• castration-resistant prostate cancer;
  • angiogenesis inhibitors;
  • cediranib;
  • AZD2171


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

What's known on the subject? and What does the study add?

  • Recent advances in the treatment of metastatic castration-resistant prostate cancer (CRPC) have resulted in improved outcomes; however, the effects have not proved to be long term, highlighting the need for new therapies, particularly in patients with docetaxel-refractory metastatic CRPC. Angiogenesis has been shown to play an important role in the development and progression of prostate cancer. Although targeting angiogenesis appears to be a rational and therapeutic approach for metastatic CRPC, identifying the appropriate subgroups that may benefit from anti-angiogenic therapy remains a challenge.
  • The study demonstrates the potential use of dynamic contrast-enhanced (DCE)-MRI variables as pharmacodynamic endpoints in predicting the clinical outcomes associated with anti-angiogenic agents such as cediranib. Further investigation into the potential predictive value of DCE-MRI variables as biomarkers for antiangiogenic therapy is warranted.


  • To assess the efficacy and toxicity of cediranib, a highly potent inhibitor of vascular endothelial growth factor receptor tyrosine kinases, in patients with metastatic castration-resistant prostate cancer (CRPC) previously treated with docetaxel-based therapy.

Patients and Methods

  • The study used a Simon two-stage trial design, which required at least two of 12 patients in the first cohort to be progression-free at 6 months.
  • We enrolled a total of 35 evaluable patients who all received cediranib 20 mg orally daily.
  • In a second cohort, 23 additional patients received prednisone 10 mg daily with cediranib.
  • Endpoints included tumour response, progression-free survival (PFS), overall survival (OS), vascular permeability via dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), and toxicity.


  • A total of 59 patients were enrolled, of whom 67% had received two or more previous chemotherapy regimens.
  • Six of 39 patients with measurable disease had confirmed partial responses and one had an unconfirmed partial response.
  • At 6 months, 43.9% of patients were progression-free; the median PFS and OS periods for all patients were 3.7 months and 10.1 months, respectively.
  • We found that the DCE-MRI variables baseline transport constant (Ktrans) and rate constant at day 28 were significantly associated with PFS in univariate analyses, but only baseline Ktrans remained significant when considered jointly.
  • The most frequent toxicities were hypertension, fatigue, anorexia and weight loss; the addition of prednisone reduced the incidence of constitutional toxicities.


  • This study demonstrated that cediranib was generally well tolerated with some anti-tumour activity in highly pretreated patients with metastatic CRPC who had progressive disease after docetaxel-based therapy.


castration-resistant prostate cancer


Eastern Cooperative Oncology Group


progression-free survival


overall survival


dynamic contrast-enhanced MRI


transport constant


rate constant


vascular endothelial growth factor


National Cancer Institute


Response Evaluation Criteria in Solid Tumours


area under the curve measured over 60 seconds


Food and Drug Administration


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Prostate cancer is the most common non-dermatological cancer in men and has the second highest cancer-related death rates in men [1]. Docetaxel and prednisone are currently the standard of therapy for patients with symptomatic metastatic castration-resistant prostate cancer (CRPC). Cabazitaxel and abiraterone are approved as second-line treatment options for patients with post-docetaxel progressive metastatic disease, but the duration of response is generally short-lived with a modest survival benefit; therefore, more effective therapeutic approaches are still needed, especially in the post-docetaxel setting.

Vascular endothelial growth factor (VEGF), a pro-angiogenic factor, and its receptors have been shown to be important in promoting tumour angiogenesis, playing a critical role in prostate cancer development and progression [2-4]. In preclinical models, inhibition of angiogenesis has been shown to be an effective target in CRPC. VEGF expression is observed in prostate tumours and in plasma and urine from patients with metastatic disease, where increased expression is associated with disease progression [5, 6]. The fms-like tyrosine kinase/kinase insert domain receptor receptors are expressed in human prostate cancers and are correlated with higher grade lesions and outcome [7]. While targeting angiogenesis appears to be a rational and therapeutic approach for CRPC [8], recent phase III trial results of VEGF pathway inhibitors bevacizumab [9] and sunitinib [10] have shown no clinical benefit, suggesting the need for predictive biomarkers to identify appropriate subgroups that may more likely benefit from this targeted therapy.

Cediranib (AZD2171) is a potent oral small molecule inhibitor of VEGFR-1, VEGFR-2 and VEGFR-3 [11, 12] and also of c-kit, to a lesser extent [12]. Cediranib has been investigated in clinical trials with multiple cancers including colorectal [13], lung [14, 15], ovarian [16] and renal [17] cancer, and glioblastoma [18, 19]. Cediranib has been reported to show activity in prostate cancer. Ryan et al. [20] reported a phase I trial that established a maximum tolerated dose of 20 mg with the dose-limiting toxicities of muscle weakness and hypertension. We conducted a phase II trial using cediranib 20 mg orally, once daily to assess the clinical efficacy and side-effect profile in patients with metastatic CRPC who have progressed after therapy with docetaxel. Previous studies have reported that serum PSA level has not been a dependable marker in assessing response using non-cytotoxic drugs [21, 22]. The present study used clinical and radiographical criteria to assess response. A second part of the study added prednisone in an attempt to limit the constitutional side effects that were associated with cediranib.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Patient Selection

