COLORECTAL CANCER
Charles Lopez, M.D., Ph.D.
Associate Professor of Medicine
Hematology and Medical Oncology
May 2009
Outline
• Background.
• Metastatic colorectal cancer.
– Chemotherapeutic and biologic agents
– Ongoing trials updates
• Adjuvant colorectal cancer.
– Current practice and ongoing trials
COLORECTAL CANCER UPDATE
2007 Estimated US Cancer Cases*
Men
766,860
Women
678,060
26%
Breast
15%
15%
Lung & bronchus
Colon & rectum
10%
11%
Colon & rectum
Urinary bladder
7%
6%
Uterine corpus
Non-Hodgkin
lymphoma
4%
4%
Non-Hodgkin
lymphoma
Melanoma of skin
4%
4%
Melanoma of skin
Kidney
4%
4%
Thyroid
Leukemia
3%
3%
Ovary
Oral cavity
3%
3%
Kidney
Pancreas
2%
3%
Leukemia
19%
21%
Prostate
29%
Lung & bronchus
All Other Sites
ONS=Other nervous system.
Source: American Cancer Society, 2007.
All Other Sites
COLORECTAL CANCER UPDATE
Epidemiology
Lifetime Probability of Developing Cancer
Men
Women
All sites
1 in 2
All sites
1 in 3
Prostate
1 in 6
Breast
1 in 8
Lung
1 in 13
Lung & bronchus
1 in 17
Colon and rectum
1 in 17
Colon & rectum
1 in 18
Urinary bladder
1 in 28
Uterine corpus
1 in 38
Lymphoma
1 in 46
Lymphoma
1 in 55
Melanoma
1 in 52
Ovary
1 in 68
Kidney
1 in 64
Melanoma
1 in 77
Leukemia
1 in 67
Pancreas
Oral Cavity
1 in 73
Urinary bladder
1 in 88
Stomach
1 in 82
Uterine cervix
1 in 135
1 in 79
* For those free of cancer at beginning of age interval. Based on cancer cases diagnosed during 2000 to 2002.
Source: DevCan: Probability of Developing or Dying of Cancer Software, Version 6.0 Statistical Research and
Applications Branch, NCI, 2005. http://srab.cancer.gov/devcan
Gastrointestinal Cancers in the United States
Small intestine
2%
Other digestive
organs
2%
Gallbladder
3%
Esophagus
6%
Anus
2%
Liver
7%
Colon
41%
Stomach
8%
Pancreas
13%
Rectum
16%
American Cancer Society. Cancer Facts and Figures 2006. Atlanta: American Cancer
Society; 2006.
Colorectal Cancer
• Major public health problem in the US and
worldwide
• Worldwide, nearly 1,000,000 cases diagnosed
each year
• In the US, 130,000-140,000 cases diagnosed
each year.
• In the US with 50,000-60,000 deaths each year
 < 40% of CRC diagnoses are localized disease
CRC = colorectal cancer.
Colorectal Cancer Facts & Figures. 2005. American Cancer Society.
Staging
TNM
Stage
I
I
IIA
IIB
IIIA
IIIB
IIIC
IV
TNM
Class
T1N0M0
T2N0M0
T3N0M0
T4N0M0
T1-2N1M0
T3-4N1M0
TanyN2M0
TanyNanyM1
MAC
A
B1
B2
B3
C1
C2/C3
C1/C2/C3
D
N2 denotes 4 or more regional
lymph nodes (Need to sample at
least 12 nodes).
(Modified Astler-Coller and New TNM Staging)
COLORECTAL CANCER UPDATE
Epidemiology
Stage I
Stage II
Stage III
Stage IV
Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: Incidence - SEER 17 Regs Limited-Use, Nov
2006 Sub (1973-2004 varying), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2007, based on
the November 2006 submission.
Risk Factors for Colorectal Cancer
Genetic
susceptibility
FAP (risk approaches 100% by age 50)
HNPCC (lifetime risk approaches 80%)
Family history
Adenoma (first degree) RR 1.72;11% lifetime risk
CRC (first degree) RR 2.62; 16% lifetime risk
Medical history
Inflammatory bowel disease (pancolitis
 8 years or left-sided colitis  15 years)
(10-20% risk)
Characteristics
Age (91% of cases occur after age 50)
Male sex (35%  in men)
Race/ethnicity (15%  in African Americans)
Obesity and diet (red meat, alcohol consumption)
Smoking
FAP = familial adenomatous polyposis; HNPCC = hereditary nonpolyposis colon cancer syndrome; CRC = colorectal cancer.
Anderson FW, et al. J Natl Cancer Inst. 2002;94:1126-1133; Colorectal Cancer Facts & Figures. American Cancer Society;
2005; Levin B, et al. NCCN. v.1.2005; Lynch HT, et al. Cancer. 1995;76;2427-2433; Petersen GM, et al. Cancer.
1999;86(suppl):2540-2550; Winawer SJ, et al. Gastroenterology. 2003;124:544-560.
Associations for Risk of Colorectal Cancer
Dietary Factors:
Dietary Fiber
Dietary Fat
Red Meat
Alcohol
Folate
Calcium and vitamin D
Non-Dietary Factors:
Body Mass Index
Physical Activity
Hormone replacement
Smoking
Aspirin
• HPFS, middle age men: BMI  25 kg/m2;Physical
activity  15 MET-hours/week;Daily folate containing
MVI;Alcohol < 15 g/day;Non-smoker;Red meat  2
servings/week (e.g.-- 3.1% of all men)
– Eliminate 71% of all colorectal cancer!
(95% CI, 33-92%)
Colon Cancer: Multi-step Model of Carcinogenesis =
biologic heterogeneity
•
•
•
•
Cancers arise in polyps
Develop over years
Multiple molecular events
Activation of oncogenes
and loss of tumor
suppressor genes
Beart R. Clinical Oncology. Abeloff M, et al. Ed. 1998 Churchill Livingstone
Evolution of a lethal cancer
Jones S. et.al. PNAS 2008;105:4283-4288
©2008 by National Academy of Sciences
COLORECTAL CANCER UPDATE
Screening
Easily Overlooked Lesions Tied to Colon Cancer
By Denise Grady
An easily overlooked type of abnormality in the colon is the most likely type to
turn cancerous, and is more common in this country than previously thought,
researchers are reporting.
lity in the colon is the most likely type to turn cancerous, and is more common in
this country than previously thought, researchers are reporting.
New York Times, March 5, 2008
COLORECTAL CANCER UPDATE
Screening
Flat Polyps
Soetinko et al. JAMA. 2008;299(9):1027-1035
COLORECTAL CANCER UPDATE
Screening
Flat Polyps
Soetinko et al. JAMA. 2008;299(9):1027-1035
Development of Systemic Treatments
for Metastatic Colorectal Cancer
2004
Cetuximab
+ irinotecan (or alone)
Bevacizumab + 5-FU-based
chemotherapy
Targeted
regimens
5-FU dominates treatment for mCRC
Investigations of 5-FU
combination and schedules
1950
1960
1970
1980
1957
5-FU introduced
Cytotoxic
regimens
5-FU = 5-fluorouracil; LV = leucovorin;
mCRC = metastatic colorectal cancer.
Holen KD, Saltz LB. Lancet Oncol. 2001; 2:290-297;
Venook A. The Oncologist. 2005;10:250-261.
Targeted
Therapies
1990
1996
Irinotecan
(2nd line)
2000
Irinotecan +
5-FU/LV
(1st line)
2000
2010
2006
Panitumumab
2004
Oxaliplatin + 5-FU/LV
(1st line)
2002
Oxaliplatin + 5-FU/LV
(2nd line)
Median OS* (mo)
Therapeutic Regimens
for the Treatment of MCRC
24
~15-20
18
~12-14
~11-12
12
6
~15-17
~4-6
0
1960s
*Both first- and second-line exposure to therapy.
MCRC = metastatic CRC; OS = overall survival.
Avastin® (bevacizumab) PI. December 2004.
Camptosar® (irinotecan) PI. July 2005.
Eloxatin™ (oxaliplatin) PI. April 2005.
Erbitux™ (cetuximab) PI. June 2004.
