#42

JUPITER did not show a significant difference in CV events, it was statistically non significant (at least MI, stroke and death from CV). It was stopped early (probably due to adverse effects of rosvustatin on glucose metabolism and new onset diabetes in treated group), the ACM at the time of trial stopping was starting to converge (maybe diabetes is more dangerous than LDL-C?)… and I would not take JUPITER as a proof of anything other than intensive rosvustatin therapy in primary prevention can cause diabetes.

#43

The JUPITER trial findings were briefly summarized in the following article from the Journal of Clinical Lipidology in 2013. The title of the article is,
Point: Why statins have failed to reduce mortality in just about anybody PDF

The authors write, "In JUPITER (Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin trial), a trial involving 17,802 participants randomized to rosuvastatin or placebo, investigators reported a highly significant reduction of nonfatal myocardial infarction and stroke; however, the only statistically significant benefit from the statin for women was from fewer revascularizations, whereas for all participants the cardiovascular mortality was not reduced.

The authors also criticize the use of Meta-Analyses, remarking that they provide “metrics of limited value to patients most interested in whether the use of statins will reduce their absolute risk of dying prematurely.”

1 Like
#44

The JUPITER trial lasted less than two years and was stopped early. Read below as to why.

The trial was stopped early, after just 1.9 years median duration, by the study’s Independent Data Monitoring Board, because the interim results met the study’s predefined stopping criteria (it had been predetermined that it would be unethical to continue the study once it became clear that the patients in one arm of the study had a significantly higher cardiovascular risk than the other arm’s patients).[1][8]

3 Likes
#45

Given the exponential growth in the number of studies published per year we shouldn’t rely (or worse, post…) anything older than 2020.

2 Likes
#46

It did.

In this trial of apparently healthy persons without hyperlipidemia but with elevated high-sensitivity C-reactive protein levels, rosuvastatin significantly reduced the incidence of major cardiovascular events.

Rosuvastatin to Prevent Vascular Events in Men and Women with Elevated C-Reactive Protein | NEJM

(at least MI, stroke and death from CV).

Wrong.

The rates of the primary end point were 0.77 and 1.36 per 100 person-years of follow-up in the rosuvastatin and placebo groups, respectively (hazard ratio for rosuvastatin, 0.56; 95% confidence interval [CI], 0.46 to 0.69; P<0.00001), with corresponding rates of 0.17 and 0.37 for myocardial infarction (hazard ratio, 0.46; 95% CI, 0.30 to 0.70; P=0.0002), 0.18 and 0.34 for stroke (hazard ratio, 0.52; 95% CI, 0.34 to 0.79; P=0.002), 0.41 and 0.77 for revascularization or unstable angina (hazard ratio, 0.53; 95% CI, 0.40 to 0.70; P<0.00001), 0.45 and 0.85 for the combined end point of myocardial infarction, stroke, or death from cardiovascular causes (hazard ratio, 0.53; 95% CI, 0.40 to 0.69; P<0.00001), and 1.00 and 1.25 for death from any cause (hazard ratio, 0.80; 95% CI, 0.67 to 0.97; P=0.02).

It was stopped early (probably due to adverse effects of rosvustatin on glucose metabolism and new onset diabetes in treated group),

It was stopped because it would be unethical to keep the patients on placebo as pointed out by @DeStrider.

the ACM at the time of trial stopping was starting to converge (maybe diabetes is more dangerous than LDL-C?)

If by “convering” you mean a statistically significant 20% reduction…

Jupiter-Front
Jupiter-Front1427×803 128 KB

While yes there was an uptick in the acm for the rosuvastatin group in year 5, so was there for the placebo group. That results in the 20% difference we see.

and I would not take JUPITER as a proof of anything other than intensive rosvustatin therapy in primary prevention can cause diabetes.

A 20% reduction in all-cause mortality is nothing but a 0.6 point increase in diabetes in the rosuvastatin group over the placebo group warrants caution?

