- Email: [email protected]
A Single-center Experience With Ganciclovir-resistant Cytomegalovirus in Lung Transplant Recipients: Treatment and Outcome
CLINICAL LUNG AND HEART/LUNG TRANSPLANTATION
A Single-center Experience With Ganciclovir-resistant Cytomegalovirus in Lung Transplant Recipients: Treatment and Outcome Anita J. Reddy, MD,a Aimee K. Zaas, MD,b Kimberly E. Hanson, MD, MHS,b and Scott M. Palmer, MD, MHSa Background: Cytomegalovirus (CMV) disease is a major cause of morbidity and mortality after lung transplantation despite ganciclovir prophylaxis. The emergence of ganciclovir-resistant CMV in lung transplant patients has been reported, although the optimal strategy for the management of these infections remains uncertain. A review of the results of glanciclovir susceptibility testing in lung transplant recipients was performed. Methods: We found 54% (113 of 210) of lung transplant patients developed CMV infection over a 4-year study period with ganciclovir-resistant CMV infection occurring in ⬎5% of patients (6 of 113). The demographic and clinical characteristics of patients who developed ganciclovir-resistant vs -sensitive CMV infection were similar, although 50% (3 of 6) patients who developed resistance were CMV mismatched (D⫹/R⫺ serology). All patients’ CMV isolates had mutations in the UL97 gene. In addition, the 3 mismatch patients also had CMV with mutations in the UL54 gene. Results: Treatment with a combination of foscarnet and ganciclovir or foscarnet alone for ganciclovir-resistant infection led to a significant reduction in virologic load in all patients (p ⫽ 0.03), although transient increases in viremia were observed in some patients early after treatment. Renal function worsened after treatment, but overall it was not significantly different from pre-treatment values (p ⫽ 0.07). Conclusions: Single or combination therapy with foscarnet is effective for treatment of ganciclovir-resistant isolates and excessive concern regarding toxicity should not preclude consideration of these treatments when clinically indicated. J Heart Lung Transplant 2007;26:1286 –92. Copyright © 2007 by the International Society for Heart and Lung Transplantation.
Cytomegalovirus (CMV) is a common and clinically significant opportunistic infection after lung transplantation.1 The incidence of CMV infection and disease after lung transplantation has varied widely among prior reports, ranging from 30% to 86%, with mortality rates of 2% to 12%,1–5 in part due to differences in CMV prophylaxis, immunosuppressive regimens and diagnostic modalities. CMV infection is costly and has indirect effects including increased risk for concurrent bacterial and fungal infections.6 – 8 CMV disease has also been associated with the development of allograft rejection9 and bronchiolitis obliterans syndrome (BOS).10 Consequently, many lung
From the Division of aPulmonary, Allergy and Critical Care Medicine, and bDivision of Infectious Diseases, Department of Medicine, Duke University Medical Center, Durham, North Carolina. Submitted May 16, 2007; revised August 21, 2007; accepted September 19, 2007. Reprint requests: Anita J. Reddy, MD, Department of Pulmonary, Allergy and Critical Care Medicine, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Desk A90, Cleveland, OH 44124. Telephone: 216-444-4505. Fax: 216-445-6024. E-mail: [email protected] Copyright © 2007 by the International Society for Heart and Lung Transplantation. 1053-2498/07/$–see front matter. doi:10.1016/ j.healun.2007.09.012
1286
transplant centers use universal prophylaxis with ganciclovir or valganciclovir to prevent or delay onset of CMV disease, especially during periods of heightened immunosuppression. Ganciclovir resistance among patients with active CMV infection is increasing among solid-organ transplant recipients, and a few reports have emerged following lung transplantation.11,12 Reported risk factors for development of ganciclovir resistance include being a seronegative recipient of a seropositive organ (D⫹/R⫺), prolonged exposure to oral ganciclovir and intense immunosuppression.12,13,16 Despite multiple reports of ganciclovir-resistant CMV disease in solid-organ transplant recipients, optimal treatment regimens are unknown. In this study, we describe our experience with treatment of ganciclovir-resistant CMV infection using either foscarnet or combination therapy with foscarnet and ganciclovir in lung transplant recipients. METHODS Patients We reviewed the results of ganciclovir susceptibility testing performed in lung transplant recipients at Duke University Medical Center between April 2002 and October 2005. This time period was selected because
The Journal of Heart and Lung Transplantation Volume 26, Number 12
all patients underwent identical initial immunosuppression and CMV prophylaxis, as well as bilateral lung transplantation. All patients who underwent primary transplantation during this time period were included in the analysis and recipients of re-transplantation (n ⫽ 4) were excluded. Patient follow-up occurred until December 2006. This study was approved by the institutional review board and relevant clinical data were extracted from medical records.
