Engineering Guide · Vendor-Neutral

BPOG vs USP <665>: Which Extractables Protocol Do You Need?

Published August 19, 2026 · 12 min read
BPOG extractables protocol vs USP 665 side-by-side: model solvents, extraction time points, and regulatory status WFI (pH 7) 0.1 M H3PO4 (pH 2) 0.5 N NaOH (pH 13) 50% EtOH / H2O 24 h 21 d 70 d at 40 °C INDUSTRY CONSENSUS BPOG Best practice (voluntary) VS bridging data C1: 50% EtOH / H2O C2: pH 3 salt C3: pH 10 buffer Low = gen chem Mod = 1 solv High = all 3 solvents COMPENDIAL · 1 MAY 2026 USP <665> Mandatory (regulatory) Solvents · Time points · Risk-tier depth
Figure 1: The BPOG protocol (left) uses four model solvents at three long time points at 40 °C as a worst-case supplier benchmark. USP <665> (right) uses up to three solutions at a single time point, scaled to the end user's component risk tier. Dashed arrows show that BPOG data usually bridges into a USP <665> assessment rather than needing a repeat study.
Quick Verdict

Use the BPOG protocol when you want a supplier-agnostic, worst-case dataset for early screening and vendor benchmarking. Use USP <665> because you have to — it is compendial and mandatory from 1 May 2026 for every plastic component that touches a US drug substance or drug product. Most manufacturers keep both: BPOG for kinetics, USP <665> for the regulatory dossier.

Key differences at a glance

Side-by-side comparison

Factor BPOG USP <665>
Regulatory status Voluntary best practice (BioPhorum consortium) Compendial — mandatory from 1 May 2026
Governing body BioPhorum Operations Group (industry consortium) US Pharmacopeia (compendial standards body)
Model solvents 4 (WFI, 0.1 M H₃PO₄, 0.5 N NaOH, 50% ethanol/water; originally 6) 3 (C1 50% EtOH, C2 pH 3 salt, C3 pH 10 buffer)
Time points 3 — 24 h, 21 d, 70 d (release kinetics) 1 — representative point in 1–21 d range
Extraction temperature 40 °C (with −25 °C freeze reference) 40 °C maximum
Risk tiering Same worst-case protocol for every component Scaled by low / moderate / high component risk
Analytical panel HS-GC-MS, GC-MS, LC-MS, ICP-MS, TOC, NVR per time point HS-GC-MS, GC-MS, LC-MS, ICP-MS, UV, TOC (orthogonal)
Who typically owns the report Component supplier (Extractables Report to end users) Drug-product manufacturer (regulatory dossier)
Cost per component (typical, third-party lab) USD 20–40k (full 3-time-point matrix) USD 8–25k (scaled by risk tier)

Values reflect typical published protocol descriptions and 2025–2026 lab quotes. Your USP <665> laboratory's current scope and your component vendor's Extractables Report take precedence.

BPOG in detail

The BPOG extractables protocol was published in 2014 by the BioPhorum Operations Group, an end-user consortium of biopharma manufacturers frustrated by the lack of comparability between vendor Extractables Reports. Before BPOG, every supplier chose their own solvents, times, temperatures, and analytical thresholds, and end users could not stack two vendor packages next to each other and decide which single-use bag or filter was cleaner. BPOG's remit was to fix that by standardising the extraction so that Sartorius, Cytiva, Thermo Fisher, and MilliporeSigma reports would all be measured against the same yardstick.

How the BPOG extraction is designed

The original 2014 protocol specified six model solvents, chosen to bracket the chemistry that a single-use component might see anywhere in a real bioprocess: water for injection at neutral pH, 0.1 M phosphoric acid to simulate acidic hold conditions, 0.5 N sodium hydroxide to represent CIP-adjacent alkaline exposure, 1 M sodium chloride for high ionic strength, 50% ethanol/water as an organic simulant, and 1% polysorbate 80 as a surfactant-containing formulation surrogate. A 2020 revision dropped the sodium chloride and polysorbate 80 solvents after multi-vendor data showed they added few unique extractables that were not already captured by the other four, leaving today's four-solvent panel.

Extraction runs at 40 °C for three cumulative time points — 24 hours, 21 days, and 70 days — plus a −25 °C, 30-minute freeze reference for storage-relevant chemistry. The multi-time-point design lets end users see whether an extractable is a fast surface release (essentially complete by 24 h) or a slow diffusion out of the polymer bulk (still climbing at 70 d), which matters for how much shows up in a real 14-day hold bag.

