SlideShare a Scribd company logo
1 of 11
Download to read offline
MANAGERIAL AND DECISION ECONOMICS
Manage. Decis. Econ. 28: 469–479 (2007)
Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/mde.1360

        The Cost of Biopharmaceutical R&D:
               Is Biotech Different?
                            Joseph A. DiMasia,* and Henry G. Grabowskib
                      a
                          Tufts Center for the Study of Drug Development, Tufts University, USA
                                   b
                                     Department of Economics, Duke University, USA




             The costs of developing the types of new drugs that have been pursued by traditional
             pharmaceutical firms have been estimated in a number of studies. However, similar analyses
             have not been published on the costs of developing the types of molecules on which biotech
             firms have focused. This study represents a first attempt to get a sense for the magnitude of the
             R&D costs associated with the discovery and development of new therapeutic biopharmaceu-
             ticals (specifically, recombinant proteins and monoclonal antibodies [mAbs]).
                We utilize drug-specific data on cash outlays, development times, and success in obtaining
             regulatory marketing approval to estimate the average pre-tax R&D resource cost for
             biopharmaceuticals up to the point of initial US marketing approval (in year 2005 dollars).
             We found average out-of-pocket (cash outlay) cost estimates per approved biopharmaceutical
             of $198 million, $361 million, and $559 million for the preclinical period, the clinical period,
             and in total, respectively. Including the time costs associated with biopharmaceutical R&D, we
             found average capitalized cost estimates per approved biopharmaceutical of $615 million,
             $626 million, and $1241 million for the preclinical period, the clinical period, and in total,
             respectively. Adjusting previously published estimates of R&D costs for traditional
             pharmaceutical firms by using past growth rates for pharmaceutical company costs to
             correspond to the more recent period to which our biopharmaceutical data apply, we found
             that total out-of-pocket cost per approved biopharmaceutical was somewhat lower than for the
             pharmaceutical company data ($559 million vs $672 million). However, estimated total
             capitalized cost per approved new molecule was nearly the same for biopharmaceuticals as for
             the adjusted pharmaceutical company data ($1241 million versus $1318 million). The results
             should be viewed with some caution for now given a limited number of biopharmaceutical
             molecules with data on cash outlays, different therapeutic class distributions for
             biopharmaceuticals and for pharmaceutical company drugs, and uncertainty about whether
             recent growth rates in pharmaceutical company costs are different from immediate past
             growth rates. Copyright # 2007 John Wiley & Sons, Ltd.




                  INTRODUCTION                                examine approaches that could reduce those costs.
                                                              If the productivity of new drug development can
The financial viability of new drug and biophar-               be improved, then more innovations may be
maceutical development depends on the expected                pursued and eventually reach the patient. The
costs of, as well as the returns to, R&D. When                Food and Drug Administration (FDA), through
R&D costs are substantial it is important to                  its Critical Path Initiative, has initiated a process
                                                              to, in part, explore how the agency, industry, and
*Correspondence to: Tufts Center for the Study of Drug
Development, Tufts University, 192 South Street, Suite 550,   academia can establish methods that would lower
Boston, MA 02111, USA. E-mail: joseph.dimasi@tufts.edu        development costs (FDA, 2004).

Copyright # 2007 John Wiley & Sons, Ltd.
470                                 J.A. DiMASI AND H.G. GRABOWSKI


   R&D costs for new drugs (including the costs of     project-level and aggregate annual expenditure
failures and time costs) have been estimated to        data for a consulting project for a biotech firm.2
average in excess of $800 million (in year 2000        We combined data by period and type of
dollars) for development that led to approvals in      compound from these two sources. We focus on
the 1990s, with a marked upward trend relative to      therapeutic recombinant proteins and monoclonal
earlier decades (DiMasi et al., 2003). These R&D       antibodies (mAbs), which are overwhelmingly the
cost estimates have used data on new drugs             two most prevalent compound types in the biotech
developed by traditional pharmaceutical firms           sector. The consulting project focused on compounds
(primarily new chemical entities). No study to         that first entered clinical testing from 1990 to 2003.
date has focused on the types of molecules that are    With compound type and period of initial clinical
developed by biotech firms. One might conjecture        testing as study criteria, we utilized data on four
that biopharmaceuticals are less costly to develop     biologics from three companies used in the earlier
because biotech firms need to be more nimble and        study and 13 compounds from the biotech firm.3
creative or that fewer safety issues arise for many       While the data on cash outlays are limited to the
biopharmaceuticals because they replace sub-           17 compounds noted above, we are able to use a
stances that exist naturally in the body. However,     much larger data set to estimate average develop-
some industry insiders estimate that costs, even for   ment times, clinical success rates, and phase
biotech firms, exceed $1 billion.1                      transition probabilities. These data are used to
   In this paper, we make a first attempt to            account for time costs and the costs of develop-
examine the magnitude of R&D costs associated          ment failures.4 We used a Tufts Center for the
with developing the types of molecules on which        Study of Drug Development (CSDD) database of
biotech firms focus. Specifically, we use drug-          biopharmaceutical compounds. The Tufts CSDD
specific cost, development time, and clinical           database is constructed from information con-
success rate data for therapeutic biopharmaceu-        tained in a number of commercial business
ticals to estimate pre-tax R&D resource costs. We      intelligence databases (PharmaProjects, R&D
then compare these results to those obtained for       Focus, and iDdb3), trade press accounts, company
development of new drugs by traditional pharma-        reports and websites, and company surveys. For
ceutical firms (DiMasi et al., 2003). Given that the    our analyses of development times, clinical success
biopharmaceutical data are, on average, more           rates, and phase transition probabilities, we used a
recent than the data used for DiMasi et al. (2003),    subset of this database. The compounds included
we estimate the difference in study periods. Our        are therapeutic recombinant proteins and mAbs that
results for biopharmaceutical development are          were first tested in humans from 1990 through 2003.
then also compared to those for traditional            There are 522 such compounds, and they include
pharmaceutical firms with costs extrapolated using      molecules that were abandoned during develop-
estimated past growth rates for pharma costs to        ment, as well as those that have attained US Food
coincide with the more recent biopharmaceutical        and Drug Administration (FDA) approval.5
study period.                                             We compare our results for biopharmaceuticals
   The rest of this paper is organized as follows.     to our previously developed estimates of R&D
The next section contains a description of the data    costs for new drugs developed by traditional
used for our analyses. Then, we describe the           pharmaceutical firms. The data underlying the
methods used to obtain our results. We further         ‘pharma’ results are described in DiMasi et al.
present our results. Finally, we summarize our         (2003). These data included cash outlays for 68
conclusions and offer some discussion of the results.   new drugs and development times, clinical ap-
                                                       proval success rates, and transition probabilities
                                                       for a larger data set of 534 new drugs.
                      DATA

Our data on project costs derive from two sources.                        METHODS
First, the sample for our study of pharmaceutical
R&D costs (DiMasi et al., 2003) contained a small      The methodology used for the analysis here is
number of biologic compounds developed by              explained in detail in DiMasi et al. (2003). We
pharmaceutical firms. Second, we obtained               shall only briefly outline the methods here.

Copyright # 2007 John Wiley & Sons, Ltd.                       Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                     DOI: 10.1002/mde
THE COST OF BIOPHARMACEUTICAL R&D                                        471

Out-of-Pocket Costs: Phase Means, Success Rates,         in the sample that have proceeded from one phase
              and Expected Costs                         to another among the set of drugs that entered
                                                         the first phase and either proceeded to the next
We refer to actual cash outlays of the firm as out-       phase or were terminated in the first phase.
of-pocket costs. We converted the data on clinical       This approach should provide reasonable esti-
period expenditures by phase and year to 2005            mates of phase transition probabilities since the
dollars using the GDP Implicit Price Deflator. We         lengths of individual phases are short relative to
determined mean costs for these molecules for            total development times. The implicit assumption
phase I, phase II, and phase III. Long-term animal       needed for such an approach is that those drugs
testing costs incurred during clinical development,      that are still active at the time of analysis will
regulatory approval submission costs, and chem-          proceed to later phases more or less in accordance
istry, manufacturing and control costs related to        with the estimated transition probabilities. The
development and incurred during clinical develop-        overall clinical success rate is then determined as
ment are subsumed in the cost estimates for the          the product of the phase transition probabilities.
clinical phases. The expenditures considered in this     Clinical success is defined as US regulatory
report for the sample of 17 molecules are only           approval for marketing.
those that were incurred prior to original market-          Expected out-of-pocket cost per investigational
ing approval.                                            drug is the weighted average of mean phase costs,
   To obtain a full R&D cost estimate that would         where the weights are the estimated probabilities
account for the costs of failures and the time cost      that an investigational molecule will enter a given
of new pharmaceutical development, we must               phase. Finally, the out-of-pocket cost per ap-
build up to one through analyses of the expected         proved new molecule is obtained by dividing the
costs for the clinical and preclinical periods. For      out-of-pocket cost per investigational molecule by
purposes of this study, by the clinical period we        the estimated clinical approval success rate.
mean the time from initial human testing of a               Preclinical costs are obtained in a manner
compound to original marketing approval. The             similar to the method we used in DiMasi et al.
preclinical period refers to activities engaged in       (2003). We examined time series data on aggregate
prior to the start of human testing. Thus pre-           preclinical and clinical expenditures for new
clinical R&D costs include expenditures for both         molecules at the firm level. These data, along with
basic research and preclinical development.              our estimated clinical period costs per investiga-
   Expected costs take into account the fact that        tional and per approved molecule, were used to
not all compounds will progress all the way              infer the corresponding values for preclinical costs.
through development to approval. We first work            The time series data on preclinical and clinical
at the investigational molecule level. For the           expenditures were linked, as was done in DiMasi
clinical period this means that we must estimate         et al. (2003), via an estimated 5-year lag between
the probabilities that a compound that enters the        the middle of the preclinical period and the middle
clinical testing pipeline will make it to each phase.    of the clinical period.7
These values can be estimated from the data in the
Tufts CSDD database of biopharmaceutical com-
                                                         Capitalized Costs: Development Times and Discount
pounds. Statistical inference using, in part, survi-
                                                                                Rate
val analysis to account for censoring of the data
has been implemented in a number of studies of           Drug development is a very lengthy process. As
drug industry success rates.6 However, given             such, there are substantial time costs to investing
lengthy drug development times, such an approach         in R&D years before any potential returns can be
requires that there be a substantial period of time      earned. We capture the time costs of drug
between when the most recent drug enters clinical        development in a single monetary measure by
testing and when the analysis is conducted. Since        capitalizing costs forward to the point of original
we must include here drugs that have entered the         marketing approval at an appropriate discount
clinical testing pipeline relatively recently, we have   rate. The discount rate used is a cost of capital
estimated success and phase attrition rates in a         estimate for a sample of firms obtained from
more mechanistic manner. We estimated a phase            applying the Capital Asset Pricing Model
transition probability to be the percentage of drugs     (CAPM). More detail on this process is explained

Copyright # 2007 John Wiley & Sons, Ltd.                         Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                       DOI: 10.1002/mde
472                                          J.A. DiMASI AND H.G. GRABOWSKI


below in the context of a discussion of the result                 phase transition probabilities shown in Figure 1.
we obtained for a biotech discount rate.                           For comparative purpose, we also reproduce the
   Capitalized costs are the sum of out-of-pocket                  phase transition probabilities for the DiMasi et al.
and time costs. To obtain time costs we not only                   (2003) study. Multiplying the phase transition
need an appropriate discount rate, but also a                      probability estimates for biopharmaceuticals
timeline over which out-of-pocket costs are                        yields an overall clinical approval success rate of
capitalized forward to marketing approval at the                   30.2% (as opposed to 21.5% for pharma). To
discount rate. Thus, we estimate average clinical                  obtain an estimate of the expected clinical period
phase and regulatory review lengths from the data                  cost per investigational molecule we need esti-
in our subset of therapeutic recombinant proteins                  mated probabilities that a molecule that enters
and mAbs. As noted above, we use the estimate in                   clinical testing will reach a given phase. Those
DiMasi et al. (2003) for the time from discovery to                values can be derived from the transition prob-
first human testing.                                                abilities and the overall clinical approval success
                                                                   rate. To be conservative, we assume a 100%
                                                                   success rate for regulatory approval submissions
                            RESULTS                                to the FDA so that the probability that a
                                                                   regulatory submission will be made is assumed to
Our focus is on biopharmaceutical development, but                 be the same as the overall clinical approval success
we will also make some comparisons to estimated                    rate. Previous studies have shown 100% success
costs for traditional pharmaceutical firms.                         rates for regulatory submissions for biophar-
                                                                   maceuticals for almost every period analyzed
                                                                   (Reichert and Paquette, 2003; Reichert, 2005).
Clinical Phase Costs per Investigational Molecule
                                                                   Altering this value within reason does not have
Table 1 shows our estimated average clinical                       an appreciable effect on the results. Applying
period phase costs for the sample of compounds                     the probabilities as weights for the mean costs
for which we obtained detailed data. Mean clinical                 yields an estimated out-of-pocket cost per
phase costs are higher than those that we had                      investigational molecule of $169 million for
obtained in our R&D cost study for traditional                     biopharmaceuticals.
pharmaceutical firms when adjusted for inflation.
For the period we analyzed, the sum of the clinical                Out-of-Pocket Clinical Cost per Approved Molecule.
period mean phase costs for biopharmaceuticals                     What we are mainly interested in are costs per
($166 million) is 14% higher than what we had                      approved new molecule. We obtain such values by
found for pharma development ($146 million).8                      dividing costs per investigational molecule by the
                                                                   estimated clinical approval success rate (30.2%).
Success Rate and Phase Transition Probabilities.                   This yields an estimate of the out-of-pocket
Using information from the Tufts CSDD bio-                         clinical period cost per approved new molecule
pharmaceutical database, we estimated the                          of $361 million for biopharmceuticals.