Reddy et al.
1287
Ganciclovir Resistance Resistance to ganciclovir was determined based on CMV DNA sequence analysis for the presence of UL97 and UL54 mutations performed at either Mayo Medical Laboratories (Rochester, MN) or Colorado University Hospital (Denver, CO). CMV genotyping was sent based on clinical suspicion of ganciclovir-resistant disease (i.e., multiple episodes of CMV disease or increasing CMV viral loads by hybrid capture studies despite treatment with ganciclovir).
CMV Prophylaxis CMV prophylaxis regimens were based on donor and recipient CMV seropositivity. Patients with D⫹/R⫺ CMV serostatus received intravenous (IV) ganciclovir at a dose of 5 mg/kg every 12 hours for 4 weeks, followed by 5 mg/kg daily for 10 weeks, then continued on valganciclovir 900 mg total daily dose for prophylaxis indefinitely. In D⫹/⫺/R⫹ patients, IV ganciclovir was administered post-operatively at a dose of 5 mg/kg every 12 hours until hospital discharge or tolerance of oral medications, at which time valganciclovir prophylaxis was continued for 1 month. D⫺/R⫺ patients did not receive any prophylaxis. All doses of IV ganciclovir or oral valganciclovir were adjusted for creatinine clearance. Immunosuppression All patients received similar initial immunosuppressive regimens, which included azathioprine, tacrolimus and prednisone. Patients also received basilixumab intraoperatively and on Day 4 post-operatively, and IV methylprednisolone peri-operatively. Clinical Protocols Patients underwent routine testing for CMV DNA in whole blood using the quantitative Hybrid Capture Test, version 2.0 (Digene; Gaithersburg, MD) at the time of follow-up visits and when clinically indicated. Patients also underwent routine bronchoscopy with bronchoalveolar lavage (BAL) and transbronchial biopsies at 1, 3 and 6 months post-transplant. All BAL samples were sent for CMV culture and biopsies were sent for immunohistopathologic staining for CMV. Clinical indications for bronchoscopy and CMV testing included increase in dyspnea, cough, fever, new radiographic changes and/or decline in forced expiratory volume of 10% on pulmonary function testing. Treatment for acute rejection included IV solumedrol for 3 days, followed by a prednisone taper. If rejection was recurrent or high grade, patients were treated with anti-lymphocyte globulin, followed by a prednisone taper along with a 3-week course of ganciclovir.
Treatment All patients with resistant isolates were treated with a combination of ganciclovir and foscarnet, or foscarnet monotherapy. The treatment regimen provided a dose of 2.5 mg/kg ganciclovir divided once or twice daily in combination with foscarnet at 90 mg/kg divided twice daily or foscarnet at 180 mg/kg divided twice daily. Patients were treated with IV therapy until clinical and virologic improvement was seen. Selection of single vs combination therapy was made at the discretion of the physician based in part on the patient’s renal function at the time of treatment. CMV immunoglobulin was also administered at the discretion of the treating physicians. Treatment Monitoring Patients were considered to be treated effectively and clinically improved when two consecutive negative CMV DNA hybrid capture tests at least 6 weeks apart with no recurrence of CMV disease as documented by CMV DNA hybrid capture testing or BAL/histopathology. Testing was routinely first performed while still on therapy, but patients underwent continued serial testing (as shown in Figure 1) after the completion of therapy. CMV viral load and serum creatinine in patients with ganciclovir-resistant CMV infection were followed periodically and as clinically indicated from the time of diagnosis of initial CMV infection until the end of the study period. Glomerular filtration rate (GFR) calculations were based on the Modification of Diet in Renal Disease formula. Statistical Analyses Patients with ganciclovir-resistant CMV isolates were compared with patients with ganciclovir-sensitive CMV isolates using Fisher’s exact test for dichotomous variables (e.g., proportion with CMV mismatch) and 2-tailed t-test for continuous variables (e.g., age). Comparisons between CMV clearance and renal function before and after treatment in patients with ganciclovir resistance were compared by paired t-test using the SIGMASTAT v3.0 program (San Jose, CA). One patient died before complete clearance of CMV, and the most recent values for
1288
Reddy et al.
