Skip Navigation

Ganciclovir Is Associated with Low or Undetectable Epstein-Barr Virus DNA Load in Cerebrospinal Fluid of Patients with HIV-Related Primary Central Nervous System Lymphoma

  1. Simona Bossolasco1,
  2. Kerstin I. Falk3,4,
  3. Maurilio Ponzoni2,
  4. Norberto Ceserani1,
  5. Fulvio Crippa1,
  6. Adriano Lazzarin1,
  7. Annika Linde3,4, and
  8. Paola Cinque1
  1. 1Clinic of Infectious Diseases, San Raffaele Hospital, Milan, Italy
  2. 2Department of Pathology, San Raffaele Hospital, Milan, Italy
  3. 3Department of Virology, Swedish Institute for Infectious Disease Control, Karolinska Institutet, Stockholm, Sweden
  4. 4Microbiology and Tumor Biology Center, Karolinska Institutet, Stockholm, Sweden
  1. Reprints or correspondence: Dr. Simona Bossolasco, Clinic of Infectious Diseases, San Raffaele Hospital, Via Stamira d'Ancona, 20, 20127 Milano, Italy (bossolasco.simona{at}hsr.it).
 
Next Section

Abstract

Background. Epstein-Barr virus (EBV) is pathogenically linked to human immunodeficiency virus (HIV)–related primary central nervous system lymphoma (PCNSL) and is found in virtually all HIV-related PCNSL cases. The objective of this study was to assess the effect of ganciclovir on EBV DNA replication in patients with HIV-related PCNSL.

Patients and methods. EBV DNA was measured by real-time polymerase chain reaction in cerebrospinal fluid and plasma samples from 25 patients with HIV-related PCNSL. Eight of these patients were receiving ganciclovir for concurrent cytomegalovirus infections.

Results. EBV DNA was detected in cerebrospinal fluid samples obtained from 15 (88%) of 17 ganciclovir-untreated patients and 4 (50%) of 8 ganciclovir-treated patients (P = .028). EBV DNA load was significantly lower for treated patients, compared with untreated patients (median value, 2.15 vs. 4.16 log copies/mL; P = .001). Analysis of sequential cerebrospinal fluid samples from 7 patients showed that EBV DNA decreased in samples obtained from 2 patients following the start of ganciclovir administration but did not decrease in samples obtained from the 5 untreated patients. In addition, patients who received ganciclovir survived longer than the untreated patients (median duration of survival, 181 vs. 72 days; P = .006).

Conclusion. The effect of ganciclovir on EBV DNA load in cerebrospinal fluid supports the hypothesis that EBV is replicating in patients with PCNSL. This observation, together with the effect of ganciclovir therapy on patient survival, suggests that this drug might be useful for the management of PCNSL.

HIV-related primary CNS lymphoma (PCNSL) is associated with Epstein-Barr virus (EBV) infection, which is detected in tumor tissues in almost 100% of cases [1, 2]. EBV DNA is found in the CSF of 90%–100% of patients with PCNSL [2]. Recently, several studies have demonstrated the presence of lytic cycle EBV proteins or transcripts in the tissue and CSF of patients with HIV-related PCNSL. This suggests the possibility of replicative infection in this type of tumor [3–7]. Antiviral agents currently in use, including ganciclovir (GCV), cidofovir, foscarnet, and, to a lesser extent, acyclovir and penciclovir, can inhibit EBV replication [8–11]. In a few instances, antiviral agents have also been used in combination with antiretroviral drugs and/or chemotherapy and radiotherapy for treatment of PCNSL [12–16].

The objective of this study was to assess the antiviral effect of GCV by measuring EBV DNA levels in CSF and plasma samples obtained from patients with HIV-related PCNSL who were treated or untreated with this drug. In addition, the difference in survival between GCV-treated and untreated patients was also evaluated.

Previous SectionNext Section

Materials and Methods

Patients and samples. CSF specimens from 25 HIV-infected patients with PCNSL that had been admitted at San Raffaele Hospital (Milan, Italy) between 1994 and 2000 were retrospectively examined. Plasma samples from 15 of these 25 patients were also available. PCNSL was diagnosed by histopathological examination of tissues obtained postmortem (10 patients); by stereotactic brain biopsy (5 patients); or by clinical criteria including presence of focal contrast—enhancing brain lesions identified by CT or MRI, lack of response to antitoxoplasmic treatment, and a hypercaptation pattern identified by thallium-201 single-positron emission CT (10 patients) (table 1) [17, 18]. CSF and plasma samples were obtained for diagnostic purposes after obtaining informed consent from the patient. Aliquots of the samples were stored at -80°C until tested.

