Introduction

Human cytomegalovirus (HCMV) infection still remains the most frequent viral complication of the post transplant period in both solid organ and hematopoietic stem cell transplant (HSCT) recipients. Of the two currently adopted strategies for prevention of HCMV disease, prophylaxis requires sustained administration of antiviral drugs to all transplanted patients with disadvantages in terms of drug toxicity,^{1, 2} while preemptive therapy requires administration of anti-HCMV drugs only to patients with HCMV infection in blood, and is based on continuous monitoring of HCMV infection in blood by either antigenemia or DNAemia.^{1, 3, 4, 5, 6}

Unlike solid organ transplant recipients, preemptive therapy in HSCT recipients is generally initiated upon the first confirmed positive result for HCMV in blood.^{4, 5} This approach has been utilized due to the high risk of HCMV interstitial pneumonia in HSCT recipients in the early post transplant period.⁷

In a recent study conducted by our group in which an ultrasensitive diagnostic assay (immediate-early mRNA detection by nucleic acid sequence-based amplification) was compared with antigenemia, it was shown that this sensitive assay led to treatment of a higher number of patients with respect to antigenemia.⁸ At this time, based on a retrospective analysis of DNAemia levels in HSCT recipients undergoing antigenemia-guided surveillance of HCMV infection,^{9, 10} we decided to compare a predetermined cutoff of 10 000 copies per ml whole blood with first antigenemia positivity, for preemptive therapy guidance of HCMV infection in a population of pediatric HSCT recipients. DNAemia was chosen for cutoff definition instead of antigenemia, since the assay is partially automatable and standardizable, and more closely reflects the actual viral replication in vivo. Results of this recently completed study showed that patients in the DNAemia arm were treated in a significantly lower number as compared to those in the antigenemia arm.¹¹

In parallel, we conducted a similar trial in adult HSCT recipients, to verify whether use of the DNAemia cutoff spared antiviral treatment in a significant portion of this patient population in comparison to antigenemia. This study was particularly stimulated by our knowledge that the frequency of HCMV infection in the post transplant period is about twice in adults (70–80% ) as compared to young (30–40% ) HSCT recipients.

Patients and methods

Patients

From December 2002 to April 2007, a total of 88 consecutive adult patients (median age 44, range 20–67, years) received an unmanipulated allogeneic HSCT from an human leukocyte antigen (HLA)-identical sibling (n=45) or an unrelated donor (UD, n=43) at the Unità di Trapianto di Midollo, Clinica Ematologica, Fondazione IRCCS (Istituto di Ricovero e Cura a Carattere Scientifico) Policlinico San Matteo, Pavia, Italy. Inclusion criteria were serological evidence of past HCMV infection in the donor (D) or recipient (R) and patient's written informed consent, while exclusion criteria were D/R seronegativity for HCMV, T-cell depletion of the graft and a life expectancy less than 2 months. Five patients were dropped from the study (two were lost to follow-up and three had protocol violations). The characteristics of the 83 patients analyzed in the study are reported in Table 1.

Table 1 - Characteristics of the 83 patients completing follow-up.

Full table

Study design

Following approval from the local ethics committee and institutional review board, a prospective, randomized, open-label trial comparing the use of a DNAemia cutoff versus first positive antigenemia for initiation of preemptive therapy of HCMV infection in HSCT recipients was carried out.

The primary end point of the study was the comparison of the number of patients treated in the two arms. Secondary end points were comparisons in the two arms of the frequency of HCMV infection, time to initiation of antiviral therapy, frequency of relapse episodes and relapse episodes requiring antiviral treatment, overall duration of antiviral treatment as well as first course of treatment. Finally, the occurrence of graft-versus-host disease (GVHD) and transplant-related mortality (TRM) was investigated in the two arms.

The sample size calculated by the ² test indicated a minimum number of 37 patients to be randomized per arm, based on a significance level of 0.05, a study power of 0.80, and the hypothesized need for treatment in 80% of patients allocated to the antigenemia arm and 50% of patients allocated to the DNAemia arm. This hypothesis was based on previous studies conducted at our institution.

