Disentangling the Cost of Orphan Drugs Marketed in the United States

The increasing number and high prices of orphan drugs have triggered concern among patients, payers, and policymakers about the affordability of new drugs approved using the incentives set by the Orphan Drug Act (ODA) of 1983. This study evaluated the factors associated to the differences in the treatment cost of new orphan and non-orphan drugs approved by the FDA from 2017 to 2021. A generalized linear model (GLM) with the Gamma log-link analysis was used to ascertain the association of drug characteristics with the treatment costs of orphan and non-orphan drugs. The results of the study showed that the median and interquartile range (IQR) drug cost was USD 218,872 (IQR = USD 23,105) for orphan drugs and USD 12,798 (IQR = USD 57,940) for non-orphan drugs (p < 0.001). Higher market entry prices were associated with biologics (108%; p < 0.001), orphan status (177%; p < 0.001), US sponsor companies (48%; p = 0.035), chronic use (1083%; p < 0.001), treatment intent (163%; p = 0.004), and indications for oncology (624%; p < 0.001) or genetic disorders (624%; p < 0.001). Higher market entry treatment cost for newly approved drugs were associated with biologics, orphan status, US sponsor companies, chronic use, therapeutic intent, and indications for oncology or genetic disorders.


Introduction
In the United States (US), orphan drugs are indicated for the treatment of rare diseases and conditions affecting fewer than 200,000 patients [1]. With an estimated 7000 orphan diseases, 1 out of every 10 Americans live with a rare condition [2]. The Orphan Drug Act (ODA) was introduced in 1983 to encourage the development of new drugs for such conditions. The orphan designation introduced by the Orphan Drug Act of 1983 allows drug manufacturers to benefit from several incentives, such as market exclusivity, fee waivers, direct funding for research and development (R&D), and tax credits that aim to boost returns on investment in orphan drug research and development [3,4].
The Increase in the demand for orphan drugs to address a growing number of rare diseases coupled with the steady increase in prices has raised concerns about the affordability of orphan drugs [5][6][7][8]. New drugs are expensive and contribute to rising healthcare costs for public and private patients [9][10][11], and the FDA orphan designation is associated with higher prices and out-of-pocket expenditures [12,13]. However, studies assessing the factors behind differences in the costs of orphan and non-orphan drugs in the US are lacking. This study evaluated the factors associated to the differences in the treatment cost of new orphan and non-orphan drugs approved by the FDA from 2017 to 2021.

Data Sources
We extracted the list of new molecular entities and therapeutic biologics approved and marketed in the US in 2017-2021 from the FDA website [14]. Vaccines, allergenic products, and blood and blood products were excluded from the study. We collected the first wholesale acquisition costs (WACs) from the IBM Micromedex RED BOOK and used the WACs at market entry as proxies for the actual acquisition costs by private payers. Pharmaceutical companies use the WACs to set the initial reference price in the Medicaid outpatient pharmacy, 340B Drug Pricing Program, and Federal Supply Schedule programs [15,16]. The Medicare Part B program also uses the WACs to set the initial prices for reimbursement of drugs used in physician offices. We collected price data at the national drug code (NDC) level and selected the lowest NDC cost per unit at market entry whenever several NDCs were available for the same active ingredient, dosage form, and strength. We selected the unit (tablet, capsule, vial, etc.,) closest to the FDA-recommended strength when a drug had several strengths. We classified the approved drugs in the following therapeutic categories [17]: genetic disorders, HIV and related comorbidities, other infectious diseases, oncology, transplants, and other areas.
We collected each drug's recommended dose and treatment duration from the first FDA approved label. When the FDA-approved label did not indicate the treatment duration, we used the median treatment duration from pivotal clinical trials listed on the label. We assumed an average patient weight of 70 kg and a body surface area of 1.75 m 2 to calculate the daily dose for adult patients, and 40 kg was used to calculate the daily dose for pediatric patients if any adjustment was needed (Appendix A, Table A1).
We calculated the treatment cost for single-use, use for less than one year, and use for one year or longer. We inflated the prices to USD 2021 using consumer price index (CPI) non-seasonally adjusted data for all US city average items and all urban consumers from the US Bureau of Labor Statistics [18].

Data Analysis
We conducted descriptive statistics for each variable included in the analysis. Then, we studied the correlations between the treatment cost of newly approved drugs at market entry and the variables, as well as between the variables themselves. We used the Chisquared test or Fisher's exact test in combination with the Phi-coefficient or Cramer's V considered in cases where both variables were categorical. If both variables were continuous, scatter plots were depicted, and Spearman or Bravais-Pearson correlation coefficients were calculated. We used the point-biserial correlation to check for correlations between categorical and continuous variables. Kruskal-Wallis test is also conducted to check for a significant difference between the means of the ordinal variables' groups (Appendix A, Table A2).

Study Outcome: The Treatment Costs of New Approved Drugs at Market Entry
We used a generalized linear model (GLM) with the Gamma log-link to assess the association between the treatment costs of newly approved drugs at market entry and potential variables: the date of first market entry, application type (New Drug Applications (NDAs), Biologic License Applications (BLAs)), country of incorporation of the sponsor company (US vs. non-US), a binary indicator for first-in-class, a binary indicator for orphan drugs, FDA review type (standard vs. priority), therapeutic intent (diagnosis, prevention, or treatment), therapeutic area (genetic disorders, HIV and related comorbidities, other infectious diseases, oncology, transplants, and other areas), age group (adult, pediatric and adult, or pediatric), and treatment duration (single-use, less than one year, or one year or longer) while addressing the right-skewed distribution of our data. We included all statistically significant variables (p < 0.05) from the bivariate analysis in the GLM. We tested for multicollinearity among independent variables in the GLM using the variance inflation factor (Appendix A, Table A3).
We used the train-test split procedure to estimate our model's performance and prevented the model from overfitting by using root-mean-square error (RMSE). We used two-tailed statistical tests and a p value of 0.05 as the significance threshold. We conducted all analyses using RStudio statistical software (version 4.0.3).

Treatment Cost of New Approved Drugs at Market Entry
The median treatment cost was USD 218,872 for orphan drugs (IQR = USD 231,057, range USD 237-USD 1,272,021) and USD 12,798 for non-orphan drugs (IQR = USD 57,940, range USD 44-USD 382,866, p < 0.001; Figure 1; Appendix A, Table A1).
Compared with non-orphan biologics drugs, the median treatment cost was 4.3 times higher for orphan therapeutic biologics (USD 264,007.88 vs. USD 61,468.75, p < 0.001) and 3.2 times higher for orphan fixed drug combinations (USD 100,177.88 vs. USD 30,895.32, p < 0.001; Table 1).
The median treatment cost was higher for orphan drugs marketed by US companies than for companies from other countries (USD 237,265 vs. USD 128,580, p = 0.005; Table 1; Appendix A, Figure A1).

Treatment Cost of New Approved Drugs at Market Entry
The median treatment cost was USD 218,872 for orphan drugs (IQR = USD 231,057, range USD 237-USD 1,272,021) and USD 12,798 for non-orphan drugs (IQR = USD 57,940, range USD 44-USD 382,866, p < 0.001; Figure 1; Appendix A, Table A1).   However, the difference in the median treatment cost for non-orphan drugs marketed by US companies and those marketed by companies from other countries was insignificant (USD 9483.82 vs. USD 15,834.14, p = 0.262; Table 1).
The median treatment cost for first-in-class approved orphan drugs was not statistically significant relative to the median treatment cost for other orphan drugs (USD 239,593.23 vs. USD 206,176.28, p = 0.322). The median treatment cost for orphan drugs that received a priority review was not significantly different from the cost for orphan drugs with standard reviews (USD 233,934.14 vs. USD 142,195.27, p = 0.053).
Although orphan drugs intended for treatment had a median treatment cost three times higher than drugs for preventive use, the difference was not statistically significant (USD 230,768.11 vs. USD 71,503.98, p = 0.190). The median treatment cost for non-orphan drugs intended for treatment indication was significantly higher than for drugs for the preventive indication (USD 18,486.88 vs. USD 2311.92, p = 0.047).
For the therapeutic areas, we identified a significant difference in the median treatment cost for oncology orphan drugs compared to non-orphan drugs (USD 220,832.30 vs. USD 156,126.94, p = 0.002; Appendix A, Figure A2). Finally, the median treatment cost across patient age groups was significantly higher for orphan drugs targeting both adult and pediatric populations than for non-orphan drugs (USD 280,152.74 vs. USD 1067.40, p < 0.001).

Factors Explaining Treatment Cost of New Approved Drugs at Market Entry
The date of market entry, priority review, and approval as first-in-class drugs were not statistically significantly associated with the mean treatment cost for newly approved drugs at market entry. However, the mean treatment cost at market entry was positively associated with biologics (110%; p < 0.001) and orphan drugs (177%; p < 0.001). Higher market entry treatment costs were also associated with drugs sponsored by US pharmaceutical companies (67%; p = 0.035), drugs intended for treatment rather than prevention (163%; p = 0.004), and treatments with a duration of one year or longer compared to single use (1092%; p < 0.001; Table 2). Among the therapeutic areas, the higher market entry treatment costs were significantly associated with drugs indicated for oncology (698%; p < 0.001) and genetic disorders (608%; p < 0.001) compared to infectious diseases ( Table 2). An RMSE was obtained for each train model and test model (RMSE-train model = 3.0 ≈ REMS-test model = 2.8), indicating a good model fit.

Discussion
This novel study assessing US treatment costs of newly approved drugs at market entry from 2017 to 2021 found that the median treatment cost was 17 times higher for orphan than non-orphan drugs. However, after controlling for the characteristics of the drug, date of market entry, therapeutic class, FDA review designation, country of the sponsor company, therapeutic intent, and treatment duration, the treatment drug cost was 2.8 times higher for orphan than non-orphan drugs.
The median treatment costs for orphan drugs exceeded USD 200,000 at market entry. Over the past 20 years, drug expenditures in the US market have increasingly shifted toward drugs that treat relatively few people [7], and the rapid growth of orphan drug approvals has raised concerns about their pricing and affordability [5]. The high costs of orphan drugs are also associated with large out-of-pocket expenditures [12,13].
The findings that the cost of new drugs is associated with orphan drugs, therapeutic biologics, therapeutic class, therapeutic intent, and long treatment duration align with prior research that found that launch prices of new drugs in the US increased faster for biologics and drugs treating rare diseases [11]. In fact, the financial burden on patients and healthcare payers results in high profits for pharmaceutical companies in marketing orphan drugs, even for a small patient populations [12,19,20].
Previous studies have pointed out that drug development is less costly for orphan than for non-orphan drugs due to smaller and fewer efficacy and safety trials, shorter FDA review time, higher marketing approval success rates, and lower marketing prices [6,12,13,21]. Since rare diseases are often serious or life-threatening, most orphan drugs qualify for designations and regulatory pathways established by Congress to expedite new drug development and FDA review and approvals [22,23]. Our study confirmed that a higher percentage of orphan than non-orphan drugs benefited from FDA-expedited designations and approval pathways.
The pharmaceutical industry has been criticized for high prices and profits from orphan drug incentives in situations that do not meet the Orphan Drug Act's original intent [24,25]. Orphan designations for marketed drugs and the division of diseases into sub-types to apply for multiple orphan designations have also been associated with delays in generics entry [26]. Moreover, results showed that drugs sponsored by US pharmaceutical companies were significantly associated with higher drug treatment costs at market entry than non-US pharmaceutical companies. The differences in prices of drugs at market entry between US and non-US pharmaceutical companies could be explained by different factors than the country of the sponsored company, such as disease severity, additional non-orphan indications, or route of administration.
Previous studies found that the year of market entry was associated with increased drug prices at US market entry [11,27]. However, our analysis showed no statistically significant association between the date of market entry and treatment drug cost, possibly due to the relatively short period evaluated in our study.
To mitigate the high price of drugs for vulnerable populations, Congress created the 340B program in 1992 that requires pharmaceutical manufacturers to provide front-end discounts (typically 30-50%) for outpatient prescription medicines that serve high numbers of uninsured and poor patients [28]. More than 40% of hospitals in the US are eligible to participate in the 340B program [29]. In 2010, The Affordable Care Act (ACA) of 2010 excluded all sales of drugs that obtained orphan drug approval from the discounts offered by the 340B program to safety-net healthcare providers [30]. Manufacturers of frequently utilized drugs, such as the best-selling drug adalimumab, may identify a new use that meets the definition of a rare disease and obtain FDA approval for an orphan drug indication, thus effectively ceasing the provision of 340B discounts for sales of the drug [31,32].
This study evaluated the cost of drug treatment but not the effectiveness of new drugs approved in the US. However, coverage and reimbursement decisions consider both factors (cost and effectiveness). From an economic perspective, orphan drugs should be subject to the same clinical effectiveness, cost-effectiveness, and budget impact analyses as nonorphan drugs [33,34]. However, healthcare organizations and insurers use special criteria when making orphan drug reimbursement decisions [33]. The economic incentive and ethical imperatives remain unresolved for ensuring access to safe, effective, and affordable treatments for patients with rare diseases [35]. Moreover, balancing the economic incentives to develop and market orphan drugs against the overall benefits and improvements in health outcomes remains critically important [36,37].
A potential reason explaining the high cost of orphan drugs is that R&D expenses for orphan drugs must be recouped from a small number of patients, resulting in high drug treatment costs per patient [21].
A previous study found that the prices of an orphan drugs in Europe were higher for conditions with low prevalence [38]. However, another study concluded that the prices of orphan drugs in the US are unlikely to be driven by the prevalence of the target disease [39]. Further studies are needed to associate disease prevalence and drug treatment costs.
The Orphan Drug Act, enacted 40 years ago, has been credited to have an important role in the development and approval of drugs for rare diseases [36]. As a result, there has been a substantial increase in the number of orphan drugs approved by the FDA, providing therapeutic options for patients with unmet medical needs. However, the high cost of these drugs creates significant financial barriers to patient access and highlights the need for a more sustainable and equitable pricing structure to ensure patients' access to affordable treatments.

Limitations
This study used the wholesale acquisition cost (WAC) at market entry as a proxy of the actual acquisition cost by private payers. Companies typically use the WAC to set the initial reference price in the Medicaid outpatient pharmacy, the 340B Drug Pricing Program, and the Federal Supply Schedule programs. As price increases in those programs are limited by the rise in the consumer price index, pharmaceutical companies do not have incentives to reduce the market entry price below the WAC. The Medicare Part B program also used the WAC to set new drugs' initial price. Public and private payers also use the WAC to estimate drug product reimbursement to pharmacies and providers.
The sample included new molecular entities and new therapeutic biologics approved by the FDA in 2017-2021. The study excluded non-therapeutic biologics and approvals of already marketed drugs. However, these exclusions do not affect the validity of the results for the stated period, although future studies on more extensive and inclusive datasets could further extend the validity of our findings. We also did not consider the potential number of users for each drug. Follow-up studies could evaluate the effect of patient population size on drug prices. We used median treatment costs for the FDArecommended dose and treatment duration. Future studies could use doses and treatment durations observed in clinical practices and average prices weighted by the number of users to better estimate the societal impact of high-cost drugs.

Conclusions
Orphan drugs were priced significantly higher than non-orphan drugs at market entry. Higher market entry treatment costs were associated with biologics, orphan status, US sponsor companies, chronic use, therapeutic intent, and indications for oncology or genetic disorders. Future research should assess whether the clinical benefits of orphan drugs justify their high costs.

Data Availability Statement:
The data used to support the findings of this study will be available upon request.

Conflicts of Interest:
The authors declare that they have no competing interests.    Variance inflation factor (VIF), standard error (SE). We checked the multicollinearity of independent variables by using VIF for each independent variable in the set of multiple regression variables. The higher the value of VIF, the higher the correlation between this variable and the rest. If the VIF value is higher than 5, it is usually considered to have a high correlation with other independent variables. However, the value of VIF for each independent variable included in our model was less than 5. Our data show low multicollinearity; it is not severe enough to warrant corrective measures.
Therapeutic Area 1 0.332 Age Group 1 Cramer's V, point-biserial, Kruskal. Test and Spearman correlation coefficients were used. Variance inflation factor (VIF), standard error (SE). We checked the multicollinearity of independent variables by using VIF for each independent variable in the set of multiple regression variables. The higher the value of VIF, the higher the correlation between this variable and the rest. If the VIF value is higher than 5, it is usually considered to have a high correlation with other independent variables. However, the value of VIF for each independent variable included in our model was less than 5. Our data show low multicollinearity; it is not severe enough to warrant corrective measures.