Targeting angiogenesis in sarcomas

1
Targeting angiogenesis in sarcomas

• Robert G. Maki, MD PhD
• Memorial Hospital
• New York, NY, USA

2
Examples of commercially approved
anti-angiogenic agents

• Taxanes
• Thalidomide
• Cyclo-oxygenase inhibitors
• PPAR-gamma inhibitors
• Bisphophonates
• Interferon-alpha
• Bevacizumab, VEGF-Trap
• Sunitinib, sorafenib

3
What is anti-angiogenic therapy?
VEGF, PDGF
X
Endothelial precursor
Tumor blood vessel
4
Angiogenesis more than one process

• Vasculogenesis
• Angiogenesis
• SPROUTING angiogenesis
• SPLITTING (intussiceptive) angiogenesis
• Arteriogenesis

5
Sprouting angiogenesis
6
Splitting / intussusceptive angiogenesis
7
Functionally speaking
Pre
8
Key angiogenic factors
erythropoietin, AT-II, endothelins,
Ribatti D et al. Pharmacol Rev 2007 59185
9
Knockout mice

• VEGF/- embryonic lethal, impaired
  vasculogenesis
• FGF-2-/- vasculogenesis intact, decreased
  smooth muscle tone
• PDGF-B-/- no microvascular pericytes, capillary
  aneurysms, endothelial hyperplasia, hemorrhage
• EphB(1,2)-/- embryonic lethal vasculogenesis
  or cardiac development impaired
• FAK-/- (in endothelial cells) embryonic lethal

VEGF targeting
10
PDGF-B/ or PDGF-B-/- mice
Pericytes missing in d12 embryos in brain
capillaries
Lindahl P et al. Nature 1997 277242
11
Familial syndromes and their mouse counterparts

• Hereditary hemorrhagic telangiectasia (AVMs
  without intervening capillaries)
• Type I Endoglin ENG, 9q33-34
• Type II activin receptor-like kinase 1
  (ACVRL1ALK1), 12q13
• Both have effects via TGF-beta signaling
• Cavernous malformations (CNS)
• 7p, 7q, 3p families (CCM1, 2, 3) mutation in
  KRIT1

• ENG-/- or ACVRL1-/- mice have defective capillary
  formation, defective smooth muscle recruitment,
  defective modeling, cardiac defects
• Conversely, CCM1-/- mice have intact capillary
  formation but have defective arterial generation,
  and die as embryos

Urness LD et al. Nat Genet. 2000
26328 Whitehead KJ et al. Development 2004 131
1437
12
We have what we have

• Taxanes angiosarcoma
• Thalidomide Uterine LMS (GOG)
• Dose escalate 200 mg 1000 mg
• n29 evaluable
• Median PFS 1.9 mo 2 patients with SD at least 6
  months
• In another small phase II study,
  ULMS/carcinosarcoma patients Rx with thalidomide
  had median PFS 1.8 months, RR0.
• Case reports of responses angiosarcoma,
  osteosarcoma
• COX inhibitors nil except desmoids (?)
• Interferon-alfa giant hemangiomas, GCT of bone,
  desmoid tumors ?, ASPS?

Fata F et al. Cancer. 1999 862034 Penel N et
al. J Clin Oncol (ASCO) 2007 25(18S)
10002 McMeekin et al. Gynecol Oncol. 2007106596
13
TNP-470

• Fumigillin analog (antibiotic from Aspergillus
  fumigatus fresenius)
• Inhibits FGF-2 associated endothelial
  proliferation
• Inhibits vasculogenesis in CAM and retinal models
• Phase I study in Kaposi sarcoma, weekly dosing
• 10-70 mg/m2 IV weekly, well tolerated at all
  doses
• n38
• 7 with PR (18)

Dezube BJ et al. J Clin Oncol 1998 161444
14
ABT-510

• Thrombospondin mimetic (1000x more potent)
• Decreased EC proliferation, EC migration,
  capillary tube formation, and vasculogenesis in
  CAM and retinal models
• Phase I n15 sarcoma patients, 1 PR, 4 SD gt 6 mo
• Randomized Phase II 20 mg sc daily v. 100 mg sc
  BID
• n88, 73 with prior chemotherapy
• 6 month PFS 34 for patients with no prior Rx,
  41 for those who had received prior chemotherapy
  
• Dose did not affect PFS rate
• Low patient CEC concentration was a predictor of
  longer time to progression, independent of
  treatment