                                                                   Out-of-Pocket Preclinical Cost per Investigational
Table 1.       Out-of-pocket Preclinical and Clinical                                 Molecule
               Period Cost per Investigational Biophar-
               maceutical Compound (In Millions of                 Preclinical cost per investigational molecule is
               2005 dollars)a                                      obtained by multiplying our estimated clinical
Testing             Mean            Probability        Expected    phase cost per investigational molecule by a ratio
phase               cost ($)        of entering        cost ($)    of preclinical to clinical expenditures obtained by
                                    phase (%)
                                                                   applying the lag noted above to the aggregate
Preclinical         59.88           100                 59.88      expenditure time series data. The aggregate
Phase I             32.28           100                 32.28
Phase II            37.69            83.7               31.55
                                                                   data, with a lag imposed, implies that clinical
Phase III           96.09            47.1               45.26      period phase costs should account for 65% of total
                                                                   out-of-pocket cost. These estimates yield an
Total                                                  168.97
                                                                   out-of-pocket preclinical cost per investigational
a
    All costs were deflated using the GDP Implicit Price Deflator.   molecule of $59.9 million, and, using a 30.2%

Copyright # 2007 John Wiley & Sons, Ltd.                                  Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                                DOI: 10.1002/mde
THE COST OF BIOPHARMACEUTICAL R&D                                               473

clinical approval success rate, a preclinical out                                    from DiMasi et al. (2003). Total clinical plus
of-pocket cost per approved new molecule of                                          approval time is 8% longer for the biopharma-
$198 million.                                                                        ceuticals, with nearly all of the difference accounted
                                                                                     for by phase I.

                Capitalized Costs                                                    Cost of Capital. In our prior analysis of tradi-
                                                                                     tional pharmaceutical firms, we utilized a cost of
As noted above, to obtain estimates that include
                                                                                     capital of 11% as a discount rate for R&D
the time costs of new drug development we need to
                                                                                     activities that were first taken into clinical trials
estimate development times and choose an appro-
                                                                                     between 1983 and 1994 (DiMasi et al., 2003;
priate discount rate.
                                                                                     Grabowski et al., 2002). This cost of capital
                                                                                     estimate was based on concepts from modern
Development Times. Our data on biopharmaceu-                                         finance theory.
tical compound development histories for the                                            Utilizing the CAPM framework (Brealey and
period analyzed yielded the mean clinical devel-                                     Myers, 2000), the firm’s cost of capital, rn ; is a
opment and approval phase lengths shown in                                           weighted average of its cost of capital on its debt
Figure 2. The phase results are averages across all                                  and equity capital.9 Given the low debt values of
compounds that completed the phase, regardless                                       large pharmaceutical firms, the equity cost of
of whether the compound was ultimately approved                                      capital becomes the key factor driving the
for marketing. For comparative purposes we                                           weighted cost of capital for the firms. In the case
also show the pharma development time results                                        of biotech firms the debt component is negligible,


                                               83.7%
                     Transition Probability




                                                           71.0%                                                   68.5%
                                                                                                       64.2%
                                                                             56.3%
                                                                                       44.2%




                                                    Phase I-II                Phase II-III         Phase III-Approval

                                                                           Biotech   Pharma

                                                Figure 1. Transition probabilities for clinical phases.




                   Biotech                          19.5            29.3               32.9            16      97.7




                    Pharma                      12.3         26.0               33.8            18.2        90.3



                                              0.0                                                                      120.0
                                                                                 Months

                                                                   Phase I    Phase II    Phase III    RR

                                                Figure 2. Clinical development and approval times.

Copyright # 2007 John Wiley & Sons, Ltd.                                                       Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                                                     DOI: 10.1002/mde
474                                 J.A. DiMASI AND H.G. GRABOWSKI


given that long-term debt after 1990 is less than       financial economists, we estimated cost of capital
1% of market valuation. Thus, for all practical         values for a sample of biotech firms at roughly five
purposes the equity cost of capital for biotech         year intervals from their 1989 estimate. The lower
firms is the same as their weighted cost of capital.     value in 2004 reflects declining value in the risk
   In the CAPM framework, investors require a           free rate and the equity premium in recent years
risk premium for holding equity in a particular         compared to the 1994–1999 period. The focus of
company. This premium is based on the relative          our analysis is on R&D projects initiated since the
riskiness to investors of that company’s assets. The    mid 1990s through the early 2000s where a 10% to
formal measure of relative riskiness is the beta        12.5% rate was observed. We therefore use the
coefficient, or the firm’s contribution to the             average of the three values for the cost of capital in
variance in the returns from a diversified portfolio     Table 2, 11.5%, as the benchmark value for biophar-
of equity shares. The CAPM assumes that                 maceuticals. We also perform simulations around this
investors hold well-diversified portfolios.              baseline value to analyze the sensitivity of the
   The CAPM implies that the expected return on         capitalized R&D cost to this cost of capital value.10
a firm’s assets (the equity cost of capital) is equal       Discussions with a few of the leading pharma-
to the risk-free rate plus a risk premium which is      ceutical firms suggest that a nominal cost of capital
positively related to the riskiness of the firm’s        in the range of 12–15% was being utilized by many
assets relative to other stock market assets            large pharma firms in 2001–2002 (Grabowski,
                                                        et al., 2002). Given a 3% rate of inflation, this
rE ¼ rf þ betaðrm À rf Þ:
                                                        would imply a 10–12% real cost of capital for
In this equation rf is the risk-free rate (the return   major pharmaceutical firms. This is roughly
in treasury bonds minus a horizon premium is            consistent with estimates of the cost of capital
typically used as a proxy for the risk-free rate);      derived from the CAPM in this period.
rm is the long-term rate of return for a market
basket of common stock (usually the S&P index);
                                                        Capitalized Costs per Investigational Molecule. We
ðrm À rf Þ is the equity premium, and beta is a
                                                        obtain capitalized costs by spreading our esti-
measure of the relative riskiness of a specific firm
                                                        mated expected out-of-pocket phase costs per
(based on a regression analysis that yields the
                                                        investigational molecule over estimated mean
covariance of returns with the overall S&P index).
                                                        phase lengths and then capitalizing them forward
   Under CAPM, a firm with a beta of one would
                                                        to marketing approval at an 11.5% discount rate
have the same riskiness as the overall S&P index,
                                                        using a representative time profile. The results are
whereas those with values greater than one are
                                                        shown in Table 3.
more risky, and correspondingly, those with betas
                                                           Preclinical capitalized cost per investigational
below one are less risky. Company specific values
                                                        molecule is obtained by spreading the out-of-
for beta can be found in Value Line’s Investment
                                                        pocket cost per investigational molecule deter-
Surveys and other security analyst publications.
                                                        mined above ($60 million) over an estimated
Betas in these sources are typically updated on a
                                                        preclinical period (52.0 months) and then capita-
periodic basis.
                                                        lizing forward to marketing approval at an 11.5%
   Myers and Shyam-Sunder (1995) examined the
                                                        discount rate over the representative time profile.
cost of capital for seven smaller biotechnology and
                                                        Doing so yields a capitalized preclinical period
specialty pharmaceutical firms for 1989. These
                                                        cost per investigational molecule of approxima-
firms had higher betas and costs of capital than the
                                                        tely $186 million. Capitalized clinical cost per
major pharmaceutical firms. The greater betas or
                                                        investigational molecule is obtained by capitalizing
riskiness exhibited by these firms were consistent
with the fact that the smaller biotech firms had
fewer commercialized products and proportio-
                                                        Table 2.   Nominal and Real Cost of Capital
nately more earlier-stage R&D projects. The
                                                                   (COC), 1994–2004
average cost of capital for the full sample of seven
biotech and specialty pharma firms was 19% in                                     1994        2000         2004
nominal terms and 14% in real terms.                    Nominal COC (%)          17.0        15.0         13.0
   Using the same methodology as employed               Inflation rate (%)         4.5         3.0          3.0
                                                        Real COC (%)             12.5        12.0         10.0
by Myers and Shyam-Sunder (1995) and other

Copyright # 2007 John Wiley & Sons, Ltd.                        Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                      DOI: 10.1002/mde
THE COST OF BIOPHARMACEUTICAL R&D                                                   475


Table 3.      Capitalized Preclinical and Clinical Period costs per Investigational Biopharmaceutical
              Compound (In Millions of 2005 dollars)a
Testing phase      Expected out-of-       Phase length      Monthly        Start of phase     End of                Expected
                   pocket cost ($)        (mos.)            cost ($)       to approval        phase to              capitalized
                                                                           (mos.)             approval (mos.)       costb ($)
Preclinical        59.88                  52.0              1.15           149.7              97.7                  185.62
Phase I            32.28                  19.5              1.66            97.7              78.2                   71.78
Phase II           31.55                  29.3              1.08            78.2              48.9                   56.32
Phase III          45.26                  32.9              1.38            48.9              16.0                   60.98

Total                                                                                                               374.70
a
 All costs were deflated using the GDP Implicit Price Deflator.
b
 Expenditures capitalized forward to the point of marketing approval for a representative time profile at an 11.5% real discount
rate. The estimated length of the approval phase is 16.0 months.

out-of-pocket clinical phase cost forward to                       14% higher for the clinical period, and 24% higher
marketing approval according to the time profile                    in total.
in Figure 2. This yields a capitalized clinical period                It may be the case, however, that the appro-
cost per investigational molecule of approximately                 priate figures for R&D costs for traditional
$189 million.                                                      pharmaceutical firms to compare with our biotech
                                                                   estimates should be much higher than those shown
                                                                   by the middle bars of Figure 4. The reason is that
        Total R&D Costs per Approved Molecule                      the biotech data are somewhat more recent than
                                                                   the data used for DiMasi et al. (2003). We
To get estimates of fully allocated total cost per
                                                                   conducted two types of comparisons to judge the
approved new molecule, we need only add
                                                                   extent to which the period is shifted. Examining
estimates of cost per approved molecule for the
                                                                   both actual approval dates for biotech compounds
preclinical and clinical periods. Applying the
                                                                   in the Tufts CSDD database and for those used in
clinical approval success rate of 30.2% for
                                                                   the DiMasi et al. (2003) sample, as well as average
biopharmaceuticals to the capitalized preclinical                  approval dates on which phase I testing began for
cost per investigational molecule noted above                      biopharmaceutical compounds and for the data in
yields a preclinical period cost per approved new
                                                                   DiMasi et al. (2003), suggested a shift of approxi-
molecule of $615 million. Similarly, applying the
                                                                   mately five years in the study periods. Thus, we
success rate to our estimate of capitalized clinical
                                                                   should consider what new drug development costs
period cost per investigational molecule yields a
                                                                   for pharma firms would be five more years into the
capitalized clinical period cost per approved
                                                                   future. In DiMasi et al. (2003) we compared costs
molecule of $626 million. Total capitalized cost
                                                                   for the current sample to those for an earlier
per approved molecule for biopharmaceuticals is                    period covered by a previous study (with more
then $1241 million. Out-of-pocket, time, and                       than a decade difference in time). We applied the
capitalized costs per approved new molecule are
                                                                   growth rates (over and above inflation) for the
shown in Figure 3.
                                                                   preclinical and clinical periods that we observed
                                                                   between our two earlier studies on pharma costs to
R&D Cost Comparisons: Biotech and Pharma. Our                      the most recent pharma data assuming a further
estimates for biopharmaceutical out-of-pocket                      five-year shift. The results are the pharma time-
preclinical, clinical, and total out-of-pocket                     adjusted values given by the third set of bars in
R&D costs are shown in Figure 4. For compara-                      Figure 4. The unadjusted figures can be viewed as
tive purposes, we also show the corresponding                      what the outcomes for pharmaceutical firms would
figures for pharma from our most recent study                       be if they had kept cost increases in the later five-
of R&D costs for traditional pharmaceutical                        year period in line with general inflation.
firms (DiMasi et al., 2003). The overall figures                        The time-adjusted out-of-pocket biotech costs
for pharma firms are significantly lower than                        for the preclinical period are still somewhat higher
those for biotech development. Biopharmaceutical                   than for pharma even with the period adjustment
costs are 46% higher for the preclinical period,                   (32% higher). However, for the clinical period and

Copyright # 2007 John Wiley & Sons, Ltd.                                  Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                                DOI: 10.1002/mde
476                                                                 J.A. DiMASI AND H.G. GRABOWSKI


                                                                                                                                            1,241




                       Millions (2004$)
                                                                                                                                    682
                                                                       615                                 626
                                                                                                                          559
                                                              417
                                                                                       361
                                                                                                 265
                                                    198



                                                       Preclinical**                          Clinical                            Total

                                                                              Out-of-pocket        Time      Capitalized

                                           * Based on a 30.2% clinical approval success rate
                                          ** All R&D costs (basic research and preclinical development) prior to initiation of clinical testing

      Figure 3. Pre-approval out-of-pocket (cash outlay) and time costs per approved new biopharmaceutical.Ã




                                                                                                                                             672

                                                                                                          522            559
                             Millions (2005$)




                                                                                                                                   452

                                                                                       361       316

                                                    198
                                                              136       150




                                                        Preclinical*                            Clinical                          Total

                                                                       Biotech       Pharma       Pharma (time-adjusted)**

                                                  * All R&D costs (basic research and preclinical development) prior to initiation of clinical testing
                                                ** Based on a 5-year shift and prior growth rates for the preclinical and clinical periods

               Figure 4. Pre-approval cash outlays (out-of-pocket cost) per approved new molecule.


in total, biopharmaceutical out-of-pocket costs are                                                clinical period and total costs are proportionately
lower than our reported pharma costs adjusted for                                                  closer to pharma costs than are out-of-pocket
a later period. Specifically, clinical period costs are                                             costs. Capitalized clinical period costs for bio-
31% lower and total costs are 17% lower for                                                        pharmaceuticals are 29% lower than for time-
biopharmaceuticals. Of course, we do not know if                                                   adjusted pharma costs. However, total capitalized
pharma costs continued to increase at the same                                                     cost per approved biopharmaceutical ($1241 mil-
rates as they had in the past.                                                                     lion) is only 6% lower than total capitalized time-
   Our main results for capitalized costs are shown                                                adjusted pharma cost ($1318 million).
in Figure 5. Capitalization increases biopharma-
ceutical costs relative to pharma costs because of a
longer development timeline and a higher cost of                                                                                CONCLUSIONS
capital.11 As a result, the capitalized preclinical
costs for biotech are proportionately higher                                                       While estimates of the level of, and trends in, R&D
(40%) relative to time-adjusted pharma costs                                                       costs for traditional pharmaceutical firms have
than are out-of-pocket costs. However, capitalized                                                 been published, to date no studies have focused

Copyright # 2007 John Wiley & Sons, Ltd.                                                                         Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                                                                       DOI: 10.1002/mde
THE COST OF BIOPHARMACEUTICAL R&D                                                       477

                                                                                                                         1,318
                                                                                                       1,241




                     Millions (2005$)
                                                                                        879                      899


                                        615                         626
                                                                              523
                                                     439
                                              376




                                          Preclinical*                       Clinical                            Total

                                                     Biotech      Pharma        Pharma (time-adjusted)**

                           * All R&D costs (basic research and preclinical development) prior to initiation of clinical testing
                         ** Based on a 5-year shift and prior growth rates for the preclinical and clinical periods

                              Figure 5. Pre-approval capitalized cost per approved new molecule.

specifically on biotech firms or particular types of                               new molecule appears to be essentially the same as
biopharmaceutical development. We have taken a                                   estimated total capitalized per approved new drug
first step toward getting a sense for the magnitude                               for traditional pharmaceutical firms. However,
of what the full R&D resource costs associated                                   total out-of-pocket costs for biopharmaceuticals
with discovering and developing biopharmaceuti-                                  were found to be somewhat lower, both out-of-
cals to the point of initial regulatory marke-                                   pocket and capitalized clinical period costs for
ting approval had been for recent years.                                         biopharmaceuticals were lower, and preclinical
Using compound-specific costs for a sample of 17                                  period costs for biopharmaceuticals were some-
investigational biopharmaceuticals from four                                     what higher.12 Determining what the actual
firms, a time series of annual preclinical and                                    growth rates in costs for pharma firms had been
clinical expenditures for a biotech firm, estimated                               in recent years awaits further study.
average development times and phase transition                                      Several caveats to our results should be men-
probabilities for over 500 therapeutic recombinant                               tioned. The results are preliminary in that the
proteins and mAbs, we estimated average pre-                                     sample size for mean phase costs is relatively small,
clinical period, clinical period, and total costs per                            although the sample sizes for development times
approved new biopharmaceutical. We found out                                     and success rates are quite large. Beyond this, the
of-pocket (cash outlay) cost estimates of $198                                   comparisons with pharma costs should be viewed
million, $361 million, and $559 million per                                      with some caution for two reasons. First, as noted,
approved new biopharmaceutical for the preclini-                                 pharma costs may not have changed to the same
cal period, the clinical period, and in total,                                   degree in recent years as they did in the past.
respectively (in year 2005 dollars). These figures                                Second, costs can vary by therapeutic class
include the costs of compound failures. Adding                                   (DiMasi et al., 2004). The distributions of inves-
time costs to cash outlays, we found cost estimates                              tigational compounds by therapeutic class for
of $615 million, $626 million, and $1241 million                                 traditional pharmaceutical firms do differ from
per approved new biopharmaceutical for the                                       the distributions by class for the recombinant
preclinical period, the clinical period, and in total,                           protein and mAb biopharmaceuticals that we
respectively (in year 2005 dollars).                                             examined. Specifically, investigational biopharma-
   Our estimates for biopharmaceuticals are higher                               ceutical molecules were more concentrated in the
than those we found for our previous study of                                    oncology and immunologic categories than were
pharma costs (DiMasi et al., 2003). However, the                                 pharma molecules for the period analyzed in DiMasi
biopharmaceutical data that we used is of a more                                 et al. (2003), while the pharma distribution was more
recent vintage. If past growth rates in R&D costs                                concentrated in the cardiovascular and neuropharma-
for traditional pharmaceutical firms are applied to                               cologic classes. It is unclear how these differences
the results in DiMasi et al. (2003), then total                                  affect the comparative results; while full clinical period
capitalized biopharmaceutical cost per approved                                  costs for new cardiovascular and neuropharmacologic

Copyright # 2007 John Wiley & Sons, Ltd.                                                      Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                                                    DOI: 10.1002/mde
478                                       J.A. DiMASI AND H.G. GRABOWSKI


drugs were found in DiMasi et al. (2004) to be about             9. The weighted average company cost of capital can
average for pharma development, not enough infor-                   be expressed in terms of the following equation:
mation was available to determine costs for oncology                rn ¼ rD ð1 À TC ÞðD=VÞ þ rE ðE=VÞ;
and immunologic drugs. Additional research is needed
                                                                    where rD and rE are the expected rates of return on
to fully resolve these issues.                                      assets of comparable riskiness for the firm’s debt and
                                                                    equity securities, respectively. TC is the firm’s corpo-
                                                                    rate tax rate, and D=V and E=V are the proportion of
                         NOTES                                      the firm’s market valuation represented by debt and
                                                                    equity securities, respectively. The debt component of
 1. Gottschalk (2004) notes that a manager at a biotech             the cost of capital is multiplied by ð1 À TC Þ; because
    company estimated that his company spends in                    interest on debt obligations is tax deductible, while
    excess of one billion dollars to get a drug to market           earnings on equity shares are not.
    (lecture to Professor Fiona Murray’s MIT Sloan              10. Financial economists suggest that the risk and cost
    Management class 15.968, ‘Building a Biomedical                 of capital of an individual R&D project will depend
    Business,’ by Bill Anderson, VP Business Planning,              on the stage of the project and, correspondingly, on
    Biogen Idec Inc., 3 December 2003).                             the amount and timing of follow-on investments
 2. The firm provided data on its R&D expenditures in                required to achieve commercial success. By contrast,
    the form required to apply the basic methodology                the estimates derived from corporate financial data
    used in DiMasi et al. (2003). The purpose was to test           by Myers and Shyam-Sunder (1995) and other
    their hypothesis that their R&D costs were in fact              financial economists represent an average cost of
    significantly lower than the estimate in DiMasi et al.           capital for a firm’s aggregate portfolio of R&D
    (2003) for traditional pharmaceutical firms.                     projects as well as their complementary capital
 3. The sample consisted of nine recombinant proteins               investments in manufacturing and marketing assets.
    and eight mAbs.                                                 Some analyses of the pharmaceutical industry have
 4. Given that we use development times and success rates           utilized a higher cost of capital for earlier stage
    for what is essentially the universe of biopharmaceu-           R&D projects based on cost of capital estimates
    ticals developed by all firms, it is not possible to infer       from firms at different stages of the life cycle. For
    what costs per approved new molecule are for the                example, the Office of Technology Assessment (U.S.
    biotech firm that provided molecule-specific cash                 Congress, Office of Technology Assessment, 1993)
    outlays to us. Company-specific success rates, in                utilized a 14.5% real cost of capital for the earlier
    particular, can have a substantial impact on total              pre-clinical stages of pharma R&D based on the
    R&D costs for a given company. One might wonder,                Myers and Shyam-Sunder (1995) biotech and small
    however, about the internal consistency of all of the           firm sample, and lower values for later stages of the
    data. It is unlikely that company-specific mean clinical         life cycle. Myers and Howe (1997) generate a ‘stair-
    phase expenditures will have an appreciable effect on            stepped’ cost of capital using a Monte Carlo
    success rates. It is also likely that mean clinical phase       simulation model and an option value approach. To
    costs for an investigational molecule of a given type           our knowledge, none of the big pharma firms use a
    and therapeutic class will not vary much across firms.           stair-stepped cost of capital in their NPV and return
    One potential concern, though, is the possibility that          calculations, but some are considering this approach.
    there were some time–cost tradeoffs for the phase data       11. The discount rate has a modest effect on total
    (Scherer, 1966). This is more of a concern for                  capitalized costs. If we use a 10.5% discount rate for
    molecules that fail in testing than for those that              biotech, then its total capitalized cost falls by 6.8%.
    succeed, since the total amount of testing for                  If we use a 12.5% discount rate for biotech, then
    molecules that are eventually approved for marketing            total capitalized cost is 7.3% higher.
    is likely to be essentially the same, regardless of         12. The higher preclinical expenditures per approved
    whether some testing is done in parallel rather than            biopharmaceutical may, to some extent, help
    sequentially. We have no reason to believe that the             explain the higher clinical approval success rates
    biotech firm in question here differed from other firms            for biopharmaceuticals.
    with regard to time–cost tradeoffs.
 5. This data set consisted of 278 recombinant proteins
    and 244 mAbs.
 6. See, for example, DiMasi et al. (1991), DiMasi (1995),                          REFERENCES
    Gosse et al. (1996), DiMasi (2001), DiMasi et al. (2003).
 7. In the absence of precise data on the length of the         Brealy RA, Myers SC. 2000. Principles of Corporate
    preclinical period for these molecules, we used the           Finance (6th edn). Irwin/McGraw-Hill: Burr Ridge, IL.
    value estimated for DiMasi et al. (2003). Managers          DiMasi JA. 1995. Success rates for new drugs entering
    at the biotech firm from which we obtained cost                clinical testing in the United States. Clinical Pharma-
    data agreed that the estimate was reasonable.                 cology and Therapeutics 58(1): 1–14.
 8. See, however, our discussion below about differ-             DiMasi JA. 2001. Risks in new drug development:
    ences in time periods between our previous study              approval success rates for investigational drugs. Clinical
    and the data used for this report.                            Pharmacology and Therapeutics 69(5): 297–307.