Figure 1
Figure 1

Epstein-Barr virus (EBV) DNA load in CSF and plasma samples obtained from patients treated with ganciclovir (GCV) and GCV-untreated patients with AIDS-related primary CNS lymphoma. Dotted line, detection limit of the PCR assay; horizontal line, median EBV DNA level.

Figure 2
Figure 2

Kaplan-Meyer survival curves for patients treated with ganciclovir (GCV) and GCV-untreated patients. Median duration of survival after onset of primary CNS lymphoma was 181 days (interquartile range, 128–251 days) for GCV-treated patients and 72 days (interquartile range, 51–92 days) for GCV-untreated patients. GCV-treated patients received GCV for a median of 272 days (interquartile range, 67–412 days) following onset of lymphoma.

Table 1
Table 1

Demographic and clinical characteristics of patients with primary CNS lymphoma (PCNSL) at the time of diagnosis.

Eight of the 25 patients were receiving GCV treatment for concomitant cytomegalovirus (CMV) infection at the time that samples were obtained (table 1). None of the 17 remaining patients received drugs active against herpesviruses in the 4 weeks prior to sampling. Twelve patients were receiving antiretroviral therapy (ART) at the time of CSF sampling, consisting of HAART in 5 patients, single-drug ART in 6 patients, and dual-drug ART in 1 patient (table 1).

Follow-up CSF specimens were available from 7 of the 25 patients; these were drawn 20–96 days after first sampling. Two of these 7 patients initiated GCV treatment, 70 and 76 days before the follow-up sampling, after drawing the first CSF sample. Two of the 5 remaining patients discontinued GCV treatment 48 and 19 days before follow-up sampling, and the other 3 were untreated.

Only treatments administered after the first CSF samples were obtained were considered for analysis of survival: 10 patients received GCV (for ⩾53 days), 2 received antiretroviral monotherapy (zidovudine or didanosine, for ⩾98 days), 2 received HAART (stavudine, lamivudine, and indinavir, for ⩾116 days), 5 received radiotherapy (total dose, 45–50 Gy), and 1 received chemotherapy (2 cycles of methotrexate, administered at a dosage of 1.5 g/m2 twice per week every 3 weeks).

Nucleic acid extraction and real-time PCR. Real-time PCR was used to measure CSF and plasma EBV DNA levels following nucleic acid extraction, according to a previously described method [19].

Statistical analysis. The Mann-Whitney U test was used to compare the EBV DNA load in the different patient groups. Spearman correlation and the χ2 test were used to test associations between continuous or categorical variables, respectively. Median values and 25th–75th percentile values are presented for continuous variables. The Cox proportional hazards regression model was used to evaluate the association between survival and continuous variables in a univariate analysis. The Kaplan—Meier method was used to estimate the distribution of survival times. The log-rank test was used for comparisons between survival curves. Negative EBV DNA PCR values were defined for statistical purposes as equal to 2 log copies/mL.

Previous SectionNext Section

Results

EBV DNA was detected in CSF samples from 15 (88%) of 17 GCV-untreated patients and from 4 (50%) of 8 treated patients (P = .028). The viral load was significantly lower in CSF samples from GCV-treated patients, compared with untreated patients (median viral load, 2.15 vs. 4.16 log copies/mL; P = .001) (figure 1). EBV DNA was found in the plasma samples from 1 (20%) of 5 GCV-treated and 3 (33%) of 9 GCV-untreated patients (P = not significant). This represented no significant difference in EBV DNA levels between the 2 groups (median EBV DNA level, 2.00 log copies/mL in both groups). EBV DNA levels in CSF samples did not correlate with plasma levels or CSF cell counts. Twelve patients were receiving ART at the time of CSF sampling. EBV DNA was detected in the CSF samples obtained from 5 (71%) of 7 patients receiving mono or dual ART, 3 (60%) of 5 HAART-treated patients, and 11 (85%) of 13 untreated patients. The levels did not differ significantly between the groups (median EBV DNA level, 3.36 vs. 3.67 log copies/mL).

