Response to letter to the editor

Abstract

Dear Sir,In the Letter to Editor submitted by Drs. Tomljenovic and Shaw [1], concern was raised regarding the inclusion of women who received only one or two doses of the HPV vaccine in the autoimmune safety surveillance, in addition to those who received three doses, in the sense that this combined population may be less sensitive for detection of serious autoimmune disorders. However, we believe it is a strength of our study that those who received only one or two doses were included in the safety evaluation. This is because those who developed an adverse reaction after the first or second vaccination might not continue to complete the three-dose series. By excluding women who only received one or two doses, we may select a population in which adverse reaction was less likely to develop. However, we agree completely with Drs. Tomljenovic and Shaw for the need of by dose analysis. In our study population of 189 629, 51 603 received three doses of the HPV vaccine (44 000 received all three doses within 12 months). Autoimmune safety signals were evaluated in our study by assessing any cluster of disease onset by age, dose sequence and timing of onset. Confirmed new onset autoimmune conditions were tabulated according to the number of doses received prior to the onset. No cluster was identified for more doses. Furthermore, in our graphical analysis (e.g. Figure 2 in Chao et al. [2]), the onset of autoimmune endocrine conditions was fairly evenly distributed by number of doses received. Drs. Tomljenovic and Shaw also raised concerns about the composition of the Safety Review Committee (SRC) in this study. The SRC was charged to evaluate the data related to three study components: (i) general safety, (ii) pregnancy outcomes and (iii) autoimmune conditions. The selection of the SRC members thus was based on the consideration of representation of expertise in vaccine safety and pharmacoepidemiologic methodologies, scientific contents related to vaccines and disease aetiology, related clinical expertise, as well as years of experience. In addition, the data they reviewed included the results that had been subjected to specialty-specific review by case review committees as detailed in our original publication. Therefore, the approach to select SRC members from the noted specialties seemed appropriate to us, and to regulatory bodies overseeing the study as well. We also would like to point out that most neurological and ophthalmologic outcomes were very rare – there were only a few confirmed new onset cases in our study despite a large population under surveillance. The small number of cases presents challenges in evaluating pattern of disease onset, as discussed in our original publication [2]. Hence, we supplemented our safety evaluation by a simulation modelling technique to compare incidence of autoimmune conditions in the vaccinated vs. the nonvaccinated women. In the case of such rare events, safety evaluation could not solely rely on clinical knowledge of the disease. Given these considerations, having a full panel of autoimmunologists, neurologist and ophthalmologist on the SRC was unlikely critical for proper evaluation of the safety signal. As mentioned in our previous letter to editor, the need to evaluate long-term safety and rare adverse events of HPV vaccine should not be overlooked. For rare events, studies designed specifically to look for these outcomes will be needed. No conflict of interest was declared.