Xeloda® (capecitabine) PI. April 2003.
Vectibix (panitumumab) PI. 2006.
1980s
1990s
2000s
5-FU
5-FU biomodulation
Irinotecan
Oxaliplatin
Cetuximab
Bevacizumab
Panitumumab
Colorectal cancer:
Evolution of 1st line therapy for metastatic
disease.
5-Fluorouracil; 5-FU
5-FU/LV Foundation.
U.S.
Europe
bolus
infusional
Mayo
monthly
Roswell
Park weekly
de Gramont
bi-weekly
First line: de Gramont vs Mayo 5-FU/LV
5FU: 425mgs/m2 x5d, q 4wks
LV: 20 mgs/m2 x5d, q 4wks
500 pts randomized
40-
p=0.0004
5FU: 400 mgs/m2(bolus) d1,2 q 2wks
LV: 200 mgs/m2 d1,2 q 2wks
5FU: 600 mgs/m2(22hrs) d1,2 q 2wks
8-
33%
20-
p=0.0012
27.6
15-
14.1
22
4-
p=0.067
13-
13.1
14%
RR (%)
PFS (wks)
OS (mos)
de Gramont et.al. JCO 15:808 (97).
Chemical Structure of Platinum Analogues
NH3
Pt
Diaminocyclohexane
(DACH) carrier
ligand
NH
NH3
O
2
C
Cl
CISPLATIN
OXALATE
hydrolysable
ligand
Pt
NH2
Cl
O
C
O
NH3
O
C
O
C
Pt
trans-l-diaminocyclohexane oxalatoplatinum
OXALIPLATIN, Eloxatin®
O
NH3
O
•Active in NCI CRC cell lines
•DNA adducts, cross-links
•Hold if CrCL<20cc/min
CARBOPLATIN
First line: FOLFOX vs infusional LV/5-FU
Oxaliplatin 85 mg/m2 d1, q 2wks
LV: 200 mg/m2; 5FU 400mg/m2 IVB
5FU 600 mg/m2 (22 hr CIV);d1,2q2wk
420 pts randomized
5FU: 400 mgs/m2(bolus) d1,2 q 2wks
LV: 200 mgs/m2 d1,2 q 2wks
5FU: 600 mgs/m2(22hrs) d1,2 q 2wks
50-
51%
p<0.0001
10-
9.0
p=0.001
18-
6.2
20-
5-
14-
#
p=0.12
16.2
14.7
22%
RR (%)
* PFS (mos)
*=primary endpoint
# OS (mos)
deGramont et.al. JCO 18:2938 (2000).
Irinotecan (CPT-11, Camptosar®)
CH3
CH2
N
O
N
N
C
O
O
N
Irinotecan hydrochloride
O
HO
O
CH2CH3
•Topoisomerase I inhibitor an enzyme that relieves torsional strain in DNA
•Hepatic metabolism. Majority excreted in bile requiring dose adjustment
•Converted by the carboxylesterase enzyme to the SN-38 metabolite, which
is 1000-fold more potent than the parent drug
First line: FOLFIRI vs infusional LV/5-FU
CPT-11 180 mg/m2+ deGramont q 2wks
or
CPT-11 80 mg/m2+HDLVB/5FU24CI qwk
385 pts randomized
deGramont q 2wks
or
500 LVB/5FU 2600mg/m2 24 CIV qwk
40-
p<0.005
8-
35%
20-
p=0.001
6.7
18-
17.4
4.4
4-
p=0.031
14-
14.1
22%
RR (%)
PFS (mos)
OS (mos)
Douillard et.al. Lancet 355:1041 (2000).
5-FU/LV Foundation.
U.S.
Europe
bolus
infusional
FOLFIRI
Mayo
monthly
Roswell
Park weekly
FOLFOX
de Gramont
bi-weekly
5-FU/LV Foundation.
U.S.
