Have we learned from our mistakes in transplantation?

Chris Watson

Addenbrooke’s NHS UNIVERSITY OF Cambridge University Hospitals Foundation Trust CAMBRIDGE Learning from mistakes

The Pancreas Forum Annual Report on Pancreas transplantation ¡A forum for learning from the mistakes of others ¡Benchmarking centres ¡ And their good practice against their peers ¡Annual meeting, now in ¡Providing data for

th National Pancreas Transplant Forum 2018

Venue: Radisson Blu Hotel, Cardiff CF10 2FL its 14 year Date: 2nd February 2018 patients

PROGRAMME

9:00 Welcome

9:15 NHSBT Annual Pancreas and Islet Transplant Report 2016/17 Claire Counter

9:45 National Data on the impact of pancreas quality and retrieval logistics on the transplant outcome Claire Counter

10:00 Pancreas Discard Rate in UK – Are we all saying the same thing? Sorina Cornateanu (Edinburgh)

10:30 Discussion

10:45- 11:00 Coffee/Tea Break

11:00 Trends in pancreas transplantation – USA Raja Kandaswamy (USA)

11:40 Trends in pancreas transplantation – UK Argiris Asderakis (Cardiff)

12:00 Discussion

12:30 Lunch

13:30 Pancreas Transplantation in Type 2 DM Sanjay Sinha (Oxford)

Raja Kandaswamy (USA)

14:15 Update on Immunotherapy Trials on Type 1 DM Colin Dayan (Cardiff)

14:45 Discussion

15:00-15:15 Coffee/Tea Break

15:15-16:15 Case presentations

Cambridge, Edinburgh, Guys, Hammersmith, Manchester, Newcastle, Oxford, Cardiff

Acknowledgements

¡Claire Counter Mistakes? ¡Questions ¡ What did we get right (if anything)? ¡ What are our mistakes? ¡ Did we learn from them? ¡Considerations ¡ Programme organisation ¡ Indications ¡ Patient Assessment ¡ Implantation techniques ¡ Development of pancreas transplantation in the UK

¡National funding 250 Sol pancreas 57 Kid + Panc ¡ Scotland 2002 200 26 36 37 29 ¡ England 2004 26 53 32 39 18 17 ¡ Wales 2006 150 20 ¡ Islets 2008 100 188 188 173 173 167 6 166 162 159 159 155 151 ¡Note: 50 102 ¡ activity was 75 157 transplants in 2015 0 Pancreas transplantation in UK vs elsewhere

4 3.7 3.7 3.5 3.1 3.1

3 2.8

2.5 2.1 2.0 1.9 1.9 1.9 1.8

2 1.7 1.5 1.5 1.5 1.4 1.5 1.4 1.1 1 0.7 0.5

0.5 0.4

0 UK USA Italy Spain Czech Czech Poland France Austria Ireland Croatia Finland Canada Norway Sweden Belgium Portugal Australia Germany Argentina Switzerland Netherlands

Global Observatory on Donation and Transplantation 2010 data https://reports.ont.es/caapmp.aspx Centre volume affects SPK outcomes

¡Pancreas transplants in EuroTransplant 2008-2013 Patient Survival Survival ¡Centre volume Big centre ¡ Big: ≥13 per year ¡ Medium: 5-12 per year Big centre Small centre ¡ Little: <5 per year Small centre

Kopp et al. Transplantation 2017;101:1247 Organisation of pancreas transplantation in the UK

¡8 centres ¡Centrally funded ¡NHSBT oversight x Edinburgh

x Newcastle

Manchesterx

Cambridge x Oxford x Cardiff x WLRTC x x Guys Centre activity in UK

70 ¡Range 5 to 62 Solitary pancreas 60 Pancreas + kidney ¡5/8 centres in “Big” category 50

40

30

20

10

0

NHSBT Centre specific data, 2016/17 What did we learn: Solitary pancreas transplantation

Pancreas graft survival Solitary pancreas transplant activity 100 2005-2010 70 90 60 80 50 70 60 40 50 30 57 40 53

% graft survival 30 20 39 36 37 N % Survival (95% CI) 32 29 20 1 Year 3 Year 5 Year 26 26 2005-2006 33 75 (56-87) 53 (35-69) 50 (32-66) 10 20 18 17 10 2007-2008 52 69 (54-80) 55 (40-67) 6 0 2009-2010 44 80 (64-90) 0 0 1 2 3 4 5 years post-transplant

