Materials and methods

Patients

Samples were taken from 26 allogeneic related BMT recipients at Namazi Hospital, Shiraz University of Medical Sciences, Shiraz, Iran between May 2002 and September 2003: 14 were male and 12 were female subjects; age range 6–42 years, median 19 years. Their underlying hematological diagnoses were leukemia (n=8), thalassemia major (n=14), Hodgkin's lymphoma (n=2) and aplastic anemia (n=2). HCMV serology status of donors (D⁺ or D^-) and recipients (R⁺ or R^-) was determined before transplantation by an ELISA method (G.E.N.E.S.I.S., England).

Surveillance samples were collected at 7–10 days prior to transplantation and then weekly from day 14 to 120 post transplantation. Clinical data were collected and analyzed prospectively by physicians who were blind to the laboratory data. Patients were conditioned according to a standard protocol. Patients with Hodgkin's lymphoma (n=2) received nonmyeloablative regimens, and others received myeloablative regimens. HCMV-related active diseases were defined according to Ljungman et al.¹³

Specimens

A total of 209 blood samples including peripheral blood mononuclear cells (PBMN) (n=177), PMN leukocytes (n=177) and plasma (n=209) from 26 BMT recipients were tested by an HCMV double primer PCR assay. Whole blood samples (n=177) were also tested by quantitative antigenemia assay.

Blood cell separation

The whole leukocyte fraction was isolated according to Stephen et al¹⁴ with a small modification. EDTA-anticoagulated blood was added to cold 0.84% NH₄Cl solution (1:4 v/v). After 10 min incubation at 4°C, the mixture was centrifuged at 3000 g for 7 min. The cell pellets were resuspended in 1 ml of cold 0.84% NH₄Cl for 5 min to rupture the remaining erythrocytes. After sedimentation, the leukocyte-enriched pellets were layered on top of 3–5 ml Ficoll–Hypaque density gradient medium (Ficoll, Lymphoprep, Nycomed, Netherlands) and centrifuged at 800 g for 25 min at room temperature. Mononuclear leukocytes were then isolated from the sediment, which contained PMN leukocytes. Cytospin slides were also obtained by centrifugation of 2 10⁵ leukocytes on glass slides at 500 g (Cytospin 3, Shandon, England). The slides were dried at room temperature and fixed in cold acetone for the pp65 antigen detection assay.

All the plasma samples and leukocyte fractions (ie PBMN and PMN) were frozen at -70°C to be used for double primer PCR assay.

HCMV pp65 antigenemia assay

HCMV antigenemia assays were performed according to the manufacturer's instructions. All samples were evaluated for the presence of pp65 antigen using Brite Turbo assay (IQ, Netherlands).

DNA extraction from leukocytes

Frozen 100 l aliquots of white blood cells (WBCs) were thawed and DNA was extracted by the boiling method and sodium acetate–ethanol precipitation. PBMN and granulocytes were heated separately at 95°C for 20 min. The samples were then centrifuged at 14 000 rpm for 10 min. The supernatants were removed and transferred to new tubes. Sodium acetate (3 M, 10 l) and absolute ethanol (300 l) were added to each tube. Tubes were centrifuged at 14 000 rpm for 10 min. The supernatant was removed and 50–100 l of TE buffer (pH 7.8) was added to the pellets.

DNA extraction from plasma

A 100 l portion of the plasma sample was used in the standard phenol–chloroform nucleic acid extraction method.¹⁵

PCR amplification

The quality of DNA preparation from all samples was checked by amplification of a 110-bp fragment of the -globin gene, using PCO₃ and PCO₄ primers.¹⁶ For qualitative HCMV DNA detection, two pairs of oligonucleotide primers specific for the fourth exon of the HCMV immediate early gene (IE) and HCMV late antigen (LA) were used in a multiplex format to generate fragments of 393 and 136 bp, respectively.¹² A 5 l portion of extracted DNA was added to 45 l of a PCR mixture. All PCR reaction mixtures contained 120 mM KCl, 24 mM Tris-HCl pH 8.3, 200 M of each deoxynucleoside triphosphate, 2 mM MgCl₂, 25 pmol of each primer and 2 U of Taq DNA polymerase in a final volume of 50 l. Thermocycling conditions were set up to an initial step or denaturation step (at 94°C for 3 min), followed by 35 cycles (at 94, 57 and 72°C), of which each lasted 1 min, the third with a 5-min extension at 72°C. The same process was repeated for amplification of the -globin sequence. Subsequently, 10 l of each amplified product was analyzed by conventional gel electrophoresis followed by ethidium bromide staining (10 g/ml). Samples were retested in the event of positive results.

Each PCR processing procedure was performed in a physically separate room to prevent contamination. Negative controls were used to ensure that carryover contamination would not occur.