Europe
bolus
infusional
IFL
Mayo
monthly
Roswell
Park weekly
FOLFIRI
FOLFOX
de Gramont
bi-weekly
First line: IFL vs Mayo 5-FU/LV
5FU: 425mgs/m2 x5d, q 4wks
LV: 20 mgs/m2 x5d, q 4wks
457 pts randomized
40-
20-
39%
p<0.001
21%
RR (%)
CPT11: 125 mgs/m2
LV: 20 mgs/m2
5FU: 500 mgs/m2; q wk 4/6
8-
4-
7.0
p=0.004
15-
14.8 p=0.04
4.3
13-
PFS (mos)
• 3rd arm: CPT-11 alone arm=LV/5FU
12.6
OS (mos)
Saltz et.al. NEJM 343:905 (2000).
5-FU/LV Foundation.
U.S.
Europe
bolus
infusional
???
FOLFIRI
FOLFOX
IFL
Mayo
monthly
Roswell
Park weekly
de Gramont
bi-weekly
First-Line Oxaliplatin in the US: N9741
Metastatic
Disease
Mayo
trash
R
A
N
D
O
M
I
Z
A
T
I
O
N
N=264
Bolus Saltz Regimen*
IFL
CPT-11: 125 mg/m2/wk x 4 wks, q 6 wks
5FU: 500 mg/m2/wk x 4 wks, q 6 wks
LV: 20 mg/m2/wk x 4 wks, q 6 wks
Infusional FOLFOX-4 Regimen†
N=267
Oxaliplatin: 85 mg/m2 d1 q 2 wks
5-FU: 400 IV/600 CI mg/m2 d1, 2 q 2 wks
LV: 200 mg/m2 d1, 2 q 2 wks
Wasserman Regimen‡
N=264
FOLFOX4
IROX
CPT-11: 200 mg/m2 d1 q 3 wks
Oxaliplatin: 85 mg/m2 d1 q 3 wks
*Saltz et al. J Clin Oncol 14:2959, 1996.
†André et al. J Clin Oncol 17:3560, 1999.
‡Wasserman et al. J Clin Oncol 17:1751, 1999.
Goldberg J Clin Oncol 22:23, 2004
Europe vs USA?
U.S.
Euro
Feb 2004: FOLFOX FDA approved
as first-line chemotherapy in U.S.
N9741: Efficacy
IFL
N=264
MS (mo)
TTP (mo)
14.8
6.9
FOLFOX4
N=267
IROX
N=264
CPT-11:
200 mg/m2 d1 q 3 wks
Oxaliplatin:
85 mg/m2 d1 q 3 wks
19.5
17.4
(p=0.001)
(p=0.04)
8.7
6.5
(p=0.0014)
ORR (%)
31
45
35
(p=0.002)
FOLFOX: Second-line approval in US 8/02; First line approval 2/04
Goldberg J Clin Oncol 22:23, 2004
FOLFIRI versus FOLFOX
progression
FOLFIRI
FOLFOX6
progression
FOLFIRI
progression
N=113
R
N=113
progression
FOLFOX6
Regimens: oxaliplatin* (100 mg/m2) or irinotecan† (180 mg/m2) IV +
LV 200 mg/m2 over 2 hours d 1, 5-FU 2,400-3,000 mg/m2 over 46 hours
Primary end point: TTP
Secondary end points: ORR, safety
Tournigand et al. J Clin Oncol. 2004
FOLFIRI verses FOLFOX6
Results
Arm A
FOLFIRI
FOLFOX
N=109
N=81
ORR %
56*
15†
Arm B
FOLFOX FOLFIRI
N=111
P
N=69
Value
4†
.68
54*
Median TTP (mo)
14.4
11.5
.65
Median Overall
Survival (mo
20.4
21.5
.9
Tournigand C et al. JCO 2004
(Similar results: Colucci G, et al. J Clin Oncol. 2005;23:4866-4875).
Molecular re-classification of metastatic
colorectal cancer
N
I.
?
II.
.
DIA
HRN
?
Molecular re-classification of metastatic
colorectal cancer
N
I.
II.
Preliminary data: gene signature predicts diffuse
metastatic disease
.
DIA
HRN
Incorporation of Targeted Agents
in the Standard Management of
Metastatic Colorectal Cancer
Agents Targeting the VEGF Pathway
Anti-VEGF
antibodies
VEGF
Soluble
VEGF
receptors
Anti-VEGFR
antibodies
P
P
P
P
VEGFR-1
Ribozymes
P
P
P
P
VEGFR-2
Small-molecule
VEGFR inhibitors
(TKIs)
Endothelial cell
–VEGF stimulates new blood vessel formation in normal tissues and tumors
–VEGF blockade normalizes tumor vasculature and improves drug delivery.
Phase III Trial of Avastin + IFL as First-Line
Therapy for MCRC (AVF2107g): OS
100
Median survival: 15.6 mo (w/o Avastin)
vs 20.3 mo (w/Avastin)
HR=0.66, P<0.001
OS (%)
80
Placebo + IFL (n=411)
Avastin + IFL (n=402)
60
1-year survival:
74% vs 63%
40
20
2-year survival:
45% vs 30%
0
0
6
12
18
Months
Error bars represent 95% CIs.
Avastin® (bevacizumab) PI. December 2004.
Data on file (SR2), Genentech, Inc. 2005.
Hurwitz et al. N Engl J Med. 2004;350:2335.
24
30
EGFR as a Therapeutic Target
The EGF Receptor: (HER1 or c-Erb-1)
EGFR a member of a subfamily of type I receptor tyrosine kinases
(including HER2, HER3 and HER4)
EGF
TGF-
HER2/3/4
EGFR
EGFR
EGFR
EGF = epidermal growth factor; TGF- = transforming growth factor-alpha; EGFR =
epidermal growth factor receptor.
Properties of ERBITUX (Cetuximab)
 IgG1 MAb (chimerized)
 Binds specifically to EGFR
and its heterodimers
 Binds to EGFR with high affinity
(Kd = 2.0 x 10–10 M):
1 log higher than the natural ligand
 Following the recommended dose
regimen (400 mg/m2 initial dose/250 mg/m2 weekly dose),
the mean half-life was 114 hours (range 75-188 hours)
MAb = monoclonal antibody; EGFR = epidermal growth factor receptor.
ERBITUX Package Insert, June 2004; Data on file. ImClone Systems Incorporated and Bristol-Myers Squibb Company; 2004.
Please see Important Safety Information including
WARNING regarding infusion reactions on slides 39 to 52.
Cetuximab activity in irinotecan refractory mCRC
(Bond Trial): Objective Response Rates
All Patients
Irinotecan-Refractory and Oxaliplatin Failure Patients
Cunningham D, et al. N Engl J Med.
2004;351:337-345.
Panitumamab +BSC vs BSC
ORR
8%
0%
• N=463
• Patients
third-line
mCRC
NCIC-CO.17
Cetuximab+BSC
N=287
mCRC failing (<6 months)
oxaliplatin and CPT11 regimens
with 5FU. No prior
NOTE: No crossover allowed
anti-EGFR Rx. Bev permitted.
Primary endpoint=OS
Median duration of f/u = 14.6 months
BSC
N=285
Jonker et. al, NEJM 357:2040-08 (2007)
Jonker et. al, NEJM 357:2040-08 (2007)
Primary endpoint
OS
Cetuximab
plus bsc
6.1 mo
1 yr
21%
p = 0.005
HR = .77
Bsc alone 4.6 mo
16%
7% cross-over from bsc-->cet
Post trial rx: 27.5% (cet) verses 23.2% (bsc)
What about in a “real world” setting?
EPIC
Sobrero et.al. AACR April 2007
CPT-11
*OS
PFS
RR
10.7 mo
3.98 mo
16.4%
+ cetuximab
mCRC failing
Oxaliplatin+/-bev
50% crossover
(p = ns)
(p < 0.0001)
N=1298
* Primary Endpoint = OS
Irinotecan: 350 mg/m2 q3weeks
Cetux: 400mg/m2 load then 250mg/m2
CPT-11
9.99 mo
2.56 mo
4.1%
What about upfront?
CRYSTAL
Primary Endpoint = PFS
PFS
FOLFIRI
+ Cetuximab
8.9 mo
RR
46.9%
(34% 1yr)
Untreated
mCRC
p=0.036
p=0.005
8.0 mo
38.7%
N=1217
FOLFIRI
(23% 1yr)
Van Cutsem E, et al. ASCO 2007. Abstract 4000.
