a comprehensive analysis of the fda 510(k) process/media/images...a comprehensive analysis of the...

A Comprehensive Analysis of the FDA 510(k) Process

Industry Practice and the Implications for Reform

John H. Linehan, Ph.D. Northwestern University

Jan B. Pietzsch, Ph.D. Wing Tech Inc.; Stanford University

National Press Club, Washington, D.C.May 24, 2011

Outline

• The Medical Device Industry and Device Development

• Introduction to the Research Study• Objectives and Methodology

• Respondent Characteristics

• Key Findings• Predictability and Interaction with FDA

• Different Impact on Large and Small Companies

• International Comparison

• Observations: Opportunities for Improvement

• Concluding Remarks

2

The Medical Device Industry and Medical Device Development

Medical Device Companies by Size

4Source: US Dept Commerce

Small companies

Medium companies

Large companies

5

Device Development Is an Iterative Process

• Medical device development is a highly iterative process

• Need to improve product continuously through frequent, positive iterations, while avoiding unnecessary iterations

• Efficient planning and execution requires predictable process

Design device / Iterate design

Prototype device

Obtain clinician feedback

Medical Device Development Functions

6

Cross-Functional Management

Marketing

Research and Development

Legal

Regulatory

Reimbursement

Manufacturing & Operations

Quality

Clinical

Sales

Impacts of Regulation on Device Development

Gate1

Pro

ject

Def

initi

on A

ccep

tanc

e/ C

once

pt C

harte

r

Gate2

Initi

al D

esig

n Ac

cept

ance

/ D

evel

opm

ent A

gree

men

t

Gate3

Fina

l Des

ign

Acce

ptan

ce/

Ram

p U

p R

eadi

ness

Gate4

Pro

duct

Lau

nch

Acce

ptan

ce/ L

aunc

h R

eadi

ness

Formulation/

Concept and Feasibility

Phase

Initiation/ Opportunity

and Risk Analysis

Design and Development/

Verification & Validation

Phase

Final Validation/

Product Launch Preparation

Phase

Product Launch and Post-Launch

Assessment

Market Analysis

Competitive Assessment

Financial Review

Legal/ IP Analysis and Filings

Early Risk Assessment

Regulatory and Clinical Path

Initial evaluation of possible

development of commercial product

Definition of design input based on customer needs and technical requirements

XDevelopment of product design and of manufacturing process; verification and

validation

XXFinal validation of manufacturing process;

preparation of product introduction

XXMarket introduction of product; continuous

improvement

D E V E L O P M E N T

Project Core Team Selection

Early Concept Selection

Customer Input / VOC

Initial Design Risk Analysis (dFMEA)

Initiate & Maintain Design History File (DHF)

Product Design Development

Clinical Validation Plan

Supplier Collaboration

Detailed Producibility Analysis

Initial Regulatory Strategy

Regulatory Strategy Update

Reimbursement Strategy Update

Initial Process FMEA (pFMEA)

Clinical Validation

Final Patent Review with R&D

Mfg/ Ops Scale Up

Full Process Qualification

DHF Completion

Obtain Regulatory Approval/ Clearance

dFMEA Update & Review

Market Launch Plan/ Forecast

Update design control docs as needed

Value Proposition Viable &

Sustainable

Technical Feasibility Proven

and Optimized

Product Risks Acceptable

Manufacturing and Value Chain Conf idence

Commercialization Readiness

Design Output meets Targets

Risk Mitigation Conf irmed

Regulatory Submission

Testing Complete

Sales Launch Final Validation

Stable Manufacturing

Process

Business Launch Plan Adjust

Cross-functional

Mgmt

Functional Groups

Marketing

Research & Development

Legal

Regulatory

Reimbursement

Manufacturing & Operations

Quality

Clinical

Gate 1 Decisions

General Project Plan & Timeline

Prototype Analysis

IP Landscape Review & Review of Filings

Initial Reimbursement Strategy

Initiate DFM (Tooling, Fixturing)

Maintain DHF & Project Timeline

Design Verification and Validation

Design Risk Analysis (dFMEA)

Customer Prototype Eval.

