European Journal of Epidemiology

, Volume 28, Issue 8, pp 705–707

No evidence of a link between multiple sclerosis and the vaccine against the human papillomavirus

Authors

  • Paolo Pellegrino
    • Unit of Clinical Pharmacology, Department of Biomedical and Clinical Sciences, University Hospital “Luigi Sacco”Università di Milano
  • Carla Carnovale
    • Unit of Clinical Pharmacology, Department of Biomedical and Clinical Sciences, University Hospital “Luigi Sacco”Università di Milano
  • Valentina Perrone
    • Unit of Clinical Pharmacology, Department of Biomedical and Clinical Sciences, University Hospital “Luigi Sacco”Università di Milano
  • Stefania Antoniazzi
    • Unit of Clinical Pharmacology, Department of Biomedical and Clinical Sciences, University Hospital “Luigi Sacco”Università di Milano
  • Marco Pozzi
    • E. Medea Scientific Institute
  • Emilio Clementi
    • Unit of Clinical Pharmacology, Department of Biomedical and Clinical Sciences, University Hospital “Luigi Sacco”Università di Milano
    • E. Medea Scientific Institute
    • Unit of Clinical Pharmacology, Department of Biomedical and Clinical Sciences, University Hospital “Luigi Sacco”Università di Milano
Letter to the Editor

DOI: 10.1007/s10654-013-9830-y

Cite this article as:
Pellegrino, P., Carnovale, C., Perrone, V. et al. Eur J Epidemiol (2013) 28: 705. doi:10.1007/s10654-013-9830-y

Cervical lesions due to human papilloma virus (HPV) types 16 and 18 infection are associated with the development of almost 70 % of cervical cancer cases worldwide [1]. The recent introduction of preventive vaccination against papilloma virus will significantly reduce the incidence of cervical cancer. Two virus-like particle (VLP)-based vaccines targeting HPV16 and HPV18 are now offered to adolescent girls and young women among various countries within different vaccine programs [1, 2]. The efficacy of these vaccines in reducing cervical lesions and the safety of this large-scale immunisation in the young adult population has been widely assessed in several clinical trials [1, 2] Furthermore both vaccines were monitored via post licensure safety surveillance study that showed a safety profile broadly consistent with pre marketing studies [1, 2].

More recently, however, cases regarding the onset or exacerbation of multiple sclerosis (MS) following HPV vaccination have been reported, indicating a possible relationship between anti HPV vaccination and MS [3, 4].

A relationship between MS and vaccination was previously reported for vaccination against hepatitis B (HB) in France, a large-scale, immunisation implemented in the early 1990s leading to the immunisation of almost 20 million of individuals. Reports on the temporal association between HB vaccination (HBV) and MS onset led to severe concerns in public opinion about the safety of this vaccine [5]. Despite further epidemiological studies failed to demonstrate this relationship and no increase in MS prevalence was observed following any vaccine introduction, public confidence was lost and HBV coverage in France remains below 25 % [5].

HBV vaccination is not dissimilar to that with HPV in terms of target populations and large-scale implementation. An early evaluation of the possible relationship between MS and HPV immunisation appears therefore fundamental, also to avoid undue confidence loss in HPV vaccination.

Sutton et al reported a case series of five Australian patients aged 16–25 years presenting MS with a CNS inflammatory disorder that occurred within 28 days following HPV vaccination [3]. In two of the described cases, MS had not been diagnosed before while three patients had already MS that was however exacerbated by HPV vaccination (Table 1). Chang et al. [4] described two cases of Clinically Isolated Syndrome (CIS) occurring 30 and 45 days following HPV vaccination. These data, while indicating a possible relationship between HPV vaccination and MS onset, do not allow determining whether there is or there is not causal association between the two events [3, 4]. Indeed the cause of MS remains unknown at present; hypotheses regarding possible roles of viral agents, and vaccines, in its pathogenesis have been proposed with none established [5].
Table 1

Demographics and clinical characteristic of literature retrieved cases

Author

Case

Vaccine

Dose

Age (years)

