To the Editor:

We read with interest the article by Herida et al,¹ entitled “Incidence of Non–AIDS-Defining Cancers Before and During the Highly Active Antiretroviral Therapy Era in a Cohort of Human Immunodeficiency Virus-Infected Patients,” published in the September 15, 2003 issue of the Journal of Clinical Oncology.

It described very clearly the changing incidence of non-AIDS defining cancers in the pre–and post–highly active antiretroviral (HAART) eras. One of its findings was that it showed a two-fold increase in the incidence of HIV-related lung cancer in men in the HAART era (standard incidence ratio [SIR], 2.05; 95% CI, 1.62 to 2.56). This figure was significantly higher than that seen in the matched HIV-negative population (SIR, 2.12; 95% CI, 1.76 to 2.65). The incidence of HIV-related lung cancer also increased significantly for women over this period (SIR, 6.59; 95% CI, 3.40 to 11.52). The authors argued that these results were predominantly as a result of the excess number of cigarettes smoked in the HIV-positive population, compared to HIV-negatives. They also suggested that the observed rise in females was because of a rise in the incidence of lung cancer as a whole in females, which was not accounted for in the control cohort, as it did not include patients diagnosed after 1996 (for which data is not yet available).

Recently published data from our group also found an increase in the incidence of HIV-related lung cancer since the introduction of HAART (SIR, 8.93; 95% CI, 4.92 to 19.98).² Together these results suggest that this fatal disease may be becoming more prevalent in people with HIV-1 infection. However, we would argue that the causes of this rise may be different from those suggested by Herida et al.¹

Although patients with HIV smoke more cigarettes than the HIV-negative population,³ it has been previously shown that this alone cannot account for the excess of lung cancers seen in HIV. This pre-HAART study assumed the whole HIV-positive population smoked, and still found an excess risk of lung cancer.⁴ Moreover, adenocarcinoma, which is the histologic subtype least strongly associated with smoking cigarettes, is most commonly seen in this population.^2,6 Additionally, there is no data to suggest that the incidence of smoking has gone up since the introduction of HAART, which would be necessary if this was the cause of the increased incidence of the cancer. In our study there was no difference between the number of pack years smoked in the pre- and post-HAART eras. It is therefore difficult to suggest smoking is responsible for this rise seen.

As people age, the incidence of lung cancers rises sharply.⁵ The study by Herida et al¹ showed the men were marginally older in the post-HAART era (3 years), and this may account for some of the increase. This finding contradicted our work, which showed there was no difference between the ages of the patients with lung cancer in pre- and post-HAART eras. The major difference between the two populations in our study was the duration of HIV before the development of cancer (2 v 11 years). We suggest that this prolonged immune suppression may be predisposing individuals to lung cancers, especially adenocarcinomas, and therefore incidence may continue to rise as people with HIV live longer.

The last point we would like to make surrounds the incidence of HIV-related lung cancer in women. This is not the first study to find an increased incidence of the disease in females.⁷ Therefore, speculating that the rise in this disease will match the as yet unavailable epidemiologic data for the whole population may be hazardous. This is especially true when one considers the strong body of evidence showing this disease to be a serious and increasing problem in men as well as women.^7-10

Authors' Disclosures of Potential Conflicts of Interest

The authors indicated no potential conflicts of interest.