Metastatic colorectal cancer....
?
CALGB/SWOG80405 Trial: First Line
Met CRC.
Study PIs: Charles Blanke, M.D./Alan Venook, M.D.
FOLFOX
FOLFIRI
Randomize
“Dealers Choice”
Bev
C225
Bev/C225
Target accrual 2300
PACCE: Panitumumab Advanced CRC Evaluation Study
Primary endpoint=PFS
N >1000
• First-line
mCRC
• No CNS
metastases
• No CV risk
factors
R
A
N
D
O
M
I
Z
E
Panitumumab +
oxaliplatin- or irinotecan-based
chemotherapy + bevacizumab
(q2 wk)
Oxaliplatin- or irinotecan-based
chemotherapy + bevacizumab
(q2 wk)
Interim (12-week) response-rate analysis first 500 patients:
response rates similar between treatment groups.
US NIH Clinical Trials Database (www.clinicaltrials.gov); accessed 8/3/06
Cairo 2
Punt et. al ASCO 2008
Capox + bev + cetux
PFS
tox
9.6 mo
82%
p=0.018
p=0.0013
N=368
mCRC
Capox + bev
N=368
10.7 mo
72%
Cairo 2
Punt et. al ASCO 2008
PFS
PFS
Capox + bev + cetux
PFS
WT-KRAS MT-KRAS
9.6 mo 10.5 mo 8.6 mo
N=368
p=0.018
mCRC
Capox + bev
N=368
p=ns
p<0.05
10.7 mo 10.5 mo 12.5 mo
CRYSTAL
Van Cutsem E, et al. ASCO 2008.
PFS
PFS
PFS
WT-KRAS MT-KRAS
FOLFIRI
+ Cetuximab
Untreated
mCRC
N=1217
FOLFIRI
10Endpoint = PFS
8.9 mo
9.9 mo
(34% 1yr)
(43% 1yr)
p=0.036
HR=0.85
p=0.017
HR=0.68
8.0 mo
8.6 mo
(23% 1yr)
(25% 1yr)
p=0.75
HR=1/07
No change in PFS or RR in FOLFIRI w/o cetuximab
Suggests KRAS a predictive marker (vs prognostic)
OPUS trial: randomized phase II FOLFOX +/- cetuximab similar results that only
WT-KRAS benefits from cetuximab (Bokemeyer et. al. ASCO 2008)
Metastatic colorectal cancer....
??????
The EGF Receptor: (HER1 or c-Erb-1)
EGFR a member of a subfamily of type I receptor tyrosine kinases
(including HER2, HER3 and HER4)
EGF
TGF-
HER2/3/4
EGFR
EGFR
EGFR
EGF = epidermal growth factor; TGF- = transforming growth factor-alpha; EGFR =
epidermal growth factor receptor.
CALGB/SWOG80405 Trial: First Line
Met CRC.
Study PIs: Charles Blanke, M.D./Alan Venook, M.D.
Bev
FOLFOX
FOLFIRI
Randomize
“Dealers Choice”
ONLY WT KRAS
C225
Bev/C225
Target accrual 2300
ADJUVANT THERAPY
TNM
Stage
I
I
IIA
IIB
IIIA
IIIB
IIIC
IV
TNM
Class
T1N0M0
T2N0M0
T3N0M0
T4N0M0
T1-2N1M0
T3-4N1M0
TanyN2M0
TanyNanyM1
MAC
A
B1
B2 ?
B3 ?
C1
C2/C3
C1/C2/C3
D
N2 denotes 4 or more regional
lymph nodes (Need to sample at
least 12 nodes).
(Modified Astler-Coller and New TNM Staging)
COLORECTAL CANCER UPDATE
Surgical Management
Zollinger, Atlas of Surgical Operations, 1993
COLORECTAL CANCER UPDATE
Surgical Management
COLORECTAL CANCER UPDATE
Surgical Management
Total Mesorectal Excision
• nerve preservation
• sharp dissection
• complete hemostasis
• intact mesorectal envelope.
