Download Free Medical Statistical Adaptation to Oncology Drug Development Evolution PowerPoint Presentation

Statistical Adaptation to Oncology Drug Development Evolution PowerPoint Presentation

DOWNLOAD MEDICAL POWERPOINT PRESENTATION

About This Presentation

Description : Check out this medical PowerPoint presentation titled "Statistical Adaptation to Oncology Drug Development Evolution: From Cut/Burn/Poison to Immunotherapy".

This medical PowerPoint presentation is about the shift in oncology drug development from traditional therapies to immunotherapy.

In recent years, there has been a significant shift in oncology drug development from traditional therapies such as surgery, radiation, and chemotherapy (cut/burn/poison) to immunotherapy. Immunotherapy harnesses the power of the patients immune system to target and destroy cancer cells. This shift in oncology drug development has also necessitated a corresponding evolution in statistical adaptation.

Traditional drug development relied on statistical methods to determine the safety and efficacy of drugs in clinical trials. These statistical methods were developed to assess the response rates and survival outcomes of patients treated with chemotherapy, radiation, or surgery. However, immunotherapy works differently, and new statistical methods are required to evaluate the efficacy and safety of immunotherapy drugs.

One of the major challenges of immunotherapy drug development is the variability in response rates among patients. Unlike traditional therapies, where response rates are generally high, immunotherapy may only be effective in a subset of patients. Therefore, new statistical methods have been developed to account for this variability and to identify patient subgroups that are more likely to benefit from immunotherapy.

Another challenge is the use of new response criteria in immunotherapy clinical trials. Traditional response criteria such as RECIST (Response Evaluation Criteria in Solid Tumors) are based on changes in tumor size, which may not be an accurate indicator of response to immunotherapy. Therefore, new response criteria such as irRECIST (immune-related Response Evaluation Criteria in Solid Tumors) have been developed to account for the unique mechanism of action of immunotherapy.

Finally, there is a need for new statistical methods to evaluate the safety of immunotherapy drugs. Unlike traditional therapies, which generally have well-defined toxicity profiles, immunotherapy can cause immune-related adverse events (irAEs) that may not be immediately apparent. Therefore, new statistical methods have been developed to monitor and assess the safety of immunotherapy drugs.

In summary, the shift in oncology drug development from traditional therapies to immunotherapy has necessitated a corresponding evolution in statistical adaptation. New statistical methods have been developed to account for the unique mechanism of action of immunotherapy, the variability in response rates among patients, the use of new response criteria, and the assessment of safety. These new statistical methods will play a critical role in the development of new and more effective immunotherapy drugs.

Tags : Oncology Drug Development, surgery, radiation, chemotherapy, immunotherapy, pharmaceuticals, pharmaceuticals PowerPoint presentations

Published on : Jun 23, 2023 | Views : 1064 | Downloads : 0

Shareable Url :

Embed Code :

Comment

Transcript

Slide 1-

Statistical Adaptation to Oncology Drug Development Evolution: From Cut/Burn/Poison to ImmunotherapyMingxiu Hu, PhD, ASA FellowSr. Vice President, Data Science and SystemsAdjunct Professor of Biostatistics, Yale University3rd OncoStat Symposium, April 26-27, 2019
Slide 2-

Topics History of Cancer Treatments Surgery/Radiation/Chemotherapy History of Immunotherapy Evolution of Oncology Drug Development Clinical Trial Design Analysis Methodology 2
Slide 3-

How does cancer develop? Cell reproduction error Sunlight or toxins Mutated by virus or genetically Aging Random error 3 Immune system Recognize and kill 99.9999% of mutant cells 0.00001% escape and lead to uncontrolled cell production–cancer
Slide 4-

Surgery/Radiation/Chemotherapy/Targeted Therapy Surgery (1700’s) and Radiation (1896) Early stage and localized cancer Not work with metastatic cancer (majority) Chemotherapy/Targeted Therapies (1946) Direct attack on cancer Kill cancer and normal cells Lots of side effects Can come back quickly Cure of cancer was only a dream Efficacy is limited A few month improvement in survival Quality of life suffers 4
Slide 5-

Early Hints of Immunotherapy Elizabeth Dashiell 17 years old, Rockefeller’s lover Cross-country train journey in the summer of 1890 Returned in August with a swollen pinkie (half size of an olive) Grow back a week after a perfect surgery Diagnosed sarcoma after second surgery Spread quickly and died on Jan 31, 1891 5
Slide 6-

Early Hints of Immunotherapy Fred Stein (1885) 31 years old, a house painter Egg-sized mass on left cheek 5 surgeries over 3 years Tumor roared back after each surgery Skin grafts left open wound and led to erysipelas infection (deadly infection in 19th century) Red rashes, raging fever, chills, inflammation, … Miraculously, his tumor melt away! 6 Infection made the tumor “hot”
Slide 7-

Coley’s Toxin Invented by Dr. William Coley in 1890’s Recipe: Meat-water + salt + peptone + bacteria Induce infection to recruit immune cells to kill cancer No understanding of immune system, genes, mutations at that time Inconsistency of efficacy reporting: Coley’s daughter: 500 remissions out of 1000 patients A 1960 report: 20 out of 93 Serious side effect (including death) 7 Setbacks for immunotherapy: 1952, Park Davis stopped production 1963, FDA not acknowledge as a proven cancer therapy 1965, American Cancer Society called it “Unproven Methods of Cancer Management”
Slide 8-

How does immune system work? Innate Immune Response Macrophages recognize and eat obvious invaders (bacteria, viruses, sick cells, etc.) Adaptive Immune Response Against unfamiliar invaders Foreign cells B-cells mark invaders using its antibodies Macrophages eat them Infect body cells to produce viruses: CD4 helper T-cell: Coordinate CD8 killer T-cell: Kill 8 Cancer: Sick body cell Not infected but mutated
Slide 9-

Long road to modern immuno-oncology therapy ~120 years from Coley’s Toxin (1890) to the first modern immunotherapy (Ipilimumab) 2011 Long time to understand our immune system: T-cells was discovered in 1968 Different types of T-cells (CD4, CD8, CD28, etc.) distinguished later Many misunderstandings about cancer and immune system: T-cells cannot see cancer: No symptoms and too similar to normal cells Setbacks for immunotherapy for many years due to this misunderstanding Even Ipilimumab was developed based on a misunderstanding Immune Response: Recognition → Activation →Attack/Kill 9
Slide 10-

Immunoncology Therapy—Checkpoint Inhibitor Ipilimumab, 2011 Mistakenly thought CTLA-4 protein is an immune response gas pedal and Ipilimumab is an agonist that promotes immune response Actually, CTLA-4 is a brake and Ipilimumab is an CTLA-4 inhibitor. Two wrongs make one right Patent was granted based on the wrong understandings Inhibit CTLA-4, unleash general activation of T-cell responses Lots of side effect 10
Slide 11-

Immunoncology Therapy—Checkpoint Inhibitor PD-1/PD-L1 Inhibitors: PD-1 inhibitors: Nivolumab and Prembrolizumab, 2014: PD-L1 inhibitors: Atezolizumab, 2016 PD-1/PD-L1 shot down T-cell attacks after immune response activation PD-1/PD-L1 inhibition is more specific and less side effect 11
Slide 12-

Immuno-oncology Therapy—CAR-T Cell Therapy CAR-T Cell Therapy: Collect a patient’s own T-cells Modify in vitro by adding CAR gene Better recognize and attack cancer Inject back to the patient So far mainly for liquid tumors (tests ongoing for solid tumors) 12
Slide 13-

Future Direction of Immunotherapies 13 INFLAMED (“hot” tumor) CD8+ T Cells infiltrated But not functional IMMUNE EXCLEDED CD8+ T Cells accumulated but not efficiently infiltrated IMMUNE DESSET CD8+ T Cells absent from tumor and its periphery Respond favorably to checkpoint Convert to inflamed phenotype with inhibition combinations
Slide 14-

Future Direction of Immunotherapies 14 Hot Tumor vs Cold Tumor Only hot tumors respond well to checkpoint inhibitors TMB=Tumor Mutation Burden
Slide 15-

Future Direction of Immunotherapies 15 IR=Immune Recognition; TIL=Tumor Infiltrating Lymphocytes; MSI=MicroSatellite Instability Future Immunotherapy will be highly personalized: TMB, PDL1, MSI, MSS, etc
Slide 16-

Special Statistical Issues with Oncology Immunotherapy Development Focus on two key issues: Delayed Effect--Non-proportional hazard Underestimate treatment effect at interim analysis How to estimate treatment effect Personalization Identify sub-populations to distinguish a new treatment Complex designs 16
Slide 17-

Delayed Effect Examples—Nivolumab in Melanoma and Renal Cell Carcinoma (RCC) 17 PFS in Melanoma OS in Melanoma PFS in RCC OS in RCC
Slide 18-

Delayed Effect—More Powerful Test Statistics 18
Slide 19-

Delayed Effect—More Powerful Test Statistics 19
Slide 20-

Delayed Effect—Power Comparison 20
Slide 21-

Delayed Effect—Sample Size Comparison 21
Slide 22-

Delayed Effect—Estimation of Treatment Effect Non-proportional hazard Single piece of HR does not capture real effect A Practical Approach (see our paper for details): Estimating separation time Estimating HR before separation time Estimating HR after separation time 22
Slide 23-

Delayed Effect—Design Consideration Group Sequential Designs: Don’t use futility or use lower futility hurdle Add more interim analyses (longer PFS and OS) Build in delayed effect to determine “optimal” timing if information available Sample/even size Adaptive Design: Build in delayed effect in determining conditional power, if information available Delay interim decision Eliminate futility zone and unfavorable zone to be conservative 23
Slide 24-

Complexity of Trial Designs Design Factors: Multiple treatment regimens (different combinations) Multiple primary endpoints (ORR, PFS, OS) Multiple tumor types (basket trial) Multiple interim analyses Interim adaptations (enrichment, sample size adaptation) Identify gene signature and confirming treatment effect 24
Slide 25-

Challenges in Identifying Effective Subgroups: Our big data problem is different Outcome is patients’ life and cannot be lightly tested Life experiment vs. social media experiment Data collection involves a long, expensive, and strictly regulated process Most variables are not predictive and predictive ones can only explain a small amount of variability Statistical methods have to be efficient, reliable, and reproducible, which is lacking Most gene signature findings are not reproducible 25