The debate! Ajay Singh YES, WE CAN Dr. Locatelli NO, WE CAN T - - PowerPoint PPT Presentation

Ajay Singh YES, WE CAN

• Dr. Locatelli

NO, WE CANT

The debate!

Should hemoglobin targets for anemia patients with chronic kidney disease be changed?

DISCLOSURES

PI of the CHOIR study, received research support and consulting income from J and J Member, Executive Steering Committee of the TREAT study, received research support and consulting income from Amgen Consultant for Rockwell, Fibrogen, GSK, Sandoz, AHSP,

• ConCert. Past income from Watson, Roche, AMAG

Member of MEDCAC Steering Committee for the National Quality Forum Member Scientific Advisory Board, National Kidney Foundation

2

Gerard Dou 1662, Louvre Doctors are men who prescribe medicines of which they know little, to cure diseases of which they know less, in human beings of whom they know nothing Voltaire (French Philosopher and Writer. One of the greatest of all French authors, 1694-1778)

The clinical context

• Patient with CKD, either non-dialysis or

dialysis

• Anemic iron replete
• What level of Hb should treatment be

initiated?

• What target Hb should you aim for?

Should Hb targets be changed?

Yes!

Initiate ESA at 9 g/dL not at 10 g/dL Aim for >9 g/dL and <12 g/dL Because the evidence supports it!

Lower Hb threshold should be 9 g/dL

Data from Normal Hematocrit in dialysis patients Data from TREAT in non-dialysis CKD patients

Upper Hb level should be left unchanged at 12 g/dL Focus should be on using lower doses of ESA

Lower Hb threshold should be 9 g/dL

• Normal Hematocrit Study
• Symptomatic, high-risk dialysis patients
• Similar mortality to the USRDS dialysis population
• Representative, generalizible study

Besarab et al. N Engl J Med 339:584-590, 1998

Normal Hematocrit Study

Randomization was to 2 Hb arms: 27-33% -- versus 39-45% Ie, >9 g/dL and <11 g/dL versus >13 g/dL and <15 g/dL

Normal Hematocrit Study

• Tested hypothesis that patients with normal

Hb 13-15 g/dL will have better outcomes than patients with Hb 9-11 g/dL

• 1233 HD patients with CAD or CHF
• Primary end-point: Death or MI

Besarab et al. N Engl J Med 339:584-590, 1998

8

Normal Hematocrit Study

Low Hct Normal Hct n 618 615 Hct 30 42 (achieved 39%) Epoetin dose 160 460 Total deaths 150 183 Non-fatal MI 14 19 RR 1.3 (0.9-1.9)

Besarab et al. N Engl J Med 339:584-590, 1998

9

Study terminated early due increased risk Higher rate of vascular access thrombosis in normal Hct group: (243 patients, or 39 percent, vs. 176 patients, or 29 percent; P=0.001).

Normal Hematocrit Study

In dialysis patients, the only well powered study, demonstrates that patients treated to a range of 9-11 g/dL have better outcomes than those treated to a range of 13-15 g/dL

Lower Hb threshold should be 9 g/dL

• What about non-dialysis patients?
• TREAT study
• Type 2 diabetic patients with kidney disease
• Over 50% of kidney disease patients have type

2 diabetes

In patients with type 2 diabetes, chronic kidney disease not requiring dialysis, and concomitant anemia, raising hemoglobin with darbepoetin alfa would lower the rates of death and cardiovascular morbidity and/or death and end- stage renal disease

Pfeffer MA, et al N Engl J Med. Oct 30. 2009

Hypothesis

13

TREAT

Pfeffer MA, et al N Engl J Med. Oct 30. 2009

TREAT tested maintaining a Hb of > 9g/dL versus 13 g/dL Achieved Hb of Hb Median: 10.6 IQR [9.9 11.3] versus Hb Median: 12.5 IQR [12.0 12.8]

TREAT: Hemoglobin levels

Mean Hemoglobin (g/dL)

9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 Placebo Darbepoetin alfa

Month

6 12 18 24 30 36 42 48

Median dose: 176 g IQR [104 305] Mean: 225 g ± 208 Median dose: 0 g IQR [0 5] Mean: 5 g ±11 Hb Median: 12.5 IQR [12.0 12.8] Hb Median: 10.6 IQR [9.9 11.3]

