Quality-by-design of nanopharmaceuticals. A state of the art

• ICH published the Guideline document: Q8 ... Host Cell Proteins Time Temperature Fermentation ... pH Protein Characteristics Bu!er Concentration...

0 downloads 54 Views 6MB Size
Quality-by-design of nanopharmaceuticals. A state of the art Thierry Bastogne

To cite this version: Thierry Bastogne. Quality-by-design of nanopharmaceuticals. A state of the art. European Commission JRC Workshop: Bridging communitie in the field of nanomedicine, Sep 2017, Ispra, Italy.

HAL Id: hal-01670000 https://hal.archives-ouvertes.fr/hal-01670000 Submitted on 21 Dec 2017

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Bridging communities in the field of nanomedicine

European Commission, Joint Research Centre (JRC) 27-28 Sep. 2017, Ispra, Italy



Quality-by-Design of Nanopharmaceuticals. A State of the Art T. Bastogne | CRAN CNRS-Univ. Lorraine | INRIA BIGS | CYBERNANO JRC, Ispra, Italy, 27-28 Sep

Contents

1.  QbD in Theory 2.  QbD in Practice (2007-2017) 3.  One perspective…

T Bastogne, JRC-Ispra 27-28 Sep 2017

7 compounds 2 parameters 3 tested values

Biomarker Compound Biomarker Concentration Shell Material Shell Thickness Shell Diameter Shell Ingredient Concentration Active Substance Concentration Active Substance Molecule Core Material Core Size Targeting Molecule Type Targeting Molecule Concentration Coating Ingredient Coating Ingredient Concentration

Unit 1 Emulsifica/on Unit 2 Polymerisa/on Unit 3 Coa/ng Unit 4 Evapora/on Unit 5 Heat

3(7x2) > 4.106

formulations

Sonica/on Time Sonica/on Amplitude Sonica/on Power Polymeriza/on Time Polymeriza/on Temperature Polymeriza/on RPM Coa/ng Concentra/on Coa/ng Time Coa/ng Temperature Evapora/on Time Evapora/on Temperature Evapora/on RPM Heat Time Heat Temperature Packaging Support Packaging Volume

6 production units 3 parameters 3 tested values

3(6x3) > 3.109

nano-products

Unit 6 Packaging

T Bastogne, JRC-Ispra 27-28 Sep 2017

1

Risk Management

A large population of possible different Nano-objects

EFFICACY:

SAFETY:

H0: Nano is not Efficient H1: Nano is Efficient

Prob[Efficacy|Data] ?

H0: Nano is not Toxic H1: Nano is Toxic

Prob[Safety|Data] ?

How to minimize the risks of bad decisions ? Quality-by-Design : an approach to estimate and control those risks ICH Q8,Q9,Q10 T Bastogne, JRC-Ispra 27-28 Sep 2017

2

Historical background Systems Engineering > 1980

•  •  •  •  •  • 

QibD Genesis > 1990

GMP 2002

cGMP 2004

QbD (ICH Q8)

Extensions

> 2005

Aeronautics & Automative Industries : Total Quality Management, Design for Six-Sigma FDA officials realized that biologics and drugs could also stand to benefit from QbD. Concept paper on 21st Century Good Manufacturing Practices. FDA produced a guidance document : « Pharmaceutical cGMPs for the 21st Century » ICH published the Guideline document: Q8 (R2): Pharmaceutical Development. Now adaptation for Biomedical Devices & Analytical Methods*

*S. Chatterjee, QbD Considerations for Analytical Methods - FDA Perspective, IFPAC Annual Meeting, Baltimore, Jan 2013

3

QbD LifeCycle A risk-based project management : •  6 main tasks •  6 main deliverables •  4 go / no go tests

Project Start

QTPP Qualitative Target Product Profile

Profile your Drug

Identify Quality Attributes

Criticity Analysis CQA Critical Quality Attributes

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology Manage Product Lifecycle

T Bastogne, JRC-Ispra 27-28 Sep 2017

4

QbD LifeCycle

Project Start

A risk-based project management : •  6 main tasks •  6 main deliverables •  4 go / no go testing

QTPP Qualitative Target Product Profile

Profile your Drug

Identify Quality Attributes

Criticity Analysis CQA Critical Quality Attributes

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

Life Scientist T Bastogne, JRC-Ispra 27-28 Sep 2017

Data Scientist

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology Manage Product Lifecycle

5

QbD-1: Profile your Nano

Project Start

QTPP Qualitative Target Product Profile

ü  ü  ü  ü  ü  ü  ü  ü  ü 

Name Dosage Form Route Of Administration Dosage Strength Pharmacokinetics Clinical Intended Use Reference Listed Drug Scale Of Production Safety Concerns

Identify Quality Attributes

Quality Target Product Profile T Bastogne, JRC-Ispra 27-28 Sep 2017

Criticity Analysis CQA Critical Quality Attributes

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

QTPP

Profile your Drug

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology Manage Product Lifecycle

6

QbD-2: Quality Attributes ?

Project Start

QTPP Qualitative Target Product Profile

To measure potential consequences we need to define relevant QA QA = physico-chemical or biological property to be controlled to ensure to get the expected quality/safety/efficacy requirement.

Profile your Drug

Identify Quality Attributes

Criticity Analysis CQA Critical Quality Attributes

High Risk

Low Risk QA

QA / CQA ?

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

CQA Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Critical Quality Attributes ?

Project End

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology

How ? Prior Risk Analysis (Failure Mode & Effect Analysis) T Bastogne, JRC-Ispra 27-28 Sep 2017

Manage Product Lifecycle

7

QbD-3: Formulation & Production Factors ? QTPP Qualitative Target Product Profile

Which are the most influent factors that could cause variability of CQA ? Figure 1: Typical bioprocess with a complicated nest of parameters influencing the final drug product Time

Fermentation Media

Temperature

Induction OD

Inoculum Stir Rate

CO2 Evolution

Load Concentration

Flow Rate

Column Height

Extraction

O2 Flow Rate

Resin Load Concentration

Bed Height

Flow Rate

Activity

Additives

Refolding

pH

Raw Materials

Buffer Concentration

Purification

Buffer Concentration

Diafiltration

Postload Wash

Flow Rate

Residence Time

Product Formulation & Process Management

Protein Characteristics

pH

Conductivity

CQA Critical Quality Attributes

Temperature

Pressure

Separation

Glucose Concentration

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Stability Quality

Filtration

pH

Host Cell Proteins

Drug

Cost

Determine Region of Quality/Safety

Purity

CPP Critical Process Parameters

Model Quality ?

Figure 2: Failure mode and effect analysis (FMEA) output

DS Design Space

CMA CPP Critical Material Attributes Critical Process Parameters n

n

io

io

at

at

ltr

ltr

Fi

#2 n

afi

m

Parameters with low RPN scores (low risk parameters)

Di

n

#1 lu

m lu Co

Co

n ti o pa

ra

Refold

Se

Pr

Se

ed

Fe

rm

en

to

r

Inoculum

Fe o d u rm c e n ti o n to r

Parameters with high RPN scores (high risk parameters) need further investigation

Process Stage

How ? Design of Experiments for Factor Screening T Bastogne, JRC-Ispra 27-28 Sep 2017

Criticity Analysis

Operator

Volume

Flow Rate

Fermentation

Profile your Drug

Identify Quality Attributes

Shaker Position

Risk Priority Number (RPN)

tHBJONBYJNVNJOGPSNBUJPOGSPN a minimum number of experiments tTUVEZFGGFDUTJOEJWJEVBMMZCZ varying all operating parameters simultaneously tUBLFBDDPVOUPGWBSJBCJMJUZJO experiments, operators, raw materials, or processes themselves tJEFOUJGZJOUFSBDUJPOTBNPOH process parameters, unlike with onefactor-at-a-time experiments tDIBSBDUFSJ[FBDDFQUBCMFSBOHFTPG key and critical process parameters contributing to identification of a design space, which helps to provide an “assurance of quality.” Proper execution of DoE within a design space is safe under QbD in bioprocess industries because work within a design space is not considered a change (1). However, some pitfalls can lead to a poorly defined design space. They can come from unexpected results, failure to take account of variability (due to assay, operator, or raw material) within a process, and the choice of parameters and their ranges considered in an experimental study, as well as errors in statistical analysis (e.g., model selection, residual analysis, transformation of response). We present here some good industrial practices based on our experience, on literature for the application of a DoE approach in bioprocess industries, and on nonbiotechnological industrial approaches (e.g., the oil and chemical industries, in which DoE and similar statistical techniques have been applied for many years). Setting “SMART” Objectives: It is

