Few areas of pharmaceutical manufacturing have attracted as much investment in recent years as high-potency containment.
The expansion of oncology pipelines and antibody-drug conjugates (ADCs) has pushed specialist manufacturers to invest in isolators, closed-transfer technologies and manufacturing environments capable of controlling occupational exposure at increasingly low concentrations.
The commercial rationale appears compelling. According to the GlobalData analysis, worldwide ADC sales rose from $1.6bn in 2017 to $13.6bn in 2024 and are forecast to reach $65.2bn in 2031. The number of active ADC programmes increased from 557 in 2020 to 1,643 in 2025, with 71% still at discovery or preclinical stage1.
At the same time, “OEB6” is appearing more frequently in the language used by specialist CDMOs to describe ultra-high-containment capability. While industry definitions vary, leading CDMO Indena classifies an OEB6 capability as being below 100ng/m³ for ADC payload and linker-payload manufacturing.
It is tempting to draw a straight line between these trends, suggesting that more ADCs and potent payloads mean that OEB6 will eventually become the next standard for HPAPI manufacturing.
However, such conclusions fail to address perhaps the most important question: are the molecules themselves actually becoming more potent?
ADC growth does not automatically mean potency growth
The unconjugated payload is among the most hazardous components handled during ADC manufacturing. Payload and linker-payload synthesis can involve highly potent cytotoxic small molecules at occupational exposure limits measured in nanograms per cubic metre. This makes the rapid expansion of ADC development directly relevant to specialist HPAPI manufacturers. Yet ADC innovation is not simply moving towards ever-more-potent cytotoxins.
Historically, ADC developers explored exceptionally powerful DNA-damaging payloads, including pyrrolobenzodiazepines (PBDs), indolinobenzodiazepines, and other agents with very high intrinsic cytotoxicity. Their potency alone did not automatically translate into more effective medicines. Reviews of the field have highlighted dose-limiting toxicities and narrow therapeutic windows that restricted the clinical development of some ultra-potent payload classes5,7.
One recent analysis of DNA-interacting ADC payloads concluded that payloads spanning roughly double-digit picomolar to single-digit nanomolar activity have been the most successful with current conjugation technologies, while still more potent classes were associated with greater clinical toxicity7.
The rise of topoisomerase-I payloads illustrates the change in thinking. Molecules such as deruxtecan and SN-38 are generally less intrinsically cytotoxic than some PBDs, calicheamicins or classical ultra-potent tubulin-targeting warheads. Their success has instead been dependent on the whole ADC architecture: linker stability, drug-to-antibody ratio (DAR), membrane permeability, bystander effect and tumour biology3,5,6.
Published comparisons show a broad range of cellular potencies among successful ADC payloads, from picomolar activity for calicheamicin and maytansinoids to low-nanomolar activity for deruxtecan3. This is an important reminder that the next generation of ADCs is not necessarily a race towards the lowest possible IC50.
It should be noted that IC50 is not an occupational exposure limit (OEL). Pharmacological potency may inform occupational toxicology, but a compound-specific OEL reflects a broader assessment of hazard, dose-response, route of exposure, toxicology, pharmacokinetics, and uncertainty. Therefore, a more effective cancer drug is not automatically a molecule with a lower occupational exposure limit.
What about conventional small-molecule oncology?
The picture is also more nuanced outside ADCs. A 2026 Genentech analysis estimated occupational exposure limits for 83 FDA-approved small-molecule kinase inhibitors. The results covered a wide range, from 50ng/m³ to 96,000ng/m³.
Most were far from the ultra-low exposure levels associated with the highest-containment manufacturing. Eighty-two per cent had estimated OELs above 1,000ng/m³, while the remaining 18% were between 50ng/m³ and 1,000ng/m³. None fell below 50ng/m³ 4.
In other words, within this group of approved oncology drugs, the analysis did not identify a general shift towards extremely low occupational exposure limits.
The comparison has its limits. Kinase inhibitors represent only one class of oncology small molecules, and estimated OELs for approved drugs do not necessarily reflect what is currently moving through early development.
Nevertheless, the data suggests that the whole oncology pipeline is not simply becoming increasingly potent. A more likely scenario is that the market is expanding at both ends. Most molecules remain within established HPAPI ranges, while a smaller group of ultra-potent compounds creates increasing demand for very high-containment capabilities.
