Sunday, February 8, 2015

HPV Vaccination and Type-1 Diabetes

There are 14+ studies showing that vaccines do not increase the chance of getting type-1 diabetes, and I've posted that list in the past. However, most of those studies were done before HPV vaccines were developed, so this posting focuses on HPV vaccines.

This posting covers the two HPV vaccines in common use in the US: Cervarix (made by GlaxoSmithKline) and Gardasil (made by Merck).  A vast amount of background information on HPV and these vaccines is available on Wikipedia:
Including HPV prevalence, death rates, etc.

Studies on HPV Vaccines and Type-1 Diabetes

It is important to remember that these vaccines are typically given to girls (and increasingly boys) at about 10-14 years old.  This is "prime time" for type-1 diabetes diagnosis as well.  So we should all expect stories that talk about someone who got an HPV vaccine and then came down with type-1 diabetes a few weeks later.  Occasionally, we'll even hear about stories where the person was diagnosed with type-1 diabetes the day of the vaccine, or maybe just a day or two later.  Because of blind bad-luck, and the relative number of people given the vaccine and diagnosed at the same time, you will see a few that overlap.

But the important question is this: do people who get HPV vaccines get type-1 diabetes at a higher rate, than those who don't?

The Huge Gardasil Study

A very large study was published in 2012, which looked at Gardasil and autoimmune diseases. It covered 190,000 people, and was done by an insurance company. Type-1 diabetes was one of the diseases they looked for specifically.

Here is the summary from the news coverage (bolding added):
They found no increase in any of 16 autoimmune disorders in the vaccinated population compared to a matched population of non-vaccinated girls and women. The 16 autoimmune disorders they looked for were:  "... immune thrombocytopenia, autoimmune hemolytic anemia, systemic lupus erythematosus, rheumatoid arthritis, juvenile rheumatoid arthritis, type 1 diabetes, Hashimoto's disease, Graves' disease, multiple sclerosis, acute disseminated encephalomyelitis, other demyelinating diseases of the central nervous system, vaccine-associated demyelination, Guillain-Barr syndrome, neuromyelitis optica, optic neuritis and uveitis."
I know some people are afraid that the same drug companies that sell the vaccine will also fund the studies showing they are safe. Having an insurance company run the study is much better. They are the people who pay for the vaccine, and they are the people who pay for any adverse side effects of the vaccine, so they are the best people to evaluate safety. Plus, they have the records to do it well. Not just anyone can review 190,000 patient records!


Large Cervarix Study

This was also a large study, about 18,000 people.  Type-1 diagnosis was specifically tracked as part of the trial's safety protocol, and the people were followed for 4 years.
In the Phase III trial, no differences were observed in the overall safety profile of the vaccine when administered to women aged 15–25 years with or without evidence of prior HPV-16/18 infection. [r3] 
Clinical Trial Record:

Don't underestimate the size (and therefore the power) of these studies.  Larger studies make it possible to detect smaller effects, and an 18,000 person study could detect even a very slight rise in the chance of a disease (and a 190,000 person study, even more so).  For comparison, phase-III studies, the largest required before a new drug is approved, are usually around 300 people for potential cures for type-1 diabetes. Obviously, both the Cervarix and Gardasil studies described above are vastly larger than that.  It's not a direct comparison, because phase-III studies are done on people who have type-1 diabetes already, while these were done on the general population.  Never the less, these are very large, powerful studies, and much larger than commonly used in testing potential type-1 diabetes cures.

Other Studies

You'll notice that there are no studies listed here that found HPV vaccines to cause higher rates of type-1.  That's because I could not find even one such study.  I looked for such a study in two ways. First, I looked for the Clinical Trials site which is run by the FDA and is required for all studies that are used for drug approval in the US.  Second I looked at "Google Scholar" which is Google's specialized academic paper search engine.  

In total, there were 60 Cervarix and 113 Gardasil studies listed on the FDA's clinical trial registry, and none of those found an increase in type-1 diabetes based on HPV vaccination.  Obviously, I did not read 170+ studies.  I did look through their titles, and then skimmed the abstracts of the ones that looked promising, and then read the abstract (or occasionally the whole paper) if the abstract looked like it contained applicable data.  I was not able to find one paper that suggested a higher rate of type-1 diabetes in people who got HPV vaccines as compared to people who did not.

