PRECLINICAL PHARMACOLOGY OF XmAbTM2513 AN FC ENGINEERED HUMANIZED ANTICD30 MAB Phil Hammond, Greg La

1
PRECLINICAL PHARMACOLOGY OF XmAbTM2513AN FC
ENGINEERED HUMANIZED ANTI-CD30 MABPhil Hammond,
Greg Lazar, Sher Karki, David Carmichael, Seung
ChuXencor, Inc. 111 W. Lemon Ave., Monrovia, CA
91016
XmAb2513
Abstract
Was humanized with an increase in affinity
Has Increased Fc? Receptor binding affinity

• Humanized Version Of Murine Mab AC10
• Cluster C epitope
• XmAb Fc Region
• Engineered for high affinity Fc? receptor binding
• Antiproliferative Against HL And ALCL Cell Lines
• Blocks CD30 Ligand Binding
• Has Improved Effector Functions
• Cytotoxicity (ADCC)
• Phagocytosis (ADCP)
• Exhibits No Complement Dependent Cytotoxicity
  (CDC) Activity
• Is Active In Subcutaneous Xenograft Models Of HL
• Well Tolerated In Toxicology Studies
• PK Suggests at Least Every Other Week Dosing

XmAb
VH, 27 mutations VL, 15 mutations
XmAb2513 is a new humanized monoclonal antibody
(mAb), to the human cell surface antigen CD30,
with an engineered Fc region to enhance
recruitment of effector cells and potentiate
anti-tumor efficacy. It is being developed for
CD30-positive (CD30) diseases such as Hodgkin
Lymphoma (HL) and anaplastic large cell lymphoma
(ALCL). XmAb2513 was derived from the murine mAb
AC10 by humanizing the variable domain using the
method of human sequence content optimization
while retaining high binding affinity for CD30.
In addition, the Fc region was engineered to
increase the binding affinity for all Fc?
receptors (Fc?Rs). Using biacore measurements,
XmAb2513 was determined to have a binding
affinity of 465 pM for CD30. The Fc engineering
increased the binding affinity of XmAb2513 for
Fc?RI, Fc?RIIa. Fc?RIIb, and Fc?RIIIa by between
3- and 26-fold when compared to the binding
affinity of an unengineered comparator
mAb. XmAb2513 retains the potent
anti-proliferative activity exhibited by the
parental antibody against HL and ALCL cell lines.
In addition, as a result of the Fc engineering,
XmAb2513 exhibited superior antibody-dependent
cell-mediated cytotoxicity (ADCC), mediated by NK
cells that primarily express Fc?RIIIa, when
compared to the unengineered mAb. The mean
efficacy (percentage of cells specifically lysed)
improvement was 4.7fold and the mean potency
(concentration giving 50 of maximal lysis)
improvement was 2.4-fold over the unengineered
mAb. XmAb2513 was also 2.1-fold more
efficacious than the unengineered mAb in
antibody-dependent cell-mediated phagocytosis
(ADCP) assays using Fc?RIIa/b and Fc?RIIIa
expressing macrophages. The in vivo anti-tumor
activity of XmAb2513 was evaluated using
subcutaneous xenograft models in SCID mice.
Statistically significant reductions in tumor
growth, together with enhanced survival, were
observed at 3 mg/kg while at 10 and 30 mg/kg
XmAb2513 was even able to eliminate established
tumors. XmAb2513 has been successfully engineered
to possess multiple mechanisms of action,
including ADCC and ADCP, with significant
improvement over those of an unengineered IgG1
mAb comparator. Additionally, XmAb2513 has
potent anti-proliferative effects and was
efficacious against HL xenografts. These in
vitro and in vivo pharmacology data provide a
rationale for the clinical testing of XmAb2513 in
patients with CD30 hematologic malignancies.
IgG1
Murine mAC10 Chimeric cAC10 Humanized xAC10
Log10 scale
Human String Content VH 0.84, VL 0.93
4x increase in antigen binding affinity
Measured by competitive AlphaScreen assay

• Binding affinities were determined using a
  biacore method.
• Antibodies were immobilized on a Protein A chip
  and receptors included in the mobile phase.
• Dissociation constants were determined from
  Langmuir curve fit.

