Cohort Study of Pioglitazone and Cancer Incidence in Patients with Diabetes Mellitus

Study type

• Study topic:
  Disease /health condition

  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Assessment of risk minimisation measure implementation or effectiveness

  Disease epidemiology
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Study drug International non-proprietary name (INN) or common name:
  PIOGLITAZONE
• Medical condition to be studied:
  Type 2 diabetes mellitus

Population studied

• Short description of the study population:
  Patients from Kaiser Permanente of Northern California (KPNC) Diabetes Registry who were followed up from January 1, 1997 to December 31, 2005 for the usage of pioglitazone and incidence of cancer.
• Age groups:
  Adults (18 to < 46 years)

  Adults (46 to < 65 years)

  Adults (65 to < 75 years)

  Adults (75 to < 85 years)

  Adults (85 years and over)
• Special population of interest:
  Other
• Special population of interest, other:
  Diabetes mellitus patients
• Estimated number of subjects:
  252467

Study design details

• Main study objective:
  To evaluate whether treatment with pioglitazone was associated with risk of incident cancer at the 10 most common sites (prostate, female breast, lung/bronchus, endometrial, colon, non-Hodgkin lymphoma NHL, pancreas, kidney/renal pelvis, rectal, and melanoma) in a cohort of patients with recognized diabetes.
• Outcomes:
  Hazard ratio of the 10 most common cancers associated with ever use of pioglitazone. Hazard ratio of the 10 most common cancers associated with time since first use, duration and dose of pioglitazone. Age, gender, calendar year-specific cancer incidence rates stratified by diabetes status. Age-, gender-adjusted incidence rates stratified by diabetes status, calendar year. Association between diabetes and risk of the 10 most common cancers for full KPNC member and survey responder.
• Data analysis plan:
  Study 1: Cox proportional hazards regression modeling was used to provide point and interval estimates of the relative hazard of the 10 most common cancers associated with ever use of pioglitazone (primary analysis) and time since first use, cumulative duration, and dose (secondary analyses). In all regression analyses, these measures of exposure to pioglitazone were treated as time-dependent covariates and time since entry into the cohort was the time scale. Study 2: Cancer incidence rates were calculated with attention to the proper allocation of at-risk person-time. The association between diabetes and risk of each of the 10 most common cancers was assessed using Cox proportional hazards regression models with control for available potential confounders: age, gender, calendar year. Similarly, Cox regression technique was used to examine the association between diabetes status and cancer risk among survey responders with adjustment for additional potential confounding variable.

Documents

Study results

AD-4833-403 (KPNC 10 common cancers) 10-yr study report

AD-4833-403-FUM 35.4 KPNC Interim Results Report-2011-12-07

Study publications

Lewis JD, Habel LA, Quesenberry CP, et al. Pioglitazone Use and Risk of Bladder…