Discussion
Although a number of previous studies have compared atorvastatin with rosuvastatin in patients with hypercholesterolemia, some did not include high-risk patients, some were local studies conducted in a single country, and many were not powered to study efficacy in terms of LDL-C lowering [
21‐
24]. PULSAR is the first, prospective, large-scale, multinational study designed to compare low doses of rosuvastatin and atorvastatin for their LDL-C-lowering efficacy in high-risk patients. The PULSAR study is part of a wider program investigating the efficacy and safety of rosuvastatin[
25]. The program was designed to address the hypothesis that the statin with the greatest efficacy for improving the atherogenic lipid profile and beneficially modifying inflammatory markers will also slow progression of atherosclerotic plaques, and consequently, result in the greatest reductions in cardiovascular morbidity and mortality[
25]. PULSAR is one of the studies designed to address the first part of the hypothesis, investigating the effects of rosuvastatin on the lipid profile. The results of the PULSAR study demonstrate that rosuvastatin 10 mg was significantly more effective than atorvastatin 20 mg at reducing LDL-C levels in high-risk patients with hypercholesterolemia.
This is consistent with findings from previous studies that have compared rosuvastatin 10 mg and atorvastatin 20 mg in patients with hypercholesterolemia. In three separate studies, one with 2431 patients with hypercholesterolemia (LDL-C ≥ 160 and < 250 mg/dL [4.1 and 6.5 mmol/L]), one with 461 patients (aged 40–80 years) with CHD and low HDL-C, and one with 263 patients with type 2 diabetes, rosuvastatin 10 mg was more effective at reducing LDL-C than atorvastatin 20 mg after 6 weeks of treatment (45.8% vs. 42.6%, 44.0% vs. 38.4%, 45.9% vs. 41.3%, respectively; all p < 0.05)[
11,
26,
27]. Furthermore, in an 8-week study of 3140 high-risk patients with hypercholesterolemia and CHD, atherosclerosis, type 2 diabetes, or a 10-year CHD risk > 20%, rosuvastatin 10 mg was also significantly more efficacious than atorvastatin 20 mg at reducing LDL-C (47.0% vs. 43.7%, p < 0.001)[
12].
Consistent with the greater LDL-C-lowering efficacy in the PULSAR study, more patients treated with rosuvastatin 10 mg achieved recommended 2003 European and NCEP ATP III LDL-C goals than with atorvastatin 20 mg. Furthermore, a greater proportion of patients at highest risk (with established CVD, type 2 diabetes, LDL-C ≥ 6 mmol/L, TC ≥ 8 mmol/L, or blood pressure ≥ 180/110 mmHg) achieved the more stringent European LDL-C goal of < 2.5 mmol/L (100 mg/dL) with rosuvastatin 10 mg than with atorvastatin 20 mg. A previous study of 2829 high-risk patients showed that 52% did not achieve a LDL-C goal of < 2.5 mmol/L (100 mg/dL) with the initial statin dose[
9]. Of the remaining patients, 55% did not have their dosage up-titrated, and of those whose treatment was titrated to a higher dose, only 31% achieved the LDL-C goal. Therefore, selecting a statin that is more efficacious at starting dose will reduce the need for dose titration and improve goal achievement, potentially leading to benefits in CVD-risk reduction.
HDL-C is thought to have a protective role against the development of atherosclerotic plaques[
28] and a low HDL-C level is considered a risk factor for CHD. Agents that improve HDL-C as well as lower LDL-C may offer additional benefits for CHD-risk reduction. In the present study, increases in HDL-C were significantly greater with rosuvastatin 10 mg than with atorvastatin 20 mg.
Several studies have shown that other lipid parameters, such as apolipoproteins and lipid ratios, are better predictors of CVD risk than LDL-C and may be used to guide therapeutic decisions[
29]. For example, in the AMORIS study of 175,553 individuals, ApoB was found to be a stronger predictor than LDL-C for risk of fatal MI[
30]. Furthermore, the ApoB/ApoA-I ratio was found to be the most effective predictor of MI in the INTERHEART study of 15,152 patients with CHD from 52 countries[
31]. In the present study, patients receiving rosuvastatin 10 mg also showed greater improvements in levels of ApoA-I, and ratios of LDL-C/HDL-C, TC/LDL-C, nonHDL-C/HDL-C, and ApoB/ApoA-I, compared with those receiving atorvastatin 20 mg. Reductions in TC, TG, nonHDL-C, and ApoB levels were similar between treatments. Thus, the results of the PULSAR study are consistent with those from previous studies comparing the effects of rosuvastatin 10 mg and atorvastatin 20 mg on lipid parameters. Rosuvastatin 10 mg significantly improved levels of HDL-C, TC, and nonHDL-C and ratios of LDL-C/HDL-C and TC/HDL-C, compared with atorvastatin 20 mg in previous studies with high-risk patients[
12,
26,
27]. Improvements in the atherogenic lipid profile may be beneficial for reducing global risk in patients with CHD[
6]. Furthermore, as part of the wider rosuvastatin clinical trial program, one study has reported that rosuvastatin 40 mg can arrest and even reverse progression of atherosclerosis, in association with reductions in LDL-C[
32]. Ultimately, to assure the clinical relevance of changes in the lipid profile, the greater efficacy of rosuvastatin in terms of LDL-C lowering must translate into reductions in morbidity and mortality. As such, several outcomes studies are now underway to assess the efficacy of rosuvastatin in high-risk patients [
33‐
35].
