CLINICAL RESEARCH Effects of SNF472, a Novel Inhibitor of Hydroxyapatite Crystallization in Patients Receiving Hemodialysis — Subgroup Analyses of the CALIPSO Trial Paolo Raggi1, Antonio Bellasi2, Smeeta Sinha3, Jordi Bover4, Mariano Rodriguez5, Markus Ketteler6, David A. Bushinsky7, Rekha Garg8, Joan Perelló9,10, Alex Gold11,12 and Glenn M. Chertow12 1 Division of Cardiology and Department of Medicine, University of Alberta, Edmonton, Alberta, Canada; 2Research, Innovation and Brand Reputation Unit, ASST Papa Giovanni XXIII, Bergamo, Italy; 3Department of Renal Medicine, Salford Royal NHS Foundation Trust, Salford, United Kingdom; 4Department of Nephrology, Fundació Puigvert and Universitat Autònoma, IIB Sant Pau, REDinREN, Barcelona, Spain; 5Nephrology Unit, Hospital Universitario Reina Sofia, IMIBIC, REDinREN, Córdoba, Spain; 6Department of General Internal Medicine and Nephrology, Robert-Bosch-Krankenhaus, Stuttgart, Germany; 7Depart- ment of Medicine, University of Rochester School of Medicine, Rochester, New York, USA; 8PharmaDRS, LLC, San Diego, California, USA; 9Research and Development, Sanifit Therapeutics, Palma, Spain; 10University of the Balearic Islands, Palma, Spain; 11Research and Development, Sanifit Therapeutics, San Diego, California, USA; and 12Department of Medicine, Stanford University, Palo Alto, California, USA Introduction: Coronary artery calcium (CAC) is highly prevalent and linked with poor outcomes in patients receiving maintenance hemodialysis, and its reduction may improve patient prognosis. SNF472, a se- lective inhibitor of hydroxyapatite crystallization, slows CAC progression in patients receiving mainte- nance hemodialysis. In this analysis, we assessed the efficacy of SNF472 in prespecified patient subgroups. Methods: In a randomized clinical trial SNF472 300 mg, SNF472 600 mg, or placebo were infused thrice weekly in 91, 92, and 91 patients receiving maintenance hemodialysis and with CAC at baseline, respec- tively. In prespecified subanalyses, the percent change in CAC volume score (CACvs) from baseline to week 52 in modified intention-to-treat (mITT) and per-protocol (PP) populations was calculated in the following subgroups: age, sex, diabetes mellitus, dialysis vintage, prior atherosclerotic cardiovascular disease, baseline use of non-calcium and calcium-based phosphate binders, calcimimetics, activated vitamin D, warfarin, and statins. Results: In the main trial, SNF472 significantly reduced CACvs progression compared with placebo (11% versus 20% mITT analyses; P ¼ 0.016; 8% vs. 24% PP analyses; P < 0.001). Treatment differences for CACvs progression were similar across all subgroups, and all interaction P values were non-significant in mITT and PP analyses. Conclusions: SNF472 treatment for 52 weeks reduced CACvs progression compared with placebo in a broad range of patients receiving maintenance hemodialysis. Future studies will determine the impact of SNF472 on cardiovascular events in this population. Kidney Int Rep (2020) 5, 2178–2182; https://doi.org/10.1016/j.ekir.2020.09.032 KEYWORDS: coronary calcification; vascular calcification; hemodialysis ª 2020 International Society of Nephrology. Published by Elsevier Inc. This is an open access article under the CC BY- NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). AC is highly prevalent and linked with an within atherosclerotic plaques and in the media layer C increased risk of morbidity and mortality in pa- tients with end-stage kidney disease (ESKD).1 In these due to dysregulation of mineral metabolism.2 CAC is associated with a variety of cardiovascular events patients, CAC accumulates in the arterial intima such as acute and chronic coronary artery syndromes, heart failure with preserved and reduced ejection frac- tion, cardiac arrhythmias, and sudden death.1 Correspondence: Paolo Raggi, Division of Cardiology, University Although the initiating stimulus may be different, of Alberta, 5A9-014, 11220 83rd Avenue NW, Edmonton, Alberta T6G 2B7, Canada. E-mail: raggip@gmail.com all calcification processes ultimately depend on the Received 22 July 2020; revised 10 September 2020; accepted 15 formation of calcium phosphate crystals and, in most September 2020; published online 4 November 2020 cases, they result in the deposition of a 2178 Kidney International Reports (2020) 5, 2178–2182 P Raggi et al.: Subgroup Analyses of CALIPSO CLINICAL RESEARCH Table 1. Patient characteristics according to prespecified blinded to