← all comparisons

Epidemiology-Based Market Sizing — model comparison

Size the US addressable patient population and rough peak revenue for a DLL3 T-cell engager (tarlatamab-like) in second-line extensive-stage small cell lung cancer.

Judged 1–5 on the expert rubric. The deterministic answer key is the same one on the environment page.

Claude Opus 4.8 frontier
anthropic · 4 tool calls
3.8
funnel_construction5
eligibility_gating2
numerical_consistency4
assumptions_revenue4
faithfulness4
flawed — The funnel construction is clean and correctly uses incidence as the basis for the queried US geography, matching the reference through step 6 (8,720). However, the agent double-counts the line gate by applying an extra 0.50 '2L line split'
open full trajectory →
final answer

Now I have everything. Let me build the funnel from incidence.

Patient Funnel — US, 2L ES-SCLC, DLL3 TCE

The dataset gives SCLC total incidence; I gate down to the eligible 2L DLL3+ pool.

Step Fraction Patients/yr
SCLC incidence (US, new/yr) 33,000
× Diagnosed 0.95 31,350
× Treated 0.85 26,648
× Extensive-stage 0.70 18,653
× DLL3-expressing 0.85 15,855
× Reach/fit for 2L therapy 0.55 8,720
× 2L line split 0.50 ~4,360

US addressable 2L ES-SCLC DLL3+ population ≈ 4,300–4,400 new patients/year.

A note on the two "2L-ish" gates: fit_for_2L_fraction (0.55) captures whether a patient is well enough to receive any second-line therapy, while the 2L line split (0.50) captures the share actually reaching/electing 2L treatment. I applied both since they describe distinct attrition (clinical fitness vs. line progression). If you view them as overlapping, the pool is larger — see sensitivity below.

Rough Peak Revenue

Effective revenue per addressable patient/year = price × persistence:
- $220,000 × 0.55 = $121,000 per patient-year

(The 8-month treated duration is already baked into the 0.55 persistence factor — ~8/12 ≈ 0.67, further haircut for early discontinuation/mortality, so I don't double-count it.)

Peak revenue, fully-captured addressable pool:
- 4,360 patients × $121,000 ≈ $528M

This represents 100% of the addressable pool — i.e., the theoretical ceiling, not a realistic share. Applying a more realistic peak share for a branded 2L agent (with competition):

Peak market share Peak US revenue
100% (full addressable ceiling) ~$530M
60% ~$315M
40% ~$210M

Sensitivity on the funnel

  • If the two 2L gates overlap (use only the 0.50 line split, drop the 0.55 fitness gate): addressable ≈ 7,900/yr → full-pool revenue ~$960M; at 50% share ~$480M.
  • Realistic central estimate: US addressable ~4,000–8,000 new patients/yr, peak revenue plausibly $300–500M at a credible (~40–60%) peak share.

Key caveats

  • Teaching/curated data — rounded figures, not a live or citable pull.
  • Incidence-based, so this is an annual flow, not a point prevalence. Because SCLC is so aggressive (short duration, high mortality), incident flow is the right basis and treated patients don't accumulate much.
  • Biggest swing factors: the 2L reach assumptions (the 0.55×0.50 stack), DLL3 testing/expression cutoffs in practice, and real-world persistence. The line-split interpretation alone moves the answer ~2×.
  • Not included: EU5 (available if you want ex-US), 1L/3L+ expansion potential (a 1L move would dramatically enlarge the pool), or pricing erosion from competition.

Want me to run the EU5 geography, or model a 1L expansion scenario for upside?

GPT (frontier) frontier
openai · 3 tool calls
3.6
funnel_construction4
eligibility_gating3
numerical_consistency4
assumptions_revenue3
faithfulness4
acceptable — The agent correctly used incidence as the funnel basis and built a clean population->treated->eligible chain, landing at ~8,000 addressable patients close to the reference 8,720. However, it chose to apply the explicit 50% 2L rate instead o
open full trajectory →
final answer

Below is a transparent US funnel and rough peak revenue sketch for a DLL3 T‑cell engager (tarlatamab‑like) in 2L extensive‑stage SCLC (ES‑SCLC) using the curated dataset.


1️⃣ Start with incidence (funnel basis = incidence)

US ES‑SCLC incidence (annual new pts):
- SCLC annual incidence: 33,000
- Extensive‑stage fraction: 70%
23,100 ES‑SCLC / year


2️⃣ Diagnosed and treated in 1L

  • Diagnosis rate: 95% → 21,945
  • Treatment rate (receive systemic therapy): 85%
    18,653 treated ES‑SCLC (1L)

3️⃣ Reach second line

Two relevant gates provided:

  • Proportion who receive 2L: 50%
  • “Fit for 2L” fraction: 55%

These are conceptually overlapping; using the explicit 2L rate (50%) for the funnel:

→ 18,653 × 50% = 9,327 2L patients


4️⃣ DLL3 expression eligibility

  • DLL3+ fraction: 85%

→ 9,327 × 85% = 7,928 eligible patients/year


✅ US Addressable Population (2L ES‑SCLC, DLL3+)

8,000 patients per year

(Rounded for peak planning purposes.)


5️⃣ Revenue Assumptions

  • Annual price: $220,000
  • Persistence factor: 0.55
  • (reflects short duration due to progression/mortality)

Effective revenue per treated patient:

[
220,000 \times 0.55 = 121,000
]

$121K per patient


6️⃣ Peak Revenue Estimate

Peak share assumption is the major swing factor.

