Neoadjuvant trials provide endpoints, such as pathological complete response (pCR) to treatment, that will potentially translate into meaningful improvements in overall survival and disease-free survival. Neoadjuvant trials need smaller sample sizes and are less expensive, and the endpoint of pCR is achieved in months, rather than years. For these reasons, the neoadjuvant setting is ideal for testing emerging targeted therapies in early breast cancer. Recently the US Food and Drug Administration has released a draft Guidance to Industry, outlining a pathway to accelerated approval for neoadjuvant breast cancer therapies using pCR. The association between pCR and outcome is clear for chemotherapy in triple-negative breast cancer and for HER2-targeted agents in HER2-positive disease, but might not hold true for other tumor subtypes such as luminal cancers. Since pCR is rarely achieved with either chemotherapy or endocrine therapy in hormone-receptor-positive breast cancer, in this setting we need to identify different intermediate endpoints, which might be translational endpoints within “window-of-opportunity,” “residual disease,” and “genome forward” trials. Prospective validation of effective noninvasive techniques for monitoring of residual disease burden could enhance the ability to identify promising targeted therapies in the neoadjuvant setting.
Breast cancer subtypes are characterized by specific molecular events. Genome-wide sequencing studies will increase our understanding of cancer biology and will identify two specific types of molecular events: the “drivers,” providing a survival- and proliferation-selective advantage, and the “passengers,” neutral to the selection process. It will be essential in the future to identify all molecular pathways that emphasize the heterogeneity and complexity of human breast cancer to explain mechanisms sustaining proliferation hallmarks of cancer and “drive” tumor progression and resistance to therapy. Disease segmentation in subtypes can offer insights to personalize treatment and the neoadjuvant setting is ideal to explore “hypothesis driven” studies. The main challenge of personalized therapy is to distinguish which patients will benefit most from a particular targeted agent from those who will not. The neoadjuvant model provides a powerful tool to better understand breast cancer biology, to detect and validate surrogate biomarkers, and to accelerate evaluation of new active drugs. We analyze the reasons for which the neoadjuvant setting should be used as a model for testing emerging targeted therapies in breast cancer.
Rationale for Drug Development in the Neoadjuvant Setting
Drug development through traditional phases in advanced-disease setting is a long and expensive process. Recently the US Food and Drug Administration (FDA) outlined a new pathway for accelerated drug approval through neoadjuvant trials (1). This pathway needs to reconsider some of the current approaches to clinical trial design. The neoadjuvant trial model provides a great opportunity to test the efficacy of new targeted agents in patients with previously untreated disease and allows to detect the subgroups of patients and tumors in which they are most effective. In the neoadjuvant setting the tumor remains in its microenvironment and is accessible for pre- and postexperimental treatment biological assessment through tissue biopsy, offering a real-time examination of tumor response and pharmacodynamics of experimental agents. The pretreatment biopsy will allow the potential identification of predictive biomarkers of response. A seminal study by Ellis et al. (2) assessed the great opportunity offered by the neoadjuvant model. To correlate the variable clinical features of estrogen-receptor-positive breast cancer with somatic alterations, they studied pretreatment tumor biopsies accrued from patients treated with aromatase inhibitor in two neoadjuvant studies by massively parallel sequencing and analysis. Eighteen significantly mutated genes were identified, including five genes (RUNX1, CBFB, MYH9, MLL3, and SF3B1) previously linked to hematopoietic disorders. Mutant MAP3K1 was associated with luminal A status, low-grade histology, and low proliferation rates, whereas mutant TP53 was associated with the opposite pattern. Moreover, mutant GATA3 correlated with suppression of proliferation upon aromatase inhibitor treatment (2). In another study, investigators performed messenger RNA gene expression analysis from pretreatment core biopsies of patients treated with ixabepilone (3). Gene expression data were available for 134 patients. Estrogen receptor (ER) gene expression (ESR1) was inversely related to pathological complete response (pCR) in breast and had a positive predictive value of 37% and negative predictive value of 92%. A 10-gene penalized logistic regression model developed from 200 genes predictive of ixabepilone sensitivity in preclinical experiments included ER and tau and had higher positive predictive value (45%) and comparable negative predictive value (89%) to ER1 (3).The tissue biopsies obtained during the