treatment allocation to assess inter-reader subgroupsa,b agreement. The study was approved by the research Combined SNF472 ethics committee of each participating institution and Placebo doses Overall Characteristics N [ 91 N [ 183 N [ 274 the trial was conducted according to the principles of Age, yr, mean  SD 64.1  8.2 63.3  9.0 63.5  8.9 the Declaration of Helsinki. Female 33 (36.2) 74 (40.4) 107 (39.0) The primary efficacy analysis model was an analysis Dialysis vintage, mo, median 34.5 (6–263) 44.3 (7–521) 41 (6–521) of covariance with the change in log score (log[week (range) 52] – log[baseline]) as the dependent variable, and a Diabetes mellitus 60 (65.9) 114 (62.2) 174 (63.5) fixed effect term for randomized treatment group as Atherosclerotic cardiovascular 45 (49.5) 70 (38.3) 11 (40.1) disease well as log(baseline) as covariates; the model was Non-calcium-based phosphate 55 (60.4) 119 (65.0) 174 (63.5) stratified by baseline Agatston CAC score. The geo- binders metric least square means and 95% confidence in- Calcium-based phosphate binders 26 (28.6) 55 (30.0) 81 (29.5) Calcimimetics 26 (28.6) 60 (32.7) 86 (31.3) tervals were estimated and back-transformed to yield Activated vitamin D 47 (51.6) 97 (53.0) 144 (52.5) mean percent change from baseline to week 52 for the Warfarin 5 (5.5) 13 (7.1) 18 (6.5) primary analysis of the combined SNF472 groups Statins 53 (58.2) 115 (62.8) 168 (61.3) compared with placebo. The primary analyses were a Values shown are n (%) unless otherwise stated. conducted on an mITT population (patients who b Patients could take both calcium-based and non–calcium-based phosphate binders simultaneously. received at least 1 dose of SNF472 or placebo and had an evaluable multidetector computed tomography at baseline and 52 weeks, or any time during follow-up if hydroxyapatite solid phase. SNF472 is an intravenous a patient was discontinued prematurely from the formulation of the hexasodium salt of myo-inositol study). The PP population included those in the mITT hexaphosphate with highly effective in vitro and who met all study entry criteria, received at least 80% in vivo inhibition of hydroxyapatite nucleation and of study drug, and had a week-52 visit and multi- crystal formation.3 In the phase 2b Calcifactant for detector computed tomography scan. We report results Acute Lung Injury in Pediatric Stem Cell Transplant in predefined subgroups at baseline: age (<65 or $65 and Oncology Patients (CaLIPSO) trial of patients years), sex, diabetes mellitus, dialysis vintage (<2, 2 receiving maintenance hemodialysis, SNF472 signifi- to <5, $5 years), prior atherosclerotic cardiovascular cantly reduced the progression of CACvs compared disease, non-calcium and calcium-based phosphate with placebo in an mITT population, and the effect binders, calcimimetics, activated vitamin D, warfarin, was confirmed in a PP population.4 In this report, and statins. we present results of prespecified patient subgroup analyses. RESULTS METHODS Table 1 shows the patient characteristics of the pla- cebo and pooled SNF472 cohorts. Table 2 shows the The study design has been published in detail before.5 baseline and week-52 serum levels of calcium, phos- Briefly, we randomized 274 adult patients (18 to 80 phorus, and intact parathyroid hormone for the placebo years old) on maintenance hemodialysis for $6 and the combined SNF472 treatment groups. In the months, with diabetes mellitus if younger than 55 primary analyses, SNF472 significantly reduced years of age, and baseline CAC Agatston score between progression of log CACvs compared to placebo 100 and 3500 units on a screening noncontrast multi- detector computed tomography to placebo, 300 mg, or Table 2. Serum calcium, phosphorus, and intact parathyroid 600 mg of SNF472 infused thrice weekly during he- hormone at baseline and week 52 modialysis for 52 weeks. The Agatston CAC score is Mean ± SD, (N) commonly used in clinical practice; however, the Analyte Placebo (N [ 90) Combined Dose (N [ 183) CACvs is more reproducible. Therefore, the CACvs was Calcium, mmol/L used to measure CAC change.6 Multidetector computed Baseline 2.18  0.17 (84) 2.22  0.19 (164) tomography imaging was performed at enrollment and Week 52 2.23  0.18 (73) 2.23  0.18 (183) Phosphorus, mmol/L at week 52, or as soon as feasible after voluntary Baseline 1.42  0.51 (32) 1.31  0.38 (60) withdrawal or kidney transplantation. An experienced Week 52 1.49  0.53 (35) 1.39  0.51 (70) radiologist blinded to patient