Project Start

Project End

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology Manage Product Lifecycle

Figure 3: Cause-and-effect (fishbone) diagram

Wash Load pH Concentration Conductivity

Bed Height pH Volume Temperature

pH

Equilibration

Conductivity Flow Rate CV

Temperature Conductivity pH Gradient Slope End Buffer Flow Rate Start Buffer

Column #1

Purity Concentration Yield

8

QbD-4: Design Space ?

Project Start

QTPP Qualitative Target Product Profile

CQA = f(CMA,CPP)

Profile your Drug

Identify Quality Attributes

Criticity Analysis

CMA

CQA Critical Quality Attributes Product Formulation & Process Management

Design Space

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology

CPP

How ? Design of Experiments for Response Surface Modeling T Bastogne, JRC-Ispra 27-28 Sep 2017

Manage Product Lifecycle

9

QbD-5: Control Strategy ?

Project Start

Control ?

QTPP Qualitative Target Product Profile

Profile your Drug

Process Identify Quality Attributes

Criticity Analysis CQA Critical Quality Attributes

CMA

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology

CPP

Manage Product Lifecycle

How ? Statistical Process Control T Bastogne, JRC-Ispra 27-28 Sep 2017

10

QbD-6: Product LifeCycle Management Control ?

Project Start

QTPP Qualitative Target Product Profile

Profile your Drug

Process Identify Quality Attributes

Criticity Analysis CQA Critical Quality Attributes

CMA

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

Requirement Assessment

Establish Control Strategy PAT Process Analytical Technology

CPP

Manage Product Lifecycle

How ? PLM Methods (Product LifeCycle Management) T Bastogne, JRC-Ispra 27-28 Sep 2017

11

In Practice ?

12

In practice ? •  •  •  • 

Bibliographic engine: Web of Science Keywords: nano, quality-by-design & drug delivery Replication: every 6 months 30 identified articles between 2007 and 2017

QdD Articles in Nanomedicine

9 8 7

T. Bastogne, “Quality-by-design of nanopharmaceuticals - A state of the art,” Nanomedicine: Nanotechnology, Biology, and Medicine. June 2017.

6 5 4 3 2 1 0 2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

This work was supported by the European Union and the ERA-NET framework under the EuroNanoMed II project NanoBiT. T Bastogne, JRC-Ispra 27-28 Sep 2017

13

Where in practice ? 1.  2.  3.  4. 

Asia (44%) USA (28%) Europ (15%) Africa & Middle East (13%)

Middle East Africa

Asia

T Bastogne, JRC-Ispra 27-28 Sep 2017

EU

USA

14

Nanoparticles were prepared using modified ionic gelation method,[21,22] where CS was dissolved in 1% acetic acid solution to a various concentration and TPP was dissolved in distilled water with various concentrations, based on the results of preliminary study. RZT was uniformly dispersed in TPP solution and this solution was added drop-wise to CS solution under continuous stirring at room temperature. RZT CS nanoparticles formed based on the principle of electrostatic attraction between positively charged primary amino groups on CS chains and charged polyanions (TPP). RZT CS nanoparticles were centrifuged Zidan et al Dovepress at 6000 rpm for 30 min (Remi R-88). The supernatant liquid Table 2 QTPP and CQA of target drug product,was gel with polymeric nanoemulsified particles, forwere injection separated and nanoparticles redispersed in PBS QTPP of a gel with polymeric nanoemulsified particles at pH 6.8 and ultrasonicated for 5 min to disaggregate the QTPP elements Target CS nanoparticles. Three nanoparticles Justification optimized batches,

1) QTPP

•  Frequency: 5/30 (16.7%) •  Since 2015

Dosage form Route of administration Dosage strength Dosage form design Pharmacokinetics Stability Drug product quality attributes

Container closure system

Alternative methods of administration

Table 1: QTPP for RZT CS nanoparticles Profile component Target Dosage form Dosage design Particle size (nm) Entrapment efficiency (%)

Drug release (h)

Justification

Nanoparticles

Novel dosage form for targeted drug delivery Sustained release For long-term nanoparticles treatment of RZT 350-650 Narrow distribution >50 Higher entrapment is better for the nanoparticulate dosage form >48 To achieve sustained drug release for long period of time

RZT: Rizatriptan, QTPP: Quality target product profile, CS: Chitosan

Hydrogel

Pharmaceutical equivalence requirement: same dosage form Injection Pharmaceutical equivalence requirement: same Journal of Advanced Pharmaceutical Technology & Research | Jul-Sep route of administration % of drug substance (% w/w) Pharmaceutical equivalence requirement: same dosage strength Polymeric nanoemulsified carriers incorporated Match reference-listed drug product into hydrogel Bioequivalent to reference-listed drug Match reference-listed drug product Equivalent or longer shelf life compared to Shelf life not 24 months at room temperature reference-listed drug product Physical attributes, identification, assay, Pharmaceutical equivalence requirement: fulfill the uniformity of content, degradation products, same quality standards as reference-listed drug residual solvents, dissolution, microbiological product quality, pH, and rheological behavior Suitable container closure system that will Vials or prefilled syringes, similar with referencesupport estimated shelf life and drug product listed drug product, acceptable for the patient integrity during the transport, Identical primary packaging as reference-listed drug product No None are listed on reference drug product labeling

A.E. Shirsat & S.S. Chitlange, 2015

2015 | Vol 6 | Issue 3

89

A.S. Zidan, 2016

CQAs of gel with polymeric nanoemulsified particles Drug product quality attributes

Target

Physical attributes T Bastogne, JRC-Ispra 27-28 Sep 2017

Is it a CQA? Justification

15

T Bastogne, JRC-Ispra 27-28 Sep 2017

T. Bastogne, 2017.

Size Encapsulation Effic Polydispersity Index Zeta Potential Amount of Release Dissolution Rate Stability Process Yield Emulsification Rate Form PK efficiency Turbidity Lag time Cytotoxicity pH Moisture Crystallinity Enthalpy Permeation Milling Time Aggregation

0%

25%

20

75%

60

Cumulative Percentage

50%

40

5 main Critical Quality Attributes (70%) 1.  NP Size 2.  Encapsulation Efficiency 3.  Polydispersity Index 4.  Zeta Potential 5.  Amount of Release 0

Frequency

100%

80

2) CQA Specification

16

50% 25% 0% Size Load. Particles

pH of Aq. Phase

Surfactant Type

Ingredient Type

Surfactant Conc.

Drug Load

2−Ingredients Ratio

Ingredient Conc.

T Bastogne, JRC-Ispra 27-28 Sep 2017

Cumulative Percentage

75%

30 20 10 0

Frequency

6 Criticial Material Attributes > 90% 1.  Ingredient Concentration 2.  Ingredients Ratio 3.  Drug Load 4.  Surfactant Concentration 5.  Ingredient Type 6.  Surfactant Type

100%

3) CMA Specification

17

T Bastogne, JRC-Ispra 27-28 Sep 2017 Screw Speed

Pump Speed

Injection Rate

Homogenization Pres.

Flow

Barrel Temperature

Aspiration Rate

Stirring Rate

Sonication Time

Processing Temp.

0%

25%

5

75% Cumulative Percentage

50%

10

15

•  No really dominant CPP •  Process dependant

Milling Rot. Speed

Bead Size

0

Frequency

100%

4) CPP Specification

18

56

B. Shah, D. Khunt, H. Bhatt et al. / European Journal of Pharmaceutical Sciences 78 (2015) 54–66

5) Prior Risk Analysis Table 1 Solubility study (n = 3). Lipids

Melting point (!C)

Drug: lipid ratio (D: L) 1:2

1:3

+ ++ ++ ++ +

++ +++ +++ +++ ++

•  Frequency: 5/30 (16.7%) •  Since 2015 Apifil Compritol GMS PA Stearic acid

62–65 65–77 55–60 52–55 69–70

+ Not clear, ++ Turbid, +++ Clear.

