A worked risk-adjusted NPV build on a Phase 2b asset, from probabilities and risked costs to mid-year discounting, decomposition and the implied rate.
Risk-adjusted net present value, or rNPV, is the standard way to value a drug asset that has not yet reached the market. The idea fits in one sentence: forecast the cash flows the asset would produce if everything worked, weight each one by the probability that the program is still alive when it occurs, and discount the result at a rate that reflects market risk only. The execution is where most models go wrong. This guide walks through the full calculation on VLT-201, an oral small molecule in Phase 2b for moderate-to-severe ulcerative colitis, owned by Veltrana Therapeutics, a fictional company built as a teaching case. Every number below ties to the prefilled workbook, which lands on an rNPV of $71.09 million at January 1, 2027.
Worked in full in The Biotech Valuation Handbook by Julian R. Sterling, with every figure reproduced in live Excel workbooks.Get the DRM-free PDF, $19.99 →
A drug in clinical development carries two very different kinds of risk. The first is technical: the trial may fail, the regulator may refuse the file, the program may be shelved. That risk is specific to the asset, it can be diversified away by an investor holding many programs, and it is best expressed as a probability. The second is market risk: the sensitivity of the asset's future cash flows to the economy and to equity markets. That risk is what the discount rate is for.
rNPV keeps the two apart. Technical risk goes into the probabilities, market risk goes into the rate, and nothing is counted twice. In the Veltrana case the rate is 11 percent, nominal and after tax, applied to probability-weighted cash flows. Practitioners commonly use something in the range of 10 to 15 percent or more for a small biotech, and roughly 8 to 12 percent for large pharma, so 11 percent is a defensible midpoint for a company with a single late-stage asset and no debt. The choice of rate is discussed at length in The Biotech Valuation Handbook, including why a single company-wide rate is usually preferable to one rate per asset.
What rNPV does not capture is managerial flexibility beyond the go or no-go gates already in the tree, the value of tax losses (Veltrana's net operating losses are not valued in the asset rNPVs), or corporate overhead. Those belong to the sum-of-the-parts, not to the asset.
Before any forecast, fix four conventions and write them on the cover sheet of the model.
The mid-year convention matters more than it looks. Discounting full years instead would push every cash flow six months later and shave a few percent off the value of an asset whose cash flows are spread over two decades. Either choice is defensible; mixing them across assets in the same portfolio is not.
VLT-201 reads out its Phase 2b at the end of 2027, with $45.00 million of remaining cost that year. If the readout is positive, Phase 3 runs from 2028 to 2030 at $105.00 million a year. The NDA is filed and reviewed in 2031, with a filing cost of $12.00 million plus $55.00 million of pre-launch commercial spend. The US launch follows in 2032 and the ex-US licensee launches in 2033.
Each gate carries a conditional probability: 35 percent from Phase 2 to Phase 3, 62 percent from Phase 3 to NDA, and 90 percent from NDA to approval. Multiplying them gives the cumulative likelihood of approval from today: 0.35 x 0.62 x 0.90 = 0.1953, or 19.53 percent. Two intermediate probabilities are needed as well: the probability of reaching Phase 3 is 35 percent, and the probability of filing is 21.7 percent (0.35 x 0.62).
Where those three rates come from, and how they compare with published benchmarks, is the subject of the companion article on probability of success by clinical phase. For the build itself, what matters is that each cash flow gets the probability of the event that must happen before it is incurred.
Costs are risked exactly like revenues. A dollar of Phase 3 spending is only spent if the Phase 2b succeeds, so it carries the 35 percent probability of reaching Phase 3. The 2027 Phase 2b cost carries 100 percent, because it will be spent whatever the result.
| Year | Risked CF ($m) | Discount factor | PV ($m) |
|---|---|---|---|
| 2027 | -45.00 | 0.9492 | -42.71 |
| 2028 | -27.07 | 0.8551 | -23.15 |
| 2029 | -36.75 | 0.7704 | -28.31 |
| 2030 | -36.75 | 0.6940 | -25.51 |
| 2031 | -14.54 | 0.6252 | -9.09 |
The 2029 and 2030 figures are simply 105.00 x 0.35 = 36.75. The 2028 figure is less negative because the $35.00 million Phase 3 start milestone from the ex-US licensee, Ostrander Pharma AG, arrives the same year, taxed at 21 percent and also weighted at 35 percent. In 2031, the $12.00 million filing cost and the $55.00 million pre-launch spend are weighted at the 21.7 percent probability of filing, which gives -14.54.
