Cancer Chemoprevention Biomarkers Jian Yu Rao M D

Cancer Chemoprevention & Biomarkers Jian. Yu Rao, M. D. Prof. Of Pathology/Epidemiology UCLA

CANCER PREVENTION • PRIMARY • STOP THE EXPOSURE • SECONDARY • INTERVENTION • TERTIARY • TREATMENT

CANCER INTERVENTION – Treating premalignant lesions • Surgical/procedure • Castration / Oophorectomy • Polypectomy • Colposcopy • Non-surgical (CHEMOPREVENTION) – Nutritional/chemical – Immuno (BCG) – Vaccine – Molecular targeted • Behavioral – Smoking cessation

S Lippman, Cancer Res. 2009 Jul 1; 69(13): 5269 -84

A Most Successful Cancer Control Story in 20 th Century Pap Test Colposcopy + CONE/LEEP Detect Early Lesions Treat Early Lesions Decreased incidence/ mortality of cervical CA

CHEMOPREVENTION • Administrating specific amounts of a particular natural or synthetic chemical in an attempt to identify agents that will prevent, halt or reverse the process of carcinogenesis • The basic assumption is that treating early stages of malignant process will halt the progression of malignancy • The key is to define early lesions, and treat the malignant field

Exposure to Carcinogen Birth Precancerous Intraepithelial Lesions, (PIN, CIN, Pa. IN. . ) Additional Molecular Event CHEMOPREVENTION Cancer

Multiyear progression from initiation and early precancerous lesions to invasive disease in major cancer target organs Kelloff et al. 2000 (Fig. 1)

THEORIES SUPPORT FOR CHEMOPREVENTION • EPIDEMIOLOGICAL EVIDENCE: • OVER 50% CANCERS HAVE NO KNOWN RISK FACTORS • NUMEROUS EVIDENCE TO DEMONSTRATE THE INVERSE RELATIONSHIPS OF SOME NUTRIENT FACTORS WITH CANCER RISKS

THEORIES SUPPORT FOR CHEMOPREVENTION (Cont. ) • EXPERIMENTAL EVIDENCE: • ALTHOUGH CARCINOGENESIS IS REGARDED AS NONREVERSIBLE PROCESS, STUDIES SHOWED THIS IS ONLY TRUE AT LATE STAGE. IN FACT, A LARGE PORTION OF THE LONG LATENCY PERIOD OF CARCINOGENIC PROCESS IS REVERSIBLE. • IN VITRO CULTURE AND IN VIVO ANIMAL STUDIES IDENTIFIED NUMEROUS AGENTS THAT CAN REVERSE, OR HALT THE CARCINOGENESIS PROCESS, PARTICULARLY AT THE EARLY STAGE.

THEORIES SUPPORT FOR CHEMOPREVENTION (Cont. ) • CLINICALLY • ADVANCES IN CERTAIN TYPES OF CANCER TREATMENT HAVE LIMITED SUCCESS IN REDUCING THE OVERALL INCIDENCE, OR EVEN MORTALITY OF CANCER.

Chemoprevention: Some Terminologies • INDIVIDUAL RISK AND STRATIFICATION • INTERMEDIATE END POINT MARKER (SURROGATE END POINT MARKER) • FIELD CANCERIZATION • MULTI-PATH OF CARCINOGENESIS

RISK STRATIFICATION • Identification of AT-RISK subjects who are also SUSCEPTIBLE to treatment: LEGEND: Not at risk to develop disease At risk of developing disease, biology A, responsive to agent X At risk of developing disease, biology B, NOT responsive to agent X

INTERMEDIATE END POINT MARKER (SURROGATE END POINT MARKER) • These are prevention biomarkers which are specifically related to early stages of carcinogenesis. • These markers are used to identify individual’s risk for developing cancer and to monitor the effectiveness of intervention methods.

FIELD CANCERIZATION • The whole field of tissue of a particular organ is exposed to the carcinogenic insult and is at increased risk for developing cancer. • Although only a few foci eventually develop malignancy, the other areas are not necessary entirely “normal”. • Most common epithelia cancers are developed through this mechanism. Examples of such cancers are: Head and neck ca, bladder ca, breast ca, lung ca, GI ca, etc.

MULTI-PATH OF CARCINOGENESIS • The current model of carcinogenesis is that cancer develops through multiple events which are not necessary through linear steps, but rather through overlapping networks.

CHEMOPREVENTION IN DIFFERENT RISK CATEGORIES Risk category Parameter General Population High Risk Agent toxicity Trivial to none Slight Selection method Public Health Clinical Other consideration Use dietary supplements may be applicable Need biomarkers From lee W. Wattenberg, P. S. E. B. M. , 1997 216: 133 -141.

Phase I Trial Objectives: • To determine the intervention’s short-term (<1 yr. ) dose-toxicity relationship • To determine the intervention’s human pharmacokinetics Design: • • Single arm, nonrandomized Multiple dose levels Less than 1 yr. duration Accrual 25 -100

Phase II Trial Objectives: • To determine the intervention’s side effects • To determine optimal recruitment methods of the target population • To determine retention of study participants to the study intervention and procedures • To determine optimal methods for the conducting of a phase III trial • To determine the effect of the intervention on biomarkers of carcinogenesis (phase II b) Design: • • Randomized, double-blind, placebo-controlled Multiple dose levels or agents One to five years in duration Accrual 100 s-1000 s

Phase III Trial Objectives: • To determine the effect of the intervention on the cancer incidence (total and specific cancer type) • To determine the effect of the intervention on death rate and disease incidence • To determine the long-term side effects of the intervention • To determine the nature history of specific biomarkers of carcinogenesis (placebo group) and the effect of the intervention agent (treatment group) on these markers. Design: • • Randomized, double-blind, placebo-controlled Multiple dose levels or agents, alone or in combination Five to ten years in duration Accrual 1000 s-10, 000 s

UNIQUE FEATURES OF CHEMOPREVENTION • Participants are usually healthy or at least “cancer free” • The degree and incidence of side effects that are acceptable are low • The end point is disease prevention, not disease response • The incidence of the study end point is low

CATEGORIES OF CHEMOPREVENTIVE AGENTS • BLOCKING CARCINOGEN METABOLISM AND EXPOSURE • INCREASE TISSUE RESISTANCE/DIFFERENTIAITON • TARGETING ONCOGENIC PATHWAYS

CATEGORIES OF CHEMOPREVENTIVE AGENTS • BLOCKING AGENTS • Prevent metabolic activation of carcinogens or tumor promoters • Enhance detoxification Glutathione-S-transferase • Trap reactive carcinogenic species: Glutathione, N-Acetylcysteine • Vaccines: HBV, HPV

CATEGORIES OF CHEMOPREVENTIVE AGENTS (Cont. ) • INCREASING TISSUE RESISTANCE • Induce tissue maturation/differentiation Pregnancy or hormonal induced maturation of terminal ducts of breast - decrease breast cancer Retinoids, DMFO, etc • Decrease target tissue function Castration - reduce risk of prostate ca • Decrease cell proliferation Low fat diet decrease epithelial proliferation rate in intestinal tract - reduce colon cancer risk

CATEGORIES OF CHEMOPREVENTIVE AGENTS (Cont. ) • PATHWAY SPECIFIC AGENTS • Cox-2 inhibitors • Anti-angiogenesis • Anti-EGFR • Hormone antagonists • Augmenting tumor suppressor functions • Inhibiting oncogenic activities (e. g. , Ras)

BRIDGING THE GAP BETWEEN CANCER TREATMENT AND PREVENTION (William WN, et al, Nat Rev Drug Discov. 2009 Mar; 8(3): 213 -25. )

CHEMOPREVENTION TO HUMANS UPDATE • BREAST CANCER • Selective Estrogen Receptor Modulator (SERM)/STAR Trial • Animal model well established • PROSTATE CANCER • SCID model established • Hormonal modulation may have potential • PCPT Trial – Finasteride (5 -a-reductase, 5 mg/day) • 2 -arm trial, 18, 882 subjects, 7 yrs • PCP=18. 4% vs 24. 8% in treated vs ctrl group • Selenium/Vit E Trial - negative

STAR TRIAL • STAR: The Study of Tamoxifen (20 mg/day) and Raloxifene (60 mg/day) for breast cancer prevention • A Phase III trial involving over 50 centers and 19, 747 high risk women age 35 or above • Both drugs reduced the risk of developing invasive breast cancer by about 50 percent. • In addition, women who took raloxifene daily had 36 percent fewer uterine cancers and 29 percent fewer blood clots than the women who took tamoxifen.