When BPOG wins

Supplier screening and vendor selection is where BPOG shines. Because the four solvents together span pH 2 to pH 13 plus 50% ethanol at 40 °C for up to 70 days, the resulting Extractables Report is broadly a worst-case envelope: if a leachable is not going to fall out of the polymer under those conditions, it is very unlikely to fall out in your gentler real-world process. That makes BPOG the sensible package for the "which vendor's 2000 L bag should we qualify" conversation, and for early-phase programmes where the process is still moving. It is also the right dataset for the sterilisation-effect and freeze-thaw studies BioPhorum is still actively updating, and for informing the "worst case" leg of a USP <1665>-style risk assessment.

USP <665> in detail

USP <665> "Plastic Components and Systems Used to Manufacture Pharmaceutical Drug Products and Biopharmaceutical Drug Substances and Products" is a compendial chapter of the US Pharmacopeia. That word "compendial" is the whole story: unlike BPOG, USP <665> carries the force of a US regulatory standard, and after two extensions its official date is finally fixed at 1 May 2026. From that date every plastic component that contacts a US-marketed drug product or drug substance must be qualified under a USP <665>-aligned assessment. Its companion chapter, USP <1665>, is informational and describes how the risk assessment should be built and defended.

How USP <665> scales testing to risk

USP <665> is not a single fixed protocol like BPOG. It is a risk-tiered framework that lets the end user match the depth of testing to the actual risk each component poses. Three risk levels are used — low, moderate, and high — and the end user (not the supplier) assigns the level based on the process step, the contact duration, the contact temperature, the process-stream chemistry, and the number of purification unit operations still to come between the component and the finished drug product. A short-contact gasket in a WFI header at ambient is low risk; a formulation-hold bag at 2–8 °C for 30 days is high risk.

The chapter then defines three standardised extraction solutions: Solution C1 is 50% ethanol/water as an organic simulant, Solution C2 is a pH 3 salt solution for acidic conditions, and Solution C3 is a pH 10 buffer for alkaline conditions. Low-risk components need only general chemistry testing (TOC, UV, pH, conductivity) — no organic profiling at all. Moderate-risk components need organic extractables profiling with C1 (50% ethanol) alone. High-risk components need the full three-solution panel. Extraction runs at up to 40 °C for a single "representative" time point selected from the 24-hour, 168-hour (7-day), or 504-hour (21-day) checkpoints, chosen to bracket the actual worst-case process contact time.

When USP <665> wins

USP <665> is the report format that will appear in every US IND, BLA, and NDA from 2026 onward. Its risk-scaled design means you are not paying for a 70-day, six-solvent extraction on a peristaltic-pump tubing segment that sees fluid for 20 minutes at 22 °C, and its single-time-point simplification cuts typical per-component lab spend by a factor of two to three compared with a full BPOG matrix. Where USP <665> unambiguously wins is the situation you actually care about: getting a US drug product through review without a compliance question on your single-use qualification package.

Pros and cons

BPOG protocol

Advantages

  • Worst-case solvent envelope (pH 2 to pH 13 + 50% ethanol) captures nearly every leachable a real process could ever mobilise.
  • Three time points (24 h, 21 d, 70 d) show release kinetics, not just a snapshot — useful for predicting long-hold leachables.
  • Standardised across suppliers, so vendor Extractables Reports for competing single-use bags can be compared line by line.
  • Widely available today — most tier-1 SU suppliers already publish BPOG packages for their major SKUs.

Disadvantages

  • Not compendial — a BPOG-only submission will draw a question from FDA/EMA reviewers once USP <665> is in force.
  • The same worst-case matrix runs for every component regardless of actual risk, so cost per component is high.
  • 70-day extraction adds calendar time to any new-component qualification — painful for late-stage change controls.
  • Model solvents (0.5 N NaOH, 50% ethanol) rarely match actual process conditions, so the data still needs interpretation before it drives a leachables prediction.

USP <665>

Advantages

  • Compendial and regulator-recognised — a single package that satisfies FDA, EMA, and PMDA reviews without additional bridging.
  • Risk-tiered scope: low-risk components skip organic profiling entirely, cutting per-component cost.
  • Single representative time point (24 h, 7 d, or 21 d) shortens qualification calendar time versus a 70-day BPOG matrix.
  • Pairs with USP <1665> guidance for a defensible, life-cycle-aware risk-assessment narrative.