This is a summary of information available in 2008 in this document:
Here is their data:
No increased risk of [new onset autoimmune diseases] was observed in the HPV-16/18 vaccine group compared with any of the control groups in any age group. ... The overall incidence of these events was low and within the range expected in the general population.  An integrated analysis [ie. review of many different studies] of over 68,000 subjects who received AS04-adjuvanted vaccines or controls demonstrated a low rate of autoimmune disorders (~0.5%), with no evidence of an increase in relative risk associated with AS04-adjuvanted vaccines. [r1]
New onset of autoimmune diseases (NOADs). ... No differences were seen in the frequency of NOADs between study groups in any age group during any follow-up period and no apparent clusters of events were detected . [r1]
In vaccinated girls aged 10–14 years, no withdrawals due to adverse events and no vaccine-related SAEs [serious adverse events] were reported over 24 months of follow-up after the first vaccine dose. [r2]

Common Complaints and Worries about HPV Vaccine Safety

A common worry that I hear is "these vaccines are not tested enough to know they are safe".

If this is an issue for you, ask yourself how many times a vaccine should be tested; in how many different clinical trials?   The FDA's answer to this question is "at least 4".  For the FDA, 4 clinical trials are the minimum level of testing (and not just for vaccines, for all new drugs).  As I said before Cervarix has already been tested in 60 separate clinical trials, and Gardasil in 113 trials.  Even if you would prefer a safety margin of 2x or 3x more trials than the FDA requires, these vaccines are way above that.

I sometimes hear people say "HPV vaccination has been associated with [bad effect X]..." 

Of course.   Every day, 100s of girls are given the HPV vaccination.  Just based on chance, some of those people are going to get the flu, get a bad grade, be diagnosed with type-1 diabetes, or be hit by a car, that very day!  (Or maybe the next week.)  With the large number of vaccinations given, even very rare events (like heart attacks and strokes in the young) are going to happen from time to time. That's what statistics are all about.

You will hear people say "I heard about person X, and they got rare cancer the week after an HPV vaccination".  They sure did.  If you vaccinate a million people a year, then some very rare things are going to happen.  In the case of HPV vaccines, there are two separate groups of people who will endlessly repeat any vague badness they can associate with HPV vaccination.  (One group is the anti-vaxxers, who object to all vaccines, and the other group is ultra conservatives, who object to vaccines against sexually transmitted diseases.)

A second issue is VAERS.  VAERS is a public database of medical events seen after vaccinations. Anyone at all can add events to VAERS.  They are not limited to entering effects on their own patients, and can enter cases where they don't know important facts.  There is no quality control or required fact checking.  VAERS was set up to find very rare side effects.  The hope was that if something very unusual happened, it would be stored in the database, and if it happened again, it could be investigated.  Unfortunately, the database is subject to manipulation, and in any case does not imply causation.  The following is a true story:

When HPV was first being used, I read in both Catholic and anti-vaccine web sites about HPV causing a girl's death.  They referred to VAERS reports.  I was able to find the report, and it went like this [4]:
VAERS ID 279592  2007-05-24
Administered by: OTH     Purchased by: OTH
Symptoms: Death Thrombosis
Information has been received from a licensed visiting nurse via a nurse practitioner. The nurse practitioner was told by a friend that a female patient was vaccinated with Gardasil and two weeks after developed a blood clot. Subsequently the patient died. The cause of death was from the blood clot. The reporting licensed visiting nurse considered the blood clot to be immediately life-threatening and disabling. Additional information has been requested. [There was no additional information included.]
In this case, the report was not made by anyone who actually knew the subject.  Nor even by someone who knew someone who knew the subject.  The licensed visiting nurse then filed a VAERS adverse effect report, although she did not include the name of the patient (she did not know it), nor the vaccine dose, nor the date of the vaccination or death.  In the legal world, this would be hearsay twice over.  In the scientific world this is too vague to be useful.  Indeed, it's not clear the event actually happened at all.  There is no required fact checking of VAERS data, .

