The same $1.5 billion peak-sales drug is worth $14.4 million in the rNPV model in this article at a 15% discount rate. At 8%, it is worth $118.3 million. Nothing about the drug changed. Only one input did.
That spread is the reason sell-side price targets for clinical-stage biotechs differ so much. Analysts look at the same trial data and patent filings. They disagree on a short list of inputs: probability of success, peak sales, years of exclusivity, discount rate, and net cash. This article ranks those inputs by how much they move value. It uses published success-rate data, statutory exclusivity rules, FDA price-erosion data, and three recent acquisitions.
The model in this article is our own. It values a hypothetical Phase II asset. It does not value a real company, and it is not investment advice. Every calculated figure carries the label “analysis” or sits in a table marked as calculated.
The Short Answer: Which Biotech Valuation Metrics Move a Price Target Most?
Four inputs move a biotech price target most: probability of success, years of market exclusivity, the discount rate, and peak sales. In our worked model, each one changes value by more than 50% across a range that real analysts use. Net cash sets the floor under the stock, and it often decides the price when the pipeline is out of favor.
The Six Findings That Matter
BIO found that 7.9% of drug programs that entered Phase I between January 2011 and November 2020 reached approval. [1][2] Other published estimates run from 7.3% to about 14%, depending on source and period. [1][2][5]
An asset that starts Phase II has a 15.1% cumulative chance of approval when we apply BIO’s phase transition rates. (Analysis, from the rates in [1].)
Risk adjustment removes 91% of the unrisked value in our model: $878.6 million before risk, $76.1 million after. (Analysis.)
Moving the discount rate from 8% to 15% cuts the model’s rNPV by 88%. One extra year of exclusivity adds 17% to 18%. (Analysis.)
Patent term extension of five years turns a $11.8 million asset into an $88.7 million asset in our example. (Analysis, using the statutory limits in [11] and [12].)
After six or more generic entrants, generic prices fall by more than 95% from the brand price. [15]
What Is rNPV and Why Do Analysts Use It for Biotech Price Targets?
rNPV means risk-adjusted net present value. It is the standard way to value a drug that has no sales yet. WIPO calls it “the most popular, and therefore de facto valuation method” for biotech assets. [7]
Definition: rNPV in Plain Terms
An rNPV model forecasts the cash flows of one drug. It multiplies each cash flow by the probability that the cash flow happens. Then it discounts the result to today. [7] A development cost in Phase III only occurs if Phase II succeeds, so the model weights that cost by the Phase II success rate. [7]
Evaluate lists the inputs for a pipeline asset: forecast sales, development and commercial costs, launch timing, patent life, probability of success, discount rate, terminal value, and royalty or licensing economics. [6] Evaluate also names three other methods that analysts use next to rNPV: comparable company analysis, precedent transaction analysis, and sum-of-the-parts valuation. [6]
How an rNPV Becomes a Per-Share Price Target
The conversion from asset value to share price is simple arithmetic. We describe it here as a method, not as a claim about any one analyst.
Value each pipeline asset with its own rNPV.
Add the asset values. This is the sum-of-the-parts figure.
Add net cash (cash and securities minus debt). Subtract the present value of unallocated overhead.
Divide by the fully diluted share count.
Two of those lines are not about science. Net cash and share count depend on financing. A company that must raise money will issue shares, and the per-share value of the same pipeline falls.
Step One: Value Each Asset Separately
Each drug has its own patents, its own indications, and its own odds. Pooling them hides the risk. If one asset carries 80% of the value, that asset sets the price target.
Step Two: Treat Net Cash as a Separate Line
Cash is certain. Pipeline value is not. A clean model keeps them apart so a reader can see what the market pays for the pipeline after it subtracts cash. Later sections show why that spread matters.
Probability of Success: How Phase Transition Rates Move Biotech Value
Probability of success is the largest single risk factor in an early-stage model. A Phase II asset has an 84.9% chance of failing before approval when we use BIO’s rates. (Analysis, from [1].)
What the BIO 2011-2020 Data Shows
BIO, Informa Pharma Intelligence, and QLS Advisors measured success rates for programs from January 2011 to November 2020. [2] The table below shows the phase transition rates for all therapeutic areas, as reproduced in a published comparison table. [1]
Of drugs that entered Phase I between January 2011 and November 2020, 7.9% reached approval. For oncology drugs, the figure was 5.3%. Source: BIO analysis, as reported in the NEWDIGS FoCUS Project research brief, Tufts Medical Center, October 9, 2023. [2]
Transition
All therapeutic areas
All oncology
Hematological oncology
Source
Phase I to Phase II
52.0%
48.8%
50.1%
[1]
Phase II to Phase III
28.9%
24.6%
27.8%
[1]
Phase III to NDA/BLA
57.8%
47.7%
60.0%
[1]
NDA/BLA to approval
90.6%
92.0%
90.0%
[1]
Likelihood of approval from Phase I
7.9%
5.3%
7.5%
[1]
The weakest step is Phase II. Only 28.9% of programs in all therapeutic areas passed it. [1] An investor who buys a Phase II company buys the step with the most failures.
