Skepticism about science and medicine

In search of disinterested science

Beyond Belief: Deadly vaccines for Africa and Asia

Posted by Henry Bauer on 2013/05/12

Here’s what’s known about Gardasil and Cervarix (see Deadly vaccines):
1. There’s no good evidence that they do anyone any good
2. There’s proof that they harm some people, at times even fatally

What then would one expect the manufacturers to do: Withdraw the vaccines?

Of course not. They would find a way to sell them despite the evidence:

Drugmakers, health groups bring poor girls vaccine (cr. AP, Linda A. Johnson)
“Two multinational drugmakers are teaming up with top global health groups to protect millions of girls in the world’s poorest countries from deadly cervical cancer.
Starting with pilot programs in eight Asian and African nations, the ambitious project ultimately is intended to inoculate more than 30 million girls in more than 40 countries by 2020. . . .
The endeavor was announced Thursday by the GAVI Alliance, a public-private partnership that’s worked with drugmakers to deliver affordable vaccines to poor countries to treat childhood illnesses that are big killers.”

The GAVI Alliance began as the “Global Alliance for Vaccines and Immunisation”. The name change may seem trivial, but it reflects the fact that it’s all about marketing and fund-raising; as with too many other “charities”, those who benefit most are the staff of the organization who enjoy well-paying jobs with ample benefits — for example, about 15% of salaries of GAVI staff are paid into retirement plans or accounts, amounting to about $4 million in 2011. If you can find in the GAVI Alliance Financial Reports  more information than that about how many employees there are, who they are, or what their compensation is, then you’re a better investigator than I am.
But what do the drug companies get out of this arrangement?

“Merck will supply its Gardasil for $4.50 per dose, and Glaxo its Cervarix for $4.60 per dose. In the U.S., the shots cost well over $100 apiece, and a three-dose series over six months is required. . . . .
The goal is for the governments of those countries to show they can set up a national system . . . to provide the vaccines over the long term . . .
Merck, based in Whitehouse Station, N.J., is providing 93 percent of the shots initially. It’s also agreed to provide more shots at an even-lower price in the future, if higher volumes of vaccines are ordered, as that would reduce production costs. . . .
In the U.S., the vaccines have become steady money makers since they were launched a half-dozen years ago, but they haven’t turned into the mega-sellers initially envisioned.”

I think it’s unlikely that Merck or Glaxo are planning to lose money on this. “Production costs” can be calculated in a number of different ways, of course, but I think one understands the present initiative best by looking at the marginal costs of producing more of these vaccines, in other words regarding the initial research and testing and marketing costs as done with and considering now only how much the actual cost of making more doses is. I would bet it’s less than $4.50-4.60 per dose.

But my disbelief at this initiative, and my cynicism and paranoia were aroused not only by the financial aspects. I was bemused — though not really surprised — to see the blatantly false claims of what these vaccines can prevent:
“The vaccines protect against the strains of human papilloma virus (HPV) . . . that most commonly cause cancer. The virus, transmitted during sex, causes cervical cancer as well as vaginal, vulvar, anal and oral cancers. The vaccines prevent roughly 70 percent of those cancers.”

There is no proof at all the vaccines prevent any cancer. All the evidence amounts to no more than that those strains of  HPV are often found in association with cervical cancer. That’s a correlation, which everyone should know is no proof of causation. The vaccines have not been in use for long enough for there to be any data about actually preventing cervical cancer.
As for “vaginal, vulvar, anal and oral cancers”, this was news to me, so I enquired further. And indeed the authorities do claim it to be so:

From the CDC:
“HPV types are often referred to as “low-risk” (wart-causing) or “high-risk” (cancer-causing), based on whether they put a person at risk for cancer.”
[This classification is guess-work. It is not known whether they cause cancer; everything is merely correlation]
”The International Agency for Research on Cancer found”
[NO: speculated]
”that 13 HPV types can cause cancer of the cervix; one of these types can cause”
[NO: pure speculation]
”cancers of the vulva, vagina, penis, anus, and certain head and neck cancers (specifically, the oropharynx, which includes the back of the throat, base of the tongue and tonsils). The types of HPV that can cause genital warts are not the same as the types that can cause cancer.
Most people who become infected with HPV do not know they have it. Usually, the body’s immune system gets rid of the HPV infection naturally within two years. This is true of both high-risk and low-risk types. By age 50, at least 4 out of every 5 women will have been infected with HPV at one point in their lives. HPV is also very common in men, and often has no symptoms” [emphasis added].
In other words, the risk from “high-risk” HPV is tiny at best. There are about 150 million women in the United States; so about 120 million (“4 out of every 5”)  had been infected with HPV at some time or other. In 2009, about 12,500 had been diagnosed with cervical cancer and 4000 died of it.

So the proportion of HPV-infected women
 who experience cervical cancer
 is 12,500/120,000,000 = 1/10,000 = 0.01%

When only 1 in 10,000 infected people experience illness, what sort of illness-causing infection is that?

As to cancers of “vulva, vagina, penis, anus, or the oropharynx (back of the throat, including the base of the tongue and tonsils)”; they “are much less common than cervical cancer. Much less is known about how many people with HPV will develop cancer in these areas.”
If they’re even less common, then of course their incidence is less than 0.01%.

***************************************************************************

This business illustrates once again how untrustworthy are the pronouncements of official institutions.

The only grounds for claiming HPV to be a cause of any cancers is that HPV has been found in some proportion of certain cancers. Such correlation does not amount to evidence that the virus causes those cancers. Indeed, the claim of causation becomes truly absurd when correlations are used to estimate what proportion of cancers are caused by infectious agents in general:
“The estimated total of infection-attributable cancer in the year 2002 is 1.9 million cases, or 17.8% of the global cancer burden. The principal agents are the bacterium Helicobacter pylori (5.5% of all cancer), the human papilloma viruses (5.2%), the hepatitis B and C viruses (4.9%), Epstein-Barr virus (1%), human immunodeficiency virus (HIV) together with the human herpes virus 8 (0.9%). Relatively less important causes of cancer are the schistosomes (0.1%), human T-cell lymphotropic virus type I (0.03%) and the liver flukes (0.02%)”
(D. M. Parkin, “The global health burden of infection-associated cancers in the year 2002”, International Journal of Cancer, 118(2006) 3030-44).
There is no actual proof of causation in any of those cases. Everything is calculated on the basis of correlations.
It’s worth recalling that the “war on cancer” declared in 1971 in the USA had funded unprecedented amounts of research by virologists searching for viral causes of human cancer; but they failed to find any at all (Peter Duesberg, Inventing the AIDS Virus [Regnery, 1996] chapter 3). Even before 1971, for some six decades some researchers had looked unsuccessfully for viral causes of human cancers after Peyton Rous had found a viral cause of chicken tumors.

Lack of substantive success in finding any human cancers demonstrably caused by a virus has now morphed into the declaration that viruses cause some proportion of cancers because they are sometimes found associated with those cancers. In other words, viruses cause cancer because correlation proves causation.
The only possible cure for this diseased interpretation might be to demand that every medical researcher be required to take an elementary course in probability and statistics during which they are required every day to write for several hours the phrase

CORRELATION NEVER PROVES CAUSATION

Advertisements

32 Responses to “Beyond Belief: Deadly vaccines for Africa and Asia”

  1. natasha tauber said

    …maybe add to this that ‘hpv’ has never actually been isolated/found/proven to exist, much like ‘hiv’, ‘hepC’ or ‘swine-bird-flu’ etc……..

    • Henry Bauer said

      natasha tauber:
      Interesting, thanks, I wasn’t aware of that. So how was HPV and its umpteen strains “identified”?

  2. Aut Aut said

    Hpv has never been proven to exist? But… isn’t it the same virus that “also” causes the warts?

  3. dankegel said

    You wrote: “The only grounds for claiming HPV to be a cause of any cancers is that HPV has been found in some proportion of certain cancers. Such correlation does not amount to evidence that the virus causes those cancers.”

    No, there’s more evidence than that. Here’s a brief history (links at hpv.kegel.com):

    In 1983, Harald zur Hausen published the paper “A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different geographic regions” in which he described a new virus, HPV 16, which was present in 11 of 18 german cervical cancer samples.

    In 1989, the paper “Immortalization of human foreskin keratinocytes by various human papillomavirus DNAs corresponds to their association with cervical carcinoma” showed that the types of HPV associated with cancer could immortalize human cells (a neccessary part of carcinogenisis).

    In 1993, the paper “Epidemiologic evidence showing that human papillomavirus infection causes most cervical intraepithelial neoplasia” showed that HPV accounted for most of the risk from the classic risk factors for cervical cancer: high number of sex partners, cigarette smoking, low age at first intercourse, and lower socioeconomic status.

    Also in 1993, the paper “The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53” showed that HPV deactivates the anti-cancer gene P53.

    In 1999, the paper “Human papillomavirus is a necessary cause of invasive cervical cancer worldwide” established that HPV is present in nearly all cervical cancer.

    Those five papers, and the accumulated understanding we have today of how HPV works, establish beyond a reasonable doubt that HPV causes nearly all cervical cancer today. As a result, Harald zur Hausen won the 2008 Nobel Prize for Medicine for his discovery.

    So, no, it’s not just correlation. We have a fairly deep understanding of how HPV causes cancer.

    You also wrote “When only 1 in 10,000 infected people experience illness, what sort of illness-causing infection is that?”

    But that’s incorrect; http://www.cancer.org/cancer/cancerbasics/lifetime-probability-of-developing-or-dying-from-cancer says that 1 in 147 American women contracts cervical cancer in her lifetime, and since essentially all cervical cancer is caused by HPV (see above), and about half of all women get infected by HPV in their lifetime (iirc), that means the chance of cervical cancer given being infected at least once by HPV is more like 1 in 74. (Assuming I haven’t bollixed up my figures.)

    • Henry Bauer said

      dankegel:

      What does “cause” mean?

      Scientists often speak of “necessary and sufficient” as the criterion if A is to be said to cause B.

      The polio virus causes polio: all cases of polio come after infection with that virus, and there are no cases of polio without infection by that virus. So too with measles. Tuberculosis. Malaria. And so on.

      Not all cases of cervical cancer are associated with HPV (e.g. “HPV 16, which was present in 11 of 18 german cervical cancer samples”). Far from all infections by HPV area associated with cervical cancer:
      “http://www.cancer.org/cancer/cancerbasics/lifetime-probability-of-developing-or-dying-from-cancer says that 1 in 147 American women contracts cervical cancer in her lifetime.”
      That’s precisely what I said:
      12,000 cervical cancers per year among ~150,000,000 American women of whom about 4 in 5 are HPV infected for some time, sooner or later. Give each woman 80 years of life. Then the chance of any one woman getting cervical cancer in her lifetime is 12,000 x 80 divided by 150 million, i.e. ~ 1 in 150. That risk is essentially negligible compared to most other risks, and the risk of death is only 1/3 of that 1/150, because there are only ~4,000 deaths among the 12,000 cases.
      (My 0.01% compared rates in any given year: in any given year only 0.01% of HPV-infected women get cervical cancer. Over a lifetime for any given woman, that becomes ~1/150).

      You cite “Epidemiologic evidence”: that’s simply correlation.

      “Immortalization of human foreskin keratinocytes by various human papillomavirus DNAs corresponds to their association [CORRELATION again] with cervical carcinoma”.

      As to “risk factors for cervical cancer”: Again, “risk factors” are measures that CORRELATE with some ailment.

      “’Human papillomavirus is a necessary cause of invasive cervical cancer worldwide” established that HPV is present in nearly all cervical cancer”:
      No. It established that HPV could be found in nearly all cervical cancer. Correlation once more (even though in at least some cases the correlation is only 11 of 18?)

      The only reason to claim that HPV causes cervical cancer is because there is a correlation, and it’s far from a necessary or sufficient correlation.

      “essentially all cervical cancer is caused by HPV (see above)”
      No: HPV is often found in women who have cervical cancer. It’s a correlation.
      Since 4 out of 5 American women (CDC, cited in my blog post) will be HPV-infected for some years during their life, the probability that someone with cervical cancer will test positive for HPV is quite high; that’s not the same thing as a high probability that someone with HPV will get cancer.

      “As a result, Harald zur Hausen won the 2008 Nobel Prize for Medicine for his discovery.”
      Nobel Prizes have been given for mistakes on more than one occasion: for treating schizophrenia by infection with malaria (1927) or by lobotomy (1949), and for discovery of the non-existent kuru “slow virus” (1976), for instance.

      Then again, please bear in mind that cervical cancer was declared in 1993 to be caused by HIV, “the virus that [supposedly but not actually] causes AIDS” — for the same reason that HPV is now said to be implicated, a correlation between finding HIV often in women who had cervical cancer.

      Modern medical practices rely quite promiscuously on the notion that correlation proves causation. “Risk factors” are nothing but correlations, yet they are treated as though they were causes. The Institute of Medicine report, “Evaluation of Biomarkers and Surrogate Endpoints in Chronic Disease” (2010), finds that none of the biomarkers ( = risk factors = correlations) that are in common use are actually reliable measures of the ailments with which they are associated; for example, cholesterol is not a valid measure of heart disease and is therefore not a cause of heart disease.

      The only way to establish whether Gardasil prevents cervical cancer, by contrast to protecting against HPV, would be to have a trial in which women vaccinated with Gardasil are followed over their lifetime together with a matched control group of non-vaccinated women. Any other data simply reflect correlations.

      Everyone agrees, in principle, that benefits of treatment should be weighed against risks. The lifetime chance of a woman dying of cervical cancer is about 1/500 (1/150 of getting the cancer, but only 1/3 of those die). By contrast, a significant number of girls have already been damaged severely enough by Gardasil as to cause compensation totaling about $6 million to have been paid out because of “26 new deaths reported between September 1, 2010 and September 15, 2011 as well as incidents of seizures, paralysis, blindness, pancreatitis, speech problems, short term memory loss and Guillain-Barré Syndrome”.

      The GAVI initiative is truly “Beyond Belief”.

      [Revised 21 May, credit Richard Karpinski]

      • Dan Kegel said

        Thanks for the reply!

        I agree, correlation != [does not equal] causation. But the fact that HPV is present in roughly 99% of cervical cancers is not the only evidence that HPV causes cancer. We also know that HPV’s E6 gene codes for a protein that inactivates P53, one of the key anticancer systems in the human cell. (See http://www.ncbi.nlm.nih.gov/pubmed/8221889 ). And we know that you can immortalize skin cells using HPV genes. So not only did we place HPV at the scene of the crime, we also found a gun in its hand, and verified that the gun was in working order and capable of killing.

        That’s a lot more evidence than we have that the vaccine has harmed any girls. Sure, there are a couple hundred reports of girls becoming sick after taking the vaccine… but that only establishes a correlation, not causation. Scientists did take that warning seriously, and studied 200,000 women, but found no increased risk of any of the reported problems. ( See http://www.ncbi.nlm.nih.gov/pubmed/21907257 and http://www.ncbi.nlm.nih.gov/pubmed/21973261 ) Sometimes bad things happen by chance.
        So it seems there isn’t even a real correlation between the vaccine and the kinds of harm you’re talking about.

        Did you see the news today that the vaccine has reduced the incidence of HPV in the generation of girls since the introduction of the vaccine? There’s a fair bit of info from the study at http://www.medpagetoday.com/PublicHealthPolicy/PublicHealth/39973
        It says that among all females 14 to 19, even those not vaccinated, infection with high risk HPV strains has gone down from 7.2% to 3.6%. That’s great, considering that half of all girls and most boys haven’t been vaccinated yet.

        My personal prediction? The vaccine (and its improved successors) will drive HPV out of the cancer business… and folks like you will claim that cervical cancer went away for some other reason, just as antivaccine groups claim that measles and polio went away without any help from vaccines. I look forward to your revisionist history in advance 🙂

      • Henry Bauer said

        Dan Kegel:

        That the vaccine decreases HPV incidence doesn’t mean that it does any good against cervical cancer or genital warts or anything else. But the evidence of harm is quite definite. The National Vaccine Injury Compensation Program doesn’t shell out money unless there has really been damage done:
        The “National Vaccine Injury Compensation Program (VICP) has awarded $5,877,710 dollars to 49 victims in claims made against the highly controversial HPV (human papillomavirus) vaccines. To date 200 claims have been filed with VICP, with barely half adjudicated. . . . [after] 26 reported deaths of young, previously healthy, girls after Gardasil vaccination in just one year. . . . Merck studied the Gardasil vaccine in fewer than 1,200 girls under 16 prior to it being released to the market under a fast-tracked road to licensure. . . . Neurologist Dr. Ian Sutton reported negative neurological side effects from Gardasil. He reported five cases of multiple sclerosis-like symptoms emerging shortly after women received the Gardasil vaccine” (U.S. Court Awards $6 Million in Damages to Gardasil Victims).

  4. Dan Kegel said

    Hi Mr. Bauer,

    You write that Gardasil was only studied on 1,200 girls under 16 before licensing.
    But http://www.fda.gov/ohrms/dockets/ac/06/briefing/2006-4222B3.pdf
    shows that the clinical trials used to support approval of the vaccine included about 12000 vaccinated girls and women, not 1200.
    (And that size is not unusual; another vaccine, Boostrix, was approved after a clinical trial of 3,000 people,
    http://www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4097B1_2.pdf )
    But studies didn’t stop then; as I wrote earlier, they monitored 200,000 girls and women after vaccination, and found no harm ( http://www.ncbi.nlm.nih.gov/pubmed/21907257 and http://www.ncbi.nlm.nih.gov/pubmed/21973261 ).
    So the safety was established to about 1 in 10,000 before licensing, and to about 1 in 200,000 after licensing. Considering that the vaccine prevents a disease that kills about 1 of every 417 American women, clearly, getting vaccinated is much safer than not getting vaccinated.

    You mention the VICP. Statistics on the federal vaccine compensation program are at http://www.hrsa.gov/vaccinecompensation/statisticsreports.html , showing how many people have been compensated for claimed injuries from various vaccines since 1989. For instance, 545 for flu, 53 for HPV, and 35 for tetanus.
    This is a no-fault insurance system of sorts, and http://en.wikipedia.org/wiki/Vaccine_court#Previous_rulings says that one claim was paid “despite several studies showing that the vaccine neither causes nor worsens the disease, and despite a conclusion by the Institute of Medicine that evidence favors rejection of a causal relationship”.
    On the whole, I don’t think one can say that, e.g. the 35 claims paid out for injuries from tetanus vaccine mean that the tetanus vaccine is unsafe, or that the 53 claims paid out for Gardasil mean that Gardasil is unsafe. And again, it’s useful to keep these numbers in perspective; there are about 30,000 cases of HPV-related cancer every year in the US. In other words, HPV causes more cases of cancer in a day than the number of people *ever* compensated by VICP for Gardasil. So even if all of the VICP compensees were actually harmed by the vaccine, which is highly unlikely, the good done by the vaccine *still* is orders of magnitude larger than the alleged harm.

    Finally, you said “That the vaccine decreases HPV incidence doesn’t mean that it does any good against cervical cancer or genital warts.” But in Australia, which has a strong HPV immunization program, the rate of genital warts has fallen more than tenfold in the under-21 crowd since vaccination started ( http://www.bmj.com/content/346/bmj.f2032, http://www.lancet.com/journals/laninf/article/PIIS1473-3099(10)70225-5/abstract ). Countries and age groups that were not vaccinated have seen no such decline in genital wart rates, and genital wart rates were holding steady before introduction of the vaccine.
    It seems quite clear that Gardasil caused the decline in genital warts.

    Are you claiming that Gardasil did *not* cause the decline in genital warts seen in Australia?

    • Henry Bauer said

      Dan Kegel:
      So we have to agree to disagree and let readers make up their own minds.
      It’s easy to dismiss a few REPORTED deaths as unimportant if one doesn’t know the people involved, and doesn’t realize that some unknown number of similarly caused deaths were never recognized as such and therefore not reported.
      The incidence of cervical cancer was quite small before Gardasil, and had been dropping for two decades, and genital warts are not particularly damaging.
      All individuals need to weigh for themselves, and their children, the possible but low-probability risk against the low-probability benefit.

  5. Dan Kegel said

    So you admit that HPV causes genital warts, and that Gardasil prevents HPV infections and therefore prevents genital warts. That’s progress. Now, how far are you from admitting that HPV causes cancer? Does the fact that one can reliably immortalize cells in the laboratory using HPV genes ( heck, you can even buy a kit to do it, http://www.biocat.com/cgi-bin/page/sub1.pl?sub1=cell_immortalization&main_group=cell_biology )help convince you? Or that disrupting the action of HPV genes kills HPV-caused cancer in the laboratory ( http://www.ncbi.nlm.nih.gov/pubmed/12087999 )?

    The number of cancer deaths from HPV far, far outweighs the number of reported deaths from Gardasil… and those reported deaths are only attributed to Gardasil by correlation. As you are so fond of saying, correlation != causation.

    Right then, I’m off. Let’s check in again when good data is available about decreasing cancer rates in Australia.

    • Henry Bauer said

      Dan Kegel:
      I admitted no such thing.
      The Gardasil-caused deaths and other injuries are NOT a matter of correlation, because the Gardasil was injected at a known time and the damage followed ina manner that indicted the Gardasil—convincingly enough to have the government agency fork out $6 million.
      There’s all the difference in the world between a correlation in a set of data gathered by observation, with no info about relative times, and the observation that A follows B.
      A classic illustration: Some eminent biostatisticians did not accept that smoking had been proven to cause lung cancer just because there was so great a correlation between lung-cancer victims and heavy smoking. The issue was decided after enough dogs had been forced to inhale tobacco smoke and subsequently developed lung cancer: the effect clearly followed the cause ina sufficiently large number of instances.

  6. Dan Kegel said

    Hang on a tic. I finally got around to reading the “About” section on your blogs. It seems that you’ve written books about how there’s no such thing as the “scientific method”, and denying that HIV causes AIDS.

    So, let’s take a step back, and establish some common ground. Do you believe that, generally, living cells work by transcribing DNA into RNA, and then into proteins, and these proteins (plus lipid bilayers) are responsible for most of a cell’s structure and function?

    • Henry Bauer said

      Dan Kegel:
      Have you read those books of mine? If not, please draw no inferences. Ignorance is not bliss, and it’s certainly not a good basis for conversation. All of my books have had very good reviews. The “Scientific Method” one has been used as a graduate text in a number of places, which is why it’s been kept in print for two decades..

      I have no quarrel with the DNA –> RNA –> protein mechanism, though it’s actually rather more complicated than that sounds, since different bits of genes get activated to work together at particular times for particular purposes, and the “junk” (non-gene) DNA serves purposes presently still being unraveled, including how processes are turned on and off and when.
      Don’t forget the mitochondria (or chloroplasts respectively) in your picture of “most” of a cell’s function, that’s where the energy centers are.

  7. Dan Kegel said

    OK, moving on to cancer. Do you agree that it’s been established that mamalian cells generally have mechanisms which repair damage to DNA, which trigger cell death when those mechanisms detect irreparable damage to DNA, and that cancer generally involves a failure or suppression of those mechanisms?

    (I’m moving gradually because I have no idea where your understanding of cancer and how viri may cause it diverges from the standard one.)

    • Henry Bauer said

      Dan Kegel:
      No, I don’t agree.
      Please cite proof (not conjectures or speculations or one-off “studies”) of apoptosis triggered by failed repair mechanisms.
      Re cancer, I like the aneuploidy theory revivified by Duesberg, who first discovered oncogenes and then realized that had been a mistake.

  8. Dan Kegel said

    OK, let’s table cancer for the moment, and move on to warts. Do you agree that [1] it’s been established that warts are generally a transmissible disease, and that [2] the disease agent is a virus, and that, [3] in humans, these viri are generally known as HPV?

  9. Dan Kegel said

    Fascinating. ( I read it at http://dermatologycentral.typepad.com/files/on-warts.pdf just now.)

    We’ll come back to warts, but for now, on to the scientific method. Do you agree that, when trying to come to an understanding of physical phenomena, that the best way to proceed is to make hypotheses about how they work, and then test those hypotheses with experiments, using techniques like double-blinding, careful documentation, and reverification by independent third parties to guard against bias, wishful thinking, and fraud? i.e. would you change your mind about how warts work if presented with sufficient experimental evidence?

    • Henry Bauer said

      Dan Kegel:
      My book, Scientific Literacy and Myth of the Scientific Method (Urbana & Chicago: University of Illinois Press) 1992 was very favorably reviewed, widely adopted in undergraduate and graduate classes, and remains in print. You should read it.
      The “scientific method” is perhaps an ideal, but it’s certainly not how science has been and is being done. “Testing hypotheses” sounds great but in practice is quite fallible because experiments deliver unclear answers and different folk interpret them differently (see Thomas Kuhn on that).
      “Reverification by independent third parties” doesn’t happen because there’s no career benefit to repeating things and repetitions typically don’t get published. And there are no independent parties to do it, there are only competitors or colleagues. And the myth of reliable peer review has been exposed by quite a number of people.

  10. Dan Kegel said

    I’ve read parts of your book (thanks to Google Books).

    To rephrase your reply, you consider the “scientific method” a good ideal, don’t think it works out often in practice due to lack of clarity or motivation (apologies if I’ve mangled it), but are still open to having your ideas disproven by really clear experiments that give unambiguous results.

    So, what experimental results would cause you to agree that warts are a transmissible disease?

    Here are a few I found, in chronological order.

    – Jadassohn, J. “Sind die Verrucae vulgares übertragbar.” Verhandl. d. deutsch. dermat. Gesellsch. 5 (1896): 497. Available as free Google Book, https://play.google.com/store/books/details?id=L3zuDwdUDvYC
    On numerous occasions, he removed warts from patients, innoculated himself or colleagues with them, and observed warts appearing at the innoculation sites.

    – “Note on the experimental transmission of warts in the dog”, McFadyean & Hobday, 1898; pg. 341, http://books.google.com/books?id=yRAbAQAAMAAJ) which said, roughtly,
    Find a dog with lots of warts in its mouth, and remove one of its warts. Then find a few non-warty dogs, scratch up an area of their lips with a sterile instrument, and rub the cut side of the removed wart against them. Then observe the dogs for a few months, and see if they develop a wart.”
    When McFadyean & Hobday did this, warts were indeed transmitted.

    – This experiment was repeated around 1900 by Cadeac in both dogs and horses (http://books.google.com/books?id=xANVAAAAYAAJ, pg. 275), with similar results.

    – “INFECTIOUS PAPILLOMATOSIS OF RABBITS : WITH A NOTE ON THE HISTOPATHOLOGY.”
    Shope RE, Hurst EW. J Exp Med. 1933 Oct 31;58(5):607-24.
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2132321/pdf/607.pdf
    “A papilloma has been observed in wild cottontail rabbits and has been found to be transmissible to both wild and domestic rabbits. The clinical and pathological pictures of the condition have been described. It has been found that the causative agent is readily filtrable through Berkefeld but not regularly through Seitz filters, that it stores well in glycerol, that it is still active after heating to 67°C. for 30 minutes, but not after heating to 70°C., and that it exhibits a marked tropism for cutaneous epithelium. The activities and properties of the papilloma-producing agent warrant its classification as a filtrable virus. Rabbits carrying experimentally produced papillomata are partially or completely immune to reinfection and, furthermore, their sera partially or completely neutralize the causative virus. The disease is transmissible in series through wild rabbits and virus of wild rabbit origin is readily transmissible to domestic rabbits, producing in this species papillomata identical in appearance with those found in wild rabbits.”

    – “The Growth in the Developing Chicken Embryo of a Filtrable Agent from Verruca Vulgaris”, Bivins, The Journal of Investigative Dermatology (1953),
    http://www.nature.com/jid/journal/v20/n6/full/jid195358a.html
    Mostly good for its references.