Copyright # 2007 John Wiley & Sons, Ltd.                                 Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                               DOI: 10.1002/mde
THE COST OF BIOPHARMACEUTICAL R&D                                            479

DiMasi JA, Grabowski HG, Vernon J. 2004. R&D costs         Grabowski HG, Vernon JM, DiMasi JA. 2002. Returns
 and returns by therapeutic category. Drug Information       on research and development for 1990s new drug
 Journal 38(3): 211–223.                                     introductions. PharmacoEconomics 20(Supplement 3):
DiMasi JA, Hansen RW, Grabowski HG. 2003. The                11–29.
 price of innovation: new estimates of drug deve-          Reichert JM. 2005. Monoclonal antibody successes in
 lopment costs. Journal of Health Economics 22(3):           the clinic. Nature Biotechnology 23(9): 1073–1078.
 141–185.                                                  Reichert JM, Paquette C. 2003. Clinical development
DiMasi JA, Hansen RW, Grabowski HG, Lasagna L.               of therapeutic recombinant proteins. BioTechniques
 1991. Cost of innovation in the pharmaceutical              35(1): 1–6.
 industry. Journal of Health Economics 10(2): 107–142.     Myers SC, Howe CD. 1997. A life cycle financial model
Food and Drug Administration. 2004. Challenge and            of pharmaceutical R&D. MIT Program on the
 Opportunity on the Critical Path to New Medical             Pharmaceutical Industry, Massachusetts Institute of
 Products. Available at: http://www.fda.gov/oc/initia-       Technology, Cambridge, MA.
 tives/criticalpath/whitepaper.pdf [accessed 21 February   Myers SC, Shyam-Sunder L. 1995. Measuring pharma-
 2006].                                                      ceutical industry risk and the cost of capital. In
Gosse ME, DiMasi JA, Nelson TF. 1996. Recombinant            Competitive Strategies in the Pharmaceutical Industry,
 protein and therapeutic monoclonal antibody drug            Helms RB (ed.). American Enterprise Institute for
 development in the United States: 1980–1994. Clinical       Public Policy: Washington, DC.
 Pharmacology and Therapeutics 60(6): 608–618.             Scherer FM. 1966. Time–cost tradeoffs in uncertain
Gottschalk AHB. 2004. Improving the efficiency of the          empirical research projects. Naval Research Logistics
 later Stages of the Drug Development Process: Survey        Quarterly 13: 71–82.
 Results from the Industry, Academia, and the              United States Congress, Office of Technology Assess-
 FDA. Dissertation for Master of Science in Health           ment. 1993. Pharmaceutical R&D: Costs, Risks
 Sciences and Technology, Massachusetts Institute of         and Rewards. US Government Printing Office:
 Technology.                                                 Washington, DC.




Copyright # 2007 John Wiley & Sons, Ltd.                           Manage. Decis. Econ. 28: 469–479 (2007)
                                                                                         DOI: 10.1002/mde

More Related Content

What's hot

Challenges for drug development jsr slides aug 2013
Challenges for drug development jsr slides aug 2013Challenges for drug development jsr slides aug 2013
Challenges for drug development jsr slides aug 2013CincyTechUSA
 
The Price Of Innovation New Drug Development Cost 2003
The Price Of Innovation New Drug Development Cost 2003The Price Of Innovation New Drug Development Cost 2003
The Price Of Innovation New Drug Development Cost 2003Dmitry Tseitlin
 
The price of innovation new estimates
The price of innovation new estimates The price of innovation new estimates
The price of innovation new estimates Dmitry Tseitlin
 
Cost on development of new drug
Cost on development of new drugCost on development of new drug
Cost on development of new drugDrAsimraza
 
Case 5.1 - DESIGNING DRUGS VIRTUALLY
Case 5.1 - DESIGNING DRUGS VIRTUALLYCase 5.1 - DESIGNING DRUGS VIRTUALLY
Case 5.1 - DESIGNING DRUGS VIRTUALLYAya Wan Idris
 
Wk 5 case 1 designing drug virtually
Wk 5 case 1 designing drug virtually Wk 5 case 1 designing drug virtually
Wk 5 case 1 designing drug virtually dyadelm
 
RyMat Drug Development Training Module
RyMat Drug Development Training ModuleRyMat Drug Development Training Module
RyMat Drug Development Training ModuleStuart Silverman
 
In silico Drug Design: Prospective for Drug Lead Discovery
In silico Drug Design: Prospective for Drug Lead DiscoveryIn silico Drug Design: Prospective for Drug Lead Discovery
In silico Drug Design: Prospective for Drug Lead Discoveryinventionjournals
 
Nanopa4
Nanopa4Nanopa4
Nanopa4gana9
 
Role of bioinformatics and pharmacogenomics in drug discovery
Role of bioinformatics and pharmacogenomics in drug discoveryRole of bioinformatics and pharmacogenomics in drug discovery
Role of bioinformatics and pharmacogenomics in drug discoveryArindam Chakraborty
 
Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...
Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...
Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...mconghuyen
 
Overcoming obstacles to repurposing for neurodegenerative disease
Overcoming obstacles to repurposing for neurodegenerative diseaseOvercoming obstacles to repurposing for neurodegenerative disease
Overcoming obstacles to repurposing for neurodegenerative diseaseLona Vincent
 
Linked data-and-sp ls-fda-spl-jamboree-092014
Linked data-and-sp ls-fda-spl-jamboree-092014Linked data-and-sp ls-fda-spl-jamboree-092014
Linked data-and-sp ls-fda-spl-jamboree-092014Richard Boyce, PhD
 
Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014
Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014
Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014Richard Boyce, PhD
 
eye of the storm
eye of the stormeye of the storm
eye of the stormJan Malek
 
Accelerating Generic Approvals by Dr Anthony Crasto
Accelerating Generic Approvals by Dr Anthony CrastoAccelerating Generic Approvals by Dr Anthony Crasto
Accelerating Generic Approvals by Dr Anthony CrastoAnthony Melvin Crasto Ph.D
 
Drug Repositioning Workshop
Drug Repositioning WorkshopDrug Repositioning Workshop
Drug Repositioning WorkshopGenna Gerla
 
The role of artificial intelligence in drug repurposing
The role of artificial intelligence in drug repurposingThe role of artificial intelligence in drug repurposing
The role of artificial intelligence in drug repurposingAboul Ella Hassanien
 

What's hot (19)

Challenges for drug development jsr slides aug 2013
Challenges for drug development jsr slides aug 2013Challenges for drug development jsr slides aug 2013
Challenges for drug development jsr slides aug 2013
 
The Price Of Innovation New Drug Development Cost 2003
The Price Of Innovation New Drug Development Cost 2003The Price Of Innovation New Drug Development Cost 2003
The Price Of Innovation New Drug Development Cost 2003
 
The price of innovation new estimates
The price of innovation new estimates The price of innovation new estimates
The price of innovation new estimates
 
Cost on development of new drug
Cost on development of new drugCost on development of new drug
Cost on development of new drug
 
Case 5.1 - DESIGNING DRUGS VIRTUALLY
Case 5.1 - DESIGNING DRUGS VIRTUALLYCase 5.1 - DESIGNING DRUGS VIRTUALLY
Case 5.1 - DESIGNING DRUGS VIRTUALLY
 
Wk 5 case 1 designing drug virtually
Wk 5 case 1 designing drug virtually Wk 5 case 1 designing drug virtually
Wk 5 case 1 designing drug virtually
 
RyMat Drug Development Training Module
RyMat Drug Development Training ModuleRyMat Drug Development Training Module
RyMat Drug Development Training Module
 
Ibogaine Research
Ibogaine ResearchIbogaine Research
Ibogaine Research
 
In silico Drug Design: Prospective for Drug Lead Discovery
In silico Drug Design: Prospective for Drug Lead DiscoveryIn silico Drug Design: Prospective for Drug Lead Discovery
In silico Drug Design: Prospective for Drug Lead Discovery
 
Nanopa4
Nanopa4Nanopa4
Nanopa4
 
Role of bioinformatics and pharmacogenomics in drug discovery
Role of bioinformatics and pharmacogenomics in drug discoveryRole of bioinformatics and pharmacogenomics in drug discovery
Role of bioinformatics and pharmacogenomics in drug discovery
 
Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...
Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...
Re-Engineering Early Phase Cancer Drug Development: Decreasing the Time from ...
 
Overcoming obstacles to repurposing for neurodegenerative disease
Overcoming obstacles to repurposing for neurodegenerative diseaseOvercoming obstacles to repurposing for neurodegenerative disease
Overcoming obstacles to repurposing for neurodegenerative disease
 
Linked data-and-sp ls-fda-spl-jamboree-092014
Linked data-and-sp ls-fda-spl-jamboree-092014Linked data-and-sp ls-fda-spl-jamboree-092014
Linked data-and-sp ls-fda-spl-jamboree-092014
 
Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014
Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014
Addressing gaps-in-clinically-useful-evidence-on-dd is-nlm-training-2014
 
eye of the storm
eye of the stormeye of the storm
eye of the storm
 
Accelerating Generic Approvals by Dr Anthony Crasto
Accelerating Generic Approvals by Dr Anthony CrastoAccelerating Generic Approvals by Dr Anthony Crasto
Accelerating Generic Approvals by Dr Anthony Crasto
 
Drug Repositioning Workshop
Drug Repositioning WorkshopDrug Repositioning Workshop
Drug Repositioning Workshop
 
The role of artificial intelligence in drug repurposing
The role of artificial intelligence in drug repurposingThe role of artificial intelligence in drug repurposing
The role of artificial intelligence in drug repurposing
 

Viewers also liked

Sia itar-written-testimony-for-hfac-hearing
Sia itar-written-testimony-for-hfac-hearingSia itar-written-testimony-for-hfac-hearing
Sia itar-written-testimony-for-hfac-hearingDmitry Tseitlin
 
Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...
Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...
Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...Dmitry Tseitlin
 
National Security Implications of virtual currency examining the potential fo...
National Security Implications of virtual currency examining the potential fo...National Security Implications of virtual currency examining the potential fo...
National Security Implications of virtual currency examining the potential fo...Dmitry Tseitlin
 
Impact investing - profit meets purpose
Impact investing - profit meets purposeImpact investing - profit meets purpose
Impact investing - profit meets purposeDmitry Tseitlin
 
Bnef neo2015 executive-summary
Bnef neo2015 executive-summaryBnef neo2015 executive-summary
Bnef neo2015 executive-summaryDmitry Tseitlin
 
LauncherOne Virgin Galactic 2013
LauncherOne  Virgin Galactic  2013LauncherOne  Virgin Galactic  2013
LauncherOne Virgin Galactic 2013Dmitry Tseitlin
 
Solar Energy Industry Forecast DOE 2008
Solar Energy Industry Forecast DOE 2008Solar Energy Industry Forecast DOE 2008
Solar Energy Industry Forecast DOE 2008Dmitry Tseitlin
 
State of innovation Thomson Reuters 2016
State of innovation Thomson Reuters 2016 State of innovation Thomson Reuters 2016
State of innovation Thomson Reuters 2016 Dmitry Tseitlin
 
Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009
 Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009 Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009
Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009Dmitry Tseitlin
 
Презентация А.Б.Чубайса для французский инвесторов
Презентация А.Б.Чубайса для французский инвесторовПрезентация А.Б.Чубайса для французский инвесторов
Презентация А.Б.Чубайса для французский инвесторовDmitry Tseitlin
 
Россия 21 века образ желаемого завтра
Россия 21 века образ желаемого завтраРоссия 21 века образ желаемого завтра
Россия 21 века образ желаемого завтраDmitry Tseitlin
 
план мероприятий по стимулированию инновационной активности предприятий
план мероприятий по стимулированию инновационной активности предприятийплан мероприятий по стимулированию инновационной активности предприятий
план мероприятий по стимулированию инновационной активности предприятийDmitry Tseitlin
 
Cтратегия РВК на 2010 год
Cтратегия РВК на 2010 годCтратегия РВК на 2010 год
Cтратегия РВК на 2010 годDmitry Tseitlin
 
Годовой отчет Роснано за 2008
Годовой отчет Роснано за 2008Годовой отчет Роснано за 2008
Годовой отчет Роснано за 2008Dmitry Tseitlin
 
Презентация А.Миллера на комиссии по модернизации 25 декабря 2009
Презентация А.Миллера на комиссии по модернизации 25 декабря 2009Презентация А.Миллера на комиссии по модернизации 25 декабря 2009
Презентация А.Миллера на комиссии по модернизации 25 декабря 2009Dmitry Tseitlin
 

Viewers also liked (16)

Sia itar-written-testimony-for-hfac-hearing
Sia itar-written-testimony-for-hfac-hearingSia itar-written-testimony-for-hfac-hearing
Sia itar-written-testimony-for-hfac-hearing
 
Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...
Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...
Ракета космического назначения «Зенит 3SL» для программы «Морской Старт» В.М...
 
National Security Implications of virtual currency examining the potential fo...
National Security Implications of virtual currency examining the potential fo...National Security Implications of virtual currency examining the potential fo...
National Security Implications of virtual currency examining the potential fo...
 