SELECT TRIAL (Lippman, et al, JAMA, 2009) • SELECT: Selenium and Vitamin E Cancer Prevention Trial – A Phase III trial involving over 400 sites and 35533 men, age >50 (for AA) or >55 (others) – 4 groups (selenium, vitamin E, selenium + vitamin E, and placebo) in a double-blind fashion between August 22, 2001, and June 24, 2004 – No significant differences (all P>. 15) in prostate cancer end points

CHEMOPREVENTION TO HUMANS UPDATE (CONT. ) • GASTRIC AND ESOPHAGEAL CANCER • A combination of beta carotene, vitamin E, and selenium may be effective in early stage lesions, but not late severe dysplastic lesions. • LUNG CANCER • Beta-carotene or alpha-tocopherol showed reverse effect in lung cancer risk in heavy smokers in Finland • Ongoing trials with COX-2 inhibitor in former smokers here at UCLA

CHEMOPREVENTION TO HUMANS UPDATE (CONT. ) • COLON CANCER • Sulindac, a nonsteroidal anti-inflammatory compound hold great promise. Others, such as Oltiparz, selenium, and antioxidants vit. E/A, etc, may also be effective. Large trials are not available. • HEAD AND NECK CANCER • Retinoids showed promising results in both animal models and human studies.

SUMMARY (CHEMOPREVENTION) • Chemoprevention is an important approach for cancer control • The goal is to treat premalignant lesions • Successful stories include cervix, breast, prostate, head and neck • Less successful stories include lung, colon, etc • No “magic” single bullet, but new molecular targets show greater promise than traditional nutritional regimens

BIOMARKERS OF CANCER • CLINICAL SETTINGS (TUMOR MARKERS) • EPIDEMIOLOGICAL AND PREVENTIVE SETTINGS (INTERMEDIATE END POINT OR SURROGATE END POINT MARKERS).

CURRENT CLINICALLY USED TUMOR MARKERS • • • PSA - Prostate Adenocarcinoma Alpha FP - Hepatoma & seminoma HCG - Choriocarcinoma CEA - Colon cancer CA 19. 9 – Pancreatic cancer CA 125 – Ovarian cancer Problem: Low sensitivity/specificity

BIOMARKERS • • Genetic susceptibility markers Markers of exposure Markers of biological effects Tumor markers

Exposure to Carcinogen Birth Precancerous Intraepithelial Lesions, (PIN, CIN, Pa. IN. . )e Additional Molecular Event Cancer Surrogate End Point Markers Genetic Suscep. Markers for Exposure Markers of Effect CHEMOPREVENTION Tumor Markers

HOW TUMOR MARKERS ARE USED CLINICALLY • Early detection and diagnosis • Predict the biological potential of cancer (prognostication) • Determine who will be likely benefited from therapy (phamacocogenetic / pharmacogenomic) • Monitor the effectiveness of therapy

CRITERIA FOR SELECTING BIOMARKER • FITS EXPECTED BIOLOGICAL MECHANISM • BIOMARKER AND ASSAY PROVIDE ACCEPTABLE SENSITIVITY, SPECIFICITY, AND ACCURACY • BIOMARKER IS EASILY MEASURED • BIOMARKER MODULATION CORRELATES TO THE END POINT – Disease incidence (for detection marker) – Disease progression (for prognostic marker) – Response to therapy (for therapeutic marker)

ASSAY VALIDITY • SENSITIVITY: – % of assay-positive cases in case group • SPECIFICITY – % of assay-negative cases in control group • PPV (Positive Predictive Value) – % of “true” assay-positive cases out of all assaypositive cases • NPV (Negative Predictive Value) – % of “true” assay-negative cases out of all assaynegative cases These numbers can be totally meaningless – always know what kind of design used (case-control vs cohort) and what is the gold standard

OTHER ASSAY ISSUES • BIOMARKER CAN BE OBTAINED BY NONINVASIVE TECHNIQUES • ASSAY IS NOT TECHNICALLY DIFFICULT • MULTIPLE MARKERS CAN BE EVALUATED SIMULTANEOUSLY IN LIMITED SAMPLE VOLUMES • COST

CATEGORIES OF MARKERS • HISTOLOGICAL AND MORPHOMETRIC MARKERS • PROLIFERATION, DIFFERENTIATION AND INVASION MARKERS • PATHWAYS SPECIFIC MARKERS • MARKERS OF GENETIC AND EPIGENETIC INSTABILITY

Issues in Using Biomarker • The observed change may not correlate with end point (cancer incidence, response to therapy, etc). • Can not measure the quality of life. • Adverse effect may not be observed in short term biomarker studies.

High Throughput Techniques • Array technology (“-omics”) – DNA chips • c. DNA array format • in situ synthesized oligonucleotide format (Affymetrix) – Proteomics – Tissue arrays • These are powerful tools and high throughput methods to “fish-out” biomarker targets, but they are not the answers themselves • Individual targets/patterns identified need to be validated

An example of our 9000 gene mouse-arrays using differential expression analysis with Cy 3 and Cy 5 fluorescent dyes.

