Post-Transplant Complications

Monitoring four herpesviruses in unrelated cord blood transplantation

Article metrics


Cord blood transplantation, which has lower risk of graft-versus-host disease than bone marrow transplantation, might have higher risk of infections. A system to quantify four herpesviruses, CMV, human herpesvirus 6 (HHV6), EBV, varicella-zoster virus using the real-time PCR assay was established and applied for prospective viral load monitoring in three recipients undergoing cord blood transplantation. CMV and HHV6 were detected in peripheral blood from all three recipients, while EBV was detected in two. Varicella-zoster virus was not detected at all. At the peak of HHV6 or CMV, each patient showed virus-related symptoms. During the pre-transplant period, CMV DNA was detected in two recipients who later developed CMV-related diseases. These observations indicate that our system is not only useful for managing herpesviruses infections in transplant recipients, but also a powerful method for clarifying the relationships between the viral load and clinical symptoms. Bone Marrow Transplantation (2000) 26, 1193–1197.


Recently, the number of cord blood transplants (CBT) from related and unrelated donors has increased to treat malignant and nonmalignant disorders. The hematopoietic stem cells found in cord blood possess higher proliferative potential than adult bone marrow cells.1 CBT also has a lower risk for severe GVHD than bone marrow or peripheral blood stem cell transplantation.1 On the other hand, recipients of CBT may have higher risk of infections than recipients of other hematopoietic stem cell transplants (HSCT) because the cord blood contains few memory T cells to respond to exogenous antigens. Nuckols2 reported that infectious complications were frequently encountered at autopsy of CBT recipients. It was also reported that a positive serostatus for CMV was associated with worsened survival in CBT recipients.3

Meanwhile, herpesvirus infections cause serious complications in the management of HSCT recipients. Herpesviruses latently infect many individuals and become reactivated when the immune system is suppressed. Therefore, quantitative analysis is essential to diagnose symptomatic herpesvirus infections.45 We have established a system to quantify CMV, EBV, and varicella-zoster virus (VZV) genomes using a ‘real-time’ quantitative PCR method.678 This method measures the accumulation of PCR products with a fluorogenic probe and real-time laser scanning in a 96-well plate, and has a very wide dynamic range.9 Since this assay does not require additional handling, such as electrophoresis, faster assays with a higher throughput are possible. In this study, we newly established a system for human herpesvirus 6 (HHV6) and quantified four herpesviruses simultaneously in three recipients who underwent unrelated CBT. With this method, we observed how each viral burden correlated with its clinical manifestations in these recipients.

Patients and methods

Patient characteristics

Three recipients of unrelated CBT were enrolled in this study.

Patient A was a 14-year-old girl with AML; she has been reported elsewhere.8 She was inducted to the first CR with chemotherapy. However, a chromosomal aberration still remained in her bone marrow, so an autologus BMT was performed after pre-conditioning with Bu and melphalan. Four months after the autologus BMT, she relapsed. Neither a HLA-identical familial nor an unrelated donor was found. A donor serologically HLA− one locus mismatched was found in a cord blood bank in the United States. The patient received a CBT from this donor. She was pre-conditioned with 10 Gy of total body irradiation (TBI), etoposide (VP-16) 30 mg/kg once daily iv on days −6 and −5 (total dose 60 mg/kg) and CY 60 mg/kg once daily iv on days −4 and −3 (total dose 120 mg/kg). To prevent acute GVHD, CsA 3 mg/kg/day by continuous iv infusion from day −1, and short-term MTX 15 mg/m2 at day +1, 10 mg/m2 on days +3, +6, +11 and +18 (total dose 55 mg/m2) were administered. She received 3.2 × 107/kg nucleated cells.

Patient B was a 2-year-old boy with AML. He received chemotherapy and had an initial CR. Six months later, he had a relapse and chemotherapy produced a second CR. A donor serologically mismatched at one locus was found in a local cord blood bank in Japan, and he received an unrelated CBT. The patient was pre-conditioned with 12 Gy of TBI, Bu 35 mg/m2 every 6 h po on days −7 and −6 (total dose 280 mg/m2), and melphalan 70 mg/m2 once daily iv on days −5, −4 and −3 (total dose 210 mg/m2). He received 1.1 × 107/kg nucleated cells. To prevent acute GVHD, CsA 3 mg/kg/day continuous iv infusion from day −1 and short-term MTX iv 15 mg/m2 on day +1, 10 mg/m2 on days +3, +6, +11 (total dose 45 mg/m2) were administered.

Patient C, a 9-year-old boy, had adrenoleukodystrophy. He had received an unrelated CBT 5 months before at another institute after pre-conditioning with 8 Gy of TBI, CY (50 mg/kg/day for 4 days), and anti-thymocyte globulin (ATG, 2.5 mg/kg/day for 4 days). However, graft rejection occurred and he was transferred to our hospital to receive another unrelated CBT. A donor serologically mismatched at three loci was selected in a local cord blood bank in Japan. He was pre-conditioned with 7.5 Gy of total lymphoid irradiation, Bu 1 mg/kg po in divided doses daily from day −10 to day −7 (total dose 4 mg/kg), CY 50 mg/kg once daily iv from day −6 to day −3 (total dose 200 mg/kg), and ATG (2.5 mg/kg once daily iv from day −5 to day −2 (total dose 10 mg/kg)). Then he received 2.6 × 107/kg nucleated cells from a donor. To prevent acute GVHD, CsA and MTX were administered in the same regimen as patient B.

For prophylaxis against viral infections, all recipients were given CMV high-titer gamma globulin at 200 mg/kg weekly iv and aciclovir was administered orally from the pre-conditioning periods at 20 mg/kg/day. They were all seropositive for CMV, HHV6, EBV, and VZV before transplant.

Sample preparation

Informed consent was obtained and EDTA-treated peripheral blood was collected every 1 to 2 weeks.

Peripheral blood leukocytes (PBLs) were prepared as previously described.8 For the PCR assay, DNA was extracted from 106 PBL, using a QIAamp Blood Kit (Qiagen, Chatsworth, CA, USA), eluted in 50 μl of distilled water. Plasma was separated from EDTA-treated whole blood by centrifugation. DNA was extracted from 200 μl of plasma in the same manner.

For the pp65-antigenemia assay, PBLs were cytospun onto a slide glass. The antigenemia assay was performed as previously described.8 The number of positive cells was expressed per 1 × 106 PBLs.

Real-time quantitative PCR assay

The primers and probe settings for CMV, EBV, and VZV were as described in previous reports.678 The PCR primers and probe for HHV6 were selected from the U31 gene (large tegment protein). The upstream primer was 5′-TTTGCAGTCATCACGAT-CGG-3′, and downstream primer was 5′-AGAGCGACAAATTGGAGGTTTC-3′. A fluorogenic probe (5′-AGCCACAGCAGCCATCTACATCTGTCAA-3′) was located between the primers. The PCR reaction was performed using a TaqMan PCR kit (PE Applied Biosystems, Foster City, CA, USA) as previously described.7 A standard graph was constructed using the CT values obtained from serially diluted PSTY05 that contained the U31 gene.10 To estimate the amount of human genomic DNA in each PBL extract and to normalize the amount of viral DNA, human β-actin DNA was quantified by real-time PCR using TaqMan β-actin Control Reagent. From the amount of human β-actin gene, the number of PBLs in each sample was estimated, and the amount of viral DNA was adjusted and expressed per 106 cells. For plasma, the copy number was expressed per ml. The detection limits of each viral DNA were approximately 10 copies/106 cells and 50 copies/ml plasma, respectively.


Using the real-time PCR assay, we serially monitored the DNA load of CMV, HHV6, EBV, and VZV in blood samples from three unrelated CBT recipients. CMV and HHV6 were detected in all of the recipients. EBV was detected in two of three recipients, while no VZV was detected. All the grafts were negative for each virus. Figure 1 shows the change of each virus load and the clinical courses of the recipients.

Patient A (Figure 1a)

Figure 1

Clinical time-course and virus loads of three recipients of cord blood transplant. DNA copy number in PBLs and plasma are expressed per 106 cells and per ml, respectively. • = CMV DNA copy number in PBLs; = CMV DNA in plasma; ▪ = HHV6 DNA in PBLs; □ = HHV6 DNA in plasma; = EBV DNA in PBLS; = EBV DNA in plasma. VZV DNA was never detected in any case. (a) patient A; (b) patient B; (c) patient C. GCV, ganciclovir; PFA, foscarnet.

Patient A developed pyrexia in the second week. A blood culture performed at day +9 revealed that she had septicemia due to Streptococcus epidermidis. With antibiotics, she recovered and the pyrexia resolved. At the fourth week, she developed pyrexia again; however, repeated blood cultures were negative for bacterial and fungal growth. Both CMV and HHV6 DNA were detected in her PBL at this time. Additionally, conventional isolation for HHV6 from her PBL was positive in the second week. After receiving ganciclovir, CMV DNA became undetectable and the fever disappeared. On the other hand, HHV6 DNA remained positive until the end point of observation.

In this recipient, the neutrophil count reached over 5.0 × 105/ml on day +17, and the platelet count reached over 20 × 106/ml without transfusion on day +93. However, karyotypic analysis in her bone marrow demonstrated a relapse of AML in the fourth week. She died 19 months after the CBT with disseminated AML.

