. Type 2 diabetes and the DiRECT Trial Mike Lean Glasgow Royal Infrmary

1. Funded by Diabetes UK to find a practical management solution for T2D, in primary care . Type 2 diabetes and the DiRECTTrial Mike Lean Glasgow Royal Infrmary MCN meeting Feb 5th 2019

2. Disclosures: Departmental research funds, support for conference attendance and fees for Advisory Boards and lecturing from Novo Nordisk, Lilly, and Cambridge Weight Plan. Medical consultancy fees from Counterweight Ltd. Shares costing £10 in Eat Balanced. Photograph if you must! All slides are posted on: www.directclinicaltrial.org.uk

3. Life-expectancy is still reduced by T2DM despite guidelines &drugs to lower glucose/HbA1c, LDL & BP Years of life lost 55 55 European Risk Factor Consortium, NEJM 2011

4. 488 drugs (70 generic compounds) are licenced for T2DMexcluding insulins, plus >25 seeking licences, and more in development (McCombie et al BMJ 2017: data from Medtrak, April 2017) 55 55 European Risk Factor Consortium, NEJM 2011

5. Weight gain/ obesity is the main driver of T2DMColditzGA et al. Ann Int Med, 1995 Adjusted RR (BMI <22 = referent) 100 90 80 70 60 50 40 30 20 10 0 <22 22- 23- 24- 25- 27- 29- 31- 33- >35 22.9 23.9 24.9 26.9 28.9 30.9 32.9 34.9 BMI (kg/m2)

6. Guidelines recommend Bariatric surgeryBut 15% (=15 kg) weight loss achieves most T2DM remissions Gastric band vs Usual diet 13% -15kg 83% remissions with >15kg loss 73% 2 year RCT Dixon et al (2008) JAMA

7. 15kg weight loss on 450kcal/d dietNormalised beta-cell function and pancreas fat Lim et al, Diabetologia 2011

8. 15kg intentional loss might normalise life expectancy with T2DM Life expectancy (mean age 64at diagnosis) 18 95% CI 16 Normal life expectancy mean 14 ↑ 95% CI 12 10 T2DM BMI>25 8 15 kg loss 14 0 2 4 6 8 10 12 Weight loss (kg) in first 12 months Lean et al Diabetic Medicine 1990 Lean et al. Diabetic Medicine, 1990

9. BEST LONG TERM RESULTS ARE WITH MOST RAPID WEIGHT LOSSVLCD reliably achieves 15kg weight loss Subjects completing 1-year in 80 studies: n = 26,455, completers = 18,199 (69%) (Franz et al JADA 2007) 15 kg

10. To maintain weight loss, behaviours must counteract both environment and physiology Obesogenic Environment Physical environment Food environment Educational environment Cultural environment Social environment Social Marketing (normalised behaviours) Obesogenic medications Biological & Physiological Adaptation Satiety signals - fall with weight loss (Leptin, PYY, CCK, amylin, insulin, GLP-1) Orexigenic signals - rise with weight loss (eg. Ghrelin) Metabolic Rate falls with energy restriction & with weight loss Leslie et al 2007; Sumithran et al 2011; Maclean 2011; Leibel et al 1995;

11. 0 – 8 weeks ■ 810kcal/d liquid formula ▲ 415kcal/d liquid formula ♦ (E) Knee exercises group ● (C) Control – no intervention ■ (D) 1500kcal/d [average one formula meal/day] Copenhagen Weight Loss in Knee Osteoarthritis trial: more liberal TDR equally effective ) D = structured food/formula maintenance programme 8-16 weeks 1200kcal Part food/part formula 2 meals /day 810 kcal/d 415 kcal/d 415kcal VLED vs. 810kcal LED No sig. difference in weight loss n=64 per group n=96 per group (Bliddal et al, secondary care, dietitian managed)

12. Counterweight-Plus feasibility pilot (n = 91, BMI 47) (820kcal Total Diet Replacement, Food Reintroduction and Maintenance) Maintaining weight loss ≥15kg at 12 months: • 33% of all 91 patients • 44% of patients with a known 12-month weight • 57% of those who lost >15kg on LELD Highly cost-effective: • 4 times more lose >15kg as with bariatric surgery Weight Change (kg) -12.4kg -16.9kg Days Lean et al, Br J General Practice (2013)

13. National NHS GuidelineScottish Intercollegiate Guidelines Network115: Management of Obesity (2010) Weight loss interventions should be targeted to improving these comorbidities; in many individuals a greater than 15-20kg weight loss (will always be over 10%) will be required to obtain a sustained improvement in comorbidity”.

