Retooling the War on Cancer
Retooling the War on Cancer
Dr. Schapira: Do you have any thoughts about novel ways for funding science? How should the relationships among academic centers, government agencies, and pharmaceutical companies or biotech companies evolve?
Mr. Leaf: We need lots of new models. Carl Nathan, MD, a microbiologist at Weill Cornell Medical College, has done some deep and creative thinking about the drug development question, particularly in regard to new antibiotics and vaccines -- for which there is obviously critical need, but little financial incentive to entice for-profit companies.
Depending on the specific requirement, one model might be a better fit than another. For example, when there's some good evidence to suggest that an off-patent molecule or natural agent may be active against a specific disease, it may make sense to try what some have called a "pull" model of drug development. In this model, a government body or even a well-funded charity essentially pays for early testing (or a share of it) or partners with a for-profit biotech or pharmaceutical company to pay for preclinical testing, or maybe phase 1 clinical trials. A model such as this might also be useful in funding the development of biomarker tests, which are expensive to vet and unlikely to offer a quick return on investment.
Another issue is the clinical trials we undertake. I think we have to start choosing trial designs that let us learn far more quickly about what's working and what isn't -- whether that's through adaptive trial designs, as Laura Esserman, Don Berry, and others are doing with the I-SPY 2 trial, or through so-called "N of 1" studies. The incredible individuality of patients -- and the often striking intratumoral heterogeneity within those patients -- makes it almost impossible to have a "big box" clinical trial that provides any real information. We need to try some new clinical trial models.
We also want to think about models that allow us to combine drugs earlier in the investigational stage. If we are talking about hitting multiple targets, we are going to have to combine targeted therapies -- sometimes before their marketing approval -- in order to see real effects. What we do now is serial testing: We will try a drug until we see resistance, and then we will try a second one that works on those resistance mechanisms, and then we will see new resistance mechanisms and so forth. What we need to do is combine investigational agents.
Again, we want to be careful about all of these new models. We don't want to necessarily charge into the unknown full-steam. But we have to try something. The current system just doesn't work.
New Models Are Needed
Dr. Schapira: Do you have any thoughts about novel ways for funding science? How should the relationships among academic centers, government agencies, and pharmaceutical companies or biotech companies evolve?
Mr. Leaf: We need lots of new models. Carl Nathan, MD, a microbiologist at Weill Cornell Medical College, has done some deep and creative thinking about the drug development question, particularly in regard to new antibiotics and vaccines -- for which there is obviously critical need, but little financial incentive to entice for-profit companies.
Depending on the specific requirement, one model might be a better fit than another. For example, when there's some good evidence to suggest that an off-patent molecule or natural agent may be active against a specific disease, it may make sense to try what some have called a "pull" model of drug development. In this model, a government body or even a well-funded charity essentially pays for early testing (or a share of it) or partners with a for-profit biotech or pharmaceutical company to pay for preclinical testing, or maybe phase 1 clinical trials. A model such as this might also be useful in funding the development of biomarker tests, which are expensive to vet and unlikely to offer a quick return on investment.
Another issue is the clinical trials we undertake. I think we have to start choosing trial designs that let us learn far more quickly about what's working and what isn't -- whether that's through adaptive trial designs, as Laura Esserman, Don Berry, and others are doing with the I-SPY 2 trial, or through so-called "N of 1" studies. The incredible individuality of patients -- and the often striking intratumoral heterogeneity within those patients -- makes it almost impossible to have a "big box" clinical trial that provides any real information. We need to try some new clinical trial models.
We also want to think about models that allow us to combine drugs earlier in the investigational stage. If we are talking about hitting multiple targets, we are going to have to combine targeted therapies -- sometimes before their marketing approval -- in order to see real effects. What we do now is serial testing: We will try a drug until we see resistance, and then we will try a second one that works on those resistance mechanisms, and then we will see new resistance mechanisms and so forth. What we need to do is combine investigational agents.
Again, we want to be careful about all of these new models. We don't want to necessarily charge into the unknown full-steam. But we have to try something. The current system just doesn't work.
Source...