Determination of PCR sensitivity

DNA extracted from 10-fold serial dilution of plasmid DNA with 1 pg/l concentration (supplied by Cinna Gennic, Tehran, Iran) was added to plasma or leukocyte samples (negative for HCMV DNA) to be used as a template for the PCR reaction. Conventional gel electrophoresis and ethidium bromide staining were then used to analyze the PCR amplification products.

Determination of PCR specificity

PCR specificity was determined using DNA extracted from the herpesvirus group including HSV-1, HSV-2, VZV, EBV and HHV-6 and from uninfected Vero and Hep-II cell lines.

Statistical analysis

Using SPSS software version 11.5, ², two-tailed Fisher's exact test, bivariate and normal correlation methods were performed to analyze data and determination of any correlation between the presence of HCMV DNA in clinical samples and appearance of HCMV-associated signs and symptoms in patients.

Results

Determining the sensitivity and specificity of PCR

The PCR detection limit, which was evaluated by gel electrophoresis and ethidium bromide staining, was 500 copies of HCMV DNA per ml of specimen. The PCR specificity was determined using DNA extracted from a herpesvirus group other than HCMV. No amplification was noted by gel electrophoresis analysis.

HCMV antibody status

HCMV antibody was positive in 23 (88%) out of 26 donors and in 24 (92.3%) out of 26 BMT recipients. Overall, the pattern of HCMV antibody in both BMT donors and recipients was R^-/D⁺=2, R⁺/D^-=3, R⁺/D⁺=21 and R^-/D^-=0.

Monitoring of HCMV infection in BMT recipients by double primer PCR assay and by antigenemia

The double primer PCR assay was applied to 563 clinical samples of leukocytes and plasma. HCMV DNA was detected in 112 (19.9%) out of 563 samples. Of the 112 positive samples for HCMV DNA, five (4.5%) samples were only positive for the LA fragment, eight (7.1%) samples were only positive for the IE fragment and 99 (88.4%) samples were positive for both IE and LA fragments. Double primer PCR assay increased the number of positive samples by an average of 11.6%. An increase in the number of leukocytes (mean leukocyte number 4.7 10³/l) also increased the likelihood of detection of both IE and LA fragments in samples. The mean leukocyte number in samples, in which IE or LA fragments were detected, was 0.8 10³ and 1.2 10³/l, respectively.

HCMV infection was detected in 16 out of 26 (61.5%) BMT patients during the monitoring period by either double primer PCR or antigenemia assay. Two HCMV-seronegative recipients received bone marrow from HCMV-seropositive donors and both of them developed HCMV disease during the study.

The double primer PCR assay detected HCMV DNA in the plasma of 12 (46.1%), PBMN of nine (34.6%) and PMN leukocytes of 14 (53.8%) patients out of the 26 BMT patients. In six (42.8%) of the 14 patients with positive PMN leukocytes, HCMV DNA was detected in their plasma samples 7–10 days before detection of DNA in PMN leukocytes. Meanwhile, the antigenemia assay was also positive (ie >4 granulocyte/2 10⁵ leukocytes) in 14 (53.8%) out of the 26 patients. In the 14 patients with HCMV disease, no other microbiological findings were detected.

The results obtained with the double primer PCR assay were compared with the antigenemia assay results, and a significant association between results of the double primer PCR assay of PMN leukocytes and antigenemia for detection of active HCMV infection in all of the patients was found (r=1).

A diagnosis of symptomatic active HCMV infection was made¹² in 14 out of 26 patients (53.8%). Of the 14 HCMV antigenemia and PMN leukocyte double primer PCR assay positive patients, 10 developed active disease including fever with leukopenia and pneumonia, and the remainder showed signs and symptoms including fever, gastritis, leukopenia and skin rash without respiratory symptoms. The clinical signs and symptoms of this group of patients are presented in Table 1. Significant association between a positive antigenemia assay and positive PMN leukocyte double primer PCR assay with development of pneumonia was identified (P=0.05). In one asymptomatic patient, HCMV DNA was detected in both plasma and PBMN leukocytes. However, HCMV infection was determined neither by the PMN leukocyte double primer PCR assay nor by the antigenemia assay.

Table 1 - Results of HCMV double primer PCR and antigenemia assays and their relation to clinical findings in bone marrow transplant patients.

Full table

Since physicians were blind to the laboratory data, in the case of HCMV disease, ganciclovir therapy was instituted on clinical grounds. In all patients treated with ganciclovir, the double primer PCR assay of PMN leukocytes and antigenemia assay became negative 10–14 days from the end of therapy. However, in some cases, HCMV DNA persisted in plasma even up to 30 days from the end of ganciclovir therapy. Only one (7.14%) of 14 HCMV symptomatic patient expired at day 90 post transplantation.

