Multi-Center Randomized Phase II Study Comparing Cediranib plus Gefitinib with Cediranib plus Placebo in Subjects with Recurrent/Progressive Glioblastoma

Background Cediranib, an oral pan-vascular endothelial growth factor (VEGF) receptor tyrosine kinase inhibitor, failed to show benefit over lomustine in relapsed glioblastoma. One resistance mechanism for cediranib is up-regulation of epidermal growth factor receptor (EGFR). This study aimed to determine if dual therapy with cediranib and the oral EGFR inhibitor gefitinib improved outcome in recurrent glioblastoma. Methods and Findings This was a multi-center randomized, two-armed, double-blinded phase II study comparing cediranib plus gefitinib versus cediranib plus placebo in subjects with first relapse/first progression of glioblastoma following surgery and chemoradiotherapy. The primary outcome measure was progression free survival (PFS). Secondary outcome measures included overall survival (OS) and radiologic response rate. Recruitment was terminated early following suspension of the cediranib program. 38 subjects (112 planned) were enrolled with 19 subjects in each treatment arm. Median PFS with cediranib plus gefitinib was 3.6 months compared to 2.8 months for cediranib plus placebo (HR; 0.72, 90% CI; 0.41 to 1.26). Median OS was 7.2 months with cediranib plus gefitinib and 5.5 months with cediranib plus placebo (HR; 0.68, 90% CI; 0.39 to 1.19). Eight subjects (42%) had a partial response in the cediranib plus gefitinib arm versus five patients (26%) in the cediranib plus placebo arm. Conclusions Cediranib and gefitinib in combination is tolerated in patients with glioblastoma. Incomplete recruitment led to the study being underpowered. However, a trend towards improved survival and response rates with the addition of gefitinib to cediranib was observed. Further studies of the combination incorporating EGFR and VEGF inhibition are warranted. Trial Registration ClinicalTrials.gov NCT01310855


D D O
O R R I I C C Multi-centre, randomised, double-blind phase II study comparing cediranib (AZD2171) plus gefitinib (Iressa, ZD1839) with cediranib plus placebo in subjects with recurrent/progressive glioblastoma Short Title/acronym: DORIC EUDRACT no: 2010-021531-13 Sponsor name & reference: University College London, UCL/10/0035 Funder name & reference: AstraZeneca, endorsed by CR-UK Clinicaltrials.gov no: (To Be Confirmed) Design: Multi-centre, randomised, double-blind phase II study Overall aim: The primary objective of the study is to determine the efficacy of cediranib in combination with oral gefitinib and cediranib alone by assessment of progression free survival (PFS) defined as the time from randomisation to first progression or death (whichever occurs first) Primary endpoint: Progression free survival (PFS) defined as the time from the date of randomisation to the date of first progression or death due to any cause, whichever one comes first. Secondary endpoints: • Overall survival (OS) • Radiographic response rate (RR) • PFS rate at 6 months (24 weeks) • 12 months survival rate • Steroid use • Time to deterioration of neurological status • Safety and tolerability Target accrual: 112 patients Planned number of sites: 6 centres Target countries: UK Treatment summary: Enrolled subjects will be randomised to receive either cediranib 30mg od orally and gefitinib 500mg od orally or cediranib 30mg od plus placebo. Each cycle of treatment lasts 6 weeks. Patients will be treated until progression (unless clinician feels patient is receiving benefit in which case treatment can continue), patient decision or the development of unacceptable toxicity. Anticipated duration of recruitment: Page no.10 expression of tissue factor which initiates pro-thrombotic conditions and these in turn contribute to the anomalous processes of neo-vascularisation that define glioblastomas (Rong et al. 2006a;Rong et al. 2006b). Disrupting the processes involved in angiogenesis is being increasingly adopted as a key element of the management of many tumour types including glioblastoma ).

Trial Schema
The use of anti-angiogenesis agents 'normalises' tumour vasculature, reducing tumour hypoxia, correcting interstitial hypertension and oedema. Each of these processes is thought to be desirable in enabling the more effective delivery of therapeutic agents ). By reducing interstitial hypertension and oedema 'anti-VEGF' treatments have demonstrated pronounced 'steroid sparing' effects. In the context of recent phase II trials the use of the monoclonal antibody bevacizumab on its own or in combination with chemotherapeutic agents has resulted in unprecedented levels of radiographic response in patients with recurrent glioblastoma, resulting in FDA approval for use in recurrent glioblastoma (Cohen et al. 2009b;Friedman et al. 2009;Vredenburgh et al. 2007). Bevacizumab has EMEA approval for use in some cancers but not glioblastoma (EMEA/CHMP/746208/2009).
In health the blood-brain barrier effectively limits the delivery of therapeutic drugs to the central nervous system. The blood-brain barrier (BBB) is formed by the interactions of the specialised endothelial cells of the cerebral microvasculature and the end-foot processes of astrocytes. It is essential in maintaining homeostasis within the central nervous system. It acts as a selective barrier to the diffusion of molecules on the basis of their size and lipid solubility. In general macro-molecular drugs are excluded from crossing the blood-brain barrier whilst small, lipid soluble agents can cross the barrier passively and small hydrophilic substances require active transport mechanisms. Targeted therapies currently include both monoclonal antibodies and small molecule inhibitors. Monoclonal antibodies are complex proteins measuring 150,000 Daltons in size whilst small molecule inhibitors are of the order of 400 Daltons (Esteva 2004). Small molecule targeted therapies have been demonstrated to cross the blood brain barrier in-vivo unlike monoclonal antibodies which are readily excluded by virtue of their size. The development of malignant lesions within the brain (primary and secondary) perturbs the function of the blood-brain barrier as is evidenced by the sequelae of vasogenic oedema, but some of the ability of the barrier to prevent macro-molecular entry is retained. There is therefore a rationale for testing of small molecule inhibitors in preference of antibodies when performing targeted therapy for glioblastoma.

Rationale for targeting EGFR
The epidermal growth factor receptor (EGFR) (ErbB-1) is a member of a family of 4 structurally related membrane spanning receptor tyrosine kinases including ErbB-3, ErbB-4 and ErbB-2 or HER-2.

DORIC
DORIC protocol version 1.0 30.11.2010 Protocol Template FINAL version 1.0 29Jan10 Page no.11 They are activated with varying efficacy by epidermal growth factor ligands. Receptor-ligand interaction leads to dimer formation and activation, which stimulates the intrinsic intracellular proteintyrosine kinase activity. The downstream targets of EGFR include the signaling proteins with important roles in cell lineage determination and cell survival. Mutations leading to the overexpression or amplification of EGFR contribute to oncogenesis by inducing cells to proliferate and to resist apoptosis. A variety of mutations affecting the expression and function of this family of receptors have been demonstrated in a variety of cancers (Nicholson et al. 2001).
In 40-50% of cases of glioblastoma the EGFR is over-expressed and co-expression of the constitutively activated mutant variant of EGFR, the epidermal growth factor variant III (EGFRvIII) is observed in nearly half of these cases (Ekstrand et al. 1991;Frederick et al. 2000;Libermann et al. 1985;Pelloski et al. 2007;Wong et al. 1987). The Epidermal growth factor receptor variant III (EGFRvIII) is the product of a common, tumour-specific mutation of the EGFR consisting of an inframe deletion of exons 2-7 (801bp) from its extracellular ligand-binding region.

Rationale for targeting both the VEGF and EGFR pathways
A network of interconnected signal transduction pathways is responsible for the development and maintenance of many solid tumours. Consequently, blockade of a single pathway may be ineffective in the long term because activation of other pathways can serve as escape mechanisms for the tumour (Scagliotti 2007).
Parallel and reciprocal pathways exist between the VEGFR and EGFR signalling cascades, clearly linking these pathways within tumours (Tabernero 2007). Therefore, dual inhibition of VEGFR and EGFR signaling cascades may be important for optimal suppression of tumour growth, and have been shown to have synergistic effects in preclinical models (van Cruijsen 2005). A randomised phase II study in non small cell lung cancer patients found a comparative effect with a more favourable toxicity profile using a combination of the VEGF antibody bevacizumab with the EGFR small molecule inhibitor erlotinib, in comparison to bevacizumab in combination with standard chemotherapy (Herbst 2007).
EGFR, as well as being involved in controlling cell proliferation and apoptosis, has also been shown to play a role in tumour angiogenesis (Ellis 2004). Activation of the EGFR pathway by either EGF or TGF increases the production of angiogenic factors (VEGF and VEGFR) in a variety of tumour cells including gastric (Akagi et al. 2003), bladder (Perrotte et al. 1999), and pancreatic cell lines . Preclinical studies using high grade glioma cell lines have similarly demonstrated that EGFR expression up-regulates the production of VEGF through mechanisms distinct from those under DORIC DORIC protocol version 1.0 30.11.2010 Protocol Template FINAL version 1.0 29Jan10 Page no.12 hypoxic regulation (Clarke et al. 2001;Maity et al. 2000;Pore et al. 2003). Conversely, inhibition of the EGFR pathway has been shown to reduce the production of angiogenic molecules in gastric, colonic, pancreatic and breast cancer cell lines (van Cruijsen et al. 2005). Of particular relevance to this study, gefitinib treatment has been shown to cause a dose-and time-dependent decrease in VEGF production in in-vitro studies of various cancer cell lines (Ciardiello et al. 2001) whilst also inhibiting the expression of other mediators of tumour-induced angiogenesis, such as cyclooxygenase-2, in squamous cell carcinoma cell lines of the head and neck (Chen et al. 2004).
Additional evidence that the EGFR and VEGFR pathways are linked has come from a recent study demonstrating that inhibition of the downstream EGFR-mediated effector, mammalian target of rapamycin (mTOR), reduces VEGF expression and capillary tube formation by endothelial cells (Bianco et al. 2008). Further support justifying the strategy of dual inhibition came from the discovery that combined inhibition of multiple targets has the potential to overcome resistance to monotherapies (Rubin and Duensing 2006). In one preclinical study, Van Cruijsen et al. demonstrated that resistance to anti-EGFR therapy could be overcome by adding antiangiogenic therapy to an anti-EGFR regimen (Van Cruijsen et al 2005).
Although some patients initially respond to EGFR TKIs, nearly all eventually acquire resistance to therapy following multiple or prolonged treatment (Camp et al. 2005). Resistance can be acquired via several mechanisms. Mutation or over-activity of EGFR-independent signaling pathways, such as the acquisition of K-ras mutation, or loss of PTEN expression are often responsible. Conformational changes in the TKI binding domain, and ligand-independent activation may also occur. The induction of higher angiogenic potential via upregulation of VEGF and other pro-angiogenic molecules by tumour cells has also been shown to be play a role (Ciardiello et al. 2004;Naumov et al. 2009;Viloria-Petit et al. 2001;Viloria-Petit and Kerbel 2004). If this is the case, then dual inhibition of both pathways may act to prevent resistance to EGFR inhibition through VEGFR.  Table 1 Trials of cediranib (AZD2171) in brain cancer (w w w .clinicaltrials.gov)
Baseline characteristics were balanced across treatments, although the lomustine arm did have a slightly higher Karnofsky Performance Status and lower steroid use compared to the other arms.
Notably early effects were seen with cediranib (as demonstrated by change in contrast-enhancing area and response rate) which were not maintained throughout the trial. Many of the patients on the lomustine arm went on to receive bevacizumab as subsequent therapy after the trial, influencing the overall survival S for these patients and thus confounding the overall survival findings. The lomustine alone arm had a relatively small number of patients in the context of a phase III trial and a few patients had a good response which influenced the overall outcome. Despite some weaknesses in design this trial showed evidence of clinical activity of cediranib on some endpoints.  Table 2 Trials of gefitinib in brain cancer (w w w .clinicaltrials.gov)

Trials using cediranib in other cancers
Phase II studies using single agent gefitinib in recurrent glioblastoma have demonstrated only modest activity in genetically unselected populations (Franceschi et al. 2007;Rich et al. 2004 and hence there is an increasing acceptance of the need to appropriately combine targeted therapies (Pillay et al. 2009).

