November 30, 2011

For Colorectal Cancer It's the Blood Sugar NOT The Insulin Level That Counts

A recently published study based on data from the Womans Health Initiative (WHI) sheds light on a question many of us have wondered about: Is the higher incidence of cancer among people with Type 2 caused by higher insulin levels (or, perhaps injected insulin) or something else?

The study was conducted by a team at The Albert Einstein College of Medicine, which some of you may remember is where Dr. Bernstein earned his M.D.. It has through the years done several low carb studies, too.

The study is summarized here:

Science Daily: High Blood Sugar Levels in Older Women Linked to Colorectal Cancer

The actual abstract is found here:

A longitudinal study of serum insulin and glucose levels in relation to colorectal cancer risk among postmenopausal women.
G C Kabat et al. British Journal of Cancer , (29 November 2011) | doi:10.1038/bjc.2011.512

From the abstract we learn that in a group of 4902 middle aged women who were tracked for 12 years, about 1.6 percent developed colorectal cancer. They had had their fasting blood sugar and fasting insulin measured at the beginning of the study and occasionally through the study. This study found that women whose fasting blood sugar was over 99.5 mg/dl (5.53 mmol/L) at the beginning of the study had a greater risk of developing colorectal cancer than those with blood sugars under 89.5 mg/dl (4.98 mmol/L). The relationship held true for subsequent blood tests too.

However--and this is what is interesting about this study--there was no relationship between these women's fasting insulin levels or the calculated HOMA values (which are believed to measure insulin resistance) and their risk of getting this cancer.

So this data would suggest that it is the high blood sugars, not the high insulin which promotes the cancer.

Before you panic because your fasting blood sugar is over 99.5 mg/dl (as is the case with many of us thanks to dawn phenomenon) remind yourself of this: from what we can see in other contexts, it isn't actually mildly elevated fasting blood sugars that damage our bodies. We see the correlation between fasting sugars and complications in studies because in people eating high carbohydrate diets, mildly elevated fasting blood sugars almost always go hand in hand with significantly elevated post meal sugars. Especially in women, who may be diabetic by glucose tolerance test for up to a decade before they would be diagnosed using a fasting glucose test. (You can see the data backing this up HERE.)

So a woman with a fasting sugar of 105 mg/dl, for example, when she eats her morning bagel with jelly, along with a sweetened latte, may easily reach a blood sugar that approaches 200 mg/dl even if it resolves fairly quickly due to a near-healthy second phase insulin release. But if diabetic you should wake up with a blood sugar of 105 mg/dl and eat an egg and bacon for your breakfast with no carbs, you may easily end up with a blood sugar of 95 an hour later, which suggests you would have a similar health outcome to a completely non-diabetic woman with a fasting blood sugar of 89 mg/dl who ends up at 95 mg/dl an hour after eating her breakfast.

It's also worth noting that the statistical measure the study calculated was "risk" not incidence--and that risk is a statistical artifact that magnifies numbers to provide a more dramatic impact. The actual increase in incididence due to elevated blood sugar was likely around 5 cases per thousand or .5%.

But on the positive side, the fact that even after using magifying statistical techniques like "risk" the researchers couldn't find the expected connection between fasting insulin levels, HOMA, and cancer gives us one more, intriguing, piece of data to answer the quesion, "Do people with diabetes get cancer because of high insulin levels (injected or natural) or because of high blood sugars?" Here at least, it looks like the answer is, as is the case with all the other diabetic complications, "It's the blood sugars, stupid!"

This is very good news because our insulin levels are very hard to control and insulin resistance may be genetic and not something we can lower. Even many devout low carbers with Type 2 continue to be insulin resistant no matter what they weigh (based on how much insulin they have to inject to lower their blood sugars.) But insulin resistant or not, we can control our blood sugars--often most effectively with insulin--and if the conclusion of this study is reinforced by findings in other studies, we may be able to relax about the potential impact on cancer of our injecting insulin to control those blood sugars.

One last note: If you have had years of exposure to higher than normal blood sugars, you might be interested in knowing that Metformin has been shown to suppress the growth of existing early colorectal growths in people without diabetes. Read about that HERE.

