Cancer Pharmacogenomics: Setting a Research Agenda to Accelerate Translation

Slide 1 of 25: Genomic Guided Clinical Trials to Evaluate the Clinical Utility of Cancer Pharmacogenomics

Geoffrey S. Ginsburg, M.D., Ph.D.
Director, Center for Genomic Medicine
Duke Institute for Genome Sciences & Policy

July 21, 2009


Slide 2 of 25: A Paradigm for Discovery and Development of Cancer PGx Biomarkers

[Image] describing a paradigm for discovery and development of cancer pharmacogenomics biomarkers.


Slide 3 of 25: Concept of a Genomic Signature

[Image] describing the concept of a genomic signature.


Slide 4 of 25: Lung Cancer Prognosis Genomic Signatures: The General Approach

[Image] describing lung cancer prognosis genomic signatures.


Slide 5 of 25: A Metagene Predictor of Recurrence

[Images] describing a metagene predictor of recurrence.

Potti A et al. NEJM. 2006; 355 (6): 570-80.
Copyright © 2006 Massachusetts Medical Society. All rights reserved.


Slide 6 of 25: Independent Validation

[Images] describing independent validation.

Potti A et al. NEJM. 2006; 355 (6): 570-80.
Copyright © 2006 Massachusetts Medical Society. All rights reserved.


Slide 7 of 25: An Opportunity to Improve Prognosis in Lung Cancer

[Images] describing an opportunity to improve prognosis in lung cancer.

Potti A et al. NEJM. 2006; 355 (6): 570-80.
Copyright © 2006 Massachusetts Medical Society. All rights reserved.


Slide 8 of 25: CALGB 30506 – A Phase III Trial to Evaluate Genomic Prognosis

[Image] describing a Phase III CALGB trial (#30506) to evaluate genomic prognosis, approved by NCI’s Cancer Therapy Evaluation Program (CTEP) in May 2009.  The key points are:

  1. Does the genomic assay accurately predict low versus high risk?
  2. Do patients predicted to be at high risk for recurrence benefit from chemotherapy?

Potti A et al. NEJM. 2006; 355 (6): 570-80.
Copyright © 2006 Massachusetts Medical Society. All rights reserved.


Slide 9 of 25: How Best to Treat? Many Choices But No Guidance

The speaker’s key point was that the regimens were developed for groups of cancer patients, not individuals. The speaker illustrated his point using an example of the National Comprehensive Cancer Network Practice Guidelines in Oncology for Non-Small Cell Lung Cancer (NSCLC) which contain details about chemotherapy regimens used in 4 randomized trials of adjuvant chemotherapy in NSCLC.


Slide 10 of 25: Signatures of Drug Sensitivity

[Image]  of signatures of drug sensitivity.

Reprinted by permission from Macmillan Publishers Ltd:Nature Medicine, Potti A et al. Nature Medicine. 2006; 12 (11): 1294-1300. Copyright 2006.


Slide 11 of 25: A Panel of Signatures to Guide the Use of Cytotoxic Chemotherapies

[Images]  of a panel of signatures to guide the use of cytotoxic chemotherapies.

Reprinted by permission from Macmillan Publishers Ltd:Nature Medicine,
Potti A et al. Nature Medicine. 2006; 12 (11): 1294-1300. Copyright 2006.


Slide 12 of 25: A Protoype for Clinical Utility Studies: Guiding Standard of Care Therapies

[Image] of a prototype for clinical utility studies for standard of care therapies.


Slide 13 of 25: A Breast Cancer Neoadjjuvant Trial

[Image] outlining trial design for a DOD-funded study of breast cancer neoadjuvant therapy initiated in June 2008.


Slide 14 of 25: EORTC 10994 Multicenter Prospective Neoadjuvant Phase III Breast Cancer Trial: Blinded Validation (n = 162)

[Image] outlining trial design for a blinded validation of a multicenter prospective neoadjuvant phase III breast cancer trial consisting of 162 participants with locally advanced/inflammatory breast cancer. The participants were randomized to two study arms. Arm A (non-taxane arm) received neoadjuvant PEC therapy and Arm B (taxane arm) received neoadjuvant ET therapy. Participants were assessed for complete pathologic response. There was also an independent validation of chemotherapy response signatures using biopsies from all participants.

Modified from Bonnefoi H et al. Lancet Oncology. 2007; 8: 1071-8.


Slide 15 of 25: EORTC 10994 Multicenter Prospective Neoadjuvant Phase III Breast Cancer Trial: Blinded Validation (n = 162)

[Image]  showing the probability of FEC or ET sensitivity for Arm A and Arm B, respectively.

Bonnefoi H et al. Lancet Oncology. 2007; 8: 1071-8.


Slide 16 of 25: Novel Paradigms for Drug Development

[Image] showing novel paradigms for drug development.


Slide 17 of 25: Surrogate Signatures for Pathway Activation Underlying the Oncogenic Phenotype

[Image] showing surrogate signatures for pathway activation underlying the oncogenic phenotype.


Slide 18 of 25: Linking Pathways Underlying the Oncogenic Phenotype with Therapeutics

[Image showing linkages between pathways underlying the oncogenic phenotype and therapeutics.


Slide 19 of 25: Pathway Signature = Drug Sensitivity Signature

[Image]  showing a pathway signature as a drug sensitivity signature.

Bild A et al. Nature. 2006; 439: 353-7. Reprinted by permission from Macmillan Publishers Ltd: Nature, copyright 2006.


Slide 20 of 25: Novel Paradigms for Drug Development

[Image] showing a pathway predictor for enrichment/targeted therapies.


Slide 21 of 25: A Multi-Step Strategy for Personalized Cancer Therapy

[Image] showing current pathway where a cancer patient receives the standard-of-care versus a genomics-guided pathway where the cancer patient would undergo a biopsy (or provide another type of tumor sample) so that recurrence prediction could be performed to help clinicians identify who to treat. Chemotherapy response prediction could also be conducted to help clinicians determine how to treat cancer patients, i.e. which chemotherapy regimen and/or targeted drugs.


Slide 22 of 25: The Duke Clinical Genomics Studies Unit: Driving Genomics Guided Trials Therapy

[Image] showing the organizational structure of the Duke Clinical Genomics Studies Unit. The Medical Director is responsible for the Clinical Genomics CRCs (and interacts with the PIs), Operations and Projects Management, and Clinical Genomics Technologies.


Slide 23 of 25: The Duke Clinical Genomics Studies Unit

[Image] showing the process the Duke Clinical Genomics Studies Unit utilizes for its research.


Slide 24 of 25: Building the Infrastructure to Make this Work

  • Biobanking
    • Coordinated efforts
    • Operational and informatics support
    • Standards
  • Genomic Technologies
    • Core laboratories
    • Economies of scale
  • Informatics
    • Reliable, interoperable EHRs
    • Integration of research, clinical, molecular data
  • Biostatistics
    • Critical shortage must be addressed
    • Physician training in quanitative skills
  • Decision Making
    • Understanding of human decision making
    • Biological, psychological and social factors
    • Education of health care professionals

Source:  Califf and Ginsburg, JAMA, 2008.


Slide 25 of 25: Opportunities to Enable Scientific and Clinical Evaluation of Genomic Markers

  • Patient registries (common and rare diseases)
    • Longitudinal follow up
    • Robust phenotypes
  • Population studies linked to EHRs
  • Prospective clinical trials
    • “Genomics Trials Cooperative Group”
  • Industry
    • Public-private partnerships
    • Sample collection in phase II-IV trials
  • A national virtual sample biorepository linked to research and clinical data

Return to Top