HCMV infection and disease were defined as previously reported.¹² Patients were randomized to either arm using a block random design for two groups on the basis of donor type (sibling versus UD). In the DNAemia arm, patients were treated upon reaching a DNA level of 10 000 copies per ml whole blood, a cutoff selected in view of results from previous studies,^{9, 10} while treatment was stopped after two consecutive negative results. Infection relapses were treated similarly. In the antigenemia arm, patients were treated immediately upon detection of two or more pp65-positive leukocytes. When a single positive leukocyte was detected, patients were treated if antigenemia positivity was confirmed at the subsequent control (that is, after 2–3 days). Therapy was discontinued upon two consecutive negative results, and relapse episodes were treated similarly. Patients in the DNAemia arm were also tested for antigenemia, while patients in the antigenemia arm were also tested for DNAemia. However, results of these assays were not transmitted to clinicians, thus not influencing therapy.

GVHD prophylaxis consisted of cyclosporine A associated with a short course of methotrexate and steroids (methylprednisolone 0.5 mg/kg per day from day +4 until day +30) for patients receiving the allograft from an HLA-identical sibling, whereas patients transplanted from an UD also received antithymocyte globulin (ATG) at the total dose of 7.5 mg/kg over 3 days before stem cell infusion. Acute GVHD was treated with steroids as first-line therapy (methylprednisolone 2–5 mg/kg per day), while patients with steroid-resistant disease were treated with extracorporeal photochemotherapy¹³ or ATG. All patients enrolled in the study were given transfusions of irradiated leukocyte-depleted red blood cells, platelets and frozen plasma from blood donors not screened for HCMV serology. In addition, all patients were given acyclovir for prophylaxis of herpes simplex virus reactivation from day - 1 through day +30. Intravenous ganciclovir (5 mg/kg, b.i.d.) was administered as preemptive therapy in both arms. Relapse episodes were treated similarly. In case of drug toxicity, foscarnet (90 mg/kg, b.i.d.) was given as a replacement for ganciclovir.

Virological follow-up

Patients were monitored for HCMV infection for at least 3 months after HSCT. As a rule, anticoagulated blood samples were collected twice a week until day +45 and weekly thereafter, unless patients showed virological evidence of active HCMV infection at the time of expected monitoring discontinuation. In this case, monitoring was continued until virus disappearance from blood. Subsequently, HCMV monitoring was performed in all patients at least at 6, 9 and 12 months post-HSCT and in the presence of clinical signs or symptoms suggestive of HCMV disease. The median follow-up of surviving patients was 15 months (range 3–41) for patients allocated to the DNAemia arm, and 19 months (range 3–50) for patients allocated to the antigenemia arm (P=NS). The D/R serostatus was determined by ELISA prior to transplantation.¹⁴

DNAemia was quantified by real-time PCR according to a procedure previously reported.¹⁵ HCMV DNA copy number was determined from DNA aliquots extracted from 10 l whole blood (NuclisensEasy MAG extractor, BioMérieux, Durham, NC, USA) according to Boom et al.¹⁶ Results were multiplied by 100 to achieve copy number per ml. The detection limit of the assay is 1000 DNA copies per ml blood. The antigenemia assay was performed by counting the number of pp65-positive leukocytes/2 10⁵ leukocytes examined.^{17, 18} In 52 patients (25 in the antigenemia, and 27 in the DNAemia arm), HCMV-specific CD4⁺ and CD8⁺ T-cell reconstitution was monitored according to a method previously reported. HCMV-specific immune reconstitution was considered to have occurred in patients reaching levels of HCMV-specific CD4⁺ and CD8⁺ T cells greater than 0.4 per l blood (that is, the minimum level observed in healthy HCMV-seropositive subjects).¹⁹

Statistical analysis

Data were analyzed as of 15 July 2007. Differences between medians were compared by using the Mann–Whitney U-test. Differences in percentages were tested using the Pearson's ² test, while the Fisher's exact test was used to evaluate differences in percentages when the total sample size was less than 30. All tests were two tailed. A Cox regression model was adopted to analyze factors potentially associated with HCMV infection or anti-HCMV preemptive therapy. Grade II–IV acute GVHD was evaluated as a time-dependent variable and considered only if occurring before HCMV infection or anti-HCMV treatment. The probabilities of HCMV infection, need for preemptive therapy, acute GVHD and 1-year TRM were calculated and expressed as cumulative incidence taking into account competing risks.^{20, 21} P-values lower than 0.05 were considered statistically significant.

Results

Detection of HCMV infection

As shown in Table 2, in the post transplantation period, HCMV–DNA was detected in blood of 32 of 42 (76% ) patients enrolled in the DNAemia arm, whereas HCMV antigenemia was detected in 35 of 41 (85% ) patients in the antigenemia arm (P=NS). The cumulative incidence curves (Figure 1a) showed a probability of HCMV infection (as detected by the relevant randomization assay) at day +360 of 76% (95% confidence intervals (CI) 64–90% ) in the DNAemia arm, and 85% (95% CI 75–97% ) in the antigenemia arm (P=0.20). However, since DNAemia was also determined in patients of the antigenemia arm, as well as antigenemia was also tested in patients of the DNAemia arm, it was possible to compare results of the two assays in the whole patient population (regardless of randomization arm). In this respect, a trend toward a higher proportion of patients with positive antigenemia (69 of 83 patients, day +360 cumulative incidence 84% , 95% CI 76–92% ) as compared to patients with positive DNAemia (57 of 83 patients, day +360 cumulative incidence 69% , 95% CI 59–79% ) was detected (P=0.07).

Figure 1.

Cumulative incidence of human cytomegalovirus (HCMV) infection (a) and preemptive treatment (b) in the two randomization arms. Numbers in parentheses indicate the actual number of infected (a) or treated (b) patients and the relevant probability at day +360 with 95% confidence intervals (CI).

Full figure and legend (89K)

Table 2 - Preemptive therapy of HCMV infection in the two randomization arms.

Full table

Preemptive therapy

The number of patients requiring antiviral treatment was significantly lower in the DNAemia arm (25 of 42 patients, 60% ) as compared to the antigenemia arm (33 of 41 patients, 80% ), as shown in Table 2 (P=0.03). The cumulative incidence curve of the probability of starting preemptive therapy at day +360 was 63% (95% CI 49–79% ) in the DNAemia arm and 80% (95% CI 69–94% ) in the antigenemia arm (P=0.02; Figure 1b). In more details (Table 2), among the 32 of 42 patients in DNAemia arm in which HCMV–DNA was detected in blood, 7 had DNA levels consistently below 10 000 copies per ml blood (and HCMV infection resolved spontaneously), whereas the other 25 reached the cutoff for starting preemptive therapy. In the antigenemia arm, among the 35 of 41 antigenemia-positive patients, 2 had a single pp65-positive leukocyte not confirmed at the subsequent control (and did not receive anti-HCMV treatment), while the remaining 33 patients matched the criteria for preemptive treatment. Thus, the number of patients treated was significantly lower in the DNAemia arm (P=0.03). Times to first HCMV detection and treatment onset were comparable in the two arms, as well as the percentage of patients experiencing HCMV relapses and the percentage of patients needing antiviral treatment for relapse (Table 2). Similarly, the median duration of the first course of treatment as well as the total duration of antiviral therapy was not significantly different between the two randomization arms (Table 2).

No case of HCMV disease was observed in either arm during the study period, except for a single patient allocated to the DNAemia arm who was affected by biopsy-proven HCMV gastritis 4 months after transplantation in the absence of virus in blood (Table 2). The gastric disease was treated with ganciclovir and resolved following a 3-week treatment course.

The cumulative incidence of grade II–IV acute GVHD as well as that of TRM was comparable in the two arms. The cumulative incidence of grade II–IV GVHD was 31% (95% CI 20–49% ) in the DNAemia and 27% (95% CI 16–44% ) in the antigenemia arm (P=NS). The 1-year TRM was 35% (95% CI 22–55% ) in the DNAemia, and 27% (95% CI 16–46% ) in the antigenemia arm (P=NS).

The following variables potentially influencing the detection of HCMV infection were analyzed (Table 3): conditioning regimen (chemotherapy based/total body irradiation), stem cell source (bone marrow/peripheral blood), donor serostatus (HCMV-seronegative/seropositive), recipient serostatus (HCMV-seronegative/seropositive), donor type (unrelated/sibling), ATG administration (yes/no), randomization arm (antigenemia/DNAemia) and acute GVHD (grade 0–I/II–IV). In multivariate analysis, factors significantly associated with the detection of HCMV infection in the entire patient population were HCMV seropositivity recipient, an UD and allocation to the antigenemia arm. In addition, HCMV seropositivity of the recipient and allocation to the antigenemia arm were the only factors statistically influencing the probability of receiving preemptive therapy in the multivariate analysis. No influence of acute GVHD on the occurrence of HCMV infection and requirement of preemptive therapy was observed (Table 3).

Table 3 - Factors potentially influencing HCMV detection in blood and preemptive treatment (multivariate analysis).

Full table

Viral load at treatment initiation

As shown in Figure 2a, upon initiation of treatment, median DNAemia levels were significantly higher in the 25 treated patients of the DNAemia arm (20 300, range 10 000–147 200, copies per ml) as compared to the 33 treated patients of the antigenemia arm (1100, range 0–68 700, copies per ml; P<0.01). Similarly (Figure 2b), levels of antigenemia were significantly higher in the DNAemia (14, range 0–300, pp65-positive leukocytes) as compared to the antigenemia arm (3, range 1–39, pp65-positive leukocytes; P<0.01).

Figure 2.

Levels of DNAemia (a) and antigenemia (b) at start of treatment in the two randomization arms. Median levels of DNAemia (20 300, range 10 000–147 200, copies per ml) and antigenemia (14, range 0–30, pp65-positive leukocytes) at start of treatment in the DNAemia arm were significantly higher than DNAemia and antigenemia levels at start of treatment in the antigenemia arm (1100, range 0–68 700, copies per ml; and 3, range 1–39, pp65-positive leukocytes, respectively).

Full figure and legend (64K)

Immune control of HCMV infection

HCMV-specific CD4⁺ and CD8⁺ T-cell reconstitution data were available for 52 patients equally distributed between the two randomization arms. The cumulative incidence of the immunological reconstitution for HCMV-specific CD4⁺ in adult HSCT recipients at 100 days after transplantation was 72% (CI 56–92% ) in the antigenemia and 68% (53–89% ) in the DNAemia arm, and for HCMV-specific CD8⁺ 84% (71–100% ) in the antigenemia and 80% (65–97% ) in the DNAemia arm. The difference between the two arms was not significant for either CD4⁺ or CD8⁺ T-cell-mediated immune response.

Discussion

In the past, preemptive therapy of HCMV infection in HSCT recipients was initiated upon first detection of virus in blood. The rationale for this approach was based on the risk of an early progression to interstitial pneumonia, a severe complication associated with a high fatality rate.⁷ Of the two assays currently employed for diagnosing HCMV infection in blood, DNAemia has been shown to correlate with the actual viral replication rate more closely than antigenemia.²² The latter assay is based on the detection of the viral protein pp65 in peripheral blood leukocytes of transplant recipients. The synthesis of this protein can occur independently of viral DNA replication.²³

Based on this knowledge and the results of a receiver–operator curve analysis,^{9, 10} we recently terminated a trial in young HSCT recipients in which a DNAemia cutoff of 10 000 copies per ml blood was compared to initial antigenemia positivity for starting preemptive therapy of HCMV infection.¹¹ Results of this study showed that a DNAemia cutoff can be safely adopted to guide preemptive therapy of HCMV infection in HSCT recipients (no case of HCMV disease was observed), while the number of patients requiring treatment with respect to antigenemia was significantly decreased.