Regulatory Submission

Product Branding

Finalize Reimburse-ment Strategy

Begin Process IQ/OQ/PQ/PPQ

Finalize Process IQ/OQ/PQ/PPQ Quality Audits

Process improvmts. as needed

Post-market Surveillance/ MDR

Product improvmts. as needed

Physician training & contd. sales efforts

Update reimbursmt. as needed

Design FreezeGate 2 Decisions Gate 3 Decisions Gate 4 Decisions

Phases

Continued Clinical Validation

Reimbursement Path

Market Opportunity

Basis for Competition

Tech, Regulatory, IP Approach

Feasible

Manageable Risk and Executional

Gaps

Gate 1 Decisions

Customer Prototype Eval

SalesSales Training Reps Attend

Surgical Cases

Phase I Phase II Phase III Phase IV Phase V

Design Outputs = Design Inputs

Patent Review

7

Introduction to the Research Study

8

Study Objective and Methodology

9

• Elicit from those engaged in medical device development, what seems to work well and how the 510(k) regulatory process could be further strengthened.

• Collect comprehensive data set to provide the basis for constructive input to strengthening the process:– Timelines

– Interactions with the agency

– Issues and challenges in current implementation

– Comparison among international regulatory programs

Approach and Study Methodology

Topic Identification

Interviews with 80+ medical device experts (industry and FDA)

Survey design

Two rounds of expert review and prioritization

Data gathering

Online surveyDec. 22–Feb. 22

Analysis and Results

Analysis and results presentation

8/2010 5/2011

10

• Target respondents: – Individuals closely involved with the 510(k) process

– Broad outreach through professional societies, industry groups, and trade media

• Survey Structure:– General part and Device-specific part

– 86 questions total

• Responses:– N=356 respondents total

– Number of respondents varied per question, as not all questions were answered by every respondent

– N per question stated for each question in graphs and appendix

11

Approach and Study Methodology

Years of 510(k)-Related Experience

12

0%

5%

10%

15%

20%

25%

30%

20 or more

15-19 10-14 5-9 2-4 Less than 2

Resp

onde

nts

Years of 510(k)-Related Experience N= 354

26%

14%

21% 21%

10%

7%

Representativeness: Breakdown by Device Type

Type of Device Actual % of FDA Applications

Survey Respondents %

Surgical, Orthopedic, and Restorative Devices 28% 37%

Cardiovascular Devices 13% 23%

Anesthesiology, General Hospital, Infection Control,and Dental Devices 23% 13%

Reproductive, Abdominal, and Radiological Devices 17% 7%

Ophthalmic, Neurological, and ENT Devices 6% 5%

Chemistry and Toxicology Devices 5% 3%

Immunology and Hematology Devices 3% 2%

Microbiology Devices 2% 1%

Other 3% 9%

Actual % FDA applications: Based on all applications to FDA in 2008-2010 (from FDA database); Survey Respondents %: Based on respondent’s statement about device field with most extensive 510(k) experience. 13

FDA’s Internal Assessment compared to Survey Responses

SE

Source: - FDA CDRH, 510(k) Working Group - Preliminary Report and Recommendations, Vol. 1, August 2010.- MDUFA Meeting Report, 2011.

NSE 8%

18%

14* Includes the following responses: De-Novo, Converted to PMA, Other

N= 240

73%

12%Other*

80%

2010

8%

SURVEY


116

15

N= 224

* SE and NSE only. Avg. duration for SE: 204 days (N=179); NSE: 279 days (N=18); Withdrawn: 330 days (N=13), with long tail.

