Final Results of a Phase 1b of Tarextumab (TRXT, OMP-59R5, anti-Notch2/3) in Combination with Nab-paclitaxel and
Gemcitabine (Nab P+Gem) in Patients (pts) with Untreated Metastatic Pancreatic Cancer (mPC): ALPINE Study
Eileen O’Reilly1, Lon Smith2, Johanna C. Bendell3, John Strickler4, Mark Zalupski5, William Gluck6, Ann M. Kapoun7, Lu Xu7, Wan-Ching Yen7, Dawn Hill7, Lei Zhou7, Jakob Dupont7, Allen Cohn8
Sloan-Kettering Cancer Center, New York, NY; 2South Texas Accelerated Research Therapeutics, San Antonio, TX; 3Sarah Canon Research Institute, Nashville, TN; 4Duke University, Durham, NC; 5 University of Michigan Medical Center, Ann Arbor, MI;
6Greenville Hospital System, Greenville, SC; 7OncoMed Pharmaceuticals Inc, Redwood City, CA; 8Rocky Mountain Cancer Center, Denver CO
OMP-PN17
TRXT showed superior anti-tumor
activity when it is combined with
Nab-P+Gem vs combined with
gemcitabine alone.
TRXT at 40 mg/kg every other
week in combination with weekly
nab-paclitaxel (10 mg/kg) and
gemcitabine (5 mg/kg) resulted in
complete tumor regression,
suggesting that TRXT at 40 mg/kg
was a maximum efficacious dose.
•
Drug exposure of TRXT at 40
mg/kg in mouse is estimated to be
equivalent of 15 mg/kg in patients.
900
600
Preferred Term
60
Days Post Treatment
0.8
0.6
0.4
mPFS in MPACT (5.5mo)
0
Waterfall Plot (n=37 evaluable)
Preferred Term
100
200
300
NOTCH3>50%
NOTCH3<50%
400
mOS in MPACT (8.5mo)
0
100
200
Time To Progression
•
•
300
400
500
Survival Days
Von Hoff et al, NEJM 2013: GEM/Nab-P: mPFS 5.5mos; mOS 8.5mos
Potentially longer survival was noted in Notch3 high pts receiving GEM/Nab-P/TRXT (at 5-15mg/kg)
Patient Variables
• The incidence of diarrhea was less than expected based on single agent data. Co-administration of
chemotherapy with TRXT may lessen GI tract goblet cell hyperplasia and thus allow for higher doses of TRXT.
• The incidence of diarrhea, nausea and vomiting were 83.3%, 83.3% and 66.7% respectively in 15 mg/kg dose
cohort compared to lower dose levels suggest TRXT-dose dependent effect. However, the severity of diarrhea,
nausea and vomiting were mostly Grade 1 or 2. Only three subjects in 15 mg/kg dose cohort required TRXT
dose reduction/discontinuation in response to Grade ≥ 2 diarrhea.
• The number of pts at each dose cohort was too small to determine if the addition of TRXT increased the
incidence of cytopenias. Platelet count recovery was noted when chemotherapy was held but TRXT was
continued.
* Data cutoff as of November 18, 2014
0
40
NOTCH3>50%
NOTCH3<50%
0.0
Baseline Characteristics (n=40)*
0.2
Notch3 low (n=12)
AEs Related to Tarextumab Occurring in
≥ 10% of Pts (n=40)*
20
Notch3 hi (n=12)
0.0
Primary: MTD
Secondary:
•
PK of Tarextumab
•
Immunogenicity
•
Safety and tolerability of the combination
Exploratory:
•
PD biomarkers, including Notch pathway related genes and proteins and
circulating tumor cells
•
Progression free survival (PFS) in pts with Notch3 high
Notch3 low (n=12)
OS
Primary: PFS: all pts & in Notch 3 high pts (25, 50, 75th percentiles)
Secondary:
•
Overall survival (OS), 12 months OS, overall response rate
(ORR) in all pts, and in Notch 3 high pts
•
Safety and tolerability of the combination
Exploratory:
•
PD biomarkers, including Notch pathway related genes and
proteins and circulating tumor cells
•
CA19-9 (≥50% reduction from the baseline) in all pts and in
Notch 3 high pts
300
0
0.8
Objectives:
Objectives:
Control mAb
Anti-Notch2/3
Gem+Abraxane
Anti-Notch2/3+Gem/Abrx
1200
50th Percentile
Notch3 hi (n=12)
Pancreas Tumor
Poorly differentiated
B-Raf mutation
