How to Cross the Border from R to D?

The Example of Conception of New Medical Devices

Lionel Pazart

INSERM CIC 1431, Besançon University Hospital,

Place Saint Jacques, 25030 Besançon cedex, France

Keywords: Translational Research, Medical Device, Innovation, R&D, Clinical Trials.

Abstract: The border between Research and Development for a new medical devices is often unclear since the process

of development of a new medical device remains non linear, with the need of feedback from trials in clinical

situation to new conception of the product. More importantly, the classification of the different steps of a

project impacts on 1/the identification of right partners for the project, 2/ state aid intensities, generally

lower for activities linked to development than for research related activities 3/impact factor of publication

related to the phase of the project. Sometimes researchers under-estimate these studies because it is thought

that, although essential to set-up new investigation tools, they do not lead to an increase of fundamental

knowledge. However, and especially in the field of medical devices, users have to face specific difficulties

due to the variability of the biological systems under study. Results obtained in translational research often

depend on this variability and new questions or scientific obstacles arise from the confrontation to the real

world. In order to address these new challenges, reverse translational research is required. Fundamental

research is then fed from the results of translational research. In this position paper, we would like to present

a useful model of medical device development through several examples of translational research to

illustrate the adequacy of research to bridging fundamental research results to the closest to the patients.

1 INTRODUCTION

Basic research rarely knows what discovery will

serve and disruptive innovations in health mostly

come from basic research whose authors have not

suspected the consequences (for instance, the

discovery of electron spin in 1922 to the MRI in

70’s). At the opposite, applied research is primarily

directed towards a specific practical objective (for

instance the long story to capture and preserve

images began with the Egyptians some ten thousand

years ago when they noted the ability of light to

transmit images). Between these both enemies’

brothers, experimental development is a systematic

work, using knowledge gained from basic research

and/or practical experience, which is directed to

produce new products or to improve substantially

those already existing. To transform basic research

results into a practical innovation, translational

research needs big efforts to conceive future

application, and requires thinking differently and

changing our mind. Research and Development for a

new medical devices is often unclear since the

process of development of a new medical device

remains non linear, with the need of feedback from

trials in clinical situation to new conception of the

product. Sometimes basic researchers neglect these

further studies because it is thought that, although

essential to set-up innovative technologies, they do

not lead to an increase of scientific knowledge.

However, and especially in the field of medical

devices, users have to face specific difficulties due

to the variability of the biological systems under

study. Variability is easily understood from one

patient to another one. But there is also the

variability of a single patient whose metabolism

evolves naturally by the time and additionally with

the therapeutic actions. Translational research has to

understand these variability and new basic research

questions arise from the confrontation to the real

world. In order to address these new challenges,

reverse translational research is required.

Fundamental research is then fed from the results of

translational research.

In this overview, we would like to present a

useful model of translational research for medical

Pazart L.

How to Cross the Border from R to D? - The Example of Conception of New Medical Devices.

DOI: 10.5220/0006803100010001

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIOSTEC 2015), pages 7-12

ISBN: 978-989-758-071-0

device development through several examples to

illustrate the adequacy of research to bridging

fundamental research results to the closest to the

patients.

2 CLASSIFICATION OF

RESEARCH ACTIVITIES

The Frascati Manual

defines R&D as “creative

work undertaken on a systematic basis in order to

increase the stock of knowledge”. The term R&D

covers in fact three activities: basic research, applied

research and experimental development.

Basic research is experimental or theoretical work

undertaken primarily to acquire new knowledge of

the underlying foundation of phenomena and

observable facts, without any particular application

or use in view.

Applied research is also original investigation

directed primarily towards a specific practical aim.

In technical fields, applied research could be

associated with the ‘industrial research’ for

developing new products or processes.

Experimental development is systematic work,

drawing on existing knowledge gained from

research and/or practical experience, which is

directed to producing new materials, products or

devices, to installing new processes, systems and

services, or to improving substantially those already

produced.

For example, the determination of the amino acid

sequence of an antibody molecule would be basic

research. Investigations undertaken to identify the

right antibody for a specific membrane protein of a

virus would be applied research. Experimental

development would then consist of designing a

biochip functionalized with the right antibody for the

disease on the basis of knowledge of its structure

and clinically testing with biological liquid of

interest (blood, urine, etc.) in order to make the right

diagnosis.