Trials targeting EGFR and VEGFR
Many clinical studies investigating combinations of VEGFR and EGFR blocking agents in a variety of tumour types, including breast, colon, renal, head and neck, non-small cell lung cancer, and adenocarcinoma of unknown primary have been undertaken Cohen et al. 2009a;Dickler et al. 2008;Hainsworth et al. 2005;Hainsworth et al. 2007;Herbst et al. 2005;Saltz et al. 2007 Table 3 Phase II and II I Trials of com bination anti-VEGF and EGFR therapy in all cancers excluding brain cancer (see Table 4) (w w w .clinicaltrials.gov)

Phase I study of cediranib and gefitinib
A phase I study of cediranib plus gefitinib in patients with advanced solid tumours showed this combination to be well tolerated with manageable side effects. Diarrhoea, hypertension, fatigue, anorexia, skin toxicity and altered thyroid function were most commonly seen adverse events. The maximum tolerated dose (MTD) of cediranib in combination with 250 mg gefitinib was found to be 30  Table 4 Trials of com bination anti-VEGF and EGFR therapy in brain cancer (w w w .clinicaltrials.gov) Early studies using single agent targeted therapies directed against EGFR in patients with recurrent glioblastoma have shown disappointing results (Franceschi et al. 2007;Rich et al. 2004 upregulation of VEGF and other pro-angiogenic molecules by tumour cells, imparting upon them a higher angiogenic potential (Ciardiello et al. 2004;Naumov et al. 2009;Viloria-Petit et al. 2001;Viloria-Petit and Kerbel 2004). This has clear implications for the single agent use of EGFR targeted therapy in the 'avidly vascular' model of glioblastoma, and is the basis of the move to develop clinical trials examining the efficacy and safety of anti-VEGF and anti-EGFR targeted therapy in combination in recurrent glioblastoma. The combined EGFR/VEGF blocking agent AEE788 was well tolerated in a Phase I study in patients with recurrent glioblastoma. Seven (27%) patients achieved stable disease and progression free survival at 6 months was 14% (Reardon et al. 2005). Another combined EGFR/VEGF blocking agent vandetanib has been studied in a phase I study in combination with etoposide in recurrent glioblastoma patients. At dose level 1, higher than expected rates of grade 4 neutropaenia were observed. Consequently the dose level has been reduced, and dose escalation continues. A phase II trial is planned when the maximum tolerated dose of vandetanib with reduced dose etoposide is determined . In a recent phase II trial Hasselbalch et al treated forty-three patients with recurrent glioblastoma with a combination of cetuximab, bevacizumab and irinotecan. Radiological responses were noted in 34% of patients, of which two had complete responses and nine had partial responses. The 6-month progression-free survival probability was 30% and median overall survival was 29 weeks (95% CI: 23-37 weeks). However this was nonsuperior compared with results with bevacizumab and irinotecan alone (Hasselbalch et al. 2010). In a phase II trial of bevacizumab plus erlotinib in twenty-five patients with recurrent glioblastoma, twelve patients (48%) achieved radiological response and 6 month progression-free survival rate was 24% (Sathornsumetee et al 2008). The combination was well tolerated, with common side effects being rash, diarrhoea, mucositis and fatigue.
Given that the recent phase I study combining cediranib with gefitinib demonstrated combination treatment to be generally well tolerated, and to show encouraging antitumour activity in patients with advanced solid tumours (van Cruijsen et al. 2010), we aim to follow on directly from this and both build on the experience with cediranib as a single agent in recurrent/progressive glioblastoma and investigate whether the efficacy of cediranib is improved with the addition of gefitinib.

Proposed Trial
This is a phase II, randomised, double blind placebo controlled study in patients with recurrent or progressive glioblastoma (WHO grade IV). Patients are to receive cediranib in combination with gefitinib or cediranib with placebo.

Primary Objective
• The primary objective of the study is to compare the efficacy of cediranib in combination with oral gefitinib with cediranib alone.

Secondary Objectives
• To document the safety and tolerability of cediranib in combination with oral gefitinib and cediranib alone.  The progression definition will be based on modified RANO criteria (Wen 2010), such that progression will be defined as the earliest time that at least one of the following occurs: As assessed by site staff at clinical assessments: 1. Clinical deterioration not attributable to other causes apart from the tumour (e.g seizures, medication adverse events, complications of therapy, cerebrovascular events, infection, and so on) or changes in dexamethasone dose.
Clinical deterioration will be defined as: • Decline in Karnofsky performance status from 100 or 90 to 70 or less, or a decline of at least 20 from 80 or less, or a decline from any baseline to 50 or less, for at least 7 • Decline in ECOG performance status from 0 or 1 to 2 or 2 to 3

Failure to return for evaluation as a result of death or deteriorating condition
In this instance, the earliest known date of death or deterioration will be used as the date of progression. Where unknown, the clinical assessment date at which the patient did not attend will be used (only for patients known to have not returned for evaluation due to death or deteriorating condition).
As determined by retrospective radiographic central review: 3. Any new lesion (short axis must be ≥10 mm on at least 2 axial slices that are 5mm apart with 0mm skip and the new lesion should be outside the original tumour volume) 4. Increase in ≥25% of sum of the products of perpendicular diameters of enhancing lesions compared with baseline scan, on stable or increasing doses of steroids (dexamethasone) compared to baseline (T1 post-contrast scan) 5. Clear progression of non-measureable disease (clinically significant increase in size with a short axis ≥10mm on at least 2 axial slices that are 5mm apart with 0mm skip) 6. Significant increase in T2/FLAIR non-enhancing lesion -on stable or increasing steroids (dexamethasone) compared with baseline or best response not caused by co-morbid events. Precise quantification of the increase in T2/FLAIR signal can be difficult and must be differentiated from other causes of increased T2 or FLAIR signal, such radiation effects, decreased corticosteroid dosing, demyelination, ischemic injury, infection, seizures, postoperative changes, or other treatment effects before making a determination of progressive disease. Changes in T2/FLAIR signal that suggest infiltrating tumour include mass effect (as determined by sulcal effacement, ventricular compression, and thickening of the corpus callosum), infiltration of the cortical ribbon, and location outside of the radiation field. The initiation of these changes can be subtle, and convincing non-contrast-enhancing growth may require one or two confirmatory scans. If non-enhancing progression is determined after retrospective review of images, the scan at which these changes were first detected should serve as the progression scan. Change in signal intensity in T2/FLAIR within the original enhancing volume would not be sufficient for evidence of tumour progression.
The date of progression will be determined as being the date of the first MRI scan fitting the criteria for progression, or the date the clinical deterioration or death was first reported, whichever is earliest.
Locally reported progression should be determined by the above criteria. However, due to the possibility of pseudoprogression in these patients, a confirmatory second scan taken 6 weeks after the first should be used when making treatment decisions based on radiological progression. Please note it is not necessary to stop the trial drug in patients who have progressed only radiologically (even if confirmed at subsequent visits) if the clinician feels that the patient is benefiting from treatment.

Secondary outcome variables
• Overall survival (OS) -date from randomisation to death due to any cause • Radiographic response rate (RR) at each assessment point • PFS rate at 6 months (24 weeks). Progression is defined in the same way as described for the primary endpoint.
• Safety and tolerability • Overall Survival rate at 12 months • Time to sustained increase (≥ 21 days) in steroid dosage by ≥ 8mg/day compared to baseline or sustained nadir (≥ 21 days) (This is defined as time from randomisation to first increase in dexamethasone dose. An increase in dexamethasone dose is defined as a change from the baseline of ≥8mg/day for at least 21 days (i.e. the increase is sustained for at least 21 days consecutively). The earliest date when the dexamethasone dose was ≥ 8mg/day compared to baseline or sustained nadir in this period will be used for comparison between the treatment groups. Sustained nadir refers to the highest steroid dose taken during a ≥ 21 days period where the patient is consistently taking a lower dose of steroids compared to the baseline dose.) • Average daily steroid dosage change from baseline until first progression

Trial activation
UCL CTC will ensure that all trial documentation has been reviewed and approved by all relevant bodies and that the following have been obtained prior to activating the trial:

Site selection
In this protocol trial "Site" refers to the hospital or site where trial-related activities are actually conducted.
Sites must be able to comply with: • Trial treatment(s), imaging, follow up schedules and all requirements of the trial protocol

Selection of Principal Investigator and other investigators at sites
Sites must have an appropriate Principal Investigator (PI) i.e. a health care professional authorised by the site and ethics committee to lead and coordinate the work of the trial on behalf of the site. Other investigators at site wishing to participate in the trial must be trained and approved by the PI. All investigators must have experience of treating glioblastoma, have consultant status and routinely present cases to a Multi-Disciplinary Team.
Once the site has been activated by UCL CTC, the PI is responsible for ensuring: • Adherence to the most recent version of the protocol Page no.28 The following documentation must be in place prior to a site being opened to recruitment by UCL CTC trial team: • Trial specific Declaration of Participation/Site Registration Form including signature and delegation log (completed by all members of the core trial team at the site (research nurses, trial administrators, lead pharmacy staff responsible for the trial), and signed off by the Principal investigator) • All relevant institutional approvals (e.g. local NHS permission) • A signed Clinical Trial Site Agreement (CTSA) between the Sponsor and the relevant institution (usually a NHS Trust) The PI, other delegated site investigators and all staff involved in the conduct of the trial at the site must be identified on the site delegation log held at site and copied to UCL CTC prior to site activation.
Sites must also have in place facilities for providing 24 hour medical advice for trial patients.

Site activation
Once the trial team at UCL CTC have confirmed that all documentation is in place a site activation letter will be issued to the Principal Investigator, at which point the site may start to approach patients.

Informed consent
Sites are responsible for assessing a patient's capability to give informed consent.
Sites are responsible for ensuring all patients have been given the current version of the patient information sheet, are fully informed about the trial and have confirmed their willingness to take part in the trial by signing a consent form. In addition to the consent form to indicate consent to the trial, there are additional consent forms to indicate that the patient is willing to donate any of the optional biological specimens for the trial (archival biopsy samples, blood samples throughout the trial, and/or a post mortem tumour sample). The PI or other delegated site staff are required to provide a full explanation of the trial and all relevant treatment options to each patient prior to trial entry. During these discussions the current detailed patient information sheet for the trial will be given to the patient. A minimum of twenty four hours must be allowed for the patient to consider and discuss participation in the trial. Written informed consent on the current version of the consent forms for the trial must be obtained before any trial-specific procedures are conducted.
For patients in this trial, consent will also be obtained from the patient's named carer, because the carer may be required to assist with CRF completion should the patient's condition deteriorate or in the event that the patient is unable to speak or write. In these circumstances the carer may give answers on behalf of the patient for questions regarding quality of life, Karnofsky performance score, health economics, and adverse events.
Site staff are responsible for: • checking that information on the consent forms are complete and legible • checking that the patient has completed/initialled all relevant sections and signed and dated the form • Checking that an appropriate member of staff has countersigned and dated the consent forms to confirm that they provided information to the patient • Checking that an appropriate member of staff has made dated entries in the patient's medical notes relating to the informed consent process (i.e. info given, consent signed etc.) • Adding the patient trial number to all copies of the consent form The original signed consent form and a copy must be stored at site (in the Investigator Site File and the patient's medical notes). A further copy must be given to the patient.
The right of the patient to refuse to participate in the trial without giving reasons must be respected.
All patients are free to withdraw at any time (see section 13.0 -withdrawal of patients).

Patient Eligibility
There will be no exception to the eligibility requirements at the time of randomisation. Queries in relation to the eligibility criteria should be addressed prior to calling/faxing for randomisation. Patients are eligible for the trial if all the inclusion criteria are met and none of the exclusion criteria applies.

Inclusion Criteria
For inclusion in the clinical trial, patients must fulfil all of the following criteria: • Provision of informed consent • Age ≥18 years • Life expectancy ≥ 12 weeks • Histological/cytological confirmation of glioblastoma (WHO grade IV) • Patients with measurable disease (contrast-enhancing tumour ≥10 mm by shortest diameter on 2 axial slices) by MRI imaging within 7 days prior to enrolment. (If patients have recently had a routine MRI scan, this should be assessed before deciding whether or not to screen the patient, and booking the screening/baseline MRI.) • Patients must have been on no steroids or a stable dose of steroids (dexamethasone) for at least 5 days before the baseline MRI o It is not essential that the entire Stupp regimen of 6 cycles of adjuvant temozolomide following chemoradiotherapy has been completed.
o The last dose of temozolomide must be more than 28 days from enrolment.
o Gliadel ® wafers are permitted, as it is part of local treatment. For inclusion in the genetic research, patients must fulfil the following criterion: • Provision of informed consent for genetic research (separate consent required for tumour biopsy, blood sample, and post mortem donations) If a patient declines to participate in any of the genetic research, there will be no penalty or loss of benefit to the patient. The patient will not be excluded from other aspects of the study described in this Clinical Study Protocol, so long as they consent to the main study.