Symptom onset (days)

Diagnosis

Sutton I [3]

1

qHPV

3

16

21

CIS

2

qHPV

2

25

16

CIS

3

qHPV

2

21

1

CDMS

4

qHPV

3

26

4

CDMS

5

qHPV

2

16

4

CDMS

Chang J [4]

1

qHPV

2

19

30

CIS

2

qHPV

1

18

45

CIS

qHPV Quadrivalent HPV vaccine, CIS clinical isolated syndrome, CDMS clinical defined multiple sclerosis

To assess the relationship between HPV vaccination and MS, we analysed the information present in the Vaccine Adverse Event Reporting System (VAERS), one of the biggest national and international databases, which collects approximately 28,000 adverse drug reactions (ADRs) following immunisation per year.

The analysis of both the domestic (i.e. United States) and non-domestic VAERS databases allowed us to retrieve 107 (mean per year: 17.8 ± 8.45) reports of MS onset (102 cases) or MS exacerbation (5 cases) between 2006 and 2008. The majority of reports regarded the quadrivalent HPV (qHPV) vaccine and almost 27 % of the cases occurred after more than 30 days from vaccination (Table 2).
Table 2

Database cases

Characteristic

N

%

Vaccine

 qHPV vaccine

102

95

 bHPV vaccine

3

3

 Unknown

2

2

Doses before onset

 1st dose

19

18

 2nd dose

25

23

 3rd dose

31

29

 Unknown

32

30

Time to symptom onset

 0–7 days

17

16

 7–14 days

10

9

 15–30 days

10

9

 31–60 days

11

10

 61–120 days

8

7

 Over 120 days

11

10

 Unknown

40

37

qHPV Quadrivalent HPV vaccine, bHPV bivalent HPV vaccine

In a search within the European Medical Agency (EMA) adverse event database, we retrieved a further 130 cases of MS (124 cases of onset, 6 cases of exacerbation) of which 123 were associated with the qHPV vaccine and 7 with the bivalent HPV vaccine. Ten additional cases were retrieved from the Australian TGA Database of Adverse Event Notifications between April 2007 and February 2013, of which nine were new MS diagnosis and one was a disease relapse.

The incidence of MS reports following HPV vaccination was not formally evaluated within any previous studies. Considering the number of qHPV vaccine doses distributed in United States between June 1, 2006 through July 30, 2012 (46 million) and the VAERS domestic reports in this time-frame (31 reports) we estimated a reporting rates (RRs) of 0.08/100,000 doses.

A similar evaluation of qHPV vaccine doses (7 million of doses) and reports from the Australian TGA database yielded a RRs of 0.14/100,000 doses. Similar results were estimated from the EMA database.

The temporal relationship between HPV immunisation and symptom onset does not provide evidence about causality relationship. In addition the population in study (young women) have a higher risk of developing MS [5], thus HPV vaccination may casually occur closely to symptom onset in a patient who was developing MS independently from vaccine. The incidence of MS in the absence of HPV vaccination in young women within 6 weeks periods of observation has been estimated to be 1/100,000 [5]. While these data are not directly comparable to those of the pharmacovigilance vaccination databases because limited to a specific region in the US, still they may provide insight about the aetiology of cases retrieved from our database research.

Even taking into account the known limitations of passive surveillance systems, i.e. under-reporting and report quality [1], the results of our analysis raise doubts about a possible relationship between HPV vaccination and MS onset or exacerbation. Concerns about the possible adverse reactions to vaccination occur on a regular basis in the public opinion. Considering the public health importance of preventing HPV infection, measures should be taken in order to dissipate concerns about possible increases in the risk of MS following HPV vaccination.

Acknowledgments

This work was supported by Grants from Agenzia Italiana del Farmaco (AIFA), the Regional Centre of Pharmacovigilance of Regione Lombardia.

Conflict of interest

The authors declare that they have no conflict of interest.

Copyright information

© Springer Science+Business Media Dordrecht 2013