Patients were considered eligible if they had: (i) histological confirmation of prostatic adenocarcinoma; (ii) metastatic CRPC with radiographic evidence of disease, with progression defined as an increase in radiographic lesions and/or PSA level rising on successive measurements; (iii) previously received a docetaxel-based chemotherapy regimen with no limit on subsequent treatments; and (iv) Eastern Cooperative Oncology Group (ECOG) performance status of 0, 1 or 2. Patients were excluded if they had: (i) QTc > 470 ms; (ii) >1+ proteinuria; (iii) uncontrolled hypertension; or d) abnormal haematological or biochemical characteristics and/or brain metastasis. Patients were not allowed to receive other anti-tumour therapy (including radiation therapy) except continuation of LHRH agonists. Concurrent use of bisphosphonates was allowed for patients with known bone metastases. All patients gave written informed consent in accordance with federal, state and institutional guidelines, and the National Cancer Institute (NCI) review board approved the study.

Study Design

This was a single institution, single-arm, open-label phase II clinical trial with part one using a two-stage optimum design, as described by Simon [23]. The study was designed before the publication of the recent Prostate Cancer Working Group 2 criteria and was based on data from previous trials of NCI patients with similar eligibility requirements. The trial targeted 30% as the desirable proportion of patients who were progression-free, according to clinical and radiographic criteria, by the 6th monthly evaluation (p1 = 0.30); and results would be considered inadequate if only a fraction consistent with 10% were progression-free by the 6 month evaluation time (p0 = 0.10). With α = 0.10 and β = 0.10 as acceptable error probabilities, initially 12 patients were enrolled and followed for progression. If 0–1 of the 12 patients was progression-free at 6 months, then no further patients were enrolled. If two or more of the first 12 patients were progression-free at 6 months, then enrolment was continued until a total of 35 evaluable patients had been entered. If five or fewer of the 35 patients were progression-free at 6 months, this would not be considered adequate responsiveness to treatment, while if six or more of the 35 patients were progression-free at 6 months, this would have indicated results consistent with an adequate progression-free survival (PFS) probability worthy of further investigation. Under the null hypothesis (10% of patients progression-free at 6 months), the probability of early termination after 12 patients had been evaluated at 6 months would be 66%.

In the second portion of the study, 23 additional evaluable patients were enrolled in the trial and received prednisone and cediranib. This was done because a large proportion of patients from the first part of the study experienced grade ≥2 fatigue, anorexia and/or weight loss, many of whom appeared to be responding to therapy. To determine if adding prednisone was associated with an improvement in treatment response, a sample size calculation was carried out after evaluation of the first 28 patients. Using a one-sided α = 0.05 Fisher's exact test, we found that 23 patients would provide 80% power to detect a difference between 19/28 (68%) patients identified with grade ≥2 fatigue, anorexia and/or weight loss from the first part of the study, and 30% of patients with the same level and type of symptoms.

The secondary objective of the present study was to measure the overall response rate and overall survival (OS). PSA progression may occur in the absence of radiological or clinical progression and response rate to therapy with cediranib. An additional objective was demonstration of a pharmacodynamic effect by cediranib in the patient and on the tumour (when possible) via dynamic contrast enhanced (DCE)-MRI) to determine early changes in tumour vascularity during treatment.

Treatment Plan and Toxicity Evaluation

Patients were given cediranib 20 mg orally for a 28-day cycle in the first stage of the study. Patients were evaluated in the clinic every 4 weeks, and radiographic assessments using CT and bone scintigraphy were obtained every 2 months. Blood tests including complete blood count, chemistry and PSA level were obtained at each monthly visit. Response and progression were evaluated using Response Evaluation Criteria in Solid Tumours (RECIST) 1.0 [24]. For bone scans, progression was defined as appearance of a new lesion, and improvement was defined as the complete resolution of at least one lesion. PSA values were recorded but not used as a criterion for progression. Echocardiograms and electrocardiograms were carried out and patients’ troponin levels were measured monthly owing to previous reports of cardiotoxicity [25]. The dose of cediranib was adjusted based on NCI Common Toxicity Criteria for Adverse Events (CTCAE version 3.0). Briefly, patients found to have grade 1 toxicity had no dose adjustment or interruptions. Cediranib was decreased by 25% for persistent grade 2 nausea, vomiting, diarrhoea or other intolerable grade 2 side effects. Cediranib was held for Grade 3 non-haematological toxicity or Grade 4 haematological toxicity until resolution of side effects ≤ Grade 1. After resolution, the dose was decreased by 25%. If a patient's side effects did not resolve, the patient was withdrawn from the study. Patients with Grade 4 non-haematological toxicity were to be withdrawn from the study. Hypertension was managed using therapy and toxicity grading specific for anti-angiogenic therapies. Proteinuria, analysed by dipstick, required confirmation with a dose adjustment of 25% for >1 g of protein in 24-h urine. All patients in the second stage of the trial received prednisone 10 mg orally daily and cediranib 20 mg orally daily. Both drugs were taken together in the morning. Cortisol levels were checked in patients who were not on steroids at the time of enrolment.