The Journal of Heart and Lung Transplantation December 2007
Figure 1. Trends in CMV viral loads in patients with ganciclovir-resistant isolates. Arrows indicate point of initiation of ganciclovir/foscarnet or foscarnet therapy. Bars indicate periods of resistant CMV treatment. Asterisk indicates episodes of acute rejection.
The Journal of Heart and Lung Transplantation Volume 26, Number 12
Reddy et al.
1289
fully treated with standard ganciclovir regimens and negative CMV hybrid capture assays were documented at the end of therapy.
CMV viral load and renal function prior to death were used for this analysis. All reported values represent median time with standard error measurements and interquartile ranges (IQRs).
Risk Factors Of the 6 patients who developed ganciclovir-resistant CMV, however, 3 (50%) recipients were D⫹/R⫹ for CMV. When the proportion of patients with CMV mismatch was compared among those who developed ganciclovir-resistant infection (n ⫽ 6, incidence 50%) to the other patients with ganciclovir-sensitive infection (n ⫽ 107, incidence 25%), a trend was observed toward an increased incidence of CMV mismatch among the patients with ganciclovir-resistant isolates (p ⫽ 0.18, Fisher’s exact test). Four of 6 patients with ganciclovirresistant CMV infection had episodes of acute rejection prior to detection of CMV. The majority of these episodes were low grade and were treated with corticosteroids, but 2 patients were treated with anti-lymphocyte globulin for persistent rejection. The rates of acute rejection, persistent rejection and use of anti-
RESULTS Demographics A total of 210 patients underwent lung transplantation between April 2002 and October 2005 at Duke University Medical Center. Among these patients, 113 patients were diagnosed with CMV viremia and/or disease (53.8%), and 6 of those with CMV disease (5.3%) exhibited ganciclovirresistant CMV infection (Table 1). The median time to detection of the resistant isolate after transplantation in all patients with a resistant isolate was 231 ⫾ 76 days (IQR 94 to 493). All patients received prior ganciclovir or valganciclovir for treatment or prophylaxis for CMV prior to detection of a resistant isolate. Three of 6 patients (50%) had episodes of recurrent CMV disease prior to diagnosis of ganciclovir-resistant disease (range of one to three episodes). These episodes were success-
Table 1. Demographic Data for Patients With Ganciclovir-resistant CMV Patient
Demographic data Age Gender Disease Transplantation CMV serostatus Clinical data Rejection episodes prior to Ganc R CMV Cytolytic therapy prior to Ganc R CMV diagnosis Therapy at time of initial CMV diagnosis Therapy at time of Ganc R CMV diagnosis Symptomatic CMV Episodes Time from transplant to Ganc R CMV (days) Ganc R CMV isolate UL97 UL54 Susceptibility Treatment Treatment for Ganc R CMV Time from CMV diagnosis to clear CMV (days)
1
2
3
4
5
6
28 F CF
48 M IPF
57 M COPD
60 F COPD
60 M COPD
61 F COPD
D⫹/R⫺
D⫹/R⫹
D⫹/R⫹
D⫹/R⫺
D⫹/R⫹
D⫹/R⫺
1
3
3
4
2
0
None
None
1
1
None
None
Valganc
Valganc
Ganc
Ganc
None
Valganc
Ganc, fosc
None
Ganc
Fosc
Ganc
Ganc
0
3
0
1
2
0
94
284
93
502
178
493
G598D/G L501L/I R-ganc, cido, S-fosc
C607F
N510S
N510S
R-ganc, S-fosc, cido
R-ganc, S-fosc, cido
R570Q A834P R-ganc (high), fosc, cido (low)
E596G/E P522L R-ganc, cido, S-fosc
Ganc, fosc
Fosc, CMV IgG
46
158
Ganc, fosc, CMV IgG 575
Ganc, fosc, CMV IgG 324
N/A, not applicable; R, resistant; S, sensitive; ganc, ganciclovir; fosc, foscarnet; cido, cidofovir; valganc, valganciclovir.