Exclusion Criteria
Any of the following is regarded as a criterion for exclusion from the study: • Patients on enzyme-inducing anti-epileptic drugs (see section 7.5) within 2 weeks prior to study enrolment. Note: Patients are eligible if they switched to non-enzyme inducing agents and discontinued enzyme-inducing agents for more than or equal to 2 weeks prior to randomisation • Inadequate bone marrow reserve as demonstrated by an absolute neutrophil count ≤1.5 x 10 9 /L or platelet count ≤100 x 10 9 /L or requiring regular blood transfusions to maintain haemoglobin >9g/dL • Serum bilirubin ≥1.5 x ULRR (except for patients with known documented cases of Gilbert's Syndrome) • ALT or AST ≥5 x ULRR Page no.32 • Greater than +1 proteinuria on two consecutive dipsticks taken no less than 1 week apart unless urinary protein <1.5g in a 24 hr period or UPC (Urine Protein: Creatinine) ratio <1.5 • History of significant gastrointestinal impairment, as judged by the investigator, that would significantly affect the absorption of cediranib or gefitinib, including the ability to swallow the tablet whole • Patients with a history of poorly controlled hypertension with resting blood pressure >150/100mmHg in the presence or absence of a stable regimen of anti-hypertensive therapy, or patients who are requiring maximal doses of calcium channel blockers to stabilise blood pressure • Any evidence of severe or uncontrolled diseases (e.g. unstable or uncompensated respiratory, cardiac, hepatic or renal disease) • Unresolved toxicity >CTC AE grade 1 from previous anti-cancer therapy (including radiotherapy) except alopecia (if applicable) • Mean QTc with Bazetts correction >470msec in screening ECG or history of familial, long QT syndrome (see Appendix 7) • Cardiac ventricular arrhythmias requiring anti-arrhythmic therapy • Significant haemorrhage (>30mL bleeding/episode in previous 3 months) or haemoptysis (>5mL fresh blood in previous 4 weeks) • Recent (<14 days) major surgery or brain biopsy. Recent craniotomy (<28 days) prior to first dose, or a surgical incision that is not fully healed • Pregnant or breast-feeding women or women of childbearing potential with a positive pregnancy test prior to receiving study medication (see section 5.1.3) • Known hypersensitivity to cediranib, gefitinib or any of its excipients • History of other malignancies (except for adequately treated basal or squamous cell carcinoma or carcinoma in situ) within 5 years, unless the patient has been disease free for 2 years and they have tissue diagnosis of the target lesion Page no.33 • Involvement in the planning and conduct of the study (applies to both UCL CTC, AstraZeneca staff and staff at the study site) • Past medical history of interstitial lung disease, idiopathic pulmonary fibrosis, drug-induced interstitial disease, radiation pneumonitis which required steroid treatment or any evidence of clinically active interstitial lung disease • Previous enrolment as part of the present study • Treatment with an investigational drug within 30 days prior to the first dose of cediranib/gefitinib • Other concomitant anti-cancer therapy except steroids (dexamethasone only) • Previous anti-angiogenesis (e.g. bevacizumab, sorafenib, sunitinib) therapy • Previous anti-EGFR treatments (e.g. cetuximab, panitumumab or small molecule tyrosine kinase inhibitors etc.) or downstream targets e.g. mTOR inhibitors.
• Patients with evidence of any intratumoural or peritumoural haemorrhage deemed significant by the treating physician • Patients who have received any form of cranial radiation within 3 months prior to study entry (excluding imaging) • Patients who have progressed within 3 months of completion of standard cranial radiation • Patients that have received radiosurgery or brachytherapy • Patients on >8mg/day dexamethasone or equivalent steroids on any day of the 2 weeks prior to randomisation

Restrictions
• For elective surgery during the study, or any procedure that carries a risk of internal bleeding, it is recommended that cediranib be stopped for 2 consecutive weeks prior to the surgical procedure. Cediranib treatment can be restarted when the surgical wound has healed. If emergency surgery is performed, precautions should be taken to minimise the potential risk of bleeding and thrombosis associated with this class of agents, cediranib should be stopped and Page no.34 close monitoring for bleeding, wound healing and thromboembolic complications should be initiated. Patients should not receive cediranib within 2 weeks of major surgery.
• Caution in the concomitant use of any medication that may markedly affect renal function. Such medications may be used with caution if deemed essential for treatment or may be continued if already in use prior to entry in the study with no effect on renal function.
• Caution should be exercised in concomitant use of any medication that may significantly affect hepatic P450 drug metabolising activity by way of enzyme induction or inhibition within 2 weeks of the first dose of cediranib or gefitinib and throughout the study period (see Appendix 3 for a list of drugs) • Patients who are blood donors should not donate blood during the study and for 3 months following their last dose of study treatment • Medicinal products that cause significant sustained elevation in gastric pH, such as proton pump inhibitors and h2 antagonists may reduce bioavailability and plasma concentrations of gefitinib and therefore, may reduce efficacy. Antacids if taken regularly close in time to administration of gefitinib may have a similar effect. These should be avoided where possible.
• Patients should take their medication at least 1 hour prior to the consumption of a meal or at least 2 hours after a meal has been ingested.

Pregnancy and Birth Control
The potential risk to patients who conceive or are pregnant whilst taking study medication is not known, although reproductive toxicity has been reported in animals administered with either gefitinib or cediranib. Women of childbearing potential will be asked to take a pregnancy test by a healthcare professional at the screening assessment within 14 days of starting trial treatment.
Women of child bearing potential will be defined as all female trial participants who are not postmenopausal. Postmenopausal females are defined as: • Natural menopausal with menses >2 years ago

Contraception
Female patients on the study must be post-menopausal, surgically sterile, sexually abstinent or use two reliable forms of contraception from starting cediranib and/or gefitinib and for 2 weeks after the last dose. Reliable methods of contraception should be used consistently and correctly, acceptable methods include barrier methods, implants, injectables, combined oral contraceptive methods, some IUDs, or partner had a vasectomy.
Male patients must use a barrier method of contraception from starting trial medications and for 2 weeks after the last dose.
In the event of a pregnancy occurring on the trial, please report immediately to the UCL CTC (see section 10.6 Pregnancy reporting).

Pre-randomisation evaluation
Patients must give written informed consent before any trial specific screening investigations may be carried out. The following assessments or procedures are required to evaluate the suitability of patients prior to entry: • Confirmation (histological/cytological) that patient has glioblastoma (WHO grade IV) • Within 7 days prior to randomisation:

Screening Log
A screening log should be maintained by the site and kept in the Investigator Site File. This should record each patient screened for the trial and whether or not they were randomised to the trial. If they were not randomised the reasons should be given. The log should be sent to UCL CTC when requested with patient identifiers removed prior to sending.

Randomisation procedures 6.1 Randomisation
Patient randomisation will be performed centrally at the UCL CTC, after the relevant site has faxed the fully completed randomisation form to UCL CTC and it has been checked by a trained UCL CTC staff member to confirm eligibility and informed consent. Randomisation must be performed prior to commencement of any trial treatment. Patient randomisation will be performed using an interactive web-based recognition system (Cenduit IWRS), with a link to an Access randomisation programme.
Randomisation will be undertaken using minimisation, and stratified by age (<65 or ≥65yrs) and resection for the recurrent disease (previous resection / no previous resection), Karnofsky performance status (>80 / ≤80). Within each stratum patients will be randomised in a 1:1 ratio to the treatment arms.
Sites should ensure that the following have been completed before randomising: • Patient consent form must be signed • Pre-randomisation evaluations should be carried out at sites as detailed in Section 5.3 (following consent if not done routinely) • The randomisation form must be fully completed prior to faxing the UCL CTC • Screening MRI scan must have been completed (a copy should be sent to UCL CTC following randomisation, see Section 7.7) On receipt of completed randomisation form at UCL CTC: • If the patient is confirmed as eligible, a patient trial number will be assigned at UCL CTC • The UCL CTC will then randomise the patient on the Cenduit interactive web-based response system (IWRS). The IWRS will send an e-mail to site staff and pharmacy, notifying them of medication codes which correspond to stock sufficient to cover the patient's first 6 weeks of treatment i.e. 6 weeks of daily cediranib plus daily gefitinib/placebo. Site staff will be responsible for using IWRS for drug resupply visits (for more information, refer to the IWRS user manual).
• A unique patient trial number will be assigned. The UCL CTC will fax a confirmation form to the site's lead research contact and also the lead pharmacist, confirming both the patient's inclusion in the trial and their trial number. The site should then enter the patient's trial number in the randomisation CRF and arrange for the MRI scan to be sent to the UCL CTC. Page no.38 • The original randomisation form should be posted to UCL CTC and a copy kept at site.
• Case report forms will be sent by UCL CTC to the main contact at site.

Randomisation telephone number:
+44 (0) Once a patient has been randomised onto the trial they must be provided with the following: • A copy of their signed consent form(s) and patient information sheet • A patient contact card. Site on-call contact details for out-of-hours medical care must be added to this card and patients informed to carry this with them at all times while on the trial The baseline assessment should be carried out on the day that study treatment begins, which must be within 2 weeks of the screening assessment. If screening laboratory and ECG tests were carried out more than 7 days prior to baseline these should be repeated at baseline. The baseline MRI should be within 7 days before randomisation. Baseline assessments should be carried out as per Section 8.1 and Section 5.2.

Interactive Web-based Recognition System (IWRS)
The IWRS will be used for randomised allocation, trial medication assignment, initial drug supply and resupply, discontinuation from study treatment, emergency code breaks (i.e. unblinding) and trial drug shipment confirmation. The IWRS technology will be managed and maintained by Cenduit, in collaboration with Fisher Clinical Services. The system is accessible via the internet 24 hours a day, 7 days a week. The IWRS user manual will be provided to all sites, and contains full details regarding the operation and use of the system.

Trial drug supply 6.2.1 Initial trial drug supply
An initial supply of cediranib, gefitinib and placebo will be shipped to each site by Fisher upon site

Resupply of cediranib & gefitinib/placebo
Resupply of cediranib and gefitinib/placebo will be managed automatically by IWRS, which will count down as patients are allocated trial stock. A resupply shipment will be triggered once a site's stock becomes sufficiently low. However, site pharmacies will be able to manually request additional stock via the IWRS. For exceptional circumstances (e.g. a site intending to randomise several patients on one day), sites will need to provide the UCL CTC with a week's notice where possible, so that IWRS is able to ensure that sufficient stock is provided to meet the site's dispensing requirements. Enrolled subjects will be randomised to receive either cediranib 30mg od orally and gefitinib 500mg od orally or cediranib 30mg od plus placebo. Each cycle of treatment lasts 6 weeks.

DORIC
Treatment will continue until confirmation of progression as defined in 2.6.1.1, patient decision or the development of unacceptable toxicity (if there is radiological progression only treatment can continue if the investigator has the opinion that the patient is receiving benefit. It is not compulsory to stop trial treatment at any indication of radiological progression, even at confirmed radiological progression on multiple scans if it is felt that the patient is receiving benefit. Investigators should be wary of pseudoprogression on MRI scans.).
If none of the above are thought to have occurred on the last day of the treatment cycle, the patient will start the next cycle of treatment the following day. Treatment pauses of up to 14 days are permitted (21 days are permitted in consultation with the CI) but should be recorded on the next CRF. Treatment delays of longer than 14 days (or 21 days in consultation with the CI) will result in patient withdrawal from the study.

Summary Treatment Schedule
Patients randomised to receive cediranib and gefitinib will receive their drugs according to the following schedule: Patients randomised to receive cediranib and placebo will receive their drugs according to the following schedule: For both arms, each cycle will last 6 weeks and treatment will continue until progression as defined in 2.6.1.1, patient decision or the development of unacceptable toxicity (if there is progression treatment can continue if the investigator has the opinion that the patient is receiving benefit). There is therefore not a set number of treatment cycles.
At the beginning of each treatment cycle, all patients will be dispensed enough drug for one cycle (6   weeks) plus an extra weeks worth of tablets in case of delayed cycle start assessments. This is to prevent unnecessary treatment delays. Unused tablets must be returned to the pharmacy for accountability and destruction.

Supplied by AstraZeneca
Pharmaceutical form 35 x film-coated tablets per bottle, containing 30mg cediranib/tablet (20mg and 15mg tablets also available where dose reductions are necessary, see section 7.3.4). Other ingredients are mannitol, dibasic calcium phosphate anhydrous, sodium starch glycolate and magnesium stearate with a film coat containing hypromellose 2910, polyethylene glycol 400, red iron oxide, yellow iron oxide, black iron oxide and titanium dioxide Dose and scheduling 30mg once daily* (1 tablet)

Mode of administration
Orally (in the morning at least one hour before food. If a meal is taken before the drug at least two hours should elapse before taking cediranib)

Storage
Less than 30°C. Protect from light, protect from moisture. Week

Status of drug
Licensed for treating Non Small Cell Lung Cancer. Not licensed for treatment indication (Glioblastoma).

Mode of administration
Orally (may be taken with or without food)

Storage
Less than 30°C

Supplied by AstraZeneca
Pharmaceutical form 100 x 250mg film-coated tablets identical to Gefitinib per bottle. The tablets contain lactose monohydrate, microcrystalline cellulose, croscarmellose sodium, povidone, sodium lauryl sulphate and magnesium stearate with a film coat containing hypromellose, macrogol 300, red iron oxide, yellow iron oxide and titanium dioxide.

