Nonmelanoma skin cancers among HIV-infected persons in the HAART era.
E. Garlassi,V. Harding,J. Weir,N. Francis,M. Nelson,T. Newsom-Davis,M. Bower
DOI: https://doi.org/10.1097/QAI.0b013e318251b004
2012-06-01
Journal of Acquired Immune Deficiency Syndromes
Abstract:To the Editors: Since the advent of highly active antiretroviral therapy (HAART), there has been increasing attention on the non–AIDS-defining cancers, including nonmelanoma skin cancers (NMSCs). Although epidemiological studies frequently exclude NMSCs, 2 studies have reported an increased relative risk for NMSCs in people living with HIV. However, little data exist on the pathological details and outcomes of NMSCs in this population. NMSCs are the most common malignancies in the United Kingdom with nearly 100,000 cases registered in 2008, although registration is known to be incomplete. More than 80% of cases occur in people older than 60 years, and 70%–80% are diagnosed as basal cell carcinoma (BCC) and most of the remaining cases are squamous cell carcinoma (SCC). Risk factors include cumulative solar ultraviolet and ionizing radiation exposure, genetic defects of DNA repair and of the Sonic hedgehog pathway, and infection with beta (cutaneous) genotypes of human papillomavirus. In addition, immunosuppression has been associated with an increased risk for NMSCs particularly among recipients of solid organ transplants. Epidemiological surveys of the incidence of non–AIDSdefining malignancies in seropositive populations usually exclude NMSCs, although 2 recent studies have reported an increased relative risk for NMSCs in people living with HIV. In a metaanalysis published in 2007, the standardized incidence ratio for NMSCs in patients with HIV/AIDS was recorded as 4.1, whereas the value was 28.6 for allograft recipients. Underreporting of NMSCs and dermatological care in diverse locations prevent an accurate epidemiological study in our cohort at the Chelsea and Westminster Hospital. The natural history of NMSCs depends on the site, size, grade, and histopathologic subtypes and the presence of perineural invasion and lymphatic infiltration. In the case of BCC, adverse histopathologic subtypes include morphoeic, infiltrative, or micronodular histologies. We have assessed the presence of these features in a cohort of HIVseropositive patients with histopathologically confirmed NMSCs. We carried out a retrospective single-center study involving HIV-positive patients diagnosed with histopathologically confirmed NMSCs. Cases were identified from a search of the histopathologic specimen database in the Chelsea and Westminster Hospital from 1996 to 2011. Full clinicopathologic details were sought for all known HIV-seropositive patients identified as having histologically diagnosed NMSCs. For the purposes of this study, HAART was defined as treatment with $2 nucleoside reverse transcriptase inhibitors in combination with at least 1 protease inhibitor or 1 nonnucleoside reverse transcriptase inhibitor, or an abacavir-containing or a tenofovircontaining regimen of $3 nucleoside reverse transcriptase inhibitors. Data regarding clinicopathologic features and outcome were analyzed using descriptive statistics. Comparison of variables between groups was performed by x2 test for nominal variables and Mann–Whitney and Kruskal–Wallis tests for nonparametric continuous variables; all P values presented are 2-sided. Overall survival was calculated from the day of NMSC diagnosis until death or the date of last follow-up. A total of 159 NMSCs, including 126 BCCs and 33 SCCs, were identified in 58 patients (57 men) with a mean age of 54 years (range: 31–77 years). In the same period of time (1996 to October 2011), 510 patients were diagnosed with cutaneous Kaposi sarcoma. At the time of diagnosis of NMSC, 76% were on HAART, the median CD4 cell count was 343 cells per cubic millimeter (range: 1–937 cells/mm3), 69% had an undetectable plasma HIV viral load, and the median duration of HIV-seropositive status was 10.6 years (range: 0.5–25 years). The clinicopathologic details of the 58 patients are shown in Table 1. There were no significant differences between patients with SCC and BCC in age (Mann–Whitney P = 0.10), duration of HIV infection (Mann–Whitney P = 0.98), CD4 count (Mann–Whitney P = 0.69), CD4 nadir (Mann–Whitney P = 0.16), or viral load (Mann–Whitney P = 0.85) at NMSC diagnosis. Nearly a quarter (23%) of the BCCs arising in individuals living with HIV infection were adverse histological subtypes (11% infiltrative, 11% morphoeic, 1% micronodular). No tumors demonstrated dermal lymphatic infiltration, and only 1 tumor showed perineural invasion. The characteristics of BCC in HIV-seropositive individuals are shown in Table 1. There was no correlation between the aggressive histological subtypes and tumor site on the body (x2 P = 0.25), HAART usage (x2 P = 0.62), undetectable HIV viral load (either ,400 copies/mL, x2 P = 0.55, or ,50 copies/mL, x2 P = 0.87), CD4 cell count (Mann–Whitney P = 0.14), duration of HIV infection (Mann– Whitney P = 0.65), length of time on HAART (Mann–Whitney P = 0.16), or CD4 nadir (Mann–Whitney P = 0.98). The median follow-up for patients with BCC was 3.6 years (range: 0.1–16 years), and the 5-year overall survival was 90% (95% confidence interval: 89% to 92%). Three patients with BCC have died: 1 patient from primary cerebral lymphoma, 1 from systemic non-Hodgkin lymphoma, and 1 from progressiveMycobacterium avium complex. SCC staging results were as follows: pT1, 80%; pT2, 3%; and pT4, 17%. Half the tumors were moderately differentiated, 41% well differentiated, and 9% poorly differentiated. One tumor demonstrated both perineural invasion and lymphatic infiltration, and this patient presented with metastatic lung disease. The characteristics of SCC in HIV-seropositive individuals are shown in Table 1. There was no correlation between tumor grade and tumor site on the body (x2 P = 0.11), HAART Presented at the 18th Annual Conference of the British HIV Association, April 18–20, 2012, Birmingham, United Kingdom. The authors have no funding or conflicts of interest to disclose.