data, including treatment Intact parathyroid hormone, assignment, reviewed the quality of all scans for pmol/L acceptability and measured CACvs. A portion of all Baseline 48.3  41.0 (84) 44.6  36.7 (161) Week 52 46.0  37.9 (71) 39.0  32.5 (149) scans (w15%) was reviewed by an expert reader Kidney International Reports (2020) 5, 2178–2182 2179 CLINICAL RESEARCH P Raggi et al.: Subgroup Analyses of CALIPSO Figure 1. Progression of coronary artery calcium volume score in patient subgroups (modified intention-to-treat analyses). CAC, coronary artery calcium; CI, confidence interval; mITT, modified intention to treat; Pi, phosphate. (mITT: 11% vs. 20%, P ¼ 0.016; PP: 8% vs. 24%, solid phase. SNF472 is a first-in-class inhibitor of P < 0.001). Figure 1 shows the difference in pro- crystallization of calcium and phosphorus, whose gression of CACvs in the prespecified subgroups in metabolism is highly dysregulated in ESKD. Several mITT analyses. Figure 2 shows the difference in factors have been associated with CAC progression. progression of CACvs in the prespecified subgroups Age is most influential, as atherosclerosis is almost in the PP analyses. All interactions were nonsig- universal among adult individuals. In the general nificant and SNF472 consistently reduced progres- population, men show an earlier onset and larger sion of CACvs in all prespecified subgroups. accumulation of CAC compared to women,7 although this difference may be attenuated in ESKD. Diabetes mellitus is a strong predictor of progression of CAC in DISCUSSION the general population and in ESKD where it is often Vascular calcification is a complex phenomenon associated with low turnover bone disease and accel- involving numerous active and passive mechanisms. erated vascular calcification.8 Prior atherosclerotic The final step requires nucleation of calcium phosphate cardiovascular disease is an obvious risk factor for the crystals that lead to the formation of a hydroxyapatite progression of CAC, whereas time on dialysis 2180 Kidney International Reports (2020) 5, 2178–2182 P Raggi et al.: Subgroup Analyses of CALIPSO CLINICAL RESEARCH Figure 2. Progression of coronary artery calcium volume score in patient subgroups (per-protocol analyses). CAC, coronary artery calcium; CI, confidence interval; Pi, phosphate. predisposes to CAC accumulation through continuing crystallization with this agent can reduce progression exposure to dysregulated mineral and bone meta- of CAC independently of comorbidities or concomitant bolism, as well as chronic inflammation and oxidative treatments. stress.9 Some medical interventions appear to affect This study had several limitations: we did not record CAC progression. Warfarin and statins, through a or analyze calcium and phosphate balance during or probable interference with natural inhibitors of calci- between dialysis sessions and we did not mandate a fication, have been associated with progression of single dialysate calcium concentration in the CALIPSO CAC.10 Large doses of active vitamin D may accelerate study. However, it is unlikely that variability in these and calcimimetics may reduce CAC progression.11,12 parameters could explain the treatment effect in a ran- The non–calcium-based phosphate binders are the domized double-blind controlled study where patient only class of drugs that have been shown in open- cohorts had similar baseline clinical characteristics. We label randomized clinical trials to reduce progression did not measure serological markers of bone metabolism. of CAC when compared to calcium-based binders.13,14 A meta-analysis of several randomized controlled SNF472 demonstrated similar efficacy in all sub- trials comparing non–calcium-based to calcium-based groups, showing that inhibition of hydroxyapatite phosphate binders suggested that the former reduce Kidney International Reports (2020) 5, 2178–2182 2181 CLINICAL RESEARCH P Raggi et al.: Subgroup Analyses of CALIPSO CAC progression and might reduce mortality in 2. Raggi P, Giachelli C, Bellasi A. Interaction of vascular and ESKD.15 Larger and longer-term studies of SNF472 will bone disease in patients with normal renal function and pa- tients undergoing dialysis. Nat Clin Pract Cardiovasc Med. be required to determine whether SNF472 can reduce 2007;4:26–33. the frequency of heart failure, stroke, sudden cardiac 3. Ferrer MD, Ketteler M, Tur F, et al. Characterization of SNF472 death, and other complications associated with coro- pharmacokinetics and efficacy in uremic and non-uremic rats nary, cerebrovascular, and peripheral arterial calcifi- models of cardiovascular calcification. PLoS ONE. 2018;13: cation in ESKD. e0197061. 