VMRFG/FEDESA SURVEY MUTUAL RECOGNITION PROCEDURE ...

C:\WINNT\Profiles\jp\Bureau\FEDESA Survey\FEDESA-VMRFG Survey.doc 25 July 2000

VMRFG/FEDESA SURVEY

MUTUAL RECOGNITION PROCEDURE FOR

VETERINARY MEDICINAL PRODUCTS1999

FINAL, 25 JULY 2000

Incomplete draft issued to VMRFG on 19 April 2000 for information prior to Infoday in May 2000

First draft issued to VMRFG and Reg.Procedures AHG for comments

22 June, 2000

Deadline for comments on first draft 17 July, 2000

Second draft issued for comments andapproval

25 July, 2000

Deadline for approval of second draft 11 September, 2000

VMRFG/FEDESA 1999 SURVEY OF

THE MUTUAL RECOGNITION PROCEDURE

EXECUTIVE SUMMARY

Following the FEDESA survey of the functioning of the Mutual Recognition Procedure (MRP)between 1995 and 1998, FEDESA and the Veterinary Mutual Recognition Facilitation Group(VMRFG) agreed to work together to continue this useful exercise in a joint survey. This reportrepresents the results of the first joint survey for procedures where day 90 fell in 1999.

• There were 27 MR procedures during 1999; for the survey 22 completed questionnaires werereceived from 11 companies. Of these, 5 were for immunological products, all for food-producinganimals. The other 17 were pharmaceutical products for both food-producing (5) and companion(12) animals, including one line extension, one abridged application, and one repeat procedure.

• Two thirds of the MRP applications were covered by just two Reference Member States (RMS)(UK and France), which had 7 procedures each, closely followed by Germany with 6. Denmark andSweden were RMS each on one occasion. The performance of these RMS was considered by theapplicants satisfactory or excellent and the choice of them was mainly determined by either theexistence of an existing national licence and/or previous satisfactory experience with the RMS.

• All 15 members states were involved as CMSs in the MRP during 1999, the average number ofCMSs per procedure being 8. The main reason for the choice was commercial and those MS withno marketing interest or existing national license were excluded.

• The overall level of withdrawals has fallen from 11% in 1998 to 8% in 1999. Withdrawals remainhigh for Denmark and were also a problem for Sweden. This reduction in the number ofwithdrawals reflects the desire on the part of CMS to improve the working of the mutualrecognition procedure where possible.

• The average number of questions being asked by CMSs was 8 per procedure (cf. 8.4 in 1998),giving an average of 65 questions per dossier. Austria, Germany, the Netherlands and Sweden haveshown a reduction in the number of questions. France and UK continue to ask 20 to 70 questionsper procedure. Belgium, Finland, Spain, Denmark and Ireland continue to ask on average morethan 10 questions per procedure. This remains a key area of concern for industry. The initiative toclassify the questions into major, minor and SPC points is considered by the VMRF group as animportant step forward to reduce the number of questions which are likely to lead to arbitration.

• The average time taken to obtain a national licence from Day 90 has increased since 1998 from 58days to 71 days. Counted from the receipt of the final translations, the average time in 1999 was 56days. Due to the incomplete data, this may not reflect the actual average time taken. This is now thekey area of concern for industry as it causes a direct financial loss by delaying product launches.

• The overall view of applicants and RMSs is that the majority of procedures were satisfactory andthere is a high level of satisfaction with the work of each party. However, industry is not satisfiedwith national practices leading to withdrawals, the lack of progress in reducing the number ofquestions at Day 60, and the lack of progress in reducing the average time to obtain a licence.

• The overall view of VMRF Group is that these results show a 92% success for these 22 mutualrecognition procedures. The main reasons for withdrawals are scientific disagreements, and theseshould be resolved by convening meetings between the parties involved. The classification ofquestions and the automatic validation are considered as important steps to improve the procedure.


CONTENTS

1 BACKGROUND ................................................................................................................. 5

2 METHOD............................................................................................................................ 6

2.1 Preparation of 1999 questionnaire ....................................................................................62.2 Data collection procedure ................................................................................................62.3 Data analysis and reporting ..............................................................................................6

3 RESULTS AND COMMENTARY...................................................................................... 6

3.1 Response to the questionnaire...............................................................................................63.2 Reasons for choice of RMS..................................................................................................73.3 National Assessment Procedure by RMS...............................................................................73.4 The Performance of RMSs and Applicants ............................................................................73.5 The Concerned Member States...........................................................................................103.6 Automatic validation..........................................................................................................113.7 Number of questions from CMSs ........................................................................................113.8 Number of questions to each part of the dossier ...................................................................123.9 VMRFG Discussion...........................................................................................................14

3.9.1 Need for a VMRFG Discussion ....................................................................................143.9.2 Performance of VMRFG Discussion.............................................................................14

3.10 Time taken to send translations and to obtain national licences ...........................................153.11 Withdrawals ....................................................................................................................153.12 Reason for withdrawal from a CMS..................................................................................173.13 Major problems reported by applicants and RMSs .............................................................17

3.13.1 Major problems encountered with the Applicants.........................................................173.13.2 Major problems encountered with the Reference Member State ....................................173.13.3 Problems encountered with the Concerned Member States...........................................17

3.14 Evaluation of work of RMS and Applicant (Question 4e and 4f) .........................................183.15 How did the Procedure function? ......................................................................................193.16 Proposals for improvement. ..............................................................................................20

4. DISCUSSION....................................................................................................................... 21

5. CONCLUSION .................................................................................................................... 23


SUMMARY STATISTICS

Data source: 22 completed procedures from 11 companies from 1999.