Results of the PULSAR study showed rosuvastatin 10 mg to be a cost-effective alternative to atorvastatin 20 mg, both in terms of cost per percentage LDL-C reduction and cost per patient achieving their NCEP ATP III or 2003 European LDL-C goal, both in a UK and US setting. These results are in line with several previous cost-effectiveness analyses, which reported rosuvastatin to be more cost-effective than atorvastatin, pravastatin and simvastatin [
36‐
38]. A recent UK study showed simvastatin 40 mg to be a cost-effective alternative to atorvastatin 20 mg, and has stressed the importance of comparing costs when considering statin therapies[
39]. The study reported the benefits of switching to more cost-effective statins in terms of potential savings to the health service[
39]. Further economic analyses of rosuvastatin are now needed to determine its potential as a more cost-effective therapy compared with other statins.
Both rosuvastatin and atorvastatin were well tolerated in this study, and none of the reported AEs were unexpected given the age and underlying medical conditions of the patient population. Most AEs were of mild or moderate severity, and were not considered to be treatment-related. The most commonly reported AE was myalgia, although none of the cases were associated with a clinically important elevation in CK (> 10 × ULN) (or even a CK > 3 × ULN). Furthermore, there were no reports of rhabdomyolysis, renal, or liver insufficiencies during the study. It is possible that the open-label design of the study could potentially have biased reporting of AEs, especially since the study was conducted at a time of high media activity related to rosuvastatin. This is supported by an NLA analysis of AE reporting rates of several statins, which found that the reporting of rosuvastatin-associated rhabdomyolysis and renal failure increased following media publicity[
19]. However, the AE reporting patterns of rosuvastatin did not differ greatly from those of other statins, once differential reporting effects were taken into account.
Results from the PULSAR study are consistent with previous studies, which have assessed the safety of rosuvastatin and atorvastatin in a range of patients with dyslipidemia[
16,
40]. In a global analysis of 12,400 patients in the rosuvastatin phase II/III clinical program, Shepherd et al[
16] found that rosuvastatin 5–40 mg had a similar safety profile to other statins, and demonstrated a favorable benefit-risk profile across this dose range. Nevertheless, there has been some concern regarding the potential toxicity of rosuvastatin, particularly in terms of renal and muscle events [41]. However, the Food and Drug Administration recently conducted a comprehensive review of available safety data from pre-clinical studies, pre-marketing clinical trials, phase IV studies and post-marketing AE reports [42]. The review concluded that rosuvastatin poses no greater risk of muscle toxicity or serious renal injury compared with other statins [42]. Furthermore, a pooled data analysis of 9416 patients in 44 clinical trials found that the safety of atorvastatin 10–80 mg was also similar to that of other statins[
40]. Results from 2 recent large-scale, independent, pharmacoepidemiologic studies conducted in The Netherlands and the USA were in accordance with these findings [
17,
18]. These studies of real-world patient data in over 96,000 patients receiving statin therapy found that the incidence of pre-defined events requiring hospitalization associated with the muscle, liver or kidneys was the same for all currently marketed statins [
17,
18].
Acknowledgements
Author MC is the Principal Investigator of the study and contributed to the concept and design of the study. Author BB contributed to the design of the study. Authors JA, J-PB, HHG, and SM are the investigators who were involved in implementation of the study. Author MP contributed to the planned statistical analyses of the study and sample size determination. Author FS was involved in data analysis and interpretation. All authors read and approved the final manuscript.
This study is sponsored by AstraZeneca. MC is currently on the speaker bureau and a consultant for AstraZeneca, Merck, Pfizer and Schering-Plough, and is involved in research sponsored by AstraZeneca and Hoffman La Roche. JA is a member of several advisory boards for AstraZeneca, has given lectures on their behalf and has participated in clinical trials sponsored by AstraZeneca. The authors FS, MP, and BB are employed by AstraZeneca. FS and BB possess shares in AstraZeneca, and MP and BB have options on AstraZeneca shares. J-PB, HHG and SM declare that they have no competing interests.
