. ,  2009, vol.  101  21(pg.  1453- 1463) 4. Taube SE.  Biomarkers in oncology: trials and tribulations,  Ann N Y Acad Sci. ,  2009, vol.  1180 (pg.  111- 118) 5. Harris L,  Fritsche H,  Mennel R, et al.  American Society of Clinical Oncology 2007 update of recommendations for the use of tumor markers in breast cancer,  J Clin Oncol. ,  2007, vol.  25  33(pg.  5287- 5312) 6. Simon RM,  Paik S,  Hayes DF.  Use of archived specimens in evaluation of prognostic and predictive biomarkers,  J Natl Cancer Inst. ,  2009, vol.  101  21(pg.  1446- 1452) 7. Burstein HJ.  Preoperative therapy as a model for translational research in breast cancer,  Cancer Invest. ,  2008, vol.  26  3(pg.  217- 221) 8. Dowsett M,  Goldhirsch A,  Hayes DF,  Senn H-J,  Wood W,  Viale G.  International web-based consultation on priorities for translational breast cancer research,  Breast Cancer Res. ,  2007, vol.  9  6pg.  R81  9. Wolff AC,  Berry D,  Carey LA, et al.  Research issues affecting preoperative systemic therapy for operable breast cancer,  J Clin Oncol. ,  2008, vol.  26  5(pg.  806- 813) 10. Daidone MG,  Benini E,  Tomasic G, et al.  Biological markers as indicators of response to primary and adjuvant chemotherapy in breast cancer,  Int J Cancer. ,  1999, vol.  84  6(pg.  580- 586) 11. Ellis MJ,  Coop A,  Singh B, et al.  Letrozole is more effective neoadjuvant endocrine therapy than Tamoxifen for ErbB-1 and/or ErbB-2-positive, estrogen receptor-positive primary breast cancer: evidence for a phase III randomized trial,  J Clin Oncol. ,  2001, vol.  19  18(pg.  3808- 3816) 12. Archer CD,  Parton M,  Smith IE, et al.  Early changes in apoptosis and proliferation following primary chemotherapy for breast cancer,  Br J Cancer. ,  2003, vol.  89  6(pg.  1035- 1041) 13. Dowsett M,  Ebbs SR,  Dixon JM, et al.  Biomarker changes following neoadjuvant anastrozole, tamoxifen, or the combination: influence of hormonal status and HER-2 in breast cancer—a study from the IMPACT trialists,  J Clin Oncol. ,  2005, vol.  23  11(pg.  2477- 2492) 14. Caudle AS,  Gonzales-Angulo AM,  Hunt KK, et al.  Predictors of tumor progression during neoadjuvant chemotherapy in breast cancer,  J Clin Oncol. ,  2010, vol.  28  11(pg.  1821- 1828) 15. Mohsin SK,  Weiss HL,  Gutierrez MC, et al.  Neoadjuvant trastuzumab induces apoptosis in primary breast cancer,  J Clin Oncol. ,  2005, vol.  23  11(pg.  2460- 2468) 16. Daidone MG,  Veneroni S,  Benini E, et al.  Howell A,  Dowsett M.  Biological markers and changes induced in their profiles following primary chemotherapy: relevance for short- and long-term clinical outcome,  ESOI Scientific Updates, Vol. 4. Philadelphia, PA: Elsevier ,  1999(pg.  53- 72) 17. Cappelletti V,  Gariboldi M,  De Cecco L, et al.  Patterns and changes in gene expression following neo-adjuvant anti-estrogen treatment in estrogen receptor-positive breast cancer,  Endocr Relat Cancer. ,  2008, vol.  15  2(pg.  439- 449) 18. Chang JC,  Wooten EC,  Tsimelzon A, et al.  Gene expression profiling for the prediction of therapeutic response to docetaxel in patients with breast cancer,  Lancet. ,  2003, vol.  362  9381(pg.  362- 369) 19. Gianni L,  Zambetti M,  Clark K, et al.  Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer,  J Clin Oncol. ,  2005, vol.  23  29(pg.  7265- 7277) 20. Tabchy A,  Valero V,  Vidaurre T, et al.  Evaluation of a 30-gene paclitaxel, fluorouracil, doxorubicin and cyclophosphamide chemotherapy response predictor in a multicenter randomized trial in breast cancer,  Clin Cancer Res. ,  2010, vol.  16  21 21. Mazouni C,  Baggerly K,  Hawke D, et al.  Evaluation of changes in serum protein profiles during neoadjuvant chemotherapy in HER2-positive breast cancer using LC-MALDI-TOF/MS procedure,  Proteomics. ,  2010, vol.  10  19(pg.  3525- 3532) 22. Taylor KJ,  Sims AH,  Liang L, et al.  Dynamic changes in gene expression in vivo predict prognosis of tamoxifen-treated patients with breast cancer,  Breast Cancer Res. ,  2010, vol.  12  3pg.  R39  23. Juul N,  Szallasi Z,  Eklund AC, et al.  Assessment of an RNA interference screen-derived mitotic and ceramide pathway metagene as a predictor of response to neoadjuvant paclitaxel for primary triple-negative breast cancer: a retrospective analysis of five clinical trials,  Lancet Oncol. ,  2010, vol.  11  4(pg.  358- 365) 24. Li X,  Lewis MT.  Huang J, et al.  Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy,  J Natl Cancer Inst. ,  2008, vol.  100  9(pg.  672- 679) 25. Yu F,  Yao H,  Zhu P, et al.  let-7 regulates self renewal and tumorigenicity of breast cancer cells,  Cell. ,  2007, vol.  131  6(pg.  1109- 1123) 26. Creighton CJ,  Li X,  Landis M, et al.  Residual breast cancer after conventional therapy display mesenchymal as well as tumor-initiating features,  Proc Natl Acad Sci U S A. ,  2009, vol.  106  33(pg.  13820- 13825) Published by Oxford University Press 2011. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI Monographs Oxford University Press

Strategies to Translate Preclinical Information to Breast Cancer Patient Benefit

Download PDF
5 pages

Loading next page...
 
/lp/oxford-university-press/strategies-to-translate-preclinical-information-to-breast-cancer-e8yjC1IrvE

References (25)

Publisher
Oxford University Press
Copyright
Published by Oxford University Press 2011.
ISSN
1052-6773
eISSN
1745-6614
DOI
10.1093/jncimonographs/lgr033
pmid
22043041
Publisher site
See Article on Publisher Site

Abstract

Abstract Despite the progress in understanding breast cancer biology, translation of basic findings into clinical applications still appears to be a complex process, and few molecular markers/signatures are in routine clinical use or currently challenged for their clinical utility. Disease complexity, certainly, represents an obstacle to successful translation, but methodological pitfalls in development and validation steps also contribute. Translational research should be planned as a round-trip from the bench to the bedside and back. The preoperative/neoadjuvant setting represents an ideal model because it allows identification and validation of treatment response predictors and of pharmacodynamic markers associated with clinical downstaging, investigations on in vivo action mechanism of drugs, and indirect validation of findings from preclinical models. Availability of well-annotated, high-quality biospecimens; standardized, reproducible, and robust assays to detect molecular markers/signatures even on few cells; prospective planning of study design; and regulatory issues adequately fitting preclinical and clinical needs represent fundamental assets for translational studies. After decades of research aimed at identifying cellular and molecular alterations correlated with, or causative of, breast cancer progression, the translation of basic findings into clinical applications as well as our understanding of breast cancer biology are still modest. Despite major achievements such as the development of powerful drugs specifically targeting molecular alterations, the development of genomic classifiers associated with metastatic potential, as well as the possibility to predict response and toxicity for several drugs, the outcome of the first decade of the genome era has not yet substantially affected the care of breast cancer patients. However, the identification of molecular intrinsic subtypes, which are associated to distinct patient clinical outcomes (1), corroborated biologically driven treatment options. In fact, hormonal therapy is the first choice treatment for luminal A, slowly proliferating estrogen receptor (ER)–positive tumors, trastuzumab is the standard of care for HER2-positive cancers, and chemotherapy with dose-dense regimens is recommended for basal-like ER-negative rapidly proliferating tumors. Several obstacles, including the complexity of disease biology itself, but also of analytical approaches used to develop molecular predictors in some ways limited a successful translation. However, a lack of preclinical focus on the intended clinical application, and some misunderstanding about the entire process of development and validation, coupled with gaps in effective and efficient collaboration among the key players involved in the different phases of translating candidate markers/signatures into routine practice, played a major role. The Round Trip of Translational Research The pathway toward personalized medicine is currently a work-in-progress, which takes advantage