Lessons Learned in T1 Research: mouse to human

1
Lessons Learned in T1 Research mouse to human

• Jean Y. Tang MD PhD
• Assistant Professor
• Department of Dermatology

2
Mechanism of disease
In vitro experiments
Animal studies
Human clinical trials
New drugs
Epidemiological studies
3
Lessons learned challenges

• Jack of all trades, Master of none
• Journals
• Conferences
• Students and trainees
• Not that many role models
• find a true believer and the experts
• Slower time to publication
• Grants enthusiasm from NIH

4

• KL2/K23 Mentors
• Ervin Epstein, Childrens Hospital Oakland
  Research Institute
• Mary-Margaret Chren, Dept of Dermatology, UCSF
• Charles McCulloch, Steve Cummings, Dept of
  Epidemiology and Biostatistics, UCSF

5
Translational Research in BCC
Screen for drugs in cell lines and Ptch1/- mice
Clinical trials in PTCH1 /- BCNS patients
Observational studies at Kaiser/UCSF
6
Summary of chemopreventive agents against BCCs
In vitro Mice Observational studies Clinical Trials
Celecoxib (oral) ? ? mixed ? in subset
Statin No No No N/A
Vitamin D3 ?? ?? ? Biomarker study
Hh Antagonist (Genentech) ??? ???? none Enrolling RCT
Itraconazole (antifungal) ?? ?? none Biomarker study
7
Challenge Lesson Learned
Celecoxib (oral) Combining mouse and RCT Statistics, finding the experts
Statins Difficulty in combining mouse and epidemiologic results (Kaiser) OK to disagree
Vitamin D3 Difficulty of doing it all yourself lab, mouse, pilot clinical trial 1Mentor and his lab SFCC, EpiBio mentors -Public databases
Hh Antagonist Collaborations with Genentech -Good to have lab skills -RCT design
Itraconazole Collaborations with basic lab FDA approved drug -Nice to collaborate -Time to start the pilot trial
8
Epidemiology of BCC

• 1 million BCC cases per yr in US
• Estimated annual incidence of 0.1 to 0.5
• Rare risk of metastasis lt 0.5
• 5th most costly cancer for Medicare
• The age-adjusted incidence per 100,000 white
  individuals 475 cases in men, 250 cases in women
  
• The estimated lifetime risk of BCC in the white
  population is 33-39 in men and 23-28 in women.
• Risk of second BCC 44 in 3 yr

9

• BCC basic science
• Almost all BCCs have mutations in PTCH1 tumor
  suppressor gene
• All BCCs have increased Hedgehog signaling

10
Hedgehog signaling pathway regulates cell
proliferation and growth
11
Basal cell nevus syndrome
Basal cell nevus syndrome are PTCH1/-
12
Ptch1/- mice mimic BCNS phenotype develop BCC
tumors after IR or UV treatment Goodrich and
Scott, Science 1997Aszterbaum, Oro, Scott,
Epstein Nature Medicine 1999
13
Mechanism of disease
Roadmap for finding new therapeutics
Hedgehog pathway
BCC cell lines
In vitro experiments
Ptch1 /- mice
Animal models
PTCH1/- Basal cell nevus syndrome patients
Human clinical trials
Patients with sporadic BCCs
Epidemiological studies
14
Summary of chemopreventive agents against BCCs
In vitro Mice Observational studies Clinical Trials
Celecoxib (oral) ? ? mixed ? in subset
Statin No No No N/A
Vitamin D3 ?? ?? ? Biomarker study
Hh Antagonist (Genentech) ??? ???? none Enrolling RCT
Itraconazole (antifungal) ?? ?? none Biomarker study
15
Genetic deletion of Cox1 or Cox2 decreases
microscopic BCCs in Ptch1/- mice
IR-treated Ptch1/- mice wild type (n24),
deleted for Cox1 (n12) or for Cox2 (n6). Mean
and SEM. plt0.05 Cox1 and Cox 2 KO Smithies
Cell 1995


16
Celecoxib decreases microscropic BCC burden in
Ptch1/- mice (plt0.05)
17

• Study design Phase II randomized,
  double-blinded, placebo controlled trial
• Subjects 60 patients with Basal Cell Nevus
  Syndrome (BCNS)
• Treatment oral Celecoxib at 200mg BID versus
  placebo for 24 months followed by 12 months of
  observation
• Primary endpoint change in BCC numbers during
  study periods