Mode of administration
Orally (may be taken with or without food)

Treatment compliance
Patients should be given clear instructions on how and when to take their study drugs. The investigator or pharmacy must retain records of all study drugs administered. The UCL CTC will check these records to confirm compliance with the protocol administration schedule.
Patients should be encouraged to record when they take their study drugs each day in patient diaries.
Any dose reductions will be documented on the CRFs, along with reasons for the dose reduction. If a patient forgets to take a tablet, and it is within 6 hours of the scheduled time, then the patient should be advised to take the tablet as soon as possible. If it is more than 6 hours after the scheduled time, then study medication should not be taken for that day. Study medication should continue as previously scheduled on the subsequent day. A patient should not take more than a single day's dose of tablets within a day. Use of study medication in doses in excess of that specified in the protocol is considered to be an overdose. Refer to section 10.2.1.1 for procedures in case of an overdose.
If the patient vomits within 30 minutes from taking the tablet or if they can identify the tablet in the vomit content, the patient can take a new tablet from the bottle. Trial medication should continue as usual on the subsequent day.

Difficulties swallowing tablets
Entry into the DORIC trial requires that a patient is able to swallow tablets whole. However, for patients who are unable to swallow tablets due to their underlying disease or who become unable to swallow the tablet whole whilst on study, and who are considered to be benefiting from therapy, tablets may be administered by alternative methods as outlined below.
Cediranib film-coated tablets have not been deliberately formulated to be dispersible tablets. However they may be administered as a dispersion in water ( injection] or purified water). Liquids other than water should not be used, and the tablets should not be crushed or ground. Care should be taken to ensure that the whole dose is administered.
The following procedure is recommended: Cediranib tablets may be dispersed in half a glass (2 fluid ounces or 50 ml) of non-carbonated drinking water. No other liquids should be used. The tablet is dropped in water, without crushing, stirred until dispersed (approximately 10 minutes) and the resultant dispersion swallowed immediately. Any residues in the glass are mixed with half a glass of water and swallowed. The liquid can also be administered through nasogastric or gastrostomy tubes.
Where patients require the dose to be administered by nasogastric or gastrostomy tubes, cediranib tablet(s) dispersed in water can be dosed with: •  Polyurethane (PUR) or Poly vinylchloride (PVC) naso-gastric, naso-intestinal or percutaneous endoscopic gastrostomy (PEG) feeding systems in conjunction with PUR syringes.
• PVC syringes are not recommended for use.
• Two system washes are conducted through the giving set to ensure the correct dose is obtained.
The tablet can be swallowed whole with some water or if dosing of whole tablets is not possible, tablets may be administered as a dispersion in water (non carbonated). No other liquids should be used. Without crushing it, the tablet should be dropped in half a glass of drinking water. The glass should be swirled occasionally, until the tablet is dispersed (this may take up to 20 minutes). The dispersion should be drunk immediately after dispersion is complete (i.e. within 60 minutes). The glass should be rinsed with half a glass of water, which should also be drunk. The dispersion can also be administered through a naso gastric or gastrostomy tube.

Pharmacy responsibilities: Labelling
The clinical study drugs (cediranib and gefitinib and gefitinib placebo) will be supplied by AstraZeneca Each container of cediranib, gefitinib or placebo will have an investigational use label permanently affixed to the outside and will be labelled in accordance with local regulations, stating that the drug is for investigational use only and should be kept out of reach of children. Instructions stating that the tablets should be taken at the same time each day will be included.
It is the investigator's responsibility to ensure the patient receives the correct dose.
All products distributed through the Interactive Web Response System (IWRS) system will be labelled with a unique medication ID number to allow drug tracking.

Storage
All investigational products must be kept in a secure place under appropriate storage conditions. A description of the appropriate storage and shipment conditions are specified on the investigational product bottle label and in the Investigator's Brochure for cediranib (AZD2171, 2010) and SmPC for gefitinib. The investigator will instruct the patient about storage requirements for study medication.
In the event of deviations from acceptable storage temperatures during the storage of trial drug at the site, the temperature excursions procedure should be followed (details will be held in the pharmacy file).

Accountability
It is the investigator/institution's responsibility to establish a system for handling study treatments, including investigational medicinal products, so as to ensure that: • Deliveries of such products from Fisher are correctly received by a responsible person.
• Such deliveries are recorded.
• Study treatments are handled and stored safely and properly.
• Study treatments are only dispensed to study patients in accordance with the protocol.
• Any unused products are accounted for and returned for destruction, or destroyed locally, in liaison with UCL CTC • The study personnel will account for all drugs dispensed.
• Certificates of delivery, destruction and return must be signed Destruction of these drugs will be conducted according to a site's local policy, at the request of

Drug accountability
Accountability for cediranib, gefitinib and placebo at participating sites is the responsibility of the Principal Investigator, who may delegate this responsibility to the local pharmacist, or other appropriately qualified personnel. The responsible person will ensure that the cediranib, gefitinib and placebo are used only in accordance with this protocol and that appropriate records are maintained.
The site pharmacy must maintain accountability records for each drug including receipt, dispensing, return and destruction of medication. Accountability logs will be provided for each site pharmacy and accountability will be carried out throughout the trial whilst patients receive study medication. Drug delivery receipts, certificates of destruction of unused/expired medication and records of returned unused/expired medication must be kept in the pharmacy file. Temperature logs should be kept in the pharmacy file or centrally.
Patients should return any unused medication at their next visit to the pharmacy (e.g when collecting their next batch of trial medication). Returns should be counted by the pharmacist and recorded on accountability logs in the trial pharmacy site files. Patients who withdraw from the trial should arrange for medication to be returned to the pharmacy for accountability and destruction in cytotoxic waste according to local procedures.
Cediranib, gefitinib and placebo are supplied for DORIC patients only and must not be used outside the context of this protocol. Under no circumstances should the P.I. or other site personnel supply cediranib, gefitinib or placebo to other investigators, patients, or clinics, or allow supplies to be used other than directed by this protocol without prior authorisation from the supplier and notification to the Sponsor.

Dose modifications
If cediranib or gefitinib is not tolerated, patients will be managed by allowing a dose reduction by one or more levels according to

Management of general toxicity
The following general guidance is recommended for management of toxicities and dose de-escalation.
• All dose changes should be documented with clear reasoning and documentation of the approach taken • Treat each of the toxicities with maximum supportive care (including withholding the experimental therapy where required) • If the symptoms promptly resolve with supportive care, consideration should be given to continuing the same dose of study medications along with appropriate continuing supportive care, providing that continued treatment with study therapy is considered medically appropriate, in the opinion of the investigator as the patient is considered to be receiving benefit. If medically appropriate, dose reductions are permitted for study medications Prophylatic antiemetics need only be given as required.

Management of toxicity attributable to cediranib
Dose interruptions should be used as the first approach to managing toxicity and dose reduction may be considered.
The events for which de-escalation of the cediranib dose is recommended include, with the A maximum of a 14-day (21-day in consultation with the CI) delay in dosing for cediranib is permitted. If a longer interruption is required due to unresolved toxicity, cediranib should be discontinued. In addition, there are certain circumstances in which study medication should be permanently discontinued (see section 13.1).

Management of proteinuria
If a patient has a change of two plus (++) from baseline on two consecutive urine protein dipstick measurements, please measure urine protein creatinine ratio or collect a 24-hour urine for total protein. A urine protein/creatinine ratio of 0.15 (urine protein and urine creatinine expressed in mg/dL) approximates a 24-hour urine protein of 150 mg/24 hours or 0.15g/24 hours, which is the upper limit of normal. If 24 hour proteinurea or UPC ratio is classified as CTCAE grade 3 or more please follow guidance in Section 7.4.1.

Management of abnormal thyroid function tests
Cediranib therapy has been associated with increases in TSH. In the majority of patients this has not resulted in reductions in either total triiodothyronine (T3) or free thyroxine (T4) to below the lower Pharmacological levothyroxine substitution should be given when clinically indicated to normalise the thyroxine level to within the normal range, and before the patient becomes clinically symptomatic.
Levothyroxine therapy should also be considered in patients with TSH increases (and thyroxine levels within the normal range), with adverse events and symptoms suggestive of hypothyroidism such as fatigue. Thyroid function and related symptoms should be monitored frequently and the dose of levothyroxine should be titrated as required (Rini et al. 2007).

Management of haematological toxicity
Thrombocytopenia, of CTC grade 1 or 2 in the majority of cases, has also been seen with cediranib monotherapy and combination therapy with other treatments. In view of this finding, close monitoring of platelets at study visits is recommended. Patients with platelet count ≤100 x 109/L at baseline should not be included in studies with cediranib and are not eligible for this trial.
Although the effect of short treatment breaks of cediranib on platelet count recovery has not been studied, it is possible that a short break of a few days may help in the situation of CTC grade 3 or grade 4 or prolonged thrombocytopenia. Treatment breaks should be taken at the discretion of the treating clinician and patient response monitored. As guidance, where platelets are <50 x 10 9 /L a dose reduction or interruption may be required. Concerns may be discussed with the Chief Investigator.

Hypertension management
Patients will be provided with blood pressure monitors at the beginning of the trial to monitor blood pressure at home on a daily basis and will be encouraged to record this in the patient diaries. Data from the diaries will be used by the site staff to complete the CRFs at each assessment.  The development of hypertension in patients treated with cediranib is generally an early event, usually occurring in the first 4 weeks of treatment. In a minority of patients hypertension can also occur later in a treatment schedule.
The rigorous monitoring of BP and adherence to the hypertension management guidelines are necessary in order to achieve optimal hypertension control. Patients on prior anti-hypertension therapy may be particularly at risk of developing moderate or severe hypertension on cediranib.
Therefore, patients with pre-existing hypertension or on anti-hypertension medications are likely to benefit from having their BP management optimised before starting cediranib.
For all BP thresholds described in this protocol, a trigger level is considered to be met if either the systolic and/or diastolic pressure reaches the threshold. If the threshold is recorded at home, it must be confirmed by a healthcare professional as defined above before commencing any treatment.
CLEAR reasons for progressing to the next step in the management protocol must be documented.
When managing mild to moderate hypertension, the following principles should be noted: • It is recommended that no more than 2 drugs are added in a 48-hour period before temporarily stopping cediranib.
The following cautions and contraindications should be noted: •

Dermatological toxicity
The cumulative exposure of trial subjects to oral VEGFR tyrosine kinase inhibitors is associated with the dose dependent development of palmar plantar erythrodysesthesia (PPE) (Porta et al. 2007;Rosenbaum et al. 2008 General measures for managing PPE associated with oral VEGFR tyrosine kinase inhibitors (Schwandt A et al. 2009): • Patients should be advised to make use of the following: cotton socks, shoes with extra support and gel insoles • Topical emollient creams and lotions should be used to minimise dryness, pruritus and discomfort • The use of oral anti-histamines can also be considered for the symptomatic management of pruritus • The use of urea containing creams can be considered for the management of areas affected by hyperkeratosis See Appendix 1 for details of recommended skin care products.

Oral toxicity
Mucositis may also occur as a schedule limiting toxicity in patients receiving cediranib.
General supportive measures to prevent/ minimise oral toxicity include the following: • The gentle brushing of teeth and the use of non-alcohol based mouthwashes and toothpastes Page no.55 The syndrome can present in a variety of non-specific ways, including headache, seizures, lethargy, confusion, blindness and other visual and neurological disturbances. Hypertension may be present, but is not necessary for the diagnosis of RPLS. MRI is the most sensitive imaging modality to detect RPLS and is recommended in suspected cases to confirm the diagnosis. RPLS is reversible upon removal of any possible precipitating factors and control of hypertension.
Active management of hypertension according to the hypertension management guidelines in this section may be expected to reduce the incidence of RPLS. However, if any case of RPLS occurs that is confirmed by imaging (CT or MRI), cediranib should be immediately discontinued in addition to any other measures to alleviate symptoms and control blood pressure.

Fatigue
Fatigue experienced by patients taking cediranib and/or gefitinib may be rapid in onset. During appointments patients should be specifically interrogated about fatigue, and estimated fatigue levels (according to CTC) should be recorded.
Patients should seek medical advice early if Grade 2 fatigue develops (moderate fatigue causing difficulty performing some activities of daily living).

Fatigue management
• In case of clinically significant fatigue in the investigators assessment, the patient may consider short treatment breaks (cediranib and gefinitib) of 2-3 days (max. 14 days allowed (21 days in consultation with the CI)). Care should be taken to ensure that the patients' nutritional status is optimised.
• Patients should be encouraged to drink plenty of fluids (1.5L/day-sips).
• Patients should be encouraged to manage fatigue by alternating periods of rest with frequent light exercise, which may improve the symptoms in some cases.
• Patients should restart treatment when symptoms have improved.

Management of toxicity attributable to gefitinib
Dose interruptions should be used as the first approach to managing toxicity and dose reduction may be considered.
The events for which de-escalation of the gefitinib dose is recommended include: • CTCAE grade 3 or higher toxicities of duration >3 days that are considered to be related to study treatment that are not responding to maximal supportive care within 48 hours, at the discretion of the investigator.
With the EXCEPTION of Page no.57 One dose reduction for the gefitinib treatment will be permitted during the study. No re-escalation of dose is permitted.
A maximum of a 14-day (21-day in consultation with the CI) delay in dosing for gefitinib is permitted.
If a longer interruption is required due to unresolved toxicity, gefitinib should be discontinued.
In addition, there are certain circumstances in which study medication should be permanently discontinued (see section 13.1).

