Using evidence to examine the safety of Gardasil

Medicine - Syringe - DrugsEvery once in a while, the debate on vaccine safety and efficacy surfaces in the news and on social media. The most infamous example remains the controversy surrounding the measles-mumps-rubella (MMR) vaccine and autism (link that has been repeatedly proven as non-existent).

Most recently, it is the human papillomavirus (HPV) vaccine Gardasil that has been put in the spotlight. An article published in the Feb. 5th,2015 edition of the Toronto Star questioned the safety of this vaccine, based on a few cases of serious adverse events, some temporally related to HPV vaccination. This article created an uproar in the medical and scientific community leading to the article being removed from the newspaper’s website. The Star’s publisher even posted a comment that concluded “in this case [the] story treatment led to confusion between anecdotes and evidence”.


Before reviewing the evidence on Gardasil effectiveness and safety, let’s go back to what the vaccine is trying to prevent: infections by certain HPV types and the consequences of such infections. HPV is the most prevalent sexually transmitted infection. More than 120 types of HPV have been identified and approximately one third of these can infect the genital tract. A dozen HPV types are classified as “high risk” (or oncogenic), because they have been linked to cancers.  HPV types 16 and 18 are the high risk types most frequently found in cancers. HPV types not classified as oncogenic are referred to as “low risk”.HPV types 6 and 11 are low risk types that cause condylomas, or genital warts.

More than 99 per cent of cancers of the uterine cervix are caused by HPV. Other types of cancers have also been associated with HPV, but to a lesser extent: cancers of the vulva, vagina, anus, penis and oropharynx. Progression to cancer is usually the result of a persistent infection of the epithelium by a high-risk HPV.  Most of the time, there will be an intermediate stage of “pre-cancer” between simple infection and invasive cancer. The precancerous stage often lasts several years, if not decades.

The quadrivalent human papillomavirus (qHPV) vaccine (Gardasil) was approved by Health Canada in 2006. This vaccine protects against infection by the HPV types most commonly associated with genital warts (HPV 6 and 11) and with precancerous and cancerous lesions (HPV 16 and 18). Its use is only prophylactic; hence, it will not treat  an HPV infection. It is currently licensed for use in Canada in females aged 9 to 45 years and males aged 9 to 26 years. The recommended vaccination schedule comprises three doses given at 0, 2 and 6 months. There is currently no recommendation for booster doses.

The efficacy of the qHPV vaccine was initially studied in Phase II and III trials where approximately 10,000 women aged 16 to 26 years received the vaccine, and were compared to a similar number who did not. The primary endpoints were: HPV 16-18 related cervical, vulvar and vaginal precancers, genital warts, and persistent infections with HPV 6, 11, 16, 18. The efficacy of the qHPV vaccine was over 95 per cent for all the endpoints when the vaccine was given to young women who had no sign of previous infection with HPV vaccine types. Longer-term follow-up of subpopulations from the initial studies confirmed sustained protection for up to 8-10 years after vaccination.


As any randomized trial, especially those done to secure homologation of a product, the qHPV vaccine phase II and III trials were conducted in strictly controlled “best scenario” conditions. Starting in 2007 in Australia, several HPV vaccination programs were implemented in resource rich settings, targeting girls, teens and young women. Data from these “real world” settings were eagerly awaited, in the hope that the trial efficacy would translate into program effectiveness. And finally such data is available.

The most exciting news for health care providers, public health planners (and anyone interested in the topic really) is that the population effectiveness appears to be excellent. The results are consistent across countries, from Canada, to Australia, Belgium, Denmark, Germany, New Zealand, the United States and Sweden.  The protection is evident in terms of important reductions in HPV infections, cervical abnormalities (both when measured by screening and diagnostic tests), and genital warts.