Cumulative Probability by Starting Phase (Calculated)
We multiplied BIO’s transition rates from each starting point to approval. The result tells you what odds to put in a model for an asset at each stage.
Asset stage today
Cumulative chance of approval, all areas
Cumulative chance of approval, all oncology
Phase I
7.9%
5.3%
Phase II
15.1%
10.8%
Phase III
52.4%
43.9%
NDA/BLA filed
90.6%
92.0%
Source: analysis. We multiplied the transition rates in [1]. Rounding can move the last digit.
The Phase III row matters most for price targets. A drug that enters Phase III has about a 52% chance, not 90%. A Phase III readout is therefore a coin flip in the all-areas data and worse in oncology. (Analysis, from [1].)
Why Do Success Rate Estimates Differ Between Sources?
Published estimates of the chance of approval from Phase I run from 7.3% to about 14%. The difference comes from period, sample, and method.
Source
Period
Likelihood of approval from Phase I
Citation
BIO / Informa / QLS
Jan 2011 to Nov 2020
7.9%
[1][2]
IQVIA (as tabulated in [1])
2010 to 2023
7.3%
[1]
IQVIA (as tabulated in [2])
2010 to 2022
13.2%
[2]
Wong, Siah and Lo (MIT)
Jan 2000 to Oct 2015
About 14%
[4][5]
Two documents list different IQVIA figures for nearly the same period: 7.3% and 13.2%. [1][2] We could not reconcile them from the documents. A model builder should state which source and which definition the model uses.
How Many Trials Sit Behind the Wong Estimate?
Wong, Siah and Lo analyzed 406,038 clinical trial entries that covered more than 21,143 compounds between January 1, 2000 and October 31, 2015. [4] They published the work in Biostatistics in 2019. [4]
Oncology Success Rates Compared With All Indications
Wong and colleagues put the oncology success rate at 3.4%, the lowest of the disease areas they studied. [4][5] BIO’s later window puts oncology at 5.3%. [2] The two numbers cover different years and use different methods, so they do not conflict. Both say that oncology carries more risk than the average drug.
Do Biomarkers Change the Odds?
Yes. Wong and colleagues found that trials that select patients with biomarkers have higher overall success probabilities than trials that do not. [4] The abstract we reviewed gives no single percentage for the gain.
How Long Does Each Development Phase Take?
Time matters because it eats patent life and delays cash. BIO’s analysis of 9,704 clinical development programs gives these average durations: Phase I, 2.3 years; Phase II, 3.6 years; Phase III, 3.3 years; and Phase III to regulatory approval, 1.3 years. [3] The sum is 10.5 years from Phase I to approval. [3]
Peak Sales: The Input Analysts Miss Most Often
Peak sales set the size of the prize. Forecasts of peak sales are also the least reliable input in the model.
How Often Do Launches Miss Analyst Forecasts?
L.E.K. Consulting tracked 450 new molecular entities. About half of the drugs launched in the last 15 years sold more than 20% below analyst estimates, according to a Fierce Pharma report on the study. [8] When L.E.K. used consensus forecasts from just before launch, 40% of products missed by more than 20%. [8]
The same report says only one-fifth of new medicines reached $1 billion in U.S. sales. [8] More than half did not reach $250 million. [8] We read these figures from a news summary. We did not have the L.E.K. report, so the method behind them is unverified.
What Does a Miss Look Like by Therapeutic Area?
Fierce reports that 38% of oncology launches missed expectations. Immunology and cardiovascular launches missed at a rate of about half, and infectious disease at about 48%. [8] Oncology launches missed least often in that data.
What Peak Sales Does a Drug Need to Break Even in the Model?
In our worked model, the drug breaks even at about $780 million in peak sales. (Analysis. The model is linear in peak sales, so we solved for zero.) Below that figure, the expected value after risk is negative.
That number shows why a $1 billion peak-sales claim is not a safe margin. A 20% miss on a $1 billion forecast puts sales at $800 million, close to break-even in this setup.
Net Price Versus List Price
Peak sales must be net sales. Rebates and discounts reduce the figure. A forecast that starts from list price overstates the cash that arrives.
Market Share and Time to Peak
Peak sales depend on share of a treated population and on how many years the ramp takes. Our model uses a five-year ramp (8%, 25%, 50%, 75%, then 100% of peak). A slower ramp pushes cash past the patent expiry and lowers value.
Patent Life and Exclusivity: The Metric That Sets the End Date of the Cash Flows
Patent life and regulatory exclusivity decide how many years of sales an rNPV model can count at full price. After that date, the model must cut revenue. This is the area where patent data has the most direct effect on a price target.
Definitions: Four Layers of Protection (Original Classification)
We use a four-layer classification. It is our own framework, not an industry standard. It helps separate dates that block competitors from dates that do not.
Layer
What it is
How long it lasts
Source
1. Composition-of-matter patent
A patent that claims the molecule or biologic itself
Ends 20 years from the earliest filing date it claims
[24]
2. Secondary patents
Patents on formulations, methods of use, dosing, or manufacturing
Each ends 20 years from its own earliest filing date, so these often expire later than layer 1 (analysis)
[24]
3. Regulatory exclusivity
FDA blocks approval or filing of competitors for a set period
See the table below
[9][10]
4. Term restoration
Patent term extension for regulatory delay; patent term adjustment for USPTO delay
Up to five years of extension, with a 14-year cap after approval
[11][12][24]
How Long Does FDA Exclusivity Last?