Revised, node-positive TNM classification for Stage III CRC (n=50,042)
IIIA: T1/2, N1
Observed 5-year
survival (%)
70
60
IIIB: T3/4, N1
IIIC: Any T, N2
p<0.0001
59.8%
50
42.0%
40
27.3%
30
20
10
0
IIIA
IIIB
Node-positive subgroups
IIIC
Greene et al (2002)
Adjuvant Therapy for Stage II Colon Cancer:
• ASCO Guidelines : 8/15/04 JCO
– “Pts with high risk features..might be considered candidates for adjuvant
therapy..” BUT,
– “Direct evidence does not support adjuvant chemotherapy..even for high
risk stage II..”
• http://www.mayoclinic.com/calcs/colon/index-ccacalc.cfm
• Estimate that Stage II trials will need >8000 patients
to provide direct evidence for benefit.
• Need: better risk stratification to identify patient
subgroups that will benefit (e.g. do-able trials).
Adjuvant Therapy for Stage II Colon Cancer
• Indirect evidence makes it reasonable to offer for
patients with high risk features who are willing to accept
toxicity for theoretical benefit.
– High risk features:
•
•
•
•
•
< 12 LN sampled
Obstruction at presentation
Perforation at tumor site
High grade tumor
LVI
• Should encourage participation in randomized trials.
MOSAIC: Treatment arms
NEJM 350(23):2343, 2004
FOLFOX4
(n=1123)
Resected Stage II/III CRC
N=2246
40% Stage II
R
LV5FU2
60% Stage III
(n=1123)
Endpoints

Primary:
– 3-yr Disease Free Survival (DFS): a
surrogate for 5-yr Survival

Secondary:
– Safety
– Overall Survival (OS)
MOSAIC
Stage III
4-year DFS
1.0
FOLFOX4 (n=672)
LV5FU2 (n=675)
Probability
0.9
70%
61%
p=0.002
0.8
0.7
0.6
0.5
0
10
20
30
DFS (months)
40
50
24% risk reduction for stage III patients in the FOLFOX4 arm
Improved 6 year overall survival for Stage III
(hazard ratio of 0.80, confidence interval [0.65, 0.97], p=0.023).
Phase III MOSAIC Trial (ITT): Clinical
Efficacy
Disease-free
survival
(DFS)
FOLFOX
LV/5FU2
Overall
76%
69%
p=0.0008
risk reduction 24%
Stage III
70%
61%
p=0.002
Stage II
86.6%
83.9%
p-value not significant
de Gramont A, et al. Proc Am Soc Clin Oncol. 2003 Abstr 1015; Andre et.al NEJM 6/04
Irinotecan Negative in Adjuvant Setting
Clinical Trial
Patients
Outcome
CALGB C89803 Stage III
(IFL vs FL)
(n=1264)
More neutropenia,
death
PETACC-3
More neutropenia,
diarrhea
(FOLFIRI vs
deGramont)
Stage II/III
(n=3278)
High-risk Stage III
(LV5FU2+/-CPT11 (n=400)
ACCORD02
Saltz et.al ASCO Ab 3500 (2004)
Van Cutsem et.al ASCO Ab 8 (2005)
Ychou et.al. Ab 3502 ASCO (2005)
More neutropenia,
nausea
Molecular re-classification of metastatic
colorectal cancer
N
I.
?
II.
.
DIA
HRN
?
EORTC 40983
3 yr PFS
FOLFOX x6
All pts
Eligilble pts
Surgery
FOLFOX x6
Potentially resectable
35.4%
36.2%
P=0.058
P=0.041
28.1%
28.1%
N=182
mCRC
No extrahepatic disease
≤4 lesions
No prior Rx
Primary endpoint
– 40% increase in PFS
(NOT designed to validate
Surgery
Pre vs post-op chemo)
N=182
Nordlinger, ASCO 2007
EORTC 40983
• More postoperative complications in FOLFOX group
25% vs. 16% (p=0.04). (e.g. liver failure, biliary fistula).
• No difference in postoperative mortality
Nordlinger, ASCO 2007
Adjuvant colorectal cancer
1. FOLFOX for stage III (MOSAIC).
-NSABPC07 (bolus 5FU+/- ox) confirms MOSAIC. FLOX with
unfavorable toxicity profile?
