The renewed population screening for cervical cancer in the Netherlands started in January 2017. Two major changes compared to the program until then, were the switch towards primary HPV screening and the introduction of a self‐sampling device for non‐responders. Implementing changes within an existing screening program requires careful preparation. Therefore, extensive verification studies were carried out during implementation of HPV testing and use of self sampling.
Our goal was to verify that the Cobas 4800 HPV test in the Dutch screening laboratories performed similar to the clinical validation; furthermore for verification of the self-sampling we wanted to establish the optimal method of sample preparation and its feasibility in the routine lab flow. Cervical swaps were collected in PreservCyt, whereas self sampling was performed using the Evalyn brush. For HPV testing the Cobas 4800 HPV test platform was implemented, making use of the p480 de- and recapping, x480 sample processing and PCR setup, and z480 PCR amplification. The desing of the verification was written out in test protocols.
Verification of HPV test: An initial test on reproducibility, precision and accuracy was carried out at the vendor’s site on a single system using a specific verification panel and subsequently repeated on all Cobas 4800 assay systems (n=15) in the 5 screening laboratories. In order to secure the technical performance of the equipment, an operational test was conducted, including two weeks fully operational running on high volume prior to the release of the systems. The HPV verification tests were successfully performed and the data fulfilled the criteria of reproducibility, precision and accuracy. All assay systems in the screening laboratories also met the functional requirements.
Verification of the self sampling: Statistical analysis of data from a previous study was used to determine the adequate liquid volume of sample processing: it appeared that processing the Evalyn brushes in standard 20 ml PreservCyt vials was sufficient and resulted perfectly in a uniform laboratory work flow for HPV testing using both cervical scrapes and self sampling. In addition, a stress test was conducted to examine whether vials including brushes could safely be mechanically processed by the HPV assay systems.
Verification has proven to be useful in introducing two major changes within the cervical cancer screening in the Netherlands. Based upon the verification results, the Cobas 4800 systems were released for use within the renewed screening program and a standard operation procedure for the processing of self-sampling devices has been established and implemented.