Disadvantages

  • No time-course data at a single time point — slow-release leachables can be under-detected relative to a real long-hold process.
  • Risk-level assignment is the end user's responsibility, so two sites can classify the same component differently and get different results.
  • Because supplier reports historically ran BPOG, most sites still need a gap analysis before a legacy dataset can be filed as USP-compliant.
  • Third-party lab capacity is tight ahead of the 2026 deadline — some queues are already stretching past 12 weeks.

Which should you choose?

The honest answer for a US-marketed programme is both: BPOG for the supplier-benchmark and worst-case leg, USP <665> for the regulatory dossier. The four scenarios below show which of the two carries the decision when they diverge.

Early-phase supplier screening

Choosing between two SU bag films or two 0.2 µm filters before you commit to a vendor. You want the widest possible solvent envelope to see which supplier has the cleaner polymer, and time-course release data.

Choose BPOG

Commercial CMC filing (US)

Every single-use component in the drug-substance and drug-product train has to appear in the BLA or NDA package with a documented extractables assessment defensible to FDA reviewers.

Choose USP <665>

Late-stage change control

You are replacing a qualified component post-approval and need to justify equivalence quickly. A BPOG dataset is often already on file from the original vendor qualification; the fastest path is to bridge it into a USP <665> format.

Bridge BPOG → USP <665>

New long-hold formulation bag

Drug product held in a single-use bag at 2–8 °C for 30 days pre-fill. Slow-release leachables at day 21+ will drive AET calculations; a single 24-hour time point could miss them.

Use BPOG kinetics, file USP <665>

Real-world use cases

How typical biopharma sites are combining the two protocols under the 1 May 2026 deadline.

CHO mAb, 2000 L
Legacy BPOG data, gap-analysed to USP <665>

Existing Sartorius and Cytiva BPOG Extractables Reports for the bags, tubing, and filters are still on file. A third-party lab reformats them into USP <665> Solution C1/C2/C3 language and only re-runs the pH 3 salt extraction for the media-hold bag, where BPOG had no direct match.

AAV, 200 L transient
USP <665> from day one

A first-in-human AAV programme launching in 2026 skips the BPOG package entirely and buys USP <665>-format reports from suppliers directly, tiered by the transient production process risk (short-contact filters low, formulation bag high).

Microbial, 5000 L E. coli
Depth filters — the bridging bottleneck

Depth-filter capsules for cell-lysate clarification sit at the top of the E&L risk stack because of the large surface area. Site uses BPOG kinetics to defend the "worst case" narrative and USP <665> high-risk-tier data (all three C solutions at 21 d) for the filing.

Cell therapy, 10 L bioreactor
Short contact, low tier

Autologous CAR-T contact times are hours, not days. Most process contact components classify as low or moderate risk under USP <665>, so only general chemistry plus C1 (50% ethanol) profiling is needed — no BPOG matrix required.

Sizing a sterile filter or capsule for USP <665> qualification?

The Filtration Calculator sizes 0.2 µm and 0.45 µm capsules from your Vmax and Pmax data so the qualification-batch surface area matches the extractables-report basis. Pairs with the BPOG-to-USP bridging workflow.

Open the Filtration Calculator

Cost and lifecycle considerations

Three cost drivers for an extractables package

Third-party lab fees (per solvent, per time point, per analytical method) + internal risk-assessment and gap-analysis labour + programme calendar cost (a 70-day BPOG extraction on the critical path can defer a batch). USP <665>'s single-time-point, risk-scaled design reduces the first two; BPOG's kinetics reduce the third by front-loading the science and avoiding late-phase surprises.

Typical 2026 lab quotes for a full BPOG matrix on one single-use component run USD 20–40k, depending on how many analytes are quantified and whether NVR/TOC is included at every time point. A USP <665> package on the same component is USD 8–15k if it lands in the moderate-risk tier (Solution C1 only, single time point) and USD 20–25k in the high-risk tier (all three Solutions at 21 d).

The bigger cost, though, is calendar. A component that fails an extractables specification on day 21 of a BPOG extraction has cost the programme three weeks before anyone knows there is a problem. Sites that ran BPOG on all their SU components in 2023–2024 are now enjoying the payoff — those datasets are largely bridging into USP <665> with only targeted gap-fill runs.