Defining exactly what is OEB6
Before asking whether OEB6 will become the new standard, the industry must confront a basic terminology problem. At present, there is no universally harmonised pharmaceutical definition of OEB6.
Occupational exposure banding is a risk-management approach. An OEL is a compound-specific airborne concentration derived to protect workers. While an occupational exposure band (OEB) groups substances into broader exposure ranges to support hazard communication and control decisions, particularly when a definitive OEL is unavailable.
For example, NIOSH uses five occupational exposure bands labelled A to E, with Band E representing the lowest exposure range. Guidance from NIOSH explicitly states that an OEB is not intended to replace a substance-specific OEL2.
Many pharmaceutical companies have developed their own internal systems. For example, a published pharmaceutical framework describes OEB5 as 0.1µg/m³ to <1µg/m³ and creates an “OEB5 Special Case” below 0.1µg/m³ — 100 ng/m³. In that framework, payload and linker-payload components of ADCs may be assigned to this special case because the warhead often drives the hazard assessment3.
Elsewhere, Roche/Genentech uses another structure. In the 2026 kinase-inhibitor analysis, its highest category is OEB4, defined as ≤0.05µg/m³, or 50ng/m³ 4.
However, Indena defines OEB5 as 0.1µg/m³–1µg/m³ (100ng/m³–1,000ng/m³) and OEB6 as below 0.1µg/m³ (<100ng/m³), as confirmed internally by Dr Andrea Gambini1. This is one company-specific banding approach rather than a harmonised industry definition. Other CDMOs can still use different thresholds.
The practical consequence is striking: the same compound with an OEL of 20ng/m³ could sit in the highest category of several different systems — OEB4, OEB5 Special Case or OEB6 — depending on the organisation assessing it.
That does not make the terminology meaningless. Yet it does mean that “OEB6 capable” is less standardised than the phrase can imply.
A marketing label — or a real manufacturing capability?
The growing use of OEB6 has a commercial dimension. The label is an effective shorthand for telling sponsors that a CDMO is comfortable operating at the extreme end of high-potency manufacturing. But describing OEB6 as only marketing would be equally inaccurate.
Behind the terminology are real engineering investments. Ultra-high-containment manufacturing requires more than a high-specification glovebox. The full process must control exposure during charging, transfer, reaction, sampling, filtration, drying, milling, discharge, cleaning, waste handling, and maintenance. The relevant containment performance also depends on batch size, frequency of interventions, physical form and dustiness.
This distinction is recognised in the latest ISPE SMEPAC guidance, which provides a standardised methodology for evaluating airborne particle emissions and surface deposition from pharmaceutical containment systems under defined conditions. The 2024 third edition explicitly expands the approach to more complex equipment systems and integrated processes rather than treating containment as the performance of an isolated piece of equipment8.
This is why the more meaningful customer question is not simply “Is your facility OEB6?”. Instead, it should be: “What containment performance can you demonstrate for the process I need to run?”
Why ultra-high containment can matter even if OEB6 never becomes the majority
There is another reason why capacity at the extreme end of containment can be valuable even if most HPAPIs remain within what many organisations define as OEB5.
During early development, toxicological information is incomplete. Exposure bands and provisional OELs can evolve as more pharmacology and toxicology data become available. A sponsor can therefore begin process development under one containment assumption and later discover that the compound requires substantially tighter controls.
If the selected manufacturing partner cannot accommodate the revised exposure requirement, a technology transfer may be needed at precisely the point where timelines, comparability and supply continuity become more important.
From this perspective, the business case for ultra-high containment is not that every molecule will eventually become OEB6. Instead, it is that a broader operating envelope reduces development risk.
Indena’s current CDMO materials state that substances are assigned a compound-specific OEL and allocated to the appropriate manufacturing line according to the required containment, with capabilities for compounds with OELs as low as 1ng/m³ 9. The company also has more than 30 years of experience in HPAPI small-molecule manufacturing, including clinical and commercial programmes, and capabilities in maytansinoids and camptothecin-derived payload chemistry9.
That positioning is more credible than claiming that OEB6 will inevitably replace OEB5. It allows the manufacturing capability to remain relevant even if the industry ultimately settles on a different nomenclature.
OEB6: the next standard, or the next specialist capability?
The available evidence does not yet support the idea that OEB6 is on a clear path to replacing OEB5 as the universal standard for HPAPI manufacturing.