Impact investing - profit meets purpose
Impact investing - profit meets purposeImpact investing - profit meets purpose
Impact investing - profit meets purpose
 
Bnef neo2015 executive-summary
Bnef neo2015 executive-summaryBnef neo2015 executive-summary
Bnef neo2015 executive-summary
 
LauncherOne Virgin Galactic 2013
LauncherOne  Virgin Galactic  2013LauncherOne  Virgin Galactic  2013
LauncherOne Virgin Galactic 2013
 
Philips Lumiled 2001
Philips Lumiled 2001Philips Lumiled 2001
Philips Lumiled 2001
 
Solar Energy Industry Forecast DOE 2008
Solar Energy Industry Forecast DOE 2008Solar Energy Industry Forecast DOE 2008
Solar Energy Industry Forecast DOE 2008
 
State of innovation Thomson Reuters 2016
State of innovation Thomson Reuters 2016 State of innovation Thomson Reuters 2016
State of innovation Thomson Reuters 2016
 
Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009
 Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009 Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009
Презентация А.Б.Чубайса на комиссии Президента 25 ноября 2009
 
Презентация А.Б.Чубайса для французский инвесторов
Презентация А.Б.Чубайса для французский инвесторовПрезентация А.Б.Чубайса для французский инвесторов
Презентация А.Б.Чубайса для французский инвесторов
 
Россия 21 века образ желаемого завтра
Россия 21 века образ желаемого завтраРоссия 21 века образ желаемого завтра
Россия 21 века образ желаемого завтра
 
план мероприятий по стимулированию инновационной активности предприятий
план мероприятий по стимулированию инновационной активности предприятийплан мероприятий по стимулированию инновационной активности предприятий
план мероприятий по стимулированию инновационной активности предприятий
 
Cтратегия РВК на 2010 год
Cтратегия РВК на 2010 годCтратегия РВК на 2010 год
Cтратегия РВК на 2010 год
 
Годовой отчет Роснано за 2008
Годовой отчет Роснано за 2008Годовой отчет Роснано за 2008
Годовой отчет Роснано за 2008
 
Презентация А.Миллера на комиссии по модернизации 25 декабря 2009
Презентация А.Миллера на комиссии по модернизации 25 декабря 2009Презентация А.Миллера на комиссии по модернизации 25 декабря 2009
Презентация А.Миллера на комиссии по модернизации 25 декабря 2009
 

Similar to Cost Of Biopharm 2007

The Price Of Innovation New Drug Development Cost 2003
The Price Of Innovation New Drug Development Cost   2003The Price Of Innovation New Drug Development Cost   2003
The Price Of Innovation New Drug Development Cost 2003Dmitry Tseitlin
 
Translational Data Science_clean
Translational Data Science_cleanTranslational Data Science_clean
Translational Data Science_cleanChris Waller
 
R&D Productivity and Costs in Today's Health Care Arena - Pat Audet
R&D Productivity and Costs in Today's Health Care Arena - Pat AudetR&D Productivity and Costs in Today's Health Care Arena - Pat Audet
R&D Productivity and Costs in Today's Health Care Arena - Pat AudetTTC, llc
 
Drug Discovery Today v9p976
Drug Discovery Today v9p976Drug Discovery Today v9p976
Drug Discovery Today v9p976Michael Weiner
 
Building a Culture
Building a CultureBuilding a Culture
Building a CultureChris Waller
 
Building a Culture of Model-driven Drug Discovery at Merck
Building a Culture of Model-driven Drug Discovery at MerckBuilding a Culture of Model-driven Drug Discovery at Merck
Building a Culture of Model-driven Drug Discovery at MerckChris Waller
 
Biotechnology Publication
Biotechnology PublicationBiotechnology Publication
Biotechnology PublicationChad Houston
 
Medication nonadherence cost and noncompliance in clinical trials
Medication nonadherence cost and noncompliance in clinical trialsMedication nonadherence cost and noncompliance in clinical trials
Medication nonadherence cost and noncompliance in clinical trialsSynegys
 
Drug Development Life Cycle - Costs and Revenue
Drug Development Life Cycle - Costs and RevenueDrug Development Life Cycle - Costs and Revenue
Drug Development Life Cycle - Costs and RevenueRobert Sturm
 
Advancing Medicine, Improving Lives. Presentation.
Advancing Medicine, Improving Lives. Presentation.Advancing Medicine, Improving Lives. Presentation.
Advancing Medicine, Improving Lives. Presentation.ALIKAUSAR58
 
New drug + Pharmacogenetics F22 TMSUc.pptx
New drug + Pharmacogenetics F22 TMSUc.pptxNew drug + Pharmacogenetics F22 TMSUc.pptx
New drug + Pharmacogenetics F22 TMSUc.pptxDrNabanitKumarJha1
 
Pfizer and Merck cooperate to fight cancer
Pfizer and Merck cooperate to fight cancerPfizer and Merck cooperate to fight cancer
Pfizer and Merck cooperate to fight cancerEvgeniia Popova
 
Lead-Optimization-in-the-Drug-Discovery-Process.pdf
Lead-Optimization-in-the-Drug-Discovery-Process.pdfLead-Optimization-in-the-Drug-Discovery-Process.pdf
Lead-Optimization-in-the-Drug-Discovery-Process.pdfBindu238662
 
McK Q Genomics 22125
McK Q Genomics 22125McK Q Genomics 22125
McK Q Genomics 22125Craig Tanio
 
Discovery of Drug and Introduction to Clinical Trial_Katalyst HLS
Discovery of Drug and Introduction to Clinical Trial_Katalyst HLSDiscovery of Drug and Introduction to Clinical Trial_Katalyst HLS
Discovery of Drug and Introduction to Clinical Trial_Katalyst HLSKatalyst HLS
 
The Challenges Of Bacterial Resistance And Natural...
The Challenges Of Bacterial Resistance And Natural...The Challenges Of Bacterial Resistance And Natural...
The Challenges Of Bacterial Resistance And Natural...Pamela Williams
 
Various process of drug development
Various process of drug developmentVarious process of drug development
Various process of drug developmentAyanpal33
 

Similar to Cost Of Biopharm 2007 (20)

The Price Of Innovation New Drug Development Cost 2003
The Price Of Innovation New Drug Development Cost   2003The Price Of Innovation New Drug Development Cost   2003
The Price Of Innovation New Drug Development Cost 2003
 
Translational Data Science_clean
Translational Data Science_cleanTranslational Data Science_clean
Translational Data Science_clean
 
R&D Productivity and Costs in Today's Health Care Arena - Pat Audet
R&D Productivity and Costs in Today's Health Care Arena - Pat AudetR&D Productivity and Costs in Today's Health Care Arena - Pat Audet
R&D Productivity and Costs in Today's Health Care Arena - Pat Audet
 
Drug Discovery Today v9p976
Drug Discovery Today v9p976Drug Discovery Today v9p976
Drug Discovery Today v9p976
 
Building a Culture
Building a CultureBuilding a Culture
Building a Culture
 
Building a Culture of Model-driven Drug Discovery at Merck
Building a Culture of Model-driven Drug Discovery at MerckBuilding a Culture of Model-driven Drug Discovery at Merck
Building a Culture of Model-driven Drug Discovery at Merck
 
Biotechnology Publication
Biotechnology PublicationBiotechnology Publication
Biotechnology Publication
 
Medication nonadherence cost and noncompliance in clinical trials
Medication nonadherence cost and noncompliance in clinical trialsMedication nonadherence cost and noncompliance in clinical trials
Medication nonadherence cost and noncompliance in clinical trials
 
Drug Development Life Cycle - Costs and Revenue
Drug Development Life Cycle - Costs and RevenueDrug Development Life Cycle - Costs and Revenue
Drug Development Life Cycle - Costs and Revenue
 
Advancing Medicine, Improving Lives. Presentation.
Advancing Medicine, Improving Lives. Presentation.Advancing Medicine, Improving Lives. Presentation.
Advancing Medicine, Improving Lives. Presentation.
 
New drug + Pharmacogenetics F22 TMSUc.pptx
New drug + Pharmacogenetics F22 TMSUc.pptxNew drug + Pharmacogenetics F22 TMSUc.pptx
New drug + Pharmacogenetics F22 TMSUc.pptx
 
Pfizer and Merck cooperate to fight cancer
Pfizer and Merck cooperate to fight cancerPfizer and Merck cooperate to fight cancer
Pfizer and Merck cooperate to fight cancer
 
US Medical Research And Dvt
US Medical Research And DvtUS Medical Research And Dvt
US Medical Research And Dvt
 
Lead-Optimization-in-the-Drug-Discovery-Process.pdf
Lead-Optimization-in-the-Drug-Discovery-Process.pdfLead-Optimization-in-the-Drug-Discovery-Process.pdf
Lead-Optimization-in-the-Drug-Discovery-Process.pdf
 
McK Q Genomics 22125
McK Q Genomics 22125McK Q Genomics 22125
McK Q Genomics 22125
 
Directed Project
Directed ProjectDirected Project
Directed Project
 
Homework
HomeworkHomework
Homework
 
Discovery of Drug and Introduction to Clinical Trial_Katalyst HLS
Discovery of Drug and Introduction to Clinical Trial_Katalyst HLSDiscovery of Drug and Introduction to Clinical Trial_Katalyst HLS
Discovery of Drug and Introduction to Clinical Trial_Katalyst HLS
 
The Challenges Of Bacterial Resistance And Natural...
The Challenges Of Bacterial Resistance And Natural...The Challenges Of Bacterial Resistance And Natural...
The Challenges Of Bacterial Resistance And Natural...
 
Various process of drug development
Various process of drug developmentVarious process of drug development
Various process of drug development
 

More from Dmitry Tseitlin

Ton Tlegram Open Network
Ton Tlegram Open NetworkTon Tlegram Open Network
Ton Tlegram Open NetworkDmitry Tseitlin
 
Clarity from above PWC 2016
Clarity from above PWC 2016Clarity from above PWC 2016
Clarity from above PWC 2016Dmitry Tseitlin
 
Отчет Фонд Сколково 25 апреля 2011
Отчет Фонд Сколково 25 апреля 2011Отчет Фонд Сколково 25 апреля 2011
Отчет Фонд Сколково 25 апреля 2011Dmitry Tseitlin
 
Technology roadmap highlights_report 2015
Technology roadmap highlights_report 2015Technology roadmap highlights_report 2015
Technology roadmap highlights_report 2015Dmitry Tseitlin
 
IGS Low cost access to space April 2016
IGS Low cost access to space April 2016IGS Low cost access to space April 2016
IGS Low cost access to space April 2016Dmitry Tseitlin
 
Satellite technologies in UK agriculture 2015
Satellite technologies in UK agriculture 2015Satellite technologies in UK agriculture 2015
Satellite technologies in UK agriculture 2015Dmitry Tseitlin
 
A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...
A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...
A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...Dmitry Tseitlin
 
Постановление Правительства "О реализации национальной технологической иниц...
Постановление Правительства "О реализации национальной технологической иниц...Постановление Правительства "О реализации национальной технологической иниц...
Постановление Правительства "О реализации национальной технологической иниц...Dmitry Tseitlin
 
ESA - 2013 space transportation
ESA - 2013 space transportation ESA - 2013 space transportation
ESA - 2013 space transportation Dmitry Tseitlin
 
Результаты РФФИ 2015
Результаты РФФИ 2015 Результаты РФФИ 2015
Результаты РФФИ 2015 Dmitry Tseitlin
 
Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...
Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...
Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...Dmitry Tseitlin
 
Space Works Nano Microsatellite Market forecast 2016
Space Works Nano Microsatellite Market forecast 2016Space Works Nano Microsatellite Market forecast 2016
Space Works Nano Microsatellite Market forecast 2016Dmitry Tseitlin
 
WIPO Global Innovation Index 2015
WIPO Global Innovation Index 2015WIPO Global Innovation Index 2015
WIPO Global Innovation Index 2015Dmitry Tseitlin
 
Agritech Funder Investing report 2015
Agritech Funder Investing report 2015Agritech Funder Investing report 2015
Agritech Funder Investing report 2015Dmitry Tseitlin
 
High technology entrepreneurs and the patent system. 2008 Berkeley
High technology entrepreneurs and the patent system. 2008 BerkeleyHigh technology entrepreneurs and the patent system. 2008 Berkeley
High technology entrepreneurs and the patent system. 2008 BerkeleyDmitry Tseitlin
 
Start-up Space Tauri Group 2016 -
Start-up Space Tauri Group 2016 -Start-up Space Tauri Group 2016 -
Start-up Space Tauri Group 2016 -Dmitry Tseitlin
 
NASA_CCO_status-2013 update
NASA_CCO_status-2013 updateNASA_CCO_status-2013 update
NASA_CCO_status-2013 updateDmitry Tseitlin
 
SolarCity Analysts Day_2015
SolarCity Analysts Day_2015SolarCity Analysts Day_2015
SolarCity Analysts Day_2015Dmitry Tseitlin
 
SolarCity Silevo_June_2014 investors
SolarCity Silevo_June_2014 investorsSolarCity Silevo_June_2014 investors
SolarCity Silevo_June_2014 investorsDmitry Tseitlin
 

More from Dmitry Tseitlin (20)

Ton Tlegram Open Network
Ton Tlegram Open NetworkTon Tlegram Open Network
Ton Tlegram Open Network
 
Marinet en 2017
Marinet en 2017Marinet en 2017
Marinet en 2017
 
Clarity from above PWC 2016
Clarity from above PWC 2016Clarity from above PWC 2016
Clarity from above PWC 2016
 
Отчет Фонд Сколково 25 апреля 2011
Отчет Фонд Сколково 25 апреля 2011Отчет Фонд Сколково 25 апреля 2011
Отчет Фонд Сколково 25 апреля 2011
 
Technology roadmap highlights_report 2015
Technology roadmap highlights_report 2015Technology roadmap highlights_report 2015
Technology roadmap highlights_report 2015
 
IGS Low cost access to space April 2016
IGS Low cost access to space April 2016IGS Low cost access to space April 2016
IGS Low cost access to space April 2016
 
Satellite technologies in UK agriculture 2015
Satellite technologies in UK agriculture 2015Satellite technologies in UK agriculture 2015
Satellite technologies in UK agriculture 2015
 
A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...
A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...
A Roadmap to Interstellar Flight Philip Lubin Physics Dept, UC Santa Barbara ...
 