FDA exclusivity runs from approval and does not depend on patents. The table lists the main types.
Exclusivity type
Length
Note
Source
New chemical entity (NCE)
5 years
Can drop to 4 years if an ANDA contains a Paragraph IV certification
[9][10]
New clinical investigation
3 years
Applies to certain changes that need new clinical studies
[9]
Orphan drug
7 years
FDA cannot approve another application for the same drug for the same disease
[9][10]
Pediatric
6 months
Attaches to existing exclusivity; does not stand alone
[9]
GAIN (antibacterial and antifungal)
5 years
Added to certain other exclusivities
[9]
Competitive generic therapy
180 days
For ANDAs only
[9]
Biologic reference product
12 years
FDA cannot accept a biosimilar application for the first 4 years
[10]
First-filer generic
180 days
Runs from first commercial marketing by the first ANDA filer
[10]
How Does Patent Term Extension Work?
Patent term extension restores some of the patent life that FDA review consumes. The statute is 35 U.S.C. 156. Three limits apply.
The extension cannot exceed five years. [11][12]
The total remaining term after approval cannot exceed 14 years. [11][12]
Each product gets one extension, and each patent gets one. [12]
The USPTO’s guidance gives the formula: extension equals the regulatory review period, minus time before the patent issued, minus time without due diligence, minus half of the testing phase (adjusted for time before issue). [11] Wiggin and Dana state the logic more simply: half of the time from the IND’s effective date to the NDA filing, plus the time from NDA submission to approval. [12]
What Does the 14-Year Cap Do?
The cap binds when a patent already has a long remaining life at approval. The USPTO says that if the remaining term plus the extension would exceed 14 years, the extension is reduced so the total equals 14 years. [11] Our table below shows the cap in effect for a patent filed at the start of Phase I.
What Does the Five-Year Cap Do?
The five-year cap binds when development is long. [11] A program with 9.2 years of testing and 1.3 years of review earns 5.9 years by formula, and the statute cuts it to five. (Analysis, using durations from [3].)
The Patent Clock Arithmetic: How Many Years of Protection Remain at Approval?
An average program spends 10.5 years from Phase I to approval. [3] A US patent lasts 20 years from filing. [24] So the filing date, relative to the start of Phase I, decides how much life remains at launch.
The table shows the result for patents filed 0 to 8 years before Phase I begins. We used BIO’s average durations for the clinical timeline. [3] We applied the five-year and 14-year limits from [11] and [12].
Patent filed this many years before Phase I starts
Years of patent term left at approval
Extension granted (years)
Years of protection after approval
0
9.5
4.5
14.0
2
7.5
5.0
12.5
3
6.5
5.0
11.5
4
5.5
5.0
10.5
5
4.5
5.0
9.5
6
3.5
5.0
8.5
8
1.5
5.0
6.5
Source: analysis. Method: remaining term = 20 minus years filed before Phase I minus 10.5. Extension = lesser of 5 years, or 14 minus remaining term. We used the start of Phase I as a stand-in for the IND effective date. We ignored reductions for time before patent issue and for lack of due diligence. These simplifications make the extension look slightly larger than a real calculation.
In this table, a patent filed three years before Phase I leaves 6.5 years at approval. With a five-year extension, it leaves 11.5 years. Early filing of the composition patent therefore matters, but it does not decide the outcome alone, because extension fills part of the gap.
What Is Patent Term Extension Worth in rNPV Terms?
In our worked model, a drug with six years of protected sales after approval has an rNPV of $11.8 million. With 11 years, it has an rNPV of $88.7 million. (Analysis; whole years only.) A five-year extension is therefore worth $76.9 million in this example, or 6.5 times the no-extension value.
Case Study: The Keytruda Patent Clock
Merck’s Keytruda (pembrolizumab) shows how a single date can carry a company’s price target. FDA approved Keytruda in September 2014 for advanced melanoma. [14] Its U.S. compound patent expires in December 2028, and Merck expects to lose European exclusivity in 2031, according to a report that cites Merck’s disclosures. [13]
Merck reported $31.7 billion in sales for Keytruda and Keytruda Qlex in 2025, against $29.5 billion in 2024. [13] Against Merck’s reported 2025 total sales of $65.0 billion, Keytruda was about 49% of revenue. (Analysis, from [13].) That is 7.5% growth for the year, calculated from the same two figures. (Analysis.)
From approval in September 2014 to a December 2028 expiry is about 14.3 years. (Analysis.) From the date of this article to that expiry is about 2.2 years.
What Does DrugPatentWatch Show for Keytruda?
DrugPatentWatch is a patent and exclusivity database for drugs and biologics. Its Keytruda record, last updated October 18, 2026, lists BLA 125514 and states that the 12-year BPCIA reference-product exclusivity has already expired. [14] It names U.S. Patent No. 8,354,509 as central and gives 2028 as the expected U.S. loss-of-exclusivity period. [14]
The same record lists other patents with estimated expiration dates from February 4, 2035 to December 22, 2040. [14] The record separates 2028 from those later dates. An analyst must read the claims to decide which patents block a biosimilar and which do not. A model that uses the latest listed date overstates protection. A model that ignores later patents may understate it. We do not claim to resolve that question here, and the record itself flags the difference between the 2028 narrative and the later table entries. [14]
What Happens to a Price Target When a Blockbuster Nears Loss of Exclusivity?
The rNPV of the marketed product shrinks toward zero as the date nears, so the target depends more on what replaces it. Merck’s July 2025 purchase of Verona Pharma, covered below, came when Keytruda made up about 46% of Merck’s 2024 revenue. [19]
Generic Entry and Biosimilar Launch: How Fast Does Revenue Fall After Loss of Exclusivity?
The more generic firms enter, the lower the price. FDA’s own analysis of drugs with first generic entry from 2015 through 2017 measured the drop. [15]
FDA Data: Generic Price Reduction by Number of Competitors
Generic competitors
Price reduction vs. brand (average manufacturer price)
Price reduction vs. brand (invoice price)
1
39%
31%
2
54%
44%
4
79%
73%
6 or more
More than 95%
More than 95%
Source: Conrad and Lutter, FDA, December 2019. [15] The data covers drugs sold through retail pharmacies. It measures each generic’s price against the brand price in the three months before generic entry. [15]
This data measures generic prices. It does not measure how much brand revenue survives. An analyst must make a separate assumption about brand share loss.
Our Loss-of-Exclusivity Assumption in the Model
Our model sets brand revenue at 25% of peak in the first year after exclusivity ends. It then sets 12%, and 6% for three more years. This is an assumption, not a result from [15]. We chose a steep drop because the FDA data shows that prices fall fast as entrants arrive. A user who models a biologic with slower biosimilar uptake should use a gentler curve.
What Happens If Exclusivity Ends Two Years Sooner?
In our model, cutting protected sales from 10 years to 8 years lowers rNPV from $76.1 million to $47.0 million, a fall of 38%. (Analysis.) Two years of lost exclusivity cost more than a third of the value.
Discount Rate: Why 8% and 15% Give Different Price Targets
The discount rate is the easiest input to change and the hardest to defend. It shifts value more than any single clinical assumption in our sensitivity range.
What Discount Rate Do Biotech Models Use?
WIPO states that large pharmaceutical firms have had a cost of capital of roughly 7% to 9% over the past 15 years, and that small biotech firms face substantially higher costs. [7] WIPO’s own case study of a cystic fibrosis drug uses 10%. [7] A JAMA Network Open commentary on cost estimates reports that a study of drug development costs from 2000 to 2018 used an 11.0% discount rate. [17]
How Much Does the Discount Rate Move Value?
Discount rate
rNPV of the model asset ($ million)
Change vs. 10% base
8%
118.3
+55%
10%
76.1
Base
12%
45.5
-40%
15%
14.4
-81%
Source: analysis. Model assumptions appear in the worked example below.
Does a Model Count Risk Twice if It Uses a High Discount Rate?
It can. WIPO says the discount rate should reflect the organization’s cost of capital and not the specific risk of the drug, because the rNPV method already captures risk in the cash flows and probabilities. [7] WIPO also says that adjusting only the discount rate for project risk hides the effect of risk. [7] A model that uses a 15% rate and also applies a 15.1% chance of approval penalizes the same risk twice, on that logic.
How Much Does It Cost to Develop a Drug, and Why Does It Matter for Valuation?
Development costs matter because the model weights them by the chance the program survives to spend them. They also set a floor for what an acquirer needs to earn back.
Cost Per Approved Drug: Two Published Estimates
Wouters, McKee and Luyten estimated a median R&D cost of $985 million and a mean of $1.3 billion per new drug. [16] They used 63 new drugs and biologics from the 335 approved in the U.S. from 2009 to 2018, based on SEC filings. [16] The London School of Hygiene and Tropical Medicine says the sample is mostly smaller companies and over-represents some disease areas. [16] The study appeared in JAMA on 4 March 2020. [16]
A JAMA Network Open commentary reports a mean of $879.3 million per new drug in 2018 dollars, or about $1.1 billion in 2023 dollars, for the years 2000 to 2018. [17] That figure includes the cost of failed trials and of capital. [17] By therapeutic area, the estimates range from about $378.7 million for anti-infectives to about $1.76 billion for anesthetics and pain drugs. [17]
Why Our Model Uses Lower Costs
Our model spends $285 million on one asset from Phase II through filing and review. That is below the per-approved-drug figures because those figures include failed programs and earlier research. The model does not need them. It already weights each cost by the chance the program reaches that stage.
Worked Example: rNPV of a Hypothetical Phase II Asset
This section shows the full calculation. A reader can rebuild it in a spreadsheet.
Methodology and Assumptions
Data analyzed: BIO phase transition rates as reproduced in [1], average phase durations from [3], the exclusivity and extension rules in [11] and [12], and the generic price data in [15]. Time period: BIO’s rates cover January 2011 to November 2020. [2] Records: the model values one hypothetical asset, “Asset X.” It does not use a real company’s data. Calculations: annual cash flows, discounted at the end of each year, each multiplied by the cumulative probability that the stage is reached.
Parameter
Value
Starting point
Beginning of Phase II
Phase II
Years 1 to 4; $60 million total; probability 100%
Phase III
Years 5 to 7; $210 million total; probability 28.9% (Phase II to III rate)
Filing and review
Year 8; $15 million; probability 28.9% x 57.8% = 16.7%
Commercial cash flows
Years 9 onward; probability 16.7% x 90.6% = 15.1%
Peak net sales
$1.5 billion
Sales ramp (years 1 to 5 after launch)
8%, 25%, 50%, 75%, 100% of peak
Operating margin
20% in launch year 1, 35% in year 2, 45% after
Tax rate
21%
Years of protected sales after approval
10 in base case
Sales after exclusivity ends
25%, 12%, then 6% for three years, of peak; margin 45%
The phase durations are rounded to whole years (Phase II 4 years, Phase III 3 years, review 1 year) for ease of calculation. BIO’s averages are 3.6, 3.3 and 1.3 years. [3]
Base Case Results
Measure
Result
Unrisked NPV
$878.6 million
Cumulative chance of approval from Phase II
15.1%
rNPV
$76.1 million
Share of unrisked value removed by risk adjustment
91.3%
Source: analysis.
Sensitivity Analysis: Which Input Moves rNPV Most?
Probability of Success Sensitivity
Probability of approval applied to sales
rNPV ($ million)
5%
-30.5
10%
22.1
15.1% (BIO, all areas, from Phase II)
75.8
20%
127.3
30%
232.5
Source: analysis. In this version we change only the probability applied to sales, and leave cost weights as in the base case, so the 15.1% row differs from the base rNPV by rounding.
Each percentage point of approval probability adds about $10.5 million. That is about 14% of the base rNPV. (Analysis.) When we replace all-area rates with BIO’s oncology rates, the Phase II asset has a 10.8% chance of approval, and rNPV falls to $36.0 million, a drop of 53%. (Analysis.)
Years of Exclusivity Sensitivity
Protected sales years after approval
rNPV ($ million)
6
11.8
8
47.0
9
62.3
10
76.1
11
88.7
12
100.2
14
120.1
Source: analysis. One more year of exclusivity adds $12.6 million to $13.8 million near the base case, or 17% to 18% of the base rNPV. (Analysis.)
Peak Sales Sensitivity
Peak net sales ($ million)
rNPV ($ million)
750
-3.5
1,000
23.1
1,500
76.1
2,000
129.2
3,000
235.4
Source: analysis. The break-even point is about $780 million.
Limits of the Model
The model is deterministic. It uses one sales path and one margin path. It has no partner, no milestone payments, and no terminal value. It treats success probabilities as independent of market size. In real programs, drugs with larger markets may attract more competition, and that changes both inputs. The model also ignores the financing cost of dilution. Treat the results as a way to rank inputs, not to value a real asset.
Net Cash, Runway and Enterprise Value: The Other Half of a Biotech Price Target
Net cash can decide a price target when investors lose faith in the pipeline. In June 2022, more than 200 public biotechs traded below the cash on their balance sheets, according to Evercore ISI analysis reported by STAT. [18]
What Is Enterprise Value in Biotech?
Enterprise value equals market capitalization plus debt, minus cash and securities. A negative enterprise value means the market values the company at less than its cash. In that case the market assigns the pipeline no value, or assumes the company will burn its cash before it returns anything to shareholders.
How Many Biotechs Have Traded Below Cash?
STAT reported that Evercore ISI identified about 30 companies with an enterprise value $100 million or more below zero. Those companies held about $14 billion in cash against a combined market capitalization of $8 billion. [18] The XBI index was down 40% for the year at that point. [18]
The gap between $14 billion of cash and $8 billion of market value is $6 billion. (Analysis, from [18].) Cash was 1.75 times market capitalization for that group. (Analysis.) A buyer that paid the full market capitalization would receive $14 billion of cash for $8 billion, and the pipelines would cost nothing.
What Is Cash Runway and Why Does It Change the Price Target?
Cash runway is the number of quarters a company can fund operations from current cash at its current burn rate. A short runway forces a financing. Financing at a price below the model value lowers the value per share. The example below shows the effect.
Illustrative Per-Share Calculation With Dilution
Suppose a company owns only Asset X from our model. It has $120 million of net cash, $30 million of present-value overhead, and 50 million diluted shares. This is a hypothetical company.
Step
Before financing
After financing
Asset X rNPV ($ million)
76.1
76.1
Net cash ($ million)
120.0
180.0
Less overhead ($ million)
-30.0
-30.0
Equity value ($ million)
166.1
226.1
Diluted shares (million)
50
70
Value per share
$3.32
$3.23
Source: analysis. The financing sells 20 million shares at $3.00 to raise $60 million. Value per share falls 2.8% because the new shares sell below the $3.32 model value. We ignore fees.