2.5. Quality target product profile (QTPP) and risk analysis of RHT SLN N.K. Garg et al. / International Journal of Pharmaceutics 517 (2017) 413–431

415

The QTPP is described as the quality properties that a drug proFig. 1. Ishikawa diagram illustrating CPP affecting on CQA of RHT SLN. Table 1 duct need to possess so as to fulfill the objectives set in target proInitial risk assessment for ACE-NLCs. duct profile as quantitative attributes. QTPP should furnish a quantitative surrogate to describe the aspects of clinical safety have either positive sign, indicating synergistic effect or negative and efficacy by determining the CQA, CPP and control strategy sign, indicating antagonistic effect. Best fitting experimental model (ICH, 2009). (linear, two factor interaction, quadratic and cubic model) was In case of RHT SLN, QTPP is a lower size and PDI with lipidic core taken statistically on the basis of comparison of several statistical is expected to facilitate transport of drug across the nasal mucosal parameters like coefficient of variation (CV), multiple correlation barriers both into the cerebral tissues and systemic circulation. coefficient (R2), adjusted multiple correlation coefficient (adjusted Lower PDI is to reduce aggregation of particle during long term staR2), predicted residual sum of square and graphically by 3D bility. Higher entrapment efficiency is to achieve higher drug loadresponse surface plot provided by Design Expert software. The ing in lipid matrix (Vora et al., 2013). The crucial step in risk level of significance was considered at p-value < 0.05. The regresassessment is to gather the entire responsible factor systematically sion analysis, linear regression plots (observed versus predicted that could influence the desired product quality. These factors value) and Pareto chart of the dependent variables were plotted were categorized hierarchically using an Ishikawa diagram High risk parameter, Medium risk parameter, Low risk parameter. using MS-Excel. (Fig.1). The parameters summarized in Ishikawa diagram assisted (Singh 2.2.4. Pseudoternary phase diagram et al.,formulation. 2011). As per the design all the prepared formulations in the identification of failure modes of SLN T Bastogne, JRC-Ispra 27-28 Sep 2017 19 2.8. Data optimization and model validation

B. Shah et al.,2015

Criticity = Severity x Frequency

N.K. Garg et al., 2017

T Bastogne, JRC-Ispra 27-28 Sep 2017 pH meter

Sec. Ion Mass Spectro.

Viscosity

Turbidimeter

IW Part. Size Analyser

Phase Contr. Microsc.

Laser Diffractometer

Flow Cytometer

Atomic Force Microsc.

Laser Doppler Velocim.

Fourier InfraR Spectro.

X−ray Diffraction

Scan. Electr. Microsc.

Diff. Scan. Calorimet.

Fluo Plate Reader

X−ray Diffraction

Trans. Electr. Microsc.

HPLC

Dyn. Light Scattering

0%

0

25%

20

75% Cumulative Percentage

50%

40

60

4 mian measurement techno. > 50% 1.  Dyn. Light Scaterring 2.  HPLC 3.  Trans. Electro. Microscopy 4.  X-Ray Diffraction Frequency

100%

Measurement Technologies

20

Design of Experiments

B: TPP (%)

Estimate

2.21

1.23 P-value 0.2771 0.0489 X5 Design-Expert® Software Estimate

0.60

2.27 Factor Coding: Actual 0.0096 P-value 0.7439 Overlay Plot X6 Estimate

6.55

0.50 P-value 0.0294 0.2813 Particle Size X7 CI Low Estimate

2.70

1.96 CI High P-value 0.2040 0.0147 Entrapment Efficiency X8 CI Low Estimate 0.10

2.56 CI High P-value 0.2222 0.8084 Analysis of variance Design Points DF 8 8 SS 2,608.02 191.50 MS 326.00 23.94 X1 = A: Chitosan F-ratio 9.77 13.52 X2 = B: TPP 0.0436 0.0277 Prob  F R2 0.9630 0.9777

