In fast-moving areas like radiopharmaceuticals and precision immunology, the biggest portfolio risk is often not uncertainty itself, but false certainty created by mismatched benchmarks.
As life sciences enters a more disciplined cycle, one pattern shows up across investor conversations, research and development (R&D) strategy discussions, and business development and licensing (BD&L) activity: momentum concentrates, and capital as well as attention flow into a smaller number of areas that appear positioned for outsized impact and near- to mid-term value creation. In parallel, partnership models are becoming more networked and global.
Those conditions create a predictable side effect: benchmarking pressure. Investors, portfolio teams, and business development and licensing (BD&L) leaders all need quick, comparable baselines. They need a definition of what “good” looks like, how risk should be calibrated, and how a program stacks up against peers.
The benchmarking trap in “hot” categories
Benchmarking is often treated as an anchor for decision-making: a way to calibrate risk, evaluate comparables, and support portfolio trade-offs. In stable therapeutic areas (TAs), historical baselines can be useful. However, in hot, fast-moving TAs, they can become traps: the faster the change, the easier it becomes to benchmark incorrectly because surface similarity between programs masks meaningful differences in trial design, patient selection, endpoint definitions, and the evolving standard of care.
This is the benchmarking trap: comparisons can look quantitative and rigorous, while in reality they are mixing things that are no longer comparable. The illusion of precision can be especially strong when there is a sense of urgency: a TA is hot, capital is competitive, and deal windows feel short.
The problem is not that benchmarks are “wrong.” It’s that a benchmark is only as useful as the assumptions behind it. In momentum areas, those assumptions tend to change quickly:
- Science shifts: new targets, new modalities, more combination strategies
- Evidence packages shift: different endpoints, earlier biomarker gating, changing patient populations
- The overall competitive landscape shifts: new entrants, discontinuations, readouts, and approvals that reshape what “good” looks like
When those shifts are not explicitly accounted for, analyses can end up with false precision: a single number that feels decisive, even when the underlying comparison is not aligned.
Oncology: radiopharmaceuticals and the comparability problem
Radiopharmaceuticals have become one of the clearest examples of a momentum area in oncology. The category spans different radionuclides and targeting strategies, and it is expanding quickly as more companies invest in targeted radioligand therapies. Radiopharmaceuticals are “hot” because, for the first time, they combine high-precision targeting with therapeutic, curative potential.
Once primarily used for imaging in diagnostics, they are now often referred to as theranostics and are capable of both locating and killing cancer cells with high selectivity, reducing the systemic toxicity typically associated with traditional chemotherapy. The fact that they can be designed to target specific markers on cancer cells and therefore deliver radiation directly to the tumor thereby limiting side effects contributes to their appeal.
Growth like we see it with radiopharmaceuticals is creating exactly the conditions where benchmarking traps emerge.
First, “radiopharmaceuticals” is not a single development pathway. Programs can differ meaningfully in radionuclide choice, tumor biology, trial setting, endpoint strategy, and operational feasibility. Benchmarks that treat radiopharma as a single bucket can create misleading averages.
Second, operational constraints can be entangled with clinical timelines. Supply and utilization of medical radioisotopes have been flagged as a structural issue in OECD Nuclear Energy Agency (NEA) work. Shortages are not theoretical: industry reporting has described actinium-225 constraints severe enough to pause late-stage work in parts of the sector.
Third, momentum increases competitive density. As more programs enter the same tumor settings, the bar for differentiation rises, and older benchmarks can understate how quickly expectations are changing. A 2025 Fierce Pharma overview, for example, highlights the rapid commercial growth projections and deal activity fueling radiopharma’s momentum.
This doesn’t necessarily make radiopharma “harder”, but it makes it more sensitive to comparability. The most useful benchmarks tend to be segmented enough to reflect the actual decision being made: specific tumor types, treatment settings, endpoint strategies, and radionuclide classes, with explicit assumptions about operational constraints.