This is where risking changes the picture most. The near-term cost is almost certain and weighs fully; the larger, later costs are cut twice, once by time and once by the chance that the program dies before they are incurred.
Peak US net sales are built as a funnel rather than taken from a market report. VLT-201 starts from 320,000 addressable US patients with moderate-to-severe disease eligible for advanced therapy. Half of them, 50 percent, are treated with an advanced therapy. VLT-201 takes a peak share of 10 percent of treated patients. A persistence and compliance factor of 75 percent converts patients into paid patient-years. The net price is $54,000 per patient-year, from a list price of $81,000 and a gross-to-net discount of 33.3 percent.
Multiplying through: 320,000 x 0.50 x 0.10 x 0.75 x $54,000 = $648.00 million of peak US net sales, or 12,000 patient-years. The launch curve then reaches 12 percent of peak in 2032, 32 percent in 2033, 55 percent in 2034, 75 percent in 2035, 90 percent in 2036 and 100 percent from 2037, the sixth year on the market.
Two policy and patent events bend the curve. Under the Inflation Reduction Act, a small molecule approved in 2031 becomes eligible for selection seven years later, and the negotiated price is assumed to apply from 2041. With a Medicare share of 25 percent and a maximum fair price 45 percent below the prior net price, US net sales fall by 11.25 percent (0.25 x 0.45), to $575.10 million from 2041. The composition-of-matter patent, with term extension, expires at the end of 2043; generics bring US sales to 25 percent of the 2043 level in 2044 and 10 percent from 2045.
Each commercial year is converted into an after-tax cash flow before any probability is applied. In 2037, the first peak year, US net sales of $648.00 million carry COGS of $32.40 million, selling and marketing of $149.80 million and medical affairs of $25.00 million, for operating income of $440.80 million. Tax of $92.57 million leaves an after-tax commercial cash flow of $348.23 million.
That figure is then multiplied by the 19.53 percent likelihood of approval: 348.23 x 0.1953 gives about 68.0. Adding the ex-US royalty of $41.22 million and the $40.00 million sales milestone due that year, both taxed at 21 percent and weighted at the same 19.53 percent, brings the 2037 risked cash flow to $80.54 million. With the 2037 discount factor of 0.3343, its present value is $26.92 million, the largest single year in the schedule.
| Year | US sales ($m) | Risked CF ($m) | PV ($m) |
|---|---|---|---|
| 2032 | 77.76 | -8.57 | -4.83 |
| 2033 | 207.36 | 18.64 | 9.46 |
| 2035 | 486.00 | 50.26 | 20.70 |
| 2037 | 648.00 | 80.54 | 26.92 |
| 2040 | 648.00 | 75.31 | 18.41 |
| 2041 | 575.10 | 65.75 | 14.48 |
| 2044 | 143.78 | 24.95 | 4.02 |
| 2050 | 57.51 | 6.26 | 0.54 |
The launch year 2032 still loses money, because the fixed commercial base of $85.00 million a year plus medical affairs exceeds first-year gross margin. From 2038 to 2040 the risked cash flow is flat at $75.31 million while the present value slides from $22.68 million to $18.41 million: that slide is discounting and nothing else. The tail after generic entry adds very little, which is normal for a small molecule with steep erosion.