Skin toxicity
Skin rash is a commonly reported outcome of treatment with gefitinib (Herbst et al. 2003;Wacker et al. 2007). Typically the rash comprises an acneiform dermatitis but also includes xerosis (dry skin), hyperpigmentation, telangiectasia, nail and hair changes. Gefitinib should be interrupted or discontinued in patients that develop severe bullous, blistering or exfoliating conditions. It is recommended that patients use sunscreen and limit sun exposure while receiving gefitinib as sunlight can exacerbate any skin reactions that may occur. See Appendix 1 for a list of recommended skin care products.

Interstitial Lung Disease
The emerging safety profile of the gefitinib, in non-small cell lung cancer patients, suggests an overall treatment related incidence of Interstitial Lung Disease (ILD) of 1% (Cohen et al. 2003). Almost a third of the cases of ILD have been associated with fatal outcomes. The reported incidence of gefitinib associated ILD has been around 2% in the post-marketing experience in Japan, around 0.3% in approximately 23,000 patients treated with gefitinib in a US expanded access program and about 1% in studies utilising gefitinib in the first-line treatment of NSCLC (with similar rates of ILD in both treatment and placebo groups). In the Iressa Pan Asia (IPASS) study in NSCLC, 2.6% of the gefitinib arm suffered ILD compared to 1.4% of the comparator arm (carboplatin/paclitaxel).
Patients with concurrent idiopathic pulmonary fibrosis whose condition worsens while receiving gefitinib have been observed to have an increased mortality compared to those without concurrent idiopathic pulmonary fibrosis (Naito et al. 2008).
Patients typically present with acute dyspnoea that progresses relatively rapidly and invariably requires inpatient admission. In some instances patients develop associated symptoms of cough and low-grade fever.
Chest radiographs are found to be normal in 10% of patients with ILD and therefore there should be a low threshold for the further investigation of the following: • Patients with symptoms of progressive breathlessness, exertion (dyspnoea) +/-a persistent non-productive cough.
• Patients with an abnormal chest radiograph

Diarrhoea
Both cediranib and gefitinib are associated with diarrhoea and therefore action should be taken to manage diarrhoea as soon as symptoms develop.
The recommendations for diarrhoea management in this study are based on guidelines from the American Society of Clinical Oncology. The guidelines suggest the aggressive management of 'treatment induced' diarrhoea to ensure that complications are avoided and that further treatments can be administered without delays (Benson et al. 2004). • Patients should be advised to stop the cediranib and gefitinib study tablets and care should be taken to prevent dehydration.
• Following evaluation, consider ciprofloxacin for 7 days particularly if the patient has a fever or is neutropenic (although ciprofloxacin is a potential CYP450 inhibitor (see Appendix 3), in this instance it will be given whilst the patient is not on study medication).
• Consider infectious causes and aetiologies such as Clostridium difficile/viral gastroenteritis.

Management of persistent (>48h) diarrhoea despite loperamide
• The physician or study nurse should see patients in this situation.
• Hospitalisation and IV fluids may be needed.
• Consider infectious causes and aetiologies such as C-diff/viral gastroenteritis.
• Consider ciprofloxacin for 7 days particularly if the patient is neutropenic or has a fever.

Dose modifications for cediranib and gefitinib for diarrhoea
Diarrhoea which is Grade 1 in severity can be managed symptomatically without necessitating interruption of the trial drugs. Diarrhoea which is ≥ Grade 2 in severity will require the trial drugs to be interrupted and restarted when the diarrhoea resolves to Grade 1 or less, with or without a dose reduction as shown in Table 10. Dose reductions of cediranib and gefitinib will be made depending on the current dose of each agent according to the flowchart in Figure 1, giving preference to the reduction of cediranib prior to gefitinib.   The cediranib should be re-introduced first and dose modifications made to find a tolerable dose before the gefitinib is re-introduced.

Renal Impairment
From the Iressa SmPC (dated 16.07.2009): No dose adjustment is required in patients with impaired renal function at creatinine clearance >20ml/min. Only limited data are available in patients with creatinine clearance ≤20 ml/min and caution is advised in these patients.
No advice concerning dose adjustments in patients with impaired renal function is given in the Cediranib Investigators Brochure (dated 08.11.10).
For the purposes of this trial it is recommended that patients with a creatinine clearance of <30ml/min have a dose interruption for both trial drugs. Dose modifications could be considered depending on the cause of impairment.

Hepatic Impairment
Increases in transaminases, which are sometimes associated with increases in total bilirubin, have been seen in cediranib clinical studies.
Patients with ≥5x ULRR AST or ALT tests, or with ≥1.5 x ULRR serum bilirubin (except for patients with known documented cases of Gilbert's Syndrome) should dose interrupt both drugs and reintroduce at investigators discretion until cause established. Liver function tests should be monitored regularly (timing at investigators discretion). In the event of elevated bilirubin, direct and indirect bilirubin fractions should be measured.
Possible alternative causes for the abnormalities should also be investigated and documented, including concomitant medications, concomitant medical conditions, alcohol or IV drug abuse, blood transfusions, and recent travel abroad. As appropriate, tests may be performed to identify possible causes of the liver injury, such as liver ultrasound/computed tomography or MRI; serology and molecular virology; clotting screen and serum albumin; and autoantibody screening.
Caution should be exercised in the use of the following drugs, which are not recommended unless deemed essential for treatment: • Medications which may markedly affect renal function (eg, vancomycin, amphotericin, ibuprofen, pentamidine) • Any medication that may significantly affect hepatic P450 drug metabolising activity by way of enzyme induction or inhibition, or which require P450 for metabolism (from 2 weeks prior to the first dose of cediranib/gefitinib) (See Appendix 3 for a list) • Medicinal products that cause significant sustained elevation in gastric pH, such as proton pump inhibitors and h2 antagonists.
• Antacids if taken regularly close in time to administration of gefitinib • Oral anticoagulants (coumadin, warfarin) or low molecular weight heparin Other medication, which is considered necessary for the patient's safety and well-being, may be given at the discretion of the investigator(s). The administration of all medication (including investigational products) must be recorded in the appropriate sections of the CRF.

Methods for ensuring blinding
The randomisation schedule that provides details of individual patient treatment will be produced by computer software that incorporates a standard procedure for generating random numbers. All study personnel will be unaware of the randomised treatment until all decisions regarding the integrity and evaluability of the data from all patients have been made and documented. The gefitinib study medication will be provided as beige, round, film-coated tablets. The active and placebo tablets for each strength will be identical. Trial medication will be labelled using a unique material pack code which is linked to the randomisation procedure. Packaging, labelling and preparation of the study medication will be performed in a way that will ensure blinding throughout the study.
The treatment groups are double blind.

Methods for unblinding
The IWRS will hold information indicating the treatment randomisation for each randomised patient; this will be available to the investigator(s) or pharmacists via the website.
The treatment code must not be broken except in medical emergencies when the appropriate management of the patient necessitates knowledge of the treatment randomisation.
The investigator(s) must document and report to UCL CTC any breaking of the treatment code. UCL CTC retains the right to break the code for SAEs that are unexpected and are suspected to be causally related to an investigational product and that potentially require expedited reporting to regulatory authorities.
If an event causes concern, the dose of the trial drug should be reduced or discontinued according to this protocol, and the event treated.
Please note that once a patient has been unblinded, they will no longer have access to trial treatment.
Unblinding should occur following one of the two processes outlined below. For any unblinding during working hours, individual treatment codes, indicating the treatment randomisation for each patient, will be available from the IWRS and accessible by the PI and/or site pharmacist, including email notification. Further procedures for these processes will be outlined in the IWRS user manual that will be provided to sites prior to trial activation.

Unblinding outside of working hours
If a patient's randomised treatment requires unblinding outside working hours i.e. outside 9am-5pm UK time Monday-Friday (and additionally UK bank holidays), sites are permitted to access the IWRS to unblind, but with the provisos outlined above i.e. that unblinding must only be undertaken where knowledge of the randomised treatment is essential to patient safety/management.
Treatment codes will not be broken for the planned analyses of data until all decisions on the evaluability of the data from each individual patient have been made and documented. The treatment arms of patients with SARs thought to be unexpected for the trial medication (potential SUSARs) will be viewed in an unblinded fashion internally at UCL CTC where required for SUSAR reporting only, in these instances site staff will not be unblinded.

hour Emergency Medical Advice
A telephone number for the AstraZeneca switchboard, which will be open 24 hours a day, 365 days a year, will be listed on the Trials Contacts list in the investigator site file. Sites should use this number in the event that emergency medical advice is required in events that the investigator is unavailable or unable to advise.

UCL CTC must be contacted BEFORE unblinding*:
DORIC Trial Coordinator 0207 679 9860 Doric@ctc.ucl.ac.uk *except in cases of out-of-hours emergencies

Other trial treatments/interventions: Imaging
Central to the collection of the outcomes of this trial is regular MRI brain scans at screening/baseline and 6-weekly thereafter.
The same MRI scanner should be used for all images for the patient throughout the trial where possible. There should not be any switching between magnets of different field strengths.
The screening/baseline scan should be performed within 7 days prior to the randomisation date. This scan will be used to determine whether or not the patient is eligible for the trial. Routine scans prior to baseline should be assessed when deciding whether or not a patient should be screened for the trial, but will not be collected for this trial.
The follow up scans must be performed +/-10 days of the due date (every 6 weeks after the start of trial treatment), irrespective of the trial treatment administration. Follow up scans should be consistent with baseline scans with regards to modality, scanner and all associated scanning parameters.
If a patient discontinues from the trial treatment, the discontinuation MRI should be completed within 10 days of discontinuation (if possible).
Patients must be on a stable dose of steroids (dexamethasone) for at least 5 days prior to baseline scan and at least 10 days prior to all follow-up scans (where possible).
The following images will be taken on a 1.5 or greater Tesla MR scanner: Week 1 Week 2 Week 3 Week 4 Week 5 Week 6  Week 1

Schedule of Assessments -further details a) Physical Examination
Including height, weight, lying and 5 minute standing blood pressure, ECOG performance status.
Patient's blood pressure needs to be monitored weekly for the first 6 weeks of treatment of cediranib +/-gefitinib. Patients/carers will monitor the patient's blood pressure daily and record this in the patient diaries. CRFs should reflect an average of the patient-reported daily values. If a patient has BP ≥150/100 mmHg at Screening, a further reading should be taken at least 24 hours later to assess eligibility. In event of hypertension, refer to hypertension management plan, see Section 7.4.1.4.

b) Baseline measurements
Lab, TSH and ECG measurements should be retested at baseline only if the screening period is greater than 7 days. Labs and ECGs must be evaluated within 7 days prior to randomisation to confirm eligibility.

c) Urinalysis tests
Urinalysis-if a patient has two consecutive one plus (+) urine protenuria dipsticks measurements at baseline, please measure urine protein/creatinine ratio (

e) Pregnancy test
Women of child-bearing potential (see section 5.1.3) must have a negative urine or serum pregnancy test at the screening visit within 14 days prior to starting trial treatment.

f) MRI timing and guidance
Baseline contrast enhanced MRI must be completed within 7 days prior to randomisation according to acquisition guidelines. Follow-up scans should be performed at scheduled time points ± 10 days (every 6 weeks from the baseline date), independent of when the follow up assessment was completed or trial treatment administered. For acquisition guidelines see Section 7.7 (Imaging).
Protocol Template FINAL version 1.0 29Jan10 Page no.72 Patients must be on a stable dose of steroids (dexamethasone) for at least 5 days prior to baseline scan and at least 10 days prior to all follow-up scans. At discontinuation the MRI should be performed within 10 days of discontinuation where possible.
For patients who are withdrawn from study treatment for reasons other than disease progression every effort will be made to continue with collecting the trial MRI scans every 6 weeks.

g) Adverse event collection duration
Adverse events will be collected from the time informed consent is given, throughout the treatment period and up to and including the 30 days after last dose of study medication. AEs thought to be related to trial treatment will be followed up until resolution or death. Any ongoing SAE at discontinuation must be monitored until resolution. SAEs occurring after 30 days following the last dose of study drug do not need to be reported to UCL CTC unless thought to be related to the study treatment.

h) Follow up assessment schedule after discontinuation
If a patient discontinues due to reasons other than progression, the patient will continue to be followed-up every 6 weeks until progression or death. If a patient discontinues due to progression, or has progressed after discontinuation, the patient will continue to be followed-up every 12 weeks for survival (via telephone call). All efforts should be made by the Site to contact the patient's GP to assess their condition, if a patient fails to attend a clinic or cannot be followed up at site. The trial will be registered with the NHS Information Centre in the event that any patients become Lost to Follow Up.
Neurological status assessments should be continued until neurological deterioration has been documented on 2 consecutive 6 weekly assessments. Dexamethasone dose per day should be collected until progression documented (as outlined in section 2.6.1.1)

i) Blood samples for biomarkers
Samples must only be taken after provision of separate consent for genetic sampling and ideally prior to administration of investigational product. Blood samples taken at discontinuation should have a record of the time passed since last dose of trial treatment. See section 21 for details of samples.

j) Post-mortem samples
Post-mortem tumour samples should be arranged if the patient has discontinued due to death, and the deceased patient has given explicit consent for this. See section 21.

k) Assessment dates during trial treatment
The cycle start assessment should be carried out on the day the patient begins the next cycle of study treatment. Clinic assessments and MRI assessments will occur at the specified time points irrespective of trial treatment holidays. E.g a 6 week cycle may only contain 3 or 4 weeks of trial treatment if a patient has had a trial treatment holiday during that cycle. If either treatment is delayed for more than 14 days due to toxicity (or 21 days if the CI has permitted a 21 day treatment holiday), the patient should cease taking the medication and continue with the cycles of the other study treatment. If both the study treatments are delayed for more than 14 days due to toxicity (or 21 days if the CI has permitted a 21 day treatment holiday), the discontinuation assessment should be performed and the patient withdrawn from the trial.

l) Time between screening and baseline
A maximum of 14 days between screening and baseline is permitted. However for patients switching from enzyme inducing anti-epileptic drugs to non-enzyme inducing anti-epileptic drugs, this can be extended to a maximum of 21 days before baseline (as a wash-out period of 2 weeks is required. See 5.1.2 Exclusion criteria). Screening assessments must be performed within 7 days before randomisation, as per section 5.2 (pre-randomisation evaluation). If more than 14 days pass between screening and baseline, the randomisation form must be updated with prerandomisation tests completed within 14 days of the baseline visit and this reassessed for eligibility at UCL CTC before the patient proceeds with the baseline assessments.

m) Concomitant medication collection after discontinuation
After discontinuation -concomitant medication for treatment of brain tumour only should be collected (e.g radiotherapy, chemotherapy, investigational drugs).