14. DiRECT trial design:Weight Management within routine Primary Care Design: Open-label, cluster-randomised by GP practices Typical T2D patients: Duration <6 years, age <65, HbA1c <10% Co-primary outcomes: • Numbers maintaining ≥15kg weight loss at 12 months • Numbers with remission of diabetes at 12 months (HbA1c <6.5%, off anti-diabetes drugs for >2 months) Powered to detect/exclude >22% remissions at 12m (n==280)

15. 1. Total Diet Replacement Nutritionally complete (vitamins & minerals) 830 kcal: 61%E carb, 13% fat, 26% protein 2. Stepped Food Reintroduction Add a ~400kcal meal every 2-3 weeks Step-counters: gradually increase PA 3. Weight Loss Maintenance Food-based diet +/- meal replacements 50%E carbohydrate, 35% fat, 15% protein Offer Relapse Management (regain >2kg) Visits 2-4 weekly at own primary care centres Programme delivered by usual primary care staff DiRECT Intervention: Counterweight-Plus Protocol Maintain PA ~30mins/ day STOP all diabetes meds STOP all BP meds Protocol for prescription based on Guideline Lean et al, Br J General Practice (2013), Leslie et al, BMC Family Practice (2016)

16. Results: participant retention Drop-outs: 12 month outcome data collected within a 100 day window from routine GP clinic records

17. Baseline data: analysed participants Total number 298 Men / women 59% / 41% Age (years) 54 (SD 7) Weight (kg) men 106 (SD 16) women 91 (SD 13) BMI (kg/m2) 35 (SD 4) Duration of T2DM (<6y) 3.1 (SD 1.7) HbA1c (mmol/mol) 59 (SD14)(7.5%) Diet alone 24% I drug 48% 2+ drugs 28% Blood Pressure 135/85 Smoking (current) 12% Former 38% Never 50% Intervention and Control groups well balanced for all criteria 40% from practices with highest deprivation

18. Baseline medical backgrounds • Diabetic Retinopathy 35 (12%) • Hypertension (BP>130/80) 169 (57%) 1 antihypertensive drug 69 (23%) 2+ antihypertensive drugs 94 (32%) • Antidepressant drugs 68 (23%) • Total prescribed drugs none 6 (2%) 1-2 47 (16%) 3-5 116 (39%) 6-9 89 (30%) 10+ 40 (13%)

19. Results: weight changes over 12 months -14.5 kg +1.9 kg +1.0 kg

20. ITT 12-month Primary Outcome Results 1st Co-Primary Outcome: ≥15 kg weight loss Intervention 36/149 (24%) p <0.0001 Control 0/149 2nd Co-Primary Outcome: Remission of diabetes* Intervention 68/149 (46%) p <0.0001 Control 6/149 (4%) *HbA1c <48 mmol/mol (<6.5%) off all anti-diabetes medication for at least 2 months

21. Remissions by 12m weight loss: entire study population 86.1% 57.1% 33.9% ≥10 kg loss: 73% are in remission 6.7% 0% None 0-5 kg 5-10 kg 10-15 kg ≥15 kg Weight loss at 12 months

22. ITT secondary outcomes: mean changes at 12m Intervention Control P Weight (kg) -10 -1 <0.0001 HbA1c (mmol/mol) -10 +1 <0.0001 % on anti-diabetes meds 22% 82% 0.0032 Systolic BP (mm Hg) -1.3 -1.7 ns % on antihypertensive meds 32% 61% <0.0001 Serum Triglycerides (mmol/l) -0.3 +0.1 <0.0001 Quality of Life (EQ5) +7.2 -2.9 0.0012

23. Taylor R, Lean M et al. unpublished (IDF 2017) Liver and pancreas in Responders (Remissions), Non-responders and Controls

24. DiRECT: Conclusions and Actions • T2DM results from excess body fat+ age. But not necessarily permanent • Almost half can achieve remission (73% with >10kg loss) • Main predictors of remission: • More weight loss, Greater age, Lower HbA1c, More anti-HTs • Warnings? More anti-diabetes meds, possibly antidepressants • Remission should be a primary aim of diabetes care • Record, recode and reward (patients and clinics) for remissions • Ethics: Offer optimal non-surgical weight management first, before prescribing (or trialling) additional treatments for T2D

25. McCombie et al BMJ 2017

26. Priorities of people currently living with T2DM The Lancet: Finer 2017 (Diabetes UK/ James Lind Alliance Priority-Setting Partnership) • Can type 2 diabetes be cured or reversed?