In the present study, we decided to compare the same DNAemia cutoff of 10 000 copies per ml blood with first antigenemia positivity in an adult HSCT recipient population showing a much higher rate of HCMV infection in the post transplant period with respect to the pediatric patient population. Our results are in agreement with those of a recent investigation²⁴ and indicate that a DNAemia cutoff can be safely used to guide preemptive therapy of HCMV infection in adult HSCT recipients. The single case of HCMV disease was relevant to a late case of HCMV gastritis occurring 4 months after transplantation in the absence of virus in blood. In addition, adoption of a cutoff of 10 000 copies per ml blood significantly decreased the number of patients requiring treatment with respect to antigenemia. Although statistical significance was not reached, the median 6-day delay in treatment initiation observed in the DNAemia arm (Table 2) appears crucial. During this time interval, HCMV infection may be spontaneously controlled in the DNAemia arm by the reconstituting recipient immune system, which has been shown to develop protective levels of HCMV-specific CD4⁺ and CD8⁺ T cells in the great majority of HCMV-seropositive patients within the first 2–3 months after transplantation.²⁵ The finding obtained in this study that both DNAemia and antigenemia were significantly higher in the DNAemia arm as compared to the antigenemia arm suggests that during the interval between first antigenemia detection and attainment of the DNAemia cutoff, HCMV infection may have exerted a stimulatory effect on HCMV-specific immune reconstitution. Finally, use of a DNAemia cutoff did not entail any difference in both acute GVHD occurrence and TRM with respect to antigenemia.

In multivariate analysis, among the different variables examined, HCMV seropositivity, an UD and the allocation to the antigenemia arm were the factors correlating with the detection of HCMV infection. In addition, HCMV seropositivity and the enrollment in the antigenemia arm correlated also with the need for preemptive therapy. We confirmed that seropositivity of the recipient correlates with HCMV infection as already shown in our previous study in pediatric patients.²⁵ The significantly higher incidence of HCMV detection in the antigenemia arm, as revealed by multivariate analysis, might depend on the lack of correlation between antigenemia positivity and viral replication. These findings, along with our previous observations showing a better correlation of DNAemia than antigenemia with clinical symptoms,²⁶ and a series of our unpublished observations showing a striking dissociation between high antigenemia and low DNAemia levels, even prior to antiviral treatment, suggest that pp65 synthesis in the absence of HCMV DNA replication may be the major factor influencing the higher risk ratio for HCMV infection in the antigenemia arm. In addition, the significantly higher incidence of treated patients in the antigenemia arm may be referred to lack of a cutoff in the antigenemia arm (increasing number of patients treated), but it could also be partially attributed to the same mechanism of pp65 synthesis.

In conclusion, although both assays may be used to safely guide preemptive therapy of HCMV infection in HSCT recipients, the use of a DNAemia cutoff of 10 000 DNA copies per ml blood resulted in a significant reduction in the number of patients preemptively treated, thus sparing unnecessary antiviral treatment in a significant proportion of patients. It cannot be excluded that a better preemptive therapy strategy could be achieved by adopting a cutoff also for the antigenemia assay instead of first antigenemia positivity (Figure 2). However, DNAemia more closely reflects the actual viral replication in vivo and the assay is automatable and standardizable. In this respect, a recent trial on HCMV DNA quantification performed by a number of Italian transplantation centers showed a variability range of 0.5–0.7 log₁₀ among the different home-brewed and commercial assay systems available (unpublished data). Thus, DNAemia should become the first choice assay for monitoring of HCMV infection and preemptive therapy in HSCT recipients, and a DNAemia cutoff of 10 000 copies per ml blood appears safe and clinically useful.