Though researchers often state that metformin inhibits cancer due to its effect on lowering insulin, this is just a guess. The belief that it is high insulin levels that cause a greater incidence of cancer among people with diabetes is far from proven and that theory is exactly the belief this study debunks.

In fact, there is evidence emerging that metformin's anti-cancer properties are independent of its effect on insulin and have to do with its ability to suppress TORC1, a cell growth factor. (For example, see THIS STUDY.)

NOTE: Since posting this I have reviewed the full text of the study and see nothing to change the conclusions discussed here. It looks very well conducted, involved people from various well-respected public health departments in universities and medical schools, and gives no hint of corporate meddling.


Karen said...

Thank you Jeannie!! Exactly why tho I think I could get off metformin I have no desire to!! Karen said...

So Kabat's study makes a good case for Metformin?


Is he the same Kabat who has been in the news for claiming that second-hand smoke isn't all that bad?

Jenny said...

The main article doesn't mention metformin. I cited metformin in this context because there has been a lot of other research lately exposing its anti-cancer properties. Some has already been discussed in research roundups on this blog.

As far as Dr. Kabat goes, I wouldn't jump to discredit him based on one sentence in a blog. This study came up with a result that was not at all what the researchers expected. And it wasn't done in people using injectibles, so it isn't likely to be a nefarious plot by the insulin companies to make us feel safe. said...

But Jenny, it's so much fun to see big bad nefarious plots everywhere.

Big Agriculture
Big Banks
Big Cancer-Causing Cell Phones
Big Cereal Companies
Big ChemTrails
Big Fast Food Chains
Big Macs
Big Oil Companies
Big Pharma
Big Vaccines

Peter said...

Thanks for posting this Jenny, a lot of food for thought. Also those of us with dawn phenomenon will agree completely with your clarification of what a FBG means in a given circumstance.

Metformin does appear to be a very interesting drug.


Fiona said...

Jenny, do you have the full study? wondering who funded it...

Jenny said...


I don't, but if someone reading has access to it and can send me a copy, I'd love to have a look at it.

Jenny said...

I have a copy of the full text now. Thanks!

Jenny said...

After having had a chance to read the full text, it looks like a very good study. What distinguishes it from others that have had conflicting outcomes is:

1. Much larger sample size.
2. Carefully screening the population so the study starts out only with people with completely normal fasting blood sugars and insulin. Also they selected a population with largely normal BMI.

3. Taking multiple test readings through 6 years.

4. Controlling for age, smoking, and family history which are known to be strong predictors of this kind of cancer.

The researchers explain that data from smaller studies of overweight people have shown more of a correlation between with insulin levels and cancer, but several came out with very similar findings to this one and the others can be explained by flaws in study design, small sample size, or relying on only one measurement at the beginning of the study.

The study authors are all associated with various universities' epidemiology departments. (Albert Einstein College of Medicine;
Department of Medicine, University of Alabama at Birmingham; Department of Preventive Medicine, Stony Brook
University Medical Center; Department of Community Medicine, School of Medicine, West Virginia University; Department of Epidemiology and Community Health, School of Medicine, Virginia Commonwealth University Medical
Center; Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London)

There is nothing here to suggest that the data were slanted by any commercial interest and no conflicts of interest are listed at the end of the article.

I think we can indeed take this as good news.

Andrea Karim said...

THANK YOU for pointing out that high sugars can occur, and then resolve themselves before the 2-hour mark. It's SO HARD to get "traditional" health educators to understand this.

Rad Warrier said...

AD said...
THANK YOU for pointing out that high sugars can occur, and then resolve themselves before the 2-hour mark. It's SO HARD to get "traditional" health educators to understand this.

The blood sugars going somewhat high and quickly getting back to normal levels before the 2-hour post prandial can happen to non-diabetics too. Spikes are not good, but I am not unduly worried about narrow spikes because it appears that it is the area under the curve that matters more regarding damages that high blood sugars can do. In other words, the time duration of the spike matters. If the time is short enough, harm done is insignificant.


Jeffreagan2001 said...

