Window of opportunity studies: Do they fulfil our expectations?

Cancer Treatment Reviews 43 (2016) 50–57

Contents lists available at ScienceDirect

Cancer Treatment Reviews journal homepage: www.elsevierhealth.com/journals/ctrv

Hot Topic

Window of opportunity studies: Do they fulfil our expectations? Sandra Schmitz 1, François Duhoux 1, Jean-Pascal Machiels ⇑ Institut Roi Albert II, Department of Medical Oncology, Cliniques Universitaires Saint-Luc and Institut de Recherche Clinique et Expérimentale (Pole MIRO), Université catholique de Louvain, Brussels, Belgium

a r t i c l e

i n f o

Article history: Received 4 November 2015 Received in revised form 21 December 2015 Accepted 22 December 2015

Keywords: Window of opportunity study Translational research Biomarkers Safety Breast cancer Head and neck cancer

a b s t r a c t Window of opportunity studies are trials in which patients receive one or more new compounds between their cancer diagnosis and standard treatment (mainly surgery). Patients are generally cancer treatment naïve. Tumor biopsies before and after the investigational treatment are collected for translational research. Similarly, anatomic and functional pre- and post-treatment imaging may be incorporated. Ideally, the investigational treatment is kept short to avoid delaying standard treatment. Window of opportunity trials may expedite drug development, improve our understanding of pharmacodynamic parameters, and help to identify biomarkers for better patient selection. They can, however, have major drawbacks including potential safety and logistical issues, delayed standard treatment, and a probable lack of patient benefit. By focusing on breast and head and neck cancers, in this paper we discuss the advantages, disadvantages and design of window of opportunity studies. Ó 2015 Elsevier Ltd. All rights reserved.

Novel cancer treatments are often investigated in unselected end-stage cancer patients [1,2]. The choice of this patient population has, however, important limitations that may impair drug development. First, through previous exposure to anti-cancer treatments, most end-stage patients will have developed multifactorial treatment resistance mechanisms. This may blind the activity of potentially active new agents and prematurely cease their development. Second, the feasibility of conducting translational research to investigate predictive biomarkers and pharmacodynamics is hampered by the ethics of obtaining iterative tumor biopsies in palliative patients. Innovative trial designs that can identify promising new compounds and predictive biomarkers are therefore needed, particularly for targeted agents [1]. The evaluation of compounds in untreated patients prior to standard treatment may resolve some of these issues. Window of opportunity studies are trials in which treatment naïve patients consent to receive one or more new compounds, or a new treatment strategy, in the period between their cancer diagnosis and the delivery of their standard treatment [3,4]. Standard treatment is usually surgery with curative intent (enabling the collection of a substantial amount of treated tumor tissues), but both chemotherapy or radiation-based therapy are plausible. Tumor biopsies are

⇑ Corresponding author at: Department of Medical Oncology, Cliniques universitaires Saint-Luc, Avenue Hippocrate 10, 1200 Brussels, Belgium. Tel.: +32 (0) 27645457; fax: +32 (0)27645428. E-mail address: [email protected] (J.-P. Machiels). 1 Equal contributors. http://dx.doi.org/10.1016/j.ctrv.2015.12.005 0305-7372/Ó 2015 Elsevier Ltd. All rights reserved.

collected before and after the investigational treatment for translational research. Similarly, anatomic and functional pre- and posttreatment imaging can be incorporated. Excluded from the definition of window of opportunity studies are neoadjuvant treatments or trials in which standard treatment (i.e. chemotherapy and/or radiotherapy) is given with or without an investigational agent with the aim of improving disease outcome [5]. In a neoadjuvant approach, definitive standard treatment (i.e. surgery or (chemo)radiation) is delayed to give the investigational agent time to produce a therapeutic response and improve the overall treatment efficacy. Marous and colleagues have recently reviewed the designs of preoperative biomarker trials in oncology [6]. They identified 56 trials. The tumor types evaluated included breast cancer (59%), prostate cancer (11%), gastric cancer (5%), non-small cell lung cancer (5%), head and neck cancer (5%), ovarian cancer (4%), pancreatic cancer (4%), gastro-intestinal stromal tumor (2%), and endometrial cancer (2%). In this review, we discuss the design as well as the advantages and disadvantages of window of opportunity studies in cancer. We illustrate the challenges associated with window studies using examples from breast and head and neck cancer trials. Window of opportunity studies: design considerations Fig. 1 depicts the general design of a window of opportunity study. Management issues associated with the trial design should be taken into consideration and include:

51

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

a) Duration of treatment

c) Trial endpoint(s)

Treatment with the investigational agent must be short, generally a few days or weeks, to avoid delaying curative treatment. Given the risk of disease progression and the potential ineligibility for curative therapy, this point is particularly important. It is even more important if the activity of the compound under investigation is unknown and if there are no postulated or recognized predictive biomarkers. The treatment duration should also take into account the pharmacokinetic and mechanism(s) of action of the investigated compound. Regarding pharmacokinetics, the drug should be administered enough time to reach steady state to generate meaningful results. Such a period of time might be an issue with drug with a prolonged half-life. Modulation of phosphoproteins is generally achieved shortly after drug administration if the drug blood levels are appropriate. However, to evaluate gene expression profile, protein expression or immune response, this could require a longer period of time.

A primary endpoint with a statistical hypothesis to determine the sample size is mandatory. In the Marous’ review, the primary endpoint was mentioned in 80% of the study and included pharmacodynamic endpoint in 58%, efficacy in 31%, and safety in 11%. The most frequent pharmacodynamics endpoint was Ki67 (73%). However, information regarding the sample size calculation was present in only 62% of the trials reported [6]. Ideally, although not always feasible, the primary endpoint should be a molecular or a functional imaging parameter that has been validated as a surrogate marker of treatment activity that impacts outcome, such as progression-free survival (PFS) or overall survival (OS). Binary endpoints with a particular cut-off to define responders and non-responders are adequate if the assay used is standardized according to international guidelines and if a particular cut-off has been linked with clinical outcome. However, in case of exploratory analyses, continuous endpoints can also be appropriate. Ki67 is the most commonly used biomarker of treatment activity in window of opportunity studies. In breast cancer, a change in Ki67 is a validated endpoint linked to treatment efficacy and long-term prognosis [9–11]. However, variability in Ki67 measurement exists and requires standardization, and Ki67 might not be an adequate surrogate marker for all compounds and cancers [12]. Other molecular endpoints, such as a decrease in phosphorylation of the targeted kinase receptor or modulation of some cell cycle regulators, may be good targets for the investigated compound but they also require standardization, validation and central analysis [13–15]. The drawback with proliferative or molecular biomarkers is tumor heterogeneity [12]. A pharmacokinetic sample should be drawn at the time of analysis of the primary endpoint to correlate the biomarker modulation with the drug levels achieved in the blood. When the primary endpoint is based on a comparison between pre-and post-treatment biopsies, the paired biopsies should be performed under the same conditions and following the same procedures in order to limit the impact of tumor heterogeneity as well as the modifications induced by the procedure itself. In this context, it is probably more appropriate to compare paired biopsies instead of comparing a pre-treatment biopsy with a surgical specimen. Pathological response, with the quantification of viable residual tumor cells in the surgical specimen, might be linked with longterm outcome and may also offer a valid endpoint [16]. Although

b) Timing The best time to implement a window of opportunity study is during the ‘preparation’ time between diagnosis and standard treatment. This period is generally used for staging procedures, pre-operative exams, operating theatre reservation, and/or radiation therapy planning. Ideally, patient consent for the study should be obtained early on in the process, sometimes even before the tumor is biopsied, to avoid having to repeat invasive procedures. The trial strategy itself therefore carries a risk of patient loss during the screening process. With head and neck cancer studies, our practice is to therefore discuss window of opportunity clinical trials with the patient at the time of clinical diagnosis [7]. This allows us to prospectively combine standard staging investigations with those required by the window study and avoids the need for certain procedures to be repeated. The acceptable delay between diagnosis and standard treatment is not well defined in the literature, but we consider that definitive treatment should be initiated within four weeks of diagnosis [3,8].

Imaging (i.e. FDGPET/CT, DCE and DW MRI); tumor biopsy, and blood and plasma collecon Within 2 weeks prior to randomizaon

Registraon and screening

Post-treatment imaging, blood and plasma collecon

R A N D O M I Z A T I O N

Arm 1 Study drug for 2 weeks

Day before surgery

Arm 2 No treatment for 2 weeks

Study endpoints:

S U R G E R Y

Tumor biopsy, blood and plasma collecon

FDG-PET:2-[fluorine-18]-fluoro-2-deoxy-D-glucose positron emission tomography ; MRI: magnec resonance imaging; DCE: dynamic contrast-enhanced; DW: diffusion weighted Fig. 1. Typical design of window of opportunity study.

1) Pathological or molecular response 2) Acvity detected through imaging(funconal or anatomical) 3) Surgical safety and toxicity follow-up (4 weeks post surgery)

52

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

Response Evaluation Criteria In Solid Tumors (RECIST) criteria have been used in some trials and tumor shrinkage has been observed in some situations, anatomic imaging endpoints are difficult to apply due to the short treatment duration [7,17–19]. Functional imaging such as 2-[fluorine-18]-fluoro-2-deoxy-D-glucose positron emission tomography (18FDG-PET) has also been used as a primary endpoint to detect early metabolic modifications [7,13]. Safety endpoints with early stopping rules should be predefined in the protocol with a special focus on surgical safety evaluated by an independent committee [7,13]. Given that patients are generally treated with curative intent, the toxicity profile, maximum tolerated dose and recommended dose of the investigational compound must be known and well-characterized before initiating a window study.

Table 1 Potential advantages and challenges of window of opportunity studies. Potential advantages

Challenges and risks

Evaluation of new compounds in humans versus pre-clinical models (microenvironment etc.) Evaluation of new compounds in treatment naïve patients This study design may speed up the pace of early drug development by understanding the pharmacodynamics of new compounds Identification of biomarkers to allow better patient selection

Short treatment duration to avoid delaying standard treatment

e) Quality control Certification of molecular analyses and quality control, together with a central review of imaging and standard treatment procedures is required, especially where multicenter studies are concerned. Functional imaging measurements are highly dependent on the degree of imaging standardization. Imaging guidelines should be defined upfront to establish standard imaging acquisition. Accreditation (i.e. EARL accreditation for 18FDG-PET), when available, should be used, and a dummy run should be performed before site activation [22]. The proposed standard treatment must also be clearly defined and subject to quality control. f) Collaboration Finally, to be successful, participating centers must have a dedicated collaborative team of oncologists, surgeons, pathologists, and scientific researchers. Patients must also be well informed of the aims of such trials given that they will probably not derive any direct benefit. Potential achievements (Table 1) As already outlined, window of opportunity studies can precipitate the development of new compounds, rapidly identify promising agents, and provide initial pharmacodynamic parameters. Among the preoperative studies reported, pharmacodynamic

Potential impact of the investigational treatment on the feasibility and safety of surgery: wound healing and surgical margins (in case of tumor response) Tumor heterogeneity Trial logistics The need for early validated surrogate markers of treatment activity

d) Control/placebo group The evaluation of the primary endpoint may be impaired by tumor heterogeneity as well as by the inherent technical variations in the imaging or molecular read-out of the parameters used as endpoints. It is therefore crucial to include a control/placebo group, ideally at randomization. This is particularly important when the study has endpoints involving molecular analyses like immunohistochemistry, DNA investigation, gene expression profiling, or functional imaging procedures. If significant modifications are found between the pre- and post-treatment analyses in the control/placebo group, the assays and endpoints will be invalid [20]. Gross and colleagues, for example, investigated gefitinib and sulindac in head and neck cancer. They found some background changes in G protein-coupled receptor-Epidermal Growth Factor Receptor (GPCR–EGFR) signaling intermediates in placebo-treated patients, precluding mechanistic conclusions. In contrast, Ki67 modifications were only found in the investigational arm and not in the placebo arm, supporting drug activity and the validity of the analysis [21]. Similarly, a small decrease in standard uptake value (SUV) max (less than 25%) was found in untreated cancer patients when two 18FDG-PET scans were performed two weeks apart [7].

