Randomized Phase II Trial of Parsatuzumab (Anti-EGFL7) or Placebo in Combination with FOLFOX and Bevacizumab for First-Line Metastatic Colorectal Cancer.

LESSONS LEARNED: These negative phase II results for parsatuzumab highlight the challenges of developing an agent intended to enhance the efficacy of vascular endothelial growth factor inhibition without the benefit of validated pharmacodynamic biomarkers or strong predictive biomarker hypotheses.Any further clinical development of anti-EGFL7 is likely to require new mechanistic insights and biomarker development for antiangiogenic agents.
BACKGROUND: EGFL7 (epidermal growth factor-like domain 7) is a tumor-enriched vascular extracellular matrix protein that supports endothelial cell survival. This phase II trial evaluated the efficacy of parsatuzumab (also known as MEGF0444A), a humanized anti-EGFL7 IgG1 monoclonal antibody, in combination with modified FOLFOX6 (mFOLFOX6) (folinic acid, 5-fluorouracil, and oxaliplatin) bevacizumab in patients with previously untreated metastatic colorectal cancer (mCRC).
METHODS: One-hundred twenty-seven patients were randomly assigned to parsatuzumab, 400 mg, or placebo, in combination with mFOLFOX6 plus bevacizumab, 5 mg/kg. Treatment cycles were repeated every 2 weeks until disease progression or unacceptable toxicity for a maximum of 24 months, with the exception of oxaliplatin, which was administered for up to 8 cycles.
RESULTS: The progression-free survival (PFS) hazard ratio was 1.17 (95% confidence interval [CI], 0.71-1.93; p = .548). The median PFS was 12 months for the experimental arm versus 11.9 months for the control arm. The hazard ratio for overall survival was 0.97 (95% CI, 0.46-2.1; p = .943). The overall response rate was 59% in the parsatuzumab arm and 64% in the placebo arm. The adverse event profile was similar in both arms.
CONCLUSIONS: There was no evidence of efficacy for the addition of parsatuzumab to the combination of bevacizumab and chemotherapy for first-line mCRC. The Oncologist 2017;22:375-e30. © AlphaMed Press; the data published online to support this summary is the property of the authors.

RCT Entities:   Population Interventions Outcomes

Discussion

EGFL7 is a vascular‐restricted extracellular matrix protein that promotes endothelial cell adhesion and survival [1], [2], [3], [4], [5]. Parsatuzumab, a humanized anti‐EGFL7 IgG1 monoclonal antibody, selectively blocks the interaction between EGFL7 and endothelial cells, thereby potentially inhibiting vascular regrowth and further reducing tumor perfusion after antiangiogenic therapy, such as vascular endothelial growth factor (VEGF) inhibition [6]. In several xenograft and genetically engineered murine tumor models, the addition of anti‐EGFL7 enhanced the antiangiogenesis, tumor growth control, and survival associated with anti‐VEGF monotherapy [7]. Favorable tolerability and evidence of pharmacodynamic modulation and antitumor activity were observed in a phase Ib trial that evaluated parsatuzumab in combination with bevacizumab and bevacizumab/paclitaxel [7], [8].

The current study was designed to evaluate the benefit of anti‐EGFL7 when added to standard mFOLFOX6/bevacizumab in first‐line metastatic colorectal cancer (mCRC); however, no improvement in progression‐free survival (PFS) associated with parsatuzumab in comparison to placebo was observed (Figs. 1 and 2). Furthermore, no PFS benefit associated with parsatuzumab was detected in subgroups defined by Eastern Cooperative Oncology Group (ECOG) performance status, prior adjuvant therapy, number of metastatic sites at baseline, KRAS genotype, or tumor EGFL7 expression level. Of 127 patients in the intention‐to‐treat population, 115 had measurable EGFL7 and were stratified as above or below the median EGFL7 level. The adverse event profiles of the parsatuzumab and placebo arms were similar to each other and consistent with the established profile of mFOLFOX6/bevacizumab in mCRC patients. There was no evidence that the concomitant administration of parsatuzumab altered the duration or intensity of treatment with the other active study drugs. The overall treatment outcomes for the study population compared favorably with the historical performance of first‐line mFOLFOX6/bevacizumab [9], [10]. Hence, it appears unlikely that any potential activity of parsatuzumab was confounded by study conduct that resulted in compromised delivery or efficacy of the reference regimen.

Figure 1.

Kaplan‐Meier estimates of progression‐free survival. Placebo (blue) = mFOLFOX6 + bevacizumab + placebo. Parsatuzumab (red) = mFOLFOX6 + bevacizumab + parsatuzumab. +, indicates censored value on graph.Abbreviations: CI, confidence interval; mFOLFOX6, modified FOLFOX6 (folinic acid, 5‐fluorouracil, and oxaliplatin).

Figure 2.

Study design.

Abbreviations: ECOG, Eastern Cooperative Oncology Group; mFOLFOX6, oxaliplatin 85 mg/m2, 5‐FU 400 mg/m2 bolus followed by 2400 mg/m2 continuous infusion over 46 hours, folinic acid 400 mg/m2; Q14D, each 14‐day cycle.

Although anti‐EGFL7 therapy was active in preclinical models, our data in patients with previously untreated mCRC suggest that anti‐EGFL7 therapy does not add significant clinical benefit in this patient population. Any further clinical development of anti‐EGFL7 is likely to require new mechanistic insights and biomarker development for antiangiogenic agents.

Trial Information

Colorectal cancer

Metastatic/Advanced

None

Phase II

Randomized

p‐value = 0.715. Difference in ORR (95% CI): −5% (−22% to 12%)

p: .548, HR: 1.17

p: .33, HR: 1.41

PFS

Safety

Tolerability

Overall Survival

Overall Response Rate

Duration of objective response

Pharmacokinetics

Immunogenicity

Inactive because results did not meet primary endpoint

Drug Information Arm A: Placebo arm

Placebo

400 milligrams (mg) per flat dose

i.v.

Every 2 weeks until disease progression or unacceptable toxicity

Bevacizumab

Angiogenesis ‐ VEGF

5 milligrams (mg) per kilogram (kg)

i.v.

Every 2 weeks until disease progression or unacceptable toxicity.

5‐fluorouracil

400 milligrams (mg) per squared meter (m2)

i.v., bolus, 2400 mg/m2 infusion

Every 2 weeks until disease progression or unacceptable toxicity

Folinic acid

400 milligrams (mg) per squared meter (m2)

i.v.

Every 2 weeks until disease progression or unacceptable toxicity.

Oxaliplatin

Platinum compound

85 milligrams (mg) per squared meter (m2)

i.v.

Every 2 weeks for 8 cycles

Drug Information Arm B: Parsatuzumab arm

Parsatuzumab

Angiogenesis

400 milligrams (mg) per flat dose

i.v.

Every 2 weeks until disease progression or unacceptable toxicity.

Bevacizumab

Angiogenesis ‐ VEGF

5 milligrams (mg) per kilogram (kg)

i.v.

Every 2 weeks until disease progression or unacceptable toxicity

5‐fluorouracil

400 milligrams (mg) per squared meter (m2)

i.v., bolus, 2400 mg/m2 infusion

Every 2 weeks until disease progression or unacceptable toxicity

Folinic acid

400 milligrams (mg) per squared meter (m2)

i.v.

Every 2 weeks until disease progression or unacceptable toxicity.

Oxaliplatin

Platinum compound

85 milligrams (mg) per squared meter (m2)

i.v.

Every 2 weeks for 8 cycles

Patient Characteristics

74 (58.7%)

52 (41.3%)

Stage I: 1 (0.8%)

Stage IIA: 4 (3.2%)

Stage IIB: 4 (3.2%)

Stage IIIA: 2 (1.6%)

Stage IIIB: 14 (11.2%)

Stage IIIC: 6 (4.8%)

Stage IV: 94 (75.2%)

Median (range): 62 (32–80)

Median (range): See Table 1

Table 1.

Baseline patient and disease characteristics

Abbreviations: AJCC, American Joint Committee on Cancer; ECOG, Eastern Cooperative Oncology Group; UICC, Union for International Cancer Control.

0 — 66 (52.0%)

1 — 61 (48.0%)

2 —

3 —

unknown —

See Table 1

Primary Assessment Method

64

62

64

n =3 4.8%

n =37 58.7%

11.9 months, CI: 9.6, 15.8 (95% CI)

9.1 months

63

63

63

n = 3 4.8%

n= 34 54.0%

12 months, CI: 9.1, 15.8 (95% CI)

19 months, CI: 17.3, 19.0 (95% CI)

9.9 months

9.2 months

Adverse Events

Pharmacokinetics/Pharmacodynamics

Assessment, Analysis, and Discussion

Study terminated before completion

Company stopped development

Not Collected

Inactive because results did not meet primary endpoint

Antiangiogenesis therapy has shown important clinical benefits, leading to approvals of multiple VEGF/VEGF receptors inhibitors in a wide variety of tumor types. In mCRC, bevacizumab has been shown to improve overall survival and other clinical endpoints when combined with fluorouracil‐based chemotherapy as first‐line and second‐line therapy, and when continued past first progression [12], [13], [14]. Complementary targeting of other angiogenesis factors is a rational strategy to improve these outcomes; epidermal growth factor‐like domain 7 (EGFL7) has emerged as such a target. EGFL7 is a vascular‐restricted extracellular matrix protein that promotes endothelial cell adhesion and survival under stress [1], [2], [3], [4], [5]. EGFL7 is produced by endothelial cells in nascent blood vessels in tumors and other proliferating tissues, but is absent or expressed at low levels in healthy quiescent vessels and in many nonvascular cell types [2], [4], [5], [15]. EGFL7 is deposited in perivascular tracks that persist after vessel regression; vessel regrowth after antiangiogenic therapy may occur along these EGFL7‐containing extracellular matrix tracks [1], [6], [16], [17], [18], [19].