    – “Warts in sheep: Identification of a papilloma virus and transmission of infection to sheep”,
    E.P.J. Gibbs, C.J. Smale, M.J.P. Lawman, Journal of comparative pathology 1975 Vol: 85:327-334. The google snippet says “This paper reports the occurrence of warts in sheep, the identification of papilloma virus and the transmission of infection to sheep and neonate hamsters by means of bacteria- and cell-free … Before inoculation, the suspensions of wart tissues were filtered through a 200 nm…”
    I haven’t read the paper yet, but that snippet is promising.

    That’s six firsthand accounts of the transmission of warts in humans, dogs, horses, rabbits, and sheep, with varying degrees of sophistication, over a period of 80 years, by six independent investigators. I’d call that independent verification. I can probably find many more.

    Given that evidence, do you still feel that warts are non-transmissible?

    • Henry Bauer said

      Dan Kegel:
      You haven’t read my book.
      This sort of hurried response on your part does not impress me as a serious attempt to argue the substantive issues.
      SIX cited references, over a period of 80 years — including about dogs, horses, rabbits, sheep, to support the notion that HPV (HUMAN papilloma viruses) cause warts?! PLEASE!!
      How many articles have been published about HUMAN warts? Where are the reviews of the subject, preferably something like the meta-analyses done by the Cochrane Collaboration?

  11. Dan Kegel said

    Henry,
    in an earlier message, you wrote “The issue was decided after enough dogs had been forced to inhale tobacco smoke and subsequently developed lung cancer”. Therefore you seem to accept animal models as valid for reasoning about lung cancer. Are you saying that animal models are ok for talking about cancer, but not about warts? If so, why?

    • Henry Bauer said

      Dan Kegel:
      You haven’t cited an animal model for warts, you cited half-a-dozen disparate sources referring to different species.
      My question remains: How many articles have been published about warts? Where are the reviews or meta-analyses of them to establish whether anything has been established?
      On just about any topic, one can find half-a-dozen publications from an 80-year span agreeing about almost anything.
      Researchers begin by surveying what has been done in their field, what the state of the art is, and that cannot be judged by a random handful references.

  12. dankegel said

    It would be hard to count all the articles about warts! Let’s just focus on oral warts in dogs for the moment. We’ll have to do without the luxury of a meta-analysis, but that’s ok,
    it’s fun reading the individual papers. There is at least a survey of research up to about 1999:

    http://researchrepository.murdoch.edu.au/6227/1/canine_papillomavirus.pdf
    Nicholls, P.K. and Stanley, M.A. (1999)
    Canine papillomavirus-A Centenary review.
    Journal of Comparative Pathology, 120 (3). pp. 219-233.

    Two of the earliest references it lists are both in
    Journal of Comparative Pathology and Therapeutics, 11.
    which is online at http://books.google.com/books?id=yRAbAQAAMAAJ

    Penberthy, J. (1898).
    Contagious warty tumours in dogs.
    Journal of Comparative Pathology and Therapeutics, 11, 363-365.
    (Several case studies suggesting that warts are contagious in dogs.)

    M’Fadyean, J. and Hobday, F. (1898).
    Note on the experimental transmission of warts in the dog.
    Journal of Comparative Pathology and Therapeutics, 11, 341-344.
    Excerpt:
    “On August 13, 1898, the cut surface of a wart, just excised with sterile instru-
    ments from the buccal mucous membrane of a young foxhound, in which
    the lining membrane of the mouth was absolutely covered with growths,
    was used to rah a small area of abraded mucous membrane on the right
    upper lip of each of three dogs — a pug and two fox-terriers. No-
    thing was noticed for about a month, when in each of the terriers
    a very small growth began to show itself.
    After six weeks a minute elevation was observed on the scari-
    fied spot on the pug. In the two fox-terriers
    the papillomata continued to increase until
    about June 10, when they had attained an
    appearance identical with those of the natural warts in the
    mouth of the foxhound…
    These experiments conclusively proved that the common papilloma
    of the dog’s mouth are transmissible, and they support the clinical
    evidence in favour of contagion being the common cause of such growths. ”

    A more modern demonstration of transmissibility given in the survey was

    http://www.ncbi.nlm.nih.gov/pubmed/13809074
    http://cancerres.aacrjournals.org/content/19/11/1188.full.pdf+html
    Cancer Res. 1959 Dec;19:1188-95.
    Canine oral papillomatosis. I. Virus assay and observations on the various stages of the experimental infection.
    CHAMBERS VC, EVANS CA.
    Excerpts:

    “MATERIALS AND METHODS
    The strain of canine oral papilloma virus used in most of
    this work originated from papillomas removed from a dog in
    Seattle, Washington, in April, 1955, and is known as strain
    N-211. The papillomas were collected in 50 per cent buffered
    glycerin in Locke’s solution and were stored at 4 ~ C. Stock
    virus was prepared from the original material and from later
    passages in the following manner: Glycerinated papillomas
    were washed 2 times in Locke’s solution, minced with scissors,
    and ground in a sterile mortar; sufficient Locke’s solution was
    added to make a 10 per cent suspension. The suspension was
    centrifuged at 2000 r.p.m, for 10 minutes. The supernatant
    fluid was collected and stored at -25 ~ C. to -40 ~ C. for use as
    stock virus. Stock virus preparations derived from papillomas
    of the first five passages were used in the experiments to be
    described. …
    Canine oral papilloma infection was produced by applying
    the stock virus to the scarified oral mucosa or by injecting it
    into the mucosa of young dogs. After inoculation the dogs
    were routinely examined at intervals of 1 week.

