| |
HIV-related lung cancer in the era of highly active antiretroviral therapy
|
| |
| |
The data that we present here confirm the increased incidence of lung cancer in a sizeable cohort, and suggest that this risk has increased since the introduction of HAARTÉÉ there was a longer time interval [compared to patients who developed lung cancer in the pre-HAART era] between testing HIV positive and developing lung cancer. We speculate that it is prolonged exposure to moderate immunosuppression [HAART does nor completely restore the immune system] that results in reduced immune surveillance and the development of tumours. The cancers were predominantly adenocarcinomas, which require less genetic damage to develop compared with other primary lung tumours. A link between HIV-related lung cancer and increased microsatellite alterations that reflect genomic instability has been reportedÉ.
Editorial note from Jules Levin: people with HIV appear to smoke cigarettes more. Smoking cigarettes is associated with increased risk for heart disease and lung cancer. Considering that HIV-infected individuals may be at greater risk for heart disease, diabetes, and perhaps lung cancer, as this study finds, it would be a great contribution to your health to stop smoking.
Early in the AIDS epidemic it was estimated that up to 36% of patients with HIV develop a malignancy during the course of their disease, and Kaposi's sarcoma (KS) and non-Hodgkin's lymphoma (NHL) accounted for 95% of these
cancers. However, large prospective epidemiological studies have found that other non-AIDS-defining tumours also occur more frequently in HIV-seropositive individuals. These tumours include Hodgkin's disease, testicular seminoma,
head and neck tumours, and anal, lip and lung cancers. Highly active antiretroviral therapy (HAART) was introduced in the mid-1990s, and resulted in a dramatic decrease in mortality and morbidity from HIV. This is mainly
the result of a reduction in opportunistic infections; however, antiretroviral drugs have also led to a decrease in the incidence of KS and NHL. As these opportunistic neoplasms decrease and patients live longer, it has been
speculated that the incidence of other malignancies that are less directly related to immunosuppression may increase.
The incidence of lung cancer in individuals with HIV has been reported to be higher than, equal to, and lower than that in the general population. Large epidemiological studies have suggested that lung cancer does occur
more frequently, but often this link has been attributed to increased smoking rates in the HIV-positive population. A number of other confounding variables have been postulated to account for the increased risk, including diminished tumour immune surveillance as a result of chronic immunosuppression associated with HIV, opportunistic lung infections, intravenous drug use, and the increasing age of HIV patients in the era of HAART. In both HIV-negative and -positive populations the incidence of lung cancer increases with age.
Consequently, as individuals with HIV survive longer, the combination of increasing age and chronic immunosuppression may result in an increasing incidence of lung cancer.
The outcome in HIV-related lung cancer patients was considerably worse than their HIV-negative counterparts before the introduction of HAART. Little is known of the impact of HAART on the incidence of lung cancer and the
course of the disease. This study used a large, carefully documented, single institution cohort to address these issues.
STUDY
Objectives: To address the impact of highly active antiretroviral therapy (HAART) on the incidence and outcome of patients with HIV-related lung cancer.
Design and subjects: Patients with HIV-related lung cancer were identified from a prospective HIV data base of 8400 patients diagnosed between 1986 and 2001. Patients diagnosed with HIV-related lung cancer before 1996 were in the pre-HAART cohort whereas the remainder were in the post-HAART cohort.
Methods: The incidence of HIV-related lung cancer in the pre- and post-HAART cohorts was compared with the age and sex-matched population of south east England. Clinicopathological features, treatments and outcomes were also
recorded.
Results: The incidence of HIV-related lung cancer increased from 0.8 (95% CI 0.2-3.2)/105 patient-years follow-up in the pre-HAART era to 6.7 (95% CI 3.1-13.9)/105 patient-years follow-up in the post-HAART era. The age and
sex-matched incidence of lung cancer in south east England was 0.75 (95% CI 0.63-0.87)/105 patient-years, suggesting that HIV-related lung cancer only occurred more frequently in the post-HAART era (relative risk 8.93, 95% CI
4.92-19.98). The patient characteristics and outcomes were similar in the pre- and post-HAART eras, although the time interval between testing HIV positive and developing HIV-related lung cancer was longer in post-HAART patients.