225 211 days*

FY 2010

SURVEY

148

77



2.2 Cycles

16N= 211; SE: 2.1 cycles (N=191); NSE: 2.8 cycles (N=20)

Withdrawals (not included in computation): 2.9 cycles (N=14)

2009 2010

2.1

SURVEYN= 211

2.0


Key Findings Predictability and Interaction with FDA

17

0%

10%

20%

30%

40%

50%

60%

Top of the list One of top 3 factors

Important Not important

Resp

onde

nts

How important are regulatory requirements to your business decision to a major investment in a new product?

Importance of Regulatory Requirements in Decision to Invest in a New Product

18

21%

N= 351

58%

19%

2%

0%

10%

20%

30%

40%

50%

60%

70%

Critically important Important Not of major importance

Resp

onde

nts

For the technologies you have worked on, how important is predictability of the regulatory process in deciding first country for market launch?

Importance of Regulatory Process Predictability for Decision about First Country for Market Launch

19N= 351

68%

20%

12%

Respondents Perceiving Substantive Changes in FDA Review Process

20

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Yes No

Resp

onde

nts

From your experience in the last 3 years, have you perceived any substantive changes in the FDA review process and/or clearance decision of a 510(k) submission?

85%

15%

N= 349

Perceived Changes in FDA’s Requirements

21

0%

10%

20%

30%

40%

50%

60%

Clinical: endpoints, duration,

sample size, post-hoc statistical analyses

Preclinical work Regulation: appropiateness

of 510(k)

"Least burdensome"

Animal studies Labeling Other

Resp

onde

nts

In the last 3 years, in what specific areas have you perceived changes in the FDA's requirements?

N= 337

58% 57% 53%49%

34%

2%

13%

Note: More than one choice possible per respondent.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Yes No Other

Resp

onde

nts

81%

12%7%

Have Guidance Documents been Critical to your Company in Preparing Successful Submissions?

22

Availability of Guidance Document has an Impact on the Ultimate Decision

11%

89%

19%

81%

0%10%20%30%40%50%60%70%80%90%

100%

Total NSE + Withdrawn SE

Device Specific: Guidance Document Existing for Technology

Device Specific: Guidance Document NOT Existing for Technology

23

(N=93)

Total: N=222

(N=129)

Clarity of Preparation Requirements for a 510(k) Submission

24

0%

10%

20%

30%

40%

50%

60%

Very clear/certain Somewhat unclear/uncertain

Very unclear/uncertain

Resp

onde

nts

Based on your understanding, what is the current level of clarity of the requirements for preparation and submission of a 510(k)?N= 354

24%

57%

19%

Respondents Perceiving Differences Between Guidance Document and FDA Review

25

0%

10%

20%

30%

40%

50%

60%

70%

Yes No

Resp

onde

nts

If an appropriate guidance existed and was used by your company during submission of a 510(k), did you perceive any difference between the guidance document and the way the FDA reviewed your submission?

N= 300

72%

28%

Reason for Perceived Difference between Guidance Document and FDA Review

26

87%

6%2%

5%

FDA asked for information beyond that required by the

guidance

FDA indicated that unnecessary information had been provided by the sponsor

Other

FDA indicated that all necessary information had been provided by the sponsor, but the way the information was presented was

deemed inadequate N= 216

9%3%

6%

15%

2%4%

61%

Proportion of Time FDA Followed Through on Matters Discussed/Directed

Never Followed Through

Did not Follow Through in 76-99% of cases




Unspecified

Did Follow Through

27(N=211)

Perceived Difference between Pre-Submission Meeting Discussion and FDA Review

FDA did “generally

follow-through”

Interaction: Questions/Requests for Information

28

38%

4%

2%

43%

13%

Perceived as not adding to

safety and effectiveness

Other

Perceived as “scientifically

justified”

Perceived as not adding to

safety

Perceived as not adding to effectiveness

Percent of Requests for Information Obtained During Days 75-90 of FDA’s 90-day Review Period