References

1. Boeckh M, Gooley TA, Myerson D, Cunningham T, Schoch G, Bowden RA. Cytomegalovirus pp65 antigenemia-guided early treatment with ganciclovir versus ganciclovir at engraftment after allogeneic marrow transplantation: a randomized double-blind study. Blood 1996; 88: 4063–4071. | PubMed | ISI | ChemPort |
2. Nguyen Q, Champlin R, Giralt S, Rolston K, Raad I, Jacobson K et al. Late cytomegalovirus pneumonia in adult allogeneic blood and marrow transplant recipients. Clin Infect Dis 1999; 28: 618–623. | Article | PubMed | ISI | ChemPort |
3. Boeckh M, Bowden RA, Goodrich JM, Pettinger M, Meyers JD. Cytomegalovirus antigen detection in peripheral blood leukocytes after allogeneic marrow transplantation. Blood 1992; 80: 1358–1364. | PubMed | ISI | ChemPort |
4. Locatelli F, Percivalle E, Comoli P, Maccario R, Zecca M, Giorgiani G et al. Human cytomegalovirus (HCMV) infection in paediatric patients given allogeneic bone marrow transplantation: role of early antiviral treatment for HCMV antigenaemia on patients' outcome. Br J Hematol 1994; 88: 64–71. | Article | ChemPort |
5. Einsele H, Ehninger G, Hebart H, Wittkowski KM, Schuler U, Jahn G et al. Polymerase chain reaction monitoring reduces the incidence of cytomegalovirus disease and the duration and side effects of antiviral therapy after bone marrow transplantation. Blood 1995; 86: 2815–2820. | PubMed | ISI | ChemPort |
6. Ljungman P, Aschan J, Lewensohn-Fuchs I, Carlens S, Larsson K, Lonnqvist B et al. Results of different strategies for reducing cytomegalovirus-associated mortality in allogeneic stem cell transplant recipients. Transplantation 1998; 66: 1330–1334. | Article | PubMed | ISI | ChemPort |
7. Enright H, Haake R, Weisdorf D, Ramsay N, McGlave P, Kersey J et al. Cytomegalovirus pneumonia after bone marrow transplantation: risk factors and response to therapy. Transplantation 1993; 55: 1339–1346. | PubMed | ISI | ChemPort |
8. Gerna G, Lilleri D, Baldanti F, Torsellini M, Giorgiani G, Zecca M et al. Human cytomegalovirus immediate-early mRNAemia versus pp65 antigenemia for guiding pre-emptive therapy in children and young adults undergoing hematopoietic stem cell transplantation: a prospective, randomized, open-label trial. Blood 2003; 101: 5053–5060. | Article | PubMed | ISI | ChemPort |
9. Lilleri D, Baldanti F, Gatti M, Rovida F, Dossena L, De Grazia S et al. Clinically-based determination of safe DNAemia cutoff levels for preemptive therapy of human cytomegalovirus infections in solid organ and hematopoietic stem cell transplant recipients. J Med Virol 2004; 73: 412–418. | Article | PubMed |
10. Gerna G, Lilleri D. Monitoring transplant patients for human cytomegalovirus: diagnostic update. Herpes 2006; 13: 4–11. | PubMed |
11. Lilleri D, Gerna G, Furione M, Bernardo ME, Giorgiani G, Telli S et al. Use of a DNAemia cut-off for monitoring human cytomegalovirus infection reduces the number of pre-emptively treated children and young adults receiving haematopoietic stem cell transplantation as compared to qualitative pp65-antigenemia. Blood 2007; 10: 2757–2760. | Article | ChemPort |
12. Ljungman P, Griffiths P, Paya C. Definitions of cytomegalovirus infection and disease in transplant recipients. Clin Infect Dis 2002; 34: 1094–1097. | Article | PubMed | ISI |
13. Salvaneschi L, Perotti C, Zecca M, Bernuzzi S, Viarengo G, Giorgiani G et al. Extracorporeal photochemotherapy for treatment of acute and chronic GvHD in childhood. Transfusion 2001; 41: 1299–1305. | Article | PubMed | ISI | ChemPort |
14. Gerna G, Revello MG, Percivalle E, Torsellini M. A 6-h microneutralization assay for human cytomegalovirus antibody by using monoclonal antibodies. Serodiagn Immunother Infect Dis 1990; 4: 243–247. | Article |
15. Gerna G, Vitulo P, Rovida F, Lilleri D, Pellegrini C, Oggionni T et al. Impact of human metapneumovirus and human cytomegalovirus versus other respiratory viruses on the lower respiratory tract infections of lung transplant recipients. J Med Virol 2006; 78: 408–416. | Article | PubMed |
16. Boom R, Sol CJA, Salimans MMM, Jansen CL, Wertheim-van Dillen PME, van der Nordaa J. Rapid and simple method for purification of nucleic acids. J Clin Microbiol 1990; 28: 495–503. | PubMed | ISI | ChemPort |
17. Gerna G, Revello MG, Percivalle E, Morini F. Comparison of different immunostaining techniques and monoclonal antibodies to the lower matrix phosphoprotein (pp65) for optimal quantitation of human cytomegalovirus antigenemia. J Clin Microbiol 1992; 30: 1232–1237. | PubMed | ChemPort |
18. Gerna G, Percivalle E, Torsellini M, Revello MG. Standardization of the human cytomegalovirus antigenemia assay by means of in vitro-generated pp65-positive peripheral blood polymorphonuclear leukocytes. J Clin Microbiol 1998; 36: 3585–3589. | PubMed | ChemPort |
19. Lozza L, Lilleri D, Percivalle E, Fornara C, Comolli G, Revello MG et al. Simultaneous quantification of human cytomegalovirus (HCMV)-specific CD4+ and CD8+ T cells by a novel method using monocyte-derived HCMV-infected immature dendritic cells. Eur J Immunol 2005; 35: 1795–1804. | Article | PubMed | ChemPort |
20. Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999; 18: 695–706. | Article | PubMed | ISI | ChemPort |
21. Klein JP, Rizzo JD, Zhang M-J, Keiding N. Statistical methods for analysis and presentation of the results of bone marrow transplants. Part I: unadjusted analysis. Bone Marrow Transplant 2001; 28: 909–915. | Article | PubMed | ISI | ChemPort |
22. Gerna G, Lilleri D, Zecca M, Alessandrino EP, Baldanti F, Revello MG et al. Rising antigenemia levels may be misleading in pre-emptive therapy of human cytomegalovirus infection in allogeneic hematopoietic stem cell transplant recipients. Haematologica 2005; 90: 526–533. | PubMed |
23. Gerna G, Sarasini A, Lilleri D, Percivalle E, Torsellini M, Baldanti F et al. In vitro model for the study of the dissociation of increasing antigenemia and decreasing DNAemia and viremia during treatment of human cytomegalovirus infection with ganciclovir in transplant recipients. J Infect Dis 2003; 188: 1639–1647. | Article | PubMed | ChemPort |
24. Verkruyse LA, Storch GA, Devine SM, Dipersio JF, Vij R. Once daily ganciclovir as initial pre-emptive therapy delayed until threshold CMV load>or =10 000 copies/ml: a safe and effective strategy for allogeneic stem cell transplant patients. Bone Marrow Transplant 2006; 37: 51–56. | Article | PubMed | ISI | ChemPort |
25. Lilleri D, Gerna G, Fornara C, Lozza L, Maccario R, Locatelli F. Prospective simultaneous quantification of human cytomegalovirus-specific CD4+ and CD8+ T-cell reconstitution in young recipients of allogeneic hematopoietic stem cell transplants. Blood 2006; 108: 1406–1412. | Article | PubMed | ChemPort |
26. Gerna G, Zavattoni M, Baldanti F, Sarasini A, Chezzi L, Grossi P et al. Human cytomegalovirus (HCMV) leukoDNAemia correlates more closely with clinical symptoms than antigenemia and viremia in heart and heart-lung transplant recipients with primary HCMV infection. Transplantation 1998; 65: 1378–1385. | Article | PubMed | ChemPort |

Acknowledgements

We thank all the technical staff of the Servizio di Virologia for performing the assays, Daniela Sartori for preparing the paper and Laurene Kelly for revision of English. This work was partially supported by grants from the Ministero della Salute, Fondazione IRCCS (Istituto di Ricovero e Cura a Carattere Scientifico) Policlinico San Matteo (Ricerca Corrente grant 80541, Ricerca Finalizzata 2003 grant 89269 and Fondazione Cariplo grant 93005).