Product type summary:

1stprocedure

Repeatedprocedure

New activeingredient

LineExtension

Abridged Other Foodanimal

Companionanimal

Pharmaceutical 16 1 5 1 1 1 5 12

Vaccine 5 0 4 0 0 1 5 0

Total 21 1 9 1 1 2 10 12

ANNUAL TRENDS

Member States' involvement in the 22 completed procedures:

Period: 1995 -1998survey

1999survey

Overall

Number of Mutual Recognition Procedures in the surveys: 20 22 42

Number used as Reference Member State (RMS): 5 5 7

Number used as Concerned Member State (CMS): 15 15 15

Average number of CMS per procedure: 8 8 8

Total number of CMS procedures: 149 170 319

Total number of withdrawals: 16 13 29

% Total withdrawn: 11% 7.6% 9%

Data on key issues

Period: 1995 -1998survey

1999survey

Average (and maximum) number of questions per application 55 (193) 65 (254)

Average number of questions (and maximum) per CMS 8.4 (72) 8.0 (73)

Average (and max.) number of days to supply translations N.A. 15 (99)

Average (and max.) number of days from translations to issueof licence

N.A. 55 (180)

Average (and max.) total number of days from day 90 toissue of a licence

71 (312)(1998 = 58)

71 (180)


VMRGFG/FEDESA 1999 SURVEY

MUTUAL RECOGNITION PROCEDURE

VETERINARY MEDICINAL PRODUCTS 1999

1 BACKGROUNDIn April 1999 FEDESA produced its first report on the functioning of the Mutual RecognitionProcedure (MRP) for the registration of veterinary medicinal products. The report was based on asurvey of 20 procedures over the period from Q4 1995 to Q3 1998 (herein referred to as the ‘1998survey’). These procedures were estimated to represent about 1/3 of the procedures that had beencompleted during this time period. The introduction to that report recognised the following pertinentfacts:

• The MRP had been in operation for over four years, and has been obligatory for multi-nationalapplications since January 1st 1998.

• During the three-year transition period, from 1995 to 1998, improvements were made to the earlyprocedural difficulties that were encountered with the more demanding new system.

• The MRP is a considerable improvement on previous procedures, and allows not only for aproduct to be launched early in the RMS, but also offers the opportunity for more rapid access tothe EU as a whole.

• The key to the improvement of any system is an accurate understanding of any problems.Consequently FEDESA had undertaken a survey to investigate the functioning of the procedure,and to set down a benchmark against which progress could be measured.

The 1998 FEDESA survey was invaluable in facilitating an open dialogue with the Veterinary MutualRecognition Facilitation Group (VMRFG) and provided a sound basis for the discussion of both thestrengths and the weaknesses of the MRP. With the underlying causes of the problem areas moreclearly defined, proposals for future improvements could be made.

The 1998 FEDESA report praised both the replacement of the multi-state procedure with the MRP, thecreation of VMRFG, the introduction of their Best Practice Guide and the performance of thereference member states. However it was critical of the attitude of CMS to the concept of ‘mutualrecognition’. The report also highlighted problems with the quality and delivery of translations, thedifferent interpretation of guidelines and legislation, different national requirements and delays inissuing marketing authorisations as particular problem areas.

The report concluded with the recommendation that FEDESA and VMRFG should continue to worktogether in the assessment of the functioning of the MRP. Therefore, in 1999, a jointVMRFG/FEDESA questionnaire was agreed, so that the experience of both the RMS and the applicantcould be collected with full transparency.

This present report has been produced jointly by VMRFG and FEDESA and represents the results ofthis first joint survey.


2 METHOD

2.1 Preparation of 1999 questionnaireThe 1998 FEDESA questionnaire was adapted to experience from the 1998 survey and to includesections for information to be provided by the Reference Member State. In addition it was agreed thatmore detained information should be requested on the number of questions received from CMSs, inparticular the number of major, minor or SPC questions, and the number to each part of the dossier. Inaddition the dates that final translations were sent at the end of the procedure were requested.

2.2 Data collection procedureAfter completion of Day 90 of the MRP, the RMS completed its section of the 1999 questionnairebefore sending it to the applicant. The Applicant completed its section, and then returned a copy ofthe questionnaire to the RMS. A copy was also sent in confidence to FEDESA. In this way both theRMS and the applicant were able to obtain immediate feedback of the views of the other party on thefunctioning of that procedure, and FEDESA was able to collate all the responses and draft a report.

The inclusion criterion was that day 90 of the procedure fell during 1999.

2.3 Data analysis and reportingThe data from the completed questionnaires were tabulated and summarised by FEDESA, andvalidated and corrected by VMRFG. The data were cleaned of all confidential and traceableinformation. A draft report was prepared and agreed through consultation with VMRFG andFEDESA's Regulatory Procedures ad hoc group. The key issues and conclusions of the survey werereported at an EMEA/VMRFG/FEDESA Info day at the EMEA on 18 to 19th May 2000 from both theperspective of industry and the perspective of VMRFG.

3 RESULTS AND COMMENTARY

3.1 Response to the questionnaireCompleted questionnaires were received from 11 international companies covering a total of 22Mutual Recognition Procedures (MRPs). The questionnaire was generally well completed, with theexception of certain sub-sections. In particular several applicants did not report the number of daysthe clock was running or stopped, and the dates that licences were issued were not always available,even from Eudratrack.

The split of the products between livestock and companion animals, and between new and old actives,is given in Table 1. There is an even split between companion animals and livestock, and the numbercontaining new active ingredients is also evenly balanced. However less than half of the proceduresoverall concerned new active ingredients. In the 1998 survey the large majority of the procedures hadconcerned products containing new active ingredients for food-producing animals. The proportion ofimmunological products remains unchanged at just less than 25% of the procedures, and the largemajority (4/5) contained new active ingredients.

Table 1: Summary of the type of products represented in the 1999 survey

1stprocedure

Repeatedprocedure

NewA.I.

LineExtension

Abridged Other Foodanimal

Companionanimal

Pharmaceutical 16 1 5 1 1 1 5 12

Vaccine 5 0 4 0 0 1 5 0

Total 21 1 9 1 1 2 10 12


3.2 Reasons for choice of RMSThe choice of RMS was mainly determined either by the existence of a recent national licence and/or aprevious satisfactory experience with the RMS. The reputation of the RMS also played an importantpart. A summary of the reasons provided by the applicants may be found in Table 2.

Table 2: Reason for choice of Reference Member State

Number Reason9 First member state to grant a national licence/existing national licence/ assessment

done for national procedure one of the most extensive ones at the time.5 Satisfactory/good previous experience with this RMS.2 Marketing policy: Product can be marketed in RMS first/strong local demand.1 Willing to be RMS for a whole range of related vaccines.1 Geographical: The CMS included Northern countries.1 RMS acted as rapporteur in the MRL procedure.1 Home country of applicant.

3.3 National Assessment Procedure by RMSThe dates for the national assessment procedure by the RMS were provided by 14 of the 22 replies,and from these the timelines for the assessment by the RMS were summarised (Table 3).

The average assessment period (clock running) was 9 months, but this could take anything from 2 to36 months. A shorter assessment period was sometimes associated with a longer clock stop period,and vice versa. The average total length of the procedures was 21 months, with a range of 7 to 38months. Some lengthy clock stops were caused by delays in the publication of MRLs.

Only 43% of the assessments (n=14) were completed in less than 210 clock days (7 months).

Table 3: National Assessment Procedure by RMS

14 replies Months clock running Months clock stopped Total length of procedure(Months)

Average 9 12 21

Maximum 36 24 38

Minimum 2 1 7

3.4 The Performance of RMSs and Applicants

(a) RMS

Two thirds of the MRPs used just two RMSs - UK and France, which had 7 procedures each.Germany was used as RMS for 6 of the procedures, so that these three countries covered 91% of theprocedures in this survey. Denmark and Sweden were each used for one procedure. The RMSs aresummarised in table 4 and Figure 1.

The performance of all the RMSs was judged by the applicants to be fairly well balanced betweensatisfactory to excellent (see table 5). France was consistently seen as "excellent" for itsadministrative efficiency and communication, both areas in which Denmark received some criticism.Germany and Sweden were consistently seen as satisfactory in all areas for pharmaceutical products,


and Germany as excellent in all areas for immunologicals. The UK performance was generally wellappreciated and consistent.

As in the 1998 survey, applicants had a very positive view of the work of the RMSs overall.

Table 4: Number of procedures for each Reference Member State

RMS: DE DK FR IR IT SE UK

1995 - 1998 3 0 8 1 1 0 7

1999 6 1 7 0 0 1 7

Figure 1: The Reference Member States for 1999 survey

DE27%

DK5%

UK31.5%

SE5%

FR31.5%

(b) Applicants

Although one company received some criticism, overall the performance of applicants was judged bythe RMSs also to be satisfactory to excellent (table 5 and Figure 3). However the balance was tippedtowards ‘excellent’ in all areas except one – scientific quality. While the overall scientific quality ofthe applicants was never seen as ‘unsatisfactory’ by the RMSs, only a minority of applications wereregarded as ‘excellent’ in this area.

