RT Journal
A1 Avorn J
T1 APproval of a tuberculosis drug based on a paradoxical surrogate measure
JF JAMA
JO JAMA
YR 2013
FD April 3
VO 309
IS 13
SP 1349
OP 1350
DO 10.1001/jama.2013.623
UL http://dx.doi.org/10.1001/jama.2013.623
AB 
Ideally, drug approvals are based on large trials that randomize thousands of patients and measure actual clinical outcomes. By contrast, the pivotal studies for bedaquiline were relatively small; the initial finding on sputum culture conversion came from an 8-week trial of 47 patients. A follow-up study enrolled another 161 patients. By 24 weeks, 79% of patients taking bedaquiline in the second study had undergone sputum culture conversion, compared with 58% in the placebo group. The difference in each substudy was not significant at later points.1- 2