Preparazione allo sviluppo del peptide: Dalla serie A all'IND

Cosa significa in realtà “preparazione allo sviluppo del peptide”.

La preparazione allo sviluppo del peptide non è un'etichetta di fase. È l'insieme di prove che un programma può fornire a un investitore, un partner o un regolatore senza avvertimenti. Un programma è pronto quando viene completata la cronologia di sintesi, dati di scalabilità, controlli di modifica, il pacchetto analitico e le decisioni sui processi sopravvivono allo stesso tempo all'esame accurato.

Il settembre 2026 Le notizie di TwoStep Therapeutics rappresentano un utile punto di osservazione: un finanziamento di serie A e un'autorizzazione IND per TS-104, descritto pubblicamente come un coniugato peptide-farmaco che accoppia un peptide legante l'integrina con MMAE, con Fase 1 previsto più tardi 2026. Ciò che quel record non rivela sono i dati CMC sottostanti. Il finanziamento e lo sdoganamento sono visibili; la riproducibilità della sintesi, i processi di espansione e la convalida analitica dietro di essi non lo sono.

Questa lacuna è il motivo per cui questa guida è costruita sulla base di documenti normativi primari e prove di professionisti piuttosto che sui documenti depositati da una società. La struttura seguente organizza la preparazione in cinque porte, dalla sequenza al candidato pronto per IND, ognuno produce un artefatto specifico che puoi richiedere al tuo programma.

Chiave da asporto: La preparazione è valutabile prima che un regolatore la valuti. Se non puoi produrre l'artefatto dietro un cancello, quel cancello non è chiuso, indipendentemente da quanto è avanzato il programma.

Prerequisiti: ciò di cui hai bisogno prima di valutare la preparazione

Servizi I cinque cancelli che seguono sono utili tanto quanto i documenti che porti loro. Prima di eseguire un'autovalutazione, assemblare quattro cose.

Registri attuali dei lotti e certificati di analisi. Non la tabella riassuntiva di una serie di diapositive: i registri effettivi del batch, con numeri di lotto, cede, e i risultati analitici come riportati.

I metodi analitici e il loro stato di validazione. Per ogni metodo, sapere se è qualificato, convalidato, o ancora in fase di sviluppo. Un metodo che non è stato convalidato non può supportare una specifica di rilascio, e il divario emergerà al Gate 4.

Il profilo del prodotto target, percorso previsto, e dose. Questi determinano i limiti di endotossina e sterilità che applicherai in seguito, quindi una revisione della preparazione senza di essi è incompleta.

Familiarità con i principali documenti a cui fanno riferimento i cancelli. I requisiti di contenuto CMC della FDA per un IND stabiliscono la fase sezione per sezione 1 aspettative, compreso il 21 CFR 312.23(UN)(7) chimica, produzione, e controlla il contenuto. Gli standard ICH e USP citati ovunque, Io Q2(R2), Io Q6B, Io Q1A(R2), USP <71>, e USP <85>, sono i punti di riferimento per la validazione, specifiche, stabilità, sterilità, ed endotossine batteriche rispettivamente. La pagina della FDA è stata letta direttamente per questo articolo e conferma quelle fasi 1 contenuto; i PDF ICH non erano leggibili in questa esecuzione, trattateli quindi come standard da consultare piuttosto che come testi qui citati.

Supponiamo che tu abbia dimestichezza con la terminologia RP-HPLC e UPLC, può leggere un cromatogramma, e può seguire i dati LC-MS senza uno strato di traduzione. Se quello non sei tu, porta qualcuno che può. Un’autovalutazione realistica su tutti e cinque i cancelli richiede da due a tre giorni di lavoro mirato, la maggior parte del tempo è stata spesa per individuare i documenti anziché interpretarli.

Prima di iniziare: il ritardo più comune non è una lacuna tecnica. Nel corso della valutazione viene scoperto che un metodo non è mai stato convalidato o che manca un record batch.

Cancello 1: Dimostrare la riproducibilità della sintesi

Entro la fine di questo cancello dovresti essere in grado di mostrare la stessa sequenza, fatto due volte, fornisce lo stesso profilo di impurità entro i limiti definiti in anticipo. La riproducibilità è la prima cosa che un partner o un revisore esamina, perché un processo che non può ripetersi non può essere convalidato.