Patient B (Figure 1b)

Patient B developed transient pyrexia and skin rashes in the second week. During this period, HHV6 DNA was detected in both PBL (105.8 copies/106 cells) and plasma (104.1 copies/ml). Conventional isolation for HHV6 was also positive in PBL. GVHD was clinically suspected, however, histological examinations of the skin showed no specific findings suggesting GVHD. Although the skin rashes and pyrexia improved, the patient had severe diarrhea and gastrointestinal bleeding from the third week. CMV DNA appeared in his PBLs and remained detectable until the ninth week. From the sixth to the ninth week, CMV DNA was detected in his plasma, suggesting a high proliferation rate of CMV. After ganciclovir and foscarnet were administered, CMV DNA gradually disappeared in accordance with clinical improvements. A rectal biopsy done at the third week did not reveal any specific findings suggesting GVHD or viral infection. Considering the existence of cell-free CMV DNA in plasma and its dynamics, it seemed that his intestinal symptoms were related to CMV infections. EBV DNA was transiently detected, but the copy number was within the normal range of healthy seropositive individuals.7

In this recipient, the neutrophil count reached over 5.0 × 105/ml on day +21. The count of platelet reached over 20 × 106/ml without transfusion on day +85. Chimeric analysis by fluorescence in situ hybridization showed that karyotype of his bone marrow was complete donor type (100% XX) on day +21 and day +100. Patient B has been in CR and good condition for 21 months.

Patient C (Figure 1c)

From the pre-transplant period, CMV DNA was detected not only in his PBL, but also in his plasma. He became pyrexic during the pre-conditioning period. Although the pyrexia resolved spontaneously, other symptoms such as diarrhea and skin rashes appeared. Since the antigenemia positive cells were detected in his blood, ganciclovir was administered, followed by foscarnet. In the ninth week, he suddenly spiked a fever and developed respiratory distress. A chest roentgenogram showed interstitial pneumonitis. The respiratory distress rapidly progressed and the patient required mechanical ventilation. He died of respiratory failure 13 weeks after the transplant. An autopsy was not done. Just before his death, HHV6 DNA was detected in his PBL and viral culture was also positive for HHV6.

In this patient, the neutrophil count reached over 5.0 × 105/ml on day +25. He was never free of either erythrocyte or platelet transfusions until his death. Chimeric analysis by fluorescence in situ hybridization showed that karyotype of his bone marrow was completely donor type (100% XX) on day +53 and 98.8% on day +84.

Comparison between CMV DNA load and pp65 antigenemia assay

We previously reported that the copy number of CMV DNA in PBL and the count of pp65-antigen positive cells were significantly correlated.8 In this study, we prospectively performed and compared both the real-time PCR and pp65 antigenemia assays (Figure 2). In all cases, the results of both assays were almost parallel throughout each patient's course.

Figure 2

Serial comparison between the CMV DNA load and the antigenemia assay. Each value is expressed per 106 PBLs. • = CMV DNA copy number; □ = pp65 antigen-positive cell count.


In the management of HSCT recipients, the control of various viral infections is critical. Specifically, herpesviruses, such as CMV, EBV, herpes simplex virus (HSV), and VZV, have been the pathogens most likely to threaten the success of transplant. Therefore, many strategies to monitor and treat these infections have been developed and applied clinically.11121314 Meanwhile, latent viruses such as CMV or EBV are sometimes detected in specimens even from healthy seropositive individuals. Hence qualitative analysis of viral load is limited in its ability to monitor or diagnose these infections and diseases. On the other hand, the real-time PCR assay presented in this study is a quantitative system that requires minimal handling, has a wide dynamic range and high reproductivity, and so is useful for the management of HSCT recipients.

In this study, HHV6 and CMV infections were observed in all three recipients of unrelated CBT. Since the cord blood contains few memory T cells to respond to exogenous antigens, it may take longer to constitute pathogen-specific immunity. Although further studies with much larger sample sizes are required to define the frequency and severity of viral infections in unrelated CBT recipients, it is likely that severe herpesvirus infections could occur in patients who receive unrelated CBT. On the other hand, VZV was not detected in any recipients. Several hypotheses have been proposed to explain this: (1) VZV usually reactivates during the late transplant phase; (2) all the recipients received aciclovir prophylactically; and (3) VZV is a nerve-tropic virus and seldom appears in peripheral blood when it is reactivated. VZV-viremia seems less common than viremia of the other three herpesviruses in HSCT recipients.