Cost component BPOG (per SU part) USP <665> (per SU part)
Third-party lab fee (typical)USD 20–40kUSD 8–25k (by risk tier)
Calendar time (elapsed)10–14 weeks4–8 weeks
Internal QA/QC hours60–100 h40–80 h (risk assessment heavier)
Regulatory acceptance for US filingRequires bridging narrativeDirect, from 1 May 2026

Vendor landscape

The extractables ecosystem splits into two camps: the single-use component vendors that supply the material and its Extractables Report, and the third-party analytical laboratories that run the studies (typically for the drug-product manufacturer, but sometimes for the SU vendor itself).

Single-use component vendors (issue Extractables Reports)

Third-party analytical labs (run USP <665> and BPOG studies)

Frequently asked questions

Does a BPOG extractables dataset satisfy USP <665>?
Often, but not automatically. The BioPhorum protocol overlaps USP <665> in solvents (50% ethanol, water, pH extremes) and temperature (40 °C), so a BPOG dataset is usually enough to inform the USP <665> assessment. But USP <665> is compendial, so you still need a documented risk assessment that maps each BPOG data point to the USP-specified extraction solution, time point, and analytical method, and back-fills any gap. Vendors and third-party labs are re-issuing BPOG data as USP <665>-format reports for exactly this reason.
When does USP <665> become mandatory?
USP <665> becomes officially effective on 1 May 2026 after two USP-issued extensions. From that date compliance is compendial and enforceable for pharmaceutical and biopharmaceutical drug products marketed in the United States. USP <1665> remains an informational companion chapter and is not itself mandatory.
What is the difference between USP <665> and USP <1665>?
USP <665> is the compendial extractables testing standard for polymeric components used to manufacture drug products and drug substances. USP <1665> is the accompanying informational chapter that describes how to plan and execute the characterization: risk assessment approach, study design, data interpretation, and life-cycle qualification. USP <665> defines the deliverable; USP <1665> explains the methodology.
What are the BPOG extractables protocol solvents?
The 2014 BPOG protocol specified six model solvents: water for injection, 0.1 M phosphoric acid, 0.5 N sodium hydroxide, 1 M sodium chloride, 50% ethanol/water, and 1% polysorbate 80. A 2020 revision removed the sodium chloride and polysorbate 80 solvents because they produced few unique extractables, leaving four: water, 0.1 M H₃PO₄ (low pH), 0.5 N NaOH (high pH), and 50% ethanol/water (organic simulant). Extraction runs at 40 °C for 24 h, 21 days, and 70 days.
What are the USP <665> extraction solvents?
USP <665> specifies up to three standardized extraction solutions: Solution C1 is 50% ethanol/water (organic simulant), Solution C2 is a pH 3 salt solution (acidic simulant), and Solution C3 is a pH 10 buffer (alkaline simulant). Low-risk components need only general chemistry testing; moderate-risk components typically require organic profiling with C1 alone; high-risk components require all three solutions. Extraction is performed at up to 40 °C for durations from 24 hours out to 21 days.
How does USP <665> classify component risk?
USP <665> uses a three-tier system — low, moderate, and high — assigned by the end user, not the supplier. The classification is derived from process step (upstream vs downstream vs final formulation), contact duration, contact temperature, process-stream chemistry, and the residual purification steps between the component and the drug product. A gasket in a WFI line at 25 °C is low risk; a formulation-hold bag at 2–8 °C for 30 days is high risk.
Which analytical methods do USP <665> and BPOG require?
Both protocols expect an orthogonal panel: headspace GC-MS for volatile organics, direct-injection GC-MS for semi-volatiles, LC-MS (usually high-resolution QTOF or Orbitrap) for non-volatile organics, and ICP-MS or ICP-OES for elemental impurities. USP <665> also mentions UV absorbance and TOC. BPOG additionally reports non-volatile residue (NVR) at each time point so the extractable release rate can be trended, which USP <665>'s single time point does not capture.
Is BPOG going away now that USP <665> is compendial?
No. BioPhorum has repositioned BPOG as a complementary best-practice framework rather than a competitor to USP <665>. Its wider solvent palette and multi-time-point kinetics still give useful data for early-phase supplier screening, sterilisation-effect studies, and worst-case leachables risk assessments. Expect vendor Extractables Reports to keep offering both formats — BPOG for supplier-agnostic benchmarking, USP <665> for the regulatory dossier.

Resources and references

Further reading