Data from Sue Madden & Lisa Mumford Indications for transplant

¡Type 1 ¡Insulin dependent type 2 ¡ BMI <30 ¡SPK: Renal failure ¡PTA: Hypoglycaemic unawareness

https://nhsbtdbe.blob.core.windows.net/umbraco-assets- corp/6525/pol185-pancreas-transplantation-patient-selection.pdf Indications for transplant: No significant difference between type 1 and type 2

Pancreas graft survival Percentage of SPK transplants by 88% type of 84%

Type 1 Type 2 Other p=0.4 Total Txs 177 141 140 145 160 157 176 151 148 150 100 2.8 2.8 3.4 3.1 4.0 4.0 4.3 6.4 4.0 8.1 95

90 96.0 97.2 97.2 96.3 96.0 96.0

Percentage 95.0 93.6 94.9 91.9 85

80

Type 2: Insulin dependent & BMI<30kg/m2 Survival by age One year graft survival following SPK Percentage of SPK transplants by age of patient transplant, 1 April 2007 to 31 March 2017

Total 0-35 36-40 41-50 51+ 89% no. txs 188 151 159 155 173 166 188 173 167 162 88% 100 85% 90 p=0.5 80 70 60 50

Percentage 40 30 20 Age group 10 0 Patient assessment

Pancreas & Kidney Isolated pancreas ¡Diabetic comorbidity 18 ¡ Cardiovascular disease 16 ¡ Previous MI 14 ¡ Peripheral vascular disease 12 ¡ Previous amputations 10 ¡ Cerebrovascular disease 8 ¡ Previous TIAs/CVAs 6 4 Percentage of patients of Percentage ¡Renal failure 2

¡ Dialysis burden 0 MI CVA Registered Major limb Digital blind amputation amputation Pre-transplant MI is associated with poorer

graft survival One year graft survival following SPK transplant , 1 April 2007 to 31 March 2017, Percentage of SPK transplants where patient had an MI pre-transplant

Total 88% no. txs 187 151 154 151 170 165 185 172 166 155 6 75% 5.2 p=0.007 5 4.5 4.6 4.5 4.2 4.0 4

2.7 3 2.4 2.1 1.8 2 % of transplants

1 Previous MI

0

Still significant factor for one year graft survival after adjusting for donor age, BMI, donor type and waiting time: p=0.012 Pre-transplant CVA is not associated with poorer graft survival Percentage of SPK transplants where One year graft survival following SPK transplant, 1 April 2007 to 31 March 2017 patient had a CVA pre-transplant 92% Total no. txs 187 151 152 151 171 165 181 172 167 154 87% 7 6.0 6 p=0.24 5.3 5.2 5 4.6 4.4 3.9 4 3.2 3.0 3 2.3 2.4

% of transplants 2

1 Previous CVA

0

Year Previous amputation does not affect graft survival

Percentage of SPK transplants where patient had an amputation pre-transplant 88%

Total Major Limb Digital no. txs 186 151 154 149 168 163 184 173 166 151 10 p=0.95 9 8 7 6 5 4

Percentage 3 2 1 0 Amputation status

Financial Year Patient survival

¡One year patient survival is not affected by ¡ Previous MI (p=0.29) ¡ Previous CVA (p=0.16) ¡ Smoker (p=0.27) ¡ Limb amputation (p=0.25) Assessment summary

¡Solitary pancreas recipients have a 50% 5 year graft survival ¡The outcomes are the same for patients with type 1 and insulin-dependent type 2 diabetes ¡There is no clear difference in graft survival by recipient age ¡A history of MI is associated with poorer graft survival ¡Previous CVA or limb / digital amputation do not influence graft survival ¡ But there may be selection bias involved here Waiting for a transplant

¡Have we got this right 100

compared to USA? 90 Waiting list 80 survival 70 % survival% 60

Category n 1 year 4 years 50 Wait List 12478 93.4% 58.7% SPK Tx 6995 95.0% 90.3% Months 40 0 12 24 36 48

Am J Trans 2004 Outcome following listing for a simultaneous pancreas & kidney in 2013 pancreas in the UK UK USA 100% 2 7 100% 3 7 80% 80%