Постановление Правительства "О реализации национальной технологической иниц...
Постановление Правительства "О реализации национальной технологической иниц...Постановление Правительства "О реализации национальной технологической иниц...
Постановление Правительства "О реализации национальной технологической иниц...
 
ESA - 2013 space transportation
ESA - 2013 space transportation ESA - 2013 space transportation
ESA - 2013 space transportation
 
Результаты РФФИ 2015
Результаты РФФИ 2015 Результаты РФФИ 2015
Результаты РФФИ 2015
 
Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...
Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...
Cоздание сверхтяжелых РН для исследования и освоения Луны и Марса - прошлое, ...
 
Space Works Nano Microsatellite Market forecast 2016
Space Works Nano Microsatellite Market forecast 2016Space Works Nano Microsatellite Market forecast 2016
Space Works Nano Microsatellite Market forecast 2016
 
WIPO Global Innovation Index 2015
WIPO Global Innovation Index 2015WIPO Global Innovation Index 2015
WIPO Global Innovation Index 2015
 
Agritech Funder Investing report 2015
Agritech Funder Investing report 2015Agritech Funder Investing report 2015
Agritech Funder Investing report 2015
 
High technology entrepreneurs and the patent system. 2008 Berkeley
High technology entrepreneurs and the patent system. 2008 BerkeleyHigh technology entrepreneurs and the patent system. 2008 Berkeley
High technology entrepreneurs and the patent system. 2008 Berkeley
 
Start-up Space Tauri Group 2016 -
Start-up Space Tauri Group 2016 -Start-up Space Tauri Group 2016 -
Start-up Space Tauri Group 2016 -
 
NASA_CCO_status-2013 update
NASA_CCO_status-2013 updateNASA_CCO_status-2013 update
NASA_CCO_status-2013 update
 
SolarCity Analysts Day_2015
SolarCity Analysts Day_2015SolarCity Analysts Day_2015
SolarCity Analysts Day_2015
 
SolarCity Silevo_June_2014 investors
SolarCity Silevo_June_2014 investorsSolarCity Silevo_June_2014 investors
SolarCity Silevo_June_2014 investors
 

Cost Of Biopharm 2007

  • 1. MANAGERIAL AND DECISION ECONOMICS Manage. Decis. Econ. 28: 469–479 (2007) Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/mde.1360 The Cost of Biopharmaceutical R&D: Is Biotech Different? Joseph A. DiMasia,* and Henry G. Grabowskib a Tufts Center for the Study of Drug Development, Tufts University, USA b Department of Economics, Duke University, USA The costs of developing the types of new drugs that have been pursued by traditional pharmaceutical firms have been estimated in a number of studies. However, similar analyses have not been published on the costs of developing the types of molecules on which biotech firms have focused. This study represents a first attempt to get a sense for the magnitude of the R&D costs associated with the discovery and development of new therapeutic biopharmaceu- ticals (specifically, recombinant proteins and monoclonal antibodies [mAbs]). We utilize drug-specific data on cash outlays, development times, and success in obtaining regulatory marketing approval to estimate the average pre-tax R&D resource cost for biopharmaceuticals up to the point of initial US marketing approval (in year 2005 dollars). We found average out-of-pocket (cash outlay) cost estimates per approved biopharmaceutical of $198 million, $361 million, and $559 million for the preclinical period, the clinical period, and in total, respectively. Including the time costs associated with biopharmaceutical R&D, we found average capitalized cost estimates per approved biopharmaceutical of $615 million, $626 million, and $1241 million for the preclinical period, the clinical period, and in total, respectively. Adjusting previously published estimates of R&D costs for traditional pharmaceutical firms by using past growth rates for pharmaceutical company costs to correspond to the more recent period to which our biopharmaceutical data apply, we found that total out-of-pocket cost per approved biopharmaceutical was somewhat lower than for the pharmaceutical company data ($559 million vs $672 million). However, estimated total capitalized cost per approved new molecule was nearly the same for biopharmaceuticals as for the adjusted pharmaceutical company data ($1241 million versus $1318 million). The results should be viewed with some caution for now given a limited number of biopharmaceutical molecules with data on cash outlays, different therapeutic class distributions for biopharmaceuticals and for pharmaceutical company drugs, and uncertainty about whether recent growth rates in pharmaceutical company costs are different from immediate past growth rates. Copyright # 2007 John Wiley & Sons, Ltd. INTRODUCTION examine approaches that could reduce those costs. If the productivity of new drug development can The financial viability of new drug and biophar- be improved, then more innovations may be maceutical development depends on the expected pursued and eventually reach the patient. The costs of, as well as the returns to, R&D. When Food and Drug Administration (FDA), through R&D costs are substantial it is important to its Critical Path Initiative, has initiated a process to, in part, explore how the agency, industry, and *Correspondence to: Tufts Center for the Study of Drug Development, Tufts University, 192 South Street, Suite 550, academia can establish methods that would lower Boston, MA 02111, USA. E-mail: joseph.dimasi@tufts.edu development costs (FDA, 2004). Copyright # 2007 John Wiley & Sons, Ltd.
  • 2. 470 J.A. DiMASI AND H.G. GRABOWSKI R&D costs for new drugs (including the costs of project-level and aggregate annual expenditure failures and time costs) have been estimated to data for a consulting project for a biotech firm.2 average in excess of $800 million (in year 2000 We combined data by period and type of dollars) for development that led to approvals in compound from these two sources. We focus on the 1990s, with a marked upward trend relative to therapeutic recombinant proteins and monoclonal earlier decades (DiMasi et al., 2003). These R&D antibodies (mAbs), which are overwhelmingly the cost estimates have used data on new drugs two most prevalent compound types in the biotech developed by traditional pharmaceutical firms sector. The consulting project focused on compounds (primarily new chemical entities). No study to that first entered clinical testing from 1990 to 2003. date has focused on the types of molecules that are With compound type and period of initial clinical developed by biotech firms. One might conjecture testing as study criteria, we utilized data on four that biopharmaceuticals are less costly to develop biologics from three companies used in the earlier because biotech firms need to be more nimble and study and 13 compounds from the biotech firm.3 creative or that fewer safety issues arise for many While the data on cash outlays are limited to the biopharmaceuticals because they replace sub- 17 compounds noted above, we are able to use a stances that exist naturally in the body. However, much larger data set to estimate average develop- some industry insiders estimate that costs, even for ment times, clinical success rates, and phase biotech firms, exceed $1 billion.1 transition probabilities. These data are used to In this paper, we make a first attempt to account for time costs and the costs of develop- examine the magnitude of R&D costs associated ment failures.4 We used a Tufts Center for the with developing the types of molecules on which Study of Drug Development (CSDD) database of biotech firms focus. Specifically, we use drug- biopharmaceutical compounds. The Tufts CSDD specific cost, development time, and clinical database is constructed from information con- success rate data for therapeutic biopharmaceu- tained in a number of commercial business ticals to estimate pre-tax R&D resource costs. We intelligence databases (PharmaProjects, R&D then compare these results to those obtained for Focus, and iDdb3), trade press accounts, company development of new drugs by traditional pharma- reports and websites, and company surveys. For ceutical firms (DiMasi et al., 2003). Given that the our analyses of development times, clinical success biopharmaceutical data are, on average, more rates, and phase transition probabilities, we used a recent than the data used for DiMasi et al. (2003), subset of this database. The compounds included we estimate the difference in study periods. Our are therapeutic recombinant proteins and mAbs that results for biopharmaceutical development are were first tested in humans from 1990 through 2003. then also compared to those for traditional There are 522 such compounds, and they include pharmaceutical firms with costs extrapolated using molecules that were abandoned during develop- estimated past growth rates for pharma costs to ment, as well as those that have attained US Food coincide with the more recent biopharmaceutical and Drug Administration (FDA) approval.5 study period. We compare our results for biopharmaceuticals The rest of this paper is organized as follows. to our previously developed estimates of R&D The next section contains a description of the data costs for new drugs developed by traditional used for our analyses. Then, we describe the pharmaceutical firms. The data underlying the methods used to obtain our results. We further ‘pharma’ results are described in DiMasi et al. present our results. Finally, we summarize our (2003). These data included cash outlays for 68 conclusions and offer some discussion of the results. new drugs and development times, clinical ap- proval success rates, and transition probabilities for a larger data set of 534 new drugs. DATA Our data on project costs derive from two sources. METHODS First, the sample for our study of pharmaceutical R&D costs (DiMasi et al., 2003) contained a small The methodology used for the analysis here is number of biologic compounds developed by explained in detail in DiMasi et al. (2003). We pharmaceutical firms. Second, we obtained shall only briefly outline the methods here. Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 3. THE COST OF BIOPHARMACEUTICAL R&D 471 Out-of-Pocket Costs: Phase Means, Success Rates, in the sample that have proceeded from one phase and Expected Costs to another among the set of drugs that entered the first phase and either proceeded to the next We refer to actual cash outlays of the firm as out- phase or were terminated in the first phase. of-pocket costs. We converted the data on clinical This approach should provide reasonable esti- period expenditures by phase and year to 2005 mates of phase transition probabilities since the dollars using the GDP Implicit Price Deflator. We lengths of individual phases are short relative to determined mean costs for these molecules for total development times. The implicit assumption phase I, phase II, and phase III. Long-term animal needed for such an approach is that those drugs testing costs incurred during clinical development, that are still active at the time of analysis will regulatory approval submission costs, and chem- proceed to later phases more or less in accordance istry, manufacturing and control costs related to with the estimated transition probabilities. The development and incurred during clinical develop- overall clinical success rate is then determined as ment are subsumed in the cost estimates for the the product of the phase transition probabilities. clinical phases. The expenditures considered in this Clinical success is defined as US regulatory report for the sample of 17 molecules are only approval for marketing. those that were incurred prior to original market- Expected out-of-pocket cost per investigational ing approval. drug is the weighted average of mean phase costs, To obtain a full R&D cost estimate that would where the weights are the estimated probabilities account for the costs of failures and the time cost that an investigational molecule will enter a given of new pharmaceutical development, we must phase. Finally, the out-of-pocket cost per ap- build up to one through analyses of the expected proved new molecule is obtained by dividing the costs for the clinical and preclinical periods. For out-of-pocket cost per investigational molecule by purposes of this study, by the clinical period we the estimated clinical approval success rate. mean the time from initial human testing of a Preclinical costs are obtained in a manner compound to original marketing approval. The similar to the method we used in DiMasi et al. preclinical period refers to activities engaged in (2003). We examined time series data on aggregate prior to the start of human testing. Thus pre- preclinical and clinical expenditures for new clinical R&D costs include expenditures for both molecules at the firm level. These data, along with basic research and preclinical development. our estimated clinical period costs per investiga- Expected costs take into account the fact that tional and per approved molecule, were used to not all compounds will progress all the way infer the corresponding values for preclinical costs. through development to approval. We first work The time series data on preclinical and clinical at the investigational molecule level. For the expenditures were linked, as was done in DiMasi clinical period this means that we must estimate et al. (2003), via an estimated 5-year lag between the probabilities that a compound that enters the the middle of the preclinical period and the middle clinical testing pipeline will make it to each phase. of the clinical period.7 These values can be estimated from the data in the Tufts CSDD database of biopharmaceutical com- Capitalized Costs: Development Times and Discount pounds. Statistical inference using, in part, survi- Rate val analysis to account for censoring of the data has been implemented in a number of studies of Drug development is a very lengthy process. As drug industry success rates.6 However, given such, there are substantial time costs to investing lengthy drug development times, such an approach in R&D years before any potential returns can be requires that there be a substantial period of time earned. We capture the time costs of drug between when the most recent drug enters clinical development in a single monetary measure by testing and when the analysis is conducted. Since capitalizing costs forward to the point of original we must include here drugs that have entered the marketing approval at an appropriate discount clinical testing pipeline relatively recently, we have rate. The discount rate used is a cost of capital estimated success and phase attrition rates in a estimate for a sample of firms obtained from more mechanistic manner. We estimated a phase applying the Capital Asset Pricing Model transition probability to be the percentage of drugs (CAPM). More detail on this process is explained Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 4. 472 J.A. DiMASI AND H.G. GRABOWSKI below in the context of a discussion of the result phase transition probabilities shown in Figure 1. we obtained for a biotech discount rate. For comparative purpose, we also reproduce the Capitalized costs are the sum of out-of-pocket phase transition probabilities for the DiMasi et al. and time costs. To obtain time costs we not only (2003) study. Multiplying the phase transition need an appropriate discount rate, but also a probability estimates for biopharmaceuticals timeline over which out-of-pocket costs are yields an overall clinical approval success rate of capitalized forward to marketing approval at the 30.2% (as opposed to 21.5% for pharma). To discount rate. Thus, we estimate average clinical obtain an estimate of the expected clinical period phase and regulatory review lengths from the data cost per investigational molecule we need esti- in our subset of therapeutic recombinant proteins mated probabilities that a molecule that enters and mAbs. As noted above, we use the estimate in clinical testing will reach a given phase. Those DiMasi et al. (2003) for the time from discovery to values can be derived from the transition prob- first human testing. abilities and the overall clinical approval success rate. To be conservative, we assume a 100% success rate for regulatory approval submissions RESULTS to the FDA so that the probability that a regulatory submission will be made is assumed to Our focus is on biopharmaceutical development, but be the same as the overall clinical approval success we will also make some comparisons to estimated rate. Previous studies have shown 100% success costs for traditional pharmaceutical firms. rates for regulatory submissions for biophar- maceuticals for almost every period analyzed (Reichert and Paquette, 2003; Reichert, 2005). Clinical Phase Costs per Investigational Molecule Altering this value within reason does not have Table 1 shows our estimated average clinical an appreciable effect on the results. Applying period phase costs for the sample of compounds the probabilities as weights for the mean costs for which we obtained detailed data. Mean clinical yields an estimated out-of-pocket cost per phase costs are higher than those that we had investigational molecule of $169 million for obtained in our R&D cost study for traditional biopharmaceuticals. pharmaceutical firms when adjusted for inflation. For the period we analyzed, the sum of the clinical Out-of-Pocket Clinical Cost per Approved Molecule. period mean phase costs for biopharmaceuticals What we are mainly interested in are costs per ($166 million) is 14% higher than what we had approved new molecule. We obtain such values by found for pharma development ($146 million).8 dividing costs per investigational molecule by the estimated clinical approval success rate (30.2%). Success Rate and Phase Transition Probabilities. This yields an estimate of the out-of-pocket Using information from the Tufts CSDD bio- clinical period cost per approved new molecule pharmaceutical database, we estimated the of $361 million for biopharmceuticals. Out-of-Pocket Preclinical Cost per Investigational Table 1. Out-of-pocket Preclinical and Clinical Molecule Period Cost per Investigational Biophar- maceutical Compound (In Millions of Preclinical cost per investigational molecule is 2005 dollars)a obtained by multiplying our estimated clinical Testing Mean Probability Expected phase cost per investigational molecule by a ratio phase cost ($) of entering cost ($) of preclinical to clinical expenditures obtained by phase (%) applying the lag noted above to the aggregate Preclinical 59.88 100 59.88 expenditure time series data. The aggregate Phase I 32.28 100 32.28 Phase II 37.69 83.7 31.55 data, with a lag imposed, implies that clinical Phase III 96.09 47.1 45.26 period phase costs should account for 65% of total out-of-pocket cost. These estimates yield an Total 168.97 out-of-pocket preclinical cost per investigational a All costs were deflated using the GDP Implicit Price Deflator. molecule of $59.9 million, and, using a 30.2% Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 5. THE COST OF BIOPHARMACEUTICAL R&D 473 clinical approval success rate, a preclinical out from DiMasi et al. (2003). Total clinical plus of-pocket cost per approved new molecule of approval time is 8% longer for the biopharma- $198 million. ceuticals, with nearly all of the difference accounted for by phase I. Capitalized Costs Cost of Capital. In our prior analysis of tradi- tional pharmaceutical firms, we utilized a cost of As noted above, to obtain estimates that include capital of 11% as a discount rate for R&D the time costs of new drug development we need to activities that were first taken into clinical trials estimate development times and choose an appro- between 1983 and 1994 (DiMasi et al., 2003; priate discount rate. Grabowski et al., 2002). This cost of capital estimate was based on concepts from modern Development Times. Our data on biopharmaceu- finance theory. tical compound development histories for the Utilizing the CAPM framework (Brealey and period analyzed yielded the mean clinical devel- Myers, 2000), the firm’s cost of capital, rn ; is a opment and approval phase lengths shown in weighted average of its cost of capital on its debt Figure 2. The phase results are averages across all and equity capital.9 Given the low debt values of compounds that completed the phase, regardless large pharmaceutical firms, the equity cost of of whether the compound was ultimately approved capital becomes the key factor driving the for marketing. For comparative purposes we weighted cost of capital for the firms. In the case also show the pharma development time results of biotech firms the debt component is negligible, 83.7% Transition Probability 71.0% 68.5% 64.2% 56.3% 44.2% Phase I-II Phase II-III Phase III-Approval Biotech Pharma Figure 1. Transition probabilities for clinical phases. Biotech 19.5 29.3 32.9 16 97.7 Pharma 12.3 26.0 33.8 18.2 90.3 0.0 120.0 Months Phase I Phase II Phase III RR Figure 2. Clinical development and approval times. Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 6. 