Response Evaluation

Patients were evaluated based on RECIST every two cycles. Patients remained enrolled in the study unless: (i) they had radiological disease progression; (ii) they had intercurrent illness requiring cessation of cediranib; (iii) they decided to withdraw from the study; or (iv) they had unacceptable side effects. Patients with bone-only disease were considered to have progressed if they had new lesions or clinical signs of progression. Bone scans with a disappearance of any bone lesion were recorded. Patients who had a rising PSA level were still included in the study if they did not meet the above criteria [22].

Dynamic Contrast-enhanced MRI

Dynamic contrast-enhanced MRI was obtained using T1-weighted gradient echo images with low-molecular-weight gadolinium chelates. DCE-MRI scans were obtained before treatment (preferably <7 days before treatment start), 24–36 h after starting daily oral dosing, and at 28 and 56 ± 2 days, subsequently doubling the time before follow-up scans (e.g. 112 and 224 days), while patients remained enrolled in the study, and depending on the study design. The initial area under the gadolinium curve measured over 60 s (iAUC60) and the transport constant (Ktrans) from a region-of-interest analysis, averaged over the whole tumour volume (using three-dimensional techniques) or over the tumour area (using two-dimensional techniques), were measured and provided basic measurements for collation of DCE-MRI data. Both iAUC60 and Ktrans require calculation of instantaneous tumour gadolinium concentration based on the change in relaxivity due to contrast uptake ΔR1. Manual region-of-interest measurements were obtained from the target lesion. Baseline and follow-up after treatment cycles of additional DCE-MRI variables, including tumour volumes, the rate constant (Kep) and the extravascular fraction Ve, were also obtained.


Recent data from breast cancer trials using cediranib and doxorubicin at the NCI identified two patients who experienced asymptomatic ejection fraction decreases that were reversible upon withdrawal of cediranib [25]. As it is difficult to determine whether this was a toxicity that was associated with VEGF inhibitor monotherapy or with concurrent regimens including VEGF inhibitors, echocardiograms were obtained for patients concurrently with CT scans and bone scans. As there are currently no guidelines on the evaluation and management of ejection fraction in the setting of cediranib use, similar criteria to those previously used for trastuzumab, which is known to decrease ejection fraction, and to those used in the NCI breast cancer trials were applied [26].

Statistical Analysis

Time to progression was defined as the time from the date of enrollment in the study to the date of first observation of disease progression, death while enrolled in the study, or to the time when the patient was removed from the study at the Principal Investigator's discretion. Other reasons for a patient being removed from the study, such as adverse events, patient decision, or intercurrent illness, were used to censor the time to progression. OS was defined as the time from the date of enrolment in the study to the date of death from any cause. Patients remaining enrolled in the study or alive at time of analyses were censored at date of last follow-up. The probability of PFS or OS as a function of time was determined using the Kaplan–Meier method.

For the DCE-MRI analyses, changes from baseline were reported as the relative change, which was the value at (day 28 – day 0)/day 0 for each of the variables evaluated. This was done because the relative changes were less associated with the baseline values than were the absolute changes; thus, the relative change x 100% would represent the percentage change from baseline. The association between DCE-MRI variables and PFS or OS as determined by the Kaplan–Meier method was assessed. Initially, in univariate analyses, the significance of a difference in a pair of Kaplan–Meier curves, or among a set of curves, was determined using the log rank test. Initial exploratory evaluations were carried out to determine the association between the variables divided at their quartiles and survival or PFS. When an evaluation of the survival or PFS analyses suggested that there would be a potentially significant effect when the data were re-divided into the two groups suggested by the data, the resulting P value was adjusted by multiplying the unadjusted value by 3 to account for the implicit comparisons involved in deciding to use a single division of the data. Cox proportional hazards model analysis was used to determine the hazard ratio of any DCE-MRI variable found to be potentially significantly associated with outcome in the univariate analyses, as well as to determine the joint effect of the variables found to be significant. Spearman correlation was used to describe the association between change in tumour volume and the DCE-MRI variables. All P values were two-tailed, and except as noted above, were reported without any correction for multiple testing, and the results of these analyses were considered hypothesis-generating.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

A total of 59 patients were enrolled in the present study: 36 patients in the first cohort and 23 in the second cohort of the trial, between February 2007 and February 2010. The study demographics and baseline characteristics of all patients are shown in Table 1. Many patients had unfavourable prognostic factors, as evidenced by visceral disease, a high Gleason score at diagnosis, and previous treatment with two or more chemotherapy regimens. Eighteen patients (31%) had received previous treatment with at least one anti-angiogenic agent. The majority (n = 47) had an ECOG performance status of 1.