Among 7 patients with repeated CSF measurements, EBV DNA decreased a median of 2.43 log copies/mL in 2 patients receiving GCV, 1 of whom had also received radiotherapy (total dose, 45 Gy) and HAART (stavudine, lamivudine, and indinavir). The viral load remained stable in 3 GCV-untreated patients (median viral load at baseline, 4.59 log copies/mL; median viral load at follow-up, 4.79 log copies/mL) and increased a median of 3.24 log copies/mL in the 2 patients who discontinued GCV treatment before the second sampling. One of these 5 patients received radiotherapy (total dose, 49 Gy), and another received chemotherapy. Trends did not differ between patients who received and those who did not receive treatments other than GCV with regard to observed EBV DNA levels.

All but 1 patient died, and the deaths appeared to be related to PCNSL in all of the cases. Overall, the median duration of survival was 85 days (interquartile range, 47–145 days) from the first CSF sampling and 104 days (interquartile range, 72–198 days) from the time of PCNSL onset (i.e., the time when PCNSL lesions appeared in neuroradiological imaging). A number of variables were analyzed with respect to survival time, starting from either the time of CSF sampling or of PCNSL onset. These included EBV DNA levels in the first CSF sample and the use of GCV, any ART, HAART, radiotherapy, or chemotherapy following the first CSF sampling. EBV DNA levels neither correlated with duration of survival as a continuous nor as a dichotomous variable using a cut-off value of 3.51 log copies/mL (the median value of EBV DNA levels in the first CSF sample). No significant correlation was observed between survival time and the receipt of ART, HAART, radiotherapy, or chemotherapy after CSF sampling or lymphoma onset. In contrast, treatment with GCV was significantly associated with a longer duration of survival measured from PCNSL onset (P < .006; figure 2).

Previous SectionNext Section

Discussion

We found that treatment with GCV was significantly associated with low or undetectable EBV DNA levels in CSF by applying a real-time PCR technique to the retrospective study of CSF samples obtained from patients with HIV-related PCNSL. Although EBV DNA detection has been used for the clinical management of HIV-related PCNSL for many years, it has only recently been suggested that EBV DNA levels might be influenced by antiviral treatments [6].

GCV is a nucleoside analogue drug that possesses potent antiviral activity against the lytic phase of EBV in vitro [10]. It is generally agreed that EBV-associated lymphomas express only latent EBV genes, and this theoretically makes antiviral treatment ineffective. However, several studies have shown that viral replicative genes are expressed in tumor cells [3–7, 20, 21]. More recently, lytic cycle mRNA was shown to be expressed in the CSF of patients with brain lymphoma [6], suggesting that these tumors, or at least some tumor cells, might undergo lytic viral replication [3, 7–21].

Although we did not measured mRNA from replicative EBV genes, our findings on the effect of GCV on EBV DNA load in the CSF of patients with HIV-related PCNSL strongly support the hypothesis that EBV replication occurs in this type of tumor. In addition, the observation that CSF EBV DNA load neither correlated with EBV DNA plasma levels nor was associated with high numbers of blood-derived CSF leukocytes is consistent with the intrathecal origin—thus tumoral origin—of the viral genomes.

There are studies suggesting the usefulness of antiviral agents for the treatment of EBV-related lymphomas [12–22]. GCV and foscarnet have been variably used in association with other treatments (including HAART, chemotherapy, or radiotherapy) in both brain and systemic HIV-related lymphomas and have been associated with clinical remission or prolonged survival time in isolated case reports [12–16]. In the present study, we found that GCV treatment was significantly associated with a longer survival time. Thus, it seems that this drug had some effects on both EBV replication in tumor cells and disease progression. However, it should also be noted that EBV DNA levels in CSF were not found to directly correlate with survival. This apparent discrepancy might be explained by the fact that some of the patients who were not treated with GCV at the time of sampling received this treatment afterwards.

The introduction of HAART to clinical practice has led to a decrease in PCNSL incidence and to prolonged survival [23–26]. In our study, we observed that HAART had no effect on EBV DNA levels in CSF and no effect on survival. It is possible that the efficacy of HAART was limited among our patients because of their poor viroimmunological response to treatment and the low number of treated patients. On the other hand, it has recently been observed that EBV levels are stable (or even increase) in peripheral blood cells and plasma after HAART initiation [27]. In agreement with our observations, a larger study has recently shown a low overall effect of HAART on survival among patients with PCNSL [28].

In conclusion, treatment with GCV was associated with both reduced EBV DNA load in CSF and prolonged survival among patients with HIV-related PCNSL. This study has the limitation of being based on a small, retrospective, observational case series in which 40% of PCNSL cases were diagnosed according to clinical criteria alone. Nevertheless, these findings might represent a rationale for evaluating, in larger controlled studies, the use of antiviral drugs associated with conventional treatment of HIV-related lymphomas.