Nevertheless, physician-reported diabetes was more frequent in the rosuvastatin group (270 reports of diabetes, vs. 216 in the placebo group; P=0.01); these events were not adjudicated by the end-point committee.

3 Likes
#47

But that is outright wrong.

The rates of the primary end point were 0.77 and 1.36 per 100 person-years of follow-up in the rosuvastatin and placebo groups, respectively (hazard ratio for rosuvastatin, 0.56; 95% confidence interval [CI], 0.46 to 0.69; P<0.00001), with corresponding rates of 0.17 and 0.37 for myocardial infarction (hazard ratio, 0.46; 95% CI, 0.30 to 0.70; P=0.0002), 0.18 and 0.34 for stroke (hazard ratio, 0.52; 95% CI, 0.34 to 0.79; P=0.002), 0.41 and 0.77 for revascularization or unstable angina (hazard ratio, 0.53; 95% CI, 0.40 to 0.70; P<0.00001), 0.45 and 0.85 for the combined end point of myocardial infarction, stroke, or death from cardiovascular causes (hazard ratio, 0.53; 95% CI, 0.40 to 0.69; P<0.00001), and 1.00 and 1.25 for death from any cause (hazard ratio, 0.80; 95% CI, 0.67 to 0.97; P=0.02).

Rosuvastatin to Prevent Vascular Events in Men and Women with Elevated C-Reactive Protein | NEJM

#48

It sure did after carefully combed (read manipulated) data analysis.

sample size was about 17+k
CV mortality

placebo: 37 events
MI 6 events
Stroke 6 events
other CV death 25 events

rosvustatin: 31 events
MI 9 events
Stroke 3 events
other CV death 19 events

Do you think above data on CV mortality really justifies stopping this trial?

Screenshot 2024-02-24 at 16.01.54

Screenshot 2024-02-24 at 16.00.36
Screenshot 2024-02-24 at 16.00.361172×890 58.5 KB

But I just don’t want to discuss this any longer, since I just wanted to point out that you can not use JUPITER for what you are claiming.
I am in no way LDL-C casual role denier, just that I have reservation in primary prevention strategies with statins since the data from healthy group is just not there. Yes, you can speculate and decide upon this and other other studies that it makes sense for yourself but you can not claim from JUPITER that primary prevention with statins reduces CV mortality.

#49

The baseline median TC for the participants in the Jupiter trial was 186 Rosuvastatin and 185 placebo. Interquartile range 168–200 Rosuvastatin, and 169–199 placebo. See page 13 of the Jupiter Trial.

https://www.nejm.org/doi/pdf/10.1056/nejmoa0807646

That is considered too low by the standards of the Hunt Study in Norway (2013).

The aim of the present study was to document the strength and validity of total serum cholesterol as a risk factor for mortality, as defined by current CVD prevention guidelines. For this purpose, we used data from a well-defined, general Norwegian population without known CVD at baseline. We focused on deaths from cardiovascular disease, IHD and death from all causes (total mortality) within a follow-up period of 10 years.

Our calculations are thereby based on information from 52 087 individuals (24 235 men and 27 852 women) aged 20–74 years and free from known CVD at baseline.

Wheelock’s article I posted above summarizes the Hunt Study table below:

Table 1. Variation in mortality due to all causes and to cardiovascular diseases (CVD) with TC in men and women

TC, Mmol/L (mg/100ml) MEN WOMEN
All-cause CVD All-cause CVD
<5.0 (<194) 1 1 1 1
5.0-5.9 (194-228) 0.77 0.8 0.92 0.9
6.0-6.9 (232-267) 0.84 0.87 0.84 0.81
>7.0 (>270) 0.98 1.05 0.72 0.74

There are more deaths at the <194 level than all the higher levels, whether all-cause or CVD.

Below is the table stratified by age:

MEN
Age ranges TC LEVELS, mmol/L(mg/100ml)
<5.0(<194) 5.0-5.9(194-228) 6.0-6.9(232-267) >7.0(>270)
20-29 1.1 0.38 0.3 0
30-39 0.8 0.57 0.72 0.47
40-49 2.22 1.38 2.27 3.37
50-59 4.54 4.93 6.22 5.74
60-69 20.31 16.2 17.37 18.47
70-74 49.18 40.37 37.93 41.25
WOMEN
20-29 0.35 0.3 0.24 0.6
30-39 0.31 0.43 0.82 0.69
40-49 0.89 1.85 1.69 1.12
50-59 2.95 3.59 3.53 3.79
60-69 22.31 10.32 10.47 9.51
69-74 31.46 22.5 21.58 19.23

If you read horizontally, you can see that low TC is the worst performer.

#50

Where did you even get those numbers from? Here is a table of the actual number of events. Clearly the rosuvastatin group did better all around.

nejmoa0807646_t3
nejmoa0807646_t3985×556 90.2 KB

Do you think above data on CV mortality really justifies stopping this trial?

The actual data sure does.

Converge would mean that two lines eventually intersect but as we can see, by year 5 there was still a 20% gap between the placebo and the rosuvastatin group. Look at the uptick of deaths in the placebo group at the last point of the graph.

#51

From table you shared in your post above. And from source in wikipedia page @DeStrider shared.
You know it is very skillfully curated data to show how rosvustain is better.

#52

My table had 83 deaths from any cardiovascular causes in the rosuvastatin group and 157 deaths from any cardiovascular causes in the placebo group so your claim of 31 and 37 events is simply wrong.
For all-cause mortality it’s 198 deaths in the rosuvastatin group and 247 in the placebo group so rosuvastatin is clearly also better in that regard.

#53

Just read carefully what it reads. Deaths are carefully hidden.

#54

“Myocardial infarction, stroke, or confirmed death from cardiovascular causes”.
I don’t know what you are getting at. Besides, all the other figures look similarly promising for rosuvastatin, even all-cause mortality.

#55

83-33-31=19 (just some simple algebra)

#56

Rosuvastatin vs Placebo
83 deaths vs 157 deaths from any cardiovascular cause
198 deaths vs 247 death from any cause

33 strokes vs 64 strokes either fatal or non-fatal
30 non-fatal strokes vs 58 non-fatal strokes
3 fatal strokes vs 6 fatal strokes

31 myocardial infarctions vs 68 myocardial infarctions either fatal or non-fatal
22 non-fatal myocardial infarctions vs 62 non-fatal myocardial infarctions
9 fatal myocardial infarctions vs 6 myocardial infarctions

71 deaths vs 145 deaths from other confirmed cardiovascular causes

Are you trying to suggest that they lied about deaths from other cardiovascular deaths? It’s not like strokes and myocardial infarctions make up 85% of cardiovascular deaths in the US like they do globally. Especially not in this particular group of people without a history of cardiovascular disease.

In 2021 in the United States, coronary heart disease (CHD) was the leading cause of deaths (40.3%) attributable to CVD in the United States, followed by stroke (17.5%), other CVD (17.1%), high blood pressure (13.4%), heart failure (9.1%), diseases of the arteries (2.6%).

https://www.heart.org/-/media/PHD-Files-2/Science-News/2/2024-Heart-and-Stroke-Stat-Update/2024-Statistics-At-A-Glance-final_2024.pdf

Also, the death from any cause is still 198 in the rosuvastatin group compared to 247 in the placebo group.

1 Like
#57

Just read. I am not implying they lied, I am claiming they presented data in a way that makes it difficult to read objectively.

It is not 83 deaths as you read it, but 19 and not 157 but 25!

As I said it is difficult to objectively evaluate this numbers as the trial was ended prematurely. We don’t know how much would the lines converge and there is also @AnUser’s argument that ACM should not be the the goal in preventative therapies. And still there is less statistical significance in this number than new onset diabetes number in the treatment group vs. placebo.

