Page 218 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Chapter 8
described previously).17 Within LUMC the Electronic Medication Record (EMR) system EZIS (version 5.2, Chipsoft) is used, which can be consulted by physicians, pharmacists and nurses. DPYD genotyping results are communicated electronically by the responsible pharmacist into the EMR and are visible for other users of the EMR.
The prospective screening program was initiated on 15 April 2013. During a kick-off meeting attended by medical oncologists and fellows, the staff was informed and agreed on the prospective program. New medical oncologists and fellows were informed about the prospective screening program during the regular introduction program for new staff members. Genotyping was performed three times per week (Monday, Wednesday and Friday) in order to minimize the lag time between sampling and test. This resulted in a turnaround time of 2 days, allowing rapid start of treatment if needed. Ethical approval by the Institutional Review Board of LUMC was not required for the current study as it evaluates standard care. Patient data from the EMR was handled following the Codes of Proper Use and Proper Conduct in the Self-Regulatory Codes of Conduct.18
Study end points
Three study end points were evaluated to determine the successfulness of the screening program that was introduced at LUMC. We evaluated:
- The ‘implementation’, in other words, requests of the DPYD tests as standard care in
daily practice;
- The proportion of test results with a dose recommendation provided by the pharmacist;
- The follow-up of the dose recommendations by oncologists, calculated as the number
of follow-ups of dose recommendations by prescribers, excluding the patients in which a follow-up was not possible (e.g., no therapy).
Study procedures
The implementation, or routinely application of the prospective (pretreatment) DPYD screening in daily practice was evaluated by determining the proportion of patients who were screened for DPYD variants when an incident prescription for 5-FU or capecitabine was given. The data were extracted from two electronic databases. The first database contains data of all patients who are genotyped for DPYD variants. The second database (EMR EZIS) contains individual patient medical records. This system is also used by oncologists to electronically prescribe 5-FU and capecitabine. Prescription data prior to the start of the study was studied as well, to ascertain that 5-FU or capecitabine prescription was indeed the first prescription for the patient. The patient identification number was used to connect data from both databases. Discrepancies between information in the queried databases were resolved by manually checking the individual electronic patient records to identify the reason of their absence in one of the two searches. After connecting the data from both databases, all patient data were anonymized. All manual changes (additional information, removal of duplicates, among others) to the queries were double checked by the two first authors (CL and MvS).
To evaluate the follow up of the recommended dose reductions by the oncologists, medical records of patients carrying a variant in DPYD were inventoried as to determine
216