We determined the success rate of new drug approval by the US FDA in two breast cancer indications, one of which used a biomarker. This allowed us to assess if biomarkers improved clinical trial risk in breast cancer. We performed a retrospective screening of industry-sponsored drug development programs registered on clinicaltrials.gov from 1998 to 2012 for HER2-positive patients compared to patients that had either failed or had been exposed to anthracycline or taxane, whose first phase I in this indication occurred no earlier than 1998. Compounds not registered on clinicaltrials.gov and studied exclusively outside the US were excluded. Twenty-nine drugs for HER2-positive patients and 28 drugs for anthracycline/taxane-exposed patients met our screening criteria. The overall success rate of new drug development in anthracycline/taxane patients was only 15 %, while in HER2-positive patients it was 23 %. However, HER2-targeted therapies underperformed compared to broad acting agents. The cost for clinical trial testing alone, when adjusted for the risk of failure, for HER2-positive breast cancer patients was $199 million, significantly lower than the cost of $274 million for anthracycline/taxane-experienced patients. The use of a validated biomarker, such as HER2, reduced clinical trial risk by as much as 50 % resulting in cost savings of 27 % in advanced and metastatic breast cancer. However, these data have to be evaluated in a context in which studies combining a novel drug with a novel biomarker not yet recognized by the FDA may actually increase clinical trial risk.