29

0%

10%

20%

30%

40%

50%

60%

70%

80%

Received during days 75-90 Other times during review process

75%

25%

Resp

onde

nts

N= 293

Interaction: Respondent’s Perspective

30

39%

Yes No

61%

26%

74%

41%

59%

Shouldhave

Already answered

N= 282

N= 275

27%

73%

Not fully clear

N= 216

N= 260

Key Findings Different Impact on Large and Small Companies

31

Key Differences between Large and Small companies

32

Small Companies

Large Companies

New product (vs. line extension) [%] 72% 35%

SE Decision [%] 61% 88%

NSE Decision [%] 13% 1%

Interaction with FDA during development process

earlier later

Pre-submission meeting with FDA sought 39% 17%

Duration of pre-IDE process [months] 10.8 7.4

Change in lead reviewer [%] 19% 10%

Total avg. review time [days] 330 177

33

Respondents perceive:Small

CompaniesLarge

Companies

Major difference with FDA’s risk assessment [%]

48% 23%

% of FDA requests already answered in original submission

53% 33%

% of FDA requests “scientifically justified” 30% 42%

FDA requests having major effect on time[%]

45% 36%

FDA requests having major or medium effect on financial resources [%]

76% 64%

Key Differences between Large and Small companies

Key Findings International Comparison

34

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 1-2 3-5 6-9 10-19 20-29 30 or more

Com

plet

ed R

evie

ws

Review Time [Months]

US, n=121

Australia, n=48

Canada, n=75

EU=115

Japan, n=27

Comparison of International Review Time from Submission to Clearance/Registration

US

JP

CANEUAUS

Length of review process in months (based on data points for “1-2”, “3-5”, “6-9”, “10-19”, “20-29”, “30+ months” for the various regulatory systems. N per country: see above. Graph shows ultimately cleared/registered devices only. 35

Major Reason to Bring a Device OUS First

36

49%22%

9%5%

Within the last 3 years, if your company chose to first bring to market a specific device OUS, what was the major reason?

N = 201

Unpredictable 510(k)

requirements

Cost of clinical trials

Quicker process

Easier process

International Comparison between EU and US

37

EU US

Considered “most predictable regulatory system” [%]

64% 8%

First regulator/”body” approached to discuss and plan submission [%]

80% 4%

Review time (submission to decision) for products not requiring clinical data [months]

2.7 5.9

Review time (submission to decision) for products requiring clinical data [months]

4.8 13.2

Moving Forward to Foster Innovation and Timely Patient Access to Safe & Effective Technologies

38

Enhance predictability

• Increase number of guidance documents

• Timely update of guidance documents

• Clear and timely communication of new FDA expectations before publication in guidance

Increase process consistency

• Increase training (particularly implementation of current regulations)

• Reduce perceived differences in agency follow-through (by enhanced communication)

• Reduce reviewer turnover39

Opportunities

Ensure efficient review process

• Preparation of clear and complete submissions

• Eliminate repeat requests of information already provided

• Timely access to meetings

• Increased use of interactive review concept

Close gap with international systems

• Continued harmonization efforts (GHTF)

• Sharing best practices (particularly on process side), while acknowledging differences in regulatory requirements

40

Opportunities

Increase attention to specific needs of small companies (while maintaining a level playing field)

• Improve opportunities for interaction

• Provide training support in areas where small companies tend to face particular challenges

Monitor effect of process changes

• Evaluate impact of any process changes through appropriate performance metrics

• Work with industry to monitor process performance over time

41

Opportunities

Respondent-Suggested Metrics to Evaluate Future Changes in the 510(k) Process

42Assuming that the FDA will make changes to the 510(k) clearance process, what primary metrics should be used to evaluate the overall performance of the revised 510(k) process?

0%

10%

20%

30%

40%

50%

60%

Predictability Appropiate alignment of

device risk and review

intensity

Review time Number of device-related

recalls

Other

N=356

Resp

onde

nts

58% 57%51%

22%

9%

Note: More than one choice possible per respondent.

