Getting a new drug through FDA approval costs roughly $879 million on average when you factor in the cost of failed attempts and the time value of money. That figure, based on a 2020 analysis by HHS using 2018 dollars, sits at the lower end of published estimates. The range is enormous: as low as $379 million for anti-infective drugs and as high as $1.76 billion for pain and anesthesia medications. What you’ll actually spend depends on whether you’re developing a brand-new drug, a generic, a biologic, or a medical device.
The $879 Million Figure, Explained
That headline number deserves some unpacking, because it includes far more than what any single successful drug costs to develop. The mean out-of-pocket cost to develop one new drug is about $173 million in 2018 dollars, including post-marketing studies. That covers everything from early lab work through clinical trials and FDA submission for drugs that actually make it to market.
But most drug candidates fail. When you spread the cost of all those failures across the few drugs that succeed, the average jumps to $516 million. Add in the opportunity cost of capital (the returns a company could have earned by investing that money elsewhere during 10 to 15 years of development), and you reach the $879 million capitalized figure. The 95% confidence interval runs from $417 million to $1.3 billion, which reflects how wildly costs vary by disease area, trial complexity, and company size.
What Clinical Trials Actually Cost
Clinical trials consume the bulk of the development budget, and Phase III trials are where the real money goes. The median cost of a single pivotal Phase III trial is $19 million, based on an analysis of 138 trials behind 59 drugs the FDA approved in 2015 and 2016. But that median masks staggering variation.
A Phase III trial for a rare metabolic disorder with just four patients cost roughly $2 million. A massive cardiovascular trial run by Novartis with over 8,400 patients cost an estimated $347 million. The main drivers are patient enrollment numbers, trial duration, the number of clinical sites, and how hard the endpoints are to measure. A cancer trial requiring years of survival follow-up across dozens of hospitals will cost orders of magnitude more than a short trial for a skin condition at a handful of clinics.
Phase I trials (small safety studies in healthy volunteers) and Phase II trials (early efficacy testing) are cheaper individually, but many candidates go through multiple rounds of each before reaching Phase III, and most never make it that far.
FDA User Fees for Drugs
Beyond development costs, the FDA itself charges substantial application fees. For fiscal year 2026 (effective October 2025), the fee to submit a new drug application with clinical data is $4,682,003. If your application doesn’t require clinical data, the fee drops to $2,341,002. On top of that, companies pay an annual prescription drug program fee of $442,213.
These fees fund the FDA’s drug review operations and increase most years. For a large pharmaceutical company spending hundreds of millions on trials, the application fee is a small fraction of total costs. For a small biotech bringing its first product to market, it can be a significant line item.
Generic Drugs Cost Far Less
If you’re filing for a generic version of an existing drug, the economics look completely different. Generic manufacturers don’t need to repeat the expensive clinical trials that proved the original drug works. Instead, they submit an abbreviated application showing their version is bioequivalent to the brand-name product.
The application fee for a generic drug is $358,247 for fiscal year 2026 (up from $321,920 in FY 2025). Total development costs for generics typically run in the low single-digit millions, a fraction of what a novel drug requires. The main expenses are formulation work, bioequivalence studies, and manufacturing setup.
Medical Device Approval Costs
Medical devices follow a separate regulatory path, and costs vary dramatically based on the device’s risk classification. Most moderate-risk devices (Class II) go through a streamlined process called 510(k) clearance, which requires showing the device is substantially similar to one already on the market. The FDA user fee for a standard 510(k) submission is $26,067 for FY 2026. Total development and testing costs for these devices typically range from a few hundred thousand dollars to a few million.
High-risk devices (Class III), like implantable heart valves or certain diagnostic systems, need premarket approval, which requires clinical evidence of safety and effectiveness. The FDA user fee alone for this pathway is $579,272 for FY 2026, and total development costs can run into the tens or hundreds of millions depending on trial requirements.
The costs don’t necessarily end at approval. The FDA mandates post-approval studies for many devices, and these carry a median cost of $2.16 million per study. Between 2005 and 2013, device manufacturers collectively spent an estimated $1.22 billion on these required follow-up studies.
Discounts for Small Companies
The FDA offers reduced fees for smaller companies, particularly in the device space. Businesses with gross receipts under $100 million (including affiliates) qualify for reduced user fees on several submission types, including premarket approval applications. Companies with receipts under $30 million can get their first premarket application fee waived entirely.
For drugs targeting rare diseases, the Orphan Drug Act provides additional financial relief. Companies developing orphan drugs can receive FDA fee waivers, a 50% tax credit on clinical testing expenses, grants for research, and seven years of market exclusivity after approval. These incentives exist because the small patient populations for rare diseases make it difficult to recoup development costs through sales alone.
Why Estimates Vary So Widely
You’ll see FDA approval cost estimates ranging from under $100 million to over $2 billion depending on the source. The differences come down to methodology. Some figures count only out-of-pocket spending on successful drugs. Others fold in the cost of every failed candidate a company tested along the way. The highest published estimates also include opportunity costs and use data heavily weighted toward large pharmaceutical companies developing complex therapies.
Therapeutic area is the single biggest variable. Developing a new pain medication costs roughly five times more than developing a new anti-infective, largely because pain trials require more patients, longer follow-up, and more sophisticated outcome measures. Oncology and cardiovascular drugs also tend toward the expensive end, while drugs for rare diseases with clear biological targets and small trial populations can cost a fraction of the average.
Company size matters too. A small biotech licensing a promising compound from an academic lab and running a focused development program will spend far less than a large pharma company running parallel programs across multiple indications. The $879 million average reflects the full spectrum, but individual experiences land all over the map.