Inizia fissando il modo in cui misuri. La purezza è definita mediante RP-HPLC o UHPLC come la percentuale del peptide bersaglio rispetto alle impurità che assorbono a 220 nm, la base di assorbimento del legame peptidico, e le principali impurezze sono sequenze di delezione generate durante la sintesi (GenScript, recuperato 2026-08-03). Quindi impostare i criteri di accettazione lotto per lotto rispetto al grado effettivamente richiesto dal programma, non è il miglior grado disponibile: i gradi pubblicati presentano un livello immunologico ≥75%., ≥85% biochimica, ≥95% di elevata purezza e ≥98% industriale, e la variabilità da lotto a lotto aumenta al diminuire della purezza, soprattutto sotto 80% (GenScript, recuperato 2026-08-03). Tratta questi gradi come categorie commerciali del fornitore, non specifiche normative.

The failure mode is a program that reports one best-batch purity and cannot reproduce it. Qualify a supplier on a single 98% lot and scale-up may reveal that the routine grade is 85%, with visible drift between batches.

Cancello 2: Aumenta la riduzione del rischio prima che sia necessario

By the end of this gate, you should be able to say whether your route survives the move from milligram to kilogram, and point to data rather than a vendor’s confidence. Scale-up is where a workable research route quietly becomes an uneconomic or non-reproducible one, so the question is not whether a CDMO can make more material, but whether it has already made it.

Ask where solid-phase synthesis stops scaling cleanly. Neuland Labs puts straightforward linear SPPS at a practical limit of roughly 30 A 40 residui, and reports that modified peptides, meaning cyclic, lipidated or conjugated sequences, need to hold at least 95% purity at kilogram scale, which in practice means multi-step preparative HPLC and controlled lyophilization (Laboratori Neuland, 2025). Impurity thresholds usually sit at or below 0.1% and require justification, non supposizione.

The failure modes practitioner literature names at scale are specific and worth testing against your own sequence: aggregation from extended coupling cycles, side reactions, yield decline as the chain lengthens, near-linear consumption of resin, protected amino acids and solvent, purification complexity, hold-time degradation losses, and quality deviations such as stereochemistry change, purity loss and batch failure that get magnified at higher volume (Laboratori Neuland, 2025).

One honest caveat on the evidence base. The quantitative peptide synthesis scale-up benchmarks available to this assessment come from vendor and legacy literature. The large-scale SPPS figures published by Andersson and colleagues in 2000 are legacy references and were not page-verified here, so treat them as background rather than as current performance data. Demand your own numbers: record di batch, controlli in-process, and at least one demonstrated campaign at the scale you intend to file.

Cancello 3: Controllare deliberatamente le modifiche

By the end of this gate you should be able to show that every non-natural modification in your candidate is specified, controlled and analytically confirmed, non scontato. A modification made is not a modification controlled.

Enumerate each modification, name the attribute it changes, and pair it with the method that confirms it. Cyclization alters conformational stability and is confirmed by peptide mapping and NMR where identity needs orthogonal support. Lipidation and PEGylation change hydrophobicity, aggregation behaviour and half-life, and are confirmed by LC-MS/HRMS and amino acid analysis. Conjugation is the case where a peptide program’s chemistry and its analytics most often diverge, because the conjugate’s components each need their own control.

TS-104 illustrates the point. BusinessWire describes TS-104 as a PIP integrin-binding peptide conjugated to MMAE, so the readiness question is not only peptide purity but conjugate consistency. Peptidi sintetici The public record does not disclose the conjugation chemistry, the drug-to-peptide ratio control or the release specifications, so those remain inferred rather than evidenced. Di

The failure mode is a modification introduced for potency that silently changes the impurity profile or the stability of the peptide, discovered only when the peptide analytical package is assembled.

Cancello 4: Costruire il pacchetto analitico

By the end of this gate you should hold a method-by-attribute package: one row per release attribute, one named procedure per row, and a validation status for each. The package is the evidence layer every earlier gate depends on, because a purity number alone tells a reviewer nothing about identity, controione, residual solvents or endotoxin.

Start from the FDA’s own CMC content requirements for an IND, which specify structure and identity evidence, the manufacturer, a general method of preparation with reagents, solvents and catalysts plus a flow diagram, acceptable limits and analytical methods for identity, forza, quality and purity, and stability information (NOI. Amministrazione degli alimenti e dei farmaci, recuperato 2026-09-18). The agency’s stated bar is that sufficient information “should be submitted to assure the proper identification, qualità, purezza, and strength of the investigational drug.” NMR, IR and UV are the conventional identity methods, HPLC carries purity and impurities, and certificates of analysis are expected alongside.