of continuous technology developments and of an open access to omics data for the recognition of genetic and environmental risk factors, the definition of prognostic signatures (even specific for the different metastatic sites), and the identification of new drug targets. Optimization of opportunities to generate new hypotheses definitely relies on a “round-trip” model, with a translation of information direct from preclinical to clinical development and back to preclinical studies passing through, and gathering information from, clinical validation. Examples of translational research in breast cancer account for success stories, with a biunivocal relation between target and drug (ie, besides ER, the “first” druggable target targeted by selective estrogen receptor modulators, also HER2 targeted by trastuzumab, HER2 and HER1/epidermal growth factor receptor targeted by lapatinib, BRCAness, and poly (ADP-ribose) polymerase inhibitors). However, also some failures have been recently observed (ie, antiangiogenic therapy in advanced cancers). Obstacles in translating preclinical data into a high level of evidence of clinical utility have been recently discussed by Richard Simon (2). Even when the clinical needs have been clearly focused, each of the three main components of the translational research pathway appears to be characterized by problems and pitfalls (Figure 1). During the preclinical phase, use of the classical hypothesis-driven approach may be characterized by some reductionism and by limitations due to inadequacy/lack of preclinical models. Conversely, the novel data-driven approach, which takes advantage of the enormous amount of information obtained by omics approaches, may instead suffer from poor understanding of biology and inadequate validation of data. Figure 1 View largeDownload slide The full circle of translational research in cancer. Figure 1 View largeDownload slide The full circle of translational research in cancer. Clinical translation of preclinical findings represents a very complex phase, where the bench-derived hypothesis has to be challenged in the context of the patient. Such phase is often retrospective and requires availability of well-annotated clinical biospecimens coupled with a strong interaction between preclinical and clinical investigators (3). Its main pitfalls include the lack of a proper validation; the use of “convenience samples” (instead of samples derived from case series of patients homogeneous in terms of clinicopathologic features, treatments, and tumor biologic features); and poor consideration of assay reproducibility, robustness, sensitivity, and specificity (4). Inadequate performance of the clinical developmental phase is often responsible for the failure of translating biological information into clinical practice. In fact, despite the huge number of articles published in the last 30 years on prognosticators or predictors of treatment response, only ER and progesterone receptor, HER2 amplification, the biochemical markers uPA and PAI-1, and the OncoType Dx (Genomic Health, Redwood City, CA) recurrence score have been recommended by the American Society of Clinical Oncology consensus panel for their clinical utility as risk selection criteria in early-stage breast cancer patients (5). The round trip for translational research leads to a final phase of clinical validation/utility, a step that should be performed in the clinical context in which the molecular marker/classifier has been identified, and indeed requires standardized and reproducible assays, a predefined statistical plan, and consecutive biological specimens from an appropriate patient population. To speed-up clinical validation and translation of biomarkers, a “prospective–retrospective” design has been recently proposed, which employs the use of archived specimens from patients entered a previous prospective clinical trial and an a priori written definition of the study protocol (6). The Preoperative Therapy Model To speed-up the identification of predictors of treatment response within the context of clinical trials, an ideal translational model can be identified in the preoperative (or neoadjuvant) setting (7) for the following reasons: 1) breast tumors are