View PDF Version - RePub - Erasmus Universiteit Rotterdam
View PDF Version - RePub - Erasmus Universiteit Rotterdam
View PDF Version - RePub - Erasmus Universiteit Rotterdam
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
INTRODUCTION<br />
Dynamic prediction of response using longitudinal profi les 81<br />
Dynamic updating of an individuals prediction of a specific outcome based on new<br />
gathered information is not routinely implemented in prediction models, but can be<br />
of great importance for the individual patient, for the clinician and the further choice<br />
of treatment. We shall introduce the general problem in the setting of therapeutic<br />
treatment of chronic hepatitis B.<br />
In the last years treatment options for chronic hepatitis B have largely been extended.<br />
Still the virus is very difficult to eliminate and only 10-36% of the treated patients<br />
remains in remission after therapy. 1,2 Peg-interferon (PEG-IFN) has proven effective,<br />
but also has its limitations regarding multiple and possible serious side-effects. 3<br />
However, it has been demonstrated that the response to a course of interferon is<br />
durable and leads to both improved survival and reduction of the incidence of hepatocellular<br />
carcinoma. 3,4 Therefore, prediction of response to PEG-IFN is of great<br />
importance. During therapy the patient is monitored at frequently scheduled followup<br />
visits and several markers are measured to anticipate continuation. Different<br />
patterns of these markers have been described, ter Borg et al. 5 , and also flares<br />
(sudden increase) of markers have been identified as possible predictors of response,<br />
Flink et al. 6 Up to now however, these measurements have not been implemented<br />
routinely to update the individual prediction of response. These may even proof<br />
helpful in guiding and supporting the patient through the long treatment and identify<br />
patients who will have little benefit by continuation of treatment. The above<br />
sketched situation is the inspiration of this paper.<br />
Our aim is to develop a dynamic prediction model to update the prognosis of the<br />
individual patient dependent on the new information that becomes available after<br />
each follow-up visit. The statistical challenge is to find a powerful tool to make use<br />
of the multivariate longitudinal profiles to predict the binary outcome variable. We<br />
introduce a new method and elaborate on existing methods. The performance of<br />
the different methods are furthermore extensively compared. Two different views<br />
are considered: (1) directly modeling the prediction of the outcome variable with<br />
the use of logistic regression techniques with repeated updates and (2) indirectly<br />
classifying individuals into an outcome category over time using Bayes’ theorem.<br />
For the direct approach (1) either the observed marker value or the subject specific<br />
pattern of the marker is used as predictor. We suggest a GEE solution directly<br />
entering the observed marker values. 7 In contrast, Maruyama et al. 8 first fitted a<br />
linear mixed effect model to obtain the subject specific patterns of the longitudinal<br />
markers and then entered the estimated random effects in a logistic regression.<br />
Since the estimated random effects have measurement errors an adjustment of the