Baker LH et al, J Clin Oncol 2005 (ASCO)
23(16S) Abstr 9013
15
Doxorubicin bevacizumab

• Phase II study, Simon 2-stage design
• Dox 75 mg/m2 day 1, Bev 15 mg/kg day 1, q21d
• n17
• Leiomyosarcoma n11
• no angiosarcomas
• 2 PR / 17 not sufficient to continue study (12)
• 65 on therapy 12 weeks
• 6 patients with G2 cardiac toxicity, 4 of 6
  recovered
• This was not an adequate test of the efficacy of
  the combination of bevacizumab and chemotherapy
• A study of bevacizumab in angiosarcoma (Wash U.,
  St. Louis)
• Bevacizumab Gemcitabine/Docetaxel (Uterine
  sarcomas)

DAdamo DR et al. J Clin Oncol 2005 23 7135
16
Sunitinib

• Are small molecules any better than antibodies?
• Phase II study 37.5 mg oral daily
• Three strata, each with Simon 2-stage design
• Those with known prior responses to kinase
  targeted agents (angio, LMS, DFSP etc)
• Other sarcomas
• Chordoma
• n44 total accrual to date
• Only one RECIST responder, DSRCT, 0/6 chordomas
  with PR on study
• near PR for GCT of bone metastatic to lung,
  SFT-HPC
• Shorter PFS for LMS compared to sorafenib

Keohan ML, George S et al. CTOS 2007
17
Pazopanib (GW-786034)

• Antagonist of VEGFR1,2,3 c-kit PDGFRA, B
• Prior Rx 1 combination or 2 single agents
• 4 strata leio, lipo, synovial, other
• Simon 2-stage design for each stratum
• n88 in 2007 n80 evaluable
• No skin toxicity! fatigue, HTN, diarrhea
• 27 / 80 had no progression at 12 weeks (1
  endpoint)
• Only liposarcoma arm did not complete accrual
  owing to low rate of PFS at 12 weeks

Sleijfer S et al. J Clin Oncol 2007 (ASCO)
25(18S) 10031
18
Sorafenib

• Phase II multicenter study, 6 arms
• MFH, LMS, Synovial, Angio, MPNST, other
• Each its own Simon 2-stage design
• n147, over 500 cycles of therapy administered
• 2 / 37 LMS with RECIST PR
• 5 / 37 angiosarcomas with PR (one became clinical
  CR)
• Toxicity skin, hypertension, fatigue dose
  reduction from 400 mg oral BID to 400 mg oral
  daily in 52
• 18 required 2nd dose reduction
• Median overall survival 12.9 months

DAdamo DR et al. J Clin Oncol (ASCO) 2007
25(18S) 10001
19
Progression Free Rate at 12 and 24 weeks
progression free at Histology
n 12 weeks 24
weeks Leiomyosarcoma 37 54 40
Angiosarcoma 44 54 24 Other
19 61 28 MPNST 14 63
26 MFH 13 29 14 Synovial Sarcoma
13 44 11
EORTC Active 2nd line agents (Ifosfamide,
DTIC) 40 14
Van Glabbeke Eur J Cancer 2002 38543
20
Heterogeneity within angiosarcomas

• Response rate for angiosarcomas of the head and
  neck is higher that that for those below the
  clavicles
• Perhaps it is reasonable to think endothelial
  cells from different anatomic sites are distinct
  from one another, as fibroblasts appear to be

Fury MG et al. Cancer J. 2005 11241 Chang HY
et al. PNAS 2002 9912877
21
Finding genes involved in sarcomagenesis
Subtype-specific genes (those driving the
phenotype)
Gene expression pattern
Differentiation (unrelatedness to normal tissue
counterpart or to stem cell)
22
Finding genes involved in sarcomagenesis
Subtype-specific genes (those driving the
phenotype)
Gene expression pattern
Differentiation (unrelatedness to normal tissue
counterpart or to stem cell)
23
Summary

• Little single agent activity of antiangiogenic
  agents in sarcomas
• Few data regarding combinations with cytotoxic
  agents
• Angiogenic agents appear useful against
  angiosarcomas
• Role of anatomical location?
• The heterogeneity of vascular tumors appreciated
  from model systems may inform future clinical
  trials of anti-angiogenic agents

24
Conclusion

• The complexity of angiogenesis implies we have
  only started to tap the potential of this
  approach to therapy for sarcomas