We thank Dr Catherine Rycroft, from Prime Medica Ltd., who provided medical writing support on behalf of AstraZeneca.
We gratefully acknowledge the investigators, their co-investigators and study coordinators and the patients who participated in the PULSAR trial. In addition to the authors, the following investigators participated in this study:
Australia: J Karrasch, Kippa Ring, Queensland; R Lehman, Geelong, Victoria; M McKeirnan, Brisbane, Queensland; PA Phillips, Bedford Park, South Australia; D Darnell, Gosford, New South Wales.
Finland: S Lehto, Kuopio; R Korhonen, Savolinnan; L Henttonen, Hameenlinna; TH Suojanen, Seinajoki; J Saltevo, Jyvaskyla; U Venesmaa, Jyvaskyla; HJ Grannas-Honkanen, Vassa; T Jerkkola; J Strand, Oulu; T Pellinen, Helsinki.
France: P Beaufils, Paris; O Dubourg, Boulongne Billancourt Cedex; JF Rosseau, Falaise; V Rohmer, Angers; D Gouet, La Rochelle; D Richter, Jarny; J-L Jacques, Mars-la-tour; G Bertrand, Metz; C Kropka, Marange Silvange; J Dillinger, Mondelange; V Castel, La Fleche; P Sauves, La Bailleul; P Roger, Bazourges/Loir; Y Leroux, St Jean De La Motte; T Le Roy, La Fleche; P Le Pennec, Cholet; D Renou, Cholet; B Favreau, Cholet; T Petella, La Tessoualle; J Tondut, La Rochelle; J Basle, Nieul Sur Mer; C Viel, La Rochelle; B Ledermann, Nimes Cedex; P Guntz, Saint Macaire en Mauges; P Remaud, Angers; A Palomba, Angers; D Marin, Briollay; A Bossu, Bouchemaine; M Frohn, Moutiers; A Bruzzese, Sainte Marie-aux-Chenes; B Dubedout, Gan; Y Leroy, Angers; G Forcada, Vincennes; P Jourde, Hostens; L Lecocq, Merignac.
Netherlands: T Koster, Gouda; JWF Elte, Rotterdam; P Viergever, Den Heder; C Janus, Hoorn; KP Bouter, s-Hertogenbosch; RP Verhoeven, Apeldoorn; S Riemens, Meppel; JGM Tans, Den Helder.
Italy: A Pujia, Cantazaro; C Sirtori, Milan; G Specchia, Milan; R Trevisan, Bergamo; G Ghirlanda, Roma; S Genovese, Milan; A Mezzetti, Chieti Scalo; A Gavazzi, Bergamo; A Zuppiroli, Firenze.
Mexico: CAA Salinas, Mexico City; R Castaneda-Limones, Mexico City; RME Santalo, Guadalajara; L Mejia-Solis, Delagacion Miguel Hidalgo; CP Romero, Mexico City; GF Salmon, Mexico City; GG Galvez, Guadalajara; LAE Sifuentes, Monterey.
USA: HE Bays, Louisville, KY; A Elashker, Kington, NY; SK Elliot, Evansvile, IN; A Farber, Los Angeles, CA; AM Heller, Dan Jose, CA; PR Honig, Philadelphia, PA; AM Jackson, Bartlett, TN; B Krzner, Baltimore, MD; E Kim, Albuquerque, NM; JD Krantzler, Pottstown, PA; AJ Lewin, Los Angeles, CA; JL Newman, Vista, CA; PH Philander, Las Vegas, NV; TL Poling, Wichita, KA; WB Smith, New Orleans, LA; TJ True, Pelham, AL; PJ Winkle, Cypress, CA; N Bittar, Madison, WI; SEE Hafi, Houston, TX; RA Khairi, Indianapolis, IN; KD Klatt, Portland, OR; TW Littlejohn III, Winston-salem, NC; TR Smith, Chesterfield, MO; NR Patel, Beavercreek, OH; JE Sutherland, Waterloo, IA; ST Olatunbosum, Chicago, IL; MM Zimmerer, Zanesville, OH; LI Gilderman, Pembroke Pines, FL; R Bressler, Tucson, AZ; V Greiff, Tucson, AZ; DJ Mikolich, Greenwich, RI; MU Weerasinghe, Rochester, NY; AO Preti, Fort Myers, FL; DJ Morin, Bristol, TN; JP Capo, Atlanta, GA; MA Canossa-Terris, Miami Beach, FL; JA Hoekstra, Richmond, VA; B Lubin, Norfolk, VA; C Chen, Riverside, CA; MC Collins, Alabaster, AL.