2. No direct evidence for stage II pts. But, indirect evidence
makes reasonable to discuss if high risk features
•
•
•
•
•
< 12 LN sampled
Obstruction at presentation
Perforation at tumor site
High grade tumor
LVI
Some ongoing adjuvant trials
Stage II CRC: E5202 Intergroup Trial
MSS
LOH 18q
FOLFOX
R
FOLFOX +
bevacizumab
Stage II
patients
MSI+
normal 18q
No therapy
Accrual goal: 3,125
NSABP C-08: mFOLFOX6 + Bevacizumab
• Patients with
resected
stage II/III
colon cancer
• Target
enrollment,
2632
R
A
N
D
O
M
I
Z
E
mFOLFOX6 +
Bevacizumab
q14 d for 12 courses
Bevacizumab
q14 d for 6 mo
(in absence of
PD)
mFOLFOX6
q14 d for 12 courses
Trial opened Sept 2004.
ASCO 2008 Safety Update: increased grade III toxicity, p=0.006
US NIH Clinical Trials Database (www.clinicaltrials.gov); accessed 8/3/06.
AVANT Trial: FOLFOX vs FOLFOX/BEV vs
XELOX/BEV
?????????????
• Patients with
stage II/III
colon cancer
• Target
enrollment,
3450
R
A
N
D
O
M
I
Z
E
FOLFOX4
FOLFOX4 +
Bevacizumab
XELOX +
Bevacizumab
US NIH Clinical Trials Database (www.clinicaltrials.gov); accessed 8/21/06.
N0147: Adjuvant Cetuximab
For WT-KRAS ONLY!!!!!!!!
• Patients with
resected
stage III
colon cancer
• Target
enrollment,
2300
R
A
N
D
O
M
I
Z
E
FOLFOX4 ±
Cetuximab
(FOLFOX then
FOLFIRI) ±
Cetuximab
FOLFIRI ±
Cetuximab
US NIH Clinical Trials Database (www.clinicaltrials.gov); accessed 8/3/06.
Reality Check......
• Cost: Shrag, NEJM, 7/2004
– Average patient with 1st line FOLFOX+ bevacizumab (8
months) followed by CPT-11 + cetuximab (4 months)=
$161,000 USD X 50,000 Stage IV cases/year
–
> $5 BILLION USD /year (JUST for
stage IV..if used for adjuvant..> $20
BILLION USD/year!!!!
– How do we save $$$?
• Lower cost
• Limit therapy (per group)
• Individualize treatments to those most likely to benefit
(per patient)
• Limit duration of therapy
Colon Cancer: Multi-step Model of Carcinogenesis =
biologic heterogeneity
•
•
•
•
Cancers arise in polyps
Develop over years
Multiple molecular events
Activation of oncogenes
and loss of tumor
suppressor genes
Beart R. Clinical Oncology. Abeloff M, et al. Ed. 1998 Churchill Livingstone
The Genomic Landscapes of Human Breast and
CRC Cancers
Wood et al. Science 2007;318:1108-13.
Human cancer is caused by the accumulation of mutations
in oncogenes and tumor suppressor genes.
To catalog the genetic changes that occur during
tumorigenesis, we isolated DNA from 11 breast and 11
colorectal tumors and determined the sequences of the
genes in the Reference Sequence database in these
samples.
Based on analysis of exons representing 20,857
transcripts from 18,191 genes, we conclude that the
genomic landscapes of breast and colorectal cancers are
composed of a handful of commonly mutated gene
"mountains"
and a much larger number of gene "hills" that are mutated
at low frequency. We describe statistical and
bioinformatic tools that may help identify mutations with a
role in tumorigenesis. These results have implications
for understanding the nature and heterogeneity of human
cancers and for using personal genomics for tumor
diagnosis and therapy.
A New Paradigm for Colorectal Cancer
Clinical Trials
Current
A
All patients receive
standard treatment (A)
patients
B
Clinical trials
survival benefit from A
Future
A
B
patients
Molecular analysis
of tumor
C
D
Choice of treatment
dependent upon
molecular
profile of tumor and
patient genotype
Stay tuned for upcoming exciting
info...........