If you'd like to read more about VAERS and HPV data, you can look here:

But do we really know about HPV Vaccine safety, when children are getting so many other vaccinations?  What if a combination causes a problem?

It's the idea that since vaccines are tested one at a time, combinations of vaccines are not tested and might be dangerous.  It's wrong, because vaccines are not tested one at a time.  They are tested "on top" (or in addition to) a normal vaccination schedule.  So the testing covers the situation where all the vaccines are given.

When phase-III trials are run for a new vaccine, they are comparing a normal vaccine schedule plus the new vaccine, to a normal vaccine schedule. (As they should.) But if the new vaccine caused a bad interaction with the existing schedule, then it would certainly turn up in the clinical trial. Similarly, if there was some problem with the total number of vaccines, then again that would show up in the testing for HPV.

But HPV Vaccination is new, so how can long term problems be known?

I think there are two good answers to this worry.  The first is simple: HPV is generally given to pre-teens and teenagers.   Type-1 diabetes is most commonly diagnosed just slightly later: in teenagers. Therefore, it makes sense that if there was a problem, it would show up in terms of higher type-1 diabetes rates within a few years (as the pre-teens who were vaccinated became teenagers who were diagnosed).  That hasn't happened.  Gardasil has been in use for over 8 years, and Cervarix for over 5 years, so it has now been enough time so vaccinated people would have progressed into the prime years for type-1 diagnosis.

If HPV caused or contributed to type-1 diabetes, we would see three strong signals: First when some girls started to be vaccinated, there should have been a "bump" in type-1 diagnosis.  There wasn't. Diagnosis continued to steadily grow at the same rate it has for decades.  Second, when the vaccine was officially recommended for girls, then again there should have been a corresponding "bump" in diagnosis for girls. This would have been particularly obvious, because it would have been in girls and not in boys. But it didn't happen. Third, when HPV was recommended for boys, then there would have been a corresponding "bump" a few years later.  Just in boys, so obvious to spot.  Again, it did not happen.

Bottom line: HPV has been in use long enough so that if it caused or contributed to type-1 diabetes, it would have been obvious (three times over), already.

A second answer to this worry, is to ask the person if Acai berries (or whatever the current health fad) have been tested "long enough"?  Has the person reviewed the long term safety tests for Vitamin-D supplements?  Organic food? A low fat diet?  The fact is that almost nothing in our lives is systematically tested for long periods of time.  If you think about it, you would need to run the test for a person's whole life to really be sure, and the results would only help the next generation, and only if nothing in the world had changed over those years to effect the results.  In short, "not tested long enough" is a bar nothing can reach, and one that is applied selectively to the things the worrier doesn't like.

More Information

SNOPES, which is a great general resource for "internet scare stories" has a good write up on Gardasil:

More References

[r1] Descamps D, Hardt K, Spiessens B, Izurieta P, Dubin G. Safety of human papillomavirus (HPV)-16/18 AS04 adjuvanted vaccine for cervical cancer prevention: integrated summary of 11 clinical trials. Presented at: 26th Annual Meeting of the European Society for Paediatric Infectious Diseases. Graz, Austria, 13–17 May 2008.

[r2] Schwarz TF, Descamps D. Long-term safety and immunogenicity of an AS04 adjuvanted cervical cancer vaccine in girls aged 10–14 years. Presented at: 26th Annual Meeting of the European Society for Paediatric Infectious Diseases. Graz, Austria, 13–17 May 2008.

[r3] Cruickshank M, Teixeira J, Gall S et al.; on behalf of the PATRICIA Study Group. Efficacy and safety of GSK's HPV vaccine in women initially seropositive or seronegative for HPV-16/18 in a Phase III trial. Presented at: EUROGIN: New Strategies of Cervical Cancer Prevention – the Reality of HPV Vaccines. Monte Carlo, Monaco, 4–6 October 2007 (Abstract SS1–2).

[r4] These examples are from anti-vaccine web sites:

Joshua Levy
publicjoshualevy at gmail dot com 
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My daughter has type-1 diabetes and participates in clinical trials, which might be discussed here. My blog contains a more complete non-conflict of interest statement. Thanks to everyone who helps with the blog.