Immunology and autoimmune: when the goalposts move
Autoimmune and inflammation are another high-momentum domain, with increasing focus on precision immunology and pathway-specific modulation rather than broad immunosuppression. The competitive landscape is also unusually dynamic, with multiple mechanism classes, rapid label expansion strategies, and shifting standards of care across diseases.
Benchmarking traps in immunology often show up in more subtle ways than in oncology:
“Therapeutic area averages” hide disease-level differences
Autoimmune is not a single market and not a single clinical reality. Benchmarks that blend across diseases such as psoriasis, inflammatory bowel disease, and rheumatoid arthritis can lead to misleading comparisons, because trial endpoints, placebo responses, and treatment histories vary significantly.
The standard of care can change quickly
In a fast-moving therapeutic area, what counted as a strong evidence package a few years ago may no longer be enough. Benchmarks need to reflect that the reference point is not static.
Mechanism classes aren’t interchangeable
Outcomes, safety trade-offs, and regulatory dynamics can differ meaningfully across mechanism classes and delivery modalities. Class-level benchmarks can still be useful, but only when the category boundaries are drawn transparently and kept current.
In this environment, an “immunology average” can be less helpful than a stratified baseline tied to the actual decision. Benchmarks become more credible when they reflect disease subtype and severity, prior biologic exposure, line of therapy, endpoint definitions and timing, and mechanism class.
Partnership models and the rising value of shared baselines
A third theme is the evolution of partnership models toward more distributed innovation ecosystems: more active search and evaluation functions, more cross-border dealmaking, and more collaboration with venture studios, incubators, and platform biotech firms.
That matters for benchmarking because dealmaking requires shared definitions. When each party brings different assumptions, negotiations can slow, diligence can sprawl, and post-deal alignment will become harder than expected.
A transparent baseline helps reduce friction. It doesn’t force agreement, but it clarifies where the disagreement actually is: is it the evidence, the assumptions, or the strategy.
Edoardo Madussi, Senior Director, Business Development & Strategic Initiatives at Intelligencia AI, said:
“For momentum areas like radiopharmaceuticals and precision immunology, the biggest risk is often false certainty. A benchmark can look clean while mixing different patients, endpoints, and competitive sets. What tends to hold up in real portfolio conversations is a consistently curated evidence base and transparent assumptions, so teams can compare programs on what truly matches and discuss risk without talking past each other.”
What separates useful benchmarks from misleading ones
Benchmarking does not need to be complicated to be reliable. A few simple practices can materially improve decision quality:
- Treat benchmarks as ranges, not single-point numbers.
- Prefer comparability over volume; smaller but truly comparable cohorts can be more decision-useful than large mixed sets.
- Make “last updated” part of credibility in fast-moving categories.
- Where operational constraints plausibly shape timelines (radiopharma is a clear example), keep them in the risk discussion rather than treating them as an afterthought.
While these practices don’t remove uncertainty, they reduce avoidable uncertainty created by mismatched definitions and stale comparisons.
Closing Thought
Hot categories attract capital for good reasons, but they also attract noise. In fast-moving therapeutic areas like oncology and immunology, the teams that stay disciplined about comparability, keep baselines current, and make assumptions explicit are often better positioned to move quickly without losing rigor, even as the science and the competitive field evolve.
Author Bio
Edoardo Madussi, Senior Director, Business Development & Strategic Initiatives at Intelligencia AI
Edoardo Madussi is a biopharma entrepreneur and strategist focused on enabling data-driven decision-making across business development, portfolio strategy, and investment functions. At Intelligencia AI, he leads cross-functional engagement, translating clinical intelligence and AI-powered modelling into actionable insights that support risk assessment, benchmarking, and strategic prioritisation across development pipelines.
Fluent in five languages and having worked across multiple geographies, Edoardo brings a global perspective and a strong ability to translate complex scientific and commercial scenarios into actionable outcomes. He also advises on clinical trial supply chain strategy and has built collaborative partnerships across biotech, pharma, and technology sectors.