Table 5: Performances of RMSs and Applicants

Performance of RMSBy applicant

Performance of applicantBy RMS

AdminEffic.

Man.Flexi

Scient.Qualit.

Transparency

Communicat.

AdminEffic.

Man.Flexi

Scient.Qualit.

Transparency

Communicat.

Excellent 11 11 8 10 13 13 14 5 12 16

Satisfactory 10 10 14 12 8 7 6 17 8 5

Not satisf. 1 0 0 0 1 1 1 0 1 0

No answer 1 1 1 1 1


Table 6: Performance summary for each Reference Member State

RMS: DE DK FR IR IT SE UK

Performanceas RMS1998 survey

Satisfactory Satisfactoryto excellent

Satisfactory Excellent Satisfactoryto excellent

Performanceas RMS1999 survey

Satisfactoryto excellent(especiallybiologicals)

Satisfactoryto

unsatisfactory

Satisfactoryto excellent

Satisfactory Satisfactoryto excellent

Figure 2

Performance of RMS

0

2

4

6

8

10

12

14

16

18

20

Excellent Satisfactory Not satisfactory No answer

Sco

re

Admin Efficiency

Man. Flexibility

Scientific Quality

Transparency

Communication

Figure 3

Performance of applicant

0

2

4

6

8

10

12

14

16

18

20

Excellent Satisfactory Not satisfactory No answer

Sco

re

Admin EfficiencyMan. FlexibilityScientific QualityTransparencyCommunication


3.5 The Concerned Member StatesAll 15 member states of the European Union were involved as Concerned Member States (CMSs) inthe MRP during 1999. The numbers of CMSs involved in the 22 procedures covered by this surveyranged from 1 to 14, amounting to 170 CMS procedures. The average number of CMSs per procedurewas 8 (1998 survey = 8).

From this sample of MR procedures it appears that there has been no change in the trend for theNordic countries (Finland and Sweden) and Ireland, to be included the least as CMSs, probably formarketing reasons. Spain, Italy, Austria, Belgium and Netherlands were included the most as CMSs,probably either as a way of overcoming lengthy national procedures or because these countries wereleast likely to have the product already registered. The involvement of each member state in these 22MRPs is summarised in Figure 4.

Figure 4: Involvement of Member States as RMS or CMS in the MRP (1999)

0

2

4

6

8

10

12

14

16

18

AT BE DE DK EL ES FI FR IR IT LU NL PT SE UK

Member States

Nu

mb

er o

f p

roce

du

res

No. withdrawn

No. of successfulprocedures as CMS

No. of procedure asRMS

The main factors determining the choice of CMSs were marketing and the absence of existingregistrations in these countries (see tables 7 and 8).

Table 7: Reason for choice of Concerned Member States

Number Reason16 Commercial/Marketing reasons1 All European countries where other presentations of the product are registered1 European countries where the product was not registered yet1 Only Member State which had not granted a licence.1 Extensive list to avoid duplicate of MR; epidemiological reason1 CMS refused to grant MA under 1st round of mutual recognition


Table 8: Reason for exclusion of Concerned Member States

Number Reasons11 No marketing interest or existing national licence3 Existing registrations already granted by national procedures1 European countries where other presentations of the product were not registered yet1 The excluded Member States could not participate in this procedure because their

existing SmPCs were not in line with that of the RMS, either following the originalMulti-state procedure or following the first line extension

1 national vaccination policy4 n/a or no reply

3.6 Automatic validationNone of the applicants considered that the automatic validation had failed; 14 applicants indicated thatit had worked and 8 that it was not applicable.

However, for 6 of the 22 procedures CMSs raised objections at validation, and in one case this led tothe withdrawal from 2 countries. The objections related to missing data, old trials and reports, quality(degradation products), indications and the production site.

3.7 Number of questions from CMSsThe average number of questions from CMSs per procedure was 8 (c.f. 8.4 in 1998).

The CMSs asking on average more than 10 questions per procedure were France, Spain, U.K.,Belgium, Finland and Ireland (see figure 5).

By comparison, in 1998 the CMSs asking on average more than 10 questions per procedure wereGermany, France, Netherlands, Spain, Sweden and U.K. In 1999 Germany, Netherlands and Swedenall reduced their average number of questions to below 10 per procedure. Germany in particular hasshown a marked reduction.