    The number, size, and appearance of papillomas in the oral mucosa depended upon the
    concentration of virus in the inoculum, the method of inoculation, the number of sites inoculated, and the time at which the observations were made.
    The application of l0 -1 to l0 -3 dilutions of N-211 strain of virus to a scarified area usually
    produced many papillomas which were first observed as small individual growths. Many addi-
    tional tumors frequently appeared during the following week. …
    SUMMARY
    Multiple simultaneous injections of graded doses of virus into the oral mucosa of young dogs
    were found to give useful assays of canine oral papilloma virus. The Infective Doses of a typical
    stock virus preparation injected in a volume of 0.05 to 0.1 ml. was approximately 10^-5.
    A study of the experimentally produced disease revealed incubation periods of 4-8 weeks in 89 of 90 dogs. The growth period, extending from the first apperance of papillomas to the first sign
    of regression, ranged from 4-8 weeks in 37 of 57 dogs and 9-21 weeks in twelve dogs.
    All tumors in each dog regressed simultaneously. …”

    And one rather recent paper it points to goes even further, and shows that
    inactivating the extract with formalin creates a vaccine that prevents transmission:

    http://www.ncbi.nlm.nih.gov/pubmed/7734063
    Pathobiology. 1994;62(4):194-8.
    A formalin-inactivated vaccine protects against mucosal papillomavirus infection: a canine model.
    Bell JA, Sundberg JP, Ghim SJ, Newsome J, Jenson AB, Schlegel R.
    Abstract
    “A formalin-inactivated canine oral papilloma homogenate was used as a vaccine to prevent infection by the oncogenic, mucosotropic canine oral papillomavirus (COPV) in beagle dogs. Twenty-six dogs received 2 doses of phosphate-buffered saline intradermally and 99 dogs received 2 doses of the inactivated vaccine. One month after the second dose all dogs were challenged with infectious COPV by scarification of the oral mucosa. All of the control dogs developed papillomas by 6-8 weeks after challenge while none of the vaccinated dogs did. This vaccine has been used successfully in approximately 60,000 line bred beagles with no untoward effects and with long-lasting protection. These data demonstrate that a systemically administered, formalin-inactivated vaccine can protect against mucosal infection by COPV and suggest approaches for the development of human papillomavirus vaccines.”

    I could cite more, but I await your comments first. It seems that the basic technique
    of demonstrating transmission (scarification followed by application of some extract
    of a wart from another dog) was established quite early, and refined over the years by
    numourous independent groups.

    So, having seen this evidence, do you agree that oral warts in dogs are transmissible?

    • Henry Bauer said

      dankegel:
      I suppose the problem is that
      1. I don’t have the interest to look into the literature myself, and I can’t trust any collection of individual papers selected by an interested party as being representative. You haven’t offered reviews or meta-analyses.
      2. I don’t see the relevance. The point you disagreed with me on is whether Gardasil and Cervarix have been proven to be of benefit. I cited official figures about damage sufficient to qualify for compensation. But there are no data showing that incidence of cervical cancer has been decreased. All that has been shown is that the vaccines protect against some strains of HPV, but it’s never been shown that HPV causes anything, because all the evidence consists merely of correlations.
      I repeat: Official claims note that HPV infections strike about 80% of women at some time or other, so roughly about 130 million American women. The incidence of cervical cancer is about one thousand times less than that, say 13,000 annually — and that’s been decreasing for 20 years.
      Furthermore, in 1993 the Centers for Disease Control pronounced HIV as the cause of cervical cancer, also on the basis of correlations, and they have not rescinded that. Are you claiming that CDC is wrong about HIV and cervical cancer?

      No amount of stuff about oral warts in dogs seems to me a plausible lead-in to the issue of the risk-benefit ratio of Gardasil and Cervarix in humans.
      As to genital warts, since they are not harmful it seems foolish to seek to guiard against them with vaccines that have been acknowledged to cause damage sometimes as great as death.

  13. dankegel said

    Here’s how oral warts being transmissible is relevant to the risk-benefit ratio of Gardasil in humans:

    Gardasil has been reported to protect against genital warts.
    You claim that warts are not transmissible, thus no vaccine can prevent them. Warts have a societal cost, therefore, whether Gardasil is of benefit depends in part on whether it does in fact prevent warts.

    Furthermore, once it is established that warts are transmissible by virus, we can then go on to look at how the virus works, and how it may be related to cancer. But we can’t do that without plowing through the wart question first.

    So the question of whether warts are transmissible has a direct bearing, in two ways, on whether Gardasil has any benefit.

    Now, I’m not asking you to read a lot of papers. I gave you direct links to a series of four papers which describe experiments which demonstrate with increasing clarity that warts (specifically, oral warts in dogs) are transmissible. I even excerpted key parts of the papers for you so you can see why they might be worth reading.

    You *did* ask for references to support the idea that warts are transmissible. It seems rather poor sport to ask for them, and then not read them once I dig them up.

    • Henry Bauer said

      dankegel:

      “Gardasil has been reported to protect against genital warts” — ANYTHING can be reported. Correlations don’t prove causation.

      “You claim that warts are not transmissible”
      I didn’t. The post(s) you commented on focused on cervical cancer.

      “Furthermore, once it is established that warts are transmissible by virus, we can then go on to look at how the virus works, and how it may be related to cancer.”
      One can speculate endlessly about how things “may be”, but the proof is in the actual facts, which can’t be known until there’s been a long and large enough clinical trial with some treated by Gardasil and others not.

      “You *did* ask for references to support the idea that warts are transmissible. It seems rather poor sport to ask for them, and then not read them once I dig them up.”
      I asked for meta-analyses or reviews, not more individual articles.