1 Like
#58

Assuming that’s the right interpretation, in that case it would be
19 deaths vs 25 deaths from other cardiovascular causes
3 deaths vs 6 deaths from strokes
9 deaths vs 6 deaths from myocardial infarctions
Overall 31 vs 37 deaths from strokes, myocardial infarctions or other cardiovascular causes.
That begs the question: where would the other 167 and 210 deaths come from? The rosuvastatin group even had less deaths from cancer (35 vs 58).

We don’t know how much would the lines converge

Or diverge as the trial was only going on for a median of 2 years.

And still there is less statistical significance in this number than new onset diabetes number in the treatment group vs. placebo.

nejmoa0807646_t4
nejmoa0807646_t4847×995 143 KB

The p value is 0.01 for new diabetes compared to 0.02 for reduction in all-cause mortality. Both are statistically significant and I could make the same argument that the numbers would converge if the trial had kept going. After all, the difference in average HbA1c was only 0.1 points.

1 Like
#59

Screen Shot 2024-02-24 at 10.59.45 AM
Screen Shot 2024-02-24 at 10.59.45 AM1056×338 42.6 KB

GHHLW8VWQAAA7pB
GHHLW8VWQAAA7pB1720×565 56.9 KB

Our findings provide genetic evidence suggesting no threshold of lowering apoB or, equivalently, LDL-C (ie, the main apoB-containing lipoprotein) for reducing risk of CAD, all-cause mortality, and CVD mortality, further supporting the concept of the lower the better. The retention of apoB-containing lipoproteins in the artery wall and subsequent release of cholesterol contents are essential for the initiation and progression of atherosclerosis.46 Our findings are consistent with RCTs8-10 showing that intensive LDL-C lowering further reduces CVD events compared with standard LDL-C lowering. However, most RCTs9-13 failed to show significant incremental benefits of intensive LDL-C lowering on all-cause mortality or CVD mortality. This discrepancy may be related to insufficient difference in LDL-C between groups, short follow-up duration, and low proportion of death from CVD in RCTs,9-13 or differing effects on mortality between LDL-C lowering therapies.47 Notably, a trial48 has shown high-dose statin therapy further reduces all-cause mortality compared with low-dose therapy in Japanese patients with the mean LDL-C less than 90 mg/dL (to convert to mmol/L, multiply by 0.0259) at baseline. Our findings are consistent with a recent MR study49 showing no threshold in the association of LDL-C with CAD, although that study used a different approach for instrument selection and lacked power to detect a significant association of LDL-C with all-cause mortality.

Paper: Dose-Response Associations of Lipids With CAD and Mortality

3 Likes
#60

To me this looks like age is a much larger correlation to CVD events than TC. Again, not disputing LDL’s role in ASCVD, but i wonder if this table was redone using “biological age” versus “chronological age” would it look different? — ie would having a “younger” physiology reduce your risk for CVD events more than lowering LDL levels to “normal”? I’m not exactly sure what specific physiology would do this: lower inflammation? More glutathione? Better glucose control / metabolic syndrome? More musculature (core, not vanity)? Obviously if we used CAC score as our measure of “youth” then this table would be highly correlated to CAC scores.

Over time, ASCVD will eventually get all of us if we don’t die of something else. Its accumulated over time so risks invariably grow over tome (and with age). But is there something else we could be doing — other than and separate from lowering LDL — that could have a significant impact on ASCVD?

1 Like
#61

That’s not my argument at all.
My argument if there was no statistical significance in ACM that doesn’t mean anything as it is hard to detect.

In this case there was, 20% decrease.
Solely looking at cardiovascular death doesn’t mean much. All of the end points were decreased for rosuvastatin. It’s a wonder drug.

Extrapolating from the graph and trying to predict where the lines are going is simply bad science. It’s like people looking at stock price and trying to predict where it goes. You know only what the graph showed.

2 Likes