No therapeutic benefit

parameters could be identified in 48% of the trials [6]. As these trials are performed in a predominantly untreated population, the ability to detect drug activity is maximized and translational research is not impaired or blinded by previous anti-cancer therapies. The collection of biopsies before and after the investigational treatment may confirm that the drug is targeting the intended pathway, help identify subgroups of patients most likely to respond to therapy before further drug development, and identify early mechanisms of drug resistance [23]. In this regard, if standard treatment is surgery, a substantial amount of post-treatment tissue can be obtained and analyzed. Anatomic and functional imaging may also detect early drug activity. This approach is also clinically relevant. Development and preclinical testing of new cancer therapies is limited by the scarcity of in vivo models that authentically reproduce tumor growth and metastatic progression [24,25]. Patient-derived tumor xenografts retain the morphological and molecular markers of the original tumor, even after serial passages across several generations of mice, supporting their use in early drug development. However, even if they can predict the activity of drugs that target tumor cells, the tumor microenvironment is rapidly replaced by mouse tissue in these models [26]. Window of opportunity studies enable new compounds to be tested in a real clinical situation where the tumor microenvironment is adequate. Today, drugs targeting this microenvironment, such as checkpoint inhibitors, are entering the clinic. Window of opportunity trials may prove useful in this setting to investigate novel therapies for which no adequate preclinical models exist.

Challenges and risks (Table 1) Due to the short administration period of the investigational treatment, it is unlikely that participating patients will derive meaningful clinical benefit from window of opportunity studies. In addition, if a control/placebo group is included, some of the patients will be exposed to the risks of additional invasive (biopsy) and non-invasive (imaging) procedures. This raises obvious ethical concerns that must be clearly explained to the patients in a wellwritten comprehensive and informative informed consent. The number of patients to be included in the control/placebo group should also be minimized. Investigators should be aware that not all patients are able to understand the concept of challenging studies nor their implications. It is therefore essential that patients be correctly educated in order to give informed consent. Arnaout and colleagues found that 19 out of 20 eligible patients (95%) agreed to optional addi-

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

tional biopsies for a window of opportunity study in breast cancer, but the investigational agent was anastrazole, a drug with wellknown activity [27]. Enthusiasm towards such trials might be different if the activity of the tested compound is unknown. Wisinski and colleagues, for example, conducted a clinical trial to assess the willingness of breast cancer patients to enrol in a window of opportunity research study [28]. Comparatively, only 26.7% of patients agreed to participate. Recruitment to such studies is also limited due to the high level of multidisciplinary coordination and logistics required, coupled with stringent inclusion and exclusion criteria. In the Arnouat trial mentioned above, only 22 patients out of 131 (16.5%) were confirmed eligible [27]. It is therefore not surprising that these trials do not accrue well. Patient safety must be considered. Glimelius and colleagues conducted a meta-analysis of 29 window trials [3]. They found no significant negative effect on overall survival for participating patients, even in the group of patients treated with ineffective or modestly active new molecular agents. Obviously, the safety characteristics of the investigational agent should be known before initiating any window trial. In the Marious report, among the 4208 patients who received a preoperative treatment, there were two deaths (0.05%) related to the study drug [6]. Treatment had to be interrupted because of adverse events in 4%. Three hundred fourteen of the 4690 patients (7%) could not undergo surgery as per protocol, but this was due to an adverse event in only 1%. Ensuring standard practice is also of upmost importance. It is crucial that there are no delays, or only a minimal delay, to standard treatment delivery. Even if major tumor regression is observed, the planned surgical procedure should not be modified so as to ensure adequate surgical margins that are not invaded by microscopic residual tumor. There is some evidence to suggest that the study procedures themselves might have an impact on outcome. In a study of postmenopausal estrogen receptor-positive breast cancer patients treated with anastrozole for 10 days, changes in gene expression were noted in both the treatment and the control group, most likely due to wound-healing processes [20]. Interestingly, the procedures have also been shown to have an impact on Ki67 levels, mainly in HER-2 positive and triple-negative tumors [29]. Finally, in the absence of validated surrogate markers of longterm activity, it is difficult to assess the real impact of a particular compound on cancer outcome. Window trials are therefore only another step in the development of a particular compound. Similarly, with translational research, although we can detect early signs of activity, it is difficult to know how to approach acquired mechanisms of treatment resistance that may take several months to develop. Tumor heterogeneity in this setting is also a limitation.

A successful example: breast cancer In breast cancer, window studies are mainly scheduled between the diagnostic biopsy and surgery of the primary tumor. As previously explained, neoadjuvant trials are not considered window studies because they significantly delay the time to surgery. This review will thus not cover neoadjuvant trials, even though some of them do incorporate early efficacy assessments (e.g. biopsy after two weeks of therapy), which could provide information similar to that derived from window studies [30]. Most window studies evaluate tumor proliferation, measured by Ki67, because a reduction in Ki67 after preoperative antiestrogen therapy is correlated with an improved outcome [31,32]. Window studies in breast cancer have mainly focused on endocrine agents, but other agents such as targeted anti-cancer therapies, statins, metformin and nonsteroidal anti-inflammatory drugs (NSAIDs) have also been tested in this setting (Table 2).

53

Most trials with endocrine agents merely study the effect of a drug on proliferation. Others go beyond this simple ‘‘effect – no effect” approach and take advantage of a window study design to gather hypothesis-generating information. For instance, Robertson and colleagues not only showed that short-term exposure to fulvestrant reduces Ki67, but that the effect was dose-dependent, thus setting the stage for larger trials using increased doses of the drug [33,34]. The group also showed a reduction in the level of estrogen (ER) and progesterone receptors (PgR), confirming the unique mode of action of fulvestrant compared to tamoxifen. Given the rapid development of targeted therapies, window studies provide an excellent framework to rapidly assess the efficacy of a drug and to investigate biomarkers of efficacy or of futility. Erlotinib is an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI). Surprisingly, short-term presurgical treatment with erlotinib led to an anti-proliferative effect that was restricted to estrogen-receptor positive patients, despite this tumor type being less likely to be EGFR-positive [35,36]. Lapatinib is a TKI targeting both the Human Epidermal Receptor-2 (HER2) and EGFR tyrosine kinases. In a window trial, it has not only shown an antiproliferative effect in HER2-positive invasive breast cancer (especially ER negative), but also in adjacent premalignant lesions – the effect of treatment on the latter lesions is obviously very difficult to assess in other types of clinical trials [37]. What is more, the window of opportunity trial design allowed for the identification of a subset of HER2 negative breast cancers that respond to lapatinib, namely tumors with high HER3 expression [9]. Other hypothesis-generating window trials have attempted to understand the roles of statins and metformin on breast cancer. Statins, for example, seem to only reduce the proliferation of tumors that express their target: 3-hydroxy-3-methylglutaryl-coe nzyme A (HMG-CoA) reductase. These drugs seem to act via the inhibition of the mitogen-activated protein kinase (MAPK) pathway [14,38,39]. In some cases, results can be confusing as they can vary across trials, mainly due to differences in sample size and a difference in the patient population. While early trials demonstrated an effect of pre-operative metformin on Ki67, it was later shown that this effect was limited to patients with endocrine resistant and HER2 positive tumors; a trend towards an increase in Ki67 was observed in the remaining patients [40–47]. These results are of course particularly important when it comes to planning large adjuvant trials. Other window studies have concentrated on the changes in selected biomarkers after short-term exposure to a drug, without focusing on proliferation. As seen with NSAIDs, this has the advantage of proving in vivo that the target of the drug is indeed reached before moving towards large and costly clinical trials [48]. In summary, window trials are very established in breast cancer. Thanks to this study design, interesting hypotheses have been generated with endocrine agents, targeted therapies and other less conventional agents. The limitations associated with other trials remain, however, valid in this setting. Small underpowered studies or those that do not take into account different breast cancer subtypes only add to the confusion and do not lead to valid conclusions.