60% Died 60% Removed 40% 76 40% Still waiting 64 20% 40 Transplanted 20% 37

0% 0% 1 year 3 years 1 year 3 years

7% die waiting UK figures are 4/13 to 3/14 DONOR TYPES The donors: DCD are as good as DBD pancreases

SPK transplants in the UK by donor type 5 year graft survival by donor type Total no. txs 188 151 159 155 173 166 188 173 167 162 100% 15 20 22 90% 25 35 33 35 81% 44 43 78% 55 80% p=0.4 70%

60% DCD 50% 173 131 137 DBD 40% 130 138 133 153 123 119 118 30%

Percentage of transplants 20% Donor type 10% 0%

Year

SPK transplants, 1st April 2007-31st March 2017 DCD pancreas use varies from 0 to 39% in spite of there being no difference in outcomes Proportion of DBD and DCD pancreases Median waiting time to transplant used by centres,1/4/14 to 31/3/17 2011/14 100% 600

80% DBD 500

DCD 400 60% 300 40% 200

20% 100

0% 0

Data from NHSBT annual reports Cold ischaemia affects graft survival: DBD donors One year graft survival following DBD Median cold ischaemia time for SPK transplant 1 April 2007 to 31 March 2017 SPK transplants from DBD donors

91% 2160 Median ~12 hrs 84% 192032 p=0.02 168028

1440(hours) 24 Median 1200time 20 ~10.5 hrs 96016 Cold ischaemic time 72012

— ≤ 9.5h 480ischaemic 8 — 9.5 - 11.5h 4 — 11.5 - 13.5 240 Cold — >13.5h 00 The plumbing Portal vein extension graft on the donor pancreas

Percentage of SPK transplants with a Venous Extension Graft Total no. txs 170 142 143 131 141 143 170 170 152 150 25.0 23.0

20.0 17.3

15.0 11.2 10.0 8.4 7.0 5.9 5.6 6.4 % of transplants 5.0 2.3 2.9

0.0 Venous extensions are associated with poorer graft outcomes

3 month graft survival 1 year graft survival 91% 88% 83% 82% p=0.002 p=0.03

Venous extension graft Venous extension graft — No 1354 — No 1354 — Yes 135 — Yes 135

Still significant factor for one year graft survival after adjusting for donor age, BMI, donor type and waiting time: p=0.03 Is there a penalty for using a pancreas with a short portal vein?

Percentage of SPK transplants where portal vein was cut short (<1cm) 88% Total no. txs 127 108 142 132 132 133 142 149 138 139 86% 40 36.4 p=0.32 35 30.3 30 25.2 25.0 26.5 25.6 25

20 16.9 16.8 17.4 16.5 15

% of transplants 10 Donor portal vein 5 — Cut short (<1cm) n= 307 — Not short (≥1cm) n=1013 0 Duct management

¡Enteric drainage ¡ Duodeno-enterostomy ¡ Side-to-side anastomosis ¡ Roux-en-Y ¡ Duodeno-duodenostomy ¡Bladder drainage Duct management technique 1 year graft survival Total no. txs 184 141 148 144 159 157 180 171 164 158 91% 100 87% 90 80 72% 70 60 p=0.003 50 40 Percentage 30 Technique 20 — Enteric: side-to-side 1306 — Enteric: Roux-en-Y 244 10 — Bladder drainage 32 0

Enteric side to side Enteric Roux Bladder Drainage Still significant factor for one year graft survival after adjusting for donor age, BMI, donor type and waiting time: p=0.01 Roux or side-to-side enteric anastomosis? 3 month graft survival 1 year graft survival

93% 90% 91% 87% p=0.08 p=0.2

Technique Technique Immunosuppression

Note: The next slides include off-label use of immunosuppression CD25 monoclonals do not affect graft survival One year graft survival

87% Percentage of SPK transplants where intended induction agent was anti - CD25 Total no. txs 184 151 157 146 157 157 180 165 166 153 p=0.8 40.0

31.7 30.6 30.0 27.2 26.1 26.1

21.7 20.0 16.4

12.1 % of transplants 11.4 9.8 10.0

0.0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Year

SPK transplants in the UK between 1 April 2007 and 31 March 2017 Campath 1H (alemtuzumab) does improve 1year graft survival One year graft survival

Percentage of SPK transplants where intended induction 89% agent was campath 1H Total 85% no. txs 184 151 157 146 157 157 180 165 166 153 80.0 70.2 67.7 67.7 70.0 65.4 66.7 66.7 p=0.02 60.3 58.5 60.0

50.0 45.0 39.9 40.0

30.0 % of transplants 20.0

10.0

0.0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Year

Still significant factor for one year graft survival after adjusting for donor age, BMI, SPK transplants in the UK between donor type and waiting time – p=0.004. Increased significance after adjusting for these 1 April 2007 and 31 March 2017 factors appears to be due to interaction between donor factors and Campath 1H. Mycophenolate One year graft survival

Percentage of SPK transplants where intended 88% immunosuppression was mycophenolate Total no. txs 188 151 157 153 172 164 187 173 167 159 99.4 100.0 79% 96.7 98.4 96.5 98.1 100.0 95.7 93.6 95.9 p=0.04 90.1 90.0 80.0 70.0 60.0 50.0 40.0

% of transplants 30.0 20.0 10.0 0.0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Year

Not significant factor for one year graft survival after adjusting SPK transplants in the UK between 1 April 2007 and 31 March 2017 for donor age, BMI, donor type and waiting time – p=0.06 Prednisolone

Percentage of SPK transplants where intended One year graft survival immunosuppression was prednisolone 89% Total no. txs 185 151 158 151 158 160 184 172 165 155 60.0 85% 51.0 48.8 p=0.01 50.0 48.1 41.1 40.0 39.4 40.0 37.3 37.0 36.0

28.5 30.0

% of transplants 20.0

10.0

0.0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Year

Still significant factor for one year graft survival after adjusting for donor age, BMI, SPK transplants in the UK donor type and waiting time – p=0.003 Increased significance after adjusting for these between 1 April 2007 and 31 March 2017 factors appears to be due to interaction between donor factors and prednisolone. One year graft survival

Percentage of SPK transplants where intended 88% immunosuppression was ciclosporin Total no. txs 185 151 159 152 159 161 186 170 166 155 20.0 85% 18.0 18.0 17.0 p=0.13 15.6 16.0 15.3 13.2 13.3 14.0 12.6 11.9 11.8 11.6 12.0 10.0 8.0

% of transplants 6.0 4.0 2.0 0.0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Year

SPK transplants in the UK between 1 April 2007 and 31 March 2017 One year graft survival

Percentage of SPK transplants where intended 88% immunosuppression was tacrolimus Total 84% no. txs 187 151 159 154 172 165 186 173 166 161 100.0 88.8 p=0.08 87.7 87.4 86.8 87.0 85.5 87.3 90.0 84.3 82.4 84.4 80.0 70.0 60.0 50.0 40.0

% of transplants 30.0 20.0 10.0 0.0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Year

SPK transplants in the UK between 1 April 2007 and 31 March 2017 Cambridge practice – learning from mistakes or anecdote based medicine?

¡Consent ¡ Reoperation is common: 30% ¡ Nerve damage ¡ Spinal ischaemia 1% ¡ Its miserable ¡ But its great 3 months later Cambridge practice – learning from mistakes or anecdote based medicine?

¡Gastroparesis ¡ placed at surgery ¡ Avoids PN Cambridge practice – learning from mistakes or anecdote based medicine?

¡Avoid betadine flush of donor duodenum ¡ UW preserves epithelium better Cambridge practice – learning from mistakes or anecdote based medicine?

¡Reperfuse the vein first ¡ Its easier to control the bleeding Cambridge practice – learning from mistakes or anecdote based medicine?

¡Appendicectomy ¡ It sits on top of the pancreas ¡ ¡ if stones: ¡ 2 cases of early acute cholecystitis Cambridge practice – learning from mistakes or anecdote based medicine?

¡Air mattress ¡ Heel ulcers common Summary

¡ The following do not ¡The following cause affect graft survival significantly worse graft survival ¡ Type 1 vs type 2 DM ¡ Prior MI ¡ Recipient age ¡ Increased cold ischaemia ¡ Prior CVA ¡ Portal vein extension grafts ¡ Prior amputation ¡ Bladder drainage ¡ CNI choice ¡ immunosuppression ¡ Non-use of Campath and MMF