Wednesday, October 5, 2011

That's just dumb.

Via Rorschach's blog I learned of a new study about HPV-associated oropharyngeal cancers. Apparently, the story's all over the place. Some of the stories are decent, if insufficiently critical. But several present the study in ways that are just...dumb.

In a typical example, the headline from Medical News Today reads "HPV Linked Oropharyngeal Cancer Rates Rise Dramatically." The first paragraph:
In the 1980s just over 16% of patients with oropharyngeal cancers tested positive to HPV, compared to over 70% during the last decade, researchers reported in the Journal of Clinical Oncology. The authors add that if the rise in incidence continues at its present pace, the incidence of oropharyngeal cancers will overtake that of cervical cancer.
Oh, good grief. Regarding the latter claim, whether it will overtake cervical cancer is irrelevant. They're both rare in the US, and cervical cancer is rare and its rates have declined dramatically due to screening which is extremely effective in prevention. That trend will continue as long as women can get and follow up on Pap tests, are educated about the need for regular screening, and are not led to a false sense of protection by the hype about the vaccine.

The only information relevant to the rise in HPV-associated oropharyngeal cancers - aside from where and over what time period you're talking about, which I'd think would be obvious but seems to be lacking in several articles - is the actual incidence of HPV-associated oropharyngeal cancers. The MNT and other articles imply that if the percentage of HPV-associated OP cancers has increased dramatically, this reflects a dramatic increase in the incidence of HPV-associated OP cancers. But that's just dumb. OP cancers are also caused by smoking and alcohol use, so changes in smoking or drinking rates leading to a lower incidence of OP cancers caused by them will change the percentage of OP cases caused by HPV regardless of changes in their actual incidence. (Indeed, the incidence of HPV-associated cancers could decline and its percentage still increase: Say there were 100 cases of OP cancer one year, and 50 each are caused by smoking and HPV. The next year the number of cases caused by smoking drops to 10, while the number of cases caused by HPV drops to 40, for a total of 50 overall. The percentage of HPV-associated cases will have risen from 50% to 80%.)

We know that these changing percentages reflect to some extent declining smoking rates in the US and the resultant decline in smoking-related OP cancers. But the authors of the study (I don't unfortunately have access, though the accompanying editorial is available, as is the supplementary data) did find an increase in the incidence of HPV-associated OP cancers over the period studied. The presentation of this increase in the media reports has also been fairly misleading, though. The "225% increase" is repeatedly reported, but it's harder to see the "dramatic" rise when you consider that the increase is from .8 cases per 100,000 to 2.6 cases per 100,000 over a 16-year period. I don't have the knowledge to fully put this in perspective, but the actual numbers are significantly less dramatic than the trumpeted percentage increase, and it's still rare.

But of course even if there are "just" several thousand cases of HPV-associated OP cancer per year in the US, and even if it's more treatable than the other kind (the most interesting aspect of the article, from what I was able to gather from the editorial), that's several thousand too many, especially if the increase continues. I can understand people suggesting a broad HPV vaccination of men as well as women.

The HPV vaccines are approved in the US for boys 9-26 for the prevention of genital warts and some cancers, though they aren't currently recommended for boys by the CDC. Some have said that this study should push the CDC in the direction of recommending the vaccines for boys as well, or even mandating them. The problem with this is that there is no evidence that the vaccine is effective against OP cancer. There's no evidence against it, as far as I know - there's no evidence at all. Merck and GSK haven't done the research, and they're apparently not going to. I was suspicious about Merck's dropping it, but the argument made in this article is plausible: there's no equivalent to Pap smears for OP precancers, so any clinical tests of vaccine effectiveness would be expensive and take a good deal of time, and tests that screened only for oral HPV infection wouldn't necessarily convince the FDA of effectiveness against precancerous lesions.

But the fact remains that the evidence doesn't exist, even if a plausible rationale does. So "I can't see the harm of [mandatory] vaccination of boys for OP cancers" remains problematic. For families making the decision, the financial costs are known, the long-term risks aren't, and the effectiveness (in this case, either short- or long-term) isn't. As a public health measure, the same considerations apply, and additionally there are concerns about the best use of resources in terms of this disease and other related diseases. If I were being cynical, I would say that Merck is in a great position here. They're not going to spend the money to find out if the vaccine is effective in preventing OP cancers, but people will push for public funds to be used for the purpose. Meanwhile, plenty of people are urging, on the basis of this research, that the vaccine be given to or mandated for as many boys as possible, and it's already approved for boys, so sales will increase.

Even if I'm not being cynical, I would simply argue that the decisions need to be made on the basis of the available evidence and on a cost and risk vs. benefit ratio.

The same disclaimer I attached to my earlier post about this applies to this one: I'm not trained in medicine, and some of my facts or analysis could well be wrong. If they are, I would appreciate corrections. A note that shouldn't be necessary: I'm not anti-vaccine, as anyone familiar with my writing should well know, and anyone who does a search for "health" on this blog can easily discover. I'm anti-thinking-every-disease-and-vaccine-profile-is-the-same, because that's just dumb.

No comments:

Post a Comment