clinically detectable, measurable, and accessible for sequential tissue sampling; 2) preoperative treatment and serial sampling offer great opportunities to study tumor biology; 3) preoperative treatment is relatively short, and results are quickly available, thus allowing a rapid assessment of treatment efficacy; 4) available techniques permit to perform comprehensive molecular analyses even on few cells. Such a clinical approach allows pre- and posttreatment sampling to evaluate mRNA and microRNA expression, gene mutation, amplification and methylation, protein expression, single-nucleotide polymorphisms, and to quantify and qualify circulating tumor cells. Pretreatment biologic findings could be challenged as predictors of treatment response as a function of tumor volume reduction and/or of pathological response. Posttreatment changes of biological profiles could instead provide information on treatment activity in terms of pharmacodynamic markers and action mechanisms for new drugs. The use of primary systemic therapy as a test bench for the identification of molecular predictors of treatment response was classified among the highest priority topics by an international Web-based consultation of breast cancer professionals to identify the issues most widely considered to be of the utmost importance (8). Overview of Translational Studies Performed in the Context of Primary Systemic Treatments About 500 articles on predictive factors have been published in the last 25 years on more than 4000 women with stage I–III breast cancers subjected to preoperative cytotoxic and/or endocrine systemic treatments. In the pregenomic era, besides ER and tumor grade, pretreatment proliferation indices and apoptosis-related markers proved to moderately predict treatment efficacy, whereas posttreatment proliferation was predictive of long-term clinical outcome (9–16), regardless of the type of primary treatment (Table 1). Over time, collection of biospecimens and biological information from all the women entering clinical trials became the rule because majority of the patients’ consent to donate leftover material for research. The possibility of performing comprehensive molecular analyses on limited amount of tumor cells or on formalin-fixed paraffin-embedded material and on blood samples allowed for identification gene expression signatures predictive of likelihood of tumor reduction and/or pathologic complete response following endocrine (17) or cytotoxic (18,19,20) treatments, as well as for detection of changes in serum protein levels associated with clinical outcome (21). Table 1 Overview of biomarker results from preoperative systemic treatment: association with clinical endpoints* Clinical endpoint  Primary chemotherapy  Primary endocrine therapy  Tumor shrinkage  High pretreatment PI  Changes in PI    Absence of ER and/or PgR  ER and/or PgR positivity    High histological grade  EGFR expression    Increased apoptotic index  HER2 overexpression    wt-p53      mdr-1, P-gp changes    DFS and/or OS  Low-posttreatment PI  Changes in PI    Low histological grade      wt-p53      Low P-gp      Low HER2 expression    Clinical endpoint  Primary chemotherapy  Primary endocrine therapy  Tumor shrinkage  High pretreatment PI  Changes in PI    Absence of ER and/or PgR  ER and/or PgR positivity    High histological grade  EGFR expression    Increased apoptotic index  HER2 overexpression    wt-p53      mdr-1, P-gp changes    DFS and/or OS  Low-posttreatment PI  Changes in PI    Low histological grade      wt-p53      Low P-gp      Low HER2 expression    * References (9–16). DFS = disease-free survival; EGFR = epidermal growth factor receptor; ER = estrogen receptor; OS = overall survival; PgR = progestrone receptor; PI = proliferation indices; P-gp = P-glycoprotein. View Large Overall, none of the reported gene signatures is robust enough to be clinically useful mainly because of the lack of appropriate validation, and translational studies often suffer from small case series and statistical faults. Such studies, however, represent a proof-of-principle that gene and protein expression profiling may predict response to preoperative systemic treatments. Novel Approaches to Turn Preclinical Information into Clinical Benefit Genomic Classifiers Derived From Cell Lines and Small Interfering RNA Screening Modern approaches to discovering treatment response predictors rely on mechanism-driven (or biology-driven) hypotheses (Figure 2). At present, such approaches resort to the use of 1) cell lines and/or xenograft models to define gene expression signatures associated with drug response and 2) functional genomics to predict chemosensitivity. The former approach has been used to better understand the temporal response to tamoxifen by identifying treatment-induced changes in gene expression in an ER-positive breast cancer xenograft model (22), and the resulting early/transient proliferation and continuous/late/estrogen response genes proved to be associated with clinical outcome in 404 women entered four tamoxifen-treated subsets. Figure 2 View largeDownload slide Functional genomics to accelerate the discovery of drug-specific response predictors. Figure 2 View largeDownload slide Functional genomics to accelerate the discovery of drug-specific response predictors. The functional genomic approach is based on silencing target genes in cultured cells to systematically study the ability of individual genes to alter tumor chemosensitivity. The combining of experimental findings with expression profile data from human tumors allows to identify clinical relevant regulators of drug sensitivity. Using such a preclinical approach, Juul et al. (23) identified a paclitaxel response metagene subsequently validated as a pathologic complete response predictor in a series of triple-negative tumors. Tumor-Initiating Cells (TICs) and Drug Resistance The concept that cancer may be a stem cell disease, arising from tissue stem/progenitor cells or driven by cancer cells with stem cell properties, has important implications for the development of new therapeutic strategies able to target the limited fraction of TICs, which share chemoresistance properties with normal stem cells. Present findings support the hypothesis that the residual posttreatment tumors are enriched of TICs because 1) classical chemotherapy appears to increase the proportion of TICs (CD44+/CD24−//low, ALDH1+) (24); 2) epithelial–mesenchymal transition markers are frequently overexpressed (25); 3) residual cancer cells exhibit a gene expression signature peculiar of breast cancer–initiating cells (26). The therapeutic implications of this model system are valuable because TICs might represent new targets, with distinct possible strategies, such as interference with self-renewal and TIC phenotype reversal through epigenetic modulation/reprogramming. Collection of Human Biological Material in Clinical Trials: Opportunities and Obstacles Access to human biological material for research is considered a major bottleneck hindering successful translation of information from bench to bedside. The potential of this tool is provided by the enormous amount of information contained in biological samples, which, along with the continuing evolution of technologies, represent a great opportunity for new treatment strategies. Biobanks are designed to maximize this research potential, with foreseeable advantages for the community and for the industry, which may be able to transform such knowledge into products, whether diagnostics or therapeutics, for health systems. Due to the complex, multidisciplinary nature of biomaterial collection and translational research, integration of biological material collection into clinical trials warrants careful up-front planning and input from a range of expertise, which take into consideration several aspects also including logistics and standard operating procedures for biological material collection, and test to control for some sources of preanalytical variability, such as those due to timing and tissue handling. In the field of biobanking research, many ethical and legal issues are now open, particularly those affecting relationships with donors, researchers, and sponsors. Today, the main problems are dealing with 1) meaning and implications of the concept of ownership of biological material and related information, and of their “donation” for research; 2) extent of informed consent; 3) withdrawal of consent