18
Baseline characteristics of study participants
are similar in two groups (mean and SD)
Placebo (N27) Celecoxib (N33)
Age 42 12 47 12
Number of BCC tumors greater than 3mm 24 27 45 76
Body Mass Index 31 1.3 30 1.1
Male 48 58
Caucasian 96 100
With gt15 BCCs at baseline 41 39
Seen at California site 59 55
19
BCNS subject with low number of BCCs at baseline
(lt15 tumors)
20
BCNS subject with high number of BCCs at baseline
(gt15 tumors)
21
How to analyze BCC development

• Regession technique Linear mixed models
• Calcuates a slope or rate (number of BCCs/yr) for
  each patient
• Compare percent change in rate of BCCs in placebo
  and celecoxib groups
• Accounts for drop-outs
• Adjust for age, gender, BCC at baseline

22
(No Transcript)
23
Celecoxib reduces BCC development in subjects
with less severe disease (lt 15 BCCs)
24
Lessons learned importance of finding a reliable
mouse model

• Ptch1/- mice are a reliable model for testing
  new anti-BCC agents in humans
• Moderate effect of celecoxib on BCCs in mice and
  in BCNS patients
• Greater effect on tumor size rather than number
  in both mice and BCNS patients

25
Lessons learned statistics

• Statistics linear mixed models for determining
  slope of BCCs in RCT
• Regression models for tumors in mice
• Go to class
• Building a database (and managing)
• Go to class

26
Lessons learned

• Long time to publication
• Journals and co-authors disagree on whether to
  present data from mice and clinical trial study
  together
• Cancer Prevention Research

27
Summary of chemopreventive agents against BCCs
In vitro Mice Observational studies Clinical Trials
Celecoxib (oral) ? ? mixed ? in subset
Statin No No No N/A
Vitamin D3 ?? ?? ? Biomarker study
Hh Antagonist (Genentech) ??? ???? none Enrolling RCT
Itraconazole (antifungal) ?? ?? none Biomarker study
28
New and relatively safe agents that decrease
Hedgehog signaling
Cyclopamine
Smo
Gli
Corcoran and Scott, Proceedings of the Natl Acad
Sci 2006 Bijlsma and Peppelenbosch, PLOS Biology
2007
29
Mechanism of disease
Statins blocks Hedgehog pathway
BCC cell lines
In vitro experiments
Ptch1 /- mice
Animal models
Human clinical trials
Statin therapy and risk of subsequent BCCs in
Kaiser cohort
Epidemiological studies
30
(No Transcript)
31
Lessons learned

• Have a good biomarker or target gene (Gli1 mRNA)
• Have a good way to measure bioavailability
  (24OHase)
• Have a reliable cell line

32
Vitamin D3 decreases Gli mRNA in BCC cells
33
(No Transcript)
34
Topical Vitamin D3 decrease BCC development by
50 (plt0.05)
35
Topical vitamin D3 decreases BCC development in
mice
Unadjusted p value Adjusted p value
Vitamin D3 -62 .008 -53 .042
1,25(OH)2D -36 .39 -33 .43
Statin -58 .10 -40 .30
Adjusted for gender and coat color of mouse
36
Lessons learned

• Have a good biomarker or target gene (Gli1 mRNA)
• Have a good way to measure bioavailability
  (24OHase)
• Pilot trial of topical and oral vitamin D on
  human BCC

37
Mechanism of disease
Vit D3 blocks Hedgehog pathway
BCC cell lines
In vitro experiments
Ptch1 /- mice
Animal models
PTCH1/- Basal cell nevus syndrome patients
Human clinical trials
Vit D3 levels in BCC pts
Epidemiological studies
38
Lesson Learned

• Translating from mouse studies to epidemiologic
  studies (skip the human clinical trial/pilot)
• Mouse to Epi (Ralph Gonzalez)

39

• The association of serum vitamin D with skin
  cancer risk in elderly men
• Jean Y. Tang1,3, Neeta Parimi2, Angela Wu1,3,
  John Boscardin1, Meg Chren1,Steven R.
  Cummings1,2, Ervin Epstein3, and Douglas C.
  Bauer1,2
• 1 University of California San Francisco, 2 San
  Francisco Coordinating Center, California Pacific
  Medical Center Research Institute, 3 Childrens
  Hospital Oakland Research Institute