The ADC market is expanding rapidly, and payload and linker-payload manufacturing creates a growing pool of projects that require advanced containment. A subset of these compounds sits at extremely low OELs and justifies investment in ultra-high-containment capacity.
But the molecular story is more nuanced than “newer equals more potent”. Some of the most intrinsically powerful ADC payloads have faced clinical toxicity limitations, while the recent success of topoisomerase-I payloads demonstrates that ADC performance can improve through smarter molecular and conjugation design rather than simply increasing warhead potency. Outside ADCs, available data from kinase inhibitors do not show a wholesale migration towards ultra-low-ng/m³ OELs4-7.
At the same time, OEB6 itself is not a harmonised standard. What one manufacturer describes as OEB6 may be classified by another as OEB5 Special Case, or even under a differently numbered internal system2-4. Therefore, the market may be evolving in a different way.
Rather than an “OEB6 market” replacing the OEB5 market, the pharmaceutical industry may be creating a larger and commercially more important ultra-potent tail at the edge of an expanding HPAPI landscape.
For specialist CDMOs, this matters. Manufacturers able to demonstrate containment at very low exposure levels can support a wider range of molecules, absorb changes in toxicological understanding during development and reduce the risk of disruptive transfers later in a programme.
Whether the industry eventually calls that capability OEB5+, OEB6, or something else may be secondary.
The next standard in HPAPI manufacturing may not be a new number. Instead, it may be the ability to demonstrate — rather than simply declare — how potent a molecule can be handled safely by a facility.
References:
[1] Alexander Love / GlobalData draft supplied by Indena. GlobalData ADC market and pipeline figures and RFP observation from Dr Andrea Gambini. Indena classification subsequently confirmed internally by Dr Andrea Gambini: OEB5 = 0.1–1µg/m³ (100–1,000ng/m³); OEB6 = <0.1µg/m³ (<100ng/m³).
[2] NIOSH, Occupational Exposure Banding Process for Chemical Risk Management; CDC occupational exposure banding resources. Five bands A–E; OEB is not intended to replace a substance-specific OEL. https://www.cdc.gov/niosh/exposure-banding/about/index.html
[3] Graham et al., “Considerations for setting occupational exposure limits for novel pharmaceutical modalities”, Regulatory Toxicology and Pharmacology (2020). Published framework includes OEB5 0.1–<1µg/m³ and OEB5 Special Case <0.1µg/m³; ADC payload/linker-payload banding is driven by warhead hazard. https://pmc.ncbi.nlm.nih.gov/articles/PMC7605856/
[4] Moudgal et al., “Considerations for establishing occupational exposure limits for small molecule kinase inhibitors in drug development”, Frontiers in Toxicology (2026). 83 oral SMKIs: estimated OEL range 0.05–96µg/m³; none below 0.05µg/m³. https://www.frontiersin.org/journals/toxicology/articles/10.3389/ftox.2026.1743558/full
[5] Payload diversification: a key step in the development of antibody–drug conjugates. Review discussing clinical limitations of ultra-potent PBD/IGN/CBI payloads and the success of less-potent Topo-I payloads enabled by improved ADC design. https://pmc.ncbi.nlm.nih.gov/articles/PMC9847035/
[6] Considerations for the design of antibody drug conjugates (ADCs) for clinical development: lessons learned. Payload potency ranges include calicheamicin, DM1/DM4, MMAE, deruxtecan and PBD. https://pmc.ncbi.nlm.nih.gov/articles/PMC10717055/
[7] Present Scenario and Future Landscape of Payloads for ADCs: Focus on DNA-Interacting Agents. Review concludes that medium-potency payloads can provide a more favourable balance with current ADC technologies and notes toxicity constraints for extremely potent classes. https://pmc.ncbi.nlm.nih.gov/articles/PMC11510327/
[8] ISPE Good Practice Guide: SMEPAC, Third Edition (2024), and 2025 Pharmaceutical Engineering overview. Standardised methodology for evaluating containment performance under defined conditions, including complex equipment systems. https://ispe.org/publications/guidance-documents/good-practice-guide-smepac-standardized-methodology-evaluation-pharmaceutical-airborne
[9] Indena CDMO Overview 2026 supplied internally: OEL capability as low as 1ng/m³; 25 years of HPAPI small-molecule experience; current clinical/commercial HPAPI portfolio and payload-linker capabilities. Internal claims require Indena approval before publication.