Table 9: Average number of questions per procedure from CMSs (plus 1998 versus 1999)


1999 Average 2 17 1 11 0.5 16 14 40 18 1 0 7 5 6 29

1999 Minimum 0 6 0 0 0 0 7 23 7 0 0 2 0 0 20

1999 Maximum 6 27 14 19 3 43 24 73 48 5 0 12 24 10 48

Comparison with 1998 data

1998 Average 5 9 16 9 0 11 7 23 5 1 0 12 1 13 39

1998 Minimum 0 2 0 2 0 0 1 7 0 0 0 5 0 8 22

1998 Maximum 13 23 31 19 1 35 14 45 14 3 0 20 6 16 66

Change from 1998to 1999 -3 +8 -15 +2 0 +5 +7 +17 +13 0 0 -5 +4 -7 -10

% Change -61% +90% -91% +20% 0% +47%+100

% +75%+267

% +29% 0% -43%+431

% -55% -25%


Figure 5: Average number of questions per procedure from CMSs

-25

-15

-5

5

15

25

35

45


Concerned Member State

Num

ber

of q

uest

ions

1998 Average

1999 Average

Change from 1998 to

3.8 Number of questions to each part of the dossierThe number of questions received to each part of the dossier has been summarised in table 9. Onaverage during 1999 applicants received 75 questions from CMSs per dossier, roughly dividedbetween major (35%), minor (26%) and SmPC (39%) questions (see figure 6). Depending upon thenumber of CMSs involved, an applicant received between 10 and 254 questions per dossier.

The majority of the questions related to two parts of the dossier: Part II quality (56%) and Part IVefficacy (26%) (see figures 7 to 9). The remaining questions related to Part I Administrative data(6%), Part IIIA Safety (11%) and Part IIIB Residues (1%).

Table 10: Summary table for number of questions by classification

For 19procedures Total Average Minimum Maximum

Total part I 50 3 0 19Total part II 492 26 0 89Total part IIIA 92 5 0 41Total part IIIB 7 0 0 3Total part IV 227 12 0 26

Major 505 27 0 108Minor 365 19 0 51SPC 556 29 0 97 P e r d o s s i e r

All questions 1426 75 10 254

N.B. The datain this tablerelates to 20proceduresbecause oneprocedure wasa repeatprocedurewith noquestions andthis data ismissing for asecondprocedure.


Figure 6: Number of questions by category Figure 7: Number of questions by dossier part

Major 35%

Minor26%

SPC39%

Total part I

6%Total part

IV26%

Total part II

56%

Total part IIIB1%

Total part IIIA

11%

Figure 8: Major questions to each dossier part Figure 9: Minor questions to each dossier part

Part I1%

Part II48%

Part IIIB1%

Part IV35%

Part IIIA15%

Part I12%

Part II68%

Part IV14%Part IIIB

1%

Part IIIA5%


3.9 VMRFG Discussion

3.9.1 Need for a VMRFG DiscussionA VMRFG discussion was necessary for 18 of the 22 procedures (82%) where major objections andconcerns were raised (figure 10). In one procedure, although no VMRFG discussion was held, it wasnoted that numerous discussions took place between the RMS and one CMS and between the applicantand the CMS, and a separate post-marketing commitment had to be agreed to for this CMS.

Figure 10

Need for a VMRFG Discussion

Yes82%

No18%

3.9.2 Performance of VMRFG DiscussionAlthough the majority of CMSs that raised major objections did attend the VMRFG break-outdiscussion sessions, in the majority of these break-out sessions one or more CMSs raising majorobjections did not attend. All applicants were immediately informed of the out come of thediscussion, and in 72% of cases the problems were resolved. In three out of the four cases where theproblems were not solved the applicants withdrew their application from some CMS to avoidarbitration.

In 50% of 18 cases where a VMRFG discussion was necessary the applicants withdrew from certainCMSs to avoid arbitration.

Table 11

Question YES NODid all the CMS who raised major objections attend? 8 10Were you immediately informed of the outcome by the RMS? 18 0Were all the problems solved? 13 5Did you withdraw your application from some CMS to avoid arbitration? 9 9


3.10 Time taken to send translations and to obtain national licencesThe average number of days taken by applicants to send final translations to each CMS was 15 days(n=116). In two cases applicants took 3 months to supply translations. However, 68% of translationswere sent with 14 days of day 90, and 91% with 28 days.

A large proportion of the dates that licences were issued were not supplied with the completedquestionnaires, and many of these dates were also not available from Eudratrack. Of 157 possiblelicences, the dates the licences were issued were supplied for 78. For several countries the data inTable 12 is based on just one or two licence dates, and therefore it may not reflect the actual averagetime taken by these countries to issue a licence.

In the 1999 survey the average time to obtain a national licence after final translations have beenreceived was 55 days. Therefore the total average time from day 90 to obtaining licence was 70 days(55 + 15). In the 1998 survey an average of 58 days was reported for this time period. Of the 78 datesreported, 26 (37%) were within the 30 days recommended in the best practice guide. Only fourcountries were able to issue licences consistently within 30 days (see figure 11).

Table 12: Days taken to obtain national licences after receipt of final translations (1999)


Average 45 85 42 19 69 24 48 20 49 180 54 58 109 49 28Max. 83 109 70 27 113 29 67 27 70 180 132 98 140 49 28

Min. 25 73 18 7 24 9 37 12 28 180 16 46 60 49 28

n 10 7 8 6 2 7 6 2 2 1 8 6 9 2 2n = number of procedures for which dates where provided and used to determine the average

3.11 WithdrawalsThe overall level of withdrawals has fallen from 11% (1998 survey) to 7% (1999 survey) (figure 12).

The recurring problems continue to affect procedures with Denmark and Sweden, with withdrawalsoccurring in more than one third of their procedures. Finland continues to have at least 10%withdrawals, and Belgium, Spain and Netherlands are also affected to a lesser extent (circa 7%).