K-Ras wt
No Treatment
50th Percentile
0.6
TRXT IV Days 1 and 15, Gem at 1000 mg/m2 on
Days 1, 8 and 15 of q28 days
1500
Tumor Volume, mm3
•
Chemo+/- mAb
(day 0-56)
Nab-P+Gem + Placebo
5 mg/kg(n=4)
TRXT in Combination with Nab-P+Gem in
Patient-Derived Pancreatic Tumor Xenografts
OS
1:1
5 mg/kg(n=4)
2.5 mg/kg (n=5)
Kaplan-Meier Analysis
PFS
1.0
7.5 mg/kg(n=6)
The maximum tolerated doses (MTDs) were: 2.5 mg/kg QW, 7.5 mg/kg Q2W and 7.5 mg/kg Q3W
Grade 3 diarrhea was the most common dose limiting toxicity (DLT), which was consistent with preclinical findings
of diarrhea from goblet cell hyperplasia. Diarrhea occurred in 81% of pts. The other commonly reported events that
were considered related to TRXT were fatigue, nausea and decreased appetite
Notch pathway and CSC pathways were reduced with TRXT treatment in serial patient tumor and surrogate tissue
samples
All pts included received TRXT with doses ranging from 5mg/kg Q2W to 15mg/kg Q2W (n=24) with GEM/Nab-P
All Pts included had PFS and OS data and had tumors evaluable for Notch3 status
Patient tumors analyzed for baseline Notch3 gene expression (potential predictive biomarker in ALPINE Ph2)
Notch3 gene expression cut-offs tested: <25%, ≥25%; <50%, ≥50% (SHOWN BELOW); <75%, ≥75%
Kaplan-Meier Analysis was performed for progression free survival (PFS) and overall survival (OS)
Nab-P+Gem + Tarextumab
10 mg/kg(n=6)
• TXRT is currently being evaluated in first-line pancreatic cancer (ALPINE study) and in first-line
extensive-stage small cell lung cancer (PINNACLE study).
• The ALPINE study is a Ph 1b/2 trial of TRXT in combination with nab-paclitaxel and gemcitabine
(Nab-P+Gem) in pts with untreated metastatic pancreatic cancer. The enrollment of the Phase
1b portion of the study has been closed and is ongoing for the Phase 2 portion of the study.
• Here we report the final clinical results of Phase 1b portion of the trial.
•
15 mg/kg
(n=12)
12.5 mg/kg(n=3)
•
•
•
•
•
0.4
•
TRXT IV Day 1 and 15, Nab-P at 125 mg/m2,
followed by Gem at 1000 mg/m2 on Days 1,
8 and 15 of q28 days
Preferred Term
Tumor Notch3 Level and Timed Endpoints
0.2
•
•
Phase 1b Dose escalation:
• Untreated metastatic pancreatic cancer
• DLT assessed within first 28 days
Randomized Phase 2
• N=124
• 1st line metastatic pancreatic cancer:
Subject Time on Study (n=40)*
PFS
• The Notch pathway plays a central role in embryonic development, the regulation of stem and
progenitor cells, and is implicated centrally in many human cancers, including pancreatic cancer,
and small cell lung cancer (SCLC).
• Notch3 leads to poor survival & chemo resistance in pancreatic cancer (Mann, PloS ONE, 2012)
• Approximately 70% of pancreatic cancer over-expresses Notch3 (by gene expression RUO test).
• Tarextumab (TRXT) is a fully human IgG2 that inhibits signaling of Notch2 and Notch3 receptors.
• TRXT has anti-cancer stem cell (CSC) and anti-pericyte activity in tumor vasculature.
• The first-in-human Ph1a trial of TRXT in patients (pts) with refractory solid tumors (Smith,
EORTC-NCI-AACR 2013) showed:
AEs Occurring in >25% of Pts (n=40)
All Grades Regardless of Relationship*
[email protected]
1.0
Study Schema and Objectives
Background
% change from baseline
1Memorial
Copies of this poster obtained through Quick Response (QR)
Code are for personal use only and may not be reproduced
without permission from ASCO and the author
80
OMP-59R5 Dose Level (mg/kg)
Gemcitabine: 20 mg/kg; Nab-P 30 mg/kg
Anti-Notch2/3: 40 mg/kg, q2wk
• Pancreatic Cancer patient factors were assessed in a COX proportional model.