neoadjuvant treatment can be useful for correlating changes of biomarkers with potential outcome. The presence of high Ki67 expression after 2 weeks of neoadjuvant endocrine therapy for estrogen-receptor-positive breast cancer was statistically associated with lower recurrence-free survival (P = .004), whereas higher Ki67 expression at baseline was not (4). The neoadjuvant setting offers also the chance to consider the pCR as a surrogate endpoint of outcome in triple-negative and HER2-positive breast cancers (5,6) and provides a strong rationale for testing new biological drugs. Based on the assumption that pCR can be used as an endpoint for a fast assessment of drug efficacy, thus allowing to save both time and money compared with conventional adjuvant trials, the FDA in 2012 released a draft guidance declaring that the agency would evaluate granting accelerated approval on the basis of pCR as surrogate endpoint of clinical benefit in neoadjuvant setting. FDA requires, after the accelerated approval, the demonstration of an improvement in disease-free or overall survival, pending the withdrawal of the indication from product labeling if confirmatory trials will not show a clinical benefit (7). Recently, the FDA approved the use of pertuzumab for neoadjuvant treatment of patients with HER2+ breast cancer (8) and this is the first example of FDA-approved drug for neoadjuvant treatment based on pCR as primary outcome (9). Therefore, the neoadjuvant setting could be used as a rapid tool to test in vivo a biological hypothesis and to demonstrate the potential activity of a specific drug.
The introduction of targeted agents in the treatment of breast cancer has resulted in new challenges for assessing response to therapy, in fact conventional response criteria are limited and may lead to misleading conclusions about a drug’s benefit. On the other hand, assessing biological endpoints can be limited by difficulty in obtaining tumor tissue before and after drug administration and by exposure to previous anticancer therapy (Figure 1). The window-of-opportunity (presurgical or phase 0) trials represent a tool to bypass these issues. In this model, a study drug is administered to women with newly diagnosed breast cancer in the time between the diagnostic breast biopsy and scheduled surgical resection. The phase 0 trials provide investigators with the opportunity to obtain tumor samples at two distinct time points and assess potential biomarkers over a short period of time. Therefore, the presurgical trials allow to evaluate quickly the pharmacokinetic and target modulation of a novel agent and provide the early demonstration of potential clinical activity of a new drug. The validation of the proliferation biomarker, Ki67, as an intermediate endpoint (4) allows using it as an appropriate surrogate marker of outcome instead of clinical response. In patients randomized in the Immediate Preoperative Anastrozole, Tamoxifen, or Combined with Tamoxifen (IMPACT) trial (10), a decrease in proliferation activity, as assessed by Ki67 expression, was found to happen after only 2 weeks of therapy and was an independent predictive factor for relapse-free survival (11). A higher inhibition of Ki67 was detected in the anastrozole arm as compared with tamoxifen or to the combination, anticipating the results of the large adjuvant Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial (12). In the study by Guix et al. (13), women with operable breast cancer were treated with the epidermal growth factor receptor tyrosine kinase inhibitor erlotinib and the results indicated that short-term treatment with erlotinib inhibited tumor cell proliferation and inhibition of proliferation occurred in ER-positive breast cancer outside of the epidermal growth factor receptor-overexpression group. Compared with pCR, changes in tumor biology were found to occur already a few days after treatment start (4). This is an important advantage because it may lead to early adjustments of individualized treatment. For this reason, biomarkers of proliferation are widely used as endpoint of outcome also in the ongoing phase 0 trials testing novel targeted agents, such as the MONALEESA-1 trial, a randomized presurgical study that will assess the biological activity of LEE011 (a CDK 4–6 inhibitor) plus letrozole versus single-agent letrozole in primary breast cancer (14). However, Ki67 has proven to be an appropriate surrogate marker for outcome in preoperative studies administering endocrine therapies (4,15,16) but may not be the most suitable endpoint for predicting outcome for other targeted therapies do not having a direct antiproliferative effects such as antiangiogenic agents. The presurgical approach with targeted agents has also other challenges, such as the optimal duration of the treatment window to detect measurable variations in the surrogate endpoints without delaying standard clinical management, the intrinsic variability, and heterogeneity of tumors—that may be underestimated in a limited tissue sample—and the safety, given that most patients are potentially curable by surgery alone.
