Quality control in generic anthelmintics: is it adequate?

We became increasingly concerned about indications of possible substandard efficacy of some generic anthelmintics, particularly after P.C. van Schalkwyk (personal communication, 1990) had found some batches of imported generic products obtained from international brokers to be poorly active, despite apparently normal physical characteristics. Therefore, considering the serious consequences this would have for sheep farming, it was decided to test the efficacy of some of the generic rafoxanide products available on the South African market. One of the three commercial formulations (of highly reputable companies) tested against a known susceptible strain of Haemonchus contortus in sheep was markedly substandard, with an arithmetic mean efficacy of 66.2% (Class B, Reinecke, 1973), compared to Class A efficacy of the other two, which also differed significantly from one another (Mann-Whitney; P = 0.01). Larger differences were found between the three products against a natural infection with a partially resistant strain of H. contortus than against the susceptible strain, with corresponding arithmetic mean efficacies of 28.7% (Class X, or ineffective), 71.3% (Class B) and 87.7% (also Class B). It is concluded that the most likely reason for the observed differences is that international brokers do not disclose the sources of supply of different batches of active ingredient (with the result that the companies buying anthelmintics from them have no way of telling when a source of supply is changed); that the efficacy of such batches differs; and that efficacy testing of individual batches in some cases is inadequate. It is suggested that registering authorities should consider simplified efficacy testing of each new batch of active ingredient before it may be marketed.