Just because the fasting insulin levels were not elevated does not mean that insulin is not the culprit. HOMA-IR is essentially the fasting glucose times the fasting insulin. If one's blood sugar is below 100 mg/dL and the fasting insulin is normal, the HOMA-IR will suggest the patient is not insulin resistant. However, as Gerald Reaven has shown time and time again in his research, a person with normal fasting glucose and fasting insulin can still be insulin resistant! It is the 3 hr insulin area under the curve that is the best measure of insulin resistance (of course, the euglycemic-hyperinsulinemic clamp or modified insulin suppression test are the gold standards for measurement of insulin resistance). My point is that it is the underlying insulin resistance and elevated levels of insulin in response to a high carb meal that is the problem. And since insulin is mitogenic, it is the most likely culprit in colon cancer as well as breast cancer. Remember, HOMA-IR is not a particularly robust measure of insulin resistance. Better to look at the fasting triglyceride/HDL ratio. If it is greater than about 3 or so, you are probably insulin resistant.
Jeff Reagan

Jenny said...

The 3 hr glucose is not necessarily a measure of insulin resistance UNLESS the insulin is also elevated. But in many people with Type 2 diabetes, by the time the blood sugars are high, the insulin is NOT elevated because most people with Type 2 have defects that limit insulin secretion, too.

High triglycerides reflect high blood sugars, not necessarily IR. There is all too much speculation re the role of insulin and insulin resistance in humans (as opposed to genetically engineered rodent "models" whose diabetes and IR proceed from genetic defects not commonly found in humans) and far too little objective high quality research.

v/vmary said...

"If you are using a basal insulin (Lantus or Levemir) there is a temptation to use a lot of it to try to lower post meal blood sugars. Your sugars will be pretty good as long as you keep eating carbs, but once you stop eating at night, the insulin keeps being absorbed into your blood stream and by 3 AM it is quite common to hypo.

However, many of us as soon as we hypo slightly, experience a burst of counterregulatory hormone intended to push the blood sugar back up to normal so by the time you wake your blood sugar is high, not low. You can tell if you have had a counterregulatory burst because it will often wake you up at 3 AM with a jolt. Your pulse will be fast and your blood pressure may be elevated too. If you measure your blood sugar it won't be low, because as soon as you get that jolt, your blood sugar will surge. That's what the jolt is meant to do."

I have been getting these jolts with fast heart beat when i try to eat low carb ala bernstein. i decided to test my blood sugar when i woke up at 2:30 and it was 82. another time it was 87 in the middle of the night. but then in the morning it was 96-98. Last night I ate higher carb, not sure how much higher, and i did not wake up in the middle of the night. and my blood sugar was still 99!! do you think if i stuck out the 2:30 wake ups, my body will get used to the lower (82-87) blood sugar in the middle of the night and stop jolting me awake as if i were hypo?

Jenny said...

If you were only in the 80s when you tested, you were probably a lot lower earlier. If you keep going low you can lose hypo awareness and have very dangerous lows. The 90s are fine. Don't risk those early morning hypos.

v/vmary said...

the thing is i don't take any diabetes medication, so it is going to be impossible for me to lose hypo awareness. i quoted you above using the insulin example because i thought perhaps a similar mechanism is going on with me with naturally. my body senses a low blood sugar level it is not used to sensing in the middle of the night (i'm speculating brought on by more lower carb eating throughout the previous day and especially in the evening), and so my liver dumps glucose into my blood to bring it to the 95-99 mark i have been getting regularly
with a higher carb diet. just as lower but better numbers in the daytime that your body isn't used to can cause you to feel hypo, perhaps the same thing happens at night. so just as you advise sticking out that uncomfortable feeling of lower but healthier blood sugars in the daytime until the body adjust, perhaps the same applies at night. my body with adjust to lower but healthier numbers in the middle of the night and stop interpreting these numbers as a crisis. i've also read other low carb ers post that when they go lower carb, they wake up in the middle of the night. or maybe my body is not used to the lower calories and thinks i'm going into starvation, so it revs up the stress hormones???

Jenny said...

If you are not taking insulin or any of the drugs that cause insulin to be secreted, then waiting it out is fine.

However, this post isn't about anything related to your question. Please let's save the comments section for comments on the posts. Thanks!