Introduction: Lymphomas continue to be a leading cause of death in HIV-positive patients (HIV+ pts) in the cART era. The aim of this study was to review our 15-year single institution experience in performing Autologous Stem Cell Transplantation (ASCT) for HIV-infected and non-HIV patients with high-risk or relapsed lymphomas, focusing on infectious complications.

Patients and Methods: We retrospectively evaluated our cohort of 28 HIV+ pts who underwent an ASCT between 2000-2015, and compared it with a well-matched control group of 39 HIV-negative pts.

Patient and ASCT characteristics are described in Table 1. All HIV+ pts were on cART. Chemomobilization was used in 60% of the HIV cohort and 84% of the controls. BEAM conditioning regimen was the most common. All transplants were performed in the same tertiary hospital JACIE-accredited SCT unit.

The primary end points were first-year cumulative incidence (CI) of infection, total infectious episodes and infection-related mortality. For the analysis, we defined 3 different time frames: 1st Pre-engraftment; 2nd from engraftment to day +100 and 3rd from day +100 until 1 year after SCT. Events occurring during the first 2 periods were considered early infections as compared to late.


All patients received antiviral and anti-PJP prophylaxis, but a significantly higher proportion of HIV+ pts were given antibacterial and antifungal prophylaxis (2/3 vs 1/3 approximately, Table 1). G-CSF support was initiated in all HIV recipients and 66% of controls, and the median days of use was longer for the HIV group (7 vs 4 days, p= 0.04). Median time to neutrophil engraftment was similar in both groups (13 vs 11 days, p=0.55); 93% of HIV pts and all control pts reached ANC > 500c/uL by day +30.

Infectious episodes (IE) are described in Table 2, divided by pathogen subtype and time frame. Globally, all infection subtypes were more common in the HIV-infected cohort at some point.

The most significant findings from the analysis are as follows. CI of early global infections: HIV 75% vs non-HIV 25% (p= 0.04), Figure 1. Median number of global infectious events: HIV 65 vs non-HIV 39 (p= 0.002; OR 1.8 [1.2-2.8]). Bacteria: CI of pre-engraftment and early bacterial infections were not different among groups (42% vs 28%, p= 0.47), but the median number of bacterial episodes was clearly different: HIV 17 vs non-HIV 12 (p= 0.08; OR 2.16 [0.96-4.8]). Fungi: CI of early fungal infections: HIV 10.7% vs non-HIV 0% (p= 0.03); minor infections were not considered. Viruses: CI of early viral infections: HIV 46% vs non-HIV 15% (p= 0.004). Median nº of early viral IE: HIV 19 vs non-HIV 6 (p= 0.007; OR 4.16 [1.43-12]). CI of late viral infections: HIV 30% vs non-HIV 11.7% (p= 0.04). CMV reactivations were by far more common in the HIV cohort (p=0,01). HIV viral load bleeps were documented in 35% of the HIV patients (most commonly in the day +30 control) and one post-transplant virological failure was diagnosed, forcing HAART substitution.

Of note, 1st year CI of infection-related mortality was 14% in the HIV group vs 0% in the non-HIV group (p= 0.01), Figure 2. Three HIV+ pts suffered early fulminant septic episodes (1 E. coli + Enterococcus, 1 Rothiamucilaginosa, 1 non-clarified - possible Stenotrophomonasmaltophilia) and a 29-year old woman in CR after a 1st line for a stage IVsB Burkitt-like lymphoma died due to a severe influenza A pneumonia. Length of admission was also significantly longer for the HIV+ pts (median days 34 vs 28, p=0.041).

Regarding long-term outcome, median follow up as of July 2016 is 82 months for the HIV+ group and 70 months for the control group: 57% and 61% of the pts in each cohort are still alive, respectively. One HIV-infected pt and 3 controls have been lost to follow-up. EFS: 1 year 71.4% vs 81.9% (ns); 5 years: 63.9% vs 66.5% (ns). Overall Survival: 1 year 75% vs 84% (ns); 5 years 66.3% vs 74.6% (ns).

Conclusion: Autologous stem cell transplantation has been proven to be feasible and effective in HIV-related lymphomas, but in our experience and despite great advances in cART and virological control, HIV+ patients are at high risk of infection and this might influence post-ASCT short-term survival. It is mandatory to focus on prophylactic and supportive measures and to choose carefully the optimal timing for transplantation.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract