19Th Expert Committee on the Selection and Use of Essential Medicines April 8‐12 2013

EXPERT REVIEW No.2 19th Expert Committee on The Selection and Use of Essential Medicines April 8‐12 2013

Expert peer review on application for SPIRONOLACTONE

Introduction

This application is proposing the inclusion of

1. Assessment of efficacy a. Have all relevant studies on efficacy been included Yes No (if no, please provide reference and information) The primary trials have all been identified through the use of the systematic review by Ezekowitz and McAlister (2009)plus the main trial for eplerenone in heart failure, EMPHASIS _HF, Zannad et al, 2011. There is one more recent systematic review and meta‐analysis (Chatterjee et al, Am J Med, 2012; 125:817‐825) that includes the eplerenone trial that is not cited in the application but does not alter the estimate of effect. Of note, both the large trials for spironolactone and eplerenone were stopped early for benefit. b. Summarize the data on efficacy, in comparison to what is listed in EML where applicable (limit to 2 to 3 sentences)

Spironolactone reduces all cause mortality in patients with NYHA Class III‐IV heart failure, as well as hospitalization. A summary of the results for the critical outcomes from the main two trials is shown below. Pooled estimates from the Chatterjee meta‐analysis are also included. There are no trials directly comparing spironolactone with eplerenone. c. Please provide any additional relevant information with reference

2. Assessment of safety a. Have all relevant studies on safety been included Yes No (if no, please provide reference and information)

As noted above the Chatterjee meta‐analysis provides updated pooled estimates for hyperkalemia and gynaecomastia. b. Summarize the data on safety, in comparison to what is listed in EML where applicable (limit to 2 to 3 sentences)

SUMMARY RESULTS

Placebo Intervention Relative effect Risk difference n/N n/N (95%CI) (95%CI) Heart failure hospitalisation EMPHASIS 253/1373 (18.4%) 164/1364 HR 0.58 - 6.4% (12.0%) (0.47, 0.70) (-9.1%,-3.7% RALES 336/841 (40.0)1 260/822 (31.6) 0.70, (0.59, 0.82) -8.4 All cause mortality Chatterjee – 438/1720 316/1820 0.74 (0.66, 0.83) spironolactone/canrenone Chatterjee - eplerenone 767/4686 649/4683 0.85 (0.77, 0.93) CV death EMPHASIS 185/1373(13.5%) 147/1363 0.76 - 2.7% (10.8%) (0.61, 0.94) (-5.1%, -0.3%) RALES 314/842 (37.3) 226/822 (27.5) 0.69 (0.58, 0.82) -9.8 Chatterjee – 403/1183 302/1370 0.75 (0.67, 0.84) spironolactone/canrenone Chatterjee - eplerenone 668/4686 554/4683 0.83 (0.75, 0.92) Fatal/nonfatal MI EMPHASIS 33/1373 (2.4%) 45/1364 (3.3) 1.32 (0.84, 2.06) -1.1 RALES2 15? 17? Hyperkalemia (reported as serious) EMPHASIS 50/1369 (3.7%) 109/1360 (8.0%) Not calculated 4.3% RALES 10/841(1.2%) 14/822(1.7%) Not calculated 0.5% Chatterjee – 77/1702 75/1640 1.80 (0.83, 3.91) spironolactone/canrenone Chatterjee - eplerenone 177/4708 293/4781 1.72 (1.19, 2.47) GYNAECOMASTIA Chatterjee – 9/1044 75/1235 6.26 (3.38, spironolactone/canrenone 11.57) Chatterjee - eplerenone 31/4682 23/4679 0.74(0.43, 1.27)

Spironolactone is associated with an increased risk of gynaecomastia, which may not be as frequent with eplerenone. However, both are associated with an increased risk of hyperkalemia, which maybe worse with eplerenone. Both products would require monitoring of potassium concentrations to ensure safe use. c. Please provide any additional relevant information with reference

3. Assessment of cost and availability a. Have relevant data on cost been provided Yes No (if no, please provide reference and information) b. Summarize the data on cost and cost effectiveness, in comparison to what is listed in EML where applicable (limit to 2 to 3 sentences)

1 Data extracted from text, not table as table reports events, not patients 2 outcome not clearly reported.

2 As stated in the application, spironolactone is widely available and costs a few cents per tablet. Eplerenone is substantially more expensive. Several cost‐effectiveness studies from high income countries show that spironolactone, compared to placebo as a add‐on therapy for heart failure patients already treated with ACE‐inhibitors and beta‐blockers, is cost effective under most assumptions. The WHO cost effectiveness threshold of three times GDP per capita is inappropriate for selecting medicines, however – prices that result from use of this threshold are unaffordable in many settngs. Eplerenone is not clearly cost‐effective when compared to placebo, and is not cost‐effective compared to spironolactone given that is no more effective (except with respect to gynaecomastia) and substantially more expensive. c. Please provide any additional relevant information with reference

NA d. Is the product available in several low and middle income countries?

YES

4. Assessment of public health need a. Please provide the public health need for this product (1‐2 sentences)

This is adequately addressed in the application and there is a significant public health need. b. Do guidelines (especially WHO guidelines) recommend this product? If yes, which ones? List 1 or 2 international preferable

Several international cardiology guidelines recommend use of spironolactone. The application does not identify any relevant WHO guidelines.

5. Are there special requirements for use or training needed for safe/effective use? If yes, please provide details in 1‐2 sentences

YES, use of spironolactone in patients with heart failure requires access to capacity to monitor plasma potassium concentrations, especially in patients with renal impairment.

6. Is the proposed product registered by a stringent regulatory authority? Yes No

7. Any other comments

8. What is your recommendation to the committee (please provide the rationale)

Spironolactone meets the criteria for an Essential Medicine, is already on the WHO EML as a diuretic and should be added to Section 12.4. I do not support listing with a ‘square box’ because of the current enormous price differential between spironolactone and eplerenone as the main other aldosterone antagonist and possible significant differences in safety profile with respect to hyperkalemia.

References

Chatterjee S, Moeller C, Shah N, Bolorunduro O, Lichstein E, Moskovits N, Mukherjee D. Eplerenone is not superior to older and less expensive aldosterone antagonists. Am J Med. 2012 Aug;125(8):817‐25. doi: 10.1016/j.amjmed.2011.12.018.