Table 1. Demographics and disease characteristics of enrolled patients (n = 59) patient demographics and clinical characteristics.
Median (range) age, years68.9 (51.4–79.8)
Race, n (%) 
White52 (88)
African-American3 (5)
Hispanic3 (5)
Asian1 (2)
ECOG performance status, n (%) 
02 (3)
148 (81)
29 (16)
Median (range) Gleason score at diagnosis8 (4–10)
Gleason score at diagnosis, n (%) 
≥832 (54)
≤726 (44)
Unclassified1 (2)
PSA level during study, ng/mL 
Median (range)175.45 (4.94–1587)
Haemoglobin, g/dL 
Median (range)11.8 (6.7–14.9)
Lactate dehydrogenase, U/L 
Median (range)235.5 (124–3344)
Alkaline phosphatase, U/L 
Median (range)141 (46–1244)
Metastases, n (%) 
Bone only19 (33)
Soft tissue and bone36 (62)
Liver9 (15)
Lung1 (2)
Pain at baseline, n (%)40 (69)
Previous chemotherapy regimens, n (%) 
119 (33)
221 (36)
≥318 (31)

Of the 59 patients who were enrolled, 39 had measurable disease. One patient in the first cohort was not evaluable owing to the development of a cord compression on day 2 of therapy and was subsequently removed from the trial. Among those with measurable disease, six patients had confirmed partial responses and one had an unconfirmed partial response. The probability of PFS was 43.9% at 6 months. After a median potential follow-up of 33.5 months, nine patients of 58 evaluable patients remained alive as of the last follow-up. The median PFS was 3.6 months for the first cohort (n = 35) and 3.7 months for the second cohort (n = 23 [P = 0.79; data not shown]). The median PFS for all patients (n = 58) was 3.7 months (Fig. 1A). The median OS was 10.9 months for the first cohort and 9.6 months for the second cohort (P = 0.23; data not shown), and for the whole cohort of 58 patients it was 10.1 months (Fig. 1B).


Figure 1. Kaplan–Meier analysis of A, PFS and B, OS for all patients (n = 58) included in the study.

Download figure to PowerPoint

Of the 59 patients enrolled in the study, 49 had DCE-MRI scans that were available for assessment: six patients were evaluated up to day 2, eight up to day 28, 16 up to day 56, 14 up to day 112, four up to day 224, and one up to day 448 (48 patients were evaluated on day 2, 43 on day 28, 33 on day 56, 18 on day 112, 5 on day 224, and one on days 336 and 448). A rapid reduction in the primary DCE-MRI variables Ktrans and iAUC60 was observed for the majority of patients within 24 h of the first dose (median reductions of 17.65 and 14.50%, respectively). The magnitude of change in either median Ktrans or iAUC60 observed after 28 days was greater overall than at 24 h (P = 0.001 and P = 0.015 respectively, Wilcoxon signed-rank test, n = 40). Based on previous studies with cediranib, we considered a >30% reduction in primary DCE-MRI variables to be representative of a positive pharmacodynamic response to anti-vascular therapy [24]. After one cycle of therapy, 60% of patients experienced a vascular response with >30% reductions in Ktrans, where the mean reduction was 66%. Figure 2A shows the distribution of patients with minor to major reductions in DCE-MRI variables at day 28. Over 65% of patients experienced at least minor reductions from baseline in Ktrans and iAUC60, observed at the end of cycle 1. The proportion of patients with reductions in DCE-MRI variables began to decrease at 56 days after the initiation of therapy, for patients who remained on cediranib (data not shown).


Figure 2. A, The percent of patients, and reductions, (mean ± SD) in DCE-MRI variables after one cycle of therapy. B, The percent change in DCE-MRI variables during four cycles of cediranib therapy (median ± 95% CI). C, Maximum percent change in lesion volume through all cycles and maximum percent change during 1–2 cycles of cediranib therapy in 43 patients. Compound daily growth rate represents the maximum percent change normalized by 28 or 56 days on therapy.

Download figure to PowerPoint

The change in vascular response to cediranib over the course of four cycles as percent reductions in DCE-MRI variables is shown in Fig. 2B. In the present study, the vascular response tended to reach a maximum at 28 days after initiation of treatment. The distribution of the largest percent change from baseline in lesion volume during 1–2 cycles of therapy is shown in Fig. 2C. Approximately 40% of patients had maximum lesion volume reductions of >25% (i.e. 28–84% decreases in lesion volume). The compound daily growth rate (%) is a more relative indicator for comparing patients who may have either fast or slow growth in lesion volume. For example, patients #1 and #4 in Fig. 3 had the same percent growth rate; however, patient #4 had a maximum reduction at 28 days compared with 56 days for patient #1. Also, the figure depicts only three patients who experienced maximum changes in lesion volume beyond two cycles (represented as spikes), for those patients who remained on therapy.


Figure 3. Kaplan–Meier estimates for PFS of patients with DCE-MRI variables. A, Ktrans (threshold defined as baseline Ktrans = 0.22). B, Kep (threshold defined as the post-therapy day 28 Kep = 0.35). Log rank P values shown have been adjusted for the effect of exploratory analyses used to determine the preferred quartile at which to divide the data for evaluation.