Previous SectionNext Section

Acknowledgments

Financial support. Istituto Superiore di Sanità (Rome, Italy).

Potential conflicts of interest. All authors: no conflicts.

  • Received September 19, 2005.
  • Accepted November 11, 2005.
Previous Section
 

References

  1. ↵
    1. MacMahon EM,
    2. Glass JD,
    3. Hayward SD,
    4. et al
    . Epstein-Barr virus in AIDS-related primary central nervous system lymphoma. Lancet 1991;338:969-74.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Cinque P,
    2. Brytting M,
    3. Vago L,
    4. et al
    . Epstein-Barr virus DNA in cerebrospinal fluid from patients with AIDS-related primary lymphoma of the central nervous system. Lancet 1993;342:398-401.
    CrossRefMedlineWeb of Science
  3. ↵
    1. Rea D,
    2. Delecluse HJ,
    3. Hamilton-Dutoit SJ,
    4. et al
    . Epstein-Barr virus latent and replicative gene expression in post-transplant lymphoproliferative disorders and AIDS-related non-Hodgkin's lymphomas. Ann Oncol 1994;5(Suppl 1):113-6.
    Abstract/FREE Full Text
    1. Bashir RM,
    2. Hochberg FH,
    3. Harris NL,
    4. Purtilo D
    . Variable expression of Epstein-Barr virus genome as demonstrated by in situ hybridization in central nervous system lymphomas in immunocompromised patients. Mod Pathol 1990;3:429-34.
    MedlineWeb of Science
    1. Pallesen G,
    2. Hamilton-Dutoit SJ,
    3. Rowe M,
    4. et al
    . Expression of EBV replicative proteins in AIDS related non-Hodgkin's lymphoma cells. J Pathol 1991;165:289-99.
    CrossRefMedlineWeb of Science
  6. ↵
    1. Weinberg A,
    2. Li S,
    3. Palmer M,
    4. Tyler KL
    . Quantitative CSF PCR in Epstein-Barr virus infections of the central nervous system. Ann Neurol 2002;52:543-8.
    CrossRefMedlineWeb of Science
  7. ↵
    1. Touitou R,
    2. Arbach H,
    3. Cochet C,
    4. et al
    . Heterogeneous Epstein-Barr virus latent gene expression in AIDS-associated lymphomas and in type I Burkitt's cell lines. J Gen Virol 2003;84:949-57.
    Abstract/FREE Full Text
  8. ↵
    1. Lin J-C,
    2. Smith M-C,
    3. Pagano J-S
    . Prolonged inhibitory effect of 9-(1,3 dihydroxy-2-propoxymethyl)guanine against replication of Epstein-Barr virus. J Virol 1984;50:50-5.
    Abstract/FREE Full Text
    1. Colby B-M,
    2. Shaw J-E,
    3. Elion G-B,
    4. Pagano J-S
    . Effect of acyclovir [9-(2-hydroxyethoxymethyl)guanine] on Epstein-Barr virus DNA replication. J Virol 1980;34:560-8.
    Abstract/FREE Full Text
  10. ↵
    1. Lin J-C,
    2. Nelson D-J,
    3. Lambe C-U,
    4. Cho E-I
    . Metabolic activation of 9([2-hydroxy-1-hydroxymethyl)ethoxy]methyl)guanine in human lymphoblastoid cell lines infected with Epstein-Barr virus. J Virol 1986;60:569-73.
    Abstract/FREE Full Text
  11. ↵
    1. Bacon TH,
    2. Boyd MR
    . Activity of penciclovir against Epstein-Barr virus. Antimicrob Agents Chemother 1995;39:1599-602.
    Abstract/FREE Full Text
  12. ↵
    1. Brockmeyer N-H,
    2. Pohl G,
    3. Mertins L
    . Combination of chemotherapy and antiviral therapy for Epstein-Barr virus-associated non Hodgkin's lymphoma of high grade malignancy in cases of HIV infection. Eur J Med Res 1997;2:133-5.
    Medline
    1. Raez L,
    2. Cabral L,
    3. Cai JP,
    4. et al
    . Treatment of AIDS-related primary central nervous system lymphoma with zidovudine, ganciclovir and interleukin 2. AIDS Research and Human Retroviruses 1999;15:713-9.
    CrossRefMedlineWeb of Science
    1. Meerbach A,
    2. Holy A,
    3. Wutzler P,
    4. De Clercq E,
    5. Neyts J
    . Inhibitory effects of novel nucleoside and nucleotide analogues on Epstein-Barr virus replication. Antivir Chem Chemother 1998;9:275-82.