In total, 11 patients (10 homosexual men, one female intravenous drug user) with HIV-related lung cancer were identified. The histopathological classification of the tumours were five adenocarcinomas, four squamous cell
carcinomas, one bronchoalveolar carcinoma and one small cell lung cancer. Seven patients had metastatic disease (including one with small cell lung cancer) and four had stage IIIB disease. Nine of the 11 patients developed lung cancer in the post-HAART (1997-2001) era, six of these patents commenced
HAART before the development of lung cancer, and the median duration of HAART was 2 years (range 2-6). Three patients who developed lung cancer in the HAART era were not receiving antiretroviral therapy as their immune function was well preserved. A wide range of antiretroviral agent combinations were used in line with the current hospital guidelines and resistance testing.
From the article it appeared that all 11 patients smoked cigarettes, although I donÕt think the article was clear about if they all were smokers. The article merely said that the median smoking history for the 11 patients was 35-40 packs per year.
Of the 11 patients in the study, 10 received active treatment (three radiotherapy to the lung primary, one radiotherapy for cranial metastases and six systemic chemotherapy) and one best supportive care. All of the patients treated with chemotherapy were diagnosed in the in the post-HAART era. Half of the patients treated with chemotherapy achieved a partial response, whereas the remaining three all had progressive disease on chemotherapy. Two of the three patients treated with pulmonary irradiation achieved partial responses, whereas the remaining patient had stable disease. The
median progression-free survivals for chemotherapy and radiotherapy were 2.5 months (range 1-13) and 3 months (range 0-3), respectively. Similarly, the overall median survivals were 4 months (range 2-15+) for chemotherapy and 5
months (range 3-5) for radiotherapy. Response rates, progression-free survival and overall survival were no better in the post-HAART cohort than in the earlier patients.
In this study HIV-related lung cancer occurred more frequently in the post HAART era, when compared with the HIV-negative population. Unfortunately, the outcome of these patients remains poor despite HAART.
Early epidemiological studies failed to demonstrate an association between HIV infection and lung cancer; however, this may have been due to the short life expectancy in the early years of the HIV pandemic. Subsequently, larger more complete studies involving over 300 000 patients have shown a definite link (RR 4.5), although this risk has been attributed by some authors to the high levels of smoking in HIV-positive populations. It should be noted that although HIV patients smoked more cigarettes per day than the general population, the
long-term exposure to tobacco was lower, as they presented with lung cancer at a much younger age.
The data that we present here confirm the increased incidence of lung cancer in a sizeable cohort, and suggest that this risk has increased since the introduction of HAART. The data also describe a prolonged interval from the acquisition of HIV to the diagnosis of lung cancer (a median of 11 years in the post-HAART era), which may explain the failure to identify an increased risk in pre-HAART studies when the life expectancy of patients with HIV infection was limited. The patients in the HAART era presented with lung cancer at a similar age, smoked as many cigarettes, and had a similar level of immunosuppression when compared with the pre-HAART studies.
This is the first time that an increase in the incidence of HIV-related lung cancer has been shown when comparing a pre- and post-HAART population. It is known that as HIV patients become older their risk of lung cancer increases;
however, the patients in this study were of a similar age at presentation to those published in previous pre-HAART series. There are two main differences between our population and those in the pre-HAART era. First, there was a
widespread use of antiretroviral therapy in our patients, and second, there was a longer time interval between testing HIV positive and developing lung cancer. We speculate that it is prolonged exposure to moderate immunosuppression
that results in reduced immune surveillance and the development of tumours. The cancers were predominantly adenocarcinomas, which require less genetic damage to develop compared with other primary lung tumours. A link
between HIV-related lung cancer and increased microsatellite alterations that reflect genomic instability has been reported. It is possible that as patients live longer in the HAART era, prolonged exposure to genomic instability predisposes
to the development of solid malignancies especially adenocarcinomas of the lung.
In conclusion, HIV-related lung cancer continues to have a poor outcome despite the introduction of antiretroviral therapy. In the era of HAART, the incidence of HIV-related lung cancer has increased in our cohort, and this may be a consequence of the marked fall in opportunistic infections and AIDS-defining tumours, coupled to improved life expectancy.
AIDS 2003; 17(3):371-375
Mark Bower; Tom Powles; Mark Nelson; Pallav Shah; Sarah Cox; Sundhiya Mandelia; Brian Gazzard
|
|
| |
| |
|
|
|