A challenging example: head and neck cancer Window of opportunity studies in head and neck cancer are in early stage development. Compared with breast cancer, there is less standardization. Curative surgery is more challenging, and any delay or complication during surgery has the potential to have a significant impact on patient outcome. Pathological endpoints as surrogate markers of long-term drug activity, such as Ki67, have not been clearly evaluated. In addition, a placebo or a control group

54

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

Table 2 Selected window of opportunity trials in breast cancer. Study

Agent investigated

Population

Primary end point

Finding

Bundred [65]

Exemestane and celecoxib

DCIS

Ki67

Takagi [66]

Letrozole

DCIS

Evaluation of intratumoral estrogens

Lee [67]

4-Hydroxytamoxifen gel versus oral tamoxifen Fulvestrant Fulvestrant versus tamoxifen versus placebo 3 dosages of tamoxifen

DCIS

Ki67

BC BC

Ki67 and other pharmacodynamic parameters Effect on ERalpha, PgR, Ki67 and apoptotic index

BC

Ki67

BC BC

Gene-expression profile Ki67

Guix [36]

Anastrazole Low-dose weekly tamoxifen vs raloxifen vs placebo Erlotinib

Exemestane reduces proliferation; no effet of celecoxib Intratumoral estrogen levels lower than in controls Similar decrease in proliferation with both agents Reduction of ER, PgR and Ki67 Reduction of Ki67 with fulvestrant and tamoxifen Reduction of Ki67 across all tamoxifen dosages Gene-expression changes can be detected No effect of the drugs

BC

Ki67

DeCensi [37]

Lapatinib

Ki67

Leary [9]

Lapatinib

DCIS and BC BC

Bjarnadottir [15,38,39] Hadad [46,47] Niraula, Dowling [43,44] DeCensi, Bonanni, Cazzaniga [41,42,45] Kalinsky [40] Schwab [48]

Atorvastatin

BC

Ki67

Metformin Metformin

BC BC

Ki67 Ki67

Metformin

BC

Ki67

Effect on Ki67 according to host and tumor characteristics

Metformin Etodolac

BC DCIS and BC

Ki67 Gene expression levels of the COX-2 and RXRalpha pathways, cyclin D1 protein level

No reduction in Ki67 Increase in COX-2 pathway gene expression, decrease in cyclin D1 protein level

DeFriend [68] Robertson [34] DeCensi [32,45] Morrogh [20] Serrano [69]

Reduction in Ki67 in ER+ tumors but not in triple negative nor HER2+ tumors Reduction in Ki67 in DIN, DH and invasive HER2+(especially ER-negative) BC Reduction in Ki67 in HER2+ and HER2-/HER3 high tumors Decrease in Ki67 only in HMG-CoA reductase-positive tumors Reduction in Ki67 Reduction in Ki67

Ki67

DCIS: ductal carcinoma in situ; DIN: ductal intraepithelial neoplasia; DH: ductal hyperplasia without atypia; BC: breast cancer; ER: estrogen receptor; PgR: progesterone receptor; HMG-CoA: 3-hydroxy-3-methylglutaryl-coenzyme A; RXR: retinoid X receptor.

Table 3 Selected window of opportunity trials in head and neck cancer. Study

Agent investigated

Primary end point

Finding

Thomas [57] Del Campo [17] Schmitz [7] Brana [62] Bauman [64] Gross [21]

Erlotinib Lapatinib Cetuximab Dacomitinib Erlotinib and dasatinib Erlotinib and sulindac

Safety and efficacy Apoptic index Safety and 18FDG-PET Gene expression profile and Ki67 RECIST Ki67

Tumor shrinkage observed in 29% No modification of the apoptic index but decrease in Ki67; response rate of 17% Safe; 90% had 18FDG-PET partial response (EORTC guidelines) No significant Ki67 modification but predictive signature identified Activity of erlotinib but no activity of dasatinib Erlotinib decreased proliferation with additive effect from sulindac

18

FDG-PET: 2-[fluorine-18]-fluoro-2-deoxy-d-glucose positron emission tomography.

is frequently missing, and the duration of treatment with the investigational agent varies from one patient to another within the same trial (Table 3). EGFR is an interesting target in squamous cell carcinoma of the head and neck (SCCHN) because EGFR is overexpressed in up to 90% of SCCHN patients [49]. EGFR overexpression is also linked with poor prognosis [50–52]. Cetuximab is a chimeric IgG1 monoclonal antibody (mAb) that specifically binds to the EGFR. When combined with radiotherapy, it improves loco-regional control and survival compared to radiotherapy alone in patients with stage III/IV SCCHN. The addition of cetuximab to platinum-based chemotherapy and 5-fluorouracil improves overall survival in the first-line treatment of incurable disease [53,54]. Unfortunately the objective response rate as monotherapy remains low at between 6% and 13%, and no predictive biomarkers of activity or treatment resistance have been identified [55,56]. To fully evaluate the activity of EGFR or Human Epidermal Receptor (HER) inhibitors in SCCHN, and to identify predictive biomarkers of activity, several research groups have evaluated these inhibitors in window of opportunity studies [7,17,57,58]. Thomas and colleagues conducted a phase II pre-operative study

