Evaluating CDMO companies in Asia Pacific comes down to six criteria: regulatory track record, capacity and scale path, modality coverage, cost structure and currency, IP protection and contract terms, and supply-chain resilience. This guide explains what to check under each, gives a reusable scorecard, and lists the red flags that stop a shortlist.

A CRM197 certificate of analysis should cover identity, purity by two orthogonal methods, nicked and aggregated forms, endotoxin, residual host cell protein and DNA, and confirmed non-toxicity. WHO guidance for pneumococcal conjugate vaccines sets a floor above 90% protein purity; GMP-grade supply with a US FDA drug master file lets a developer reference the data directly.

Extrapolation lets a biosimilar be approved for indications of the reference product that were not studied in the comparative clinical trial. Regulators grant it when the totality of evidence, led by analytical similarity, shows the same mechanism of action, pharmacokinetics and immunogenicity apply in each condition of use. It is justified indication by indication, never assumed.

Biosimilar interchangeability is a separate US designation layered on biosimilarity. Under PHS Act section 351(k)(4), the product must be expected to produce the same clinical result as the reference in any given patient, and switching must add no risk. FDA's 2019 guidance expected switching studies; a 2024 draft update proposes they are generally unnecessary.

IND and BLA filing sit at opposite ends of one CMC timeline. The IND (21 CFR 312) must show the product is safe to dose and consistently made; the BLA (21 CFR 601) must show a validated process, a justified specification and real-time stability. Everything between is phase-appropriate build-up.

CMC regulatory compliance for a biologic means that the cell substrate, process, analytics, specifications and stability program each follow the matching ICH Q guideline and are documented in CTD Module 3 (3.2.S and 3.2.P). Comparability after changes follows ICH Q5E, and post-approval change management follows ICH Q12.

HRV 3C protease is a cysteine protease from human rhinovirus that cuts the sequence Leu-Glu-Val-Leu-Phe-Gln↓Gly-Pro, releasing a protein from its affinity tag while leaving the target intact. It works at 4 °C and is available in R&D and GMP grades. This article explains the mechanism, reaction conditions, applications, and when the enzyme grade matters.

Bioconjugation links a payload to an antibody through a linker. Stochastic methods react native lysines or interchain cysteines and give a distribution of drug-to-antibody ratios; site-specific methods place the payload at defined positions and give a near-homogeneous product. This article compares the two and explains what each demands in analytics, HPAPI handling and supply chain design.

Bispecific antibody CDMO services differ from monoclonal antibody services in cell line strategy, purification depth, analytical scope and risk. These eight questions, with the answers a bispecific-ready partner should be able to give, let sponsors compare proposals on substance rather than on price per gram before a cell line is built.

Bispecific antibody development adds three CMC problems that a monoclonal antibody never has: four chains must pair correctly, mispaired and half-antibody species must be removed, and two binding activities must each be controlled. This guide covers the formats, the engineering fixes, the purification and analytical strategies, and what regulators expect from a bispecific program.

Microbial fermentation is the right route for non-glycosylated proteins, peptides, enzymes and plasmid DNA: Escherichia coli and Pichia pastoris reach high cell density in one to five days at a fraction of mammalian media cost. The trade-offs are endotoxin, inclusion bodies and the absence of human-type glycosylation, each managed by a specific downstream step.

Fusion protein purification is harder than antibody purification because the two fused domains rarely share a stability window, the linker between them is a protease target, and the construct often lacks a platform affinity handle. The solutions are construct-specific capture, controlled tag removal, and analytics that resolve clipped and misfolded variants before they reach the specification.

Recombinant protein production and purification is a chain of seven decisions: which host expresses the gene, how the culture is fed and when it is harvested, how cells are removed, which chromatography steps capture and polish the protein, whether refolding is needed, how the final buffer is set, and which assays prove purity and identity.

Recombinant protein drug substance manufacturing turns a characterized cell bank into frozen bulk protein that meets a release specification. Under cGMP it is a fixed sequence of unit operations: seed train, production culture, harvest, capture, viral clearance, polishing, UF/DF and bulk fill, each bounded by in-process controls and closed by ICH Q6B release testing.

Commercial manufacturing adds obligations that clinical supply never tested: uninterrupted delivery, inspection by every market's regulator, and a cost of goods that sets the margin. Evaluate a CDMO on four lines of evidence: installed and expandable capacity, approvals held by the actual site, a documented transfer-and-PPQ path, and business-continuity design across sites and cell banks.