Proteomics • Examine protein level expression in a high throughput manner • Used to identify protein markers/patterns associated with disease/function • Different formats: – SELDI-TOF (laser desorption ionization time-of-flight): the protein-chip arrays, the mass analyzer, and the data-analysis software – 2 D Page coupled with MALDI-TOF (matrix-assisted laser desorption ionization time-of-flight) – Antibody based formats

A, GTE (20 g/ml) 3. 5 4. 5 5. 1 5. 5 6. 0 7. 0 8. 4 p. I 9. 5 3. 5 4. 5 5. 1 5. 5 6. 0 7. 0 9. 53. 5 4. 5 5. 1 5. 5 8. 4 6. 0 7. 0 8. 4 9. 5 MW (k. Da) 217 116 98 8 55 2 5 11 37 9 7 10 17 16 6 30 12 9 8 1 1 5 11 13 2 7 10 6 18 16 12 14 15 B, GTE (40 g/ml) 3. 5 4. 5 5. 1 5. 5 6. 0 7. 0 8. 4 18 14 3 20 13 17 3 15 4 4 p. I 9. 5 3. 5 4. 5 5. 1 5. 5 6. 0 7. 0 8. 4 9. 5 MW (k. Da) 217 116 98 55 5 1 10 11 12 14 30 1 5 13 17 37 20 19 11 10 18 16 12 14 15 20 15 4 Time: 48 hr GTE: - 16 24 hr 48 hr + + 17 13 18

Tissue Array • Provide a new high-throughput tool for the study of gene dosage and protein expression patterns in a large number of individual tissues for rapid and comprehensive molecular profiling of cancer and other diseases, without exhausting limited tissue resources. • A typical example of a tissue array application is in searching for oncogenes amplifications in vast tumor tissue panels. Large-scale studies involving tumors encompassing differing stages and grades of disease are necessary to more efficiently validate putative markers and ultimately correlate genotypes with phenotypes. • Also applicable to any medical research discipline in which paraffin-embedded tissues are utilized, including structural, developmental, and metabolic studies.

Bladder Array Gelsolin HE

Lessons learned from SELDI -TOF • Initial study on patient serum from cancer patients (ovarian, prostate, etc) versus cancer showed very promising results (nearly 100% sensitivity/specificity to separate cancer from normal) – Used case-control design – Only 2 group-comparison (cancer vs. normal) – No validation • However, recent validation studies were rather disappointing

Biomarker-Directed Targeted Design • Increase the efficiency of the trial, but depends on: – The performance of the biomarker test (sensitivity/specificity) – Size of the treatment effect for targetnegative patients

BIOMARKER STUDY DSEIGN a. Untargeted Design: Treatment Register Randomize Control b. Untargeted Design: Treatment Register Test Biomarker Randomize + Control

BIOMARKER STUDY DSEIGN Biomarker by Treatment Interaction Design: Treatment Biomarker Randomize + Register Control Test Stratify Biomarker Treatment Biomarker Randomize Control

BIOMARKER STUDY DSEIGN Biomarker Based Strategy Design: Biomarker + Treatment A Test Biomarker - Treatment B Register Randomize No Biomarker Evaluation Treatment B

BIOMARKER STUDY DSEIGN Modified Biomarker Based Strategy Design: Biomarker + Treatment A Test Biomarker - Treatment B Register Randomize Treatment A No Biomarker Randomize Evaluation Treatment B

GENETIC SUSCEPTIBILITY MARKERS • Genome Wide Association (GWA) studies help to identified genetic markers for cancer – Prostate: Chromosome 8 q 24 (Gudundsson, et al, Nature genetics/Yeager, et al, Nature Genetics, 2007) – Lung: Chromosome 15 q 25 (nicotinic acetylcholine receptor subunits) (Huang, et al, Nature 2008/Amos, et al, Nature Genetics, 2008/Thorgerisson, et al, Nature genetic, 2008) • Genes identified in these locus may also be the targets for chemopreventive drug development

SUMMARY • No “magic” single marker, or a pattern of marker, that will address all issues about cancer • Know exactly what the marker is used for, and the criteria/gold standard for marker may be different: – Early detection - precancer lesion – Diagnosis - cancer – Therapeutic selection/monitoring - response – Prognostic indication - progression • “-omics” is a useful commodity to find marker. However, more simplified method should be used to detect marker clinically

Thank You!!!