Pricing the big three financial risks: clinical, regulatory and reimbursement

The three cliffs, in order:

1. Clinical: the Phase III readout (does the drug work?)

2. Regulatory: FDA or EMA approval (will the agency allow it to be sold?)

3. Reimbursement: coverage and price (will payers actually pay for it?)

Each cliff has a probability and a value-if-failed. That is all you need to price the risk.

Cliff one: the Phase III readout

Phase III is the large, confirmatory human trial that regulators require before approval. In oncology it is brutal.

The most cited benchmark comes from the BIO / Informa / QLS Clinical Development Success Rates study. Across 2011 to 2020, oncology had among the lowest probabilities of success of any therapeutic area. The commonly quoted figure for Phase III to approval in oncology sits around 35 to 40 percent (estimate, as of that dataset). Phase III success alone (hitting the primary endpoint) is higher, but readouts still fail often.

Let us say OncoX has a 60 percent probability of a positive Phase III readout.

Pricing the clinical cliff

The valuation swing is not symmetric. On success, the asset gains regulatory-stage value. On failure, a single-asset oncology company typically collapses toward cash value minus wind-down costs.

Worked calculation:

• Value if readout positive (moves to filing): $650 million
• Value if readout fails: $120 million (residual cash, other programs, breakup value)
• Probability of success: 60 percent

Expected value going into the readout:

The value at risk from the clinical event is the drop on failure:

That $530 million evaporates in seconds. This is why single-asset biotechs trade at a steep discount to their rNPV before a readout: the market is holding a probability-weighted number, not the success case.

Cliff two: the regulatory decision (the CRL)

Assume OncoX reads out positive and files. Now the FDA (US Food and Drug Administration) or EMA (European Medicines Agency) reviews it.

The feared outcome is a Complete Response Letter (CRL): an FDA letter stating the application will not be approved in its current form. A CRL is not always fatal (it can cite manufacturing or data-package fixes), but it means delay, extra cost, and sometimes a required new trial.

Historical approval rates for drugs that reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.Voir la définition complète → filing are high (often quoted around 85 to 90 percent for first-cycle-plus-resubmission, estimate). But a CRL still costs time. Time is cash burn plus lost patent-protected sales years.

Pricing the CRL

Say OncoX has a 90 percent probability of approval and a 10 percent chance of a CRL that delays launch by two years.

Cost of a two-year delay:

• Extra cash burn: 2 years x $60 million = $120 million
• Lost exclusivity value: peak sales pushed later, patent clock unchanged. Assume this erodes rNPV by roughly $150 million (fewer protected years at peak).

Probability-weighted regulatory drag:

Smaller than the clinical cliff, but note the asymmetry: the base case (approval) is priced in, so a CRL is a nasty negative surprise, not a symmetric bet.

🎬 [VIDEO: "What is a Complete Response Letter (CRL)?" - youtube.com - a short plain-English explainer on FDA CRLs and what they mean for a drug's path to market]

Cliff three: reimbursement

Approval means you can sell the drug. It does not mean anyone pays your asking price. This is the cliff finance people underweight most.

In the US, CMS (Centers for Medicare & Medicaid Services) sets Medicare coverage and, since the Inflation Reduction Act (IRA) of 2022, negotiates prices for selected high-spend drugs after a set period on market. In England, NICE (National Institute for Health and Care Excellence) runs cost-effectiveness appraisals; a negative NICE recommendation can effectively block NHS reimbursement, shrinking your UK revenue to near zero.

NICE uses a cost per QALY threshold. A QALY (quality-adjusted life year) is one year of life in perfect health. NICE's standard range is roughly £20,000 to £30,000 per QALY (higher for end-of-life and some cancer drugs; estimate, historically stated). If your drug exceeds the threshold, you either cut price or get rejected.

Pricing the reimbursement cliff

The reimbursement risk is rarely all-or-nothing. It usually hits price, which flows straight to peak sales and therefore rNPV.

Suppose OncoX's $1.2 billion peak sales assumed a certain price. Now model payer scenarios:

| Scenario | Probability | Peak sales | Asset value |

|---|---|---|---|

| Full price, broad coverage | 45% | $1.2B | $900M |

| Price cut / restricted label | 40% | $0.8B | $600M |

| Negative NICE + IRA pressure | 15% | $0.5B | $375M |

Expected value:

Compared with the $900 million best case, reimbursement risk alone shaves off about $199 million of expected value. That is larger than the CRL drag and comparable in importance to clinical risk, yet it sits furthest from the science and often gets the least analytical attention in a diligence.

Stacking the cliffs: the compound survival math

The mistake is pricing each cliff in isolation. To value the asset today, you multiply the survival probabilities.

Using our numbers:

• Clinical success: 60%
• Regulatory success: 90%
• Reimbursement (full or partial, weighted): use the expected reimbursement value directly

Probability of reaching the market at all:

0.60 x 0.90 = 0.54 (54% chance the drug is ever sold)

Then apply the reimbursement-weighted value of $701 million to that path, and the failure branches to their residual values. A simplified today value:

Value ~= (0.54 x 701) + (0.46 x ~120)
      ~= 379 + 55
      ~= $434 million

The company that dreams of $900 million is, on a probability-weighted basis, worth closer to $434 million today. The gap of roughly $466 million is the aggregate price of the three risks. Each dollar of that gap is a specific, nameable event.

The due-diligence checklist

When you assess a single-asset or lead-asset biotech, price each cliff explicitly:

• Clinical: What is the phase-specific probability of success for this indication (not the industry average)? Is the endpoint one the FDA has accepted before?
• Regulatory: Is there precedent for this mechanism? Any prior CRLs? Is the manufacturing (CMC) package clean? CMC issues are a common, underestimated CRL trigger.
• Reimbursement: What price does the rNPV assume, and does that price survive a NICE cost-per-QALY test and IRA negotiation exposure? Is peak sales built on list price or net price?
• Cash runway: Does cash reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.Voir la définition complète → past the next binary event? A company that must raise capital before its readout will dilute or die. Runway is a risk multiplier.

Key takeaways

• Value each binary event separately, then multiply survival probabilities. OncoX's compound survival to market was only 54 percent, cutting a $900 million dream to roughly $434 million today.
• Clinical is the biggest single cliff in oncology. A failed Phase III can wipe out $500 million-plus in seconds; oncology Phase III to approval sits around 35 to 40 percent (estimate, BIO/Informa/QLS data).
• Reimbursement risk rivals clinical risk in dollar terms and gets the least attention. A negative NICE decision or IRA price pressure attacks peak sales directly, and in our model erased about $199 million of expected value.
• Cash runway is a risk multiplier. If money runs out before the next readout, dilution or failure is forced regardless of the science.

This lesson is educational and not investment advice. All company figures are illustrative; benchmark ranges are estimates tied to their cited sources.