Translational Research involves, for the National

Institutes of Health (NIH), “the extensive body of

work required to move a discovery from bench to

bedside” and Wikipedia definition insists on the

capacity of translational research to shorten the time-

frame to reach the market. Accordingly to these

definitions, translational research covers applied

research and experimental development till the

market launch of the product. Some others add in

healthcare fields the feedback loop “from the bed to

the bench”

3 CHALLENGES OF R&D ON

MEDICAL DEVICES

With the great diversity of the medical devices from

crutches to programmable pacemakers it is not

feasible to subject all medical devices to the same

development scheme. Much specificity of MDs vs

drugs should be taken into account

, and

methodological adaptation could be performed to try

to exercise discretion to propose a feasible

methodology:

• Clinical investigation is particularly needed to get

CE mark for class IIb and III MDs

. First tests in

human need to answer to essential requirements and

to assume the safety of the device by in vitro, on

bench and in vivo (animals) tests. For the other ones

(ex: glove, eyes occlusion plasters, conductive gels,

non-invasive electrodes, image intensifying screens)

predictability of performance could be a useful

manner to answer the question.

• The interest of product may concerns either

therapeutic, diagnostic, or compensation of disability

area and the methodology of clinical assessment

should therefore be adapted classically to the main

objective of the study. For instance, clinical trials of

diagnostic tests are sometimes divided into

exploratory phases, challenge phases and advanced

phases to see how effective and how accurate the

tests are

. In all cases, a distinction should be made

between the clinical proof of efficacy and safety in

order to get the market approval and the place in the

diagnostic or therapeutic armamentarium in order to

define the price or the reimbursement of a MD. In

the latest intent, randomized controlled trials are

generally conducted to compare a new intervention

or strategy to the classical one.

• The level of innovation; should the new MD be

considered as an incremental evolution or an

evolution of rupture ? Minor or incremental changes

on an existing medical device are the most frequent

type of innovation activity in companies. Activities

leading to minor, incremental changes or adaptations

should in principle not be counted as R&D activities,

unless they are part of, or result from, a formal R&D

project in the firm.

• The equivalence with a predicate; substantial

equivalence means that the new device is at least as

safe and effective as the predicate

. This concept can

be applied to many products including high-risk

products, such as coronary stent or hip prosthesis.

To prove the equivalence, technical bench tests and

preclinical study could be done. Production of

specific clinical data could be limited to a cohort

study in order to retrieve the similar results of

predicate. However, this applies only if the

equivalence criteria are not affected claim, clinical

and technical data and environment.

• The operator/MD interaction; the clinical benefit

may depend not only on MD itself but also the

performance of the medical team (operator

dependent nature, learning curve) and the technical

platform, this organizational dimension is an

element which must be taken into account in the

early investigations of a new MD; trials should

incorporate this learning curve by providing a first

acquisition phase, in the number of subjects required

for example, and / or any interim analyzes. Another

possibility is to use a sequential adaptative Two

stages design (i.e Fleming methods)

• The diversity of use; one or more studies are

needed to develop the implementation of a new MD

and describe different operator (medical staff or the

patient himself), operating times, the technical

facilities and personnel skill to the success of the

procedure.

• The reduced life cycle. The clinical assessment

should be realized in short-term monitoring, on

technical and clinical intermediate parameters.

Nevertheless, a long term monitoring should be

performed till failure occurrence for all patients who

were implanted with an old version of MD

(particularly for implantable devices like cardiac

prosthesis, breast implants, cochlear implants etc.).

• The small size of target population. Of particular

methodological solutions can be proposed:

conducting multicentre clinical trials in Europe

(within ECRIN network for instance), or exhaustive

survey of patients through national or international

• The short track of development; a lot of MDs could

be developed with few technical experimental tests

to get the Proof of Concept without clinical test, like

for instance dental impression materials, tubes used

for pumping the stomach, urinary catheters intended

for transient use etc. For other MDs category, the

absence of an animal model to test preclinical MD

and the futility to test it on healthy volunteers

contribute to go quickly to the patient, for instance

for hip prosthesis or implantable analgesic pump.

4 PRACTICAL SITUATIONS OF

TRANSLATIONAL RESEARCH

4.1 “Optical Biopsy”

Invasive biopsy is still today the reference diagnostic

technique of a lot of skin or mucosa pathologies

(inflammation, tumours). Nevertheless, several

situations of diagnosis should be kept as

conservative as possible. Consequently, non-

invasive imaging methods (ultrasounds, computed

tomography, magnetic resonance imaging) have

been developed for clinical use. Based on the

principle of white-light interferometry and

developed initially in 1991 for in-vivo imaging of

the human eye

, OCT was investigated by a large

number of groups worldwide. With regards to

penetration depth and resolution, OCT could be a

perfect trade-off between ultrasound and confocal

microscopy. The use of optically pumped based on

specific swept sources for OCT was first

demonstrated in 2011 but since that time, the

threshold towards the use of low-cost electrically-

pumped devices is still not crossed.

