Randomized phase 2 study of pegylated SN-38 (EZN-2208) or irinotecan plus cetuximab in patients with advanced colorectal cancer


  • Presented at the Gastrointestinal Cancers Symposium; January 19-21, 2012; San Francisco, California.



Irinotecan is cytotoxic in patients with advanced colorectal cancer (CRC). SN-38 (10-hydroxy-7-ethyl-camptothecin) is the active metabolite of irinotecan. Attachment of polyethylene glycol (PEG) polymer chains (pegylation) to SN-38 (EZN-2208) increases the solubility, exposure, and half-life of SN-38. Preclinical studies demonstrated superior in vitro efficacy of EZN-2208 when it was tested in irinotecan-refractory human CRC cell lines.


Patients with metastatic or locally recurrent CRC who had previously received 5-flurouracil (5-FU), oxaliplatin, and irinotecan were assigned to receive EZN-2208 monotherapy (9 mg/m2 on days 1, 8, and 15 every 28 days for patients with KRAS-mutant tumors only [arm A]), and patients with KRAS wild-type tumors were randomized (2:1) to receive either EZN-2208 plus cetuximab (400 mg/m2 loading dose on day 1 followed by 250 mg/m2 weekly starting on day 8 [arm B]) or irinotecan 125 mg/m2 on days 1 and 8 every 21 days plus cetuximab at the same doses indicated above (arm C).


The overall response rate and progression-free survival were 0% and 1.8 months, respectively, in arm A; 10.7% and 4.9 months (95% confidence interval [CI], 3.2-5.8 months), respectively, in arm B; and 14.3% and 3.7 months (95% CI, 2.1-5.8 months), respectively, in arm C. EZN-2208 was well tolerated in combination with cetuximab. No statistically significant difference in survival was observed between arm B (9.8 months; 95% CI, 7.2-11.2 months) and arm C (9.1 months; 95% CI, 6.0-13.0 months).


EZN-2208, either as monotherapy or in combination with cetuximab, was well tolerated in patients with refractory CRC. Overall survival and progression-free survival were similar in the cetuximab plus irinotecan arm and the EZN-2208 arm. Cancer 2013;119:4223–4230. © 2013 American Cancer Society.


Irinotecan is a topoisomerase I inhibitor that is associated with an overall survival (OS) benefit when used to treat patients with locally advanced or metastatic colorectal cancer (CRC). It has proven efficacy in treatment-naive patients[1] and in treatment-refractory patients with CRC.[2] The active metabolite of irinotecan is SN-38 (10-hydroxy-7-ethyl-camptothecin). Because of its poor solubility, SN-38 is not suitable for parenteral administration.[3] Pegylation of drugs may lead to the creation of a new compound with superior solubility, pharmacokinetics, and clinical efficacy compared with the parent drug[4]; this has been demonstrated as a clinically effective treatment strategy, as evidenced by pegylated granulocyte-colony-stimulating factor (G-CSF),[5] pegylated liposomal doxorubicin,[6] and pegylated-interferon α-2b.[7]

EZN-2208 is a pegylated conjugate of SN-38 that is more water soluble, has a longer circulating half-life, and produces greater therapeutic exposure of SN-38 compared with the parent compound, irinotecan.[8] In vitro studies have demonstrated that EZN-2208 is associated with activity in tumor xenografts (including the human CRC cell line HT-29) and that it has efficacy in irinotecan-refractory tumors.[9] In preclinical studies, EZN-2208 induced tumor regression in neuroblastoma models[10] and was associated with significant tumor responses in human B-cell non-Hodgkin lymphoma xenograft models.[11] It also has been demonstrated that EZN-2208 inhibits angiogenesis by down-regulating heat-inducible factor 1α (HIF-1α).[12]

