Assessment of the efficacy and safety of pre-emptive anti-cytomegalovirus (CMV) therapy in HIV-infected patients with CMV viraemia

Abstract

A number of studies have demonstrated that cytomegalovirus (CMV) viraemia is a strong predictor for CMV end-organ disease (EOD) and death in HIV-infected patients. We assess the efficacy and safety of pre-emptive anti-CMV therapy (PACT) for preventing these events. We performed a retrospective study of all HIV-infected patients seen in our institution who had detectable CMV viraemia in 2007. Seventy-one patients with advanced HIV disease (median CD4 cell count = 61 cells/mm³) were studied. Sixteen patients received PACT (mainly valganciclovir). Patients who received PACT had lower CD4 cell counts and higher blood CMV DNA levels. The cumulative incidence of CMV EOD and death at one year was 44% and 21% in patients with and without PACT, respectively (p = 0.013). Both PACT and high blood CMV DNA levels were significantly associated with CMV EOD and death in unadjusted analysis. In adjusted analyses, only blood CMV DNA levels remained significantly associated with the risk of CMV EOD and death, whereas PACT was associated with a non-significant trend towards reduced CMV EOD or death (hazard ratio: 0.25, p = 0.13). Five patients with PACT experienced severe drug-related adverse events. In conclusion, the use of PACT in HIV-infected patients with CMV viraemia could improve outcome but is associated with significant toxicity.

Keywords

Cytomegalovirus CMV HIV AIDS opportunistic infection pre-emptive treatment valganciclovir toxicity

Introduction

The use of antiretroviral therapy has dramatically reduced the incidence of opportunistic infections in patients with HIV infection, including the incidence of cytomegalovirus (CMV) disease.¹ However, a significant number of patients still remain at risk of CMV disease because they either fail or discontinue antiretroviral therapy or present late to care with advanced HIV disease and remain immunosuppressed despite the introduction of antiretroviral therapy.² This burden of CMV disease remains particularly high in low and middle income settings, where the incidence of CMV retinitis for example has not changed in the past decade, despite the increased availability of antiretroviral therapy.³

A number of studies have demonstrated that the detection of CMV in blood or plasma is a strong predictor for CMV end-organ disease (EOD) and death in HIV-infected individuals, even in the era of antiretroviral therapy, both in high and low income settings.^4–10 The risk of developing CMV EOD or death also increases with higher levels of CMV viraemia.^5,11–13 In patients with solid organ or bone-marrow transplants, universal CMV prophylaxis and/or pre-emptive anti-CMV therapy (PACT) are recommended to prevent the occurrence of CMV EOD. In patients with HIV infection however there are no such recommendations, since most trials assessing CMV prophylaxis were performed at the beginning of the era of combined antiretroviral therapy and yielded conflicting results.^14–18 Two recently conducted studies assessing the efficacy of PACT also yielded different conclusions.^19,20 We therefore wished to review in our centre the efficacy and safety of PACT when used in HIV-infected patients with CMV viraemia.

Methods

Study design

In this retrospective study, we identified using the electronic database of the laboratory of virology, all HIV-infected patients in our institution with a positive CMV viraemia between 1 January 2007 and 31 December 2007.

We then reviewed the clinical charts of these patients, excluding those being treated for CMV EOD. We recorded for each patient the following data: age, gender, country of origin, risk factors for HIV infection, CDC clinical stage, previous occurrence of opportunistic infections, co-infections with HCV or HBV and nadir CD4 T-cell count.

Baseline laboratory evaluations included CD4 cell count, plasma HIV RNA levels and results of blood polymerase chain reaction (PCR) assays for CMV. Baseline was defined as either the time of the first positive CMV viraemia for patients without PACT or the time of the last positive PCR assay before PACT for those receiving PACT. Antiretroviral therapy and immunosuppressive regimens (steroids, chemotherapy) at baseline were also recorded.

PACT was defined as an antiviral treatment including at least one anti-CMV agent (ganciclovir, valganciclovir, foscarnet or cidofovir) given orally or intravenously in a patient with CMV viraemia without evidence of CMV EOD. Due to the retrospective design of the study the duration of PACT was not pre-defined but left to the physician’s discretion.

CMV EOD was diagnosed using the ACTG criteria for clinical events.¹⁹ Briefly, CMV retinitis was defined by typical retinal lesions including white areas with or without haemorrhages and diagnosed by an experienced ophthalmologist using indirect ophthalmoscopy. CMV gastrointestinal disease was defined by the combination of suggestive clinical symptoms (abdominal pain, diarrhoea), visualization of endoscopic lesions (mucosal erythema, erosion or ulceration) and biopsies showing histological evidence of CMV infection by characteristic intranuclear inclusions and/or antigen detection.

Patients were followed until the occurrence of CMV EOD, death or 31 st December 2008. Due to the small size of the study and the role of CMV viraemia on both EOD and death, we decided to assess the efficacy of PACT on the combined outcome of CMV EOD or death, whichever occurred first.

In order to assess the safety of PACT, we also recorded grade 3 and 4 drug-related adverse events using the National Institute of Health Division of AIDS tables for grading adult adverse experiences (neutropaenia grade 3: < 750/mm³, grade 4 < 500/mm³; anaemia grade 3 < 7.4 g/dl, grade 4 < 6.5 g/dl, hypokalaemia grade 3 < 2.5 mmol/l, grade 4 < 2 mmol/l; hypophosphataemia grade 3 < 1.9 mg/dl, grade 4 < 1 mg/dl), and events leading to PACT discontinuation or blood transfusion, or granulocyte colony-stimulating factor or erythropoietin administration.

CMV viraemia

Real-time quantitative PCR for CMV DNA in whole blood was performed as previously described. This assay was run daily in our institution, both for patients with HIV infection and for solid and bone-marrow transplants. The primers and probe are located in the UL 123 exon 4 gene.^21,22 The sensitivity of this assay is >100 genomes per ml.

Statistical analysis

Data are presented as count (percent) or median (interquartile range [IQR]) or median (range). Imbalance between baseline characteristics of patients with and without PACT was expressed in terms of standardized difference, which is 100 (m_a − m_b)/√{(s_a²+ s_b²)/2}, where for each covariate m_a and m_b are the sample means in the PACT and no PACT groups, respectively, and s_a² and s_b² are the corresponding sample variances.

The primary study endpoint, the cumulative incidence of CMV EOD or death was estimated by Kaplan-Meier product-limit estimator, among patients with or without PACT. The cumulative of incidence of CMV EOD and death separately were additionally estimated using usual methodology in a competing risks framework. The association of PACT and baseline characteristics with the hazard of CMV EOD or death was analysed using Cox proportional hazards model. Both unadjusted analyses and analyses adjusted for CMV viral load and CD4 T-cell counts were performed. Results are reported in terms of hazard ratio (HR) and 95% confidence interval (95%CI). Using the adjusted Cox model, predicted probabilities of CMV EOD or death were derived with Ederer estimate for the whole cohort of patients, had none of these patients received PACT or had all of them received PACT. All tests were two-sided, and the significance threshold was set at 0.05. Analyses were performed using R.2.6.2 statistical software (The R Foundation for Statistical Computing, Vienna, Austria).

Results

Study population

Among 2617 HIV-infected patients followed at our institution in 2007, 71 (2.7%) had at least one positive blood CMV DNA assay during the study period, excluding those being treated for CMV EOD, and were eligible for this study.

Of these 71 patients, 16 (22.5%) received PACT. These patients received valganciclovir alone (n = 12), ganciclovir (n = 1), ganciclovir followed by valganciclovir (n = 1), foscarnet (n = 1) or a combination of ganciclovir, valganciclovir, foscarnet and cidofovir (n = 1). The median duration of PACT was 21 days (IQR 8.5–30.5).

Baseline patients’ main characteristics are presented in Table 1. Most patients were men (80.3%) and had advanced HIV infection with a median CD4 cell count of 61 cells/mm³, 72% having experienced a previous AIDS-defining event. Most patients originated from sub-Saharan Africa (44%), but also Europe (35%) and North Africa (8.5%). The main risk factors for HIV transmission were heterosexual contacts in 61%, and homosexual sex in 35%.

Table 1.

Baseline characteristics of the 71 patients with blood CMV viraemia.

	PACT		Standardized difference (%)^a	All patients
Characteristics	Yes n = 16	No n = 55	Standardized difference (%)^a	All patients
Median age (IQR) yrs	44 (37–50)	43 (37–48)	24.0	43 (25–68)
Male gender, n (%)	13 (81)	44 (80)	3.1	80.3
Median nadir of CD4 cell count (IQR*) cells/mm³	19 (8–36)	26 (12–68)	−37.6	25 (11–65)
HCV co-infection, n (%)	1 (6)	2 (4)	11.8	3 (4)
HBV co-infection, n (%)	2 (12)	7 (13)	−0.7	9 (13)
History of CMV disease, n (%)	2 (12)	2 (4)	32.1	4 (6)
Stage C of CDC, n (%)	14 (88)	37 (67)	49.0	51 (72)
Antiretroviral therapy, n (%)	8 (50)	30 (55)	−8.9	38 (54)
Immunosuppressive treatment, n (%)	5 (31)	8 (15)	39.6	13 (18)
Median CD4 count, cells/mm³ (IQR*)	26 (11–72)	80 (24–150)	−33.5	61 (20–144)
CD4 count <50 cells/mm³, n (%)	11 (69)	24 (44)	51.3	35 (49)
Median HIV RNA log copies/ml (IQR*)	4.1 (2.6–5)	4.3 (2.4–5.4)	−10.73	4,33 (2.39–5.35
Plasma HIV viral load, n (%)
<50	1 (6)	6 (11)	−16.4	7 (10)
<400	4 (25)	16 (29)	−9.0	20 (28)
Median blood CMV DNAx log cp/ml (IQR*)	5.0 (4.7–5.3)	3.2 (2.7–3.6)	333.0	3.4 (2.9–4.4)
Blood CMV viral load, n (%)
≤3 log	0 (0)	25 (45)	−127.9	25 (35)
3–4 log	0 (0)	22 (40)	−114.4	22 (31)
4–6 log	16 (100)	8 (15)	339.7	24 (34)

IQR: interquartile range.

Standardized difference between treated and untreated patients.

Four patients had a history of CMV retinitis (in 1996, 2001 [for 2 patients] and 2004). Median CMV DNA level was 3.38 log₁₀ copies/ml (IQR, 2.86–4.35 log₁₀ copies/ml), and only 38 (54%) patients were receiving antiretroviral therapy.

There was a clear imbalance between the two patient groups at baseline, with those receiving PACT being more advanced (lower CD4 cell count, higher proportion of previous AIDS-defining events) and having higher blood CMV viral loads. Indeed, all patients who received PACT had a CMV viral load of at least 4 log₁₀ copies/ml.

Incidence of CMV EOD and death

Median patients’ follow-up was 14 months (range, 2–27 months) and similar in both groups. Nineteen patients met an endpoint (CMV EOD or death). The one-year overall incidence of CMV EOD disease and death was 26% in this study, and the median time from CMV viraemia to event was 75 days (Table 2 and Figure 1).

Table 2.

Outcomes of patients with or without pre-emptive anti-CMV treatment.

	Pre-emptive anti CMV treatment
	Yes n = 16	No n = 55
Median follow-up (range), months	15 (9–21)	14 (2–27)
CMV disease, n	4	4
Median time to CMV disease (range), months	1.5 (11–2.5 d)	2.5 (1–2.5 d)
Death, n	4	7
Median time to death (range), months	4 (6–15 d)	5.5 (2–9)

d: days.

Figure 1.

Cumulative incidence of CMV EOD or death in patients receiving PACT or not. Unadjusted (left) and adjusted (right) analyses.

Eight patients developed CMV EOD during follow-up (6 developed retinitis and 2 gastrointestinal CMV diseases) with a cumulative incidence of 11.3%. Eleven patients (cumulative incidence: 15.5%) died during follow-up. None of the deaths were attributed to CMV but only one autopsy was performed. The causes of death were due to lymphoma (2), non-AIDS defining cancer (2), aspiration pneumonia with sepsis (1), HIV encephalitis (1), progressive multifocal leukoencephalopathy (1), cachexia (1), metabolic disorders (1) and unknown cause (2).