M&A Comparables: What Acquirers Paid for Biotechs in 2025
Precedent transactions give a market check on an rNPV. Three large biotech acquisitions in 2025 give public price data. Each deal used a different structure.
Deal Table: Johnson & Johnson, Merck and Pfizer
Acquirer and target
Announced
Price per share
Headline value
Other documented terms
Source
Johnson & Johnson / Intra-Cellular Therapies
January 13, 2025
$132.00 cash
About $14.6 billion equity value
Release cites Caplyta peak-year sales potential of more than $5 billion, not risk-adjusted, including partner sales
[21]
Merck / Verona Pharma
July 9, 2025
$107 cash per ADS
About $10 billion total transaction value
23% premium to July 8 close; 39% premium to 60-day volume-weighted average price
[19][20]
Pfizer / Metsera
Final terms November 7, 2025
$86.25 ($65.60 cash plus $20.65 contingent value right)
$10 billion, with about $7.6 billion in cash
Novo Nordisk bid $8.5 billion on October 30; Pfizer first agreed to $4.9 billion plus CVRs in September
[22]
BioSpace corrected its headline on the Metsera deal from $9 billion to $10 billion. [22] We use the corrected figure.
What Do the Deal Numbers Imply? (Calculated)
Price Paid Relative to Company-Cited Peak Sales
Johnson & Johnson paid about $14.6 billion. [21] The company’s release cited peak-year potential of more than $5 billion. [21] The ratio is at most 2.9 times peak sales. (Analysis.) The $5 billion figure is a company projection and includes partner sales. This is one data point, not a market rule.
Run-Rate Multiple for Verona
Verona’s Ohtuvayre had $71 million in sales in the first quarter of 2025, up 95% from the prior quarter. [19] Annualized, that is $284 million. The $10 billion deal value is about 35 times that run rate. (Analysis, from [19]. Annualizing one quarter of a launch drug understates future sales.) The price rests on growth, not on current revenue.
What the Premium Says About the Pre-Deal Run
A 23% premium on $107 implies a July 8 close of about $87. A 39% premium to the 60-day average implies that average was about $77. (Analysis, from [19][20]; premiums are rounded.) The stock closed about 13% above its 60-day average before the offer. The market had already priced part of the value.
How Do Contingent Value Rights Change a Deal Price?
A contingent value right pays only if a stated event happens. In the Metsera deal, the CVR was $20.65 of the $86.25 per-share price, or 23.9%. (Analysis, from [22].) Cash was 76.1%. A price target that compares a market price to the headline deal value overstates the certain payment by the CVR amount.
What Size of Deal Do Big Buyers Target?
Merck’s CEO described deals in the range of $1 billion to $15 billion as the company’s focus, according to Fierce Pharma. [19] An rNPV that implies a value far above that range raises a question about who the buyer would be.
Policy Risk: How Medicare Negotiation Timing Affects Exclusivity Assumptions
Medicare price negotiation limits price after a set number of years, whether or not a patent remains. A model for a Medicare-heavy drug needs a price cap date that is earlier than the patent expiry.
When Can a Drug Be Selected for Negotiation?
KFF reports that a small-molecule drug must be at least 7 years past FDA approval, and a biologic at least 11 years past licensure, on the date CMS publishes the selected-drug list. [23] The gap is four years. (Analysis, from [23].) A small molecule therefore reaches the negotiation window four years sooner than a biologic.
The first negotiated prices took effect January 1, 2026 for 10 Part D drugs. [23] Prices for 15 more Part D drugs take effect January 1, 2027. [23] Prices for 15 Part B and Part D drugs take effect January 1, 2028. [23] KFF says that, starting in 2027, up to 20 additional drugs per year will be selected. [23]
How Did the 2025 Law Change Orphan Drug Treatment?
The 2025 reconciliation law widened the orphan drug exclusion in two ways. Drugs designated for several rare diseases are now excluded, and time on the market as an orphan-only drug no longer counts toward the 7-year or 11-year clock for a drug that later wins a non-orphan indication. [23] KFF says the changes delayed selection of Keytruda and Opdivo beyond 2026. [23]
What Does This Mean for a Price Target?
Analysts should add two dates to a model: the patent or exclusivity expiry, and the first date the drug could face a negotiated price. The second date can come first. For an orphan drug with several indications, the clock rules above change when that second date arrives.
How to Read a Biotech Price Target: Six Questions to Ask
Use these questions on any sell-side note. Each one maps to a finding in this article.
Which success-rate source does the note use, and from which phase? Estimates for Phase I range from 7.3% to about 14%. [1][2][5]
Are peak sales net of rebates, and what is the ramp? About half of launches since 2004 missed estimates by more than 20%, per L.E.K. [8]
Which patent date ends the cash flows, and does it include extension? In our model, a five-year extension changes value 6.5-fold. (Analysis.)
What discount rate does the note use, and does the rate also carry clinical risk? WIPO says the rate should reflect cost of capital, not drug risk. [7]
What are the net cash and share count, and at what date? The dilution example above moves value per share by 2.8%. (Analysis.)
