Overspending driven by oversized single dose vials of cancer drugs

How big is the problem?

We examined the top 20 cancer drugs that are dosed by body size and packaged in single dose vials (based on 2016 projected sales), which collectively account for 93% of all sales of such drugs. We calculated the total amount of leftover drug and resulting 2016 US revenues for each drug using the method shown in fig 1⇓. In brief, we estimated how often vial sharing occurred by examining how often claims filed with the Medicare program included amounts of drug that did not total the full contents of the vial. We then calculated the most efficient way to combine available vial sizes to achieve the lowest US Food and Drug Administration approved dose in a representative sample of the US population derived from the National Health and Nutrition Examination Survey.7 After correcting for vial sharing percentage, and adjusting the population to mirror a cancer patient population, we apportioned projected 2016 US revenues to administered or leftover drug.8 9When calculating the effect of vial sharing we assumed that doses that were not multiples of available vial sizes had no leftover drug, an assumption that made our estimates of leftover drug conservative.

Fig 1 Study flowchart

• Download figure
• Open in new tab
• Download powerpoint

Table 1⇓ shows the leftover drug from the packaging approaches for the 20 drugs. We estimate total US revenue from these drugs to be $18bn (£12.5bn; €16bn) in 2016, with 10% or $1.8bn from discarded drug. The extent and cost of leftover drug varies according to market size and available vial sizes. For example, in 2016, 7% of $3.9bn in rituximab sales will be on discarded drug, totaling $254m, while 33% of $697m in carfilzomib sales will be discarded, totaling nearly as much, $231m. Sensitivity analyses suggested our results were robust. If every person received the highest dose approved by the FDA, revenue from discarded drugs falls to $1.4bn; if every cancer patients weighed 10% less than the survey participants, the estimate rises to $2bn.

The proportion of drug left over varies from 1% to 33%. Between these extremes are drugs such as bevacizumab, which comes in both 100 mg and 500 mg vials, and ipilimumab, which comes in both 40 mg and 100 mg. About 9% and 7% of these drugs, respectively, is left over. Yet small percentages can still lead to large dollar amounts. The October 2015 Medicare Average Sales Price files show that a dose of ipilumumab might cost $29 000,10 meaning that the 7% left over would generate an additional $2000 in revenue for the company for each vial sold.

How drug quantity affects profits and waste

The effect of different approaches to packaging for single dose vials is illustrated by the two drugs bendamustine and bortezomib. Bendamustine, a drug for leukemia, is sold in a broad array of single dose vials (25, 45, 100, and 180 mg) that can be combined to reach its dose of 100 mg/m² nearly precisely (fig 2⇓). Vial combinations cover every 5 mg interval across the typical adult dose range of 110 mg to 310 mg, with the exception of 130 mg and 155 mg. We calculate that only 1% of bendamustine is wasted. Bortezomib on the other hand, a drug to treat multiple myeloma, is available in the US in only a 3.5 mg vial, much larger than the average required dose, which we calculate to be 2.5 mg based on the drug’s dose of 1.3 mg/m² and the average weight of a cancer patient. Our estimate is that 27% to 30% of bortezomib sales in the US are related to leftover drug equating to $309m. The large vial size of bortezomib seems to be unique to the US market. The drug is sold in 1 mg vials in the UK.11

Fig 2 Distribution of FDA approved dose (green histogram) in the US population of cancer patients, and available combinations of full vial contents (red lines) to achieve that dose for bortezomib (top) and bendamustine (bottom)

• Download figure
• Open in new tab
• Download powerpoint

Pembrolizumab provides another example of how vial sizes can influence revenues. When it was initially approved in the US in September 2014, the drug was sold in 50 mg vials (as a powder that needs to be reconstituted into a liquid). But in February 2015 the manufacturer introduced a larger 100 mg vial (as a liquid) and stopped distributing the 50 mg vials to the US market. Five months later, in July 2015, pembrolizumab was approved in Europe, where it is sold in the smaller 50 mg vials as a powder.

The increased revenue from the change is substantial. Consider a 70 kg patient who requires a dose of 140 mg (the drug is dosed at 2 mg/kg). When the drug was sold in 50 mg vials, reaching the desired dose would require three 50 mg vials and leave 10 mg unused. But with only 100 mg vials available, 60 mg is left over. According to the Medicare Not Otherwise Classified October 2015 file, which lists Medicare’s reimbursement rates for these drugs, each milligram of pembrolizumab costs around $50. In this example the change in vial size alone increases the revenues for the company from leftover drug by sixfold, from $500 to $3000, for a single dose. We estimate that the additional revenue to the company from the packaging change over the next five years will be $1.2bn, which comes on top of the $1.2bn they would have gained from leftover drug with the 50 mg package (table 2⇓). Similarly, by only selling bortezomib in the US in the larger 3.5 mg vials rather than the 1 mg vials sizes available in Europe, the manufacturer, Millennium, will increase its 2016 US revenues by $130m (data not shown).11

How can we stop the waste?

Regularly and systematically discarding expensive drugs is antithetical to efforts to reduce spending on healthcare services that provide no value. Policy makers should therefore explore approaches that would reduce or eliminate paying for leftover drug. Current regulatory standards could be viewed as contradictory, or at least as ambiguous (box). The FDA calls on companies to balance vial contents so that leftover drug is minimized yet they should also provide enough drug that more than one vial is rarely needed for a single dose.17 Guidance on vial sharing is also inconsistent. The Centers for Medicare and Medicaid Services essentially encourages it; the Centers for Disease Control and Prevention states that it is unsafe (box).18 19

Several policy options merit exploration. Regulators could require manufacturers to provide drugs in a reasonable set of size options to ensure the amount of wasted drug is low, say 3%. This is achievable, as table 3⇓ shows. If all of our suggestions were adopted, it would lower revenue from leftover drug from $1.8bn to $400m and, including the reductions to doctor and hospital mark-ups on leftover drug, would save around $2bn in total. An alternative would be to leave manufacturers free to select their vial sizes but also require them to refund the cost of leftover drug. This could be achieved through certified disposal and a virtual return.

One pattern sometimes seen in clinical practice is to round up doses to the quantity in the full vial, thus changing dosing from body sized based to “flat” or “fixed” dosing. The approach is problematic not only because it leads some patients to receive too high a dose and others too low when compared to the FDA approved dose, but also because it does not reduce spending on leftover drug. It merely changes clinician behavior from discarding leftover drug to infusing leftover drug into patients.

Policy makers should also revisit the current FDA guidance on the appropriate packaging of infused drugs in single dose vials and encourage the FDA, CDC, Centers for Medicare and Medicaid Services, and US Pharmacopeial Convention to reconcile their views on vial contents and vial sharing. Such steps could lead to savings for our healthcare system without sacrificing health outcomes. Opportunities to eradicate waste of this kind are rare.