Adding the present values from 2027 to 2050 gives the rNPV of VLT-201 at January 1, 2027: $71.09 million. On its own the total says little. Split by source, it tells the story of the asset.
| Component | PV | Probability applied |
|---|---|---|
| Risked development costs | -$137.04 million | 100, 35 and 21.7 percent by stage |
| Risked US commercial cash flows | $175.22 million | 19.53 percent |
| Risked licence income | $32.92 million | 35 and 19.53 percent |
| rNPV | $71.09 million | Net value today |
The decomposition shows that the risked commercial value is not much larger than the risked cost of reaching the market, and that the partner's milestones and royalties make up a meaningful share of the net. A positive rNPV that is the residual of two large numbers of opposite sign will swing hard in percentage terms. In the case, moving the peak share between 8 and 12 percent moves the rNPV between $21.37 million and $120.81 million; moving the discount rate between 9 and 13 percent moves it between $114.57 million and $37.79 million.
The value also jumps on data. If the Phase 2b is positive, the Phase 2 cost is sunk, the probability of success through approval becomes 55.8 percent (0.62 x 0.90) and the value at January 1, 2028 rises to $360.92 million. That readout is the main value inflection for the asset and, in practice, for the company.
Set every probability to 100 percent and discount the same cash flows at 11 percent, and VLT-201 is worth $718.85 million, roughly ten times its rNPV. The discount rate that brings that unrisked NPV down to $71.09 million is 23.4 percent. That is the rate an analyst using the old risk-adjusted discount rate shortcut would need to reach the same answer for this asset, at this date.
The implied rate is worth reporting, because investors who think in hurdle rates can relate to it. It should never be reused as an input. It belongs to one asset at one stage; it falls sharply after a positive readout. A high rate also crushes distant cash flows most, when the clinical risk is concentrated in the next three to four years, and it shrinks development costs in present value for the wrong reason. rNPV reduces late costs because they may never be incurred, which is the right reason.
The free rNPV drug asset valuation model is an Excel workbook (rNPV_Drug_Asset_Model.xlsx) prefilled with VLT-201. Each input in this article sits in a labeled cell, the annual schedule is built with live formulas, and the decomposition and the implied rate recalculate as you change any assumption. Replace the probabilities, the patient funnel and the timeline with your own asset and the structure carries over. The full method, with chapters on peak sales, exclusivity, the IRA, sensitivities and the sum-of-the-parts, is in The Biotech Valuation Handbook.
Use a rate that reflects market risk only, because technical risk is already in the probabilities. Practitioners commonly use roughly 8 to 12 percent for large pharma and 10 to 15 percent or more for small biotechs. The worked example uses 11 percent, nominal and after tax. Moving it to 9 or 13 percent changes the VLT-201 rNPV to $114.57 million or $37.79 million.
Yes. A cost is only incurred if the program reaches the stage that requires it. In the example, each $105.00 million year of Phase 3 is weighted at the 35 percent probability of reaching Phase 3, while the 2027 Phase 2b cost is weighted at 100 percent because it will be spent whatever the readout.
Because it assumes every gate is passed. For VLT-201 the unrisked NPV is $718.85 million against an rNPV of $71.09 million, a gap explained almost entirely by the 19.53 percent cumulative likelihood of approval. The implied discount rate that closes the gap, 23.4 percent, is an output to report, not an input to reuse.
Yes. The rNPV drug asset valuation model is a free download, prefilled with VLT-201, with every input in a labeled cell. Change the probabilities, timeline, patient funnel, price and exclusivity dates to value your own asset; the schedule, decomposition and implied rate update automatically.
This guide is one arc of The Biotech Valuation Handbook. The book values a whole clinical-stage company: probabilities of success, development costs, a patient-based peak sales forecast, gross-to-net, launch curves, exclusivity and the IRA, the asset cash flow, the discount rate, sensitivities, two licensing deals and the sum-of-the-parts.
Paperback PDF edition, $19.99 Book page and companion workbooks
Free download: rNPV Drug Asset Valuation Model (Excel).
Illustrative figures computed by the book's model; the company, its products and every dollar figure are fictional. Rules and benchmarks as read in November 2026; education only, not investment advice.