The population impact is especially telling when considering genital warts, as it is the HPV related disease with the shortest latency between infection and clinical manifestation (as such this is where the impact was expected to appear first), with a few reports finding no genital warts in cohorts of women vaccinated at a young age, within a program with a high coverage rate. More excellent news is that there appears to be some cross-protection, with qHPV offering protection against certain HPV types not directly targeted by the vaccine. Finally, herd immunity has already been demonstrated, as young unvaccinated women and heterosexual men in settings where there is a high vaccination coverage rate for young women appear to be indirectly protected also.

As with any other new vaccine or treatment, the safety of the qHPV vaccine must be thoroughly investigated. Immediate side effects were well documented in initial trials. In women 9 to 26 years, the most common side effects reported in the Gardasil monograph were at the injection site: pain (83.9%), swelling (25.4%) and erythema (24.7%). Among systemic adverse reactions, the most frequent were headaches (28.2%), pyrexia (13.0%) and nausea (6.7%). The frequency of the systemic adverse reactions was similar in the Gardasil and placebo groups.

Up to 0.8 per cent of trial participants reported a serious adverse reaction but only 0.04 per cent were considered related to vaccination. The same trials also recorded data on rare, often delayed, adverse events. There was no difference between the vaccinated and the control groups in regards to development of autoimmune disorders.

A total of 40 deaths out of 29,323 trial participants were recorded. The main cause of death was motor vehicle accident (5 in vaccine group and 4 in control group). Other causes were drug overdose/suicide (2 in vaccine group and 6 in control group), gunshot wound (1 in vaccine group and 3 in control group) and pulmonary embolism/deep vein thrombosis (1 in vaccine group and 1 in control group).


Several research teams, many independent from the drug company that markets the qHPV vaccine, have investigated whether the vaccine can be linked to adverse reactions. Specific attention has been given to auto-immune disorders, which can be grouped into three categories: rheumatologic/autoimmune disorders (such as systemic lupus erythematosus, rheumatoid arthritis, immune thrombocytopenia), autoimmune neurological/ ophthalmic disorders (such as multiple sclerosis, demyelinating diseases, Guillain-Barré syndrome) and autoimmune endocrine conditions (such as type 1 diabetes, Hashimoto’s disease, Graves disease). A recent review of studies investigating those outcomes by de Vincenzo and colleagues, concluded that there was no evidence of an increase in any of those adverse events, in populations vaccinated with the qHPV vaccine.

Closer to home, Canadian data was recently published by Harris et al. regarding the results of their study on the adverse events occurring in Ontario’s school-based HPV program. Between September 2007 and December 2011, more than 690,000 doses of Gardasil were administered through this program.

The adverse effects reported were 19.2/100,000 doses of Gardasil. The most common adverse effects following injection were ‘allergic reaction – dermatologic’ (25%), ‘rash’ (22%) and ‘local/injection site reaction’ (20%). There were 10 serious adverse effects reported: two cases of anaphylaxis (none of which met the criteria from the Brighton Collaboration), two cases of seizures (one of which was subsequently diagnosed as epilepsy), one case of thrombocytopenia and one death (secondary to an undiagnosed underlying cardiac condition).

In summary, the body of evidence underlining the effectiveness and the safety of the qHPV vaccine at the population level is growing. The fact that multiple studies, with different designs, conducted in different populations, reach the same conclusions strengthens our confidence in the findings. When pondering the safety of the vaccine, we should remember that before the introduction of the qHPV vaccine, young women across Canada, every year, were diagnosed with autoimmune disorders, were experiencing symptoms that were difficult to explain and treat, and even a few died of unknown causes.  As q-HPV vaccine programs are rolled out, it is to be expected that some of those same health events will occur, and some will be temporally related to HPV vaccination. For this reason, evaluating the safety of the vaccine is challenging. The scientific community must remain vigilant and continue monitoring the safety of the vaccine, using rigorous methods.

Although some patients may not want to hear their health care providers’ advice on qHPV vaccine, many will be receptive if we explain the extremely positive results we can already see with vaccination programs, and take the time to answer their questions related to vaccine safety. We should remember that a positive recommendation by a health care provider remains an important factor influencing the decision to be vaccinated against HPV.