How to translate the basic knowledge to a

practical application in healthcare?

A first way could be to fix the possible

application fields (for instance skin biopsy) and ask

the specialists (here university dermatologists) about

the possible clinical use with the technical

characteristics of the future device concerning the

spatial resolution of the system, the field of view and

imaging magnification. The design parameters will

be selected according to the system specifications

and technological constraints, for instance a

miniature (< 15cm3), low cost OCT imager

providing cross-sectional 3-D tomograms with a

depth around 0.5 mm, axial and transverse

resolutions of 5 µm and imaging field of 5x5 mm2.

Of course, specialists could imagine possible clinical

applications

such as superficial baso-cellular

cancer, follow up healing after an injury or surgery,

assessment of new wound dressing or graft,

determination of the degrees of skin burns, the local

efficacy and tolerance of topical treatment etc. But

the usefulness of such an OCT imager remains

questionable, and clinicians are doubtful of pictures

interpretation since they have no experience

feedback about such imaging. The learning curve is

for the moment very slow with those new technics

(new images, new colors, new field of view…),

which is currently a real limiting factor for the

diffusion of those technologies.

A second approach could be clinical use based

specifications. First of all, dermatologists are invited

to express their will. In this way, they claim for a

new device able to provide detection of early skin

cancer by discerning diseased and healthy skin, and

helping the practitioner to accurately determine the

margins for resection, which is usually affordable by

the examination of the overall architecture of

epidermis and identifying the number of atypical

cells per unit of area.

Figure 1.

Thus, the parameters of new OCT microsystem

design have to be determined by examining the

biomedical application requirements as well as the

instrumental characteristics of selected

interferometric architecture including array-type as

well as high-speed camera requirements. The

Medical ISO13485 methodology requires also a

Risk Analysis of the final product. It must be

initiated with every participant and especially the

future users. A Functional Analysis can then

describe what we expect from the MD and split it in

building blocks.

Current works are trying to improve the

accuracy, resolution, penetration depth of these

devices. Manufacturers and researchers should focus

their insights on the easiness of recording,

measuring and analyzing, the daily practice in doctor

office, their reliability, and the prize.

In summary, the best way could be analyzing the

constraints of available techniques, defining the

needs from the end-user (medical) point of view and

adapting research program to conciliate both

requirements. The following scheme tries to

represent this approach:

Figure 2.

4.2 Screening at Birth

Routine screening by capillary blood sample at birth

concerns several diseases in France. The lateral edge

of the feet was chosen as sampling area by scientific

societies, on an anatomical removal of the main

neurovascular bundles and to avoid the risk of

osteomyelitis of the calcaneus, previously found

with bites to the posterior heel. The method is

painful for the newborn and quantitative failure

often leads to sample a second time.

Figure 3.

How to improve the quality and the capacity of

screening at birth, particularly to reduce newborns’

pain?

The first way consists to search available

techniques on the shelves, then to try to adapt them

to the need.

Micro-needles array appears to be a good solution to

replace the lancet (see picture). This matrix would

be applied on the heel as a patch, the multitude of

micro-needle (deemed not painful) replacing the

wide blade of the lancet.

Figure 4.

But a lot of questions emerge to adapt this

technology to the heel of newborns:

¾ How deep to prick ?

¾ Which density of needle should be

compose the network?

¾ Which size for the channel, if any ?

A better understanding of the distribution of

capillary networks could improve the specifications

for a new device based on micro-needles array. To

acquire these data, we conducted a clinical study

using ultrasound (device Dermacup Atys) and

videocapillaroscopy (device Moritex, MS-500C

Micro-Scopeman) on both sides (lateral and medial

edges) of the heel of 62 newborns according to

gestational age at birth. The parameters of the

microcirculation were obtained by ultrasonography

(depth of dermis) and capillaroscopy: capillary

density and distribution, inter-capillary distance and

average diameter of the capillaries. The results show

that on average, the density of the capillary network

is 60 capillaries / mm2, the inter-capillary distance

of 155 microns and a diameter of 22 microns.

Another result shows that the capillary network is

oriented mainly parallel to the lateral edge of the

foot and less on the medial edge.

From our study, the results of capillaroscopy and

skin ultrasound will help determine the right micro-

needles array configuration as follows:

¾ The area provided for the needle plane is 25

mm², and the number of micro-needles

depend on the density of capillaries and of

the inter-capillary distance

¾ The depth of the dermis specified the

maximum depth of the micro-needles.

