Prompt
How do I choose between different biologics CDMO providers for upstream capacity, downstream recovery, and sterile fill-finish?
Latest observation
Choosing a biologics CDMO is mostly about matching what you need to outsource with where the technical and operational risk sits in your program.
A practical way to compare providers is to evaluate them separately for:
- Upstream capacity
- Downstream recovery/purification
- Sterile fill-finish
Then score each provider on a common set of criteria: technical fit, scale, quality/compliance, capacity, speed, and commercial terms.
1) Start with your own program requirements
Before comparing vendors, define:
- Molecule type: mAb, bispecific, Fc-fusion, enzyme, viral vector, etc.
- Expression system: CHO, HEK, microbial, etc.
- Development stage: process dev, tox, Phase 1/2/3, commercial
- Batch size and annual volume needs
- Required titer, yields, purity, and product quality attributes
- Formulation and storage needs
- Sterility assurance expectations
- Geography/regulatory markets: FDA, EMA, MHRA, PMDA, etc.
- Timeline and risk tolerance
- Whether you need one integrated CDMO or best-of-breed vendors
This determines whether you need a provider that is strong in one unit operation or one that can handle the whole end-to-end chain.
2) How to evaluate upstream capacity
Upstream is about whether the CDMO can reliably produce enough high-quality biomass or harvestable product.
Key evaluation criteria
- Platform fit
- Do they have an established process platform for your molecule type?
- How much process customization is required?
- Bioreactor scale and flexibility
- Available scales: bench, pilot, 2kL, 5kL, 10kL+, etc.
- Single-use vs stainless steel options
- Ability to run multiple campaigns in parallel
- Real capacity, not just installed capacity
- Ask about actual utilization and scheduled backlog
- How much capacity is reserved for existing clients?
- Process performance
- Typical titers and productivity ranges
- Batch success rates
- Contamination history
- Hold times and in-process controls
- Tech transfer capability
- Experience transferring from your process or platform
- Documentation quality and speed of PPQ readiness
- Supply chain resilience
- Media, resins, single-use components, filters, and buffer supply
- Dual sourcing and inventory strategy
- MSAT and analytics support
- Strength of process sciences, automation, and data systems
Questions to ask
- What is your current upstream occupancy rate?
- How many campaigns can you run simultaneously for my molecule class?
- What are your typical cycle times from inoculation to harvest?
- What is your average batch failure rate over the last 2–3 years?
- What titer range do you support at commercial scale?
- How quickly can you scale from development to GMP?
Red flags
- Installed capacity that is heavily booked
- No clear platform for your modality
- Overpromising on scale without data
- Weak raw material sourcing or frequent equipment constraints
3) How to evaluate downstream recovery
Downstream often determines purity, yield, cost of goods, and scalability.
Key evaluation criteria
- Experience with your product class
- Different purification challenges for mAbs, fragments, viral proteins, enzymes, etc.
- Purification train design
- Protein A, ion exchange, HIC, mixed-mode, filtration, viral clearance, polishing
- Yield and purity performance
- Step yields
- Aggregate and impurity clearance
- Host cell protein and DNA removal
- Scalability
- Can small-scale methods translate to commercial equipment?
- Resin lifetime and lifetime validation
- Virus clearance and safety
- Proven viral inactivation and filtration strategy
- Validation package quality
- Process robustness
- Sensitivity to feed variation, pH, conductivity, temperature
- Analytical capability
- Method development, characterization, release testing, comparability support
- Consumables and resin strategy
- Resin availability, reuse strategy, and supply constraints
Questions to ask
- What are your typical overall recovery yields for similar molecules?
- How many downstream platforms do you have already validated?
- What is your approach to viral clearance validation?
- How do you handle resin lifetime studies and cleaning validation?
- What is your track record for HCP/DNA clearance at high loads?
Red flags
- Generic downstream claims without product-class specifics
- No demonstrated validation experience
- Limited analytical support
- Unclear resin sourcing or proprietary bottlenecks
4) How to evaluate sterile fill-finish
Fill-finish is about sterility, integrity, and reliable delivery of drug product.
