1587MO - How much does it cost to research and develop a new medicine? A systematic review and evaluation of 40 years of literature

Background

The biopharmaceutical industry faces challenges related to its research & development (R&D) productivity. At the same time, some pricing policies for new products, particularly those concerning anticancer and orphan medicinal products, have been perceived as non-transparent or even outrageous, resulting in increased resistance by policy makers to reimburse them. In this context, a controversial theme has been the cost of bringing a new molecular entity (NME) to market. We reviewed, and critically assessed, the studies providing estimates of the pre-launch research & development (R&D) cost per NME.

Methods

A full systematic literature review of publications estimating the (pre-launch) R&D costs was conducted. 22 articles with 45 cost estimates were included (three focus on oncology and 16 include cancer alongside other therapeutic areas). We appraised their quality by evaluating 16 factors covering three domains: (1) how the drug samples, success rates, and development times used for cost estimation were obtained; (2) potential sources attributing to the variation in R&D costs; and (3) the cost components.

Results

Estimates of the total average capitalized R&D costs vary widely, from $161million to $4,539 million (2019 USD), with cancer drugs marking the top. We found evidence that the magnitude of these estimates has increased over time, but it is not related to study quality. In addition, average costs mask important differences, e.g. estimations suggested positive skewness for oncological drugs, with an average capitalized R&D cost between $944 and $4,539 million, while a median between $788 and $2,818 million (2019 USD). “Potential sources of variation” was the domain that shows the lowest quality scores.

Conclusions

Due to the heterogeneity of the methodologies and the variability (e.g., by therapeutic area) of the results, caution must be exercised when applying the estimated R&D cost averages. Given the variability of pre-launch drug R&D cost estimates, a standardized framework specifying the factors that ought to be considered in cost estimation seems warranted, and we propose one such here.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

German Cancer Research Center (DKFZ).

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.