A phase 1 clinical trial of EZN-2208 in solid tumors determined that the recommended phase 2 monotherapy dose when administered every 3 weeks was 10 mg/m2 without the use of G-CSF and 16.5 mg/m2 with G-CSF support.[13] The toxicity observed was similar to that of irinotecan, including diarrhea, neutropenia, nausea, and thrombocytopenia as the most commonly observed side effects. Of 39 patients who received EZN-2208, among the 17 patients who had received prior irinotecan therapy, 7 had a best overall response of either an unconfirmed partial response or stable disease that lasted a minimum of 4 months. A second phase 1 study evaluating the weekly dosing of EZN-2208 (on days 1, 8, and 15 every 28 days) determined that the recommended phase 2 dose was 9 mg/m2.[14]

EZN-2208 is associated with superior in vitro efficacy compared with irinotecan. In human CRC HT-29 tumor xenograft models, EZN-2208 demonstrated effective antitumor inhibition in an irinotecan-resistant mouse model, and multiple doses of EZN-2208 were associated with a cure in 90% of the animals.[9] Tumors frequently acquire resistance to topoisomerase inhibitors by overexpression of the efflux transporter ABCG2 (ATP-binding cassette, subfamily G).[15] Preclinical studies have indicated that EZN-2208 can overcome ABCG2-mediated resistance, possibly because of the increased drug delivery of SN-38 to tumors, by means of enhanced permeability and retention.[16-18]

Given the demonstrated preclinical efficacy of EZN-2208 and its tolerability and efficacy in phase 1 testing, we undertook a phase 2 study of EZN-2208 monotherapy in patients with Kirsten rat sarcoma viral oncogene homolog (KRAS)-mutant, metastatic colorectal cancer who had prior exposure to irinotecan, and we also conducted a randomized phase 2 study of cetuximab plus either irinotecan or EZN-2208 in patients with KRAS wild-type tumors who also had prior irinotecan exposure but who were cetuximab-naive. The primary objectives were to determine the overall response rate as well as progression-free survival (PFS) (the latter in arms B and C only); secondary objectives were to evaluate the duration of response, OS, and the safety and tolerability of EZN-2208 when combined with cetuximab.


Patient Selection

Patients with surgically unresectable (either metastatic or locally advanced) CRC were eligible for protocol participation. Inclusion criteria were as follows: age ≥18 years, histologic (or cytologic) diagnosis of unresectable (locally advanced and/or metastatic) colon or rectal adenocarcinoma with measurable disease according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1,[19] prior treatment with 5-fluouracil (5-FU), irinotecan, and oxaliplatin (alone or in combination) with evidence of disease progression either while receiving or within 3 months of receiving each agent, radiographic evidence of disease progression; an Eastern Cooperative Oncology Group (ECOG) performance status <2 with adequate organ function, and willingness to have a tumor tested for KRAS mutation status if it had not already been tested before study entry. Prior antiepidermal growth factor receptor therapy was an exclusion criterion, but prior bevacizumab therapy was allowed. The study was approved by the responsible institutional review boards and followed the ethical principles of Good Clinical Practice in accordance with the Declaration of Helsinki. Participants provided written and verbal consent before study entry. Prophylaxis for antiepidermal growth factor receptor skin toxicity (principally xerodermatitis and acneiform rash), including skin moisturizer, topical steroids (1% hydrocortisone cream), and doxycycline 100 mg twice daily, was administered from weeks 1 through 6 beginning on day −1 with the option of continuing thereafter at the investigator's discretion, as evaluated in a previous clinical trial.[20]

Study Design and Treatment

Patients who had tumors that harbored KRAS mutations were enrolled on arm A and received EZN-2208 at a dose of 9 mg/m2 intravenously over 60 minutes on days 1, 8, and 15 every 28 days, and standard serotonin antagonist antiemetics were administered before infusion. Patients who had wild-type KRAS tumors were randomized to arm B or arm C. Treatment in arm B consisted of EZN-2208 (at the same dose and schedule as arm A) combined with cetuximab at a loading dose of 400 mg/m2 on day 1, followed by 250 mg/m2 weekly, infused over 1 hour, with standard antiemetics and antihypersensitivity prophylaxis. Patients who were randomized to arm C received irinotecan 125 mg/m2 intravenously over 1 hour on days 1 and 8 every 21 days combined with cetuximab (at the same dose and schedule as arm B). The doses of irinotecan and cetuximab were derived from previous clinical trials.[21] A central randomization scheme using an interactive web-based response system was used. Randomization between arms B and C was at a 2:1 ratio using a permutated blocks method, with a block size of 6. This was a randomized, open-label study.