After some prototypes adaptation, we performed

several tests on animals. The results show that a

network of 8 micro-needles could be acceptable, and

avoid any “fakir effect” of the skin. But these micro-

needles must penetrate about 1 mm in the heel of the

newborn and three applications of the matrix are

needed to achieve a 96% probability of blood

collection. Under these conditions, it is difficult to

talk about withdrawal without pain.

A second way to solve the problem of pain consists

firstly to understand the mechanisms of this

painful process to provide input for improvement in

terms of medical devices and also to explore new

avenues for screening at birth.

A systematic clinical observation of blood collection

steps at the 72 th hour of life was conducted on a

sample of 50 newborns (PREVMAL study). The

purpose of this observation was to get confirmation

of the frequency of pain and when it appears and on

another hand to understand the factors behind its

occurrence on which further action could be taken.

The anatomical data from videocapillaroscopy and

ultrasound were also collected to be correlated with

results of observation of the act of screening in terms

of pain (DAN scale) and quantity of blood obtained.

89% of newborns have expressed a pain

. It

appeared that the pain is mainly observed when

pressure is applied on the newborn's heel to collect

the blood on the blotter paper and secondarily at the

heel prick with the lancet. So, the most painful is not

the bite but the pressure of the foot (p = 0.0005). The

pain at the sting of the lateral edge appears less

important than the sting of the medial edge, but this

result should be confirmed by a more powerful study

with more cases. No correlation was found between

pain and deep dermis, or density, or the diameter of

the capillary.

Considering these results, we were committed to

finding new methods of blood collection,

particularly to avoid pressure on the heel and the

proliferation of bites to get the blood in sufficient

quantity on the blotter paper. We have therefore

developed a system provided with a micro-machined

nozzle. This tip is applied to the heel, after bite of

the lancet. The blood viscosity properties and the

geometry of the tip make possible to maintain the

blood captive inside a reservoir. This tip is then

stamped on the blotter paper. Tests have shown that

a volume of less than 800 µl of blood sufficient to

properly soak the spaces provided on the blotter

paper.

From clinical needs to fundamental research, this

approach could be summarized as follow:

Figure 5.

To go on with the development of the new device, it

is planned to conduct clinical trials on parallel

groups for ethical reasons, with prototypes and

classic screening methods.

5 CONCLUSION

To conclude, in this position paper we considered

examples of the conception of MD either based on

technology availability or clinical use requirements.

In both approaches, identification of clinical useful

technical characteristics are critical issues to faster

the development of new medical device. As

mentioned in the introduction of the paper,

translational research activities are sometimes under-

estimated and postponed because they do not lead to

an increase of “scientific” knowledge.

However, translational research covers applied

research and experimental development and it is

essential to set-up new tools especially in the field of

medical devices. It isn’t only a question of semantic

as illustrated few years ago in a similar congress, but

a way of thinking.

If we design a clinical study to describe the

capillary network of newborn, without any objective

other than knowledge, this study should be qualified

of basic research. If the same study is intended to

complete technical specifications for a new

screening device, it becomes applied research. The

translation is there, behind the aims of the study

protocol: what might be the usefulness of study

results? Basic Researchers should be aware of this

new paradigm, even if they haven’t to focus their

attention on application anymore. Louis Pasteur

perfectly summarized this necessary connection:

“There is not basic research on a side and applied

research on the other. There are research and

applications thereof, united to each other as the fruit

of the tree is joined to the branch that has worn.”

ACKNOWLEDGEMENTS

The studies presented in this paper are funded by

public grants, under the European Commission's 7th

Framework Program, or French Health Ministry and

French National Agency for Research.

REFERENCES

OECD, Frascati Manual 2002: Proposed Standard Practice

for Surveys on Research and Experimental

Development, The Measurement of Scientific and

Technological Activities, OECD Paris.

Parquin F, Audry A and Giens XXVII N° 1 group

members. Constraints and specificities of the clinical

evaluation of medical devices. Thérapie 2012 Juillet-

Août; 67 (4): 301–309

Annex IX from 93/42/EEC concerning medical devices as

amended by Directive 2007/47/EC

ISO 14155:2011 Clinical investigation of medical devices

for human subjects - Good clinical practice

Food and Drugs Administration, Premarket Notification

(510k)

Fujimoto J. et al., Biomedical Optical Imaging, Oxford

University Press, 2009.

Sattler E, Kästle R, Welzel J: 2013 Optical coherence

tomography in dermatology, J. Biomed. Opt. 18(6),

061224

Robieux C., Sainthilier J.M., Vidal C. & al. Study of the

correlation between the structure of the capillary

network of the heel of newborns and pain during

screening at birth, J Invest Dermatol 2011, 131:2144