Opportunities and challenges in oncology targeted drug development : an assessment of the use of prevalence and companion diagnostic performance thresholds to guide clinical trial strategies

Targeted, especially stratified or biomarker-guided, therapies offer significant advantages over traditional oncology therapies in certain settings. Selecting patients most likely to respond to a drug increases the therapeutic efficacy while reducing toxicities and may accelerate regulatory approval since smaller clinical trials are needed to demonstrate benefit. Several drugs, including vemurafenib and crizotinib have demonstrated these benefits along with commercial success. However, significant risk exists for the drug developer since approval may be threatened if they fail to meet unclear and differing yet parallel requirements for both the drug and the required companion diagnostic. Tumor biology is also increasingly complex since recent studies suggest that there are limited numbers of individual driver mutations, complicated interactions throughout signaling pathways as well as extensive tumor heterogeneity, all of which will challenge the effectiveness of targeted therapies. Clinical trial strategy decisions can greatly impact the success of a targeted therapy due to these challenges. While therapeutic efficacy is considered important, biomarker prevalence and companion diagnostic performance have been shown to be as important, yet more informative at the time decisions are made. I hypothesized that common prevalence and companion diagnostic performance thresholds are being used to guide biomarker-guided clinical trial strategy decisions for targeted oncology therapies. Seventeen interviews with preclinical, clinical or translational leads were conducted across a focused set of ten "pathway-modifying" cancer drug programs (CDK4/6, MDM2 and P13KP inhibitors) that reflect the biological complexity of future targeted therapies. These interviews provided empirical data as to how biomarkers are being incorporated into current clinical trial decisions. All respondents were planning to use a companion diagnostic for their program, however, the use of biomarkers varied significantly. For those programs with ongoing clinical trials in phase I and II, 54% (n=7/13) were pursuing a biomarker-guided strategy while 46% (n=6/13) were using an initial all-comers strategy. This fairly equal split separated when compared by phase where trials in phase I and 1/11, 60% (n=6/10) were using an all-comers strategy but for those trials in phase II (n=3), all were using biomarker-guided strategies. A key finding of the interviews was that 66.7% (n=4/6) were planning biomarker enrichment as part of expansion plans. Disproving my hypothesis, however, common thresholds for neither biomarker prevalence nor companion diagnostic performance were being used to guide these decisions. Biomarker prevalences of 50-100% were stated as potentially appropriate for an all-comers strategy. Companion diagnostic performance thresholds were even less influential as only a few respondents provided a general range of desired sensitivity and specificity. This study found that actions of drug developers are not necessarily following the emerging recommendations for targeted therapies due to the significant challenges of biomarker and companion diagnostic development.

Department dc:contributor.department

Harvard--MIT Program in Health Sciences and Technology.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2012

Author dc:creator

• Vital, Heather Stacey Tomkinson

Advisor dc:contributor.advisor

• Ernst R. Berndt and Keith T. Flaherty.

• Harvard--MIT Program in Health Sciences and Technology.

Statement dc:rights

• M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/78156

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/78156