Concluding Remarks

43

Funding Source

44

Outreach Partners

45

Research Team

Investigators:

John H. Linehan, Ph.D.

Jan B. Pietzsch, Ph.D.

Research Team:

Marta G. Zanchi, Ph.D.

Abigail Garner, M.S.

Remy Durand, M.S.

Brett Kuekan, M.S.

46

Study website @ www.510k.net

47

Resource Center @ www.510k.net

• 510(k) Basics

• FDA, Government and Medical DevicesCDRH, ODE and OIVD documents, Medical Device User Fee and Modernization Act (MDUFMA) and US House of Representatives: Committee on Energy and Commerce

• FDA Guidance Documents relating to 510(k) regulatory process

• Workshops & Conferences - Webinars, TownHall and Public mtgs

• Literature - published articles pertaining to 510(k) process

• FDA Training and Continuing Education Courses

• Institute of Medicine of the National Academies (IOM)Links to agendas, webcast, presentations and reports from Meetings 1, 2 and 3 relating to 510(k)

• International Regulations 48

Appendix:Background and Additional Data

49

Previous study forming Basis for current Work:“Medical Device Development Models”

50

Journal Articles:

• Review of the U.S. Medical Device Regulation

J. Med. Devices, 283 – 292, 2007

• Stage-Gate Process for the Development of Medical Devices

J. Med. Devices, 021004-1 to 021004-152009

Pietzsch, J. B., Pate-Cornell, M. E., Yock, P. G., Aquino, L. M., Linehan, J. H.

Respondent Characteristics

51

0%

5%

10%

15%

20%

25%

30%

5 or less 6-9 10-19 20-29 30 or more

Resp

onde

nts

Number of 510(k)s during career

510(k) Submissions Involved - During Career

52

17% 17%

27%

12%

27%

How many 510(k) submissions have you been involved with over the course of your career? N= 353

510(k) Submissions -- Last 3 years

53

0%

5%

10%

15%

20%

25%

30%

35%

1-2 3-5 6-9 11-19 20 or more

Resp

onde

nts

Number of 510(k)s submitted over last 3 years

3%

16%

22%

36%

23%

N= 349

Employment Status

54

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Device manufacturer:

Large (500+ employees)

Device manufacturer:

Small (<100 employees)

Device manufacturer: Medium (100-

499 employees)

Regulatory consultant

Other

Resp

onde

nts

46%

26%

12%8% 9%

N= 351

Timelines and Performance Metrics

55

Development Process Time

• The results show that the average length of the product development process is slightly less than two years (21.7 months). (N=216) (Definition of product development process:: From initiation of formal design controls to submission of the 510(k) application; for line extensions and changes: from begin of R&D work for changes to submission of the 510(k))

• For devices that ultimately received clearance, the development process length is stated as 18.5 months (N=168), compared to almost double the time (34.6 months (N=31)) in devices that ultimately received a NSE decision or were withdrawn by the sponsor (a statistically significant difference).

56

Differences in Review Time between ODE, OIVD

• Survey results do not suggest statistically significant differences between review times among ODE divisions, and between ODE and OIVD

57

Predictability and Interaction

58

Importance of Regulatory Requirements for Business Decisions

• 98% of respondents state that regulatory requirements are either important or very important to their business decision to a major investment in a new product. 21% state it is the top of the list in their investment decision making, and 58% state it is one of the top 3 factors for their decision-making. (N=265)

• Respondents of small firms and those with longer industry experience rank the importance generally higher than their peers. (Basis: N=73 for small companies; N=121 for large companies)

59

Perceived Changes and Response

• 85% of the respondents have perceived substantive changes in the FDA review process over the course of the last 3 years. (N=349)

• 64% of respondents state that the product development R&D process at their company has changed because of 510(k) requirements. (N=260)

60

Respondents Indicating “Internal Guidelines” Cited as Justification

61

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Yes No

Resp

onde

nts

In working with a reviewer, has the reviewer cited "internal guidelines" to justify a request for information?