16

Pfeffer MA, et al N Engl J Med 2009; 361:2019-203

TREAT: Composite endpoints

Pfeffer MA, et al N Engl J Med. Oct 30. 2009

17

Sensitivity Analysis around endpoints (Source: Amgen)

18

10 20 30 40 50 6 12 18 24 30 36 42 48

Patients With Events (%) Months

Fatal and Nonfatal Stroke

HR: 1.92 (1.38 2.68) P < 0.001 Darbepoetin alfa 101 (5.0%) Placebo 53 (2.6%)

19

MACE Analysis (Source FDA)

20

Stroke Analysis

21

Type of stroke in TREAT DPO Control Overall 101 (2.6%) 101 (5.0%) Non Hem 38 74 Hemorrhagic 8 13 Unknown 7 14

• Dr. Locatelli might argue that in TREAT, stroke was

not a primary endpoint and TREAT was not powered for reduction in the rate of stroke, the highly significant P value associated with the increased hazard of a stroke with darbepoietin treatment should be ignored.

Stroke was a pre-specified cardiovascular outcome. Stroke safety signal is meaningful, dont need to power for safety signal TREAT well powered study

N=1026 Hazard ratio 1.30 (95% C: 1.06, 1.58)

Malignancy in TREAT

Darbepoetin alfa Placebo P-value Overall Cancer-related AE 139/2012 6.9% 130/2026 6.4% 0.53 Deaths attributed to cancer 39/2012 1.9% 25/2026 1.2% 0.08 Subgroup: Baseline History of malignancy (n = 348) All cause mortality 60/188 31.9% 37/160 23.1% 0.13 Deaths attributed to cancer 14/188 7.4% 1/160 0.6% 0.002

23

In non-dialysis patients, TREAT demonstrates that patients maintaining a Hb>9 g/dL have improved outcomes over those treated to a range of >13 g/dL

Minimal Improvement in QOL Hb >9 g/dL FACT-Fatigue Score at 25 Weeks

30.2 Darbepoetin alfa n = 1762 30.4 Placebo n = 1769 P = 0.002 54.7% 4.2 ± 10.5 2.8 ± 10.3 P < 0.001 49.5% FACT-Fatigue range: 0: most fatigued, to 52: least fatigued Less fatigue Mean Change

25

Patient Reported Outcomes Supportive Analysis

Domains Darbepoetin alfa n = 1138 Placebo n = 1157 P-value Energy 2.6 ± 9.9 2.1 ± 9.7 0.20 Physical Function 1.3 ± 9.2 1.1 ± 8.8 0.51

Short-Form 36 Mean Change at 25 Weeks in 2 Domains

26

What Dr. Locatelli might say

A cautious approach of treating anemia to a lower target 9 to 12 g/dL with conservative doses of ESA might return us to a pre-ESA era. On the contrary: Hb of 9 g/dL is not returning us to the pre-ESA era of Hb s of 5 to 9 g/dL TREAT and other studies show minimal QOL benefit of rasing Hb above 9 g/dL Individualizing treatment is essential; besides, no-one is arguing that we abandon ESAs

Hb g/dL Hb g/dL

10 10 11 11

% of % of Patients Patients

13 13 12 12 9 9 Low Low ESA ESA dose dose Singh et al CJASN 2010

• Dr. Locatelli might dismiss the TREAT study

He might argue that since subjects assigned to the placebo rescue arm achieved an Hb very close to the levels recommended in the guidelines, the Hb guidelines should not be changed.

But, the target Hb and achieved Hb are two different concepts

• Target Hb should be >9 g/dL, the achieved Hb

will likely by in the 10 to 11 g/dL range The guidelines should be changed to recommend, In most non-dialysis CKD patients anemia can be treated without the use of ESAs; in those in whom treatment is necessary the target Hb should be >9 g/dL and <12 g/dL .

Is there risk with achieving Hb >12 g/dL YES!

• CHOIR and CREATE
• Non-dialysis patients
• Diabetic and non-diabetic
• CREATE very under-powered

1432 patients, 130 centers, US only Epoetin-alfa Randomization High target Hgb (13.5 g/dl) n=715 Low target Hgb (11.3 g/dl) n=717 Median f/u 16 months Singh et al,New Engl J Med 2006; 355:2085-98

30

Doses of Epoetin alfa and Hb

N Hb 13.5 709 693 659 623 578 530 500 452 370 310 258 189 132 97 79 65 52 27 11 Hb 11.3 707 691 655 621 577 549 526 479 393 333 262 189 141 95 73 54 43 27 12 N Hb 13.5 709 693 659 623 578 530 500 452 370 310 258 189 132 97 79 65 52 27 11 Hb 11.3 707 691 655 621 577 549 526 479 393 333 262 189 141 95 73 54 43 27 12 N Hb 13.5 709 693 659 623 578 530 500 452 370 310 258 189 132 97 79 65 52 27 11 Hb 11.3 707 691 655 621 577 549 526 479 393 333 262 189 141 95 73 54 43 27 12 N Hb 13.5 709 693 659 623 578 530 500 452 370 310 258 189 132 97 79 65 52 27 11 Hb 11.3 707 691 655 621 577 549 526 479 393 333 262 189 141 95 73 54 43 27 12

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Epoietin-alfa Dose (U) and 95% C.I.

2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Epoietin-alfa Dose (U) and 95% C.I.

2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Epoietin-alfa Dose (U) and 95% C.I.

2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Epoietin-alfa Dose (U) and 95% C.I.

2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Mean Weekly Dose 11,215 units Mean Weekly Dose 6276 units

N Hb 13.5 710 667 632 600 558 507 485 433 367 306 252 194 139 95 81 67 49 31 13 Hb 11.3 707 672 625 603 549 528 510 471 384 334 250 182 141 101 75 60 45 30 13 N Hb 13.5 710 667 632 600 558 507 485 433 367 306 252 194 139 95 81 67 49 31 13 Hb 11.3 707 672 625 603 549 528 510 471 384 334 250 182 141 101 75 60 45 30 13 N Hb 13.5 710 667 632 600 558 507 485 433 367 306 252 194 139 95 81 67 49 31 13 Hb 11.3 707 672 625 603 549 528 510 471 384 334 250 182 141 101 75 60 45 30 13 N Hb 13.5 710 667 632 600 558 507 485 433 367 306 252 194 139 95 81 67 49 31 13 Hb 11.3 707 672 625 603 549 528 510 471 384 334 250 182 141 101 75 60 45 30 13

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Hemoglobin (g/dL) and 95% C.I.

9.5 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 15.5

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Hemoglobin (g/dL) and 95% C.I.

9.5 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 15.5

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Hemoglobin (g/dL) and 95% C.I.

9.5 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 15.5

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

Mean Hemoglobin (g/dL) and 95% C.I.

9.5 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 15.5

Study Month 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Achieved Mean Hb 12.6 g/dL Mean Hb 11.3 g/dL Singh et al,New Engl J Med 2006; 355:2085-98

31

Primary Outcomes

Primary Composite Endpoint: Death, MI, CHF hosp (no RRT) and/or stroke

Randomized Treatment

Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL Kaplan-Meier Failure Estimate (%) 0% 5% 10% 15% 20% 25% 30%

Months from Randomization

6 12 18 24 30 36

715 717 587 594 457 499 270 293 55 44 101 107 3

Hazard ratio 1.337 (1.025, 1.743) P= 0.0312

125 97

32

Achieved Mean Hb 12.6 g/dL

Randomized Treatment Hemoglobin Target 13.5 g/dL Hemoglobin Target 11.3 g/dL

0% 5% 10% 15% 20% 6 12 18 24 30 36 0% 5% 10% 15% 20% 6 12 18 24 30 36 0% 5% 10% 15% 20%

Months from Randomization

6 12 18 24 30 36 0% 5% 10% 15% 20%

Months from Randomization

6 12 18 24 30 36

CHF Hospitalization (where RRT did not occur) Myocardial Infarction Stroke Death

p = 0.0727 p = 0.0674 p = 0.9803 p = 0.7836

Components of the Primary Endpoint

Hazard ratio 1.483 (0.969, 2.268) Hazard ratio 1.409 (0.967, 2.054) Hazard ratio 1.010 (0.454, 2.249) Hazard ratio 0.915 (0.484, 1.729)

65 deaths

33

Hb 12.6 g/dL Hb 12.6 g/dL

CHOIR: QOL

• 3 instruments
• LASA
• KDQ
• SF-36

QOL Increased in both groups No difference between groups

Singh et al,New Engl J Med 2006; 355:2085-98

34

CREATE: Primary endpoint Time to first CV events (105 events)

Events: 58 vs 47 HR=0.78 (0.531.14) Log rank test p=0.20 Drueke TB, et al N Engl J Med. 355:2071-84, 2006

35

CREATE: Secondary endpoints Time to dialysis

n at risk Group 1 301 281 255 211 162 115 62 35 Group 2 302 293 269 243 199 138 82 33 6 12 18 24 30 36 42 48 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Study month Probability Group 1 Group 2 Events: 127 vs 111 HR=0.76 (0.590.98) Log rank test p=0.034

Gp 1 Gp 2

36

There is increased risk >12 g/dL

• higher rate of stroke in TREAT
• higher rate of death and CVD in CHOIR
• trend to worse outcome in CREATE

• Dr. Locatelli might use the heterogeneity in

clinical signals between CHOIR, CREATE and TREAT to argue that these studies somehow invalidate each other. Various differences in the 4 RCT s might provide an explanation: » the study population » the use of different ESA doses » the trials used different ESAs.

Class effect

Epoetin-alfa (CHOIR) Darbepoetin-alfa (TREAT) Epoetin-beta (CREATE)

Exposure to higher and different doses of ESA CHOIR higher hemoglobin arm received median

• f 10,952 units per week.

TREAT median dose of 8800 units per week in the darbepoetin treated arm CREATE, median dose of 5000 units of epoetin- beta per week was used in the higher hemoglobin arm. Enrollment of different populations? TREAT, all diabetics CHOIR 50% diabetics CREATE 25% diabetes mellitus

39

The only rational conclusion that one can reach is that the Hb in dialysis and non-dialysis patients should be 9 to 12 g/dL

It is a capital mistake to theorize before you have all the evidence. It biases the judgment. The Adventures of Sherlock Holmes, "A Scandal in Bohemia

A Response Based Algorithm for Treatment for CKD anemia in non-dialysis patients

41

CKD patient Hb<9 g/dl Define Responder Status: ESA Test Dose No Response No change (<2%) in Hb at 1 mo Response Change (>2%) in Hb at 1 mo Individualize therapy based

• n symptoms

No ESA /hold ESA escalation Search for and treat causes of non-responsiveness

• eg. Infection, failing allograft

Solomon SD et al. N Engl J Med 2010;363:1146-1155

Conclusions:

Locatelli versus Singh

What explains Increased Risk

• High Hb hyperviscosity? Activation of

endothelial cells/ platelets

• Rise in Hb
• Activation of heterodimeric ESA receptor
• Exposure to high doses of ESA
• Iron exposure
• Hypertension in ESA treated patients

44

CHOIR: Epo Dose and Outcome

Szczech LA, et al: Kidney Int 74: 791798, 2008

45

Mechanisms of ESA toxicity

rHuEPO stimulates erythropoiesis by activating EPO receptors

membrane

Homodimeric EPO Receptor rHuEPO Heterodimeric EPO Receptor Bone Marrow Heart, Brain, Kidney, endothelial cells

Response Based Approach

Guide-post should be the placebo arm of TREAT

Rescue patients if Hb falls below 9 g/dL

ESAs should be administered as a test dose

Initiate rescue therapy with darbepoetin alfa with a single dose of 0.45 g/kg. no dose escalation if there is no response Individualize treatment for patients who respond with ESA and iron.

Individualize anemia treatment above a Hb of 9 g/dL

For example: If patients feel good at 9.8 g/dL with low dosage of ESA, leave them there. If treatment to 12.0 g/dL is needed to abrogate fatigue then so be it. Until more evidence, upper limit should be 12 g/dL

48