Lazar et al, Mol Immunol. 2007
Lazar et al. PNAS 2006
Increases cytotoxicity for Fc?RIIIa allotypes
Is anti-proliferative against HL and ALCL
Mediates enhanced cytotoxicity NK cell dependent
Blocks CD30 Ligand Binding
Efficacy Improved 4.7 1.4 - Fold Potency
Improved 2.4 1.4 - Fold
Anti-human IgG (10x excess)
Karpas 299 (ALCL)
Anti-CD30 (100 ng/ml)
LDH Release Assay
Cell Counting Assay
Cytotoxicity
L540 (HD)
2.2
2.0
1.8
Differences in anti-proliferative effects between
anti-CD30 antibodies have been attributed to
different epitope clusters (Horn-Lohrens et al)
AC10 is a cluster C antibody while 5F11 is a
cluster A antibody to CD30. To determine antibody
effects on cell proliferation, either Karpas299
or L540 cells were grown for 4 days in the
presence of antibody at varying concentrations
with 10x molar access of cross linking antibody.
Cell growth was measured using an ATP dependent
luminescence assay.
Proliferation (RLU x107)
1.6
XmAb2513
1.4
Antibody dependent cell-mediated cytotoxicity was
measured by both lactate dehydrogenase (LDH)
release and cell counting (Guava). Human PBMC
effector cells were purified from a Leukopack
using a ficoll gradient. L540 Hodgkin Lymphoma
target cells were seeded into 96-well plates at
10,000 (LDH) or 15,000 (Sorting) cells/well and
opsonized using antibody at the indicated
concentration. Effector cells were added at 25x
ET and the plate incubated at 37?C for 4 hrs
prior to assay. For LDH, data were normalized to
maximal (Triton X100 lysis of target cells alone)
and minimal (PBMCs alone) lysis. Reported
cytotoxicity was derived from LDH data.
cAC10-IgG1

• CD30L binding assay was performed using L540
  cells expressing CD30.
• Fixed concentrations of antibody-crosslinked
  his-CD30L were added as indicated in the legend.
• A dose response curve of XmAb2513 was also
  added as indicated on the X axis.
• Binding of CD30L to cells was detected using
  flow cytometry and reagents for the cross-linking
  antibody.

xAC10-IgG1

• Antibody dependent cell-mediated cytotoxicity
  (ADCC) assays were performed with donor PBMCs
  isolated from Leukopaks.
• The target for these data was the HL L540 cell
  line.
• Donor allotype was determined by sequencing
  after PCR from genomic DNA.

5F11
1.2
BSA
hIgG
1.0
0.1
1
10
100
1000
Antibody concentration (ng/ml)
Increases phagocytosis for Fc?RIIa allotypes
Was well tolerated in pharmacokinetic and
toxicology studies
Caused Tumor Regression in L540 Xenograft
Increased Survival in an L540 Xenograft
ADCP Improved 2.2 0.7 - Fold
Low Range
High Range
Tumor regressions 0/9 0/9 4/9 5/9

• Was well tolerated on a Q5D X6 dosing regimen
  in cynomolgus monkeys.
• Terminal T½ was between 12.6 And 17.1 days.
• Exposure to XmAb2513 was proportional to dose
  (after 1st dose).

Conclusions
The XmAb2513 engineered antibody has been shown
to have enhanced potency and efficacy as compared
to IgG1 antibodies. This was observed both in
ADCC assays as well as in antiproliferation
assays where the antibody crosslinking required
for an antiproliferative effect was mediated by
Fc receptor binding. Published results from
clinical trials with bispecific antibodies
directed to CD30 provide clinical evidence in
Hodgkin Lymphoma that enhanced recruitment of
effector function is a successful means of
generating a cytotoxic antibody (Hartmann, 2001
Borchmann, 2002). However, manufacturing
limitations prevent bispecifics from being
practical in widespread use. Preclinical results
with XmAb2513 support further testing in the
clinic to validate the role of enhanced Fc
effector function.
For animal 3503 tumor began to regrow on day 43.
On day 45 animal 3508 no longer had measurable
tumor
References
Donor Duplicates
Donor Duplicates
Lazar et al. A molecular immunology approach to
antibody humanization and functional
optimization. Mol Immunol. 2007 Mar44(8)1986-98
Lazar et al. Engineered antibody Fc variants
with enhanced effector function. Proc Natl Acad
Sci U S A. 2006 Mar 14103(11)4005-10 Borchmann
et al. Phase 1 trial of the novel bispecific
molecule H22xKi-4 in patients with refractory
Hodgkin lymphoma. Blood. 2002 Nov
1100(9)3101-7. Hartmann et al. Anti-CD16/CD30
bispecific antibody treatment for Hodgkin's
disease role of infusion schedule and
costimulation with cytokines. Clin Cancer Res.
2001 Jul7(7)1873-81

• Female mice from the ICR-SCID strain were
  injected subcutaneously with 5x107 cells of the
  L540 HL line.
• Animals were randomized to treatment groups when
  established tumors of 50-100 mm3 were measured by
  microcaliper.
• Dosing was performed by intraperitoneal
  injection every 4 days for 10 doses (Q4Dx10).
• Animals were sacrificed when tumors reached 1500
  mm3, for humane considerations, or at the
  completion of the study on day 70.

• Antibody dependent cell-mediated phagocytosis
  (ADCP) assays were performed with
  monocyte-derived macrophages cultured from donor
  PBMCs isolated from Leukopaks.
• The target for these data was the HL L540 cell
  line.
• Donor allotype was determined by sequencing
  after PCR from genomic DNA.
• No H/H donors were identified in the donor
  populations used for these studies.

• During the study, the tumors of several animals
  regressed to the point of being unmeasurable or
  in fact undetectable.
• For all except one of the animals, indicated in
  the 10 mg/kg group, the effect was durable.

Acknowledgement
We wish to thank the many Xencor employees who
supported this work.