with erlotinib given for a median of 20 days (range: 18–30) [57]. Del Campo and colleagues conducted a neoadjuvant lapatinib trial in which therapy-naïve patients were randomized to receive lapatinib or placebo for two to six weeks before chemoradiation [17]. Both trials were pioneering, but the duration of treatment differed from one patient to another resulting in less standardized translational procedures and variations in the timing of imaging and tissue sample collection. Clinical and molecular activity associated with these compounds was also detected. Cetuximab has also been investigated in the pre-operative window period in treatment-naïve SCCHN patients selected for primary curative surgery [7]. The results demonstrated that pre-operative cetuximab infusion is safe, induces a high rate of 18FDG-PET response, and decreases tumour pEGFR, pErk and KI67 expression. A correlation between DSUVmax and residual tumor cellularity in the resected specimens was also found suggesting that 18FDG-PET may be an interesting endpoint. In the preoperative SCCHN studies with erlotinib or lapatinib described above, significant activity was also detected by 18FDG-PET [17,57]. Unfortunately, a comparison between these study results is difficult due to the absence of consensus guidelines concerning the evaluation of 18FDG-PET

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

responses in SCCHN. Despite this, the data collectively support the investigation of 18FDG-PET as an early marker of treatment activity if standardization can be performed between different trials and centers (i.e. EARL accreditation) [22]. 18FDG-PET can detect modifications in tumor composition, even in the absence of tumor shrinkage, and could therefore be a more sensitive technique to detect drug activity than conventional imaging where response is based on size modifications evaluated by RECIST criteria. These findings reinforce the fact that RECIST criteria are probably not the best way to evaluate the efficacy of targeted agents, particularly when the duration of treatment is short [59]. In contrast, diffusion weighted magnetic resonance imaging (DW MRI) may represent an interesting functional imaging alternative [60]. Another goal of window trials is to identify predictive biomarkers of drug activity. Thomas and colleagues performed gene expression profiles on tumor biopsies collected at baseline and after treatment with erlotinib. On the baseline tumor biopsies, they did not find statistically significantly differentially expressed genes between responders and non-responders. In contrast, they identified cyclin-dependent kinase 2-interacting protein (CINP) as a potential marker of erlotinib efficacy because its expression decreased only in patients who responded to treatment [61]. Dacomitinib, an irreversible oral pan–HER tyrosine kinase inhibitor was also investigated. Dacomitinib was given for 7–11 days before surgery in patients with resectable untreated oral carcinomas [62]. The investigators described a 47 gene signature that was highly expressed in responders. No differences were observed between the pre- and post-treatment tumor biopsies with regards to Ki67, between patients treated with dacomitinib or placebo, nor between responders and non-responders. The results were based, however, on only 10 evaluable patients. The Pittsburgh group investigated cetuximab in a window study and focused on the immunologic effects of the drug. They concluded that HLA class I upregulation was significantly associated with clinical response in cetuximab-treated SCCHN patients [58]. The same group also reported a patient who had a near complete pathological response after 13 days of erlotinib therapy in a window of opportunity trial. By whole exome sequencing, they found an activating MAPK1 E322K mutation [63]. Paradoxically to lung cancer, this mutation seems to enhance EGFR phosphorylation and erlotinib sensitivity in SCCHN cells. Window trials can only investigate molecular mechanisms occurring early to avoid delaying standard surgical curative treatment for ethical reasons. Other designs are needed to fully explore the long-term acquired drug resistance mechanisms. However, recently Schmitz and colleagues compared gene expression profiles in pre- and post-treatment biopsies and found that, in some patients, cetuximab could induce the activation of cancerassociated fibroblasts and favor epithelial-mesenchymal transition [23]. However, no correlation with clinical activity could be demonstrated in this small trial. Other targeted agents, which block the HER family receptors, are currently under investigation including KTN3379 (NCT02473731), a human anti-ErbB3 (HER3) monoclonal antibody; and afatinib, a pan–HER inhibitor (NCT01415674 and NCT01538381). The safety and pharmacodynamics of combined inhibition of the EGFR pathway and other potential targets in SCCHN are also under investigation. Gross and colleagues conducted a randomized, double-blind, placebo-controlled window trial of erlotinib with or without sulindac, a nonselective COX inhibitor, versus placebo [21]. Thirty-four paired tumor biopsies were obtained out of the 39 evaluable patients. There was a significant trend (p = 0.0185) in ordering of Ki67 reduction (erlotinibsulindac > erlotinib > placebo), which was the first endpoint of the study. Furthermore, low baseline pSrc correlated with greater

55

Ki67 reduction (R2 = 0.312, P = 0.024). Bauman and colleagues showed in a four-arm phase 0 trial, which investigated erlotinib, dasatinib and combined treatment versus placebo, a significant decrease in tumor size in the erlotinib arms (p = 0.0014) without additive effects of dasatinib. High baseline pMAPK expression was associated with a reduction in tumour size (p = 0.03), and high baseline pSTAT3 levels were associated with tumor progression (p = 0.03) [64]. A study combining cetuximab and/or IMC-A12, a human IgG1 monoclonal antibody to the insulin-like growth factor I receptor, is ongoing (NCT00957853). Medications blocking immunologic targets, like MEDI6469 (NCT02274155), an antiOX40 antibody; or AMG 319 (NCT02540928), a small molecule inhibitor of PI3Kd that could release the immune breaks within the tumor, are also the focus of window study designs in order to explore feasibility and pharmacodynamic parameters. Finally, cetuximab is being studied in combination with the immunologic sensitizer VTX-2337, a TLR8 agonist (NCT02124850), in order to investigate modulations of immune biomarkers and to relate them to anti-tumor response. Conclusions Window of opportunity studies are challenging and require considerable multidisciplinary collaboration and logistical support. Particular attention should also be paid to patient safety: the investigational drug must be well tolerated and have a known and acceptable safety profile. The trial should also ideally not interfere with the timely delivery of standard treatment. While window of opportunity studies require considerable support, we strongly believe that they offer an unbiased and unique opportunity to learn about the molecular activity, clinical efficacy, and mechanism(s) of action of new compounds. Dissecting the molecular pathways implicated in treatment response or resistance in tumor tissues obtained from these trials provides researchers with invaluable tools. These can be used to generate clinically meaningful hypotheses about drug combinations to increase treatment efficacy, or to develop biomarkers with a view to further tailoring and personalizing cancer therapy. Conflict of interest statement JP Machiels is an advisor for Boehringer-Ingelheim and MSD. He has received grant research from Novartis, Pfizer, Merck Serono, Janssen, and Bayer. F. Duhoux and S Schmitz have no conflict of interest Acknowlegement The authors wish to thank Aileen Eiszele for writing assistance. References [1] Postel-Vinay S, Collette L, Paoletti X, Rizzo E, Massard C, Olmos D, et al. Towards new methods for the determination of dose limiting toxicities and the assessment of the recommended dose for further studies of molecularly targeted agents–dose-Limiting Toxicity and Toxicity Assessment Recommendation Group for Early Trials of Targeted therapies, an European Organisation for Research and Treatment of Cancer-led study. Eur J Cancer 2014;50:2040–9. [2] Eisenhauer EA, O’Dwyer PJ, Christian M, Humphrey JS. Phase I clinical trial design in cancer drug development. J Clin Oncol 2000;18:684–92. [3] Glimelius B, Lahn M. Window-of-opportunity trials to evaluate clinical activity of new molecular entities in oncology. Ann Oncol 2011;22:1717–25. [4] Tsao AS. Current readings: window-of-opportunity trials for thoracic malignancies. Semin Thorac Cardiovasc Surg 2014;26:323–30. [5] Graham PJ, Brar MS, Foster T, McCall M, Bouchard-Fortier A, Temple W, et al. Neoadjuvant chemotherapy for breast bancer, is practice changing? A population-based review of current surgical trends. Ann Surg Oncol 2015;22:3376–82.

