CROI: Smoking increases cancer-associated HPV viral load

This article is more than 17 years old. Click here for more recent articles on this topic

Smoking increases the rate of viral reproduction in gay men who have anal infections with cancer-causing variants of the human papilloma virus (HPV), a German study has found.

The association between smoking and viral load was particularly strong in men who had not yet developed precancerous changes (dysplasia) in the anal or rectal epithelium (skin or mucosa), suggesting that smoking may possibly speed the development of so-called squamous cell intraepithelial lesions (SIL). High HPV viral loads have previously been shown to be associated with an increased risk of anal cancer.

Physicians took 1,806 anal swabs from 267 HIV-positive gay men during an anal dysplasia screening programme in Cologne and Bochum, Germany.


human papilloma virus (HPV)

Some strains of this virus cause warts, including genital and anal warts. Other strains are responsible for cervical cancer, anal cancer and some cancers of the penis, vagina, vulva, urethra, tongue and tonsils.

squamous intraepithelial lesion (SIL)

This term is used to describe the detection of abnormal cells that have been ‘transformed’ by HPV into a possibly pre-cancerous state. According to the degree of cell change this will be called low-grade or high-grade SIL (LSIL or HSIL). If SIL is detected, a colposcopy will usually be ordered.


The result of a statistical test which tells us whether the results of a study are likely to be due to chance and would not be confirmed if the study was repeated. All p-values are between 0 and 1; the most reliable studies have p-values very close to 0. A p-value of 0.001 means that there is a 1 in 1000 probability that the results are due to chance and do not reflect a real difference. A p-value of 0.05 means there is a 1 in 20 probability that the results are due to chance. When a p-value is 0.05 or below, the result is considered to be ‘statistically significant’. Confidence intervals give similar information to p-values but are easier to interpret. 


Cells that look abnormal under a microscope but are not cancer.


The last part of the large intestine just above the anus.

One hundred and thirty-seven on the men (51.3%) were smokers, and patients’ self-declared smoking levels correlated well with levels of cotinine, a metabolite of nicotine, in the blood.

Levels of infection with HPV, cancer-associated types of HPV, and the two most common cancer-associated types, 16 and 18, were significantly higher in men who smoked (p=0.001). In total 8% of non smokers and 87.5% of smokers had HPV infection, 70.1% and 82.7% respectively had infections with high-risk varieties of HPV, and 48% and 54.2% had infections with HPV type 16, the most common cancer-causing type.

SIL and particularly high-grade SIL (HSIL) were significantly more common in smokers. Thirty-five per cent of smokers had SIL compared with 26.6% of non smokers, and 17.6% of smokers had HSIL compared with 4.4% of non-smokers.

Results from swabs taken from inside the rectum (intra-anal) and around the anus (perianal) were compared and it was found that intra-anal swabs detected consistently higher rates of HPV infection and SIL – so the following results are for intra-anal swabs only.

Viral loads for HPV subtypes 16 and 18 were determined by a PCR viral load assay and expressed as the number of copies of virus relative to the number of copies of an activated anti-HPV antibody gene. As antibodies typically limit the spread of HPV infection, a viral load greatly in excess of antibody level indicates an infection that the immune system is failing to contain. HPV relative viral loads increased with the level of anal dysplasia: median viral loads were 0.3 in normal swabs, 2.1 in low-grade SIL (LSIL) and 106.8 in HSIL.

Median and mean intra-anal relative viral load levels of HPV type 16 were 0.5 and 62 in non-smokers (p=0.0023) and 4 and 395 in smokers (p=0.0017). The fact that the mean viral load was consistently higher than the median indicates that the distribution of viral loads was skewed; in other words a minority of patients had exceptionally high HPV viral loads.

The difference in viral load between smokers and non-smokers was greatest in those with normal cytology (cell swab) results. The difference did not reach statistical significance in patients with LSIL and HSIL. But in patients with normal swabs, median and mean relative viral load results were 0.09 and 1.8 in non-smokers and 0.7 and 108.6 in smokers (p=0.004), indicating that HPV viral reproduction is greatly accelerated in a minority of smokers. This in turn implies faster progression to SIL and possibly cancer.

This difference was still significant when adjusting for age, CD4 count, HIV viral load, AIDS diagnosis and whether patients were on antiretroviral drugs.

Results were similar for HPV type 18, the second most common cancer-causing variant.

Median and mean HPV-18 relative viral loads were 0.3 and 4.6 in non-smokers and 0.8 and 81.5 in smokers, and for those with normal swabs (i.e. no SIL), median and mean relative viral loads were 0.07 and 1.2 in non-smokers and 0.3 and 10.2 in smokers.

The investigators comment that since high HPV16 loads have previously been shown to be associated with an increased risk for anogenital cancers, gay men co-infected with HIV and HPV should be encouraged to refrain from smoking.


Brockmeyer N et al. Elevated anal HPV16 DNA loads in HIV+ men who smoke. Fourteenth Conference on Retroviruses and Opportunistic Infections, Los Angeles, abstract 872, 2007.