and its consequences, protection of personal data, availability of information useful for the patients and for their family members; 4) guarantees for the community (ie, availability of research results and technologies derived) and individuals (confidentiality of data). The guarantees should affect: Function of the biobank and its procedures (public interest, independence, transparency, confidentiality and professional secrecy, explicit agreements of transfer, and standards of conservation). Decision-making processes and players involved (institutional review boards, ethics committees, boards of trustees including representatives of associations of patients and citizens). Information and communication to all stakeholders (citizens and their associations, health institutions, public and private research institutes, industry, and political institutes). Rights of donors (personal information, use of the samples, and consent withdrawal). Conclusions The most important message derived from decades of translational research in breast cancer relies on focusing on clinically relevant needs, with a strong, intense, and continuous collaboration among preclinical and clinical investigators. Such an effort should be pursued at each step of the round trip from the bench to the bedside, and back. Instrumental to such a translational model are: 1) availability of well-annotated and high-quality biological specimens from patients entered in clinical trials; 2) standardized, reproducible, and robust assays to detect molecular markers/signatures even on limited tumor material and/or on formalin-fixed paraffin-embedded material; 3) quality assessment and assurance of reagents and test systems, including preanalytical processing, commercial kits, and equipment; 4) prospective planning of the study design, even using a prospective–retrospective approach (5); and 5) regulatory issues adequately fitting the need of translational studies. Moreover, critical issues that cannot be considered independently are analytical performance and clinical utility because the clinical value of a test cannot disregard assay reliability, and assay performance cannot overlook its clinical use (3). The neoadjuvant setting represents an ideal model to directly challenge predictive value and clinical utility of molecular markers/signatures identified on specimens of patients entered clinical trials, as well as to validate genomic classifiers derived from preclinical experimental models and functional genomic studies. Funding This work was supported by grants from the Associazione Italiana per la Ricerca sul Cancro (AIRC, the Italian Association for Cancer Research, grant number 4532 to M.G.D.) and by the Italian Ministry of Health (Special Project on Female Cancers). This work was in part supported also by funds obtained through an Italian law that allows taxpayers to allocate the “5 × 1000” share of their payments to support a research institution of their choice. References 1. Morrow PK,  Hortobagyi GN.  Management of breast cancer in the genome era,  Annu Rev Med. ,  2009, vol.  60  2(pg.  153- 165) 2. Simon R.  Lost in translation: problems and pitfalls in translating laboratory observations to clinical utility,  Eur J Cancer. ,  2008, vol.  44  18(pg.  2707- 2713) 3. Taube SE,  Clark GM,  Dancey JE,  McShane LM,  Sigman CC,  Gutman SI.  A perspective on challenges and issues in biomarker development and drug and biomarker codevelopment,  J Natl Cancer Inst. ,  2009, vol.  101  21(pg.  1453- 1463) 4. Taube SE.  Biomarkers in oncology: trials and tribulations,  Ann N Y Acad Sci. ,  2009, vol.  1180 (pg.  111- 118) 5. Harris L,  Fritsche H,  Mennel R, et al.  American Society of Clinical Oncology 2007 update of recommendations for the use of tumor markers in breast cancer,  J Clin Oncol. ,  2007, vol.  25  33(pg.  5287- 5312) 6. Simon RM,  Paik S,  Hayes DF.  Use of archived specimens in evaluation of prognostic and predictive biomarkers,  J Natl Cancer Inst. ,  2009, vol.  101  21(pg.  1446- 1452) 7. Burstein HJ.  Preoperative therapy as a model for translational research in breast cancer,  Cancer Invest. ,  2008, vol.  26  3(pg.  217- 221) 8. Dowsett M,  Goldhirsch A,  Hayes DF,  Senn H-J,  Wood W,  Viale G.  