Patient diaries
Patients will be provided with a diary and encouraged to record daily blood pressure readings, compliance in taking study medication, and dexamethasone dose taken. These will be read by the site staff at assessment visits to assist with completing the CRFs. Patient diaries should be completed until progression is documented and returned to the site staff at each clinic visit.

Blood pressure monitors
Patients with be provided with a blood pressure monitor at the beginning of the trial (supplied to the site by either the Chief Investigator or the UCL CTC trial coordinator at site activation).
Protocol Template FINAL version 1.0 29Jan10 Page no.75

Data Management Guidelines
Data will be collected from sites on version controlled case report forms (CRFs) designed for the trial and supplied by UCL CTC. Data entered onto CRFs must reflect source data at site.

Completing Case Report Forms
All CRFs must be completed and signed by staff who are listed on the site staff delegation log and authorised by the PI to perform this duty. The PI is responsible for the accuracy of all data reported in the CRF.
Once completed the original CRFs must be sent to UCL CTC and a copy kept at site. All entries must be clear and legible. The use of abbreviations and acronyms must be avoided.

Corrections to CRFs
Any corrections made to a CRF at site must be made by drawing a single line through the incorrect item ensuring that the previous entry is not obscured. Each correction must be dated and initialled.
Correction fluid must not be used. The amended CRF must be sent to UCL CTC and a copy retained at site.

Missing Data
To avoid the need for unnecessary data queries CRFs must be checked at site to ensure there are no blank fields before sending to UCL CTC. When data is unavailable because a measure has not been taken or test not performed, enter "ND" for not done. If an item was not required at the particular time the form relates to, enter "NA" for not applicable. When data are unknown enter the value "NK" (only use if every effort has been made to obtain the data).

Data Queries
Data arriving at UCL CTC will be checked for completeness, accuracy and consistency. Queries on incomplete, inaccurate or inconsistent data will be sent to the data contact at site. When responding to a query, site staff should attach an amended copy of the case report form held at site and send to DORIC DORIC protocol version 1.0 30.11.2010 Protocol Template FINAL version 1.0 29Jan10 Page no.76 UCL CTC, keeping a copy at site. All amendments must be initialled and dated. All data which requires query will be queried at least once

Timelines for data return
Sites must complete and return CRFs to UCL CTC as soon as possible after patient visit and within one month of the patient being seen. The registration form should be faxed to the UCL CTC asap when completed to enable randomisation (see section 6.0). SAE reports should be sent via fax according to the timelines in section 10.2.2 (SAEs).
Sites who persistently do not return data within the required timelines may be suspended from recruiting further patients into the trial by UCL CTC.

Adverse Event (AE)
Any untoward medical occurrence or effect in a patient treated on a trial protocol, which does not necessarily have a causal relationship with a trial treatment. An AE can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding), symptom or disease temporally associated with the use of a trial treatment, whether or not related to that trial treatment.

Adverse Reaction (AR)
All untoward and unintended responses to a trial treatment related to any dose administered. A causal relationship between a trial treatment and an adverse event is at least a reasonable possibility, i.e. the relationship cannot be ruled out.

Serious Adverse Event (SAE) or Serious Adverse Reaction
An adverse event or adverse reaction that at any dose: • Results in death • Is life threatening (the term "life-threatening" refers to an event in which the patient was at risk of death at the time of the event. It does not refer to an event that hypothetically might have caused death if it were more severe) • Requires in-patient hospitalisation or prolongs existing hospitalisation • Results in persistent or significant disability/incapacity • Is a congenital anomaly or birth defect • Is otherwise medically significant (e.g. important medical events that may not be immediately life-threatening or result in death or hospitalisation but may jeopardise the patient or may require intervention to prevent one of the other outcomes listed above)

Suspected Unexpected Serious Adverse Reaction (SUSAR)
A serious adverse reaction, the nature or severity of which is not consistent with the applicable trial treatment information.

All Adverse Events (AEs)
All adverse events that occur between informed consent and 30 days post last trial treatment administration must be recorded in the patient notes and the trial CRFs. Information regarding dates of event onset and resolution, outcome, severity and causality for the trial treatment must be recorded. AEs thought to be caused by trial treatment will be followed up until resolution or death.
Those meeting the definition of a Serious Adverse Event (SAE) must also be reported to UCL CTC using the trial specific SAE Report (see section 10.2.2 (Serious Adverse Events) for details).
Pre-existing conditions do not qualify as adverse events unless they worsen.

Overdoses
All accidental or intentional overdoses, whether they result in an adverse event or not, must be recorded in the patient notes and CRFs. Overdoses resulting in an adverse event are classified as SAEs and must be reported to UCL CTC according to SAE reporting procedures. 'Other medically significant event' must be ticked as the classification of serious and the fact that an overdose has occurred must be clearly stated. See also section 10.2.2 (Serious Adverse Events).
Sites must inform UCL CTC immediately when an overdose has been identified. See also section 11.0 (Incident Reporting and Serious Breaches).

Adverse Event Term
An adverse event term must be provided for each adverse event, preferably using the term as listed In those cases where the CTCAE criteria do not apply, severity should be coded according to the following criteria: 1 = Mild (awareness of sign or symptom, but easily tolerated) 2 = Moderate (discomfort enough to cause interference with normal daily activities) 3 = Severe (inability to perform normal daily activities) 4 = Life threatening (immediate risk of death from the reaction as it occurred) 5 = Fatal (the event resulted in death)

Causality
The PI, or other delegated site investigator, must perform an evaluation of causality for each adverse event.
As this is a placebo-controlled trial, the evaluation of causality must be performed as if the patient is on active treatment.
Causal relationship to each trial treatment must be determined as follows: •

None
There is no evidence of any causal relationship.

Unlikely
There is little evidence to suggest a causal relationship (e.g. because the event did not occur within a reasonable time after administration of a trial treatment). There is another reasonable explanation of the event (e.g. the patient's clinical condition, other concomitant treatments).

Possible
There is some evidence to suggest a causal relationship (e.g. because the event occurs within a reasonable time after administration of a trial treatment). However, the influence of other factors may have contributed to the event (e.g. the patient's clinical condition, other concomitant treatments). •

Probable
There is evidence to suggest a causal relationship and the influence of other factors is unlikely. There is clear evidence to suggest a causal relationship and other possible contributing factors can be ruled out.

Serious Adverse Events (SAEs)
All SAEs that occur between the signing of informed consent and 30 days post the last trial treatment administration (or after this date if the site investigator feels the event is related to the trial treatment) must be submitted to UCL CTC by fax within 1 business day of observing or learning of the event, using the trial specific SAE Report. All sections on the SAE Report must be completed. If the event is not being reported within 1 business day to UCL CTC, the circumstances that led to this must be detailed in the SAE Report to avoid unnecessary queries.

Events which do not require immediate reporting on an SAE Report
The following events do not require immediate reporting on an SAE Report for this trial, but must be recorded in the relevant section(s) of the CRF: • disease progression* • disease related deaths* * Investigators should take care to distinguish these from side effects of the trial drugs (e.g bleeds).
Please note that hospitalisation for elective treatment or palliative care, or problems with home carers, does not qualify as an SAE.

SAE Follow-Up Reports
All SAEs must be followed-up until resolution, and until there are no further queries. The PI, or other delegated site investigator, must provide follow-up SAE Reports if the SAE had not resolved at the time the initial report was submitted.

SAE Processing at UCL CTC
On receipt of the SAE Report, UCL CTC will check for legibility, completeness, accuracy and consistency. Expectedness will be evaluated, to determine whether or not the case qualifies for expedited reporting, using the list of expected adverse events in the current IB for cediranib and SmPC for gefitinib.
The CI, or their delegate (e.g. a clinical member of the TMG), will be contacted to review the SAE and to perform an evaluation of causality on behalf of UCL CTC. If UCL CTC has considered expectedness difficult to determine, the CI, or their delegate, will be consulted for their opinion at this time.
UCL CTC will send line-listings of all SAEs to AstraZeneca every 3 months.

Emergency unblinding of the treatment allocation
See section 7.6 for details.

SUSARs
If the event is evaluated by either the site or UCL CTC as a Suspected Unexpected Serious Adverse

Informing Sites of SUSARs
UCL CTC will inform all PIs of all potential SUSARs that occur on the trial. PIs will receive a quarterly line listing which must be processed according to local requirements.
UCL CTC will forward reports regarding SUSARs that have occurred on other trials using cediranib to all PIs. These must be processed according to local requirements and filed with the IB.

Clinical Review
UCL CTC will provide safety information to the TMG and IDMC on a periodic basis for review. Should the outcome of the review result in upgrading/downgrading of SAEs to SUSARs and vice versa, UCL CTC will provide relevant reports to the MHRA and the REC.

Additional Safety Monitoring at UCL CTC
UCL CTC will also monitor safety data for any trial related events that are not considered related to the trial treatment. Should any trial procedures appear to be resulting in adverse events, the Trial Management Group will be contacted for their opinion. If it is declared necessary to review the conduct of the trial, UCL CTC will inform the MHRA and the REC as appropriate.
If UCL CTC detects a higher incidence in rare events than is stated in the IB/SmPC for the trial treatment, a report detailing the finding will be submitted to the MHRA and REC.

Pregnancy
If a female patient or a female partner of a male patient becomes pregnant at any point during the trial, a completed trial specific Pregnancy Report must be submitted to UCL CTC by fax within 1 business day of learning of its occurrence. Consent to report information regarding the pregnancy must be obtained from the mother. The Pregnancy Monitoring Information Sheet and Consent Form for the partners of trial patients must be used for this purpose.

SAEs During Pregnancy
Any SAE occurring in a pregnant patient must be reported using the trial specific SAE Report, according to SAE reporting procedures. See section 10.2.2 (Serious adverse events) for details.

Annual Safety Reports
UCL CTC will submit Annual Safety Reports to the MHRA and REC. This will commence one year from the date of CTA approval obtained for the trial. The final Annual Safety Report will be submitted in the year following trial closure. See section 14.1 (End of Trial) for details.

Incident Reporting
Organisations must notify UCL CTC of all deviations from the protocol or GCP immediately. UCL CTC may require a report on the incident(s) and a form will be provided if the organisation does not have an appropriate document (e.g. Trust Incident Form for UK sites).
UCL CTC will assess all incidents to check if they meet the definition of a serious breach.

Serious Breaches
Systematic or persistent non-compliance by a site with GCP and/or the protocol, including failure to report SAEs occurring on trial within the specified timeframe, may be deemed a serious breach.
In cases where a potential or actual serious breach has been identified, UCL CTC will inform the MHRA within 7 calendar days of becoming aware of the breach.

Trial Monitoring and oversight
Participating sites and Principal Investigators must agree to allow trial-related on-site monitoring, including Sponsor audits and regulatory inspections by providing direct access to source data/documents as required. Patients are informed of this in the patient information sheet and are asked to consent to their medical notes being reviewed by UCL CTC on the consent form.

