Computed tomography (CT) detected lung cancer in nine of 442 HIV-positive smokers at least 40 years old who had chest CT in a nationwide French study. CT helped spot these cancers at younger ages than the targeted range in the general population, and two-thirds of the cancers were stage I or II.
Lung cancer screening with low-dose chest CT lowers mortality in general-population smokers 55 to 74 years old. Among people with HIV infection, lung cancer has the highest incidence and mortality of all non-AIDS malignancies. Smoking remains the main cause of lung cancer in the general population and in people with HIV infection. Until the French study, the potential value of CT screening for chest cancer remained uncertain in people with HIV, who have high rates of other pulmonary morbidities that may lead to false-positive nodules and increased use of potentially harmful invasive diagnostic procedures.
To address these issues, French investigators enrolled HIV-positive people at least 40 years old with a nadir CD4+ count <350 cells/mm3. All participants had smoked at least 20 pack-years but may have quit in the past three years. The study excluded people with current active cancer, AIDS illness or lung infection in the past two months. Everyone had at least two years of follow-up after a single low- to moderate-dose chest CT. A positive image led to a standardized diagnostic workup. Clinicians diagnosed lung cancer by histological analysis of biopsy specimens. The primary study outcome was the number of histologically proven lung cancers diagnosed by the single chest CT.
Between February 2011 and June 2012, 442 people had a chest CT. Median age stood at 49.8 years; 84% of participants were men; 98% were taking combination antiretroviral therapy (ART); 90% had a viral load below 50 copies/mL; and median smoking experience was 30 pack-years. Median nadir and current CD4+ counts were 168 and 574 cells/mm3.
Through a median 24.4 months of follow-up after CT, clinicians diagnosed 10 lung cancers. CTs in nine of these 10 people showed positive nodules; eight of these nine had histologically proven lung cancer, and the ninth had a highly probable lung cancer indicated by a 10-mm nodule. Of the nine CT-detected lung cancers, four were stage IA, one stage IB, one stage IIA and three stage IV. Two of the three people with stage IV cancer had delayed diagnostic procedures after CT detection of a positive nodule.
Prevalence of CT-detected lung cancer in this population stood at 2.03% (95% confidence interval [CI] 0.90 to 3.80). The researchers calculated that 49 patients (95% CI 26 to 111) would have to undergo CT to detect one case of lung cancer. CT screening led to only 18 invasive procedures in 15 patients. These procedures caused no serious adverse events and permitted alternative histologic diagnoses in four cases.
Among 404 participants who made the planned two-year follow-up visit, 402 had complete smoking data, including 368 (91% of 404) who smoked at the baseline visit. Of those 368 baseline smokers, 74 (20%) had stopped smoking at their last visit. Among active smokers at baseline and the last visit, 57 (19%) had a positive baseline CT, compared with 19 people (26%) who had stopped smoking at the last visit, but that difference lacked statistical significance (P = .23).
The researchers stress that eight of 10 diagnosed lung cancers in this HIV population occurred in people younger than 55 years old. These cancers would have been missed if CT had not been performed until age 55 in accordance with national guidelines. Performing CT in people older than 45 would not have resulted in any missed lung cancers.
The French team concludes that "lung cancer screening [with CT] should benefit [persons living with HIV] younger than 55 years who smoke and who have a history of significant immunodeficiency, possibly starting as early as 45 years."