Clarity of Guidance Documents Related to Modification/Design Changes

62

0%

10%

20%

30%

40%

50%

60%

Yes No

Resp

onde

nts

Regarding modification and design changes: Do current guidance documents ("Deciding when to submit...") provide reasonable clarity about when to file a new 510(k) application vs. add a 'letter-to-file'?

Respondents Indicating Changes in R&D Process Due to 510(k) Requirements

63

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Yes No

Resp

onde

nts

Over the last 3 years, has the product development R&D process at your company changed because of 510(k) requirements?

Predictability & Guidance Documents

• Only 41% of respondents stated that guidance existed for their device. (N=239)

• 81% of respondents felt that specific requests from the FDA were unexpected. (N=351)

• Respondents who felt that requirements were very unclear/uncertain, were substantially more likely to receive a NSE decision or to withdraw their application (30%) compared to those who felt requirements were very clear/certain (9.1%). (Basis: N=40 for unclear; N=55 for very certain/clear)

• Of those who ultimately received a NSE decision or decided to withdraw their application, 71% said that there was not a guidance document for their device. (N=34)

64

Predictability & Guidance Documents

• For modification and design changes: 41% of respondents state that current guidance documents do not provide reasonable clarity about when to file a new 510(k) compared to "letter-to-file". (N=352)

• 31% of respondents do not feel confident (7% not confident at all, 24% not very confident) about correctly deciding when to use a special 510(k) vs. a traditional 510(k). (N=352)

65

Predictability

• More than half of the respondents (57%) felt that their own risk assessment differed from that of the FDA (32% state a major difference; 25% a minor difference). (N=230)

• Where major differences were perceived, 33% of the applications ultimately resulted in a NSE decision or withdrawal (Basis: N=73), compared to 4.0% in cases where no difference in risk assessment was perceived (Basis: N=99).

• Of those who received requests for additional information from the FDA, respondents felt that the requests were "scientifically justified" in only 38% of cases. (N=248)

66

Interaction: Pre-Submission Meetings

• 31% of respondents sought interaction with the FDA prior to submitting their application. Respondents with less 510(k) experience (<5 yrs.) sought interactions only half as often as respondents with long-term experience (>15 years), a statistically significant difference. (N=242)

• Respondents who sought a pre-submission meeting with the FDA state they were able to obtain a meeting in a reasonable time frame in 43% of cases. In 57% of the cases, it was perceived as difficult to obtain a meeting. (N=221)

• Of those who ultimately received a NSE decision or withdrew their application, 47% had interacted with the FDA prior to submission. (N=225)

67

• 61% of respondents feel that FDA did generally follow-through on matters discussed and directed at pre-submission meetings (N=211)

– Large companies evaluated the FDA more positively in this regard (74% vs. 54%, statistically significant)

– Less experienced responders (<5 yrs. 510(k) experience) evaluated FDA more positively in this regard than more experienced responders (>15 yrs. experience) (63% vs. 51%)

– For respondents who state that the FDA did generally not follow through, differences exist in the assessment of the degree to which FDA does not follow through. Large companies state that FDA did not follow through in 54% of cases, small companies in 72%. (N=82)

Interaction: Pre-Submission Meetings

68

Interaction: Pre-IDE Process

• 48% of respondents whose device required clinical data reported interaction with the FDA in a pre-IDE process. (N=73)

• On average, the pre-IDE interaction period with FDA was 8.6 months long. (N=35)

• Respondents from large companies tend to experience shorter duration of the pre-IDE process (7.4 months) than respondents from small companies (10.8 months). (N=23)

69

Interaction: Meetings during Review

• Respondents who reported difficulty in obtaining a meeting in a reasonable timeframe during the 510(k) review also reported statistically significantly higher NSE/withdrawal rates (23%) compared to those able to obtain meeting in reasonable time frame (11%). (N=222)

70

Interaction: Change in Lead Reviewer

• Respondents experienced changes in the lead reviewer in approx. 14% of their submissions in the last three years. (N=346)

• There is a statistically significant difference between small and large companies, where small companies experience double the turnover that large companies do (19.3% vs. 9.7%). (Basis: N=155 large comp.; N=91 small comp.)

