• familial pancreatic cancer
• screening
• surveillance
• acid

We read with interest the letter by Harinck et al1 as a response to our paper2 and are grateful for the opportunity to respond.

Harinck et al offer the criticism that the trial was not prospective as suggested by the title of our paper and the median number of examination visits was only two, with a maximum of 1 year apart. As described in our paper this was a peer-reviewed, board-approved trial that was designed as a prospective study for a period of at least 10 years in cooperation with the Institute of Biometry and Epidemiology of the Philipps-University Marburg. Since the study is still ongoing the number of screened individuals at risk (IAR) is continually increasing. The median number of MRI per patient was 4 (2–12) and of endoscopic ultrasonography (EUS) examinations 2 (1–7), maximally 4.5 years but not 1 year apart. The median observation time per patient was 24 (1–66) months, and the overall follow-up was 152 patient years, respectively. As stated in detail in the methods section every IAR received examinations with two different types of MRI techniques per visit, which explains the different numbers of investigations.

Harinck et al suggest that our low diagnostic yield (1.3%) compared to the studies by Canto et al (10%,3) and Poley …