      • Dan Kegel said

        Let’s look back at comment #8 above and your reply:

        I wrote:
        OK, let’s table cancer for the moment, and move on to warts. Do you agree that [1] it’s been established that warts are generally a transmissible disease, and that [2] the disease agent is a virus, and that, [3] in humans, these viri are generally known as HPV?

        You replied:
        [1] No. Perhaps you are not familiar with Lewis Thomas, “On Warts”?
        [2] So, ditto, again NO.
        [3] And once more, tritto, NO.
        Cite the establishing sources

        So, yes, you claimed that warts are not transmissible, and you asked for the establishing sources, not meta-analyses.

        I have not found any meta-analysis yet. I did find one review, and gave you a link to it, plus to the four most germane papers it links to.

        I really do think it’s worth establishing some common ground, and the first place our opinions seem to diverge is on the question of whether warts are transmissible. Let’s look at that together and then move forward to the bigger questions.

      • Henry Bauer said

        Dan Kegel:
        “So, yes, you claimed that warts are not transmissible”
        No. I didn’t accept that they have been shown to be transmissible, which is not the same thing.

        “you asked for the establishing sources, not meta-analyses”.
        Let’s not get into rhetorical devices and distractions. It has been obvious to me for a long time (and is in the book you haven’t read) that no research articles constitute proof in themselves, so reviews and meta-analyses are the only things worth looking at. I presumed that you also understand thees elementary things about science when I asked for sources that might be potentially convincing.

        But nowhere have I said that I regard warts as pertinent to my posts about Gardasil and cervical cancer. My posts cite official numbers about deaths and damage from Gardasil and about the negligible incidence of cervical cancer by comparison to incidence of HPV. You haven’t raised any points against the clear implications of those data.

        “I really do think it’s worth establishing some common ground, and the first place our opinions seem to diverge is on the question of whether warts are transmissible. Let’s look at that together and then move forward to the bigger questions.”
        Maybe you think so, but I don’t. I just don’t see any relevance of warts. Even if you could convince me that they are transmissible; and even if you could convince me that they are caused by HPV, it would not establish that HPV causes cervical cancer.

  14. dankegel said

    You write “reviews and meta-analyses are the only things worth looking at”. I did point you at one review, “Canine papillomavirus-A Centenary review.”, but alas, you did not think it was worth looking at.

    I agree with you that Cochrane-style, low-bias meta-analyses are great at filtering out crap studies.
    Picking those as your standard for whether something has been shown to be true is great; I wish more people would do it.

    For instance, you might apply that standard to whether Gardasil causes death. To my knowledge, there has been no solid paper, let alone meta-analysis, that shows that it does. (I haven’t seen any of the trial transcripts from vaccine court, but I doubt they rise to the standard of reproducible research.) Thus I don’t accept that harm has been shown, and if you applied your standard uniformly, you wouldn’t, either.

    You wrote “about the negligible incidence of cervical cancer by comparison to incidence of HPV”. I believe the official statistics are that over a lifetime, about 1 in 140 American women will be diagnosed with cervical cancer, and approximately 80% of American women will be infected with HPV. That’s a 1:120 ratio, which does not seem negligible to me.

    Ioannidis’s paper, “Why most published research findings are false” (http://www.ncbi.nlm.nih.gov/pubmed/16060722) lays out several reasons why there are so many crap research papers out there. In particular, it says
    “The smaller the effect sizes in a scientific field, the less likely the research findings are to be true”,
    “The greater the number and the lesser the selection of tested relationships in a scientific field, the less likely the research findings are to be true”, and
    “The greater the flexibility in designs, definitions, outcomes, and analytical modes in a scientific field, the less likely the research findings are to be true.”

    In the case of whether canine oral warts are transmissible, the effect size is large (and plainly visible to the naked eye), there is only one binary relationship tested (whether an extract of wart can cause warts), and there is essentially only one study design and outcome in the papers cited: you put wart juice on a scarified area of a dog’s lip, and see if warts arise within about 90 days. The main variation is how the wart juice is purified. I would wager that Ionnidis would think this not an area ripe for fraud, bias, or data mining.

    The path from “warts are transmissible” to “warts are caused by papillomavirus” to “the E6 gene of the papilloma virus alone can immortalize cells” to “HPV causes cervical cancer” starts with, well, the question of whether warts are transmissible. I’m starting to suspect that you are hesitating to take that first step because you don’t like where it leads.

    • Henry Bauer said

      dankegel:
      “The path from “warts are transmissible” to “warts are caused by papillomavirus” to “the E6 gene of the papilloma virus alone can immortalize cells” to “HPV causes cervical cancer” starts with, well, the question of whether warts are transmissible. I’m starting to suspect that you are hesitating to take that first step because you don’t like where it leads.”

      No. The only way to prove that HPV causes cervical cancer would be to infect with HPV and observe the results. Can’t do that with human beings, of course, so the next best thing would be a sufficiently large and long prospective trial with controls, presumably Gardasil-treated ones, and eventually compare the rates of incidence.

      All you’re proposing is a long chain of inferences about possibility which can never constitute proof. I don’t take your first step because it’s irrelevant.

      I won’t address all your other numerous other logical or factual fallacies, I’ll just repeat: Something associated with a disease in 1 out of 120 instances does not seem like a cause, certainly not a sufficient cause. Why did the 119 NOT get cancer if they harbor the cancer-causing virus?

      When a government agency disburses $7 million to compensate for damage, it seems highly likely to me that they applied the equivalent of a meta-analysis, or quite direct evidence, before making the payments.

      You have consistently back-pedaled from the claims in my blog posts. Trying to start with warts is an impicit admission that you can’t gainsay the data and conclusions in my posts. PLease peddle your red herrings to someone else.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s