R-ganc, S-fosc, ido Ganc, fosc 64
Ganc, fosc, CMV IgG N/A
1290
Reddy et al.
lymphocyte globulin were similar in those patients with ganciclovir-sensitive CMV. Chronology of Infection The median time to CMV diagnosis was 91 days in the resistant CMV group, with a median of 2.5 prior rejection episodes prior to diagnosis of CMV disease (Table 1). The median time to detection of a resistant isolate after lung transplantation in the D⫹/R⫺ sub-group was longer than in the D⫹/R⫹ sub-group (493 days vs 178 days). Treatment Five patients received combination therapy with ganciclovir and foscarnet, whereas 1 patient received foscarnet alone for treatment of ganciclovir-resistant CMV. The average duration of therapy for these patients was 189 days. Four of 6 patients had two episodes of acute rejection, each prior to isolation of a resistant CMV isolate. Of these patients, only 2 received cytolytic therapy (Patients 3 and 4). Outcomes All patients with clinically resistant CMV infection were genotyped for the presence of UL97 and UL54 mutations. Interestingly, all patients were found to have mutations in the UL97 gene. All 3 patients who were CMV mismatched also had the UL54 mutation (Table 1). There did not seem to be a pattern associated with a particular nucleotide mutation within each gene and outcomes. CMV viral loads before and after treatment for ganciclovir-resistant CMV were also determined (Figure 1). All patients treated with ganciclovir/foscarnet or foscarnet alone had significant declines in CMV viral loads compared with the beginning to the end of therapy (p ⫽ 0.03, paired t-test), demonstrating that either single or combination therapy with foscarnet was an effective treatment in the management of lung transplant patients with documented ganciclovir-resistant CMV. Large increases in viral load were apparent in 2 patients between treatment periods (Patients 1 and 3). We also found that CMV viral load increased slightly shortly after the initiation of therapy in 3 of 6 patients (Patients 1, 2 and 4) prior to improvement in viral load. In these 3 patients, it took longer to clear CMV (mean: 315 days vs 111 days). All patients with ganciclovir-resistant infection were treated with combination ganciclovir/foscarnet or foscarnet alone. GFR was calculated for patients before and after treatment for ganciclovir-resistant CMV. There was no significant difference in GFR in patients before and after treatment (p ⫽ 0.07, paired t-test). Two patients were noted to have transient marked worsening of renal function during treatment, with some subsequent improvement of GFR. The remaining 4 patients did have dimin-
The Journal of Heart and Lung Transplantation December 2007
ished GFR after therapy for CMV. Worsening of GFR was not associated with monotherapy vs combination therapy. Five of 6 patients with resistant isolates were treated effectively and showed clinical improvement. Median days to resolution of CMV in all patients was 223 ⫾ 220 days (IQR 64 to 575 days). Four of 6 patients with ganciclovir-resistant isolates were treated with CMV intravenous immunoglobulin (IVIg), and there was no difference in time to clear CMV in patients treated with immunoglobulin. Mismatch patients appeared to have delayed clearance of CMV (89 ⫾ 60 days in the D⫹/R⫹ sub-group vs 450 ⫾ 177 days in the D⫹/R⫺ sub-group). In the D⫹/R⫺ sub-group, 1 patient died unexpectedly from unknown causes prior to documented clearing of infection, although virologic response to treatment was noted. Mortality from all causes in the resistant group was 16.7%, which was not significantly different from the 22.4% in the sensitive group ( p ⫽ 0.939; log-rank test). In this study we have reported our experience with ganciclovir-resistant CMV in lung transplant recipients. In prior reports, ganciclovir resistance was associated with a greater number of prior CMV episodes, increased exposure to cumulative intravenous ganciclovir therapy, use of induction anti-thymocyte globulin, as well as decreased survival.12,15 Other important factors for anti-viral resistance include D⫹/R⫺ status, the intensity of immunosuppression, and sub-optimal anti-viral prophylaxis.13–16 In our analysis, similar rates of acute rejection, persistent rejection and anti-thymocyte