• 62% of the respondents who did experience a change in the lead reviewer feel that it had a somewhat negative or very negative impact on their review process. (N=194)

71

Interaction: Questions/Requests for Information

• Of those who received requests for additional information from FDA, respondents felt that the requests were "scientifically justified" in only 38% of cases. (N=248)

• Of those requests that were deemed "not scientifically justified" by respondents: – 10% were qualified as such because the data had already been

provided

– 72% because respondents were of the opinion that the data did not add information regarding safety or effectiveness

• 7% not to safety

• 3% not to effectiveness

• 62% to neither

• A significant difference exists between large company and small company perception (42% vs. 30% scientifically justified). (Basis: N=118 large comp.; N=60 small comp.)

72

Interaction: Requests for Information

• 75% of the total FDA questions during the review were received by sponsors during days 75-90 of the 90-day review cycle; results show that there is a statistically significant correlation between this percentage and the ultimate decision (83% for NSE/Withdrawal vs. 73% for SE). (N=293)

• Respondents state that 27% of the FDA questions they received in the last 3 years were not fully clear to them. (N=282)

• Respondents state that, of the last 10 requests they received by the FDA, 4 of the 10 were already addressed in the original submission. (N=275)

– Small companies report that 53% of requests were already answered, whereas large companies report only 33%, a statistically significant difference. 73

Effect of Requests for Information on Time and Resources

• 41% of respondents state that FDA requests for information had a major effect on time, 32% state a medium effect, and 22% a minor effect. 6% of respondents state no effect on time. (N=195)

• Respondents from small companies feel that requests have a larger effect on time compared to respondents in large companies (45% vs. 36% state major effect). (Basis: N=49 small comp.; N=92 small comp.)

• More than 2/3 of respondents state that FDA requests had a major effect (36%) or medium effect (33%) on financial resources. (N=195)

• As can be expected, this figure is higher for respondents from small companies (76%) compared to respondents from large companies (64%). (Basis: N=49 small comp.; N=92 small comp.) 74

Clinical Data Requirements

75

Clinical Data Requirements

• 34% of respondents state that their most recent 510(k) required clinical data. (N=233)

• 41% of those requiring clinical data state that the predicate technology did not require clinical data. In this cohort, 50% ultimately received a NSE decision or withdrew their application, as compared to 25% in those whose predicate did also require clinical data. (N=76)

• The average number of patients ("n") enrolled in clinical trials was 292 (192 for small comp.; 595 for large comp.) (N=59)

• 74% of the trials were controlled (randomized, historical controls, etc.) (68% in small, 82% in large companies) (N=59)

• The average length of follow-up in these studies was 7.9 months (10.4 mths. in small, 7.3 mths. in large comp.) (N=55) 76

Differences between small and large companies

77

• The survey responses suggest statistically significant differences between product innovation in small companies and large companies:

• Small companies tend to be involved with completely new products twice as often as large companies. Large companies have a significantly higher percentage of line extensions compared to completely new products. (N=241)

• This difference is consistent with a statistically significantly higher number of successful applications of large companies compared to small companies (89% vs. 75%). (N=163)

• The development process in small companies (26.6 months) is significantly longer than the process in large companies (17.7 months). (Basis: N=57 small comp.; N=103 large comp.)

Key Differences observed between large and small companies

78


• Responses suggest small companies generally tend to interact with the agency earlier in the development process than large companies. (Basis: N=29 small comp.; N=23 large comp.)

• Respondents from small companies tend to seek pre-submission meetings with the FDA more often than large companies (38.9% vs. 17.3%). (Basis: N=88 small comp.; N=146 large comp.)