Download figure to PowerPoint

The Ktrans, Kep and iAUC60 measurements at day 0 (baseline), day 28, and the difference between the day 28 and the day 0 measurements (day 28 – day 0), were evaluated for their association with tumour volume. There was no evidence of any association between change in tumour volume or relative change in volume and the DCE-MRI variables evaluated (data not shown). These same measurements were divided into quartiles and the relationship between these variables and PFS and OS was determined. When associations were identified, the data were re-divided at the quartile associated with a difference in the prognostic ability of the variables, and the association with outcome was re-determined. Only baseline Ktrans and Kep at day 28, divided as indicated, were found to be significantly associated with PFS in univariate analyses (Fig. 3), while none of the variables examined were significantly associated with OS. As shown in Table 2, using Cox hazards models for each variable individually, a baseline Ktrans > 0.22 was significantly associated with a lower probability of PFS than a Ktrans of <0.22 (P = 0.02), with a hazard ratio of 3.26 (95% CI: 1.22–8.76). A Kep at day 28 of >0.35 was associated with significantly lower probability of PFS than a Kep of <0.35 (P = 0.02), with a hazard ratio of 2.40 (95% CI: 1.14–5.03). When the two DCE-MRI variables were considered jointly, Kep at day 28 lost its significance in the presence of baseline Ktrans.

Table 2. Cox model analysis for the individual effects of baseline Ktrans < 0.22 vs >0.22, Kep at day 28 < 0.35 vs >0.35, and for the joint effects of these variables.
VariableDFParameter estimateseChi-squarePr > Chi-squareHR95% HR CI
  1. DF, degrees of freedom; HR, hazard ratio.

Ktrans at day 010.8740.3785.3400.0212.3961.142–5.029
Kep at day 2811.1820.5045.5060.0193.2621.215–8.759
Joint effect of the parameters       
Ktrans at day 011.3890.6953.9900.0464.0111.026–15.672
Kep at day 2810.3480.4320.6480.421.4170.607–3.306

All patients who received treatment were analysed for toxicity. Patients received a median (range) of 3 (<1–16) cycles. Of the 58 patients, 30 (52%) discontinued therapy because of disease progression, 10 (17%) owing to toxicity, and 16 (28%) as a result of the physician's or the patient's own decision, but were not required by the protocol to discontinue. One patient discontinued therapy because of intercurrent illness not related to prostate cancer and one patient died during treatment from a cerebral haemorrhage complicated by disseminated intravascular coagulation. Protocol-required dose reductions for the management of cediranib toxicity were necessary for five patients.

Table 3AA shows the most common grade 2 toxicities that occurred in >10% of patients and all grade 3 and 4 toxicities. Significant grade 2 adverse events included hypertension (43%), fatigue (33%), anorexia (31%) and weight loss (27%). The intensity of these events was usually mild to moderate. Most side effects were of short duration and resolved without incident. Severe grade 3 toxicities included fatigue (10%), dehydration (10%), elevated alkaline phosphatase (9%) and muscle weakness (7%). There was a low overall incidence of Common Toxicity Criteria grade 4 toxicity that included thrombosis/embolism (n = 2) and CNS haemorrhage (n = 1). The addition of prednisone in the second stage of the study reduced the overall incidence of grade 2 constitutional toxicities of fatigue (43% in cohort 1 vs 17% in cohort 2), anorexia (34% in cohort 1 vs 26% in cohort 2), and weight loss (31% in cohort 1 vs 17% in cohort 2) as well as grade 2 dehydration, grade 2 prolonged QTc, and grade 3 elevated alkaline phosphatase (Table 3BB).

Table 3A. NCI Common Toxicity Criteria grade 2, 3 or 4 toxicities possibly, probably or definitely related to cediranib.
Toxicity*Grade 2 n (%)Grade 3 n (%)Grade 4 n (%)
  1. *Toxicities occurring in >10% of patients and any grade 3 and 4 toxicities in patients who received study drug (n = 58).

Hypertension25 (43)0 (0)0 (0)
Fatigue19 (33)6 (10)0 (0)
Anorexia18 (31)2 (3)0 (0)
Weight loss15 (27)2 (3)0 (0)
Hypothyroidism13 (22)0 (0)0 (0)
Dehydration10 (17)6 (10)0 (0)
Prolonged QTc10 (17)2 (3)0 (0)
Nausea10 (17)1 (2)0 (0)
Diarrhoea8 (14)0 (0)0 (0)
Hypoalbuminaemia8 (14)0 (0)0 (0)
Proteinuria8 (14)0 (0)0 (0)
Elevated alkaline phosphatase6 (10)5 (9)0 (0)
AST6 (10)2 (3)0 (0)
Vomiting6 (10)1 (2)0 (0)
Hyperbilirubinemia5 (9)1 (2)0 (0)
Muscle weakness3 (5)4 (7)0 (0)
Haemoglobin, anaemia3 (5)1 (2)0 (0)
Lymphopenia2 (3)3 (5)0 (0)
Platelets2 (3)1 (2)0 (0)
Alanine transaminase2 (3)1 (2)0 (0)
Pain1 (2)2 (3)0 (0)
Hypoxia0 (0)2 (3)0 (0)
Thrombosis/embolism0 (0)1 (2)2 (3)
Hyponatremia0 (0)1 (2)0 (0)
Renal failure0 (0)1 (2)0 (0)
Hypokalemia0 (0)1 (2)0 (0)
Partial thromoboplastin time0 (0)1 (2)0 (0)
CNS haemorrhage0 (0)0 (0)1 (2)
Table 3B. Number of toxicities in each cohort.
ToxicityCohort 1 (n = 35)Cohort 2 (n = 23)
Grade 2Grade 3Grade 2Grade 3
Weight loss11240
Prolonged QTc8121
Elevated alkaline phosphatase4520
Aspartate transaminase3230
Muscle weakness2311