globulin were observed among patients with ganciclovir-resistant CMV and ganciclovir-sensitive CMV disease. All patients who received anti-thymocyte globulin did have 3 weeks of prophylaxis with ganciclovir. Based on these findings, it is difficult to conclude that resistant isolates occur as a result of treatment for acute rejection. In contrast, it appears that all lung transplant recipients have high rates of acute rejection and that other factors likely contribute to the development of ganciclovir resistance. CMV-mismatch patients represented 50% of the ganciclovir-resistant population as compared with only 25% of the ganciclovir-sensitive patients. Although not statistically significant, these findings are consistent with an increased risk for resistance in CMV in mismatch patients. Interestingly, we also found that mismatch patients were diagnosed with a resistant isolate at a later time than non-mismatch patients, which could be attributed to longer periods of prophylaxis therapy in this subset of patients. Thus, duration of ganciclovir prophylaxis might also independently, or in conjunction with CMV donor–recipient serology, contribute to the risk of development of resistant isolates. In addition, we found that all mismatch patients had mutations in both UL97 phosphotransferase and UL54
The Journal of Heart and Lung Transplantation Volume 26, Number 12
viral DNA polymerase genes, whereas all non-mismatch patients only had UL97 mutations. Mismatch patients were also noted to clear CMV infection more slowly than non-mismatch patients, possibly due to higher levels of ganciclovir resistance associated with the UL54 mutation. This may reflect the decreased efficacy of ganciclovir in comparison to foscarnet due to the presence of the UL54 gene in these patients, which is associated with increased levels of ganciclovir resistance.11,17 Another potential mechanism could be related to lack of prior immunologic memory to CMV in these patients, which appears necessary to efficiently clear CMV infection.18,19 The development of higher levels of ganciclovir resistance due to UL54 mutations in this sub-group could also be attributed to longer periods of prophylaxis in mismatch patients. These results have direct clinical implications and suggest that CMV mismatch patients with ganciclovir-resistant CMV require prolonged courses of therapy to fully eradicate the infection. Treatment regimens for patients confirmed to have ganciclovir-resistant CMV infection in our study were given a combination of ganciclovir and foscarnet or foscarnet alone. Combination therapy has been shown to be synergistic in patients with low-level resistant isolates.20 Both of these treatment regimens were effective and resulted in a significant clinical and virologic improvement. Conclusions regarding utility of CMV IVIg and foscarnet monotherapy are limited by the small number of patients in this group. Furthermore, during initial treatment for resistant disease, CMV viral loads were noted to increase in some patients, but this did not alter the clinical outcome of these patients. Boeckh suggested that the continued increase of viral load during therapy in the setting of profound underlying immunosuppression leads to very short in vivo replication times of CMV, and the overwhelming replication is not immediately and completely blocked by anti-viral treatment.21 Our results suggest that checking the viral load at a time early after initiation of treatment may not be beneficial and an early increase in viral load is not necessarily an indication of long-term treatment failure or success. Concern over selecting the appropriate treatment for resistant isolates is warranted due to potentially serious side effects, including decline in renal function. We demonstrated that combination therapy led to a virologic response in every patient with non-significant reductions in GFR, although we recognize that the absence of a difference between pre- and post-treatment values in GFR could represent a lack of power due to our small sample. This suggests that adequate treatment should be initiated with careful monitoring of renal function, rather than avoiding the use of the appropriate regimen for fear of developing renal insufficiency.
Reddy et al.