Given high unmet need in 2L ES‑SCLC and limited competition:

  • Conservative peak share: 50%
  • Upside scenario: 70%

🔹 Base Case (50% peak share)

Treated patients:
[
8,000 \times 50\% = 4,000
]

Revenue:
[
4,000 \times 121,000 ≈ \$484M
]

~$0.5B US peak revenue


🔹 Upside Case (70% peak share)

[
8,000 \times 70\% = 5,600
]

[
5,600 \times 121,000 ≈ \$678M
]

~$0.7B US peak revenue


🎯 Final Answer

Metric Estimate
US addressable 2L DLL3+ ES‑SCLC patients ~8,000/year
Effective revenue per patient ~$121K
Peak US revenue $0.5–0.7B

If you'd like, I can:
- Layer in ex‑US (EU5)
- Model impact of moving to 1L
- Add probability‑adjusted revenue
- Compare to ADC competition scenarios
- Or convert this into a quick investor-style valuation framework

Claude Haiku 4.5 small
anthropic · 4 tool calls
2.6
funnel_construction3
eligibility_gating2
numerical_consistency3
assumptions_revenue2
faithfulness3
None — The agent uses the correct incidence basis (33,000) and proper diagnosis/treatment/DLL3/fit fractions, but it OMITS the extensive-stage 70% gate entirely, inflating the funnel; this is a missing required gate. Its addressable number (6,236)
open full trajectory →
final answer

Perfect! Now let me build the addressable patient population funnel and calculate peak revenue.

US Addressable Population & Peak Revenue: DLL3 T-cell Engager in 2L ES-SCLC

Funnel (Addressable Patient Population)

Stage Count Notes
US ES-SCLC Incidence 33,000 Annual new cases (incidence-based funnel)
Diagnosed & Treated 26,675 × 95% diagnosis rate × 85% treatment rate
DLL3 Expressed 22,674 × 85% DLL3+ prevalence
Fit for 2L 12,471 × 55% clinically fit for second-line
Progress to 2L 6,236 × 50% actually reach second-line (line split)
Addressable 2L Population 6,236 patients/year Ready for DLL3 T-cell engager

Peak Revenue Estimate

Metric Value
Addressable patients/year 6,236
Annual price (list) $220,000
Treatment persistence 0.55 (55% of year paid)
Effective price/patient ~$121,000 (= $220k × 0.55)
Peak Annual Revenue ~$755 million

Key Drivers & Assumptions

  • Incidence-based: Funnel starts from annual new ES-SCLC diagnoses (33k/year in US)
  • DLL3 gating: ~85% of SCLC express DLL3; 55% are clinically fit for second-line
  • 2L population: ~50% of treated patients eventually reach second-line (high mortality in SCLC)
  • Persistence (0.55): Reflects rapid disease progression and short median treated duration (~8 months); only ~55% of the annual price is realized per patient due to discontinuation/death
  • Peak scenario: Assumes market capture at scale; real peak would depend on market share, competition, and uptake velocity

This aligns with tarlatamab's labeled indication and the inherent mortality/progression dynamics of ES-SCLC.

GPT-4o mini small
openai · 3 tool calls
2.4
funnel_construction2
eligibility_gating2
numerical_consistency3
assumptions_revenue2
faithfulness3
flawed — The agent skipped the diagnosis (95%) and treatment (85%) gates entirely, applying subtype fractions directly to raw incidence, which inflates the addressable population (10,803 vs reference 8,720). More critically, it never applied any pea
open full trajectory →
final answer

To size the US addressable patient population and estimate the rough peak revenue for a DLL3 T-cell engager (similar to tarlatamab) in second-line treatment of extensive-stage small cell lung cancer (ES-SCLC), we'll use the following key data:

Epidemiology Information:

  • Incidence of ES-SCLC: 33,000 new cases per year.
  • Prevalence: 38,000 total living patients.
  • Diagnosis Rate: 95% (of patients diagnosed).
  • Treatment Rate: 85% (of diagnosed patients that receive treatment).

Subtype Prevalence:

  • Fraction of extensive-stage: 70% of SCLC cases are extensive-stage.
  • Fraction of cases expressing DLL3: 85%.
  • Fraction fit for second-line therapy: 55%.

Pricing Information:

  • Annual therapy price: $220,000.
  • Average treated duration: 8 months.
  • Persistence factor: 55% (the fraction of a year that patients remain on therapy due to discontinuation or mortality).

Calculating Addressable Patient Population:

  1. Total Incidence of SCLC:
    [
    \text{Total Incidence} = 33,000 \text{ patients/year}
    ]

  2. Patients with Extensive-stage SCLC:
    [
    \text{Extensive-stage incidence} = 33,000 \times 0.70 = 23,100 \text{ patients/year}
    ]

  3. Patients expressing DLL3:
    [
    \text{DLL3-positive patients} = 23,100 \times 0.85 \approx 19,635 \text{ patients/year}
    ]

  4. Patients fit for second-line therapy:
    [
    \text{Fit for 2L therapy} = 19,635 \times 0.55 \approx 10,803 \text{ patients/year}
    ]

Rough Peak Revenue Calculation:

  1. Effective Revenue per Patient per Year:
    [
    \text{Effective Price} = 220,000 \times 0.55 \approx 121,000 \text{ USD/patient/year}
    ]

  2. Total Addressable Revenue:
    [
    \text{Total Revenue} = 10,803 \times 121,000 \approx 1,308,763,000 \text{ USD}
    ]

Summary:

  • US Addressable Patient Population for DLL3 T-cell engager in second-line ES-SCLC: 10,803 patients/year
  • Rough Peak Revenue Estimate: Approximately $1.31 billion/year

This estimate provides a useful insight into the potential market opportunity for DLL3 T-cell engagers in the specified patient population.