The one-year cumulative incidence of CMV EOD and death was significantly higher in patients with PACT as compared to patients without PACT (44% vs. 21%, respectively, p = 0.013) (Figure 1 left panel).

Also, six patients (3 in each group) developed AIDS-related-diseases during follow-up with a median time to occurrence of 3 months. At the end of follow-up 87.5% of patients in the PACT group and 85.5% in the other group were receiving antiretroviral therapy.

Baseline risk factors for CMV EOD and death

In unadjusted analysis, PACT and blood CMV viral load were significantly associated with the risk of CMV EOD and death (hazard ratio [HR], 3.2 [p = 0.013] and 2.28 [p = 0.0002], respectively). Antiretroviral therapy was not associated with a significant reduction of risk of CMV EOD and death (HR, 0.72 [p = 0.48]). However, in adjusted analyses for baseline CD4 cell count and CMV viral load, only blood CMV DNA levels remained significantly associated with the risk of CMV EOD and death (HR, 4.47 [p = 0.0018]), and there was a trend toward a benefit for PACT in reducing the risk of CMV EOD and death rate, although this was not statistically significant (HR, 0.25 [p = 0.13]) (Table 3, Figure 1).

Table 3.

Factors associated with CMV EOD and death in adjusted and unadjusted analysis.

	Unadjusted analysis		Adjusted analysis
	HR (95%CI)	p	HR (95%CI)	p
Pre-emptive anti-CMV treatment	3.20 (1.28–7.97)	0.013	0.25 (0.04–1.50)	0.13
Blood CMV VL (log 10 cps/ml)	2.28 (1.47–3.53)	0.0002	4.47 (1.74–11.5)	0.0018
Blood CMV VL>4 log₁₀ cps/ml	3.85 (1.54–9.61)	0.00039
CD4 cell counts (continuous)	0.999 (0.994–1.00)	0.53	1.002 (0.998–1.01)	0.40
CD4 <100 cells/mm³	2.33 (0.77–7.03)	0.13
Plasma HIV VL >5 log₁₀ cps/ml	1.31 (0.53–3.26)	0.56
Antiretroviral therapy	0.72 (0.29–1.78)	0.48
Immunosuppressive treatment	1.26 (0.42–3.81)	0.68

HR: hazard ratio; 95% CI: 95% confidence interval; VL: viral load; cp: copies.

Safety of PACT

Safety data were available for 14 of the 16 patients who received PACT, and severe drug-related adverse events occurred in 5 (36%). Three patients treated with valganciclovir experienced grade 3 or 4 neutropaenia (n = 2), or grade 3 anaemia (n = 1), all received haematopoietic growth factors and one received a blood transfusion. Of note, only one patient in the PACT group received zidovudine, but no safety data were available for this patient. Two patients treated with foscarnet experienced grade 3 and 4 metabolic adverse events (hypokalaemia (n = 2), hypophosphataemia [n = 1]).

Discussion

In this retrospective cohort study of 71 HIV-infected patients with low CD4 cell counts and detectable blood CMV DNA levels, we observed in the adjusted analysis a non-significant trend (HR = 0.25, p = 0.13) towards a reduced incidence of CMV EOD and death among patients who received PACT. Also, the use of PACT was associated with a significant rate of drug-related adverse events, in particular bone-marrow and metabolic toxicities that may preclude its use, along with the cost of this strategy.

These results are in agreement with those reported by Wohl et al. who did not observe a lower incidence of CMV EOD in patients receiving PACT in the setting of a placebo-controlled randomized trial.¹⁹ In that trial, the 12-month rate of CMV EOD was 14% among viraemic patients receiving placebo, very similar to the 11.3% reported in our study, and the 13.7% reported by Deayton et al.¹² The death rate was high in the Wohl’s trial and there was also no clear indication of a survival benefit of PACT with valganciclovir among patients with CMV viraemia.¹⁹ A recent retrospective study conducted in treatment-naïve patients with CMV viraemia in Japan reported however a lower incidence of CMV EOD following CMV pre-emptive therapy without any survival benefit.²⁰ Use of antiretroviral therapy in this study was however deferred by a median of 2 months.