Residual Disease Trials
Not all patients treated with neoadjuvant chemotherapy do achieve pCR (17). In this population the study of residual disease and the evaluation of biomarkers modifications induced by the treatments is becoming of great interest as it allows identifying potential markers of treatment resistance and offers new development fields of molecular target agents. Some trials are ongoing in the residual disease setting for patients not achieving pCR after neoadjuvant treatment (Figure 2). Based on data of EMILIA study, which demonstrated that trastuzumab emtansine (TDM-1) improved survival of patients with HER2-positive metastatic breast cancer (18), a global trial (KATHERINE NCT 01772474) is evaluating TDM-1 as an alternative to continuation of trastuzumab in women with residual disease following neoadjuvant chemotherapy (19). Based on the results of a randomized phase II study that demonstrated an improved progression-free survival in patients with hormone-receptor-positive metastatic breast cancer (20), palbociclib (PD-0332991), a cyclin-D kinase 4/6 inhibitor, is being tested in addition to endocrine treatment in patients with a high clinical-pathological stage (CPS) + estrogen receptor status (E), grade (G), staging (CPS-EG score) (21) and no pCR (PENELOPE study) (22). The Hoosier Oncology Group is assessing the benefıt of cisplatin with or without a PARP inhibitor, rucaparib (Clovis), in women with triple-negative or ER-positive/BRCA mutant breast cancer who have residual disease in the breast after neoadjuvant chemotherapy (NCT01074970) (23). If successful, these trials would institute a new approach for drug development. Molecular and genetic profiling of residual disease after neoadjuvant therapy is an exciting field of study because it allows to address new agents against potential targets of resistant disease with the opportunity to improve outcome for patients with poor prognosis after standard therapy.
Genome Forward Trials
The development of high-throughput sequencing technologies improved knowledge of genetic alterations that underlie breast cancer and induced advances in the field of therapeutic researches. The detection of new predictive biomarkers together with the implementation in drug development is generating a new era of clinical studies in which the patients’ enrollment is stratified based on biomarkers that could potentially predict and augment the response to targeted therapy. In the genome-driven or “genome forward” neoadjuvant trials the activity and efficacy of new drugs are prospectively tested in a population defined on the basis of specific biomarkers that might be useful to target identification and to understand the resistance mechanisms (Figure 3). An interesting feature of these trials is an early response evaluation in which the biopsies and imaging assessments provide the opportunity to understand biology of response and resistance to new targeted agents and, accordingly, to continue the treatment only in the subgroups of patients in which a tumor growth arrest is obtained. An example of biomarker-stratified trial is the ongoing I-SPY 2 neoadjuvant trial, a randomized phase II trial designed to assess the incremental benefit of new targeted agents added to conventional chemotherapy (24).