In contrast to the 1998 survey, in 1999 UK, Greece, and Germany were not affected, while Italy wassubject to 13% withdrawals. Austria, France, Ireland, Luxembourg, and Portugal were not subject towithdrawals in either of the surveys.

Table 14: Summary table of withdrawals

CMS: AT BE DE DK EL ES FI FR IR IT LU NL PT SE UK TOTALS

Withdrawn 0 1 0 5 0 1 1 0 0 2 0 1 0 2 0 13

Totalprocedures

15 14 10 12 9 16 8 5 7 15 12 15 13 5 3 180

% Withdrawn1999 Survey

0% 7% 0% 42% 0% 6% 13% 0% 0% 13% 0% 7% 0% 40% 0% 7%

% Withdrawn1998 Survey

0% 8% 13% 50% 14% 8% 10% 0% 0% 0% 0% 18% 0% 17% 20% 11%


Figure 11: Average days to issue licence by Member States after receipt of final translations

0

20

40

60

80

100

120

140

160

180

200


Member states

Num

ber o

f day

sAverage

Max.

Min.

Figure 12: Number of withdrawals

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%


Member States

% W

ith

dra

wn

% Withdrawn 1998 Survey

% Withdrawn 1999 Survey


3.12 Reason for withdrawal from a CMSThe 1999 Joint VMRFG/FEDESA Survey of the MRP did not record the reasons for withdrawals perse, as this is the subject of a separate study by the VMRFG (insert VMRFG summary table??).However the reasons will be reflected in the problems reported in the following sections.

3.13 Major problems reported by applicants and RMSs

3.13.1 Major problems encountered with the ApplicantsThere were no major problems reported by the RMS with the applicants. One comment was receivedas follows:

• Due to the dates clashing with CVMP there was some confusion over whether the applicationwould be mutually recognised at the end of the procedure.

3.13.2 Major problems encountered with the Reference Member StateThere were a very few major problems reported by the applicant with the RMS. The comments thatwere received are as follows:

• None, excellent collaboration• Communication difficulties between RMS and one CMS leading to misunderstandings.• A few communication errors occurred (dates), but were adequately corrected.• We were not told that we had to have exactly the same indications in CMS as in RMS – the

original product had wider indications in RMS compared to CMS• The questions asked led to conditions for licensing that clearly went beyond the time path set

in legislation

3.13.3 Problems encountered with the Concerned Member StatesThe responses to question 4d: Major problems encountered with the individual Concerned MemberStates, are summarised below.

No problems were reported by the RMS in 11 (50%) of the procedures.

No problems were reported by the applicants in 8 (36%)of the procedures.

1. Problems reported by the RMS:

• 2: CMS raising objections did not attend the VMRFG meeting• 1: Different opinions about how to formulate the indication.• 1: Impossibility to discuss result of CMS assessment of applicant’s answer.• 1: Different opinions about the prudent use of a product.• 1: Problems about combination of products, application was withdrawn.• 1: Lack of full clinical documentation.• 1: Acceptance of the potency test and stability data for product.• 1: Concern about safety and efficacy.• 1: Problems about additional studies.• 1: Misunderstood question• 1: Impossibility to authorize product, because of national law.• 1: Existing vaccine, with updated dossier, treated as a completely new product


2. Problems reported by the Applicant:

• 4: Technical arguments were rejected.• 4: Problems concerning the indications and SPC (insistence that a dosage chart and a

particular statement should be included in the SPC).• 3: Problems concerning questions.• 3: Communication problems.• 2: problems with live modified vaccines or the live primer vaccine• 2: problems because of a national principle or law, CMS has not implemented all the EU law.• 1: Excessive interpretation of political texts on antibiotic resistance.• 1: Problems about interpreting guidelines on justification of a combination product.• 1: Difficulties due to CMS raising major objections but not attending VMRFG.• 1: Lack of transparency.

3.14 Evaluation of work of RMS and Applicant (Question 4e and 4f)The responses to question 4e and 4f: Evaluation of work of RMS and Applicant are summarisedbelow.

1. Comments on Applicants by RMS:

• Positive Commentso 3: Pre-submission discussions allowed procedure to function well.o 2: No problemso 1: Applicant was very responsive; made a valid effort to answer CMS’s questions

conscientiously.o 1: CMS had expressed concerns during the first procedure but these had been resolved

before the Repeat Use’ was started.o 11: Communications with Applicant was very good.o 1: Dossier well presentedo 1: The people involved performed very well.

• The Dossiero 3: Not all parts of the dossier were of uniform excellent quality.o 1: Validation postponed due to problem with DMF.o 1: Could have improved compliance to data requirements as set down in guidance.o 1: Dossier was not updated.

