pioglitazone and risk of bladder cancer

There have been conflicting reports about piogitazone and bladder cancer over the years. The FDA just came out with a specific updated review and warning (see http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm532772.htm

for the brief statement and http://www.fda.gov/Drugs/DrugSafety/ucm519616.htm for the document). In 2010 the FDA commented that there was the potential for bladder cancer based on some human and animal studies, but they did not conclude there was an increased risk, the review was ongoing, and that patients should not stop taking pioglitazone. In 2011 they did require the manufacturer to include the bladder cancer warning.

--the updated 2016 warning:

--pioglitazone should not be given to those with active bladder cancer

--clinicians should “carefully consider the benefits and risks before using pioglitazone in patients with a history of bladder cancer"

--patients should contact their clinician if they see red urine, new/worsening urge to urinate, pain on urination

--we should report adverse events related to the use of pioglitazone, online at www.fda.gov/MedWatch/report

--the prior recommendations/concerns were based on the somewhat human inconsistent data and animal studies.

    ​-- the PROactive study , a double-blind study of 5238 patients randomized to pioglitazone vs placebo along with existing diabetic drugs. 4873 completed the final trial visit and 73.9% enrolled in the observational study with median followup to 12.8 years (see Dormandy JA. Lancet 2005; 366: 1279, and the 10-year followup Erdmann E. Diabetes Obes Metab 2016; 18:266)

        --there was an increase in bladder cancer cases during the trial in those on pioglitazone, with RR2.83 (1.02-7.85) which did not persist in the 12.8 year followup

    --the 10-year prospective industry-supported study using the Kaiser Permanente of North California database of diabetic patients registered between 1997-2002 and followed til 2013.​

       --the 5-year interim results found no significant increased risk for bladder cancer with pioglitazone

       --but, there was a trend to increased risk of bladder cancer with increased pioglitazone dose and duration

    --However, a recent study evaluated the UK Clinical Practice Research Datalink (CPRD) from 2000-2013 of 145,806 patients (of whom 10,951 initiated pioglitazone), with mean followup of 4.7 years:

                --622 received diagnosis of bladder cancer

                --the fully adjusted HR for pioglitazone was 1.63 (1.22-2.19), a 63% increase, and with increasing risk with increasing dose or duration of pioglitazone

--so, the FDA is reinforcing their initial concern about bladder cancer risk, with a continued warning that the drug label indicate that there may be an increased bladder cancer risk with pioglitazone

Commentary:

--the PROactive study found, as a secondary analysis, that cardiovascular outcomes in diabetics were 16% less frequent than continuing with their regular meds (see comments in http://gmodestmedblogs.blogspot.com/2015/12/empagliflozin-good-and-bad.html ). I have been prescribing it to a few people who were insulin-averse but needed better diabetic control, and with some success (in general I am more afraid of drugs like pioglitazone which are not-so-specific and affect important prevalent enzyme systems in the body, in this case PPAR-a, peroxisome proliferator-activated receptor- alpha system,  which regulates hepatic lipid metabolism. but the clinical data seemed pretty good for patients who had exhausted the regular oral drugs). and a more recent study found pioglitazone to decrease cardiovascular events in patients with insulin resistance and prior TIA or ischemic stroke (see http://gmodestmedblogs.blogspot.com/2016/04/diabetes-update-ahrq-and-pioglitazone.html  )

I find now that I am often able to convince patients who are insulin- and even needle-averse to try one of the weekly GLP-1 agonists (which are all injected). And I do think the data are quite good on their diabetes efficacy, their limited effects on the rest of the body/higher specificity, and their clinical outcomes. For example, a study was published on liraglutide finding decreased cardiovascular events (see http://gmodestmedblogs.blogspot.com/2016/06/liraglutide-decreases-cardiovascular.html  which evaluates that study and explains my positive take on GLP-1 agonists).  In my clinical practice, I initially try metformin, but then proceed to a GLP-1 agonist as a second agent if the patient needs further diabetic control and is willing