International web-based consultation on priorities for translational breast cancer research,  Breast Cancer Res. ,  2007, vol.  9  6pg.  R81  9. Wolff AC,  Berry D,  Carey LA, et al.  Research issues affecting preoperative systemic therapy for operable breast cancer,  J Clin Oncol. ,  2008, vol.  26  5(pg.  806- 813) 10. Daidone MG,  Benini E,  Tomasic G, et al.  Biological markers as indicators of response to primary and adjuvant chemotherapy in breast cancer,  Int J Cancer. ,  1999, vol.  84  6(pg.  580- 586) 11. Ellis MJ,  Coop A,  Singh B, et al.  Letrozole is more effective neoadjuvant endocrine therapy than Tamoxifen for ErbB-1 and/or ErbB-2-positive, estrogen receptor-positive primary breast cancer: evidence for a phase III randomized trial,  J Clin Oncol. ,  2001, vol.  19  18(pg.  3808- 3816) 12. Archer CD,  Parton M,  Smith IE, et al.  Early changes in apoptosis and proliferation following primary chemotherapy for breast cancer,  Br J Cancer. ,  2003, vol.  89  6(pg.  1035- 1041) 13. Dowsett M,  Ebbs SR,  Dixon JM, et al.  Biomarker changes following neoadjuvant anastrozole, tamoxifen, or the combination: influence of hormonal status and HER-2 in breast cancer—a study from the IMPACT trialists,  J Clin Oncol. ,  2005, vol.  23  11(pg.  2477- 2492) 14. Caudle AS,  Gonzales-Angulo AM,  Hunt KK, et al.  Predictors of tumor progression during neoadjuvant chemotherapy in breast cancer,  J Clin Oncol. ,  2010, vol.  28  11(pg.  1821- 1828) 15. Mohsin SK,  Weiss HL,  Gutierrez MC, et al.  Neoadjuvant trastuzumab induces apoptosis in primary breast cancer,  J Clin Oncol. ,  2005, vol.  23  11(pg.  2460- 2468) 16. Daidone MG,  Veneroni S,  Benini E, et al.  Howell A,  Dowsett M.  Biological markers and changes induced in their profiles following primary chemotherapy: relevance for short- and long-term clinical outcome,  ESOI Scientific Updates, Vol. 4. Philadelphia, PA: Elsevier ,  1999(pg.  53- 72) 17. Cappelletti V,  Gariboldi M,  De Cecco L, et al.  Patterns and changes in gene expression following neo-adjuvant anti-estrogen treatment in estrogen receptor-positive breast cancer,  Endocr Relat Cancer. ,  2008, vol.  15  2(pg.  439- 449) 18. Chang JC,  Wooten EC,  Tsimelzon A, et al.  Gene expression profiling for the prediction of therapeutic response to docetaxel in patients with breast cancer,  Lancet. ,  2003, vol.  362  9381(pg.  362- 369) 19. Gianni L,  Zambetti M,  Clark K, et al.  Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer,  J Clin Oncol. ,  2005, vol.  23  29(pg.  7265- 7277) 20. Tabchy A,  Valero V,  Vidaurre T, et al.  Evaluation of a 30-gene paclitaxel, fluorouracil, doxorubicin and cyclophosphamide chemotherapy response predictor in a multicenter randomized trial in breast cancer,  Clin Cancer Res. ,  2010, vol.  16  21 21. Mazouni C,  Baggerly K,  Hawke D, et al.  Evaluation of changes in serum protein profiles during neoadjuvant chemotherapy in HER2-positive breast cancer using LC-MALDI-TOF/MS procedure,  Proteomics. ,  2010, vol.  10  19(pg.  3525- 3532) 22. Taylor KJ,  Sims AH,  Liang L, et al.  Dynamic changes in gene expression in vivo predict prognosis of tamoxifen-treated patients with breast cancer,  Breast Cancer Res. ,  2010, vol.  12  3pg.  R39  23. Juul N,  Szallasi Z,  Eklund AC, et al.  Assessment of an RNA interference screen-derived mitotic and ceramide pathway metagene as a predictor of response to neoadjuvant paclitaxel for primary triple-negative breast cancer: a retrospective analysis of five clinical trials,  Lancet Oncol. ,  2010, vol.  11  4(pg.  358- 365) 24. Li X,  Lewis MT.  Huang J, et al.  Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy,  J Natl Cancer Inst. ,  2008, vol.  100  9(pg.  672- 679) 25. Yu F,  Yao H,  Zhu P, et al.  let-7 regulates self renewal and tumorigenicity of breast cancer cells,  Cell. ,  2007, vol.  131  6(pg.  1109- 1123) 26. Creighton CJ,  Li X,  Landis M, et al.  Residual breast cancer after conventional therapy display mesenchymal as well as tumor-initiating features,  Proc Natl Acad Sci U S A. ,  2009, vol.  106  33(pg.  13820- 13825) Published by Oxford University Press 2011.

Journal

JNCI Monographs – Oxford University Press

Published: Oct 1, 2011

There are no references for this article.