1291
Efficacy of the chosen regimen is also important. At our institution, both regimens of combination ganciclovir/foscarnet and foscarnet alone significantly reduce CMV viral loads, as well as provide a significant clinical improvement, although the small numbers make any direct comparison of these two regimens impossible in our current analysis. A possible approach to treatment could be to choose therapy based on the gene mutation present. Patients with UL97 mutations could be treated with the ganciclovir/foscarnet combination, whereas patients with UL54 mutations could be treated with foscarnet or cidofovir due to higher levels of ganciclovir resistance found to be associated with this gene. Alternatively, one could begin treatment with combination therapy in all patients given that our results suggest that this treatment is effective in patients with both UL97 and UL54 mutations. There are several limitations to our study, including the small number of patients with proven ganciclovirresistant CMV disease at our center. It is also possible that cases of ganciclovir resistance were undetected due to lack of genotype testing, which was not performed in all patients with CMV disease. Finally, only 1 patient was treated with foscarnet alone for ganciclovirresistant CMV, thus not allowing us to compare combination treatment with ganciclovir/foscarnet to foscarnet alone. Despite these limitations, our study represents one of the largest case series addressing outcome data in lung transplant patients who have undergone a uniform periand post-transplant regimen. We have shown that mismatch patients develop resistant disease at a later time and also have mutations in genes that confer high-level ganciclovir resistance compared with non-mismatch patients. Furthermore, our study is the first to quantify the effect of treatment of ganciclovir-resistant CMV upon a patient’s virologic response and renal function. Ganciclovir-resistant CMV is increasing in prevalence in solid-organ transplant patients. Additional studies are needed to define the optimal treatment in patients who have ganciclovir-resistant isolates, although our results suggest that combination ganciclovir and foscarnet is a reasonable treatment option in patients with ganciclovir-resistant CMV. The overall approach to CMV prophylaxis remains controversial in lung transplantation. Additional multicenter studies are needed to define the relative risks and benefits of prolonged ganciclovir prophylaxis as compared with pre-emptive approaches, particularly with regard to the development of CMV resistance. REFERENCES 1. Fishman JA, Rubin RH. Infection in organ-transplant recipients. N Engl J Med 1998;338:1741–51.
1292
Reddy et al.
2. Alexander BD, Tapson VF. Infectious complications of lung transplantation. Transplant Infect Dis 2001;3:128 –37. 3. Calhoon JH, Nichols L, Davis R, et al. Single lung transplantation: factors in postoperative cytomegalovirus infection. J Thorac Cardiovasc Surg 1992;103:21– 6. 4. Pollard RB. Cytomegalovirus infections in renal, heart, heart– lung, and liver transplantation. Pediatr Infect Dis J. 1988:7(suppl): S97–102. 5. Weinberg A, Hodges TN, Li S, Cai G, Zamora MR. Comparison of PCR, antigenemia assay and rapid blood culture for detection and prevention of cytomegalovirus disease after lung transplantation. J Clin Microbiol 2000;38:768 –72. 6. Snydman DR. Infection in solid organ transplantation. Transplant Infect Dis 1999;1:21– 8. 7. Falagas ME, Arbo M, Ruthazer R, et al. Cytomegalovirus disease is associated with increased cost and hospital length of stay among orthotopic liver transplant recipients. Transplantation 1997;63:1595–601. 8. McCarthy JM, Karim MA, Krueger H, Keown PA. The cost impact of cytomegalovirus disease in renal transplant recipients. Transplantation 1993;55:1277– 82. 9. Duncan SR, Paradis IL, Yousem SA, et al. Sequelae of cytomegalovirus pulmonary infections in lung allograft recipients. Am Rev Respir Dis 1992;146:1419 –25. 10. Heng D, Sharples LD, McNeil K, et al. Bronchiolitis obliterans syndrome: incidence, natural history, prognosis and risk factors. J Heart Lung Transplant 1998;17:1255– 63. 11. Erice A. Resistance of human cytomegalovirus to antiviral drugs. Clin Microbiol Rev 1999;12:286 –97. 12. Kruger RM, Shannon WD, Arens MQ, et al. The impact of ganciclovir-resistant cytomegalovirus infection after lung transplantation. Transplantation 1999;68:1272–9.