Then attach validation expectations to each row. Io Q2(R2) sets the validation characteristics by procedure purpose: assay and potency procedures need specificity, precisione, precisione, linearity, range and robustness, with LOD and LOQ generally not expected; quantitative impurity procedures add LOD and LOQ; limit tests centre on specificity and LOD; identity centres on specificity and selectivity. The acceptance ranges most programs apply are roughly 98.0–102.0% recovery for assay accuracy, %RSD ≤2% for assay repeatability, 80–120% recovery for impurity accuracy, %RSD ≤5% for impurity precision and %RSD ≤10% at the LOQ. Treat those ranges as a starting convention rather than a fixed requirement, and confirm the parameter sets against the guideline text itself before you commit them to a specification.

For the specification itself, ICH Q6B’s specification framework is the useful template even for a synthetic peptide: identity via a highly specific test or more than one if needed, purity and impurities via method-dependent tests that may require a combination of methods, and potency via an assay reflecting biological activity. Acceptance criteria should follow process capability, method performance, product characterization and clinical relevance rather than a universal test list.

Impurity limits deserve their own row. In the FDA-adjacent framework commonly applied to synthetic peptides, peptide-related impurities present in both test and reference product should be no higher than the comparator, new peptide-related impurities above roughly 0.10% should be identified and justified, and new impurities above 0.5% are generally not acceptable. That framework was not read directly from a primary source in this review, so treat the thresholds as a question to put to your regulatory lead rather than a settled limit.

Sterility and endotoxin are the two attributes most often left until late, and both have arithmetic you can run now. The endotoxin limit formula L = K/M uses K = 5 EU/kg for most parenterals and 0.2 EU/kg for intrathecal products, with M the maximum bolus dose per kg per hour (ChemVerify’s summary of the USP bacterial endotoxins limit, calculated from the maximum human dose, 2026-04-12). The worked example is worth internalizing: 100 µg of peptide at 3 EU/mg given to a 25 g mouse yields 0.3 Unione Europea, above the 0.125 EU threshold for that dose. A limit that looks comfortable per milligram can fail at the dose you actually intend to use.

Sterility testing follows USP <71> sterility test requirements, which call for two media (Soybean-Casein Digest Medium at 20–25 °C and Fluid Thioglycollate Medium at 30–35 °C), a minimum 14-day incubation, and either membrane filtration through a ≤0.45 µm filter or direct inoculation, with direct inoculation limited to 10% of the media volume. Membrane filtration as the preferred sterility method is the regulatory gold standard where the product permits it: filtrazione, rinsing with inactivating agents such as lecithin, polysorbate 80 or sodium thiosulfate, then culture, with visual examination typically at days 3, 5, 7 E 14. Viscous or non-filterable products can use direct inoculation, but plan the method choice before the GMP batch, non dopo.

Per Suggerimento: Build the package as a matrix, not a report. Columns for attribute, metodo, validation status, acceptance criterion and source of the criterion. A reviewer who can read one table and see that every Phase 1 attribute has a validated method will spend their questions on your chemistry instead of your documentation.

A capability check worth running against any supplier: can they run RP-HPLC or UPLC purity, LC-MS identity and LAL endotoxin testing against the same specification you intend to file, and can they show the validation reports for each? MOL Changes supports peptide analytical workflows across purity, identity and endotoxin testing, and the relevant question is whether the method set maps to your specification rather than whether a certificate exists. Ask for the method-by-attribute matrix and the validation summaries behind it.

Cancello 5: Prendere le decisioni di sviluppo del processo

By the end of this gate, the process is defined, controlled and transferable, not merely reproducible in one laboratory. Process development is what turns a repeatable synthesis into a defensible manufacturing route: fix the parameters the impurity profile depends on, write down hold times and in-process controls, and build the stability package on the conditions the ICH stability conditions an IND-enabling package is built on (Io Q1A(R2), 2003) specify. For a drug substance intended for refrigerated storage, that means long-term storage at 5 °C ± 3 °C per 12 months and accelerated storage at 25 °C ± 2 °C / 60% RH ± 5% RH for six months; the general accelerated condition is 40 °C ± 2 °C / 75% RH ± 5% RH for six months.

Regulatory expectations are moving in the same direction. On 28 Luglio 2026 the FDA issued 17 revised draft product-specific guidances for certain generic peptide products, updating recommendations on ANDA submissions for recombinantly, synthetically or semi-synthetically produced peptides, test della risposta immunitaria innata, the FDA’s revised draft guidance on peptide impurity thresholds, higher order structure assessment and biological activity assessment, and is withdrawing its May 2021 guidance on highly purified synthetic peptide products referring to rDNA-origin listed drugs, with a revision planned in 2026. These guidances are ANDA-facing and do not govern IND-stage novel peptides, but they signal where impurity and structure expectations are heading. The failure mode is a process that passes in one facility and cannot be transferred, because the parameters controlling the impurity profile were never written down.

