2018 Section 6 - Laryngology, Voice Disorders, and Bronchoesophalogy

S. E. Attwood et al.

Scatterplots for levels of gastrin (Figure 1) for individual patients from baseline to last value (SOPRAN study) or 5 years (LOTUS study) con fi rm that the results are not skewed by any extreme values. No gastric carcinoids or adenocarcinomas were observed in patients during the course of either study. Cardiovascular events In the SOPRAN study, eight patients in the omeprazole group and two in the open ARS group were reported to have had fatal cardiac SAEs. 8 Five of the eight patients in the omeprazole group had a cardiac medical history and/or cardiovascular medication at baseline (none were taking clopidogrel). The patients in the open ARS group who had a fatal cardiac SAE did not have any relevant medical history or concomitant medications. Cerebral infarction was reported in fi ve patients in the omeprazole group in the SOPRAN study and none in the open ARS group (Table 2). Cerebrovascular accident was reported in three patients (one in the esomeprazole group and two in the laparoscopic ARS group) in the LOTUS study ( < 3 reports per treatment arm so not shown in Table 3). Transient ischaemic attacks occurred in six patients in the SOPRAN study (three in the omep- razole and three in the open ARS group; Table 2). Reports of angina pectoris and cardiac failure were more common in patients in the SOPRAN study [angina pectoris: 13 patients (4.4%); cardiac failure: 8 patients (2.3%); Table 2] than in the LOTUS study [angina pec- toris: 4 (0.8%); Table 3; cardiac failure: 1 (0.2%); < 3 reports per treatment arm so not shown in Table 3]. Myocardial infarction was reported more commonly in the omeprazole group than in the open ARS group in the SOPRAN study (Table 2), while in the LOTUS study, there were fi ve myocardial infarctions in each treatment group (Table 3). With respect to the interaction between PPIs and clopidogrel, data from the SOPRAN and LOTUS studies were examined to identify patients who had been treated concomitantly with clopidogrel. Such an analysis was not part of the original protocol. Nevertheless, in the SOP- RAN study, it was found that six patients had received treatment with clopidogrel (four in the omeprazole group and two in the open ARS group). No cardiac events were recorded for these patients during clopido- grel treatment. In the LOTUS study, 17 patients had received clopidogrel at some point during the study (8 patients in the esomeprazole group and 11 patients in the laparoscopic ARS group). Of these, one patient (in the esomeprazole group) suffered a cardiac event (angina

decrease in zinc levels (Table 1) over time in both treat- ment groups in the LOTUS study, but levels appeared to stabilise by 5 years. No trends towards a reduction in iron-binding capacity (measured only in the LOTUS study) were observed (Table 1). Overall, the results did not indicate any development of malabsorption or anae- mia in either study. Scatterplots for levels of vitamin B 12 (Figure 2) for individual patients from baseline to last value (SOPRAN study) or 5 years (LOTUS study) revealed no extreme values. Neoplasms – benign and malignant The total number of neoplasms [SOC ‘ Neoplasms- benign, malignant and unspeci fi ed (including cysts and polyps) ’ ] diagnosed during the SOPRAN study did not differ substantially between treatment groups (11 in the omeprazole group and 9 in the open ARS group). In the open ARS group, three deaths due to neoplasms were observed: two lung cancers and one pancreatic cancer. One of these patients was diagnosed with lung cancer after randomisation and was hence never operated on. In the omeprazole group, one death due to a malignant brain neoplasm was observed. No patients developed gastric cancer, but one patient in the open ARS group who had long-segment Barrett ’ s oesophagus at baseline developed oesophageal carcinoma during the study. With the exception of two rectal cancers (both in the omepra- zole group and considered not related to study drug by the investigator), the types of cancer were spread among several different organ systems. In the LOTUS study, benign, malignant and unspeci- fi ed neoplasms (including cysts and polyps) were reported more frequently in the esomeprazole group (11 in the esomeprazole group and 5 in the laparoscopic ARS group). With the exception of four patients with pancreatic carcinoma in the esomeprazole group [two of which had a fatal outcome (women aged 64 and 66 years)], the tumours were spread among several dif- ferent organ systems. It should be noted that, in the SOPRAN study, one case of pancreatic cancer was reported in the open ARS group and none in the omep- razole group. Gastrin and chromogranin A There was a rapid increase in gastrin (Table 1) in the fi rst year of follow-up in the PPI group in both studies, after which the values plateaued. A similar pattern was observed for serum chromogranin A levels (measured in the LOTUS study), which increased mainly during the fi rst year and remained elevated at 5 years (Table 1).

Aliment Pharmacol Ther 2015; 41: 1162 – 1174 ª 2015 John Wiley & Sons Ltd

183