27. Referring practices Cost within DiRECT £1100 per patient entered £2500 per remission Draft Business Model200 referrals (1-to-1)80 remissions120 improvedCost = £170k pa Current cost to NHS of a T2D patient = £2500 p.a. Xin et al, Lancet DE, 2018 Referring practices Referring practices Referring practices Referring practices

28. Changing how we manage diabetes since Dec 2017:As of September 2018… England • First CCG (N. Tyneside) has commissioned a DiRECT/ Diabetes Remission Service for 270 patients Scotland • 5 -13 Health Board already using Counterweight Plus • New £42m allocated for sustainable intensive weight management for T2DM • ADA/EASD Joint statement has promoted Remission of T2DM as a management target, and described the DiRECT intervention • New T2DM Remission trials planned in US, India, Nepal, Oman, New Zealand, Europe • Interest in trial in Lebanon, Abu Dhabi, Saudi…….

29. The ‘No Doubts Diet’!830 kcal/day the cheap, culturally resonant, way https://www.directclinicaltrial.org.uk/Documents/The%20Lean%20Team%20No%20Doubt%20Diet%20plan.pdf

30. Preserving metabolic health:Improving weight-loss maintenance and diabetes remissions • Physical activity • Meal replacements • Intermittent ‘fasting’ • Lower carbohydrate diet? • Higher carbohydrate diet? • Low carbon emission diet • Personal trainer • Community support • Plate Model (Armstrong & Lean 1992) 145o

31. Health by stealth: the Eat Balanced Pizza

32. Low-Carb vs High-Carb diets for T2DM?Meta-analyses for Body Weight (A) and HbA1c (B) DOM, 2019 No significant effects for LDLc, HDLc, TC, BP, or attrition rate. TG fell 0.13mm0l.l more on LCD

33. Low Glycaemic Load Diets: Either reduce total carbohydrate Or choose slow-release low GI carbohydrates Or combine with other foods and nutrients eg. • Fats • Amylose-rich non-digestible carbohydrate • Legumes

34. Association between SSB and T2DM attenuated by adiposity: reduced ORs (black bars) and increased p values (grey bars)(Han TS, 坚持-专注 (Jim) & Lean MEJ: European J Nutr 2018)

35. Thank you http://www.directclinicaltrial.org.uk/ • GP practices and patients • Academic & clinical colleagues • Ethical and R&D committees • Cambridge Weight Plan • Counterweight Ltd • Diabetes UK, and funding donors

36. http://www.directclinicaltrial.org.uk/

37. http://www.directclinicaltrial.org.uk/

38. Predictors of Weight loss and Remission Sex!More men achieved ≥15kg weight loss than women (33% vs 14%, p=0.008), but remissions were similar (p=0.33). Agewas not associated with achieving ≥15kg weight loss (p=0.36),however older patients had more remissions (17% aged <50 y; 61% aged 60-65 y) Baseline HbA1c was not associated with weight loss (p=0.49), but remission more likely with lower HbA1c (OR 28% lower per % point HbA1c, p=0.038). Longer diabetes duration more likely to achieve ≥15kg weight loss(OR 31% greater per year (p=0.032), but duration was not associated with remission. Neither outcome was associated with socioeconomic deprivation, smoking or alcohol intake.

39. Predictors of Weight loss and T2D Remission Loss of ≥15kg was more likely with greater baseline weight (<90kg: 8%; ≥110kg: 40%; trend p=0.024) or BMI (<30kg/m2: 9%; ≥40kg/m2: 31%; p=0.027). • But neither weight (p=0.93) nor BMI (p=0.26) associated with T2D remission. Prior anti-diabetic and anti-hypertensive drugs not associated with weight loss. • Remission less likely with more anti-diabetic drugs (OR 0.43 per drug, p<0.001) • Remission more likely with more antihypertensives(OR 1.37 per drug, p=0.045) Blood pressure was not associated with weight loss, • Remission more likely with higher systolic (p=0.017) & diastolic BP (p=0.013).. Antidepressant use may impair ≥15kg weight loss (p=0.085) and remission (p=0.064).