Generic manufacturers can increase their portfolio of generic sterile injectable drugs they can manufacture by filing an abbreviated new drug application (ANDA) with the FDA, which must be approved before the manufacturer can market the generic drug. An ANDA does not require that the applicant conduct preclinical or clinical trials to demonstrate safety or efficacy, instead the generic drug must be shown to be bioequivalent to the innovator drug. More ANDA approvals mean that manufacturers have more drugs to choose to manufacture on their existing capacity and therefore, that manufacturers may substitute newer drugs for other drugs. Alternatively, they may increase the rate at which they make use of their existing manufacturing capacity, though this would further reduce the elasticity of supply.
Figure 3 below shows the number of ANDA approvals for sterile injectable drugs from 1995 to 2010. As the table indicates, there was a substantial increase in the number of new injectable ANDA approvals beginning in 2008. A substantial number of these approvals were for products that during the approval process were still under patent protection (so called Paragraph 3 and Paragraph 4). These products tend to have higher margins than those where competition is already well established (Paragraph 1 and 2).
Between 2006-2010, while the overall market for sterile injectable oncology drugs increased by 14%, the number of eaches sold by generic drug manufacturers increased by nearly 30% (Figure 4). Over this period, the overall generic injectables market (including both oncology and other products) expanded by 52%. Some of this expansion was offset by reductions in brand manufacturers’ production of these drugs.
To get a better sense of how this shift in production in the market took place, we worked with the FDA to examine how the match between manufacturers and drugs evolved over this period. Figures 5 (brand) and 6 (generic) decompose the changes in manufacturer-drug pairs in the branded and generic markets over the 2006-2010 period. A manufacturer-drug pair consists of a specific manufacturer making a specific drug in a given year. For example, if Manufacturer A produced drugs a and b in 2010 and Manufacturer B produced drugs a and c in 2010, there would be 4 manufacturer-drug pairs (Aa, Ab, Ba, Bc) in 2010.
In 2006, there were 57 manufacturer-drug pairs in the branded sterile injectable oncology drug market in the US. As Figure 5 shows, between 2006-2007, 52 of those pairs continued to be supplied in the market, one new pair was introduced, and 5 pairs exited. The branded market overall is quite stable. In each year between 2006-2010, the number of changers (entrances or exits) in the market was below 11.5% of the previous year’s market. At the end of the period, the number of manufacturer-drug pairs in the market was 5% below the 2006 level.
Figure 6 shows the same decomposition for the generic industry. This sector is much more dynamic. Between 2006-2010, the number of generic manufacturer-drug combinations in the sterile oncology market increased by 45%. In every year, including 2010 (at the height of the drug shortage), the number of new combinations in the market (a manufacturer producing a drug that it had not previously produced) exceeded the number of exits. In 2007, the number of changers in the industry was 27% of the number of manufacturer-drug combinations in the prior year.
Expansion of the scope of production is also evident in the decisions of leading manufacturers to increase future manufacturing capacity. Several leading manufacturers of generic sterile injectable drugs indicated that they are upgrading existing facilities or building new facilities to serve this market. According to news reports and discussions with manufacturers, Hospira is investing $65 million in capital improvements in sterile injectable drug manufacturing sites, Teva is opening a new manufacturing site, and Ben Venue is opening a new, expanded facility to replace an older manufacturing facility. These investments will increase capacity in both older and newer generic sterile injectable drugs.
Unfortunately, this new capacity is unlikely to come on line for at least another 18 months. Meanwhile, when there is little excess manufacturing capacity, producing a new drug will often require manufacturers to reduce or stop production of another drug or to operate at a much higher than normal level of capacity utilization.
The data above suggest that the overall quantity and the range of drugs produced by generic sterile oncology manufacturers over the past 5 years has increased substantially, while the capacity of sterile injectable oncology manufacturing facilities has remained stable. In consequence, capacity utilization is likely to have increased significantly over the past 5 years. Shortages have been concentrated in drugs where the volume of sales and drug prices were declining in the years preceding the shortage, suggesting that manufacturers are diverting capacity from shrinking lines of business to growing ones.