November 15th, 2011
SATURN Regression Trial: Gorilla Statin and Statin King Battle to a Draw
Larry Husten, PHD
An epic battle comparing the two most potent statins — the reigning king atorvastatin versus “gorilla statin” rosuvastatin — has ended with a quiet draw. Results of SATURN (Study of Coronary Atheroma by InTravascular Ultrasound: Effect of Rosuvastatin Versus AtorvastatiN) were presented at the AHA on Tuesday and published simultaneously in the New England Journal of Medicine.
Stephen Nicholls and colleagues randomized 1039 patients with coronary disease to rosuvastatin or atorvastatin for 2 years and measured the progression of atherosclerosis using intravascular ultrasound (IVUS). As anticipated, after 2 years, lipid parameters were more favorable with rosuvastatin than with atorvastatin:
- LDL: 62.6 mg/dL with rosuvastatin versus 70.2 mg/dL with atorvastatin (p<0.001)
- HDL: 50.4 versus 48.6 mg/dL (p=0.01)
However, there was no significant difference in the percent atheroma volume (PAV), the primary efficacy endpoint, which decreased by 0.99% with atorvastatin and 1.22% with rosuvastatin (p=0.17). A secondary endpoint, normalized total atheroma volume (TAV), significantly favored rosuvastatin (-6.39 mm(3) versus -4.42 mm(3), p=0.01). Regression of atherosclerosis, as measured by both PAV and TAV, occurred in a majority of patients in both treatment groups.
Although the trial failed to find a difference between the agents, “it does show that high-dose, intensive statin therapy can be administered safely and can promote regression of atherosclerotic plaque to a greater extent than has previously been reported,” the authors conclude.
I agree that this was effectively a negative study, despite the emphasis during the presentation on the modest superiority of rosuvastatin on serum cholesterol values and total atheroma volume. I think more such comparative effectiveness studies are very helpful and critical in providing cost-effective clinical care. This study will come into particular use when lipitor becomes generically available and price starts to drop. (May take some time, however http://tinyurl.com/6nb8afn)
Given that evidence supports continued cardioprotection with higher dose statins, even at very low LDL levels, it makes sense to place patients with established CAD on maximum dose as long as they can be tolerated and afforded. That said, a big interest of mine with regards to this trial was the side effect profile listed on table 4 of the table. I was happy to see the general rates to be very low of elevated CPK and liver tests. Is there any reason to not max out all of our CAD patients on statins as long as they can tolerate? Especially if/when Lipitor becomes a $4 generic?
Dr. Labarthe, who performed the review after the study, commented on the high drop-out rate (>20%), and how this is further reason to consider the results as null. He pointed out the importance of knowing how the drop outs were different in characteristics from those that remained in the trial, and reasons as to why they refused to get an IVUS a second time. (Given the non-trivial risks of IVUS, I don’t blame them for dropping out myself!) I thought about this for a minute: one assumption is that the dropout is non-differential, and not biased towards one drug, thereby preserving randomization. However, it is hypothetically possible that one of the statins may preferentially drive decision-making based on its effects on patient symptoms etc (whether through CVD benefits on angina or side-effects of the drugs). Therefore, I agree that this point is worth special consideration.
Overall, a very interesting study with clinical implications.