The pathogenecity of HHV6 in HSCT recipients has been studied but not fully elucidated. Some studies suggested that HHV6 might cause skin rashes, fever and central nerve system manifestations in patients who receive HSCT.151617 The dynamics of the viral DNA load suggest that HHV6 may be related to the fever and rash that occurred in patient B. In patients B and C, CMV DNA was detected in the pre-transplant periods and both patients later developed CMV-associated symptoms. CMV reactivation and diseases often appear 1 to 2 months after HSCT. Meanwhile, Limaye et al18 reported CMV diseases before engraftment in bone marrow transplant recipients. The antigenemia assay could not be performed before engraftment because the leukocyte counts were too low. However, the real-time PCR could be done at this time, owing to high sensitivity of this assay and quantification of the human β-actin gene. The reactivation of CMV might occur more often than we expect before or during the early periods after HSCT. Evaluating the reactivation state of CMV in the pre-transplant period might be useful for predicting CMV diseases in HSCT recipients. This system is not only useful for managing herpesviruses infections in HSCT recipients, but also a powerful method for clarifying the relationships between viral load and clinical symptoms.


  1. 1

    Wagner JE . Umbilical cord transplantation Leukemia 1998 12: S30–32

  2. 2

    Nuckols JD . Autopsy findings in umbilical cord blood transplant recipients Am J Clin Pathol 1999 112: 335–342

  3. 3

    Gluckman E, Rocha V, Boyer-Chammard A et al. Outcome of cord-blood transplantation from related and unrelated donors. Eurocord Transplant Group and the European Blood and Marrow Transplantation Group New Engl J Med 1997 337: 373–381

  4. 4

    Shinkai M, Bozzette SA, Powderly W et al. Utility of urine and leukocyte cultures and plasma DNA polymerase chain reaction for identification of AIDS patients at risk for developing human cytomegalovirus disease J Infect Dis 1997 175: 302–308

  5. 5

    Boeckh M, Boivin G . Quantitation of cytomegalovirus: methodologic aspects and clinical applications Clin Microbiol Rev 1998 11: 533–554

  6. 6

    Kimura H, Wang Y, Pesnicak L et al. Recombinant varicella-zoster virus glycoproteins E and I: immunologic responses and clearance of virus in a guinea pig model of chronic uveitis J Infect Dis 1998 178: 310–317

  7. 7

    Kimura H, Morita M, Yabuta Y et al. Quantitative analysis of Epstein–Barr virus load by using a real-time PCR assay J Clin Microbiol 1999 37: 132–136

  8. 8

    Tanaka N, Kimura H, Iida K et al. Quantitative analysis of cytomegalovirus load using a real-time PCR assay J Med Virol 2000 60: 455–462

  9. 9

    Heid CA, Stevens J, Livak KJ, Williams PM . Real time quantitative PCR Genome Res 1996 6: 986–994

  10. 10

    Yalcin S, Karpuzoglu T, Suleymanlar G et al. Human herpesvirus 6 and human herpesvirus 7 infections in renal transplant recipients and healthy adults in Turkey Arch Virol 1994 136: 183–190

  11. 11

    Einsele H, Ehninger G, Hebart H 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

  12. 12

    Kanda Y, Chiba S, Suzuki T et al. Time course analysis of semi-quantitative PCR and antigenaemia assay for prevention of cytomegalovirus disease after bone marrow transplantation Br J Haematol 1998 100: 222–225

  13. 13

    Ljungman P . Immune reconstitution and viral infections after stem cell transplantation Bone Marrow Transplant 1998 21: S72–74

  14. 14

    O'Reilly RJ, Small TN, Papadopoulos E et al. Adoptive immunotherapy for Epstein–Barr virus-associated lymphoproliferative disorders complicating marrow allografts Springer Semin Immunopathol 1998 20: 455–491

  15. 15

    Kadakia MP, Rybka WB, Stewart JA et al. Human herpesvirus 6: infection and disease following autologous and allogeneic bone marrow transplantation Blood 1996 87: 5341–5354

  16. 16

    Yoshikawa T, Suzuki K, Ihira M et al. Prediction of human herpesvirus 6 infection after allogeneic bone marrow transplantation Blood 1998 92: 2597–2599

  17. 17

    Cone RW, Huang ML, Corey L et al. Human herpesvirus 6 infections after bone marrow transplantation: clinical and virologic manifestations J Infect Dis 1999 179: 311–318

  18. 18

    Limaye AP, Bowden RA, Myerson D, Boeckh M . Cytomegalovirus disease occurring before engraftment in marrow transplant recipients Clin Infect Dis 1997 24: 830–835

Download references


We thank Prof K Watanabe, Department of Pediatrics, Nagoya University School of Medicine, for helpful advice, and Prof K Yamanishi, Department of Microbiology, Osaka University School of Medicine, for a gift of PSTY05. This work was supported by a grant from Japan Society for the Promotion of Science (JSPS-RFTF97L00703) and also by a grant from Ministry of Health and Welfare, Health Science Research Grants, Research on Immunology, Allergy and Organ Transplantation (H11-Menneki-006).

Author information

Correspondence to N Tanaka.

Rights and permissions

Reprints and Permissions

About this article


  • real-time PCR
  • CMV
  • human herpesvirus 6 (HHV6)
  • EBV
  • cord blood transplantation (CBT)

Further reading