New analysis uncovers relationship between low LDL-cholesterol levels
and cancer

July 23, 2007  Michael O'Riordan

Boston, MA - Results from a new analysis, initially designed to
determine whether there was a correlation between the extent to which
statins lowered LDL-cholesterol levels and liver and muscle toxicity,
suggests that the cardiovascular benefits of achieved levels of LDL
cholesterol might be offset by an increased risk of cancer [1]. In an
analysis of patients enrolled in large, randomized statin trials,
investigators observed a "significant and linear relationship" between
achieved LDL levels and the risk of new cancer cases.

"The statin trials have clearly shown that statin therapy, overall,
reduces cardiovascular risk," said senior investigator Dr Richard
Karas (Tufts-New England Medical Center, Boston, MA). "These findings
don't change that. They're based on the same studies. But a component
of that, perhaps one of the costs of that, is a relationship between
the LDL lowering and cancer risk."

Speaking with heartwire, Karas said there is concern about how the new
findings will be reported and interpreted, especially if the message
causes some patients to stop taking statins. "What we're always doing
in terms of trying to take care of patients is balance benefit and
risk," he said. "This analysis was really focused on trying to enhance
our understanding of the risk side of that equation. It has produced a
provocative and interesting result that raises a lot of new
questions . . . but it's a complicated message, and the conclusion
people will jump to if they are not being careful is that statins
cause cancer. We don't know that, and our data don't show that."

In an editorial accompanying the published study, which appears in the
July 31, 2007 issue of the Journal of the American College of
Cardiology, Dr John LaRosa (State University of New York Downstate
Medical Center, Brooklyn) observes that the benefits of all drugs come
with a price [2]. While these new data do not provide definitive
answers to the question of cancer risk associated with lowering
cholesterol levels, he said the finding highlights an issue that has
been repeatedly raised: does the process of lowering LDL, particularly
to very low levels, introduce hazards of its own in either causing or
accelerating the progress of cancer?

LaRosa, who is the chair of the steering committee of the Pfizer-
sponsored Treating to New Targets (TNT) study, said the findings
should be viewed only as hypothesis generating. While clinicians
should continue to be vigilant in ensuring the benefits of statins
outweigh the risks, this vigilance should not deny the benefits of LDL
lowering to those who need it, he said, a sentiment echoed by the
study authors. "The bottom line is that at the current time clinical
practice should not change," said Karas.

Initially only looked at muscle and liver toxicity
Speaking with heartwire, Karas said the cancer association was a
surprise and initially wasn't even on his group's radar. The Boston
researchers conducted their analysis examining the relationship of the
degree of LDL-cholesterol lowering to liver toxicity and
rhabdomyolysis in 23 randomized, controlled trials assessing statin
therapy, including, among others, the Scandinavian Simvastatin
Survival Study (4S), West of Scotland Coronary Prevention Study
(WOSCOPS), Long-term Intervention with Pravastatin in Ischemic Disease
(LIPID) study, the Heart Protection Study (HPS), and the more recent
trials of intensive vs moderate lipid-lowering therapy such as PROVE-
IT, TNT, and IDEAL.

Investigators also studied the effect of drug dosage on liver and
muscle toxicity. In the first analysis, no matter how investigators
looked at the extent of LDL reduction, as a relative or absolute
reduction or achieved LDL-cholesterol levels, they observed no
relationship between how much the cholesterol was lowered and the risk
of liver or muscle toxicity. In the second analysis, however, when
they looked at muscle toxicity on the basis of the dose of the drugs,
they found that the higher the dose of statins used in the study, the
higher the risk of toxicity to the liver.

"Overall, statins are very safe, but the safety implication here is
that it does matter how you lower LDL-cholesterol levels," said Karas.
"In other words, the high-dose statins are associated with a higher
risk of side effects. This does then have implications for how we
practice medicine, the question being, and it's just a question, as to
whether or not we might be better off using multiple medications, all
at modest doses, to try to get the cholesterol targets that we want to
get to, to minimize side effects and maximize the benefit."

Cancer findings
Considering these findings worthy of publication, the group submitted
the manuscript to the Journal of the American College of Cardiology,
but, as noted in an editorial comment by journal editors Dr Anthony
DeMaria and Ori Ben-Yehuda [3], the researchers were asked to include
cancer in the analysis because "this was the other major side effect
often feared from statin therapy." Of the 23 statin therapy trials, 13
studies included the number of patients with newly diagnosed cancer.
Overall, there was no significant relationship between percent and
absolute reductions in LDL-cholesterol levels. There was, however, a
highly significant inverse relationship between achieved LDL-
cholesterol levels and rates of newly diagnosed cancer (R2=0.43,
p=0.009). The researchers found one additional incident of cancer per
1000 patients with low LDL levels when compared with patients with
higher LDL levels. The new cancers were not of any specific type or
location.