Key evaluation criteria
- Aseptic capability
- Is the site true aseptic fill-finish or reliant on terminal sterilization?
- Barrier technology: isolators vs RABS vs traditional cleanrooms
- Container closure formats
- Vials, syringes, cartridges, BFS, etc.
- Compatibility with your formulation and market needs
- Line flexibility and speed
- Small-batch clinical fill and large-scale commercial fill
- Changeover efficiency
- Sterility assurance and contamination control
- Environmental monitoring performance
- Media fill history
- Investigations and deviation handling
- Formulation and drug product services
- Lyophilization capability
- Cold chain, frozen storage, handling of sensitive proteins
- Visual inspection and packaging
- Inspection capabilities
- Serialization and labeling support
- Regulatory inspection history
- FDA/EMA inspection outcomes
- Warning letters, consent decrees, or recurring observations
- Capacity and scheduling
- Line availability is often a bottleneck
- Ask how far out slots are booked
Questions to ask
- What aseptic technology do you use, and why?
- How many media fills have you completed successfully?
- What is your deviation and sterility failure history?
- Can you support my dosage form and container closure system?
- How far out is commercial fill capacity booked?
Red flags
- Weak contamination control history
- Limited experience with your dosage form
- No lyophilization if you need it
- Long lead times for line slots
- Poor regulatory track record
5) Decide whether to use one CDMO or multiple providers
Use one integrated CDMO if:
- You want simpler tech transfer and accountability
- Your program is early stage or resource-constrained
- The provider has strong capabilities across all three areas
- Speed and coordination matter more than best-in-class specialization
Use multiple providers if:
- One vendor is strong in upstream, another in fill-finish
- Your product needs specialized downstream expertise
- You want to reduce single-point-of-failure risk
- You need regional manufacturing redundancy
- Capacity is tight in one segment
Tradeoff
- Integrated CDMO: simpler, fewer handoffs, but may be less specialized
- Best-of-breed network: more flexibility and expertise, but more complex management and tech transfer risk
6) Build a scorecard
A simple weighted scorecard helps compare providers objectively.
Example criteria and weights
- Technical fit: 25%
- Quality/compliance: 20%
- Real capacity and scalability: 20%
- Timeline and responsiveness: 15%
- Commercial terms: 10%
- Geographic/regulatory fit: 10%
Then score each provider for:
- Upstream
- Downstream
- Fill-finish
You may find one provider is:
- Excellent in upstream
- Average in downstream
- Weak in fill-finish
That often argues for splitting the program rather than forcing a single vendor.
7) Due diligence checklist
Request:
- Facility description and equipment list
- Capacity calendar and current utilization
- Recent audit reports or inspection summary
- Quality metrics: deviations, OOS, contamination, right-first-time
- Regulatory history
- Process platform data relevant to your molecule class
- Project plan and tech transfer timeline
- Named team bios
- Business continuity and supply chain plans
- Change control and comparability procedures
- Sample MSA/SOW and pricing model
8) Practical decision rule
If you want a simple rule of thumb:
- Choose the CDMO with the strongest proven track record in your highest-risk step
- Make sure the other steps are at least “good enough” and capacity-secure
- Avoid paying for “full service” if one part is clearly weak
- Don’t choose based on installed capacity alone; choose based on available, qualified capacity
If you want, I can also give you:
- a vendor comparison scorecard template,
- a list of due-diligence questions by function, or
- a decision matrix for single-vendor vs split-vendor outsourcing.
Observation history
Unlock observation history
Create a free Obsurfable account to unlock historical responses, model comparisons, and deeper brand intelligence.
How did Obsurfable measure this prompt?
Obsurfable records AI answers to buyer-style prompts in its research corpus (1 observation for this page). Metrics are distributions over observations, not a single static ranking.
Which AI systems does Obsurfable collect answers from?
OpenAI, ChatGPT, Google, Gemini, Google AI Mode, Anthropic, Claude, Perplexity, Grok, DeepSeek, Mistral, Copilot, and Meta AI.