-- JoAnnAlvarez - 11 Nov 2011

Predicting cardiac effects in breast cancer therapy

Meeting notes

  • Ask Carrie to check value of baseline_adiponectin of ">125000".

2012 March 23

  • They have re-run all the neureglin samples. Now they have about 80 observations. Most of the detection limit problem is solved. About 5 of them were over the limit, and they used extrapolation to estimate those. I recommended they describe this process in thier publication.
  • Want to look at the following outcome: maximum change in ejection fraction over follow up period. Emailed to clarify.
  • The variables measuring this are echoef1, echoef2, .... They each have an associated date.
  • Should we use a longitudinal model?
  • Look at plot of the trajectories over time.
  • Research question: Do neureglin, cardiovascular risk factors, baseline physical activity, and beta blockers/ACE inhibitors predict ejection fraction?
  • Want to make one variable indicating either beta blocker, ACE inhibitor, or both, to save degrees of freedom.
  • Carrie will email me when she is finished updating the database.

2011 December 1

  • Discussed how to obtain one measurement from the IPAQ, which is assessed at baseline. There are three parts: activity, sports, and leisure. Plan on summing the three index scores for now. (See the redcap form for the var names.) Carrie will check the IPAQ manual to verify that this is the validated way to combine these.
  • il6 and il1 beta are normally below the detection limit in healthy patients.
  • I talked about some possible ways of handling the detection limit problem.

2011 November 11

  • Prospective observational study of 200 patients.
  • At baseline, the following is measured: ejection fraction (EF), ~6 biomarkers from blood samples, DNA, exercise questionnaire from IPAQ. The IPAQ yields a numeric score.
  • Patients then begin chemotherapy. They either get an anthrocyclin, in which case treatment lasts ~ 4 months, or another drug, in which case treatment lasts 2 years.
  • Then the patients will have four exercise questionnaires over time (not IPAQ). These are measured in MET hours per week.
  • At a final follow up, all the measurements taken at baseline are re-collected.
  • Patients are followed by phone for a total of 5 years.
  • The main outcome is cardiac dysfunction as measured by EF at the last follow up visit: difference in volume between contraction and relaxation. An EF of above 55% is considered normal.
  • Ask Frank how to deal with any patients who stop chemo early due to side effects. (Few of these are expected. The side effects could or could not be related to cardiac problems.)
  • We also have longitudinal EF measured at different times because of standard of care. How to include this data?
  • Two initial questions: how should we change the measurement of exercise in the interim? Stop collecting it? Give pedometers? Which biomarkers are important?
  • To answer the first question: see if bl exercise affects fu EF (control or interact for bl EF) and see if bl exercise affects fu biomarkers.
  • To answer the second question, see if all biomarkers affect the final EF. Control for bl EF and age.
  • May want to include interaction between bl EF and exercise.
  • Results in 2-3 weeks!
Topic revision: r5 - 23 Mar 2012, JoAnnAlvarez

This site is powered by FoswikiCopyright © 2013-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback