Fluke or Trend? Metformin vs Cancer

Ovarian cancer is one of the deadliest cancers known to womankind.  It's development is slow & insidious w/o any good means of screening with adequate sensitivity & specificity (statistical terms used to determine if any given test is good at picking out a condition and correctly ignores those w/o said condition).  As a result, most cases of ovarian cancer are found late in its course, unlike w/breast cancer for which modern mammography is able to detect minute microfoci of calcifications long before they metastasize (of course, this raises other issues which we'll gloss over today).  But as a result of this confluence of factors, we don't have anything (effective yet safe) to throw at it.

So I woke up yesterday to find a case-control study published early online in Cancer in which the authors concluded that metformin use was associated with better survival in patients w/ovarian cancer compared to non-use.  Wow!  Talk about a game changer!  Here's a cheap, readily available medication w/minimal side effect profile and risk of adverse events.  Sure, it carries a Black Box warning but realistically, unless you have bad kidneys, the likelihood of developing lactic acidosis is very small.  Well, at least it's small enough to rate this medication as a first-line option in diabetes.  Best yet, while metformin helps lower high blood sugars (by improving insulin sensitivity), when used in isolation w/o most any other drug, especially insulins & sulfonylureas, it cannot cause hypoglycemia (dangerously low blood sugar).

Does this mean we need to put every woman suffering from ovarian cancer on metformin?  Absolutely not!  You can't make a presumption of cause & effect based upon a case-control study.  Furthermore, it would definitely be a case of jumping the gun to consider therapeutic changes based upon a study of just 72 cases (ovarian cancer w/diabetes on metformin) compared to 143 controls (ovarian cancer w/o diabetes or metformin).  A further subset analysis was performed on 61 cases of epithelial ovarian cancer w/diabetes on metformin compared to 178 controls (epithelial ovarian cancer w/o diabetes or metformin).

Where does this leave us?  Obviously, the 1st question to address is whether this is a fluke or a trend.  If you read the article, you'll see that this is definitely not a fluke.  The authors looked for this result based upon a prior study published in April 2005 in the British Medical Journal that took notice of new findings in cellular signaling pathways.  Bottom line: there's definitely some science behind this "new" finding.

Follow @alvinblin
Tweet