The Journal of Heart and Lung Transplantation December 2007
13. Limaye AP, Corey L, Koelle DM, David CL, Boeckh M. Emergence of ganciclovir-resistant cytomegalovirus disease among recipients of solid-organ transplants. Lancet 2000;356:645–9. 14. Mylonakis E, Kallas WM, Fishman JA. Combination antiviral therapy for ganciclovir-resistant cytomegalovirus infections in solid-organ transplant patients. Clin Infect Dis 2002;34: 1337– 41. 15. Bhorade SM, Lurain NS, Jordan A, et al. Emergence of ganciclovirresistant cytomegalovirus in lung transplant recipients. J Heart Lung Transplant 2002;21:1274 – 82. 16. Limaye AP, Raghu G, Koelle DM, et al. High incidence of ganciclovir-resistant cytomegalovirus infection among lung transplant recipients receiving preemptive therapy. J Infect Dis 2002; 185:20 –7. 17. Mousavi-Jazi M, Hokeberg I, Schloss L, et al. Sequence analysis of UL54 and UL97 genes and evaluation of antiviral susceptibility of human cytomegalovirus isolates obtained from kidney allograft recipients before and after treatment. Transplant Infect Dis 2001;3:195–202. 18. Ganepola S, Gentilini C, Hilbers U, et al. Patients at high risk for CMV infection and disease show delayed CD8⫹ T-cell immune recovery after allogeneic stem cell transplantation. Bone Marrow Transplant 2007;39:293–9. 19. La Rosa C, Limaye AP, Krishnan A, Longmate J, Diamond DJ. Longitudinal Assessment of cytomegalovirus (CMV)-specific immune responses in liver transplant recipients at high risk for late CMV disease. J Infect Dis 2007;195:633– 44. 20. Drew WL. Is combination antiviral therapy for CMV superior to monotherapy? J Clin Virol 2006;35:483– 8. 21. Boeckh M. Rising CMV PP65 antigenemia and DNA levels during preemptive antiviral therapy. Haematologica 2005;90:439.
A Single-center Experience With Ganciclovir-resistant Cytomegalovirus in Lung Transplant Recipients: Treatment and Outcome Anita J. Reddy, MD,a Aimee K. Zaas, MD,b Kimberly E. Hanson, MD, MHS,b and Scott M. Palmer, MD, MHSa Background: Cytomegalovirus (CMV) disease is a major cause of morbidity and mortality after lung transplantation despite ganciclovir prophylaxis. The emergence of ganciclovir-resistant CMV in lung transplant patients has been reported, although the optimal strategy for the management of these infections remains uncertain. A review of the results of glanciclovir susceptibility testing in lung transplant recipients was performed. Methods: We found 54% (113 of 210) of lung transplant patients developed CMV infection over a 4-year study period with ganciclovir-resistant CMV infection occurring in ⬎5% of patients (6 of 113). The demographic and clinical characteristics of patients who developed ganciclovir-resistant vs -sensitive CMV infection were similar, although 50% (3 of 6) patients who developed resistance were CMV mismatched (D⫹/R⫺ serology). All patients’ CMV isolates had mutations in the UL97 gene. In addition, the 3 mismatch patients also had CMV with mutations in the UL54 gene. Results: Treatment with a combination of foscarnet and ganciclovir or foscarnet alone for ganciclovir-resistant infection led to a significant reduction in virologic load in all patients (p ⫽ 0.03), although transient increases in viremia were observed in some patients early after treatment. Renal function worsened after treatment, but overall it was not significantly different from pre-treatment values (p ⫽ 0.07). Conclusions: Single or combination therapy with foscarnet is effective for treatment of ganciclovir-resistant isolates and excessive concern regarding toxicity should not preclude consideration of these treatments when clinically indicated. J Heart Lung Transplant 2007;26:1286 –92. Copyright © 2007 by the International Society for Heart and Lung Transplantation.
