Recently I was honored to co-author a letter to the editor of JAMA with Eric Topol, MD and Stuart Kauffman, PhD,MD in response to an article calling for more randomized trials. The letter was to have been an article and had other co-authors, but JAMA requested a response which by their editorial guidelines understandably limits co-authorship. Here is the response to the article (below) and a link to the article itself.
Study Design and the Drug Development Process
To the Editor
A Viewpoint by Dr Djulbegovic and colleagues1 claimed not only that randomized clinical trials (RCTs) aremore ethical but that greater use of randomized designs throughout the drug development process would “improve the efficiency, ie, enable faster development ofnew, successful treatments.” However,RCTs are outdated for several reasons. First,RCTs are inadequate to evaluate cancer therapies. Genomic analysis is uncovering the tremendous heterogeneity of what previously were considered single diseases. Genomic analysis of cancers of individual patients is disclosing the large number of mutations, and thus targets, within one person. Developing RCTs for targeted therapies would be difficult and timeconsuming, prolonging the wait for effective treatments.
Second, RCTs are a less efficient and accurate method of establishing which treatments work.
A recent study demonstrated that RCTs fail to predict or improve outcomes when evaluating multicausal diseasenetworks or treatments orwhen assessing which particular interactions are relevant.2 Rather than increase the use of RCTs, we propose an increase in the use of N-of-1 studies, which are based on simulations that rapidly sort through billions of possible interactions
at the clinical, genomic, and biological levels to arrive at predictive models of multicausality. This approach has identified candidate markers, which have been successfully used in clinical trials.3 These studies also have predicted the most effective treatments for patients based on a particular genotype
and medical history when RCTs have not.4 Increasing the use of RCTs would add to the cost and time required to developand use new products. Other analytic methods more quickly and precisely match patients to treatment.
1. Djulbegovic B, Hozo I, Ioannidis JPA. Improving the drug development
process: more not less randomized trials.JAMA. 2014;311(4):355-356.
2. Eppstein MJ, Horbar JD, Buzas JS, Kauffman SA. Searching the clinical fitness
landscape.PLoS One. 2012;7(11):e49901.
3. Wu C-C, D’Argenio D, Asgharzadeh S, Triche T. TARGETgene: a tool for
identification of potential therapeutic targets in cancer.PLoS One.
2012;7(8):e43305.
4. Garnett MJ, Edelman EJ, Heidorn SJ, et al. Systematic identification of
genomic markers of drug sensitivity in cancer cells.Nature.
2012;483(7391):570-575.
Study Design and the Drug Development Process
To the Editor
A Viewpoint by Dr Djulbegovic and colleagues1 claimed not only that randomized clinical trials (RCTs) aremore ethical but that greater use of randomized designs throughout the drug development process would “improve the efficiency, ie, enable faster development ofnew, successful treatments.” However,RCTs are outdated for several reasons. First,RCTs are inadequate to evaluate cancer therapies. Genomic analysis is uncovering the tremendous heterogeneity of what previously were considered single diseases. Genomic analysis of cancers of individual patients is disclosing the large number of mutations, and thus targets, within one person. Developing RCTs for targeted therapies would be difficult and timeconsuming, prolonging the wait for effective treatments.
Second, RCTs are a less efficient and accurate method of establishing which treatments work.
A recent study demonstrated that RCTs fail to predict or improve outcomes when evaluating multicausal diseasenetworks or treatments orwhen assessing which particular interactions are relevant.2 Rather than increase the use of RCTs, we propose an increase in the use of N-of-1 studies, which are based on simulations that rapidly sort through billions of possible interactions
at the clinical, genomic, and biological levels to arrive at predictive models of multicausality. This approach has identified candidate markers, which have been successfully used in clinical trials.3 These studies also have predicted the most effective treatments for patients based on a particular genotype
and medical history when RCTs have not.4 Increasing the use of RCTs would add to the cost and time required to developand use new products. Other analytic methods more quickly and precisely match patients to treatment.
1. Djulbegovic B, Hozo I, Ioannidis JPA. Improving the drug development
process: more not less randomized trials.JAMA. 2014;311(4):355-356.
2. Eppstein MJ, Horbar JD, Buzas JS, Kauffman SA. Searching the clinical fitness
landscape.PLoS One. 2012;7(11):e49901.
3. Wu C-C, D’Argenio D, Asgharzadeh S, Triche T. TARGETgene: a tool for
identification of potential therapeutic targets in cancer.PLoS One.
2012;7(8):e43305.
4. Garnett MJ, Edelman EJ, Heidorn SJ, et al. Systematic identification of
genomic markers of drug sensitivity in cancer cells.Nature.
2012;483(7391):570-575.