A total of 38 patients had echocardiograms. The median (range) ejection fraction on study entry was 65 (43–79)%. No patient had any clinical signs or symptoms of congestive heart failure and there were no clinically significant decreases in ejection fraction.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Cediranib has been shown to inhibit the growth of bone and brain metastases in a preclinical model of advanced prostate cancer [27]. A phase I study of cediranib was previously conducted in 83 patients with advanced solid tumours, with the primary objective of determining the maximum tolerated dose. The most common dose-limiting toxicity was hypertension (n = 7) occurring at doses of ≥20 mg [28]. A separate phase I dose-escalation study was performed in patients with CRPC and showed good tolerability overall and an acceptable cediranib toxicity profile when given at therapeutic doses [20]. Doses of 1, 2.5, 5, 10, 15, 20 and 30 mg were administered to 26 patients. Dose-limiting toxicities occurred at the 30 mg dose and the maximum tolerated dose was identified as 20 mg. Although safety, tolerability and DCE-MRI pharmacodynamic responses were the main objectives of the phase I study, an objective clinical response was observed in one patient whereas four patients experienced PSA level reductions after drug discontinuation.

In the current phase II study of patients with metastatic CRPC who have progressed after docetaxel therapy, cediranib, given at 20 mg daily, was generally well tolerated. The primary objective of the present study was met, as the probability of remaining progression-free at 6 months was 43.9%. The most common adverse events reported were hypertension, fatigue, anorexia and weight loss. Unlike Ryan et al. [20], we did not observe any grade 3 hypertension at the 20 mg dose in the present study. The most common grade 3 toxicities included fatigue, dehydration and elevated alkaline phosphatase. Grade 3/4 hromboembolism was observed in three patients and one patient experienced a grade 4 CNS haemorrhage that resulted in death. The addition of prednisone in the second cohort reduced the overall incidence of constitutional toxicities.

Until recently, life-prolonging therapies for patients with metastatic CRPC were limited to docetaxel-based regimens [29], which received US Food and Drug Administration (FDA) approval in 2004 as first-line chemotherapy for metastatic CRPC, whereas mitoxantrone had previously been approved based on improvements in quality of life [30]. In 2010–2011, three more FDA-approved therapies were added to the CRPC treatment armamentarium: the autologous cellular immunotherapy product sipuleucel-T (indicated for men with minimal or no symptoms) [31], the chemotherapy agent cabazitaxel [32] and the targeted therapy abiraterone acetate (indicated for men with disease progression after docetaxel chemotherapy) [33]. Although it appears that treatment options for patients with metastatic CRPC have increased and outcomes have improved, specifically for those in the post-docetaxel setting, the duration of PFS and OS still remains relatively short.

The median OS for the current study involving cediranib was 10.1 months and was shorter than that reported for phase III trials involving cabazitaxel (OS = 15.1 months) [32] or abiraterone (OS = 14.8 months) [33]; however, direct comparisons among the three studies should be tempered by the fact that patients in the present study were exposed to more previous chemotherapy regimens, thereby potentially affecting the benefit derived from cediranib. Although all three studies evaluated patients who had already had progressive disease after docetaxel, the majority of patients (67%) in present study of cederanib had undergone more than two previous chemotherapy regimens, including 31% who had undergone three or more. Patients in the abiraterone phase III trial were limited to only two previous chemotherapy regimens with only 30% of these patients having had more than one cytotoxic therapy [33]. Similarly, only 15% of patients enrolled in the cabazitaxel phase III trial had had exposure to more than one previous chemotherapy regimen [32].

The present study used DCE-MRI to investigate the effect of once-daily 20 mg dosing on tumour vascular permeability. The primary DCE-MRI variables iAUC60 and Ktrans revealed rapid and sustained reductions from baseline up to two or more cycles in the majority of patients. The results support the hypothesis that cediranib has effects on tumour vasculature, as evidenced by statistically significant reductions in gadolinium uptake, by as much as 97%, and across large proportions of patients in the study. Moreover, vascular permeability and perfusion of gadolinium remained decreased up to day 112; however, no correlation was found between RECIST and changes in DCE-MRI primary or exploratory variables. High baseline Ktrans and post-therapy (day 28) Kep were found to be significantly associated with PFS, indicating that DCE-MRI variables may prove effective as pharmacodynamic predictive biomarkers of clinical outcome for cediranib, as previously shown for sorafenib [34, 35]. The clinical relevance of these findings and their predictive value remain to be determined and validated prospectively, especially given the recent studies showing higher morbidity in patients receiving anti-angiogenic therapy [36].