4. Raggi P, Bellasi A, Bushinsky D, et al. Slowing progression of cardiovascular calcification with SNF472 in patients on he- DISCLOSURES modialysis: results of a randomized phase 2b study. Circula- tion. 2020;141:728–739. PharmaScribe, LLC, provided medical writing assistance with financial support from Sanifit Therapeutics, S.A. 5. Bellasi A, Raggi P, Bover J, et al. Trial design and baseline characteristics of CaLIPSO: a randomized, double-blind pla- PR has served as a consultant to Sanifit. AB has served cebo-controlled trial of SNF472 in patients receiving hemo- as a consultant to Sanifit and received lecture fees from dialysis with cardiovascular calcification [e-pub ahead of Sanofi-Genzyme, Vifor-Fresenius-Renal Pharma, Abbvie, print] Clin Kidney J. 2019. https://doi.org/10.1093/ckj/sfz144. and Amgen. JB has served as a consultant to and 6. Callister TQ, Cooil B, Raya SP, et al. Coronary artery disease: received lecture fees from Sanifit, Sanofi-Genzyme, improved reproducibility of calcium scoring with an electron- Vifor-Fresenius-Renal Pharma, Abbvie, and Amgen; and beam CT volumetric method. Radiology. 1998;208:807–814. received lecture fees from SHIRE. MR has received lec- 7. McClelland RL, Chung H, Detrano R, et al. Distribution of ture fees from Amgen, Kyowa-Kirin, Sanofi, and Vifor. coronary artery calcium by race, gender, and age: results from the Multi-Ethnic Study of Atherosclerosis (MESA). Cir- MK has served as a consultant to Sanifit, Amgen, Medice, culation. 2006;113:30–37. Sanofi, and Vifor. SS has served as a consultant to 8. Galassi A, Spiegel DM, Bellasi A, et al. Accelerated vascular Sanifit, Vifor Fresenius, and Napp. DAB has served as a calcification and relative hypoparathyroidism in incident consultant to Sanifit, Tricida, Relypsa/Vifor/Fresenius, haemodialysis diabetic patients receiving calcium binders. Sanofi/Genzyme, and Amgen; has stocks or options in Nephrol Dial Transplant. 2006;21:3215–3222. Tricida and Amgen; and currently receives grants from 9. Kooman JP, Kotanko P, Schols AM, et al. Chronic kidney the National Institutes of Health and the Renal Research disease and premature ageing. Nat Rev Nephrol. 2014;10: Institute. RG is a former employee of Sanifit Therapeu- 732–742. tics. JP is an employee of Sanifit Therapeutics; has stocks 10. Chen Z, Qureshi AR, Parini P, et al. Does statins promote or options in Sanifit Therapeutics; and has patents vascular calcification in chronic kidney disease? Eur J Clin Invest. 2017;47:137–148. related to SNF472. AG is an employee of Sanifit Thera- peutics and has stocks or options in Sanifit Therapeutics. 11. Raggi P, Chertow GM, Torres PU, et al. The ADVANCE study: a randomized study to evaluate the effects of cinacalcet plus GMC has served as a consultant to Akebia, AMAG, low-dose vitamin D on vascular calcification in patients on Amgen, Ardelyx, AstraZeneca, Gilead, Reata, Sanifit, and hemodialysis. Nephrol Dial Transplant. 2011;26:1327–1339. Vertex; has stocks or options in Ardelyx, CloudCath, 12. Urena-Torres PA, Floege J, Hawley CM, et al. Protocol Cricket, Durect, Outset, and Physiowave; and has adherence and the progression of cardiovascular calcification received research funding from Amgen and Janssen. in the ADVANCE study. Nephrol Dial Transplant. 2013;28: 146–152. ACKNOWLEDGMENTS 13. Chertow GM, Burke SK, Raggi P. Sevelamer attenuates the This work was supported by Sanifit Therapeutics, S.A. progression of coronary and aortic calcification in hemodi- alysis patients. Kidney Int. 2002;62:245–252. Data shown in this manuscript were presented in part at the ERA-EDTA 2020 Virtual Congress, and at the Kidney 14. Block GA, Spiegel DM, Ehrlich J, et al. Effects of sevelamer and calcium on coronary artery calcification in patients new Week 2020 Virtual Congress. to hemodialysis. Kidney Int. 2005;68:1815–1824. 15. Jamal SA, Vandermeer B, Raggi P, et al. Effect of calcium- REFERENCES based versus non–calcium-based phosphate binders on 1. Nelson AJ, Raggi P, Wolf M, et al. Targeting vascular calcifi- mortality in patients with chronic kidney disease: an updated cation in chronic kidney disease. J Am Coll Cardiol Basic systematic review and meta-analysis. Lancet. 2013;382:1268– Transl Sci. 2020;5:398–412. 1277. 2182 Kidney International Reports (2020) 5, 2178–2182