56

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

[6] Marous M, Bièche I, Paoletti X, Alt M, Razak AR, Stathis A, et al. Designs of preoperative biomarkers trials in oncology: a systemic review of the literature. Ann Oncol 2015. pii: mdv378. [Epub ahead of print]. [7] Schmitz S, Hamoir M, Reychler H, Magremanne M, Weynand B, Lhommel R, et al. Tumour response and safety of cetuximab in a window pre-operative study in patients with squamous cell carcinoma of the head and neck. Ann Oncol 2013;24:2261–6. [8] Primdahl H, Nielsen AL, Larsen S, Andersen E, Ipsen M, Lajer C, et al. Changes from 1992 to 2002 in the pretreatment delay for patients with squamous cell carcinoma of larynx or pharynx: a Danish nationwide survey from DAHANCA. Acta Oncol 2006;45:156–61. [9] Leary A, Evans A, Johnston SR, A’Hern R, Bliss JM, Sahoo R, et al. Antiproliferative effect of lapatinib in HER2-positive and HER2-negative/ HER3-high breast cancer: results of the presurgical randomized MAPLE trial (CRUK E/06/039). Clin Cancer Res 2015;21:2932–40. [10] Dowsett M, Smith I, Robertson J, Robison L, Pinhel I, Johnson L, et al. Endocrine therapy, new biologicals, and new study designs for presurgical studies in breast cancer. J Natl Cancer Inst Monogr 2011;2011:120–3. [11] Dowsett M, Smith IE, Ebbs SR, Dixon JM, Skene A, A’Hern R, et al. Prognostic value of Ki67 expression after short-term presurgical endocrine therapy for primary breast cancer. J Natl Cancer 2007;99:167–70. [12] Denkert C, Budczies J, von Minckwitz G, Wienert S, Loibl S, Klauschen F. Strategies for developing Ki67 as a useful biomarker in breast cancer. Breast 2015. http://dx.doi.org/10.1016/j.breast.2015.07.017. [13] Owonikoko TK, Ramalingam SS, Miller DL, Force SD, Sica GL, Mendel J, et al. A translational, pharmacodynamic, and pharmacokinetic phase IB clinical study of everolimus in resectable non-small cell lung cancer. Clin Cancer Res 2015;21:1859–68. [14] Ma T, Galimberti F, Erkmen CP, Memoli V, Chinyengetere F, Sempere L, et al. Comparing histone deacetylase inhibitor responses in genetically engineered mouse lung cancer models and a window of opportunity trial in patients with lung cancer. Mol Cancer Ther 2013;12:1545–55. [15] Feldt M, Bjarnadottir O, Kimbung S, Jirstrom K, Bendahl PO, Veerla S, et al. Statin-induced anti-proliferative effects via cyclin D1 and p27 in a window-ofopportunity breast cancer trial. J Transl Med 2015;13:133. [16] Fontanella C, Loibl S, von Minckwitz G. Clinical usefulness and relevance of intermediate endpoints for cytotoxic neoadjuvant therapy. Breast 2015. http:// dx.doi.org/10.1016/j.breast.2015.07.020. [17] Del Campo JM, Hitt R, Sebastian P, Carracedo C, Lokanatha D, Bourhis J, et al. Effects of lapatinib monotherapy: results of a randomised phase II study in therapy-naive patients with locally advanced squamous cell carcinoma of the head and neck. Br J Cancer 2011;105:618–27. [18] Lara-Guerra H, Waddell TK, Salvarrey MA, Joshua AM, Chung CT, Paul N, et al. Phase II study of preoperative gefitinib in clinical stage I non-small-cell lung cancer. J Clin Oncol 2009;27:6229–36. [19] Altorki N, Lane ME, Bauer T, Lee PC, Guarino MJ, Pass H, et al. Phase II proof-ofconcept study of pazopanib monotherapy in treatment-naive patients with stage I/II resectable non-small-cell lung cancer. J Clin Oncol 2010;28:3131–7. [20] Morrogh M, Andrade VP, Patil AJ, Qin LX, Mo Q, Sakr R, et al. Differentially expressed genes in window trials are influenced by the wound-healing process: lessons learned from a pilot study with anastrozole. J Surg Res 2012;176:121–32. [21] Gross ND, Bauman JE, Gooding WE, Denq W, Thomas SM, Wang L, et al. Erlotinib, erlotinib-sulindac versus placebo: a randomized, double-blind, placebo-controlled window trial in operable head and neck cancer. Clin Cancer Res 2014;20:3289–98. [22] Boellaard R, O’Doherty MJ, Weber WA, Mottaghy FM, Lonsdale MN, Stroobants SG, et al. FDG PET and PET/CT: EANM procedure guidelines for tumour PET imaging: version 1.0. Eur J Nucl Med Mol Imaging 2010;37:181–200. [23] Schmitz S, Bindea G, Albu RI, Mlecnik B, Machiels JP. Cetuximab promotes epithelial to mesenchymal transition and cancer associated fibroblasts in patients with head and neck cancer. Oncotarget 2015. http://dx.doi.org/ 10.18632/oncotarget.5924. [24] Cree IA, Glaysher S, Harvey AL. Efficacy of anti-cancer agents in cell lines versus human primary tumour tissue. Curr Opin Pharmacol 2010;10:375–9. [25] Johnson JI, Decker S, Zaharevitz D, Rubinstein LV, Venditti JM, Schepartz S, et al. Relationships between drug activity in NCI preclinical in vitro and in vivo models and early clinical trials. Br J Cancer 2001;84:1424–31. [26] Delitto D, Pham K, Vlada AC, Sarosi GA, Thomas RM, Behrns KE, et al. Patientderived xenograft models for pancreatic adenocarcinoma demonstrate retention of tumor morphology through incorporation of murine stromal elements. Am J Pathol 2015;185:1297–303. [27] Arnaout A, Robertson S, Kuchuk I, Simos D, Pond GR, Addison CL, et al. Evaluating the feasibility of performing window of opportunity trials in breast cancer. Int J Surg Oncol 2015;2015:785793. [28] Wisinski KB, Faerber A, Wagner S, Havighurst TC, McElroy JA, Kim K, et al. Predictors of willingness to participate in window-of-opportunity breast trials. Clin Med Res 2013;11:107–12. [29] Gandini S, Guerrieri-Gonzaga A, Pruneri G, Serrano D, Cazzaniga M, Lazzeroni M, et al. Association of molecular subtypes with Ki-67 changes in untreated breast cancer patients undergoing pre-surgical trials. Ann Oncol 2014;25:618–23. [30] Gao Q, Patani N, Dunbier AK, Ghazoui Z, Zvelebil M, Martin LA, et al. Effect of aromatase inhibition on functional gene modules in estrogen receptor-positive breast cancer and their relationship with antiproliferative response. Clin Cancer Res 2014;20:2485–94.