• The factors assessed were: Nab-P treatment; Tumor Notch3; CA19-9 tumor marker; baseline serum albumin
• All 40 ALPINE Ph1b pts (G/TRXT & G/Nab-P/TRXT) included in model (all pts got TRXT from 2.5-15mg/kg
Q2W)
PFS
GEM/TRXT
GEM/Nab-P/TRXT
•
•
•
PR
SD
PD
Total
0
6 (75%)
2
8
11 (38%)
10 (35%)
8
29
OS
Tumor assessments every two cycles ± 5 days (56±5 days); unconfirmed responses
Green Bars: GEM + TRXT pts; Purple Bars: GEM/Nab-P + TRXT pts
Pts received TRXT from 2.5 to 15mg/kg Q2W
* Data cutoff as of November 18, 2014,
Correlation of TRXT Efficacy with Notch 3
Expression: Patient-derived Pancreatic Tumor
Xenografts
Anti-Notch2/3 Efficacy in patient-derived
pancreatic tumor xenografts
Tumor Notch3 Level and
RECIST Response
Tarextumab Treatment Disposition (n=40)*
Preferred
Term
32 of 40 patients in ALPINE Ph1b with baseline tumor from a metastatic site evaluable for
Notch3 gene expression testing (Notch3 RUO-validated RT-PCR assay, Almac Diagnostics)
Summary
• TRXT (anti-Notch2/3) in combination with Nab-P+GEM was well tolerated up to 15
mg/kg every other week. One DLT of Grade 3 diarrhea occurred in one of the 12
subjects treated at 15 mg/kg dose cohort.
Tumor-expressed Notch3 in Responder
vs. Non-responder Xenografts
p = 0.0092
• Diarrhea, fatigue and anemia were the most common TRXT treatment-related
toxicities, and the events were mostly Grade 1 or 2, and easily managed with
supportive care.
• The Phase 2 dose of TRXT in combination with Nab-P+Gem is15 mg/kg Q2W
• RECIST response (unconfirmed) to GEM/Abrax/TRXT (at 5-15mg/kg) was 38%
• The median PFS and OS values for GEM/Abrax/TRXT (at 5-15mg/kg) were 5.6mos
and 11.6mos for all patients treated, respectively
Non-responder
• The tumor NOTCH3 biomarker (gene expression) was assessable in 32 of 40 pts
Responder
*p<0.05 vs. Gem as a single agent
•
10 primary patient-derived pancreatic tumor xenografts were tested for efficacy in response to TRXT (OMP-59R5,
Anti-Notch2/3) plus Gemcitabine
a.
b.
c.
•
Significant correlation was found between the levels of tumor-derived Notch3 and the efficacy of TRXT. Responder
tumors have higher levels of Notch3 compared to non-responders when treated with Anti-Notch2/3 plus Gemcitabine
d.
e.
Pt withdrew consent after one dose of TRXT due to rapid clinical deterioration
Pt died of gastrointestinal perforation associated with underlying disease
Pt discontinued TRXT treatment with Grade 2 recurrent tarextumab treatment-related infusion reaction, but continued
Nab-P+Gem until disease progression on study
Pt with G3 diarrhea that lasted for 5 days
One pt with recurrent Grade 3 or 4 diarrhea (considered related TRXT), one with concurrent illness (unrelated to study
treatment)
* Data cutoff as of November 18, 2014,
• Consistent with what has been reported in first-in-man study: diarrhea, fatigue, nausea and decreased appetite
were commonly reported events that were considered related to TRXT. However, the incidence of diarrhea
was less than expected in contrast to single agent data.
• Grade 3 TRXT treatment-related adverse events were not common.
• The frequency of Grade 3 diarrhea was less than expected even at the dose up to 15. One pt in 10 mg/kg
dose cohort had Grade 3 diarrhea lasting less than 48 hours, and managed with Imodium. Two pts in 15
mg/kg had Grade ≥ 3 diarrhea, one of them had Grade 3 diarrhea that lasted for 5 days
* Data cutoff as of November 18, 2014
• Higher response rate and longer survival was noted in patients with Notch3 high
tumors receiving GEM/Nab-P/TRXT (at 5-15mg/kg). These observations must be
verified in the placebo-controlled, randomized Ph2 setting.
• The randomized, double blinded Ph2 portion of the ALPINE study is currently
ongoing and enrolling at 36 clinical sites in the United States and data is anticipated
in early 2016 for both the intent to treat population, as well as, the tumor Notch3
biomarker positive patients