Friday, May 11, 2012

Could Anti-diabetic drugs increase the risk of pancreatic cancer?



Intro

While exploring the complex link between diabetes type II and pancreatic cancer, I have taken a so called “chicken and the egg” approach to my thoughts, what causes what. While diabetes type II has been linked as a consequence of pancreatic cancer in the near past, new studies look to diabetes and ant diabetes medications as a player in the role of pancreatic carcinogenesis.


Methods

The study looked at 973 patients with pancreatic adenocarcinoma over a four-year study starting in 2004. In the group of patients with pancreatic cancer, 259 were known to have diabetes. This group was compared to a group of 863 control patients without pancreatic cancer, of which, 109 were known diabetics. The patients’ diabetes oral medication use (**this does not include insulin) was surveyed and broken into four research groups: use of insulin secetagogues, use of thiazolidindiones (TZD), Metformin, and other anti-diabetes medication. Subsequently, pancreatic cancer risk was then “estimated using unconditional logistic regression analysis”. The participants without cancer self reported their diabetic medical history. This included diagnosis time and medications used and the duration of their use. The duration of diabetes (until time of study) was broken into three categories: less than two years, 2 to 5 years and greater than 5 years.

Note: age, sex, race, use of tobacco and alcohol products, BMI, family history, duration of diabetes, and insulin use were all statistically factored into the study.

Results

Non-diabetics had an adjusted odds ratio (AOR) of developing pancreatic cancer of 1.0. Patients who did not take oral medication had a 0.56 AOR. Those who took insulin secreatgogues had an increased AOR of developing pancreatic cancer at 1.78. Those who took thiazolidindiones had a similar risk of developing pancreatic cancer to non-diabetics with an AOR of 1.08. Interestingly, those who used Metformin, a medication that helps control the amount of glucose in the blood, had a lowered risk for developing pancreatic cancer with an AOR of 0.38. The same is true when metformin is combined with insulin secreatgogues and thiazolidindiones with AORs of 0.64 and 0.39.











Mechanism of Metformin. Why Metformin?

Metformin has direct antineoplastic (stops new/abnormal cell growth) effects. Metformin activates AMP activated protein kinase (AMPK) that sways the AMP to ATP ratio favoring AMP. Additionally, AMPK inhibits the mTOR pathway; this in turn down regulates cell proliferation. AMPK also is involved with cell polarity and cell division.  Metformin has direct effects within the signal transduction cascade (post receptor) that disrupt the PI3K/Akt/mTOR signaling pathway that regulates cell division. This stops the cancer hallmark of increased cell proliferation.

Implications, application and the future

My partner and I have been focusing intensely on the mechanism by which pancreatic cancer gives rise to diabetes type II and possibly using diabetes as an early screening biomarker for the development of pancreatic cancer. I examined this article as a way to see the full picture of pancreatic cancer. While I was mentally arguing about if pancreatic cancer gave rise to diabetes or vice versa I didn’t even consider the myriad of factors that come with having diabetes, the biggest being anti-diabetes medicine. With Metformin being so effective at regulating cell proliferation I look to Metformin or a Metformin derivative as a potential hinderance against cancer’s sneaky ways.

Thoughts and critics

…this study was impressive. To me the most impressive aspects of this study was the sample size and the participants. The type of patient controls was also, in my opinion, done impeccably well. Cases and controls were “matched by age (± 5 years), sex, and race”; this ensured that there is not genetic bias based on age, sex, or race. Additionally, each of these factors was normalized out using stats test with a p value of less than 0.05. My one critic of the validity of data is that doctors presented all medical information that was used in the study from the cancer patients group, while the participants in the non-cancer group provider their own medical data. To me it seems that they should have looked at a medical chart. Another concern for me is that only 26.6% of patients with pancreatic cancer in this study had diabetes type II. This is especially concerning because in the article that I previously blogged about, Probability of Pancreatic Cancer Following Diabetes: A Population-Based Study, the researchers claimed that 80% of pancreatic cancer patients also reported having diabetes. This leads to me to think that either the previous article had some major design flaws that I did not catch, or the current article was especially selective when choosing participants. Either way it makes me uncomfortable.