In the screening phase of I-SPY2, potential candidates undergo core breast tumor biopsies for screening which includes MammaPrint along with ER, PgR, and HER2 assessment. Patients who are ER+/MammaPrint high, ER−, or HER2+ are eligible for the study. Eligible patients are then randomized to either standard neoadjuvant chemotherapy with weekly paclitaxel (plus trastuzumab for HER2+) or paclitaxel combined with one of several investigational agents followed by four cycles of doxorubicin/cyclophosphamide. Biomarkers are used to identify signatures for experimental regimens. The primary endpoint is pCR at the time of surgery. Patients on treatment undergo serial tissue collection and imaging. Regimens are dropped if they do not improve pCR rates for any biomarker signature. The adaptive design of this trial, selecting the subgroups of patients more responsive, allows accelerating the investigational agents’ development into subsequent phase III trial, to increase the efficiency and to limit the exposure to potentially harmful drugs. In another ongoing trial, patients with clinical stage II or III luminal-type breast cancer (ER+ HER2− breast cancer) are biopsied at baseline for PIK3CA gene sequencing. They receive an estrogen deprivation therapy (aromatase inhibitor plus an LHRH-A if premenopausal). After 28 days, the patients with PIK3CA mutant tumors will be eligible to receive an AKT inhibitor (25) and those with PIK3CA wild-type tumors will be eligible to receive a CDK4/6 inhibitor trial (26). Repeated tumor biopsies will be taken before the start of either MK-2206 or PD-0332991 on cycle 1 day 1 and after 2 weeks of combination therapy for the evaluation of early response biomarkers (Ki67). Those with resistant tumors (Ki67 >10%) will go off therapy early to avoid futile treatment. Those with Ki67-responsive tumors (Ki67 <10%) will receive a total of 4 months of study drug therapy followed by definitive breast surgery to evaluate pathological response. The genome forward trials are the challenge of the future and could provide the answers to the request of patients to receive personalized treatments.
The “Oncogene Revolution” has led to an explosion of molecularly targeted therapeutics in preclinical and clinical development over the last decade. It is estimated that there are more than 800 targeted anticancer therapies currently in various stages of clinical development. Disappointingly, historical data indicate that only 5% of these investigational agents will ultimately progress to registration for widespread use. These high attrition rates have multiple causes, including lack of efficacy and excessive toxicity. In particular, when patients are selected for phase III trials based on histopathology alone, a targeted drug with a 5%–10% single-agent response rate runs a high risk of failure. Recent efforts to systematically sequence cancer genomes have revealed that individual tumors frequently harbor multiple “driver” somatic mutations that confer growth advantage and positive selection. The increasing identification of specific somatic mutations and other genetic aberrations that drive cancers leaves us on the threshold of a new era of “personalized cancer medicine,” in which specific biomarkers will be used to direct targeted agents only to those patients deemed most likely to respond. The potential medical, scientific, and economic benefits of such a personalized approach to cancer therapy are immense and self-evident. The neoadjuvant setting offers a unique opportunity to accelerate this process. Testing new agents in potentially curable patients requires a series of considerations. Firstly, there is the need to carefully select the patients at higher risk who would not benefit from standard therapy alone and for whom is ethically rational an investigational therapy. Secondly, there is the need to find the appropriate surrogate markers of outcome for every novel agent and for every subgroup of breast cancer. For example, we know that pCR is a good surrogate of outcome in ER−\HER2+ and ER−\HER2− breast cancers but it is not helpful for patients with ER+ disease (27). It is not feasible a correct assessment of an increased benefit of a new agent in absence of a valid response marker. Finally, breast cancer is a multiclonal disease and a single biopsy might underestimate the molecular aberrations driving the cancer in each individual patient. For this reason, it is important to integrate the detection of mutations with other analytic approaches such as the “liquid biopsies.” Correlation of qualitative and quantitative analysis of circulating DNA with pathological response to neoadjuvant therapy might anticipate potential resistance to biological agents and allow to early switch to non-cross-resistant therapies (28). Processes to identify which drugs and methods are appropriate to evaluate toxicity in real time are the cardinal point to do not lose the right way in the neoadjuvant model. Proof-of-concept clinical trials that establish the value of matching targeted treatments to rare molecular alterations in breast cancer and other malignancies is beyond the scope of any single pharmaceutical sponsor, cancer treatment facility, or national cancer agency and will ultimately require international collaboration (29).
- polymerase chain reaction
- chemotherapy regimen
- genes, erbb-2
- neoadjuvant therapy
- receptor, erbb-2
- united states food and drug administration
- breast cancer
- endocrine therapy
- residual tumor
- surrogate endpoints
- complete remission
- triple-negative breast cancer
- her2 positive
- molecular targeted therapy