[31] Dowsett M. Optimizing the implementation of future treatment using surrogate end-points. Breast Cancer Res 2008;10(Suppl. 4):S26. [32] Decensi A, Robertson C, Viale G, Pigatto F, Johansson H, Kisanga ER, et al. A randomized trial of low-dose tamoxifen on breast cancer proliferation and blood estrogenic biomarkers. J Natl Cancer Inst 2003;95:779–90. [33] Di Leo A, Jerusalem G, Petruzelka L, Torres R, Bondarenko IN, Khasanov R, et al. Final overall survival: fulvestrant 500 mg vs 250 mg in the randomized CONFIRM trial. J Natl Cancer Inst 2014;106:djt337. [34] Robertson JF, Nicholson RI, Bundred NJ, Anderson E, Rayter Z, Dowsett M, et al. Comparison of the short-term biological effects of 7alpha-[9-(4,4,5,5,5pentafluoropentylsulfinyl)-nonyl]estra-1,3,5, (10)-triene-3,17beta-diol (Faslodex) versus tamoxifen in postmenopausal women with primary breast cancer. Cancer Res 2001;61:6739–46. [35] Rimawi MF, Shetty PB, Weiss HL, Schiff R, Osborne CK, Chamness GC, et al. Epidermal growth factor receptor expression in breast cancer association with biologic phenotype and clinical outcomes. Cancer 2010;116:1234–42. [36] Guix M, Granja Nde M, Meszoely I, Adkins TB, Wieman BM, Frierson KE, et al. Short preoperative treatment with erlotinib inhibits tumor cell proliferation in hormone receptor-positive breast cancers. J Clin Oncol 2008;26:897–906. [37] Decensi A, Puntoni M, Pruneri G, Guerrieri-Gonzaga A, Lazzeroni M, Serrano D, et al. Lapatinib activity in premalignant lesions and HER-2-positive cancer of the breast in a randomized, placebo-controlled presurgical trial. Cancer Prev Res 2011;4:1181–9. [38] Bjarnadottir O, Kimbung S, Johansson I, Veerla S, Jonsson M, Bendahl PO, et al. Global transcriptional changes following statin treatment in breast cancer. Clin Cancer Res 2015;21:3402–11. [39] Bjarnadottir O, Romero Q, Bendahl PO, Jirstrom K, Ryden L, Loman N, et al. Targeting HMG-CoA reductase with statins in a window-of-opportunity breast cancer trial. Breast Cancer Res Treat 2013;138:499–508. [40] Kalinsky K, Crew KD, Refice S, Xiao T, Wang A, Feldman SM, et al. Presurgical trial of metformin in overweight and obese patients with newly diagnosed breast cancer. Cancer Invest 2014;32:150–7. [41] Bonanni B, Puntoni M, Cazzaniga M, Pruneri G, Serrano D, Guerrieri-Gonzaga A, et al. Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial. J Clin Oncol 2012;30:2593–600. [42] Cazzaniga M, DeCensi A, Pruneri G, Puntoni M, Bottiglieri L, Varricchio C, et al. The effect of metformin on apoptosis in a breast cancer presurgical trial. Br J Cancer 2013;109:2792–7. [43] Dowling RJ, Niraula S, Chang MC, Done SJ, Ennis M, McCready DR, et al. Changes in insulin receptor signaling underlie neoadjuvant metformin administration in breast cancer: a prospective window of opportunity neoadjuvant study. Breast Cancer Res 2015;17:32. [44] Niraula S, Dowling RJ, Ennis M, Chang MC, Done SJ, Hood N, et al. Metformin in early breast cancer: a prospective window of opportunity neoadjuvant study. Breast Cancer Res Treat 2012;135:821–30. [45] DeCensi A, Puntoni M, Gandini S, Guerrieri-Gonzaga A, Johansson HA, Cazzaniga M, et al. Differential effects of metformin on breast cancer proliferation according to markers of insulin resistance and tumor subtype in a randomized presurgical trial. Breast Cancer Res Treat 2014;148:81–90. [46] Hadad S, Iwamoto T, Jordan L, Purdie C, Bray S, Baker L, et al. Evidence for biological effects of metformin in operable breast cancer: a pre-operative, window-of-opportunity, randomized trial. Breast Cancer Res Treat 2011;128:783–94. [47] Hadad SM, Coates P, Jordan LB, Dowling RJ, Chang MC, Done SJ, et al. Evidence for biological effects of metformin in operable breast cancer: biomarker analysis in a pre-operative window of opportunity randomized trial. Breast Cancer Res Treat 2015;150:149–55. [48] Schwab RB, Kato S, Crain B, Pu M, Messer K, Weidner N, et al. A window-ofopportunity biomarker study of etodolac in resectable breast cancer. Cancer Med 2015;4(10):1583–8. [49] Kalyankrishna S, Grandis JR. Epidermal growth factor receptor biology in head and neck cancer. J Clin Oncol 2006;24:2666–72. [50] Ang KK, Berkey BA, Tu X, Zhang HZ, Katz R, Hammond EH, et al. Impact of epidermal growth factor receptor expression on survival and pattern of relapse in patients with advanced head and neck carcinoma. Cancer Res 2002;62:7350–6. [51] Wheeler DL, Iida M, Kruser TJ, Nechrebecki MM, Dunn EF, Armstrong EA, et al. Epidermal growth factor receptor cooperates with Src family kinases in acquired resistance to cetuximab. Cancer Biol Ther 2009;8:696–703. [52] Rubin Grandis J, Melhem MF, Gooding WE, Day R, Holst VA, Wagener MM, et al. Levels of TGF-alpha and EGFR protein in head and neck squamous cell carcinoma and patient survival. J Natl Cancer Inst 1998;90:824–32. [53] Bonner JA, Harari PM, Giralt J, Azarnia N, Shin DM, Cohen RB, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567–78. [54] Vermorken JB, Herbst RS, Leon X, Amellal N, Baselga J. Overview of the efficacy of cetuximab in recurrent and/or metastatic squamous cell carcinoma of the head and neck in patients who previously failed platinum-based therapies. Cancer 2008;112:2710–9. [55] Baselga J, Trigo JM, Bourhis J, Tortochaux J, Cortes-Funes H, Hitt R, et al. Phase II multicenter study of the antiepidermal growth factor receptor monoclonal antibody cetuximab in combination with platinum-based chemotherapy in patients with platinum-refractory metastatic and/or recurrent squamous cell carcinoma of the head and neck. J Clin Oncol 2005;23:5568–77. [56] Herbst RS, Arquette M, Shin DM, Dicke K, Vokes EE, Azarnia N, et al. Phase II multicenter study of the epidermal growth factor receptor antibody cetuximab