474 J.A. DiMASI AND H.G. GRABOWSKI given that long-term debt after 1990 is less than financial economists, we estimated cost of capital 1% of market valuation. Thus, for all practical values for a sample of biotech firms at roughly five purposes the equity cost of capital for biotech year intervals from their 1989 estimate. The lower firms is the same as their weighted cost of capital. value in 2004 reflects declining value in the risk In the CAPM framework, investors require a free rate and the equity premium in recent years risk premium for holding equity in a particular compared to the 1994–1999 period. The focus of company. This premium is based on the relative our analysis is on R&D projects initiated since the riskiness to investors of that company’s assets. The mid 1990s through the early 2000s where a 10% to formal measure of relative riskiness is the beta 12.5% rate was observed. We therefore use the coefficient, or the firm’s contribution to the average of the three values for the cost of capital in variance in the returns from a diversified portfolio Table 2, 11.5%, as the benchmark value for biophar- of equity shares. The CAPM assumes that maceuticals. We also perform simulations around this investors hold well-diversified portfolios. baseline value to analyze the sensitivity of the The CAPM implies that the expected return on capitalized R&D cost to this cost of capital value.10 a firm’s assets (the equity cost of capital) is equal Discussions with a few of the leading pharma- to the risk-free rate plus a risk premium which is ceutical firms suggest that a nominal cost of capital positively related to the riskiness of the firm’s in the range of 12–15% was being utilized by many assets relative to other stock market assets large pharma firms in 2001–2002 (Grabowski, et al., 2002). Given a 3% rate of inflation, this rE ¼ rf þ betaðrm À rf Þ: would imply a 10–12% real cost of capital for In this equation rf is the risk-free rate (the return major pharmaceutical firms. This is roughly in treasury bonds minus a horizon premium is consistent with estimates of the cost of capital typically used as a proxy for the risk-free rate); derived from the CAPM in this period. rm is the long-term rate of return for a market basket of common stock (usually the S&P index); Capitalized Costs per Investigational Molecule. We ðrm À rf Þ is the equity premium, and beta is a obtain capitalized costs by spreading our esti- measure of the relative riskiness of a specific firm mated expected out-of-pocket phase costs per (based on a regression analysis that yields the investigational molecule over estimated mean covariance of returns with the overall S&P index). phase lengths and then capitalizing them forward Under CAPM, a firm with a beta of one would to marketing approval at an 11.5% discount rate have the same riskiness as the overall S&P index, using a representative time profile. The results are whereas those with values greater than one are shown in Table 3. more risky, and correspondingly, those with betas Preclinical capitalized cost per investigational below one are less risky. Company specific values molecule is obtained by spreading the out-of- for beta can be found in Value Line’s Investment pocket cost per investigational molecule deter- Surveys and other security analyst publications. mined above ($60 million) over an estimated Betas in these sources are typically updated on a preclinical period (52.0 months) and then capita- periodic basis. lizing forward to marketing approval at an 11.5% Myers and Shyam-Sunder (1995) examined the discount rate over the representative time profile. cost of capital for seven smaller biotechnology and Doing so yields a capitalized preclinical period specialty pharmaceutical firms for 1989. These cost per investigational molecule of approxima- firms had higher betas and costs of capital than the tely $186 million. Capitalized clinical cost per major pharmaceutical firms. The greater betas or investigational molecule is obtained by capitalizing riskiness exhibited by these firms were consistent with the fact that the smaller biotech firms had fewer commercialized products and proportio- Table 2. Nominal and Real Cost of Capital nately more earlier-stage R&D projects. The (COC), 1994–2004 average cost of capital for the full sample of seven biotech and specialty pharma firms was 19% in 1994 2000 2004 nominal terms and 14% in real terms. Nominal COC (%) 17.0 15.0 13.0 Using the same methodology as employed Inflation rate (%) 4.5 3.0 3.0 Real COC (%) 12.5 12.0 10.0 by Myers and Shyam-Sunder (1995) and other Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 7. THE COST OF BIOPHARMACEUTICAL R&D 475 Table 3. Capitalized Preclinical and Clinical Period costs per Investigational Biopharmaceutical Compound (In Millions of 2005 dollars)a Testing phase Expected out-of- Phase length Monthly Start of phase End of Expected pocket cost ($) (mos.) cost ($) to approval phase to capitalized (mos.) approval (mos.) costb ($) Preclinical 59.88 52.0 1.15 149.7 97.7 185.62 Phase I 32.28 19.5 1.66 97.7 78.2 71.78 Phase II 31.55 29.3 1.08 78.2 48.9 56.32 Phase III 45.26 32.9 1.38 48.9 16.0 60.98 Total 374.70 a All costs were deflated using the GDP Implicit Price Deflator. b Expenditures capitalized forward to the point of marketing approval for a representative time profile at an 11.5% real discount rate. The estimated length of the approval phase is 16.0 months. out-of-pocket clinical phase cost forward to 14% higher for the clinical period, and 24% higher marketing approval according to the time profile in total. in Figure 2. This yields a capitalized clinical period It may be the case, however, that the appro- cost per investigational molecule of approximately priate figures for R&D costs for traditional $189 million. pharmaceutical firms to compare with our biotech estimates should be much higher than those shown by the middle bars of Figure 4. The reason is that Total R&D Costs per Approved Molecule the biotech data are somewhat more recent than the data used for DiMasi et al. (2003). We To get estimates of fully allocated total cost per conducted two types of comparisons to judge the approved new molecule, we need only add extent to which the period is shifted. Examining estimates of cost per approved molecule for the both actual approval dates for biotech compounds preclinical and clinical periods. Applying the in the Tufts CSDD database and for those used in clinical approval success rate of 30.2% for the DiMasi et al. (2003) sample, as well as average biopharmaceuticals to the capitalized preclinical approval dates on which phase I testing began for cost per investigational molecule noted above biopharmaceutical compounds and for the data in yields a preclinical period cost per approved new DiMasi et al. (2003), suggested a shift of approxi- molecule of $615 million. Similarly, applying the mately five years in the study periods. Thus, we success rate to our estimate of capitalized clinical should consider what new drug development costs period cost per investigational molecule yields a for pharma firms would be five more years into the capitalized clinical period cost per approved future. In DiMasi et al. (2003) we compared costs molecule of $626 million. Total capitalized cost for the current sample to those for an earlier per approved molecule for biopharmaceuticals is period covered by a previous study (with more then $1241 million. Out-of-pocket, time, and than a decade difference in time). We applied the capitalized costs per approved new molecule are growth rates (over and above inflation) for the shown in Figure 3. preclinical and clinical periods that we observed between our two earlier studies on pharma costs to R&D Cost Comparisons: Biotech and Pharma. Our the most recent pharma data assuming a further estimates for biopharmaceutical out-of-pocket five-year shift. The results are the pharma time- preclinical, clinical, and total out-of-pocket adjusted values given by the third set of bars in R&D costs are shown in Figure 4. For compara- Figure 4. The unadjusted figures can be viewed as tive purposes, we also show the corresponding what the outcomes for pharmaceutical firms would figures for pharma from our most recent study be if they had kept cost increases in the later five- of R&D costs for traditional pharmaceutical year period in line with general inflation. firms (DiMasi et al., 2003). The overall figures The time-adjusted out-of-pocket biotech costs for pharma firms are significantly lower than for the preclinical period are still somewhat higher those for biotech development. Biopharmaceutical than for pharma even with the period adjustment costs are 46% higher for the preclinical period, (32% higher). However, for the clinical period and Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 8. 476 J.A. DiMASI AND H.G. GRABOWSKI 1,241 Millions (2004$) 682 615 626 559 417 361 265 198 Preclinical** Clinical Total Out-of-pocket Time Capitalized * Based on a 30.2% clinical approval success rate ** All R&D costs (basic research and preclinical development) prior to initiation of clinical testing Figure 3. Pre-approval out-of-pocket (cash outlay) and time costs per approved new biopharmaceutical.Ã 672 522 559 Millions (2005$) 452 361 316 198 136 150 Preclinical* Clinical Total Biotech Pharma Pharma (time-adjusted)** * All R&D costs (basic research and preclinical development) prior to initiation of clinical testing ** Based on a 5-year shift and prior growth rates for the preclinical and clinical periods Figure 4. Pre-approval cash outlays (out-of-pocket cost) per approved new molecule. in total, biopharmaceutical out-of-pocket costs are clinical period and total costs are proportionately lower than our reported pharma costs adjusted for closer to pharma costs than are out-of-pocket a later period. Specifically, clinical period costs are costs. Capitalized clinical period costs for bio- 31% lower and total costs are 17% lower for pharmaceuticals are 29% lower than for time- biopharmaceuticals. Of course, we do not know if adjusted pharma costs. However, total capitalized pharma costs continued to increase at the same cost per approved biopharmaceutical ($1241 mil- rates as they had in the past. lion) is only 6% lower than total capitalized time- Our main results for capitalized costs are shown adjusted pharma cost ($1318 million). in Figure 5. Capitalization increases biopharma- ceutical costs relative to pharma costs because of a longer development timeline and a higher cost of CONCLUSIONS capital.11 As a result, the capitalized preclinical costs for biotech are proportionately higher While estimates of the level of, and trends in, R&D (40%) relative to time-adjusted pharma costs costs for traditional pharmaceutical firms have than are out-of-pocket costs. However, capitalized been published, to date no studies have focused Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 9. THE COST OF BIOPHARMACEUTICAL R&D 477 1,318 1,241 Millions (2005$) 879 899 615 626 523 439 376 Preclinical* Clinical Total Biotech Pharma Pharma (time-adjusted)** * All R&D costs (basic research and preclinical development) prior to initiation of clinical testing ** Based on a 5-year shift and prior growth rates for the preclinical and clinical periods Figure 5. Pre-approval capitalized cost per approved new molecule. specifically on biotech firms or particular types of new molecule appears to be essentially the same as biopharmaceutical development. We have taken a estimated total capitalized per approved new drug first step toward getting a sense for the magnitude for traditional pharmaceutical firms. However, of what the full R&D resource costs associated total out-of-pocket costs for biopharmaceuticals with discovering and developing biopharmaceuti- were found to be somewhat lower, both out-of- cals to the point of initial regulatory marke- pocket and capitalized clinical period costs for ting approval had been for recent years. biopharmaceuticals were lower, and preclinical Using compound-specific costs for a sample of 17 period costs for biopharmaceuticals were some- investigational biopharmaceuticals from four what higher.12 Determining what the actual firms, a time series of annual preclinical and growth rates in costs for pharma firms had been clinical expenditures for a biotech firm, estimated in recent years awaits further study. average development times and phase transition Several caveats to our results should be men- probabilities for over 500 therapeutic recombinant tioned. The results are preliminary in that the proteins and mAbs, we estimated average pre- sample size for mean phase costs is relatively small, clinical period, clinical period, and total costs per although the sample sizes for development times approved new biopharmaceutical. We found out and success rates are quite large. Beyond this, the of-pocket (cash outlay) cost estimates of $198 comparisons with pharma costs should be viewed million, $361 million, and $559 million per with some caution for two reasons. First, as noted, approved new biopharmaceutical for the preclini- pharma costs may not have changed to the same cal period, the clinical period, and in total, degree in recent years as they did in the past. respectively (in year 2005 dollars). These figures Second, costs can vary by therapeutic class include the costs of compound failures. Adding (DiMasi et al., 2004). The distributions of inves- time costs to cash outlays, we found cost estimates tigational compounds by therapeutic class for of $615 million, $626 million, and $1241 million traditional pharmaceutical firms do differ from per approved new biopharmaceutical for the the distributions by class for the recombinant preclinical period, the clinical period, and in total, protein and mAb biopharmaceuticals that we respectively (in year 2005 dollars). examined. Specifically, investigational biopharma- Our estimates for biopharmaceuticals are higher ceutical molecules were more concentrated in the than those we found for our previous study of oncology and immunologic categories than were pharma costs (DiMasi et al., 2003). However, the pharma molecules for the period analyzed in DiMasi biopharmaceutical data that we used is of a more et al. (2003), while the pharma distribution was more recent vintage. If past growth rates in R&D costs concentrated in the cardiovascular and neuropharma- for traditional pharmaceutical firms are applied to cologic classes. It is unclear how these differences the results in DiMasi et al. (2003), then total affect the comparative results; while full clinical period capitalized biopharmaceutical cost per approved costs for new cardiovascular and neuropharmacologic Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 10. 478 J.A. DiMASI AND H.G. GRABOWSKI drugs were found in DiMasi et al. (2004) to be about 9. The weighted average company cost of capital can average for pharma development, not enough infor- be expressed in terms of the following equation: mation was available to determine costs for oncology rn ¼ rD ð1 À TC ÞðD=VÞ þ rE ðE=VÞ; and immunologic drugs. Additional research is needed where rD and rE are the expected rates of return on to fully resolve these issues. assets of comparable riskiness for the firm’s debt and equity securities, respectively. TC is the firm’s corpo- rate tax rate, and D=V and E=V are the proportion of NOTES the firm’s market valuation represented by debt and equity securities, respectively. The debt component of 1. Gottschalk (2004) notes that a manager at a biotech the cost of capital is multiplied by ð1 À TC Þ; because company estimated that his company spends in interest on debt obligations is tax deductible, while excess of one billion dollars to get a drug to market earnings on equity shares are not. (lecture to Professor Fiona Murray’s MIT Sloan 10. Financial economists suggest that the risk and cost Management class 15.968, ‘Building a Biomedical of capital of an individual R&D project will depend Business,’ by Bill Anderson, VP Business Planning, on the stage of the project and, correspondingly, on Biogen Idec Inc., 3 December 2003). the amount and timing of follow-on investments 2. The firm provided data on its R&D expenditures in required to achieve commercial success. By contrast, the form required to apply the basic methodology the estimates derived from corporate financial data used in DiMasi et al. (2003). The purpose was to test by Myers and Shyam-Sunder (1995) and other their hypothesis that their R&D costs were in fact financial economists represent an average cost of significantly lower than the estimate in DiMasi et al. capital for a firm’s aggregate portfolio of R&D (2003) for traditional pharmaceutical firms. projects as well as their complementary capital 3. The sample consisted of nine recombinant proteins investments in manufacturing and marketing assets. and eight mAbs. Some analyses of the pharmaceutical industry have 4. Given that we use development times and success rates utilized a higher cost of capital for earlier stage for what is essentially the universe of biopharmaceu- R&D projects based on cost of capital estimates ticals developed by all firms, it is not possible to infer from firms at different stages of the life cycle. For what costs per approved new molecule are for the example, the Office of Technology Assessment (U.S. biotech firm that provided molecule-specific cash Congress, Office of Technology Assessment, 1993) outlays to us. Company-specific success rates, in utilized a 14.5% real cost of capital for the earlier particular, can have a substantial impact on total pre-clinical stages of pharma R&D based on the R&D costs for a given company. One might wonder, Myers and Shyam-Sunder (1995) biotech and small however, about the internal consistency of all of the firm sample, and lower values for later stages of the data. It is unlikely that company-specific mean clinical life cycle. Myers and Howe (1997) generate a ‘stair- phase expenditures will have an appreciable effect on stepped’ cost of capital using a Monte Carlo success rates. It is also likely that mean clinical phase simulation model and an option value approach. To costs for an investigational molecule of a given type our knowledge, none of the big pharma firms use a and therapeutic class will not vary much across firms. stair-stepped cost of capital in their NPV and return One potential concern, though, is the possibility that calculations, but some are considering this approach. there were some time–cost tradeoffs for the phase data 11. The discount rate has a modest effect on total (Scherer, 1966). This is more of a concern for capitalized costs. If we use a 10.5% discount rate for molecules that fail in testing than for those that biotech, then its total capitalized cost falls by 6.8%. succeed, since the total amount of testing for If we use a 12.5% discount rate for biotech, then molecules that are eventually approved for marketing total capitalized cost is 7.3% higher. is likely to be essentially the same, regardless of 12. The higher preclinical expenditures per approved whether some testing is done in parallel rather than biopharmaceutical may, to some extent, help sequentially. We have no reason to believe that the explain the higher clinical approval success rates biotech firm in question here differed from other firms for biopharmaceuticals. with regard to time–cost tradeoffs. 5. This data set consisted of 278 recombinant proteins and 244 mAbs. 6. See, for example, DiMasi et al. (1991), DiMasi (1995), REFERENCES Gosse et al. (1996), DiMasi (2001), DiMasi et al. (2003). 7. In the absence of precise data on the length of the Brealy RA, Myers SC. 2000. Principles of Corporate preclinical period for these molecules, we used the Finance (6th edn). Irwin/McGraw-Hill: Burr Ridge, IL. value estimated for DiMasi et al. (2003). Managers DiMasi JA. 1995. Success rates for new drugs entering at the biotech firm from which we obtained cost clinical testing in the United States. Clinical Pharma- data agreed that the estimate was reasonable. cology and Therapeutics 58(1): 1–14. 8. See, however, our discussion below about differ- DiMasi JA. 2001. Risks in new drug development: ences in time periods between our previous study approval success rates for investigational drugs. Clinical and the data used for this report. Pharmacology and Therapeutics 69(5): 297–307. Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde
  • 11. THE COST OF BIOPHARMACEUTICAL R&D 479 DiMasi JA, Grabowski HG, Vernon J. 2004. R&D costs Grabowski HG, Vernon JM, DiMasi JA. 2002. Returns and returns by therapeutic category. Drug Information on research and development for 1990s new drug Journal 38(3): 211–223. introductions. PharmacoEconomics 20(Supplement 3): DiMasi JA, Hansen RW, Grabowski HG. 2003. The 11–29. price of innovation: new estimates of drug deve- Reichert JM. 2005. Monoclonal antibody successes in lopment costs. Journal of Health Economics 22(3): the clinic. Nature Biotechnology 23(9): 1073–1078. 141–185. Reichert JM, Paquette C. 2003. Clinical development DiMasi JA, Hansen RW, Grabowski HG, Lasagna L. of therapeutic recombinant proteins. BioTechniques 1991. Cost of innovation in the pharmaceutical 35(1): 1–6. industry. Journal of Health Economics 10(2): 107–142. Myers SC, Howe CD. 1997. A life cycle financial model Food and Drug Administration. 2004. Challenge and of pharmaceutical R&D. MIT Program on the Opportunity on the Critical Path to New Medical Pharmaceutical Industry, Massachusetts Institute of Products. Available at: http://www.fda.gov/oc/initia- Technology, Cambridge, MA. tives/criticalpath/whitepaper.pdf [accessed 21 February Myers SC, Shyam-Sunder L. 1995. Measuring pharma- 2006]. ceutical industry risk and the cost of capital. In Gosse ME, DiMasi JA, Nelson TF. 1996. Recombinant Competitive Strategies in the Pharmaceutical Industry, protein and therapeutic monoclonal antibody drug Helms RB (ed.). American Enterprise Institute for development in the United States: 1980–1994. Clinical Public Policy: Washington, DC. Pharmacology and Therapeutics 60(6): 608–618. Scherer FM. 1966. Time–cost tradeoffs in uncertain Gottschalk AHB. 2004. Improving the efficiency of the empirical research projects. Naval Research Logistics later Stages of the Drug Development Process: Survey Quarterly 13: 71–82. Results from the Industry, Academia, and the United States Congress, Office of Technology Assess- FDA. Dissertation for Master of Science in Health ment. 1993. Pharmaceutical R&D: Costs, Risks Sciences and Technology, Massachusetts Institute of and Rewards. US Government Printing Office: Technology. Washington, DC. Copyright # 2007 John Wiley & Sons, Ltd. Manage. Decis. Econ. 28: 469–479 (2007) DOI: 10.1002/mde