Parsatuzumab (also known as MEGF0444A) is a humanized IgG1 monoclonal antibody that selectively blocks the interaction between EGFL7 and endothelial cells [6]. In preclinical models, the addition of anti‐EGFL7 enhanced the antiangiogenesis, tumor growth control, and survival associated with anti‐VEGF monotherapy [7]. Favorable tolerability and promising evidence of pharmacodynamic modulation and antitumor activity was observed in a phase Ib trial that evaluated parsatuzumab in combination with bevacizumab and bevacizumab/paclitaxel [8]. These results led to concurrent phase II trials of parsatuzumab in combination with bevacizumab and chemotherapy in patients with mCRC in this study and in patients with advanced non‐small cell lung cancer in another study (manuscript in preparation), respectively.

In this study, 127 patients with previously untreated mCRC who were not candidates for curative‐intent metastasectomy and had no contraindications to bevacizumab were randomized to receive parsatuzumab or placebo in addition to mFOLFOX6/bevacizumab every 2 weeks until disease progression or unacceptable toxicity. Oxaliplatin was capped at 8 cycles in order to minimize discontinuation of the study regimen due to chemotherapy‐related adverse events, as the duration of treatment with bevacizumab appears to be important to maximize its therapeutic benefit [10]. The protocol‐specified primary analysis was performed after the occurrence of 62 PFS events and a minimum of 12.5 months of follow‐up for all patients. The PFS hazard ratio was 1.17 (95% confidence interval [CI], 0.71–1.93; p = .548), with median PFS of 12 months for the parsatuzumab arm versus 11.9 months for the placebo arm. An exploratory analysis that included time points subsequent to metastasectomy for the 10 patients (6 in the placebo arm and 4 in the parsatuzumab arm) who became eligible for resection while on study treatment was also performed. The results of this sensitivity analysis were similar to those of the primary analysis (PFS hazard ratio of 1.11; median PFS of 12.9 months for the parsatuzumab arm versus 12.6 months for the placebo arm). With a total of 27 deaths reported, the immature overall survival (OS) hazard ratio was 0.97 (95% CI, 0.46–2.1; p = .943). The overall response rate was 59% in the parsatuzumab arm and 64% in the placebo arm. Furthermore, the PFS hazard ratio was not statistically significant in subgroups defined by ECOG performance status (0 vs. 1), history of adjuvant therapy (yes vs. no), or number of metastatic sites at baseline (1 vs. >1) or KRAS genotype (wild‐type vs. mutant); however, KRAS status was available for only 64 of the 127 patients. Based on a prior phase Ib study in which high tumor EGFL7 expression was found to be associated with lack of response (data on file), subgroup analysis was also performed based on EGFL7 expression measured in archival tumor specimens (above median vs less than or equal to median), but with no significant difference in PFS hazard ratio observed.

The adverse event profiles of the parsatuzumab and placebo arms, including the number of protocol‐specified adverse events of interest and events leading to treatment discontinuation, were similar to each other and consistent overall with the established profile of mFOLFOX6/bevacizumab in mCRC patients [12]. There was no evidence that the concomitant administration of parsatuzumab altered the duration or intensity of treatment with the other active study drugs. No difference in bevacizumab, 5‐fluorouracil, or oxaliplatin pharmacokinetics was observed between the treatment arms. Moreover, the overall treatment outcomes for the study population compared favorably with the historical performance of first‐line mFOLFOX6/bevacizumab [9], [10]. Hence, it appears unlikely that any potential activity of parsatuzumab was confounded by study conduct that resulted in compromised delivery or efficacy of the reference regimen.

These data highlight the challenge in achieving meaningful improvement in front‐line outcomes for patients with mCRC, a disease for which no new therapeutic class has been introduced since the U.S. Food and Drug Administration approvals of bevacizumab (anti‐VEGF) and cetuximab (anti‐epidermal growth factor receptor) in 2004. These phase II results for parsatuzumab underscore the difficulty of developing agents whose mechanism predicts (1) activity only in combinations (i.e., with bevacizumab) but not as a single agent and (2) enhanced survival in the absence of increased response rates. Neither validated pharmacodynamic biomarkers that reflect modulation of the targeted pathway nor strong predictive biomarker hypotheses were available to guide the development of parsatuzumab. Despite intensive efforts, such biomarkers for anti‐VEGF agents in colorectal cancer have remained elusive. Any further clinical development of anti‐EGFL7 is likely to require new mechanistic insights and biomarker development for antiangiogenic agents.