New prospective, well-powered, randomized studies assessing the efficacy of PACT in patients with CMV viraemia are therefore needed to answer this important question in a context of a persistently high burden of CMV disease, in particular in low and middle income countries.

In our study, the only significant factor independently associated with an increased risk of CMV EOD and death was the level of CMV viraemia (HR: 4.47, p = 0.0018 in the adjusted analysis). This result is consistent with previous findings.^5,11–13 In this model, however, the severity of the immune deficiency was not associated with the risk of CMV EOD or death.

Our study has however a number of limitations. It is a retrospective study with a small sample size and few events recorded, having therefore limited power to identify significant prognostic factors associated with CMV EOD or death. We combined CMV EOD and death as the primary endpoint in our study because CMV viraemia has been consistently associated with an increased risk of CMV EOD and death. However this combined endpoint did not allow the analysis of the two endpoints separately due to the small study size. Finally, we could not really assess the causes of death in our patients since autopsies were not performed (except in one patient).

A more effective intervention to prevent CMV EOD and death in patients with advanced HIV infection is undoubtedly to control HIV replication as early as possible. Indeed, as demonstrated by Zolopa et al. in a randomized controlled trial, early initiation of antiretroviral therapy was associated with a reduction of AIDS-related events and death.²³

Such a benefit of antiretroviral therapy was not seen in our study in the adjusted analysis, but this could be due again to the small study size, the fact that 54% of patients were already receiving antiretroviral therapy at baseline and 86% at the end of follow-up, and the occurrence of CMV EOD and deaths that were too early in our study (median of 75 days) to allow the benefits of antiretroviral therapy to be evidenced.

In conclusion, our study confirmed that in patients with low CD4 counts, CMV viraemia is a strong predictor of CMV EOD and death, and the risk increases with increasing CMV viral load. In this retrospective study from a high-income setting, we could not demonstrate that a short course of PACT offered survival benefit in patients with CMV viraemia, and PACT was associated with significant toxicity. However, in analysis adjusted for level of CMV viraemia, there was a non-significant trend towards reduced CMV EOD disease and death with PACT. An appropriately powered prospective trial is needed to determine if a strategy for PACT can be devised that will reduce AIDS-related CMV EOD and mortality.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Steininger

Puchhammer-Stockl

Popow-Kraupp

. Cytomegalovirus disease in the era of highly active antiretroviral therapy (HAART). J Clin Virol 2006; 37: 1–9.

Wohl

Zeng

Stewart

. Cytomegalovirus viremia, mortality, and endorgan disease among patients with AIDS receiving potent antiretroviral therapies. J Acquir Immune Defic Syndr 2005; 38: 538–544.

Ford

Shubber

Saranchuk

. Burden of HIV-related cytomegalovirus retinitis in resource-limited settings: a systematic review. Clin Infect Dis 2013; 57: 1351–1361.

Casado

Arrizabalaga

Montes

. Incidence and risk factors for developing cytomegalovirus retinitis in HIV-infected patients receiving protease inhibitor therapy. Spanish CMV-AIDS Study Group. AIDS 1999; 13: 1497–1502.

El Amari

Combescure

Yerly

. Clinical relevance of cytomegalovirus viremia. HIV Med 2011; 12: 394–402.

Salmon-Ceron

Mazeron

Chaput

. Plasma cytomegalovirus DNA, pp65 antigenaemia and a low CD4 cell count remain risk factors for cytomegalovirus disease in patients receiving highly active antiretroviral therapy. AIDS 2000; 14: 1041–1049.

Spector

Wong

Hsia

. Plasma cytomegalovirus (CMV) DNA load predicts CMV disease and survival in AIDS patients. J Clin Invest 1998; 101: 497–502.

