Portal de Periódicos da CAPES

Neoplasms and infections as the main causes of death in patients in complete response to HIV‐related non‐Hodgkin lymphoma in the combination antiretroviral therapy era: a study out of a series of 146 patients

2013; Wiley; Volume: 162; Issue: 2 Linguagem: Inglês

10.1111/bjh.12355

1365-2141

José‐Tomás Navarro, Maria João Baptista, Mireia Morgades, Cristina Tural, Fuensanta Millá, Evarist Feliú, Josep‐María Ribera,

CNS Lymphoma Diagnosis and Treatment

The widespread use of combination antiretroviral therapy (cART) has dramatically improved the prognosis of human immunodeficiency virus (HIV)-related non-Hodgkin lymphoma (NHL) (Kaplan, 2012). The longer survival of HIV-infected patients following lymphoma treatment put them at risk of death due to causes other than lymphoma (Ribera et al, 2012). Moreover, the incidence and proportion of deaths related to non-AIDS-defining cancers (NADC), mostly infection-related cancers and other non-AIDS-related events, is increasing among the HIV-infected population in the cART era, (Silverberg et al, 2009; Lifson et al, 2010; Nguyen et al, 2010). Regardless of HIV infection, NHL patients have an elevated risk of developing second malignancies and late non-neoplastic toxicities, such as infections, which can eventually cause death (Hemminki et al, 2008; Ng et al, 2011). There is scarce information about the causes of death after successful treatment of HIV-related lymphomas and their relationship with immune status and HIV-RNA load. We conducted a retrospective study in a series of 146 adult patients with HIV-related NHL, diagnosed in a single institution between 1985 and 2010, to compare the causes of death among patients with HIV-related NHL in complete response (CR), treated with and without cART. Two groups of patients were considered: those who started cART prior to lymphoma diagnosis or concomitantly with chemotherapy, and those not receiving cART. Demographic, HIV infection and lymphoma data of each case were collected. In those patients who achieved CR to lymphoma, the following variables were also recorded: lymphoma relapse, cause of death, and HIV viral load and CD4 lymphocyte count at the last follow-up. When comparing causes of death between groups, patients with an unknown cause were not considered. The Mann-Whitney U-test and the Chi-square test were used as appropriate for comparison of CD4 lymphocyte counts and HIV load between dead and alive patients in the cART group. Viral load was categorized as negative or positive. The probability of overall survival (OS, defined as the time between diagnosis and death from any cause) and disease-free survival (DFS, defined as the time between CR and relapse, last follow-up or death in CR) was calculated by the Kaplan-Meier method and compared using the log-rank test. Out of 146 patients, 139 were eligible with a median (range) follow-up of 8·24 (0·68–15·78) years. Forty patients (62%) in the cART group and 11 (26%) from the non-cART achieved CR (P < 0·001). In the cART group 33 patients received chemotherapy alone and seven chemotherapy plus adjuvant radiotherapy, whereas in the non-cART, 10 patients were treated with chemotherapy and one with chemotherapy plus radiotherapy. The comparison between these two groups is shown in Table 1. The causes of death were significantly different in the two groups (Table 1). Four patients died of a second malignancy in the cART group (hepatocarcinoma, anal carcinoma, carcinoma of vagina and Kaposi sarcoma), whereas none did so in the non-cART group. All patients developing second neoplasms had diffuse large B-cell lymphoma and had been treated with CHOP (cyclophosphamide, doxorubucin, vincristine, prednisolone) without radiotherapy. Four patients died of infection in the cART group [Pneumocystis jiroveci (n = 2), sepsis by Klebsiella oxytoca and Staphylococcus epidermidis (n = 1), and infection of unknown origin (n = 1)], with the same number of deaths by infection being reported in the non-cART group [Pneumocystis jiroveci (n = 1), progressive multifocal leucoencephalopathy (n = 1), cytomegalovirus colitis (n = 1), and infection of unknown origin (n = 1)]. Patients in the cART group who died had lower CD4 lymphocyte counts and higher HIV load than those who remain alive (Table 2). 0·108 (0·009–0·199) N = 9 0·398 (0·082–1·064) N = 24 We observed different causes of death after successful treatment for HIV-related NHL between patients who did or did not receive cART. Among the former group, second neoplasms and infections were the main causes of death (Table 1). On the other hand, lymphoma relapse and infections were the main cause of death among non-cART patients. Our study focussed on patients in CR because they presumably lived longer enough to die from causes other than lymphoma. All patients who had second neoplasms belonged to the cART group and the median time from CR to death was 112·27 months (range 11·17–149·17), while the median time from CR to death of lymphoma relapse was 12·33 months (range 7–25·90). Patients who died of infection had a similar median time from CR to death in both groups: 9·23 months (1·47–57·77) in the non-cART group and 6·85 months (4–17·07) in the cART group. In our study, the neoplasms causing death were mostly infection-related NADC. Since the introduction of cART, the spectrum of diseases affecting HIV-infected individuals has moved from AIDS-related diseases to non-AIDS-defining diseases, especially infection-related NADC (Silverberg et al, 2009; Nguyen et al, 2010). As well as longer survival, there might be additional factors predisposing HIV-related NHL patients in remission to develop non-AIDS defining diseases in the cART era. In our series, all patients on cART who died showed bad control of HIV infection (Table 2). Persistent low CD4+ cell counts and failure to suppress HIV-RNA during cART have been associated with an increased risk of infection-related NADC and higher mortality in HIV-infected patients with cancer (Achenbach et al, 2011; Kesselring et al, 2011). These data and our results suggest that the poor response to cART puts patients at risk of developing NADC, especially infection-related ones. Surprisingly, we did not find second malignancies related to NHL therapy, such as acute myeloid leukaemia, bladder cancer, or lung cancer (Ng et al, 2011). One patient from our series died of anal carcinoma. A significantly increased risk of anal cancer has been reported after chemotherapy for NHL in patients with and without HIV infection (Hemminki et al, 2008; Morton et al, 2010), probably facilitated by human papylomavirus infection. Data regarding infection as a cause of death did not show differences between patients with or without cART: the frequency, the immunodeficiency status, and the time from lymphoma response to death were similar in both groups. In summary, this study showed that patients with successful treatment of HIV-related NHL in the cART era are at risk of death not only by infections or AIDS-defining cancers but also by infection-related NADC. Poor control of HIV-infection seems to be an important factor favouring the development of infection-related NADC. Screening strategies for early detection of cancer should be performed among HIV-infected individuals in CR from NHL, especially in those with poor control of HIV infection. JTN and MJB performed the research; JTN, MJB and MM analysed the data; JTN and JMR designed the research study; CT, FM and EF participated in the study and provided the care of the patients; JTN and JMR wrote the paper. Ayuda para el fomento de la investigación clínica independiente, EC11-041 from Ministerio de Sanidad, Política Social e Igualdad and RD12/0036/0029 from RTICC, Instituto Carlos III.