Per Suggerimento: Keep a single process-development decision record that lists every fixed parameter, hold time, in-process control and acceptance criterion, so a receiving site can reproduce the impurity profile rather than rediscover it.

Errori comuni che bloccano un programma peptidico

The most consequential mistake is qualifying a supplier on a single best-batch certificate of analysis. A CoA proves one lot passed; it says nothing about whether the next ten will. Ask for a lot-to-lot trend across at least three consecutive campaigns before you treat a vendor as qualified. The failure mode is silent: release testing passes on the qualification lot, then drifts past specification at the scale you actually need.

The second mistake is treating a research-scale route as scale-ready. Solid-phase synthesis stops scaling cleanly at the point where resin loading, coupling efficiency and purification yield stop moving together, and a gram-scale route tells you nothing about that boundary. Demand kilogram data, or at minimum a documented scale-up plan with the parameters that will be re-verified.

The third is assuming a modification is controlled because it was made. Making a conjugate is not the same as controlling it. Without a method that resolves the modified from the unmodified species, you cannot tell a controlled process from a lucky one.

The fourth is building the analytical package to a purity number rather than to the attribute-level method set. Io Q2(R2) sets validation characteristics by procedure purpose, so a single purity assay cannot cover identity, potenza, impurities and stability-indicating change.

The fifth is deferring stability until after the IND-enabling package is assembled. Stability defines the shelf life and the retest period, and it takes calendar time you cannot compress later. Start it while the process is still being locked.

One honest limitation: the public TwoStep record evidences the financing and the IND clearance. The readiness claims that follow from it are inferred from the framework, not disclosed by the company.

Risultati: che aspetto ha un candidato pronto per l'IND

Peptide development readiness is not a feeling about a promising sequence. It is a set of artifacts your team can hand to an investor, a partner or a regulator without a rewrite. The Series A coverage corroborating the size and date of TwoStep’s $62.5M raise (Yahoo Finance, 2026-09-10) shows capital arriving before the CMC package exists. Peptide IND-enabling studies are what close that gap.

[VISUAL: before-after, “promising sequence” (sequenza, one research-scale batch, a single CoA) beside “IND-ready candidate” (reproducible lot-to-lot data, kilogram-scale evidence, a modification control table, a method-by-attribute analytical package, a defined and transferable process with a stability package)]

Structure the checklist around ICH Q6B’s specification framework: identità, purity and impurities, potenza, with acceptance criteria set from process capability, method performance and product characterization. Q6B was written for biotechnological products, so treat it as a framework rather than a peptide-specific rule. The stability data should already sit at the ICH stability conditions an IND-enabling package is built on: refrigerated long-term at 5 °C ± 3 °C per 12 mesi, and accelerated at 25 °C ± 2 °C / 60% RH ± 5% RH per 6 mesi. These figures come from a summary of Q1A(R2) rather than the guideline PDF, so confirm them on the guideline page before you cite them internally. Produzione di peptidi

Success looks like lot-to-lot purity inside specification, impurity thresholds at or below 0.1% with a stated justification, methods validated to ICH Q2(R2) acceptance ranges, endotoxin below the K/M limit, and sterility passing USP <71>. The stretch goal is extending the package toward higher order structure and biological activity, the assessments FDA’s 2026 draft guidances now emphasize.

Quantitative benchmarks of this kind vary by source and by program, and the figures above are drawn from vendor and legacy literature rather than audited performance data. Nothing here is regulatory advice or a substitute for your program’s own regulatory strategy.

Prossimi passi

If the five gates above map cleanly onto your program, the next useful conversation is a technical one, not a commercial one. Bring the artifacts this framework asks for: lot-to-lot purity data across at least three consecutive lots, the modification control table with its release limits, and the method-by-attribute analytical package with its qualification status. A method-level review of those three items usually surfaces the gaps that would otherwise appear during an IND review cycle, when they are far more expensive to close.

MOL Changes publishes this article as a peptide manufacturer with a commercial interest in peptide quality standards. The framework above is drawn from primary regulatory documents and the cited sources, not from the company’s own data.

Nothing in this article constitutes regulatory advice or a substitute for your program’s own regulatory strategy. Consult a qualified professional before making regulatory or CMC decisions.

For the primary documents, read FDA’s guidance on INDs for phase 1 studies of drugs, the ICH Q2(R2) analytical validation guideline, ICH Q6B specifications for biotechnological products, Io Q1A(R2) test di stabilità, and the United States Pharmacopeia chapters on sterility and bacterial endotoxins. If you would like a method-level review of your readiness package, talk to a technical expert about the specific artifacts your program has assembled.