Association of human papillomavirus and systemic sclerosis: A population based cohort study

Systemic sclerosis (SSc) is an uncommon rheumatic disease characterised by fibrosis of skin and vasculopathy.1 Human papillomavirus (HPV) is known as an independent risk factor for multiple autoimmune disorders and malignancies.2, 3 Previous study revealed that co-infections of multiple HPV subtypes were near two times more frequent in the SSc group.4 A significant positive association between self-reported abnormal Pap test and diffuse skin involvement in SSc patients were addressed in a Canadian study.5 However, no previous studies on the epidemiological relationship between HPV infection and SSc has been investigated. To assess the relationship between HPV infection and SSc, we analysed the information present in the 1997-2013 Taiwan's National Health Insurance Research Database. The baseline characteristics among groups are presented in Table 1. Demographic data between the HPV study group and comparison group were analysed by the chi-squared (χ2) tests. Cox proportional hazard regression model was applied to estimate the hazard ratio and 95% CI. Kaplan-Meier curve was presented to calculate the cumulative incidence of SSc in follow-up months. Time-to-event analysis and sub-group analysis were also conducted to further integrate our study. Both HPV infection and SSc were identified by ICD-9 codes between January 1997 and December 2013. We only included patients with at least two outpatient claims or one inpatient claim one year within the index date or the first date of SSc diagnosis. Index date was the first day of HPV diagnosis. We included 135 764 cases of patients newly diagnosed with HPV from 2003 to 2013, and matched the controls at index date and selected by propensity score matching. Similar distributions reduce the heterogeneity and selection bias. Individuals in both the study and comparison groups were followed from their index visit until an SSc event, withdrawal from the NHI programme, or till the endpoint of our study 31 December 2013. The exclusion criteria of our study include: diagnosis of HPV infection before 2003, SSc diagnosis before index date and patients did not receive HPV treatment within 3 months after the index date. After propensity score matching, there were 25 cases of SSc found in the HPV group (n = 135 764), and 23 SSc cases in the compare group (n = 135 764). In our time-to-event analysis, the incidence density of SSc was similar in the HPV-infected cohort compared with the control group (0.27 vs 0.26) per 100 000 person-months. After adjusting demographic and co-morbidities variables listed in Table 1, the adjusted hazard ratio (aHR) of developing of SSc was 1.13 (95%CI 0.57-2.21). The Kaplan-Meier curve (Figure 1) showed an insignificant cumulative probability (%) of SSc among two groups (log-rank P = .7880). In the subgroups divided by follow-up time, a significant risk of SSc was found in group diagnosed with HPV within the first year after index date (aHR = 3.44, 95% CI: 1.12-10.61), and the risk was lowered 1 year after index date (aHR = 0.56, 95% CI: 0.25-1.26). In subgroup analysis, the interaction was insignificant either in age or in sex stratification (P for interaction = .893, .385, respectively). Additionally, the individuals with comorbid diseases (systemic lupus erythematous, polymyositis and rheumatic arthritis) developed a significantly higher aHR of SSc. Limitations in our study included the absence of other immunosuppressive medication besides from corticosteroids. In conclusion, HPV infection did not remain as an independent risk factor for developing SSc after adjusting for baseline characteristics, comorbidities, and co-medications. Age 40 years and above, female sex and autoimmune comorbidities were found to be risk factors. Even considering the known limitations of retrospective cohort studies, results of our analysis may raise consensus on the association between HPV and SSc. The authors would like to thank all colleagues who contributed to this study. The authors declare that they have no conflict of interest. Study concept and design: MLC and JCCW. Acquisition, analysis and interpretation of the data: JCCW and JYH. Drafting of the manuscript: MLC. Critical revision of the manuscript for important intellectual content: JCCW, JYH, YMH.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.