The Italian Cancer and AIDS Registries Linkage Study was used to compare the incidence of lung cancer in persons with HIV/AIDS (PWA) before and after the introduction of highly active antiretroviral therapy (HAART). Twenty-one lung cancers were identified in 12 104 PWA. Incidence rates of lung cancer in PWA exceeded those in the general population, but no increase emerged after HAART. Population-based studies with long follow-up are needed to monitor trends in non-AIDS defining cancer in PWA.
In a previous issue of this journal, Bower and colleagues  suggested that lung cancer in HIV patients may be occurring more frequently after highly active antiretroviral therapy (HAART) has become available.
We have used data of the Italian Cancer and AIDS Registries Linkage (CARL) Study [2,3] to provide additional information on this topic. The general design of our record linkage study has already been described [2,3]. Briefly, it is worth remembering that 19 independent cancer registries were active in Italy in the mid-1990s, covering a population of 12.7 million inhabitants (i.e. 23% of the total Italian population) [4,5]. Cancer registries varied both in size, ranging from approximately 190 000 to nearly 1.9 million population covered, and in the number of available years of registration [4–6]. An ad-hoc software was developed to perform a reliable record linkage procedure, according to strictly confidentiality restrictions .
People under study were aged between 15 and 69 years at the time of AIDS diagnosis, reported a legal residence in areas covered by cancer registries, had a diagnosis of AIDS after 1985, and were reported with lung cancer in periods deemed complete at both registries (i.e. in most instances through the end of 1998).
At cancer registries, lung cancers were recoded according to the International Classification of Disease, 10th revision, and they were further subdivided according to whether histological or cytological confirmation (henceforth referred to as histological confirmation) was available or diagnosis had been made otherwise (i.e. clinical, instrumental diagnosis, etc.) . Cancer registry coordinators reviewed all records on the histological type and site of cancers. All lung cancers, even in the absence of histological confirmation, are included in the study because even in the general population a relevant proportion (25%) of lung cancer is not histologically confirmed [4,5].
For each person with HIV/AIDS (PWA), the relevant time period [i.e. person-years (py) at risk] was computed between 6 months before the AIDS diagnosis and the date of cancer diagnosis or death. The period was censored 5 years after AIDS diagnosis to reduce inaccuracies from losses at follow-up. Incidence rates (IR) were calculated using observed lung cancer cases and person-years of observation in PWA. The incidence rates were based on the AIDS population in the study period and corresponding 95% confidence intervals (CI) were computed .
To evaluate time trends, IR were computed separately in two periods: (i) 1985–1996, before the introduction of HAART; (ii) 1997–1998, i.e. after the introduction of HAART.
Overall, 12 104 individuals (78% men and 22% women) were diagnosed with AIDS in areas covered by one of the 19 cancer registries for a total of 18 753 py, and 21 lung cancers were identified, 11 of whom were histologically confirmed (Table 1); four were squamous cell carcinoma, two were adenocarcinoma, and five had other histologies.
Four patients developed lung cancer in the post-HAART era (Table 1). IR of AIDS-related lung cancer was 10.7 (95% CI 6.2–17.2) per 10 000 py in 1985–1996 and 14.1 (95% CI 3.7–36.4) in 1997–1998. No statistically significant change in IR emerged after the introduction of HAART overall and in separate strata of sex, age, and exposure category (Table 1). Moreover, no difference was found with respect to histologically confirmed lung cancer.
Lung cancer incidence in PWA was also compared with the incidence in the general Italian population. Expected numbers of lung cancers were computed between 6 months before AIDS diagnosis and 5 years after AIDS diagnosis in each cancer registry from sex and age-specific (5 years) incidence [5,6] and were compared with observed numbers among PWA by means of standardized incidence ratios (SIR) and 95% CI . An overall SIR of 7.4 (95% CI 4.6–11.3) emerged and was similar before (7.3; 95% CI 4.2–11.7) and after (7.9; 95% CI 2.1–20.4) the introduction of HAART. In the same period of observation, SIR of lung cancer between three and six among PWA have already been reported in different countries [3,9–11]. The reason for such excess is not fully understood. In our data, a greater excess of lung cancer was found among intravenous drug users (23.9; 95% CI 11.9–43.0) a notoriously heavily smoking population, than in other exposure categories (4.2; 95% CI 2.0–7.7). This finding points to a role of smoking in the explanation of lung cancer excess [12,13].
Although our study is based on a period after HAART too short to allow any definitive conclusion, it shows no impact of HAART on the IR of lung cancer in Italy. An increase in lung cancer IR after HAART may derive from the prolongation of survival in PWA and the improvement of health conditions after 1996, which may allow a larger number of lung cancers to emerge or to be accurately diagnosed. It is likely that, before 1996, a non-negligible proportion of lung cancer cases might have been missed as a result of severe respiratory tract comorbidity. The problem of the misclassification of non-Hodgkin's lymphoma or Kaposi's sarcoma, when localized in the lung, must also be borne in mind.
In conclusion, improvements brought by HAART are clearly opening the door to an absolute increase in a number of chronic conditions, including different cancer sites or types. Carefully designed cohort and record linkage studies can allow the quantification of changes in cancer incidence and, possibly, the elucidation of the role of HIV and non-viral factors in the onset of cancer. For the time being, increased survival in PWA adds priority to targeted preventive strategies, smoking cessation in the first place.
1. Bower M, Powles T, Nelson M, Shah P, Cox S, Mandelia S, et al
. HIV-related lung cancer in the era of highly active antiretroviral therapy. AIDS
2. Franceschi S, Dal Maso L, Arniani S, Lo Re A, Barchielli A, Milandri C, et al
. Linkage of AIDS and cancer registries in Italy. Int J Cancer
3. Dal Maso L, Franceschi S, Polesel J, Braga C, Piselli P, Crocetti E, et al
. Risk of cancers in persons with AIDS in Italy. Br J Cancer
4. Zanetti R, Gafà L, Pannelli F, Conti E, Rosso S (editors). Cancer in Italy. Incidence data from Cancer Registries
, 3rd volume, 1993–1998. Rome: Il Pensiero Scientifico Editore; 2002.
5. Parkin DM, Whelan SL, Ferlay J, Teppo L, Thomas DB (editors). Cancer incidence in five continents
, Vol. VIII, no 155. Lyon: IARIARC Science Publishers; 2002.
6. Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J (editors). Cancer incidence in five continents
, Vol. VIII, no 143. Lyon: IARC Science Publishers; 1997.
7. Dal Maso L, Braga C, Franceschi S. Methodology used for ‘‘Software for Automated Linkage in Italy’’ (SALI). J Biomed Informat
8. Breslow NE, Day NE (editors). Statistical methods in cancer research
, Vol. II. The design and analysis of cohort studies
, no. 82. Lyon: IARC Science Publishers; 1987. pp. 60–71.
9. Frisch M, Biggar RJ, Engels EA, Goedert JJ, for the AIDS Cancer Match Registry Study Group. Association of cancer with AIDS-related immunosuppression in adults. JAMA
10. Grulich A, Li Y, McDonald A, Correll PKL, Law MG, Kaldor JM. Rates of non-AIDS defining cancers in people with HIV infection before and after AIDS diagnosis. AIDS
11. Dal Maso L, Serraino D, Franceschi S. Epidemiology of AIDS-related tumours in developed and developing countries. Eur J Cancer
12. Serraino D, Pezzotti P, Dorrucci M, Alliegro MV, Sinicco A, Rezza G. Cancer incidence in a cohort of human immunodeficiency virus seroconverters. Cancer
13. Serraino D, Boschini A, Carrieri P, Pradier C, Dorrucci M, Dal Maso L, et al
. Cancer risk among men with, or at risk of, HIV infection in southern Europe. AIDS
:553–559.#m AcknowledgementsThe authors would like to thank Mrs Ilaria Calderan for editorial assistance.
Appendix: Cancer and AIDS Registry Linkage Study
Pierluca Piselli (IRCCS ‘L. Spallanzani', Rome, Italy); Claudia Braga, Gary Clifford (IARC Lyon, France); Emanuele Crocetti (Registro Tumori Toscano, Firenze, Italy); Fabio Falcini (Registro Tumori della Romagna, Forlì, Italy); Paola Zambon, (Registro Tumori del Veneto, Padova, Italy); Roberto Zanetti (Registro Tumori Piemonte, Torino, Italy); Marina Vercelli (Registro Tumori Ligure, Genova, Italy); Paolo Crosignani (Registro Tumori Lombardia, Provincia di Varese, Italy); Massimo Federico (Registro Tumori della Provincia di Modena, Italy); Vincenzo De Lisi (Registro Tumori della Provincia di Parma, Italy); Francesco La Rosa (Registro Tumori Umbro, Italy); Stefano Ferretti (Registro Tumori della Provincia di Ferrara, Italy); Mario Budroni (Registro Tumori della Provincia di Sassari, Italy); Silvano Piffer (Registro Tumori della Provincia di Trento, Italy); Ettore Conti (Registro Tumori di Popolazione della Provincia di Latina, Italy); Adriano Giacomin (Registro Tumori della Provincia di Biella, Italy); Gianni Vicario (Registro dei Tumori del Friuli-Venezia Giulia, Italy); Francesco Bellò (Registro Tumori dell'Alto Adige, Italy); Franco Pannelli (Registro Tumori Provincia di Macerata, Camerino, Italy); Rosario Tumino (Registro Tumori Ragusa, Italy); Mario Fusco (Registro Tumori della Provincia di Napoli, Italy); Giovanni Rezza (Istituto Superiore di Sanità, Roma, Italy).