•  Many inconsistencies between DoE methods and objectives

13.20 0.3454 3.68

2.84

2.00

1.16

0.023 0.3670

7.97 Particle Size CI: 0.5485

650

4.683 0.0441

2.042 0.0302

0.001 Overlay Plot

3.217 0.9712 0.1050

1.508 0.0639

35.87 0.0561

0.014 0.5523

41.08 0.0402

0.011 0.6450

2.883 0.1313 Entrapment Efficiency CI:

Entrapment Efficiency CI: 50 Entrapment Efficiency: 50

8 Particle Size: 350 102,978.02 12,872.30 7.68 0.0406 0.9534

1.100 0.4889

6 1.517 Particle Size: 350

8 0.030 0.004 0.695 0.6982 0.6495

2.175 0.0256

3.908 0.0050

Particle Size CI: 350

0.016 0.5099

14.92 0.2961

67.15

0.3575

1.325 0.0860

8 1,944.723 243.090 10.358 0.0402 0.9650

8 347.433 43.429 13.122 0.0289 0.9722

0.32 Actual Factor 1.32 response. aX1–X82.16 3 3.84 4.68and Notes: Bold values reflect significant factors that affect the corresponding are Eudragit S100, HP-B-CD, and drug loadings (mg), volumes of organic C: Stirring Speed = 900.00 aqueous phases (mL), ultrasonication time (s) and amplitude (%), and level of MgCl2 as a stabilizer (%), respectively. Abbreviations: ANOVA, analysis of variance; PDI, polydispersity index; DF, degree of freedom; SS, sum of squares; MS, mean of squares; F-ratio, model mean square A: Chitosan (%) divided by error mean square; Prob  F value, probability of obtaining an F-ratio as large as what is observed; R2, coefficient of multiple determination for predicted versus measured values.

•  A good software is necessary but not Figure 8: Design space for rizatriptan loaded chitosan nanoparticles enough ! Expertise is needed

2015-Shirsat

Table 6: Validation of design space of RZT CS nanoparticles Formulation

•  Confidence of the results requires to code 21 apply strictly validation procedures. 22

23 •  Only 5/30 papers have really implemented a cross-validation step

Composition (% w/v) X2 TPP X1 CS 2.74

3.68

3.30

2.83

2.96

3.17

Response Y1 Y2 Y1 Y2 Y1 Y2

-

Particle size Entrapment efficiency Particle size Entrapment efficiency Particle size Entrapment efficiency

RZT: Rizatriptan, CS: Chitosan, TPP: Tripolyphosphate, SE: Standard error

Predicted value

Experimental value

565.58 62.08 422.42 64.28 479.16 64.21

570.24 63.14 437.21 63.78 491.24 65.8

2016-Zidan

Table 7: Accelerated stability results for RZT CS nanoparticles Batch number

T Bastogne, JRC-Ispra 27-28 Sep 2017

21 22

Entrapment efficiency* (% w/w) Initial 3 months 99.57±2.4

96.54±1.59

Particle size (nm) Initial 3 months

Zeta potential (mV) Initial 3 months

Initial

570.24

+33.56

0.247

584.14

+32.68

21

Figure 3 Pareto charts of the main effects of variables on the investigated responses. 101.52±2.3 98.14±2.45 +35.6 +34.81 Notes: X1–X8 are Eudragit S100, HP-B-CD and drug loadings (mg), volumes of 437.21 organic and aqueous448.75 phases (mL), ultrasonication time (s) and amplitude (%), and level 0.256 of

Project Start

And after ?

Profile your Drug

•  The Design Space is not the ultimate goal. The last part of the QbD lifecyle is totally forgotten. •  No control strategy •  No continuous quality management •  Difficulty to implement on-line measurement technologies •  Another community: production & control engineering T Bastogne, JRC-Ispra 27-28 Sep 2017

Identify Quality Attributes

Criticity Analysis CQA Critical Quality Attributes

Product Formulation & Process Management

Identify Risk Factors

Criticity Analysis CMA Critical Material Attributes

Determine Region of Quality/Safety

CPP Critical Process Parameters

Model Quality ?

DS Design Space

Project End

Requirement Assessment

Establish Control Strategy

Manage Product Lifecycle

?