To heartwire, Karas said the findings are paradoxical in light of
recent meta-analyses concluding there is no significant increase in
the risk of cancer with statin therapy. Karas stressed, however, that
the new findings are observational, hypothesis generating, and in no
way definitive. "This is an association at this point," he said. "It
might have nothing to do with cause and effect. The best analogy is to
say that I have a dog, and every time an airplane goes over my house,
my dog goes out into the backyard and barks at the plane. That
airplane has never landed in my yard. Now we could say there is a very
strong association between my dog barking and planes not landing in my
yard, but there certainly is no cause and effect. "Even if the risk of
cancer is increased with statin use, Karas said clinicians would have
to balance the magnitude of that risk with the benefits of statin
therapy.

Dr Thomas Pearson (University of Rochester School of Medicine, NY),
who was not affiliated with the study, told heartwire that the results
should be interpreted carefully.

"In many ways it is d?j? vu all over again," said Pearson. "In the
1970s, there were several papers describing higher risks from cancer
in those with the lowest cholesterol levels. This is from the
prestatin era. Old-time clinicians will tell you that monitoring
cholesterol levels is useful, such that when they start to fall, you
should look for a cancer. Indeed, the MRFIT study looked at this and
showed the shorter the interval between the cancer diagnosis and the
blood test, the lower the cholesterol."

For this reason, said Pearson, one should assume the very low
cholesterol levels in patients with cancer on statins are due to a
cancer-low cholesterol link rather than a statin-cancer association.

In his editorial, LaRosa points out that no single form of cancer
predominates, "so the effect of low achieved LDL would have to have
been one that stimulates neoplasia in a variety of tissues." In
addition, the effect of low LDL-cholesterol levels would have to be
unusually rapid, given that most statin trials lasted five years or
less, in producing new cancers.

Most important, LaRosa said it is possible to further address the
cancer risk without further trials. Karas and colleagues had access to
just 13 studies reporting the number and type of incident cancers.
Other data sets have not been published, but an analysis of other LDL-
lowering trials, including statin studies, other drug studies, diet,
or even ileal-bypass surgery trials, supported by a neutral source
such as the National Institutes of Health, could determine whether
there is a significant problem with low LDL-cholesterol levels that
needs to be addressed, said LaRosa.

Study generated scrutiny and discussion
In their editorial comment, DeMaria and Ben-Yehuda write that in the
five years they have been editors at Journal of the American College
of Cardiology, no other manuscript stimulated such intense scrutiny
and discussion.

"Given the growing public angst regarding the safety of prescription
medications, all were concerned that the paper contained great
potential for harm and good," they write. "In the end, we agreed to
publish the article with as much caution and perspective as possible."

The editorialists point out the manuscript has limitations, as noted
by Karas and colleagues, including the retrospective use of summary
data, use of data from clinical trials rather than real-world
experience, and the lack of standardization of adverse events in the
individual trials. Given the limitations, and the potential that
physicians, patients, and the media might misinterpret the results,
several associate editors argued that the paper not be published.

"However, in the final analysis, the consensus was that these findings
could not be ignored, that they did indeed warrant further
investigation, and that they should be aired in public," write DeMaria
and Ben-Yehuda.

Karas reports that he has received research grants from AstraZeneca
and Kos Pharmaceuticals; has served on the speakers' bureaus and/or
received honoraria from Kos, AstraZeneca, Merck, and Pfizer; and has
served as a consultant to Kos. LaRosa reports honoraria for speaking
or consulting from Pfizer, Bristol-Myers Squibb, and Merck.

Sources

Alsheikh-Ali AA, Maddukuri PV, Han H, Karas RH. Effect of magnitude of
lipid lowering on risk of elevated liver enzymes, rhabdomyolysis, and
cancer. J Am Coll Cardiol 2007; 50:409-418. DOI: 10.1016/j.jacc.
2007.02.073. Available at: http://content.onlinejacc.org.

LaRosa JC. Means and ends of statins and low-density lipoprotein
cholesterol lowering. J Am Coll Cardiol 2007; 50:419-420. DOI: 10.1016/
j.jacc.2007.04.044. Available at: http://content.onlinejacc.org.

DeMaria AN, Ben-Yehuda O. Low-density lipoprotein reduction and
cancer. J Am Coll Cardiol 2007; 50:421-422. DOI: 10.1016/j.jacc.
2007.06.003. Available at: http://content.onlinejacc.org.

Marilyn