    MedlineWeb of Science
    1. Schmidt W,
    2. Anagnostopoulos I,
    3. Scherubl H
    . Virostatic therapy for advanced lymphoproliferation associated with the Epstein-Barr virus in an HIV-infected patients. N Engl J Med 2000;342:440-1.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Schneider U,
    2. Ruhnke M,
    3. Delecluse HJ,
    4. Stein H,
    5. Huhn D
    . Regression of Epstein-Barr virus-associated lymphoproliferative disorders in patients with acquired immunodeficiency syndrome during therapy with foscarnet. Ann Hematol 2000;79:214-6.
    CrossRefMedlineWeb of Science
  17. ↵
    1. Jaffe ES,
    2. Harris NL,
    3. Stein H,
    4. Vardiman JW
    . WHO classification of tumors of haematopoietic and lymphoid tissues. Lyon, France: IARC Press; 2001.
  18. ↵
    1. Castagna A,
    2. Cinque P,
    3. d'Amico A,
    4. Messa C,
    5. Fazio F,
    6. Lazzarin A
    . Evaluation of contrast-enhancing brain lesions in AIDS patients by means of Epstein-Barr virus detection in cerebrospinal fluid and 201thallium single photon emission tomography. AIDS 1997;11:1522-3.
    MedlineWeb of Science
  19. ↵
    1. Enbom M,
    2. Strand A,
    3. Falk KI,
    4. Linde A
    . Detection of EBV, but not HHV-8, DNA in cervical secretions from Swedish women by real-time PCR. Sex Transm Dis 2001;28:300-6.
    CrossRefMedlineWeb of Science
  20. ↵
    1. Katz B,
    2. Raab-Traub N,
    3. Miller G
    . Latent and replicating forms of Epstein-Barr virus DNA in lymphomas and lymphoproliferative diseases. J Infect Dis 1989;160:589-98.
    Abstract/FREE Full Text
  21. ↵
    1. Rea D,
    2. Fourcade C,
    3. Leblond V,
    4. et al
    . Patterns of Epstein-Barr virus latent and replicative gene expression in Epstein-Barr virus B cell lymphoproliferative disorders after organ transplantation. Transplantation 1994;58:317-24.
    MedlineWeb of Science
  22. ↵
    1. Rabkin CS
    . Editorial response: the need of investigations of prophylactic regimens to prevent AIDS-associated Non-Hodgkin's lymphoma. Clin Infect Dis 2000;30:762-3.
    FREE Full Text
  23. ↵
    1. Skiest D-J,
    2. Crosby C
    . Survival is prolonged by highly active antiretroviral therapy in AIDS patients with primary central nervous system lymphoma. AIDS 2003;17:1787-93.
    CrossRefMedlineWeb of Science
    1. Hoffmann C,
    2. Tabrizian S,
    3. Wolf E,
    4. et al
    . Survival of AIDS patients with primary central nervous system lymphoma is dramatically improved by HAART-induced immune recovery. AIDS 2001;15:2119-27.
    CrossRefMedlineWeb of Science
    1. McGowen JP,
    2. Shah S
    . Long-term remission of AIDS-related primary central nervous system lymphoma associated with highly active antiretroviral therapy. AIDS 1998;28:952-4.
  26. ↵
    1. Corales R,
    2. Taege A,
    3. Rehm S,
    4. Schmitt S
    . Regression of AIDS related CNS lymphoma with HAART [abstract MoPpB1086]. XIII International AIDS conference (Durban, South Africa, July 2000).
  27. ↵
    1. Righetti E,
    2. Ballon G,
    3. Ometto L,
    4. et al
    . Dynamics of Epstein-Barr virus in HIV-1-infected subjects on highly active antiretroviral therapy. AIDS 2002;16:63-73.
    CrossRefMedlineWeb of Science
  28. ↵
    1. Cingolati A,
    2. Cinque P,
    3. Fratino L,
    4. et al
    . 12th Conference on Retroviruses and Opportunistic Infections (Boston, 22–25 February 2005). Does HAART really improbe survival of patients with AIDS-associated primary brain lymphoma [abstract 907]?
| Table of Contents

This Article

  1. Clin Infect Dis. (2006) 42 (4): e21-e25. doi: 10.1086/499956
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most