Discussion

Because seroprevalence of HCMV is very high (85–98%) and the primary infection is uncommon in BMT recipients, antiviral therapy is usually based on the clinical picture and laboratory diagnosis. Successful pre-emptive HCMV therapy in BMT patients depends on availability of a sensitive, specific and appropriate diagnostic test for detection of active HCMV infection. The antigenemia assay is an indicator of an active HCMV infection, but its use is limited to blood samples, which contain PMN leukocytes. This test should be performed within a 4-h period.

The aim of this study was to evaluate the sensitivity and specificity of a double primer PCR assay of plasma and leukocytes in BMT recipients and compare the PCR assay with the antigenemia assay and with the clinical situation.

We demonstrated that there is a positive direct correlation between positive HCMV antigenemia and HCMV DNA detection in PMN leukocytes by the PCR assay. Similar results have been already reported using real-time PCR.^{9, 17, 18}

In this study, the PCR assay was easily capable of detecting 500 HCMV genomes/ml of clinical samples without nested PCR or hybridization assays. The sensitivity of multiplex PCR assays for detection of HCMV DNA in clinical samples of renal transplant recipients has already been described.¹² It has been shown that the method is highly sensitive for identification of patients at risk of clinically significant infection by HCMV.

The overall incidence of HCMV viremia in this study was high: 61.5% of patients demonstrated at least one positive test sample. A possible explanation for this high rate is that positive results might be increased by using a double primer PCR assay. We have demonstrated that use of a double primer assay increases the number of positive samples by an average of 11.6% compared with using a monoprimer PCR assay. Similar results have been reported¹¹ using the multiplex HCMV PCR assay for detection of active HCMV disease in renal transplant patients.¹¹ We confirmed that an increase in the total number of leukocytes will increase simultaneous detection of IE and LA fragments in samples. In three samples from the 14 patients with HCMV disease, IE fragments, and in two other samples, LA fragments were solely detected. In the remaining samples, both IE and LA fragments were detected concurrently. There have been many studies evaluating the clinical utility of a PCR-based assay for detection of HCMV DNA.^{19, 20, 21} Many of these investigations have shown that the assay may be sensitive but has not enough clinical specificity to detect HCMV in patients who do not have active HCMV disease. However, we demonstrated a significant association between double primer PCR assay results of PMN leukocytes and antigenemia results in relation to clinical status of BMT patients. Furthermore, whenever the double primer PCR assay of PMN leukocytes is positive, development of active HCMV disease is very likely.

In our study, there was an asymptomatic patient with HCMV infection in whom HCMV DNA was detected in both his plasma and PBMN leukocytes. However, HCMV infection was diagnosed neither by the PMN leukocyte double primer PCR assay nor by the antigenemia assay. Our findings confirm that viral DNA detection in PMN leukocytes is associated with active HCMV infection. In this study, 14 out of 26 BMT recipients developed active HCMV infection, but the clinical features of HCMV disease were different in this group of patients. Of the symptomatic patients, 10 patients had HCMV pneumonia and the rest had other signs and symptoms such as fever, gastritis, leukopenia and skin rash without respiratory manifestations.

The data in this study have shown the simultaneous presence of HCMV DNA in more than 75% of plasma samples and leukocytes of BMT patients. The data also indicate that molecular detection of active HCMV infection is more sensitive when a double primer PCR assay is applied to PMNL rather than to plasma samples. Compared with the antigenemia assay, the relative sensitivities of the PCR assay with PMN leukocytes and plasma were 100 and 87.5%, respectively. The specificities of the PCR assay with PMN leukocytes and with plasma samples were 100 and 85.7%, respectively.

In leukopenic BMT recipients, plasma samples would be a good mechanism for monitoring HCMV activity. Because plasma PCR is usually positive in more than 42% of patients a few days before detection of HCMV DNA in PMN leukocytes, monitoring of HCMV activity by plasma PCR might lead to earlier prescribing of an antiviral drug and a reduction in mortality.

In conclusion, our results suggest that detection of HCMV DNA in PMN leukocytes of BMT patients by a double primer PCR assay might be an alternative method to an antigenemia assay. Serial blood samples positive with a double primer PCR assay or antigenemia assay are an indication for pre-emptive therapy in this group of patients. The double primer PCR assay indicates an active HCMV infection compared with antigenemia assay and the need to start antiviral therapy. However, quantitative PCR methods are necessary for monitoring antiviral treatment.

References

References

Acknowledgements

This project was financially supported by Tarbiat Moddares University, Tehran Iran and Shiraz University of Medical Sciences, Shiraz, Iran.