PAT Process Analytical Technology

22

Conclusion

•  The Quality-by-Design approach is more and more adopted in the nano-community mainly in India and USA. •  Nevertheless, some important parts, e.g. control strategy & quality management, are still ignored. •  Statistical tools exist but they are not always used correctly à educational effort is needed. •  QbD success relies on the synergistic relationships between chemists, physicists, biologists, statisticians and engineers. T Bastogne, JRC-Ispra 27-28 Sep 2017

23

Towards a new Cardio/Neuro-Toxicity Testing Model for Nano-Products •  CiPA1: FDA, HESI, CSRC, SPS, EMA, Health Canada, Japan NIHS, PMDA •  Objective: revise the current guidelines for evaluating a pharmaceutical drugs tendency to induce cardiac arrythmias (ICH S7B). 1

Specificity> 80% Sensitivity> 80% AUC=0.78

0.9 0.8 0.7

+

+

TPR Sensitivity

A

B

0.6 0.5 0.4 0.3 0.2

C

0.1

Multi-Electrode Arrays Impedance-based Assays Patch Clamp

iPSC human-iPSC-Derived Cardiomyocites

Signal Processing Machine Learning

0

SVM Solution ROC Curve

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

FPR 1-Specificity

1.  CiPA: Comprehensive in vitro Proarrythmia Assay 2.  J. D. Strickland, W. R. Lefew, J. Crooks, D. Hall, J. N. Ortenzio, K. Dreher, and T. J. Shafer, “In vitro screening of metal oxide nanoparticles for effects on neural function using cortical networks on microelectrode arrays,” Nanotoxicology, vol. 10, no. 5, pp. 619– 628, 2016. T Bastogne, JRC-Ispra 27-28 Sep 2017

24

Special thanks to my collaborators … •  •  •  •  •  •  •  • 

M. Beckler, L. Doerr, N. Fertig (Nanion, D) [1,4] A. Fouassier (Ncardia, NL-D) [3] L. Guo (Frederick Nat Lab, NIH/ NCI, US) [5] F. Atienzar, A. Deleaunois, J.-P. Valentin (UCB, B) [3] P. Voiriot, A. Durand-Salmon (Cardiabase, F) [2] L. Batista, P. Guyot (Cybernano, F) [1,2,3,4,5] M. Barberi-Heyob (CRAN, CNRS, F) A. Gégout-Petit (INRIA BIGS, F)

EU-NCL Satellite member

[1] L. Bastista, L. Doerr, M. Beckler, N. Fertig, and T. Bastogne, “Coupled impedance & field potential data analysis of in vitro cardiomyocyte assays,” in Proc of the SPS Annual Meeting, (Berlin, Germany), September 24-27 2017. [2] P. Guyot, P. Voiriot, S. Papelier, L. Batista, and T. Bastogne, “A comparison of methods for delineation of wave boundaries in 12 lead ecg,” in Proc of the SPS Annual Meeting, (Berlin, Germany), September 24-27 2017. [3] L. Bastista, T. Bastogne, F. Atienzar, A. Delaunois, and J.-P. Valentin, “A data-driven modeling method to analyze cardiomyocyte impedance data,” in Proc of the SPS Annual Meeting, (Berlin, Germany), September 24-27 2017. [4] P. Guyot, L. Batista, E. Djermoune, J.-M. Moureaux, L. Doerr, M. Beckler, and T. Bastogne, “Compar- ison of compression solutions for impedance and field potential signals of cardiomyocytes,” in Proc of the 44-th Annual Conf. Computing in Cardiology, (Rennes, France), September 24-27 24-27 2017. [5] L. Guo, M. Furniss, J. Hamre, L. Batista, T. Bastogne, Z. Yan, J. Wu, S. Eldridge, and M. Davis, “Assessing functional and structural cardiotoxicity in cultured human ipsccardiomyocytes,” in Proc of the SPS Annual Meeting, (Berlin, Germany), September 24-27 2017.

To sum up …

•  QbD = Hollistic approach of drug development •  From predefinites objectives to full-scale production •  Risk-based approach

A good Tool for QbD is not enough !

•  Guidance ≠ Methodology •  Needs an efficient Collaboration between users •  Requires a Statistical Background •  Prior Risk Analysis •  Design of Experiments •  Multivariate Analysis •  Control Theory

Practibility for Nanomedicine ?