Assessment of Toxicity and Response

Toxicity was graded by using the Common Terminology Criteria for Adverse Events, version 3.0 (CTCAE).[22] Dose adjustments were made depending on the toxicity observed with each treatment cycle. Patients who received cetuximab and experienced a CTCAE grade >2 hypersensitivity reaction had treatment with cetuximab permanently discontinued.

Within 14 days before treatment, patients were required to undergo a complete history and physical examination, including date and stage of CRC and of metastases, prior cancer therapy, site(s) of metastases, performance status, and review of symptoms. Patients also underwent the following tests: complete blood count, urinalysis, renal and hepatic function tests, electrocardiogram, pregnancy test for premenopausal women, and electrolytes. Tumor assessments with chest roentgenogram, computerized tomography, or magnetic resonance imaging were required within 1 month before the start of treatment, and all other tests were done no more than 2 weeks before the start of therapy. The overall response rate was defined as the number of complete and partial responses documented and subsequently confirmed. During the course of treatment, history and physical examination, concomitant medications, hepatic and renal function tests, electrolytes, and toxicity assessment were performed at the beginning of each cycle. Weekly assessments of complete blood count, platelet count, and differential were collected. Assessment of response, including radiographic assessment, was performed approximately every 8 weeks. EZN-2208–associated neutropenia was managed with dose reductions. Growth factor support was not allowed on this protocol.

Statistical Plan

It was anticipated that 220 patients would be required to be enrolled to obtain 191 evaluable patients; it was estimated that approximately 36% of patients enrolled would have a tumor with a KRAS mutation based on prior data.[23] Patient outcomes were analyzed as treated, response assessable, or on an intent-to-treat basis. PFS was analyzed using the Kaplan-Meier product-limit method,[24] and the estimated median PFS and corresponding confidence intervals (CIs) were calculated.


Baseline Patient Characteristics

In this study, we evaluated 292 patients between July 2009 and July 2011 (see Fig. 1, Consolidated Standards of Reporting Trials [CONSORT] diagram). Seventy-one patients failed screening, 221 patients were assigned or randomized, and 213 patients received treatment. Patient demographics are listed in Table 1.

Figure 1.

This is a Consolidated Standards of Reporting Trials (CONSORT) diagram of the current study. KRAS indicates Kirsten rat sarcoma viral oncogene homolog; EZU-2208, pegylated SN-38 (10-hydroxy-7-ethyl-camptothecin).

Table 1. Baseline Patient Characteristics
 No. of Patients (%)
 KRAS MutantKRAS Wild Type
CharacteristicArm A: EZN-2208Arm B: EZN-2208 + CetuximabArm C: Irinotecan + Cetuximab
  1. Abbreviations: 5-FU, 5-fluorouracil; ECOG, Eastern Cooperative Oncology Group; EZN-2208, pegylated SN-38 (10-hydroxy-7-ethyl-camptothecin); PD, progressive disease; PR, partial response; SD, stable disease.

Patients treated93 (100)80 (100)38 (100)
No. evaluable for response82 (88)75 (94)35 (92)
Age: Median [range], y59 [21–81]62 [32–81]61 [35–72]
Men44 (47)43 (54)21 (55)
Women49 (53)37 (46)17 (45)
ECOG performance status   
037 (40)35 (44)16 (42)
154 (58)45 (56)22 (58)
22 (2)0 (0)0 (0)
Disease type   
Locally advanced0 (0)2 (3)1 (3)
Metastatic93 (100)78 (98)37 (97)
Prior irinotecan   
Yes92 (99)78 (98)37 (97)
No1 (1)2 (3)1 (3)
Prior response to most recent irinotecan3 (3)13 (17)3 (8)
PD within 3 mo of 5-FU, oxaliplatin, irinotecan   
Yes90 (97)77 (96)38 (100)
No3 (3)3 (4)0 (0)
Best response before study enrollment   
PR4 (4)8 (10)2 (5)
SD20 (22)20 (25)7 (18)
PD66 (71)48 (60)28 (74)
Not evaluated3 (3)4 (5)1 (3)