40
(No Transcript)
41
Table 3 Association of increasing serum 25(OH)D
levels with non-        melanoma skin cancer  
Adjusted for age (continuous variable), BMI
(continuous variable), season of blood draw, and
clinic site Adjusted for age, BMI, season of
blood draw, clinic site, outdoor walking activity
(continuous variable), and cigarette smoking
(yes/no)
42
(No Transcript)
43
Summary of chemopreventive agents against BCCs
In vitro Mice Observational studies Clinical Trials
Celecoxib (oral) ? ? mixed ? in subset
Statin No No No N/A
Vitamin D3 ?? ?? ? Biomarker study
Hh Antagonist (Genentech) ??? ???? none Enrolling RCT
Itraconazole (antifungal) ?? ?? none Biomarker study
44
Cyclopamine
Smo
Gli
Williams and Wang , Proceedings of the Natl Acad
Sci 2003
45
Placebo cream does not reduce BCC tumors
Day 0
Day 20
Day 35
46
Untreated BCC
47
Topical Hh Antagonist BCCs decreases BCC tumors

• Day 0

Day 35
Day 20
48
(No Transcript)
49
Hh Antagonist cream decreases Gli1 mRNA in BCC
tumors
S
Tu
Re
Re
Re
Tu
Tu
S
S
Tu
S
Tu
S
Tu
Tu
Tu
Hh Antagonist cream
Placebo
Lesson benefits of collaborator
50
Lesson Benefits of UCSF Cancer Center core labs
51
Topical Cur-414 treated BCC
52
(No Transcript)
53
Lessons learned mechanism of disease

• Topical and oral HH antagonists significantly
  reduce murine BCCs by decreasing tumor
  proliferation and/or inducing follicular
  differentiation.
• We know how to collect tumors, do these assays,
  get to mechanism of disease in future trials

54
(No Transcript)
55
Investigator sponsored trial in BCC prevention

• Genentech Hh antagonist GDC 0449 150mg daily
• Phase II, placebo controlled RCT in 41 Basal Cell
  Nevus Syndrome subjects
• Primary endpoint change in BCCs at 12 mo and 18
  mo

56
New and relatively safe agents that decrease
Hedgehog signaling
Cyclopamine
Smo
Gli
J Kim and P Beachy, Stanford
57
Mechanism of disease
Small molecule library screen for inhibitors of
Hedgehog pathway
Cell based assays
In vitro experiments
Ptch1 /- mice
Animal models
Patients with sporadic BCCs
Human clinical trials
58
Day 4 irritation and necrosis, 11 mm
Day 0, BCC A 10mm
Day 11 residual BCC, 3mm
Itraconazole178
59
Cyclodextrin5460
Day 0 BCC B (11 mm), BCC D (10 mm), BCC G (9mm)

Day 7 BCC B (11 mm), BCC D (10 mm), BCC G (9mm)

60
Lessons learned

• Collaboration with another basic lab
• Post-doc fellows who are experts at specific
  assays (5 years) vs training someone new
• Focus on getting the first pilot clinical trial
  of itraconazole on human BCCs
• Measure Gli1mRNA in BCCs
• Measure Ki67
• Paired t-test (at biopsy and at excision)
• New opportunities/markets

61
Lessons learned

• Easier to translate FDA approved drug or a drug
  manufactured and tested by Pharma - already have
  paid the
• New agent (vitamin D) investigator pays
• Efficacy
• Stability
• GMP grade for human
• IND

62
Challenge Lesson Learned
Celecoxib (oral) Combining mouse and RCT Statistics, finding the experts
Statins Difficulty in combining mouse and epidemiologic results (Kaiser) OK to disagree
Vitamin D3 Difficulty of doing it all yourself lab, mouse, pilot clinical trial 1Mentor and his lab SFCC, EpiBio mentors -Public databases
Hh Antagonist Collaborations with Genentech -Good to have lab skills -RCT design
Itraconazole Collaborations with basic lab FDA approved drug -Nice to collaborate -Time to start the pilot trial
63

• Childrens Hospital Oakland
• Ervin Epstein
• Po Lin So
• Tony Zheng Xiao
• Elana Shpall
• Angela Wu
• Kris Chang
• Yefim Khaimsky
• UCSF EpiBiostat
• Charles McCulloch
• Ralph Gonzalez
• Steve Hulley
• Steve Cummings
• Doug Bauer
• Neeta Parimi
• John Boscardin
• Michael Kohn

• Kaiser Division of Research
• Maryam Asgari
• Genentech
• Fred de Sauvage
• Tracy Tang
• Chris Callahan
• UCSF Derm
• Meg Chren
• Dan Bikle
• Loretta Chan
• Funding Sources
• NRCC CTSA KL2
• NIAMS K23
• Prevent Cancer Foundation
• American Skin Association