On-site monitoring
The degree of on-site monitoring will be proportionate to the objective, purpose, phase, design, size, complexity, blinding, endpoints and risks associated with the trial. UCL CTC will determine the appropriate level and nature of monitoring required for the trial. Risk will be assessed on an ongoing basis and adjustments made accordingly.
Sites will be sent a letter in advance confirming that a routine monitoring visit is due, and including a list of the documents that are to be reviewed, interviews that will be conducted, planned inspections of the facilities, who will be performing the visit and when the visit will likely occur.
On-site monitoring will be conducted by the UCL CTC Trial Monitor/Trial Coordinator. The frequency and level of monitoring will be specified in the Trial Monitoring Plan, which will be kept in the Trial Master File and regularly updated in accordance with the aforementioned risk assessments. However, the following key areas will form the core of any on-site monitoring that occurs: consent; eligibility; capturing and reporting of Serious Adverse Events; capturing of trial endpoints; data queries; filing of essential documents in the Investigator Site File and/or trial-specific patient files; drug accountability.

Monitoring report
Following a monitoring visit, the Trial Monitor/Trial Coordinator will provide a report to the site, which will summarise the documents reviewed and a statement of findings, deviations, deficiencies, conclusions, actions taken and actions required. The Principal Investigator at each site will be responsible for ensuring that monitoring findings are addressed (this may be delegated to an appropriate member of staff).
Protocol Template FINAL version 1.0 29Jan10 Page no.87

Self-assessment monitoring
Participating sites may be requested to complete a self-assessment monitoring report periodically, at a frequency detailed in the Trial Monitoring Plan. This report may include, but is not limited to, Investigator Site File and Pharmacy File document version checklists, progress of local approval for amendments and SAE review.
Responses to self-assessment monitoring will be reviewed at the UCL CTC to identify areas of noncompliance/fraud and to indicate training needs. Findings may trigger an on-site monitoring visit.

Central monitoring
Sites will be requested to submit screening logs and staff delegation logs to UCL CTC on request and these will be checked for consistency and completeness.
Eligibility of all patients entered in the trial will be checked by an appropriately trained UCL CTC staff member prior to randomisation.
Data stored at UCL CTC will be checked for missing or unusual values (range checks) and checked for consistency over time. If any problems are identified data queries will be issued to the site. Sites are required to resolve any queries and update the relevant CRF as required. All changes must be initialled and dated. The amended version must be sent to UCL CTC and a copy retained at site.
Copies of the drug accountability logs will be collected at UCL CTC for all trial patients. Sites will be required to submit logs on request. A proportion of these will be monitored centrally to ensure completeness and correlation with data captured in the CRF.
Patients will be flagged using the NHS Information Centre (please see section 8.2 for further details).
MRI scans will be submitted to UCL CTC for retrospective review for the trial, conducted by imaging experts from University College London Hospital, led by a member of the Trial Management Group. Page no.88 sent a letter in advance outlining the reason(s) for the visit, a list of the documents that are to be reviewed, interviews that will be conducted, planned inspections of the facilities, who will be performing the visit and when the visit will likely occur. UCL CTC will assess whether it is appropriate for the site to continue participation in the trial and whether the incident(s) constitute a serious breach. See section 11.0 (Incident Reporting and Serious Breaches) for details.

Trial Management Group (TMG)
The TMG will include the Chief Investigator, clinicians and experts from relevant specialities and DORIC trial staff from UCL CTC (see page 1). The TMG will be responsible for overseeing the trial.
The group will meet regularly and will send updates to Principal Investigators (via newsletters or at Investigator meetings) and to the NCRI Brain Tumour Clinical Studies Group.
The TMG will agree protocol amendments on behalf of the PIs prior to submission to the REC and MHRA. All PIs will be kept informed of substantial amendments through their nominated responsible individuals.

Independent Trial Steering Committee (ITSC)
The role of the ITSC is to provide overall supervision of the trial and ensure that it is conducted in accordance with GCP and the Protocol. The ITSC will review the recommendations of the Independent Data Monitoring Committee and, on consideration of this information, recommend any appropriate amendments/actions for the trial as necessary. The ITSC acts on behalf of the funder and Sponsor.

Independent Data Monitoring Committee (IDMC)
The role of the IDMC is to provide independent advice on data and safety aspects of the trial.
Meetings of the Committee will be held regularly (yearly or based on any safety considerations or lack of efficacy considerations). The IDMC is advisory to the ITSC and can recommend premature closure of the trial to the ITSC. The IDMC charter will define the specific responsibilities of the IDMC.

Role of UCL CTC
UCL CTC will be responsible for the day to day coordination and management of the trial and will act as custodian of the data generated in the trial (on behalf of UCL). UCL CTC is responsible for all duties relating to pharmacovigilance which are conducted in accordance with section 10.0 (Pharmacovigilance).

Withdrawal of patients
In consenting to the trial, patients are consenting to trial treatment, assessments, follow-up and data collection.

Withdrawal from Trial Treatment
A patient may be withdrawn from all trial treatment whenever continued participation is no longer in the patient's best interests, but the reasons for doing so must be recorded. Reasons for discontinuing treatment may include: • Voluntary discontinuation by the patient who is at any time free to discontinue his/her participation in the study, without prejudice to further treatment. At the time of discontinuation, the investigator should establish if the patient is willing to be followed for disease progression.
• Safety reasons as judged by the investigator and/or UCL CTC • Severe non-compliance to protocol as judged by the investigator and/or UCL CTC • Incorrect enrolment (i.e the patient does not meet the required inclusion/exclusion criteria) of the patient if there are safety reasons as judged by the investigator and/or UCL CTC • Patient lost to follow-up • Disease progression (unless, in the investigator's opinion the patient is receiving benefit from study treatment. It is not compulsory to stop trial treatment following radiological progression only. Where progression is radiologically determined only, a second confirmatory scan taken 6 weeks later should be assessed before withdrawal, to rule out pseudoprogression).
• Pregnancy • Toxicity grade 3 or above causing an IMP treatment delay for >14 days (or >21 days if CI has permitted a 21 day treatment holiday), or any other unacceptable toxicity • Patients should not receive trial drugs within 2 weeks of major surgery. Where possible, trial drugs should be suspended for 2 weeks to allow clearance before major surgery.
• Interstitial lung disease (ILD) • Severe change in liver function In these cases patients should continue to have trial assessments for the purposes of follow-up. If a patient wishes to withdraw from trial treatment, sites should explain the importance of remaining on trial follow-up, or failing this of allowing routine follow-up data to be used for trial purposes and for allowing existing collected data to be used.

Procedures for discontinuation from treatment
Patients who discontinue the trial treatment should always be asked about the reason(s) for their discontinuation and the presence of any adverse events. If possible, they should be seen and assessed by an investigator(s). The patient should return investigational products.
If all study treatment is stopped during the study, the Principal Investigator/Sub investigator will perform the best possible observation(s), test(s) and evaluation(s) as well as give appropriate treatment and all possible measures for the safety of the patient. In addition, they will record on the Case Report Form (CRF) the date of withdrawal, the reasons, and treatment at the time of withdrawal.
If only one study treatment is stopped during the study, the patient will continue with the trial assessments as before. Reasons for stopping the first drug must be recorded on the CRF. The discontinuation assessment is only to be carried out upon discontinuation of both trial drugs. At withdrawal, all on-going SAEs must be followed until resolution, unless in the Investigator's opinion, the condition is unlikely to resolve due the patient's underlying disease.
All new AEs occurring for up to 30 days after the last dose of any study medication must be recorded in the CRF. Ongoing AEs thought to be caused by trial treatment will be followed up until resolution or death. All SAEs occurring up to 30 days after the last dose of any study medication must be followed until resolution.
Patients who are not receiving treatment, but are being followed up for tumour assessments should also be followed up for procedure related SAEs, these should be reported on an SAE report form (see section 10.2.2 SAEs). AE collection beyond the 30 days post treatment is not required for patients who have progressed and are being followed up for survival only.
For patients who are withdrawn from study treatment for reasons other than disease progression every effort will be made to continue with collecting the trial MRI scans every 6 weeks. Follow up clinic assessments should be carried out every 6 weeks as detailed in section 8.
If disease progression has been documented and the patient has withdrawn from study treatment, the long-term follow up information should be collected at least every 12 weeks by telephone contact with the patient, patient's family, or by contact with the patient's physician (see section 8). Please note that withdrawal from trial treatment is not compulsory for patients with disease progression, if the investigator feels that the patient is receiving benefit from trial treatment.

Withdrawal of Consent to Data Collection
If a patient explicitly states they do not wish to contribute further data to the trial their decision must be respected and recorded on the relevant CRF. In this event details should be recorded in the patient's hospital records, no further CRFs must be completed and no further data sent to UCL CTC.

Losses to follow-up
If a patient moves from the area, every effort should be made for the patient to be followed up at another participating trial site and for this new site to take over the responsibility for the patient, or If a patient is lost to follow-up at a site every effort should be made to contact the patient's GP (if consented) to obtain information on the patient's status. Patients lost to follow up will be tracked by UCL CTC via the NHS Information Centre.

Withdrawal from Genetic Research
A patient may withdraw consent from this genetic research at any time, independent of any decision concerning participation in other aspects of the main study described in this protocol. Voluntary discontinuation by the patient will not prejudice further treatment.

Procedures for discontinuation from genetic aspects of the study
Patients who discontinue from the study should always be asked specifically whether they are withdrawing or continuing their consent for this genetic research. It must be established whether the patient: • Agrees to the genetic sample and any DNA extracted from the sample being kept for genetic research in the future.
• Withdraws consent for the sample to be kept for genetic research in the future and wishes the sample to be destroyed. Destruction of the sample (or the DNA extracted from the sample) will only be possible so long as the particular sample is traceable. In the event that genetic research has already been performed, UCL CTC will retain the results and associated data for regulatory reasons but these will not be used in any subsequent analyses.
Depending on which samples the patient had originally consented to give, the above should be checked for the use of tissue biopsy samples, blood samples and / or post-mortem tumour samples.
The investigator is responsible for providing written notification to UCL CTC of any patient who has withdrawn consent for the use or future donation of any sample taken for genetic research, and specifying which particular samples. UCL CTC will provide written confirmation to the investigator of

End of Trial
For regulatory purposes the end of the trial will be 6 months from the date the last patient finished trial treatment (the day after the date that the last trial drug was taken, i.e the first day that the patient is no longer taking either trial drug (not including treatment holidays of <14 days, or 21 days in consultation with the CI. If a patient has a treatment holiday which then is extended to withdrawal from the trial, the date of finishing trial treatment will be the first day of the treatment holiday)), or the date the last patient dies, whichever is sooner, at which point the 'declaration of end of trial' form will be submitted to participating regulatory authorities and ethical committees, as required.
If patients are alive following the end of trial, they will be followed up every 12 weeks for survival as per section 8.1 (Schedule of assessments).

Archiving of Trial Documentation
At the end of the trial, UCL CTC will archive securely all centrally held trial related documentation for a minimum of 5 years. Arrangements for confidential destruction will then be made. It is the responsibility of Principal Investigators to ensure data and all essential documents relating to the trial held at site are retained for a minimum of 5 years after the end of the trial, in accordance with national legislation and for the maximum period of time permitted by the site.
Essential documents are those which enable both the conduct of the trial and the quality of the data produced to be evaluated and show whether the site complied with the principles of Good Clinical Practice and all applicable regulatory requirements. UCL CTC will notify sites when trial documentation held at sites may be archived. All archived documents must continue to be available for inspection by appropriate authorities upon request.

Early discontinuation of trial
The trial may be stopped before completion as an Urgent Safety Measure on the recommendation of the ITSC or IDMC (see section 12.5.2 ITSC and section 12.5.3 IDMC). Sites will be informed in

Quality Assurance
All MRI scans used to produce the outcome data for the final analysis will undergo a retrospective central review at the end of the trial as per section 7.7, before analysis. Local treatment decisions will be based on local imaging results.

Sample size calculation Statistical methods and determination of sample size
In the UK there is no standard second line treatment. Lomustine based treatments may be given but have a median PFS of only 2 months (8.32 weeks) (Yung et al. 2000). However, on the basis of the FDA approval for bevacizumab in GBM we suggest that the AVF3708g and NCI 06-C-0064E studies serve as the comparator for outcomes for this phase II study:

AVF3708g and NCI 06-C-0064E studies
The AVF3708g study was a phase II trial in 167 patients (85 treated with single-agent bevacizumab) with glioblastoma at first or second relapse (Friedman, 2010). Of note only the efficacy data from the bevacizumab monotherapy arm (n=85) was used to support FDA drug approval. Findings were supported by the Phase II trial of bevacizumab in 48 patients with relapsed glioblastoma (Study NCI 06-C-0064E) (Kriesl, 2009).
• AVF3708g study -objective response rate was 28.2 % (95% CI: 18.5 -40.3%) • AVF3708g study -median response duration was 5.6 months (95% CI: 3.0-5.8 months) • AVF3708g study -the 6-month progression-free survival (PFS6) rate was 42.6% (97.5% CI: 29.6-55.5%) • AVF3708g study -the 6-month PFS rates for patients at first and second relapse were 46.4% and 27.8% respectively • AVF3708g study -the median PFS in first and second relapse was 4.4 and 3.1 months respectively • AVF3708g study -the median overall survival was 9.2 months (95% CI: 8.2-10.7 months), median overall survival at first and second relapse was 9.1 months and 9.2 months respectively • Study NCI 06-C-0064E -response rate was 35% percent by Macdonald criteria (one CR, 16 PRs) • Assuming a median PFS of 4 months with cediranib (as equivalent to bevacizumab) and an expected median PFS of 6 months on the combination cedirinib plus gefitinib arm a total of 112 patients are required to achieve 110 events in order to detect a hazard ratio of about 0.67 between the two arms.
This assumes an accrual period of 18 months, maximum follow up of 36 months, lambda 1= 0.115, lambda 2=0.1733,and a one sided type I error of 10% with approximate power of 80%. The sample size was generated using Nquery Version 7.0.

Population for analysis
The primary population will be the Intention to Treat (ITT) population defined as all patients randomized irrespective of having been administered at least one dose of randomized study medication . All primary and secondary efficacy endpoints will be analysed using this population.
The safety population will be all patients randomized and having taken at least one dose of randomized treatment. The primary endpoint will also be analysed using this population. This endpoint will be analysed using a Cox proportional hazards model. All stratified covariates will be included in the model as well as additional covariates which are deemed to influence the response.

Statistical Analysis of the primary endpoint
These will be stated a priori in the statistical analysis plan (SAP). The 90% confidence interval for the Hazard ratio will be derived which will provide an estimate of the plausible ranges of the treatment effect. A Kaplan Meier plot of Survival probabilities over time will also be generated. Patients who take any concomitant medications which are likely to influence the primary endpoint will be censored at the date of starting the medication (if available). Patients who have not progressed or died will be censored at the last date of contact.
The primary endpoint will be analysed using both the ITT and Safety Populations. Additional analysis will be conducted as is deemed appropriate and documented in the SAP. An exploratory analysis will also be carried out using the unmodified RANO criteria (Wen et al 2010).

Statistical Analysis of the Secondary Endpoints
The secondary endpoints will be subject to an ITT analysis only.

• Overall survival (OS)
This is defined as time from date of randomization to date of Death due to any cause. This endpoint will be analysed in a similar way to the primary endpoint • Radiographic response rate (RR) at each assessment point The response rates will be summarised by treatment groups in terms of counts and percentages. The following definition of responses will be used.
• • PFS rate at 6 months (24 weeks) Progression is defined in the same way as described for the primary endpoint. This endpoint will be analysed in the same way as the primary endpoint. A comparison of the PFS rate at 6 month/ 24 weeks will be made and the 95% confidence interval will be generated for the difference in the PFS rate adjusted for any covariates. A survival plot will also be generated.
• Overall Survival rate at 12 months.
This will be analysed in the same way as the PFS and OS endpoints (above) • Time to sustained increase in total daily steroid dose by ≥ 8mg/day compared to baseline or sustained nadir (≥ 21 days) This is defined as time from randomization to first increase in dexamethasone dose. An increase in dexamethasone dose is defined as a change from the baseline of ≥8mg/day for at least 21 days (i.e. the increase is sustained for at least 21 days consecutively). The earliest date when the dexamethasone dose was ≥ 8mg/day compared to baseline or sustained nadir in this period will be used for comparison between the treatment groups. Sustained nadir refers to the highest steroid dose taken during a ≥ 21 days period where the patient is consistently taking a lower dose of steroids compared to the baseline dose. Daily dexamethasone dose will be recorded until progression.
This endpoint will be analysed as a time to event endpoint similarly to the OS and PFS. Additional covariates may also be included in the analyses for adjustment.
• Average daily steroid dosage change from baseline until first progression

• Average number of progression/steroid free days
For the above 2 endpoints the mean steroid dosage prior to treatment will be considered as the patient's baseline. The number of progression/steroid free days is the number of days that a patient is known not to have used any steroids (dexamethasone) prior to progression. This will be summarized by treatment group.

• Time to deterioration of neurological status or death
This endpoint will be summarized by treatment group. Neurological status based on investigator's neurological examination prior to and during treatment will be recorded on the CRF.
Patient's last pre-dose assessment will be defined as baseline. Patients without a baseline assessment will be excluded from analysis. Time to neurological status deterioration will be defined as the date of randomization to the date of first neurological status worsening in comparison to baseline (first of 2 confirmatory reports at 2 consecutive visits, 6 weeks apart) as assessed by the clinician, or until date of death, whichever is first.

• EORTC QLQ-C30
Each domain will be summarised. The change from baseline will be summarized. For specific domains, a model based analysis will be undertaken (such as Mixed effects models). The scoring of the items/scales will be as described in the EORTC QOL Scoring Manual. Patients without a baseline assessment will be excluded from the analysis.

(visual disorder, motor dysfunction, communication deficit, and drowsiness)
Patients with recurrent glioblastoma commonly experience neurological symptoms. Therefore, improvement in visual disorders, motor dysfunction, communication deficits and drowsiness between the treatment arms for patients with neurological symptoms relative to baseline will be summarized by treatment group. Improvement in symptoms should also be concurrent with stable or decreasing use of steroids (dexamethasone). Visual disturbances will be assessed using

Statistical Analysis of Safety Data
The following safety endpoints below will all be summarized using appropriate summary statistics.
Changes from baseline may be conducted as is deemed appropriate. Safety data will be analysed and reported on an ongoing basis for IDMC meetings (where data is available).
• Adverse Events (AE) • Vital signs including blood pressure (BP) • Electrocardiogram (ECG) parameters • Laboratory findings (clinical chemistry, haematology, urinalysis, TSH, T3 and T4) • Physical examination including neurological examination Safety data will be summarized as deemed appropriate. Relevant changes from baseline may be reported and summarized by treatment group. SAEs will be summarized by grade and treatment group. Safety data will be analysed and reported on an ongoing basis for IDMC meetings (where data is available).

Stopping criteria
A treatment-related grade ≥3 toxicity rate of 30% in any treatment group along with any other clinical data agreed by the Independent data Monitoring Committee would be used as a guide to consider stopping the trial. As soon as 17 patients out of 56 (30%) in one of either arm suffer a grade ≥3 treatment-related event, this would prompt us to review the safety data more frequently. If at least 24 patients (43%) have a grade ≥3 treatment-related event, this would require the IDMC to consider stopping the trial early for safety reasons. The IDMC will be able to compare the toxicity rates between the treatment arms. All safety data will be reviewed regularly. This will be at least every 6 months by members of the TMG in a blinded fashion and at every IDMC meeting. Should stopping be considered due to safety resons, the IDMC will meet at the earliest convenient time. Blood samples (plasma and serum) will be collected and assessed for biomarkers of disease activity, effect of cediranib and gefitinib, and angiogenesis. Archival tumour samples (from the primary tumour and/or any recurrent disease) will be collected from patients who provide informed consent for the retrospective analysis of biomarkers. Biomarkers will be investigated for possible correlation with key clinical outcomes and for the effects of the study medication.
Appropriate summaries of absolute levels and changes from baseline in soluble biomarkers will be produced, as well as correlates with clinical efficacy. Summaries of biomarkers in archival tumour samples and appropriate correlates will be produced. Additional analysis may be conducted as deemed appropriate.

Handling missing data
Missing data will be left as missing and no imputation will be carried out. Further details where relevant will be outlined in the SAP.

Interim Analysis
No formal interim analysis is planned. However data is reviewed on an ongoing basis for the IDMC. In addition, conditional power estimates may be provided to determine futility of the trial based on the observed and assumed hazard ratio. This will be calculated at 50% of the required events and 75% of the events stated in the sample size section. In general a conditional power of < 20% is suggestive of futility, however clinical judgment will also be taken into account before concluding formal futility of the trial.

Ethical and Regulatory Approvals
In conducting the Trial the Sponsor, UCL CTC and Sites shall also comply with all laws and statutes, as amended from time to time, applicable to the performance of clinical trials including, but not limited to: •

Ethical Approval
The trial will be conducted in accordance with the World Medical Association Declaration of Helsinki

Regulatory Approval
A Clinical Trial Authorisation (CTA) has been granted for the trial.
The trial will be conducted at approved trial sites in accordance with the trial protocol and the terms of the CTA granted by the MHRA.

Local Site Approval
The Lead Comprehensive Local Research Network (CLRN) has given NHS permission following global governance checks. Local checks will be undertaken by local CLRNs associated with individual trial sites.
Evidence of local Trust R&D approval must be provided to UCL CTC prior to site activation. The trial will only be conducted at sites where all necessary approvals for the trial have been obtained.

Protocol Amendments
UCL CTC will be responsible for gaining ethical and regulatory approvals for all amendments made to the protocol and other trial-related documents. Once approved, UCL CTC will ensure that all amended documents are distributed to sites and CLRNs as appropriate.
Site staff will be responsible for acknowledging receipt of documents and for gaining local Trust R&D acknowledgement for all amendments and approval for substantial amendments, and for providing UCL CTC with evidence of this.

Patient Confidentiality & Data Protection
Patient identifiable data, including initials, date of birth, and NHS number will be required for the randomisation process and will be provided to UCL CTC. UCL CTC will preserve patient confidentiality and will not disclose or reproduce any information by which patients could be identified. Data will be stored in a secure manner and UCL CTC trials are registered in accordance with the Data Protection Act 1998 with the Data Protection Officer at UCL.

Indemnity
University College London holds insurance to cover participants for injury caused by their participation in the clinical trial. Participants may be able to claim compensation if they can prove that UCL has been negligent. However, as this trial is being carried out in a hospital, the hospital continues to have a duty of care to the participant of the trial. University College London does not accept liability for any breach in the hospital's duty of care, or any negligence on the part of hospital employees. This applies whether the hospital is an NHS Trust or not. This does not affect the participant's right to seek compensation via the non-negligence route.
Participants may also be able to claim compensation for injury caused by participation in this trial without the need to prove negligence on the part of University College London or another party.
Participants who sustain injury and wish to make a claim for compensation should do so in writing in the first instance to the Chief Investigator, who will pass the claim to the Sponsor's Insurers, via the Sponsor's office.
Hospitals selected to participate in this trial shall provide clinical negligence insurance cover for harm caused by their employees and a copy of the relevant insurance policy or summary shall be provided to University College London, upon request.

Translational Research
An exploratory translational part of the research will look at blood, tumour and post-mortem samples for which additional and optional consent will be obtained from patients.
The objectives of the translational research are: • To investigate the relationship between the effects of cediranib on soluble angiogenesis biomarkers and clinical efficacy • To obtain blood samples for pharmacogenetics (genetic variation giving rise to different response to the IMPs) • To obtain archival tumour biopsy samples for biomarker analysis and potential sub-group stratification for: o IDH1 and 2 (Yan et al. 2009) o MGMT methylation status o EGFR, EGFRvIII and PTEN (Mellinghoff et al. 2005) • To obtain post-mortem tumour samples for research into genomic sequence and/or gene expression patterns associated with response or resistance to cediranib and/or gefitinib.
As consent for this part of the research is optional, it is not known whether sufficient data will be generated to explore all of the above objectives. Patients will consent to donate samples for future research, consent will be generic and enduring and include consent for genetic research. Consent for blood and tumour samples will be separate from post-mortem samples.
Blood samples will be taken at time-points outlined in the schedule of assessments (section 8.1). Staff members taking samples should be appropriately qualified to do so. These will include serum samples and whole blood samples in EDTA. Samples will be stored locally at sites in freezers and collected in batches (remaining frozen) for storage at the laboratory in UCL.
Archival tumour biopsies will be collected from patients where they have given consent for this. Post mortem tumour samples will be collected where patients have given additional consent for this. These samples will be collected and stored at the laboratory in UCL.

DORIC
DORIC protocol version 1.0 30.11.2010 Protocol Template FINAL version 1.0 29Jan10 Page no.111 Sample collection, storage and shipping arrangements will be described in a separate Laboratory Manual. Sample labels will refer to patient trial number, but not to patient names.
Translational research will be retrospective to the trial. Translational research will be in line with the current evolving literature at the end of the trial and available resources and where possible will also include the following: EGFR mutation analysis, amplification of 4q (VEGFR2, PDGFR, c-Kit), and VEGF/VEGFR polymorphisms. Where possible, DNA and/or RNA will be extracted from blood and tumour samples and analysed using methods such as whole-genome sequencing, microarray analysis and gene methylation profiling. Any additional objectives for the translational research will be submitted for additional ethics approval.
Extraction of post-mortem samples will be carried out by local pathologists and will only be available at sites where this has been agreed in the clinical trial site agreement. Families of patients who have consented to donation of post-mortem samples will need to contact the local investigator at the site to arrange this. Post-mortem tissue samples taken will be of the tumour and brain only.  Table 12 Samples proposed for collection for translational research If there might be any problems with sample collection at the site, the DORIC trial coordinator should be contacted prior to site activation.