Treatment Duration and Toxicity

Patients in arm A received a median of 2 treatment cycles (range, 1-10 treatment cycles), patients in arm B received a median of 4 treatment cycles (range, 1-24 treatment cycles), and patients in arm C also received a median of 4 treatment cycles (range, 1-22 treatment cycles). The median duration of drug exposure in 7 weeks in arm A, 15 weeks in arm B, and 10 weeks in arm C.

Objective Radiographic Response Rate, Overall Survival, and Time to Progression

No radiographic responses were observed in the EZN-2208 monotherapy arm (ie, in those patients who had received prior irinotecan chemotherapy, who had progressed within 3 months of prior irinotecan and oxaliplatin chemotherapy, or who had KRAS-mutant tumors). Seventy-five patients (94%) were evaluable from arm B (EZN-2208 plus cetuximab), and 35 patients (92%) were evaluable from arm C (irinotecan plus cetuximab). No complete responses were observed, and no statistically significant differences in OS or time to progression were observed between arms B and C. The results of the investigator-determined radiographic response for arms B and C are provided in Table 2. The Kaplan-Meier PFS and OS curves are provided in Figure 2 and Figure 3, respectively. The 6-month PFS rate was 37% in arm B (95% CI, 27%-48%) and 29% in arm C (95% CI, 14%-44%).

Figure 2.

Kaplan-Meier survival curves illustrate (Top) progression-free survival and (Bottom) overall survival in arm A (pegylated 10-hydroxy-7-ethyl-camptothecin [EZN-2208] monotherapy).

Figure 3.

Kaplan-Meier survival curves illustrate (Top) progression-free survival and (Bottom) overall survival in arm B (pegylated SN-38 [10-hydroxy-7-ethyl-camptothecin] [EZN-2208] plus cetuximab; blue line) and arm C (irinotecan plus cetuximab; green line).

Table 2. Overall Response Rate, Progression-Free Survival, Overall Survival, and 6-Month Progression-Free Survival in Arms B and C
VariableArm B: EZN-2208 + CetuximabArm C: Irinotecan + Cetuximab
  1. Abbreviations: CI, confidence interval; EZN-2208, pegylated SN-38 (10-hydroxy-7-ethyl-camptothecin); NR, not reached; ORR, overall response rate; OS, overall survival; PFS, progression-free survival.

Confirmed ORR: No. (%) [95% CI], %8 (10.7) [3.7–17.6%]5 (14.3) [2.7–25.9%]
Response duration: Median [95% CI], mo(5.6) [3.6–7.7%](9.5) [2.7 to NR%]
PFS: Median [95% CI], mo(4.9) [3.2–5.8%](3.7) [2.1–5.8%]
OS: Median [95% CI], mo(9.8) [7.2–11.2%](9.1) [6.0–13.0%]
6-Mo PFS [95% CI], %(37) [27–48%](29) [14–44%]

Symptom Severity

EZN-2208 was well tolerated, both as monotherapy and when combined with cetuximab. The grade 3 and 4 adverse events observed when patients received cetuximab in combination with EZN-2208 were not statistically significantly different from those observed when patients received cetuximab combined with irinotecan, as indicated in Table 3.

Table 3. Grade 3 and 4 Adverse Reactions (From First Dose to 30 Days Within in the Last Dose of the Drug) That Occurred in ≥5% of Any Arm
 No. of Patients (%)
Adverse ReactionArm A: EZN-2208Arm B: EZN-2208 + CetuximabArm C: Irinotecan + Cetuximab
Diarrhea6 (6)15 (19)9 (24)
Abdominal pain6 (6)6 (8)4 (11)
Dehydration5 (5)8 (10)3 (8)
Fatigue9 (10)5 (6)2 (5)
Nausea3 (3)7 (9)2 (5)
Vomiting3 (3)7 (9)2 (5)
Back pain3 (3)3 (4)2 (5)
Anorexia2 (2)4 (5)0 (0)
Febrile neutropenia0 (0)5 (6)1 (3)
Constipation1 (1)2 (3)2 (5)
Pain1 (1)0 (0)2 (5)
Catheter-related infection0 (0)0 (0)2 (5)
Acneiform rash, all grades1 (1)34 (42)18 (47)
Neutropenia28 (30)28 (35)6 (16)
Anemia9 (10)1 (1)5 (13)
Hypokalemia6 (6)8 (10)1 (3)
Leukopenia7 (8)7 (9)0 (0)
Alkaline phosphatase elevation3 (3)2 (3)3 (8)
Thrombocytopenia0 (0)2 (3)2 (5)
Hyponatremia0 (0)0 (0)2 (5)


Improvement in cancer therapeutics may be achieved through improved drug delivery of agents with proven, albeit modest, anticancer efficacy. SN-38, a metabolite of irinotecan, is 100 to 1000 times more active than its parent compound.[25] The poor solubility of SN-38 makes it unsuitable as a therapeutic agent. Pegylation is a method of linking polyethylene glycol to a small molecule; the consequence is an improvement in the circulating half-life of the compound coupled with associated decreased immunogenicity.

Pegylation may also be associated with superior solubility, leading to more favorable pharmacokinetics with an associated reduction in unwanted side effects. EZN-2208 was developed as a conjugate by linking SN-38 at the carbon 20 position with a 4-arm polyethylene glycol (PEG) through a glycine linker. EZN is significantly more water soluble than SN-38; the aqueous solubility of EZN-2208 is 6.7 mg/mL equivalents of SN-38, compared with 0.0072 mg/mL for SN-38.[26]

In this study, EZN-2208 was not associated with objective radiographic responses in patients with KRAS-mutant, metastatic CRC who had received prior irinotecan therapy. It is noteworthy that the patients in arm A (KRAS-mutant tumors) had proven resistance to irinotecan and oxaliplatin; currently, the efficacy of EZN-2208 in patients with KRAS-mutant CRC who are irinotecan-naive is not known. In a randomized comparison with irinotecan plus cetuximab, EZN-2208 plus cetuximab treatment was not associated with a statistically significant difference in PFS or OS. It is not clear why this agent did not demonstrate superior efficacy compared with irinotecan; however, this may be related to inadequate tumor drug distribution or possibly to unfavorable pharmacokinetics.

Drug delivery to its target will remain an important aspect of cancer treatment and may be 1 of the reasons for drug resistance. Enhanced formulations of agents with established efficacy, with a view to optimizing pharmacokinetics, safety, and efficacy of anticancer agents, should remain an investigative priority. Monotherapy studies of EZN-2208 have been completed evaluating the efficacy of the drug in the treatment of patients with breast cancer[27] and pediatric tumors (including neuroblastoma)[28]; in addition, there is an ongoing combination study of EZN-2208 combined with the antivascular epithelial growth factor monoclonal antibody bevacizumab in solid tumors.[29] Further studies will be necessary to completely evaluate the clinical efficacy of this therapeutic strategy in the treatment of patients with solid tumors.


This study was funded by Enzon Pharmaceuticals, Inc.


Tanios Bekaii-Saab has received compensation as a consultant from Bristol-Myers-Squibb, Amgen, Genentech Inc., Lilly & Company, and Sanofi SA; has received grants from Pfizer Inc. and Oncolytics Biotech Inc.; and has received compensation from Polaris for work on the Data and Safety Monitoring Board. Christopher R. Garrett, Tanios Bekaii-Saab, Theresa Ryan, George A. Fisher, Sally Clive, Petr Kavan, Einat Shacham-Shmueli, and Richard M. Goldberg have received research funding from Enzon Pharmaceuticals, Inc. Aby Buchbinder is an employee of Enzon Pharmaceuticals, Inc. and owns stock in the company.