Although targeting angiogenesis appears to be a rational therapeutic approach for CRPC, there are still major obstacles to identifying the appropriate subgroups that may benefit more from these agents. Monotherapy with sunitinib, another small molecule tyrosine kinase inhibitor, despite supportive phase II data in patients with progressive metastatic CRPC after docetaxel treatment [37], recently failed in a phase III trial of the same patient population [10]. It would be reasonable to expect that the use of angiogenesis inhibitors in combination with chemotherapy would prove to be more efficacious than monotherapy with anti-angiogenic agents; however, results from another recent phase III study showed that the addition of bevacizumab may add little clinical benefit in OS to docetaxel in chemotherapy-naïve patients with CRPC [9]. The role that anti-angiogenic agents have in docetaxel-refractory mCRPC remains to be determined with the most promising candidate being cabozantinib, an oral small molecule inhibitor of multiple kinase signalling pathways including c-MET and VEGFR2. A phase II study investigating the use of cediranib in combination with or without dasatinib, an oral Src family kinase inhibitor, in patients with docetaxel-refractory metastatic CRPC is currently underway ( identifier NCT01260688) as well as a phase II study of docetaxel with or without cediranib in chemotherapy-naïve patients with CRPC (NCT00527124). The challenge lies in how to best combine these agents and how to measure the responses seen with these treatments.

In conclusion, this phase II study showed that cediranib, 20 mg daily, in 58 patients with docetaxel-refractory metastatic CRPC, results in perturbations in tumour vasculature, >30% PFS at 6 months, and a median OS of 10.1 months. There remains a clear need to define the most appropriate treatment approach for patients with metastatic CRPC in light of new second-line treatment options becoming available, especially those who have progressive disease after docetaxel therapy. Future studies will need to focus on determining the optimum sequence and combination of new anti-angiogenics and other investigational agents with conventional chemotherapy. Further investigation into the potential predictive value of DCE-MRI parameters as biomarkers for anti-angiogenic therapy is warranted.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

We thank the nursing staff of the NCI and the fellows of the Medical Oncology Branch at the NCI for their care of our patients, Cynthia Graves for data management assistance, and the Cancer Therapy and Evaluation Program for sponsoring the trial. Most importantly, we appreciate the patients with cancer who enrol in investigational trials to advance the knowledge of this disease.