Micol

Buchy

Guerrier

. Prevalence, risk factors, and impact on outcome of cytomegalovirus replication in serum of Cambodian HIV-infected patients (2004-2007). J Acquir Immune Defic Syndr 2009; 51: 486–491.

Fielding

Koba

Grant

. Cytomegalovirus viremia as a risk factor for mortality prior to antiretroviral therapy among HIV-infected gold miners in South Africa. PLoS One 2011; 6: e25571–e25571.

10.

Durier

Ananworanich

Apornpong

. Cytomegalovirus viremia in Thai HIV-infected patients on antiretroviral therapy: prevalence and associated mortality. Clin Infect Dis 2013; 57: 147–155.

11.

Chevret

Scieux

Garrait

. Usefulness of the cytomegalovirus (CMV) antigenemia assay for predicting the occurrence of CMV disease and death in patients with AIDS. Clin Infect Dis 1999; 28: 758–763.

12.

Deayton

Sabin

Johnson

. Importance of cytomegalovirus viraemia in risk of disease progression and death in HIV-infected patients receiving highly active antiretroviral therapy. Lancet 2004; 363: 2116–2121.

13.

Erice

Tierney

Hirsch

. Cytomegalovirus (CMV) and human immunodeficiency virus (HIV) burden, CMV end-organ disease, and survival in subjects with advanced HIV infection (AIDS Clinical Trials Group Protocol 360). Clin Infect Dis 2003; 37: 567–578.

14.

Brosgart

Louis

Hillman

. A randomized, placebo-controlled trial of the safety and efficacy of oral ganciclovir for prophylaxis of cytomegalovirus disease in HIVinfected individuals. Terry Beirn Community Programs for Clinical Research on AIDS. AIDS 1998; 12: 269–277.

15.

Feinberg

Hurwitz

Cooper

. A randomized, double-blind trial of valaciclovir prophylaxis for cytomegalovirus disease in patients with advanced human immunodeficiency virus infection. AIDS Clinical Trials Group Protocol 204/Glaxo Wellcome 123-014 International CMV Prophylaxis Study Group. J Infect Dis 1998; 177: 48–56.

16.

Salmon-Ceron

Fillet

Aboulker

. Effect of a 14-day course of foscarnet on cytomegalovirus (CMV) blood markers in a randomized study of human immunodeficiency virus-infected patients with persistent CMV viremia. Agence Nationale de Recherche du SIDA 023 Study Group. Clin Infect Dis 1999; 28: 901–905.

17.

Spector

McKinley

Lalezari

. Oral ganciclovir for the prevention of cytomegalovirus disease in persons with AIDS. Roche Cooperative Oral Ganciclovir Study Group. New Eng J Med 1996; 334: 1491–1497.

18.

Youle

Gazzard

Johnson

. Effects of high-dose oral acyclovir on herpesvirus disease and survival in patients with advanced HIV disease: a double-blind, placebo-controlled study. European-Australian Acyclovir Study Group. AIDS 1994; 8: 641–649.

19.

Wohl

Kendall

Andersen

. Low rate of CMV end-organ disease in HIV-infected patients despite low CD4+ cell counts and CMV viremia: results of ACTG protocol A5030. HIV Clin Trials 2009; 10: 143–152.

20.

Mizushima

Nishijima

Gatanaga

. Preemptive therapy prevents cytomegalovirus end-organ disease in treatment-naive patients with advanced HIV-1 infection in the HAART era. PloS one 2013; 8: e65348–e65348.

21.

Gouarin

Vabret

Scieux

. Multicentric evaluation of a new commercial cytomegalovirus real-time PCR quantitation assay. J Virol Methods 2007; 146: 147–154.

22.

Leruez-Ville

Ouachee

Delarue

. Monitoring cytomegalovirus infection in adult and pediatric bone marrow transplant recipients by a real-time PCR assay performed with blood plasma. J Clin Microbiol 2003; 41: 2040–2046.

23.

Zolopa

Andersen

Powderly

. Early antiretroviral therapy reduces AIDS progression/death in individuals with acute opportunistic infections: a multicenter randomized strategy trial. PloS one 2009; 4: e5575–e5575.