S. Schmitz et al. / Cancer Treatment Reviews 43 (2016) 50–57

[57]

[58]

[59]

[60]

[61]

[62]

[63]

and cisplatin for recurrent and refractory squamous cell carcinoma of the head and neck. J Clin Oncol 2005;23:5578–87. Thomas F, Rochaix P, Benlyazid A, Sarini J, Rives M, Lefebvre JL, et al. Pilot study of neoadjuvant treatment with erlotinib in nonmetastatic head and neck squamous cell carcinoma. Clin Cancer Res 2007;13:7086–92. Srivastava RM, Trivedi S, Concha-Benavente F, Hyun-Bae J, Wang L, Seethala RR, et al. STAT1-induced HLA class I upregulation enhances immunogenicity and clinical response to Anti-EGFR mAb cetuximab therapy in HNC patients. Cancer Immunol Res 2015;3:936–45. van Oosterom AT, Judson I, Verweij J, Stroobants S, Donato di Paola E, Dimitrijevic S, et al. Safety and efficacy of imatinib (STI571) in metastatic gastrointestinal stromal tumours: a phase I study. Lancet 2001;358:1421–3. Schmitz S, Rommel D, Michoux N, Lhommel R, Hanin FX, Duprez T, et al. Dynamic contrast-enhanced computed tomography to assess early activity of cetuximab in squamous cell carcinoma of the head and neck. Radiol Oncol 2015;49:17–25. Thomas F, Delmar P, Vergez S, Rochaix P, Hennebelle I, McLoughlin P, et al. Gene expression profiling on pre- and post-erlotinib tumors from patients with head and neck squamous cell carcinoma. Head Neck 2013;35:809–18. Brana I, She D, Chau NG, Pham N-A, Kim L, Sakashita S, et al. Preoperative window-of-opportunity (WOO) study of dacomitinib (Dac) in patients (Pts) with resectable oral cavity squamous cell carcinoma (OCC): generation of a gene signature (DGS) as a predictor of Dac activity. J Clin Oncol 2014;32:5s. Van Allen EM, Lui VW, Egloff AM, Goetz EM, Li H, Johnson JT, et al. Genomic correlate of exceptional erlotinib response in head and neck squamous cell carcinoma. JAMA Oncol 2015;1:238–44.

57

[64] Bauman JE, Duvvuri U, Gooding WE, Gross ND, Song J, Yarbrough WG, et al. Erlotinib, dasatinib, erlotinib-dasatinib versus placebo: a randomized, doubleblind window study in operable head and neck squamous cell carcinoma (HNSCC). J Clin Oncol 2014;32:5s. [65] Bundred NJ, Cramer A, Morris J, Renshaw L, Cheung KL, Flint P, et al. Cyclooxygenase-2 inhibition does not improve the reduction in ductal carcinoma in situ proliferation with aromatase inhibitor therapy: results of the ERISAC randomized placebo-controlled trial. Clin Cancer Res 2010;16:1605–12. [66] Takagi K, Ishida T, Miki Y, Hirakawa H, Kakugawa Y, Amano G, et al. Intratumoral concentration of estrogens and clinicopathological changes in ductal carcinoma in situ following aromatase inhibitor letrozole treatment. Br J Cancer 2013;109:100–8. [67] Lee O, Page K, Ivancic D, Helenowski I, Parini V, Sullivan ME, et al. A randomized phase II presurgical trial of transdermal 4-hydroxytamoxifen gel versus oral tamoxifen in women with ductal carcinoma in situ of the breast. Clin Cancer Res 2014;20:3672–82. [68] DeFriend DJ, Howell A, Nicholson RI, Anderson E, Dowsett M, Mansel RE, et al. Investigation of a new pure antiestrogen (ICI 182780) in women with primary breast cancer. Cancer Res 1994;54:408–14. [69] Serrano D, Lazzeroni M, Gandini S, Macis D, Johansson H, Gjerde J, et al. A randomized phase II presurgical trial of weekly low-dose tamoxifen versus raloxifene versus placebo in premenopausal women with estrogen receptorpositive breast cancer. Breast Cancer Res 2013;15:R47.