Does a precedent deal support the number? Compare the implied value with the three 2025 deals and note CVRs. [19][21][22]
How DrugPatentWatch Data Fits Into a Valuation
DrugPatentWatch supplies the patent and exclusivity dates that feed the exclusivity input in the model. Analysts use that kind of record to find which patents are listed for a drug, which expire first, and which generic or biosimilar firms are active. The Keytruda record discussed above lists patents with estimated expiration dates and labels some as low-certainty, based on patent claims searches. [14]
A database record is a starting point. Check the claims in the patent itself. For a small molecule, check the FDA Orange Book. For a biologic, check the Purple Book. The Keytruda record notes that Keytruda is not in the Orange Book because it is a biologic tracked in the Purple Book. [14] The date that matters is the one a court would enforce against a named challenger.
Key Takeaways
BIO’s data shows 7.9% of drugs that entered Phase I from 2011 to 2020 reached approval. [1][2]
A Phase II asset has a 15.1% cumulative chance of approval on BIO’s rates, and a Phase III asset has 52.4%. (Analysis, from [1].)
Risk adjustment removed 91.3% of unrisked value in our model. (Analysis.)
Each point of approval probability added about $10.5 million to a $76.1 million rNPV. (Analysis.)
Each extra year of exclusivity added 17% to 18% of base value. (Analysis.)
Moving the discount rate from 8% to 15% cut value by 88%. (Analysis.)
Six or more generic entrants push generic prices more than 95% below the brand price. [15]
Three 2025 acquisitions priced at $10 billion to $14.6 billion, with one deal carrying a 23.9% CVR share. [19][21][22]
FAQ
What is the difference between rNPV and a standard DCF for a biotech?
rNPV multiplies each cash flow by the chance it happens. A standard DCF discounts the cash flows without that weighting. WIPO says that a single blended discount rate hides the complexity of drug development, and that rNPV considers a range of future scenarios explicitly. [7]
How many years of exclusivity does a new biologic get?
A reference biologic gets 12 years of exclusivity from first licensure, and FDA cannot accept a biosimilar application for the first 4 years. [10] Patents can run beyond or end before that period. The Keytruda record says its 12-year period has expired. [14]
Does orphan designation add to a drug’s protection?
Orphan drug exclusivity is 7 years, and FDA cannot approve another application for the same drug for the same disease in that period. [9][10] Orphan status also affects Medicare negotiation timing under the 2025 law. [23]
What does a contingent value right add to a takeout price?
It adds a payment that depends on a future event. In the Pfizer and Metsera deal, the CVR was $20.65 of the $86.25 price. [22] A model should value the CVR at its probability, not at face value.
What multiple of peak sales do acquirers pay?
One documented data point: Johnson & Johnson’s $14.6 billion price is at most 2.9 times the company’s $5 billion-plus peak-sales figure for Caplyta. (Analysis, from [21].) The figure is not risk-adjusted and includes partner sales, so a single deal does not set a rule. [21]
How much does the first generic entrant cut the price?
FDA found that one generic competitor was associated with a 39% price reduction against the brand using average manufacturer price data, and 31% using invoice prices. [15] Two competitors gave 54% and 44%. [15]
When does a drug become eligible for Medicare price negotiation?
A small-molecule drug becomes eligible at 7 years past FDA approval. A biologic becomes eligible at 11 years past licensure. [23] CMS measures the clock on the date it publishes the selected-drug list, and it uses the earliest approval if a drug has several. [23]
Do biomarkers raise the chance of approval enough to matter in a model?
Wong, Siah and Lo found that trials that select patients with biomarkers have higher overall success probabilities. [4] In our model, each percentage point of approval probability is worth about $10.5 million of a $76.1 million rNPV. (Analysis.) A few points of gain therefore matter.
Is 7.9% still the right baseline for a new program?
Not always. IQVIA’s 2010 to 2023 data, as tabulated, shows a Phase II to Phase III rate of 36%, against 28.9% in BIO’s 2011 to 2020 data. [1] IQVIA’s Phase I to Phase II rate is lower, at 45% against 52.0%. [1] A model should name its source, period and starting phase.
Where can an analyst check a drug’s exclusivity dates?
Start with FDA’s exclusivity guidance, which lists each exclusivity type and its duration. [9] Small molecules appear in the Orange Book. A biologic such as Keytruda is tracked in the Purple Book instead. [14] Commercial databases such as DrugPatentWatch compile exclusivity data with patent listings, and each record should be checked against the primary source. [14]
References
NEWDIGS FoCUS Project. (2025). Pipeline Analysis Model comparison of likelihood of approval with BIO and IQVIA benchmarks [Supplementary table]. Nature Reviews Drug Discovery, file d41573-025-00036-8. https://media.nature.com/original/magazine-assets/d41573-025-00036-8/50677254
Tufts Medical Center, NEWDIGS FoCUS Project. (2023, October 9). Are cell and gene therapy programs a better bet? [Research brief]. https://newdigs.tuftsmedicalcenter.org/wp-content/uploads/2023/10/NEWDIGS-Success-Rate-Comparison-2023F210v056.pdf
n-side. (2022). What’s the average time to bring a drug to market in 2022? (Citing BIO analysis of 9,704 programs, January 2011 to November 2020). https://www.n-side.com/en/insights/whats-the-average-time-to-bring-a-drug-to-market-in-2022/
Wong, C. H., Siah, K. W., & Lo, A. W. (2019). Estimation of clinical trial success rates and related parameters. Biostatistics, 20(2), 273-286. https://doi.org/10.1093/biostatistics/kxx069. Abstract: https://pubmed.ncbi.nlm.nih.gov/29394327/
MIT Sloan School of Management. (n.d.). Measuring risks and rewards in drug development: New research from MIT shows success rates of clinical trials are higher than previously thought [Press release]. https://mitsloan.mit.edu/press/measuring-risks-and-rewards-drug-development-new-research-mit-shows-success-rates-clinical-trials-are-higher-previously-thought
Evaluate. (2026, June 3; modified 2026, September 17). How to value a biotech. https://www.evaluate.com/blog/how-to-value-a-biotech/
World Intellectual Property Organization. (n.d.). Intellectual property valuation in biotechnology and pharmaceuticals: The income approach. https://www.wipo.int/web-publications/intellectual-property-valuation-in-biotechnology-and-pharmaceuticals/en/3-the-income-approach.html
Weintraub, A. (2021, January 20). Failure to launch? Half of drugs rolled out since 2004 didn’t live up to sales forecasts: Report. Fierce Pharma. https://www.fiercepharma.com/pharma/half-drugs-launched-last-15-years-failed-to-meet-wall-street-s-expectations-report
U.S. Food and Drug Administration. (n.d.). Frequently asked questions on patents and exclusivity. https://www.fda.gov/drugs/development-approval-process-drugs/frequently-asked-questions-patents-and-exclusivity
Congressional Research Service. (2019, May 17). In Focus IF11217 [Table of exclusivity periods]. https://www.everycrsreport.com/files/2019-05-17_IF11217_e7095f4d16c3c07ce7a40e80a5edaac73af5195a.html
U.S. Patent and Trademark Office. (n.d.). Manual of Patent Examining Procedure, section 2758 (35 U.S.C. 156(c)(3) and (g)(6)). https://www.uspto.gov/web/offices/pac/mpep/s2758.html
Wiggin and Dana. (2003, June). Patent term extensions extend revenues. BioInsights, 6. https://www.wiggin.com/publication/bioinsights-june-2003-issue-6-patent-term-extensions-extend-revenues/
OncoDaily. (2026, February 26; modified 2026, March 12). Merck splits oncology business unit as Keytruda patent expiry nears. https://oncodaily.com/technology/merck461936
DrugPatentWatch. (2026). Keytruda (BLA 125514) biologics record [Database record, last updated October 18, 2026; accessed October 18, 2026]. https://www.drugpatentwatch.com/p/biologics/tradename/KEYTRUDA
Conrad, R., & Lutter, R. (2019, December). Generic competition and drug prices: New evidence linking greater generic competition and lower generic drug prices. U.S. Food and Drug Administration. https://fda.gov/media/133509/download
Wouters, O. J., McKee, M., & Luyten, J. (2020). Estimated research and development investment needed to bring a new medicine to market, 2009-2018. JAMA, March 4, 2020. https://doi.org/10.1001/jama.2020.1166. Summary: https://www.lshtm.ac.uk/newsevents/news/2020/average-cost-developing-new-drug-could-be-15-billion-less-pharmaceutical
JAMA Network Open. (2024). Commentary on Sertkaya et al., Costs of drug development and research and development intensity in the US, 2000-2018. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2820566
Garde, D. (2022, June 1). More and more biotechs are worth less than their checking accounts. STAT. https://www.statnews.com/2022/06/01/more-and-more-biotechs-are-worth-less-than-their-checking-accounts/
Fierce Pharma. (2025, July 9). Merck boosts respiratory portfolio with $10B acquisition of Verona and potential COPD blockbuster Ohtuvayre. https://www.fiercepharma.com/pharma/merck-sweetens-its-portfolio-10b-acquisition-verona-and-potential-copd-blockbuster-ohtuvayre
RTW Biotech Opportunities. (2025, July 10). Merck to acquire Verona Pharma [Regulatory news service notice]. https://www.ticker.app/lse/RTW/rns/2025-07-10/Merck-to-acquire-Verona-Pharma/4492Q
Johnson & Johnson. (2025, January 13). Press release on the acquisition of Intra-Cellular Therapies. Business Wire. https://www.businesswire.com/news/home/20250113969534/en/
BioSpace. (2025, November 7; corrected November 8). Pfizer wins Metsera bidding war over Novo with final $10B price tag. https://www.biospace.com/business/pfizer-wins-metsera-bidding-war-over-novo-with-final-9b-price-tag
Cubanski, J., & Neuman, T. (2026, March 11). Key facts about Medicare drug price negotiation. KFF. https://www.kff.org/medicare/key-facts-about-medicare-drug-price-negotiation/
U.S. Patent and Trademark Office. (n.d.). Manual of Patent Examining Procedure, section 2701: Patent term. https://www.uspto.gov/web/offices/pac/mpep/s2701.html
