A new study published in JAMA Oncology investigated the pricing of new cancer drugs to determine what impacts the drug’s cost most: (1) a relative health benefit such as a very small improvement in overall survival, or (2) a novel advance in treatment such as targeting a new mechanism of action that results in a significant increase in overall survival without putting patients at undue risk.

51 oncology drugs approved by the U.S. Food and Drug Administration between January 1, 2009, and December 31, 2013, were identified and examined.

The drugs approved based on disease response rate (e.g., tumor shrinkage) were priced the highest, with median costs per year of treatment of $137,952. Drugs approved on the basis of overall survival were less expensive (median cost, $112,370 annually). And those approved on the basis of progression-free survival (no significant increase in cancer), or disease-free survival (no evidence of cancer recurrence) were least expensive, with a median cost of $102,677 per year.

When analyzing drug price in relation to improvements in either progression-free survival or overall survival, there was no significant relationship between cost and the percentage of improvement. In addition, there was “little difference in the median wholesale price of 21 novel drugs and 30 next-in-class drugs approved over a 5-year period (next-in-class drugs, $119 765; novel drugs, $116 100).” The findings suggest that drug prices are not based on novelty.

Implications

Drug prices (including oncology drugs) are rising faster than other sectors of the health care system, with the cost of new cancer drugs regularly exceeding $100,000 annually. The study’s authors argue that novel drugs offering significant benefits to patients may well justify high drug costs. However, the study results do not support this conclusion.

Source: Mailankody S, Prasad V. Five Years of Cancer Drug Approvals: Innovation, Efficacy, and Costs. JAMA Oncol. Published online April 02, 2015. doi:10.1001/jamaoncol.2015.0373.


Further Reading: