What does the Orphan Drug Tax Credit tell us about the Costs of Clinical Trials?

By James Love

Summary:

  • The number of patients enrolled in the trials used to support the registration of novel orphan product are significantly smaller that non-orphan products.  One measure of this is the difference in the enrollment of trials cited in the FDA drug trials snapshots.
  • Since 2015, the average number of trials cited in the FDA trials snapshots for novel drugs were 439 for orphan products, and 2,736 for non-orphans.
  • Data from the Orphan Drug Tax Credit provides insights into the costs of drug development, or more specially, the costs of the clinical trials used to support an FDA approval.
  • From 2010 to 2016, the average qualifying trial costs claimed for the orphan drug credit was $86 million to $102 million, per FDA approved orphan indication (assuming 2 to 3 year average years of lag between the credit claimed and the approval date).  Companies were able to take a credit of $43 to $51 million, on average, for each FDA approval.
  • The $86 to $102 million in pre-credit outlays is far lower than the average of $965 million on trial costs for a new drug approval, estimated by DiMasi and others in 2016.  Some of the differences are explained by the smaller trials for orphan drugs and other differences in methodologies, although both figures include the costs of failed trials and exclude pre-clinical or cost of capital costs.
  • In 2013, the last year for which we have actual rather than projected data on the credit (from the IRS Statistics of Income), the total amount of the credit from all 132 corporate tax returns that claimed the credit was just over $1 billion, nearly the same amount as the DiMasi estimate of $965 million for a single drug. But in 2013, the FDA granted 265 orphan designations and approved 33 orphan indications, including 8 novel products which were approved for an orphan drug lead indication.
  • The data from the orphan drug tax credit illustrates the large gap between the known facts about the costs for R&D for orphan drug development, and the astronomically larger R&D costs claimed by DiMasi (and frequently quoted by other researchers, policy makers and journalists) as averages that should guide policy making.
  • These data underline the need for greater transparency of R&D costs, and more sophistication and realism by policy makers regarding the costs of research and development for drugs qualifying as orphan products.
  • The data from the orphan drug tax credit also provides additional perspective on the estimates of drug development costs provided by Vinay Prasad and Sham Mailankody in their 2017 JAMA paper.


When the GOP House leadership rolled out their proposed tax reform bill, one of the surprises was the proposal to eliminate the Orphan Drug Tax Credit (ODTC). The Senate tax bill rolled out a few days later proposed cutting the credit, by as much as half.  It’s not obvious how wedded the GOP is to reducing or eliminating the credit, which is popular with the rare disease community and is a significant subsidy for the development of drugs for cancer and other severe illnesses. However, the proposed repeal has created space to take a new look at the ODTC, and to consider long overdue reforms.

I will present suggestions for reforms in another blog, but first I wanted to provide some background on what the ODTC is, how it works, and also to examine what the credit equal to 50 percent of the cost of qualifying clinical trials can tell us about drug development costs.

For those not familiar with the Orphan Drug Act and the Orphan Drug Tax Credit, here are some key provisions.

The Orphan Drug Designation

The Orphan Drug Act provides several benefits to drug manufacturers, all of which are contingent upon an FDA orphan drug designation.

The orphan designations are given for a combination of the drug and the qualifying disease or condition.

In the current process, a company asks the FDA for the designation. The request must provide a description of “the drug” and “a discussion of the scientific rationale to establish a medically plausible basis for the use of the drug for the rare disease or condition.”

Any disease or condition that has less than 200,000 patients (for a vaccine, less than 200,000 patients per year) in the United States will qualify as an orphan.
A disease that involves 200,000 or more patients can also qualify, if “there is no reasonable expectation that costs of research and development of the drug for the indication can be recovered by sales of the drug in the United States.”

The population to be served can be defined by a disease, like ovarian cancer, which has fewer than 200,000 patients, or as a “medically plausible” subset of patients for a more common disease, when the use of the drug is not appropriate outside the subset. Companies have obtained orphan designations for drugs used to treat “subsets” of patients with common diseases like lung, breast and prostate cancers.
One drug can receive multiple designations. For example, the BMS drug Opdivo has already received 13 FDA orphan designations, including 4 that have been approved by the FDA.

A drug that is approved for a non-orphan indication can still receive one or more orphan designations, for different uses of the same drug.

Benefits of the Orphan Designation

Once a designation is approved by the FDA, the drug developer is entitled to several benefits, including most importantly a reduction in FDA user fees, an expedited approval process, seven years of exclusive rights to market the product (even if there are no patents involved), and a tax credit equal to 50 percent of the costs of “qualifying” clinical trials that occur after the date of the designation and before FDA marketing approval for that indication.

The Orphan Drug Tax Credit

The credit can be used to directly offset federal income tax liability. If necessary, there is a 20 year carryforward allowance for the credit, making the credit a bankable asset.
Here are a few important nuances:

  1. Qualifying trial costs must normally be incurred in the United States. However, a company can claim the credit for foreign trials if a sufficient testing population is unavailable in the United States.
  2. The credit cannot be claimed for expenses that were funded through federal grants or research contracts.

The Disappearing Means Test

As noted above, to obtain an orphan designation, a drug manufacturer must “demonstrate that there is no reasonable expectation that development and production costs will be recovered from sales of the drug for the orphan indication in the United States,” but only for a case where the population to be treated is 200,000 patients or more.

When the Orphan Drug Act first passed in 1983, the test regarding the economic viability of the drug was applied to any orphan designation. The Act originally read as follows:

Public Law 97-414

“SEC. 526. (a)(2).  For purposes of paragraph (1), the term ‘rare disease or condition.” condition’ means any disease or condition which occurs so infrequently in the United States that there is no reasonable expectation that the cost of developing and making available in the United States a drug for such disease or condition will be recovered from sales in the United States of such drug. Determinations under the preceding sentence with respect to any drug shall be made on the basis of the facts and circumstances as of the date the request for designation of the drug under this subsection is made.

This language was modified in the 1984 amendments, in Public Law 98-551, so that any drug used to treat a population of less than 200,000 patients automatically qualified, and the test regarding expected profits only applied to drugs with 200,000 or more patients.

As a consequence of the 1984 amendments, the Orphan Drug Tax Credit became an open ended subsidy for any drug that could qualify.

The Costs of the Credit are Growing

Over the years, both the absolute number of new drugs that qualify and the percentage of novel drugs that qualify have increased sharply.

In FY 2013, the Department of the Treasury estimated the cost of the credit (to the Treasury) at $1 billion. In FY 2017, the estimated cost of the credit was $2.28 billion. By FY 2022, the credit is estimated to cost $5.9 billion, a 20.9 percent compound annual rate of growth. By 2027, the Treasury estimates the cost of the credit to continue to increase by 20.9 percent per year, to $15.29 billion.

The Treasury’s $75 billion estimate of cost of the credit over ten years presents a challenge to those who want to restore the credit in the tax package, since this will require offsetting adjustments to the taxes for other taxpayers, and an increase to the deficit.

What Does the Credit Tell Us about the Costs of Clinical Trials?

The IRS has published the amount claimed for the Orphan Drug Tax Credit from 1983 to 2013, and the FDA publishes data on orphan designations and approvals. One simple calculation is to divide the amount of credit in a given year by the number of designations or approvals. In 2013, there were 265 orphan designations, and 33 approvals of those designations. According to the IRS, the amount of the credit claimed on corporate returns was $1,078,314,000. This works out to $4 million per designation, and $32.7 million per approval. The approval number can be thought of as risk adjusted number, representing half of the costs of the trials.

But using the same year for approvals or designations can be misleading. The trials often take more than one year, and are typically initiated well before the approvals, which would not be an issue for the calculation, except that there has been a sharp increase in both designations and approvals. It is better to look at the relationship between the designations and the approvals with a time lag, to get a more realistic view of the amount of the credit being claimed for each orphan designation or approval.
Given what is known about the lags between designation and approval (see this earlier analysis), and the expedited nature of orphan approvals, a lag of two to three years may be appropriate.

In Table 1, the cost of the ODTC per designation and per approval is presented, for five cases — the lag periods of 0 to 4 years. The years 2010 to 2013 are the IRS Statistics of Income estimates of the amount claimed for the Orphan Drug Tax Credit on Corporate returns (available here). The years 2014 to 2016 are the Treasurer estimates of the amount of the deduction taken by all taxpayers, pursuant to the Congressional Budget Act of 1974 (Public Law 93–344), which requires that a list of “tax expenditures’’ be included in the budget (available here).

Table 1: Amount of ODTC (in millions of USD) per approval of an orphan designation

Year ODTC Orphan Designations No lag One year lag Two year lag Three year lag Four year lag
2010 $647.80 198 $3.27 $3.83 $3.90 $5.35 $4.50
2011 $755.46 202 $3.74 $3.82 $4.47 $4.55 $6.24
2012 $906.46 197 $4.60 $4.49 $4.58 $5.36 $5.46
2013 $1,078.31 265 $4.07 $5.47 $5.34 $5.45 $6.38
2014 $1,210.00 292 $4.14 $4.57 $6.14 $5.99 $6.11
2015 $1,460.00 356 $4.10 $5.00 $5.51 $7.41 $7.23
2016 $1,700.00 333 $5.11 $4.78 $5.82 $6.42 $8.63

For these seven years, the amount of the credit per orphan designation ranged from $3.9 million to $6.1 million for the 2 year lag case, and $4.6 million to $7.4 million for the three year lag case.

In other words, corporations claimed from $7.8 million to $14.8 million in qualifying trial expenditures for every orphan designation, and took credits equal to 50 percent, or $3.9 million to $7.4 million.

Note that a qualifying expenditure would exclude outlays that occurred before the FDA designation, or were conducted outside of the United States if the drug manufacturer did not claim a “sufficient testing population is unavailable in the United States.” The average also includes cases where a designation was obtained, and no trials were conducted.

In Table 2, the cost of the ODTC per FDA approval of an orphan designation is presented.These numbers provide insight into the costs of trials, on a risk adjusted basis, since the credit is divided by the number of designations that received an FDA marketing approval, and includes both the trials that were successful and those that failed or development was abandoned.

Table 2: Amount of ODTC (in millions of USD) per approval of an orphan designation

Year ODTC Orphan Approvals No lag One year lag two year lag three year lag four year lag
2010 $648 15 $43 $32 $38 $40 $27
2011 $755 26 $29 $50 $38 $44 $47
2012 $906 26 $35 $35 $60 $45 $53
2013 $1,078 33 $33 $41 $41 $72 $54
2014 $1,210 49 $25 $37 $47 $47 $81
2015 $1,460 48 $30 $30 $44 $56 $56
2016 $1,700 39 $44 $35 $35 $52 $65
Average 2010-2016 $34 $37 $43 $51 $55

For the two year lag case, the average amount of the credit claimed per FDA approval of an orphan drug designation was from $43 million. For the three year lag case, the average credit per approval was $51 million.

This would make the outlays prior to the credit, equal to twice that amount, or $86 to $102 million spent on qualifying clinical trials, including the costs of failures, for each FDA approval of an orphan indication.

This is a large number, but only a fraction of the $965 million used by Joseph DiMasi in his widely quoted 2016 paper, for the cost of Phase 1-3 trials, adjusted for the risks of failures. (J.A. DiMasi et al. / Journal of Health Economics 47 (2016) 20–33).
Since both figures include the risks of failures, it is worth reflecting on the large differences.

One factor was noted above — some of the orphan drug trial outlays did not qualify for the credit.

Secondly, the orphan approvals are for designations, including often for new uses of existing drugs.

Third, and I think more consequential, the enrollment in trials for orphan approvals were often very small, and certainly small compared to non-orphans. To appreciate the differences in trial sizes, consider data from novel drugs approved by the FDA from 2015 to October 31, 2017.

Since 2015, the FDA has published a Drug Trials Snapshot for each novel drug. The snapshot focuses on the trials the FDA relied upon to evaluate the safety and efficacy of new drug approvals. It is not necessarily complete, and for example, and often excludes data from phase 1 trials, but the methodology is the same for all drugs, and one can compare novel drugs that are first approved for an orphan indication to non-orphans.
Table 3 provides data on the enrollment of patients in trials cited in the FDA Drug Trials Snapshot, for all novel drugs approval by the FDA from 2015 to October 31, 2017.

Table 3: Enrollment of patients in trials cited in the FDA Drug Trials Snapshot, for novel drugs approved by the FDA from 2015 to October 31, 2017

Number of drugs Total enrollment in trials cited in snapshot Average patents enrolled
Year* Non- Orphans Orphans Non- Orphans Orphans Non- Orphans Orphans Ratio, Non- Orphan to Orphan
2017 21 14 36,446 3,357 1,736 240 7.2
2016 13 8 29,493 1,853 2,269 232 9.8
2015 23 20 90,018 13,221 3,914 661 5.9
Totals 57 42 155,957 18,431 2,736 439 6.2

*Through October 31,2017

For the trials cited in the FDA Drug Trials Snapshot in this period, the average number of patients enrolled was 2,736 for the non-orphan drugs, and 439 for the orphan products.

The stark disparities in trial sizes could explain the much of difference between the DiMasi estimate and the Orphan Drug Tax Credit outlays. There are other differences as well, including the fact that DiMasi may be including the costs of trials for multiple indications on the same drug, and the data in Table 2 only concerns the costs of the credit per single approved indication.

In any case, the differences remain significant. Note that in 2013, the last year for which we have data from the IRS Statistics of Income, the total amount of the credit from all 132 corporate tax returns that claimed the credit was just over $1 billion, nearly the same amount as the DiMasi estimate of $965 million for a single drug. But in 2013, the FDA granted 265 orphan designations and approved 33 orphan indications, including 8 novel products approved as orphans for the lead indication.

The data from the orphan drug tax credit illustrates the large gap between the known facts about the costs for R&D for orphan drug development, and the astronomically larger R&D costs claimed by DiMasi (and frequently quoted by other researchers, policy makers and journalists) as averages that should guide policy making.

The data from the orphan drug tax credit also provide additional perspective to the estimates of drug development costs provided by Vinay Prasad and Sham Mailankody in their 2017 JAMA paper.  The Prasad and Mailankody paper reported data on R&D costs for 10 drugs, including 9 orphan products, and included not only clinical outlays, but also pre-clinical and capital costs in their estimates.  The lower estimates of drug development costs by Prasad and Mailankody surprised some observers who considered DiMasi’s higher estimates representative of the average drug development costs. Given the strong evidence that orphan drug trial costs are much lower than non-orphans, one should not be surprised at the lower Prasad/Mailankody estimates, and we also note that one shortcoming of the Prasad/Mailankody paper is that they ignore the subsidy that was provided by the orphan drug tax credit.

Finally, it is good to keep in mind how important orphan products are these days.  From 2010 to 2016, 75 percent of all novel cancer drugs were initially approved as orphan drugs.


Data on the tax credit and orphan approvals and designations since 1983 is available here.

Be Sociable, Share!

Leave a Reply

Your email address will not be published. Required fields are marked *