This project was funded in whole or in part with federal funds from the NCI, National Institutes of Health, under Contract No. HHSN261200800001E (SDS). This work was supported by the Intramural Research Program of the Center for Cancer Research, NCI, National Institutes of Health. Trial Registration ID: NCT00436956 This is a US Government work. There are no restrictions on its use. The views expressed within this paper do not necessarily reflect those of the US Government.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  • 1
    Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010; 60: 277300
  • 2
    Choy M, Rafii S. Role of angiogenesis in the progression and treatment of prostate cancer. Cancer Invest 2001; 19: 181191
  • 3
    Figg WD, Kruger EA, Price DK, Kim S, Dahut WD. Inhibition of angiogenesis: treatment options for patients with metastatic prostate cancer. Invest New Drugs 2002; 20: 183194
  • 4
    Macpherson GR, Ng SS, Lakhani NJ et al. Antiangiogenesis therapeutic strategies in prostate cancer. Cancer Metastasis Rev 2002; 21: 93106
  • 5
    Ferrer FA, Miller LJ, Andrawis RI et al. Vascular endothelial growth factor (VEGF) expression in human prostate cancer: in situ and in vitro expression of VEGF by human prostate cancer cells. J Urol 1997; 157: 23292333
  • 6
    Duque JL, Loughlin KR, Adam RM et al. Plasma levels of vascular endothelial growth factor are increased in patients with metastatic prostate cancer. Urology 1999; 54: 523527
  • 7
    Pallares J, Rojo F, Iriarte J et al. Study of microvessel density and the expression of the angiogenic factors VEGF, bFGF and the receptors Flt-1 and FLK-1 in benign, premalignant and malignant prostate tissues. Histol Histopathol 2006; 21: 857865
  • 8
    Antonarakis ES, Carducci MA. Targeting angiogenesis for the treatment of prostate cancer. Expert Opin Ther Targets 2012; 16: 365376
  • 9
    Kelly WK, Halabi S, Carducci M et al. Randomized, double-blind, placebo-controlled phase III trial comparing docetaxel and prednisone with or without bevacizumab in men with 20 metastatic castration-resistant prostate cancer: CALGB 90401. J Clin Oncol 2012; 30: 15341540
  • 10
    Michaelson MD, Sengeløv L, Saad F et al. Randomized, placebo-controlled, phase III trial of sunitinib in combination with prednisone (SU+P) versus prednisone (P) alone in men with progressive metastatic castration-resistant prostate cancer (mCRPC). J Clin Oncol 2011; 29 (Suppl. 15): 4515
  • 11
    Takeda M, Arao T, Yokote H et al. AZD2171 shows potent antitumor activity against gastric cancer over-expressing fibroblast growth factor receptor 2/keratinocyte growth factor receptor. Clin Cancer Res 2007; 13: 30513057
  • 12
    Wedge SR, Kendrew J, Hennequin LF et al. AZD2171: a highly potent, orally bioavailable, vascular endothelial growth factor receptor-2 tyrosine kinase inhibitor for the treatment of cancer. Cancer Res 2005; 65: 43894400
  • 13
    Robertson JD, Botwood NA, Rothenberg ML, Schmoll HJ. Phase III trial of FOLFOX plus bevacizumab or cediranib (AZD2171) as first-line treatment of patients with metastatic colorectal cancer: HORIZON III. Clin Colorectal Cancer 2009; 8: 5960
  • 14
    Goss GD, Arnold A, Shepherd FA et al. Randomized, double-blind trial of carboplatin and paclitaxel with either daily oral cediranib or placebo in advanced non-small-cell lung cancer: NCIC clinical trials group BR24 study. J Clin Oncol 2010; 28: 4955
  • 15
    Ramalingam SS, Belani CP, Mack PC et al. Phase II study of Cediranib (AZD 2171), an inhibitor of the vascular endothelial growth factor receptor, for second-line therapy of small cell lung cancer (National Cancer Institute #7097). J Thorac Oncol 2010; 5: 12791284
  • 16
    Matulonis UA, Berlin S, Ivy P et al. Cediranib, an oral inhibitor of vascular endothelial growth factor receptor kinases, is an active drug in recurrent epithelial ovarian, fallopian tube, and peritoneal cancer. J Clin Oncol 2009; 27: 56015606
  • 17
    Mulders P, Hawkins R, Nathan P et al. Cediranib monotherapy in patients with advanced renal cell carcinoma: results of a randomised phase II study. Eur J Cancer 2012; 48: 527537
  • 18
    Batchelor TT, Duda DG, di Tomaso E et al. Phase II study of cediranib, an oral panvascular endothelial growth factor receptor tyrosine kinase inhibitor, in patients with recurrent glioblastoma. J Clin Oncol 2010; 28: 28172823
  • 19
    Dietrich J, Wang D, Batchelor TT. Cediranib: profile of a novel anti-angiogenic agent in patients with glioblastoma. Expert Opin Investig Drugs 2009; 18: 15491557
  • 20
    Ryan CJ, Stadler WM, Roth B et al. Phase I dose escalation and pharmacokinetic study of AZD2171, an inhibitor of the vascular endothelial growth factor receptor tyrosine kinase, in patients with hormone refractory prostate cancer (HRPC). Invest New Drugs 2007; 25: 445451
  • 21
    Aragon-Ching JB, Jain L, Gulley JL et al. Final analysis of a phase II trial using sorafenib for metastatic castration-resistant prostate cancer. BJU Int 2009; 103: 16361640
  • 22
    Scher HI, Halabi S, Tannock I et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol 2008; 26: 11481159
  • 23
    Simon R. Optimal two-stage designs for phase II clinical trials. Control Clin Trials 1989; 10: 110
  • 24
    Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205216
  • 25
    Denduluri N, Berman AW, Vatas U et al. Cardiotoxicity after doxorubicin and AZD2171, an oral vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitor, in patients (pts) with breast cancer. J Clin Oncol 2007; 25 (Suppl. 18): 554
  • 26
    Tan-Chiu E, Yothers G, Romond E et al. Assessment of cardiac dysfunction in a randomized trial comparing doxorubicin and cyclophosphamide followed by paclitaxel, with or without trastuzumab as adjuvant therapy in node-positive, human epidermal growth factor receptor 2-overexpressing breast cancer: NSABP B-31. J Clin Oncol 2005; 23: 78117819
  • 27
    Yin JJ, Zhang L, Munasinghe J, Linnoila RI, Kelly K. Cediranib/AZD2171 inhibits bone and brain metastasis in a preclinical model of advanced prostate cancer. Cancer Res 2011; 70: 86628673
  • 28
    Drevs J, Siegert P, Medinger M et al. Phase I clinical study of AZD2171, an oral vascular endothelial growth factor signaling inhibitor, in patients with advanced solid tumors. J Clin Oncol 2007; 25: 30453054
  • 29
    Tannock IF, de Wit R, Berry WR et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: 15021512
  • 30
    Tannock IF, Osoba D, Stockler MR et al. Chemotherapy with mitoxantrone plus prednisone or prednisone alone for symptomatic hormone-resistant prostate cancer: a Canadian randomized trial with palliative end points. J Clin Oncol 1996; 14: 17561764
  • 31
    Kantoff PW, Higano CS, Shore ND et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010; 363: 411422
  • 32
    de Bono JS, Oudard S, Ozguroglu M et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet 2010; 376: 11471154
  • 33
    de Bono JS, Logothetis CJ, Molina A et al. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med 2011; 364: 19952005
  • 34
    Hahn OM, Yang C, Medved M et al. Dynamic contrast-enhanced magnetic resonance imaging pharmacodynamic biomarker study of sorafenib in metastatic renal carcinoma. J Clin Oncol 2008; 26: 45724578
  • 35
    Kelly RJ, Rajan A, Force J et al. Evaluation of KRAS mutations, angiogenic biomarkers, and DCE-MRI in patients with advanced non-small-cell lung cancer receiving sorafenib. Clin Cancer Res 2011; 17: 11901199
  • 36
    Ranpura V, Hapani S, Wu S. Treatment-related mortality with bevacizumab in cancer patients: a meta-analysis. JAMA 2011; 305: 487494
  • 37
    Sonpavde G, Periman PO, Bernold D et al. Sunitinib malate for metastatic castration-resistant prostate cancer following docetaxel-based chemotherapy. Ann Oncol 2010; 21: 319324