• Responses to list of questionso 1: Response to questions was poor and not argued sufficiently strongly.o 1: The technical quality of the Quality and Ecotox responses could be improved.

• Other commentso 2: Applicant did not totally understand the spirit of MRPo 1: Insufficient applicant’s involvement during MRP.o 1: Technical discussion reduced to the minimum.

2. Comments on RMSs by Applicants:

• Positive Commentso 4: Very satisfied with RMS.o 1: Overall approach very co-operative.o 1: Effective analyses on strengths-weaknesses.o 1: Assessment report was made available to the Applicant for comments.o 1: Successful formal meetings by telephone conference.o 1: Very close monitoring of Applicant’s responses by RMSo 1: Excellent collaborations with administrative and scientific staff.o 1: We were impressed how intensive the RMS defended the dossier.


• Other commentso 4: The preparation of a consolidated list of questions by the RMS would have been

helpful.o 1: Further involvement of RMS in discussion with CMS would have welcomed.o 1: Communication problems, which were later adequately corrected.o 1: Did not agree with revising the dossier.o 1: Late receipt of questions.o 1: the CMS was unable to grant the MA on the “first round “ of mutual recognition

3.15 How did the Procedure function?The responses to question 6a: How did the Procedure function, are summarised in Figure 13.In general the outcome of the procedures were satisfactory, and several procedures were reported asvery efficient. However difficulties were reported with a significant number of procedures, includingtwo procedures that were regarded to have functioned very badly.

Figure 13: How did the procedure function overall?

0

2

4

6

8

10

12

14

Very well Well Somedifficulties

Poorly

Res

po

nse

s

Response from RMS

Response from applicant


3.16 Proposals for improvement.The responses to question 6b: Proposals for improvement, are summarised below:

1. Comments from RMSs:

• 3: Restrict objections to real safety concerns.• 2: Access to CMS: Attending of all CMS at the VMRFG• 2: Technical involvement of the applicant and fully updated dossiers to be improved.• 1: Use version numbers for SPC to avoid confusion.• 1: Constant approach in 3 way discussions.• 1: More time for the applicants to respond to questions.

2. Comments from Applicants:

• 5: Improve communication: Applicant should be advised for revisions to the updated dossier;Greater involvement of RMS in discussions with CMS; Send out copy of assessment report toapplicant, so that we know what CMSs refer to.

• 3: Possibility to accept MAs from those CMSs giving positive opinions.• 1: Attending of all CMS at the VMRFG• 1: All Member States should trust each other’s assessment reports.• 1: Better adherence to European guidelines and less to national principles.• 1: The delay to answer the questions after day 60 should be extended to 15 days.


4. DISCUSSIONThe 1999 joint VMRFG/FEDESA survey of the functioning of the mutual recognition procedure hasbeen a success. Despite the fact that it was a retrospective survey for most respondents (the 1999questionnaire was not sent out until September 1999), replies were received for 22 out of a possible 27procedures. Furthermore the questionnaire was filled in reasonably well by the respondents, perhapswith the exception of section 4a concerning the dates of the mutual recognition phase. Consequentlydata from this section has not been analysed or included in this report.

The product mixAlthough the even split between pet and livestock products (50/50), and the split betweenpharmaceuticals and vaccines have remained constant (78/22), there is one striking differencesbetween the product mix for the 1995-1998 sample and the 1999 sample. In the 1998 survey 75% ofthe procedures concerned new active ingredients. In 1999 this has fallen to less than half of theprocedures. The underlying causes of this should be investigated, because if it is linked to a real dropin the number of new chemical entities reaching the marketplace, then this could have significantimplications for the future of the animal health industry.

However for vaccines practically all the procedures in 1998 and 1999 concerned new activeingredients. Presumably this is directly related to the fact that the national review of existing licencesfor vaccines is still on-going, thus making their inclusion in mutual recognition procedures difficult.

The RMS-Applicant relationshipThe overall pattern in the use of RMSs has remained unchanged, whereby UK and France continue toaccommodate 75% of all procedures, and together with Germany, just 3 countries accounting foraround 90% of procedures. Only four other MSs have acted as RMS: Ireland, Sweden, Italy andDenmark.

The 1999 survey collected data on the time taken for the national assessment procedure by the RMS.Although not all the data were supplied, it is interesting to learn that the average clock running timewas 9 months (with a range of 2 to 36 months), and only 43% of these procedures were completedwithin the required 7 month assessment period (210 clock days). With an average clock stop time of12 months (range 1 to 24 months), this means the average time taken to obtain the licence in the RMSwas 21 months (range 7 months to 38 months).