• Respondents from large companies tend to experience shorter duration of the pre-IDE process (7.4 months) than respondents from small companies (10.8 months). (Basis: N=15 small comp.; N=8 large comp.)

• Respondents from large companies report total review times that are statistically significantly shorter (approx. 177 days)) than respondents from small companies (approx. 330 days)). (Basis: N=110 large comp.; N=55 small comp.)

79


• Small company respondents report major differences in risk assessment with the FDA at a rate more than double that of large company respondents (48% vs. 23%), a statistically significant difference. (Basis: N=58 small comp.; N=110 large comp.)

• Respondents from large companies report a statistically significantly higher rate of own predicates compared to respondents from small companies. Higher rates of using other company's predicates as opposed to own predicates is correlated with statistically significantly higher NSE/ withdrawal rate. (Basis: N=52 small comp.; N=106 large comp.)

80

International Comparison

81


• 68% of respondents felt that predictability of the regulatory process was critically important in deciding in which country to launch first. (N=351)

• 64% of respondents felt that the EU's CE-marking process is the most predictable regulatory system, compared to 8% for US FDA's process. (N=351)

• 80% of respondents state that the first regulator/”body” they approached to discuss and plan their submission was one of the EU notified bodies; only 4% approached the FDA as the first regulator (Cardiovascular and IVD companies selected US FDA at higher rates of 8% and 16%). (N=167)

82


• A statistically significant difference exists between respondents from small companies (89% EU first) and large companies (72% EU first). (Basis: N=46 small comp.; N=79 large comp.)

• 65% of the devices in the device-specific part of the survey were CE-marked first. (N=192)

• Of the devices that required clinical data, 90% were CE-marked (Europe) first, compared to only 51% of devices not requiring clinical data (statistically significant). (N=62)

• When comparing process length of the CE registration process with the US FDA's clearance process, the data suggest that current process times in the US are more than twice as long as the comparable time in the EU (13.2 vs. 4.8 months for products requiring clinical data; 5.9 vs. 2.7 months for products not requiring clinical data). (Basis: N=115 EU; N=120 US) 83


• Respondents in the device-specific part of the survey report a mean length of 3.3 months for the CE-marking process (from submission to registration/CE-mark). (N=115)

• A statistically significant difference exists between respondents from large companies who report substantially shorter process length (2.1 month) compared to respondents from small companies (5.0 months). (Basis: N=33 small comp.; N=46 large comp.)

• Respondents who, within the last 3 years, chose to bring to market a specific device OUS first, state in 49% of cases unpredictable 510(k) requirements as the major reason , followed by cost of clinical trials (22%), quicker process OUS (9%), and easier process (5%). (N=201)

84

Industry Improvement Opportunities

85


• Respondents state that in 39% of the applications they submitted in the last 3 years, their company could have improved the initial submission. This response is consistent across company sizes; respondents with less experience (<5 yrs.) respond higher rates (46% compared with those with more experience (5-14 yrs.: 36%; >15 yrs. 39%). (N=260)

• Respondents, on average, state that 26% of the FDA reviewers' questions should have been anticipated by the company. There is a statistically significant difference between respondents with longer 510(k) experience (> 15 yrs.) who state that only 21% of questions should have been anticipated, compared to 30% stated by respondents with less experience. (N=216)

86


• Respondents state that, of the last 10 requests they received by the FDA, 4 of the 10 were already addressed in the original submission. This suggests improvement potential for both FDA and industry. (N=275)

87

FDA Improvement Opportunities

88

Reviewer Experience & Training

• 53% of the respondents opined that a correlation between tenure/training of the reviewer and the number of requests for additional info exists. (N=294)

• Those who state a correlation have a statistically significantly higher rate of NSE decision/withdrawal (70% vs. 50% in SE). (N=153)

• Of those that noted a correlation between tenure/training, 92% felt that lower tenure led to more questions. (N=153)

89