However this does not appear to have damaged the excellent relationship that usually exists betweenthe applicant and the RMS. As in 1998, this survey has recorded a high level of satisfaction with theRMS, and is now able to confirm that the RMS is equally satisfied with the work of the applicant.This was evident not only in the ‘performance scores’ (see 3.4), but also in the absence of any ‘majorproblems’ on both sides (with the exception of the occasional ‘communication’ problem – see 3.12),and the many positive ‘comments’ that were received. In particular applicants appreciated theproactive approach of the RMS to stand-by their assessment and defend their opinion during themutual recognition phase. This direct contact between the RMS and the applicant is a key strength ofthe MRP and should be jealously guarded during any future procedural reviews.

Where there was criticism of the applicant by the RMS it usually concerned either the quality of thedossier, or the quality of the response to the list of questions. Conversely, the most commoncomplaint from applicants was that the RMS did not consolidate the list of questions, or provided thelist of questions late.

The number of questions from CMSsThe 1998 survey identified the number of questions received by the applicant from CMSs at day 60 asa major concern for industry


Although the average number of questions being asked by CMSs has not decreased from 1998 to1999, it is apparent that a more positive approach that has been adopted by several MSs. Furthermore,the VMRFG has taken the initiative to:a) classify the questions into major, minor or SPC points (this has enabled the 1999 survey to

provides more detailed information on the type of questions being asked and to which sections ofthe dossier).

b) make a commitment in the December 1999 Table of Decisions to reduce the number of questions.

The VMRFG Discussion and withdrawalsIt is apparent that mutual recognition would not work without the VMRFG. For example 82% (n=18)of procedures a VMRFG discussion was required to resolve issues, and in the majority of cases(13/18=72%) the RMS has successfully defended the product. Never the less, applicants stillwithdrew from certain CMSs in 50% of these 18 procedures to avoid arbitration.

However the overall level of withdrawals has decreased from 11% in 1998 to 8% in 1999. The greaterpart of this success must be attributed to VMRFG, as there is no evidence that applicants are beingmore selective in their choice of CMSs (the average number of CMSs per procedure has remainedconstant).

Obtaining a national licenceThe time taken by national agencies to issue the licence after day 90 was a second key area of concernidentified in the 1998 survey, as the average time was found to be 58 days. However it was not knownwhat fraction of this delay was caused by companies supplying final translations of product literatureto the agencies. Therefore the 1999 questionnaire was modified to capture this information.

The average time from day 90 to obtaining the licence has increased from 58 days in 1998 to 71 daysin 1999, with an average of 15 days being taken by companies to supply the final translations, and anaverage of 56 days being taken by agencies to issue the licence. However this may not reflect the trueaverage as a significant number of dates were not available at the time of the survey.

Proposals for improvement of the Mutual Recognition ProcedureThe most common proposal for improving the MRP from RMSs was to restrict the questions asked byCMSs to just those of real safety concerns and for CMSs to show greater commitment to resolving theissues by attending the VMRFG meetings. RMSs also suggested that greater technical involvement ofthe applicant would be useful, and for applicants to make sure that the dossiers were always fullyupdated before entering into mutual recognition.

For applicants the greatest improvement would be improved communications, which covered allaspects of the mutual recognition phase. For example, not all RMSs got fully involved in thediscussions with CMSs, and some applicants did not receive a copy of the assessment report, whichwould have helped them understand the CMSs concerns. Applicants also suggested that thearbitration procedure should allow marketing to proceed in those CMSs with no objections; that CMSsshould adhere to European guidelines and not national principles.

Both one RMS and one applicant also suggested that more time was needed for the applicant torespond to the list of questions. However, many applicants appreciate the tight time lines of theprocedure, and argue that it is not the time available that is the problem, but the sheer volume ofquestions that is received.

Overall overviewWhen asked for their overall view of how well the MRP functioned, both RMSs and applicants gave avery similar view. The majority (67%) of procedures were considered to have functioned well or verywell. A few had “some difficulties”, and 10% had serious problems.


Proposals for improvement to the questionnaire

Section 1a: the RMS should fill in the Procedure number and the product name.

Section 4a: This section should be reviewed, and should include the following change:“Receipt of last dossier and fees to CMSs” should read “Receipt of last dossier and fees by CMSs”.

Section 4d: The upper section first column heading (“With Reference member State”) should bedeleted. The second column heading should read – “Major problems experienced between the RMSwith the applicant”

5. CONCLUSIONThe mutual recognition procedure is working reasonably well. The relationship between the RMS andthe applicant remains strong, with a good working relationship and a high level of agreement on boththe strengths and weaknesses of the MRP.

The VMRFG has continued to make considerable effort to improve the functioning of the MRP, andhas taken several important initiatives during 1999. This has resulted in a reduction in the number ofwithdrawals. Never the less there is still room for improvement in the key problem areas identified in1998. These include the number of questions from CMSs and the time taken by some agencies toissue the product licence.

It is anticipated that with the on-going initiatives of the VMRF group, and the on-going dialogue withindustry, the year 2000 survey will reveal a further consolidation of this progress.

VMRFG and FEDESA would like to thank all those who took part in this survey.

VMRFG/FEDESA SURVEY MUTUAL RECOGNITION PROCEDURE ...

Documents