Cytomegalovirus (CMV) is a common and clinically significant opportunistic infection after lung transplantation.1 The incidence of CMV infection and disease after lung transplantation has varied widely among prior reports, ranging from 30% to 86%, with mortality rates of 2% to 12%,1–5 in part due to differences in CMV prophylaxis, immunosuppressive regimens and diagnostic modalities. CMV infection is costly and has indirect effects including increased risk for concurrent bacterial and fungal infections.6 – 8 CMV disease has also been associated with the development of allograft rejection9 and bronchiolitis obliterans syndrome (BOS).10 Consequently, many lung
From the Division of aPulmonary, Allergy and Critical Care Medicine, and bDivision of Infectious Diseases, Department of Medicine, Duke University Medical Center, Durham, North Carolina. Submitted May 16, 2007; revised August 21, 2007; accepted September 19, 2007. Reprint requests: Anita J. Reddy, MD, Department of Pulmonary, Allergy and Critical Care Medicine, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Desk A90, Cleveland, OH 44124. Telephone: 216-444-4505. Fax: 216-445-6024. E-mail: [email protected] Copyright © 2007 by the International Society for Heart and Lung Transplantation. 1053-2498/07/$–see front matter. doi:10.1016/ j.healun.2007.09.012
1286
transplant centers use universal prophylaxis with ganciclovir or valganciclovir to prevent or delay onset of CMV disease, especially during periods of heightened immunosuppression. Ganciclovir resistance among patients with active CMV infection is increasing among solid-organ transplant recipients, and a few reports have emerged following lung transplantation.11,12 Reported risk factors for development of ganciclovir resistance include being a seronegative recipient of a seropositive organ (D⫹/R⫺), prolonged exposure to oral ganciclovir and intense immunosuppression.12,13,16 Despite multiple reports of ganciclovir-resistant CMV disease in solid-organ transplant recipients, optimal treatment regimens are unknown. In this study, we describe our experience with treatment of ganciclovir-resistant CMV infection using either foscarnet or combination therapy with foscarnet and ganciclovir in lung transplant recipients. METHODS Patients We reviewed the results of ganciclovir susceptibility testing performed in lung transplant recipients at Duke University Medical Center between April 2002 and October 2005. This time period was selected because
The Journal of Heart and Lung Transplantation Volume 26, Number 12
all patients underwent identical initial immunosuppression and CMV prophylaxis, as well as bilateral lung transplantation. All patients who underwent primary transplantation during this time period were included in the analysis and recipients of re-transplantation (n ⫽ 4) were excluded. Patient follow-up occurred until December 2006. This study was approved by the institutional review board and relevant clinical data were extracted from medical records.
Reddy et al.
1287
Ganciclovir Resistance Resistance to ganciclovir was determined based on CMV DNA sequence analysis for the presence of UL97 and UL54 mutations performed at either Mayo Medical Laboratories (Rochester, MN) or Colorado University Hospital (Denver, CO). CMV genotyping was sent based on clinical suspicion of ganciclovir-resistant disease (i.e., multiple episodes of CMV disease or increasing CMV viral loads by hybrid capture studies despite treatment with ganciclovir).
CMV Prophylaxis CMV prophylaxis regimens were based on donor and recipient CMV seropositivity. Patients with D⫹/R⫺ CMV serostatus received intravenous (IV) ganciclovir at a dose of 5 mg/kg every 12 hours for 4 weeks, followed by 5 mg/kg daily for 10 weeks, then continued on valganciclovir 900 mg total daily dose for prophylaxis indefinitely. In D⫹/⫺/R⫹ patients, IV ganciclovir was administered post-operatively at a dose of 5 mg/kg every 12 hours until hospital discharge or tolerance of oral medications, at which time valganciclovir prophylaxis was continued for 1 month. D⫺/R⫺ patients did not receive any prophylaxis. All doses of IV ganciclovir or oral valganciclovir were adjusted for creatinine clearance. Immunosuppression All patients received similar initial immunosuppressive regimens, which included azathioprine, tacrolimus and prednisone. Patients also received basilixumab intraoperatively and on Day 4 post-operatively, and IV methylprednisolone peri-operatively. Clinical Protocols Patients underwent routine testing for CMV DNA in whole blood using the quantitative Hybrid Capture Test, version 2.0 (Digene; Gaithersburg, MD) at the time of follow-up visits and when clinically indicated. Patients also underwent routine bronchoscopy with bronchoalveolar lavage (BAL) and transbronchial biopsies at 1, 3 and 6 months post-transplant. All BAL samples were sent for CMV culture and biopsies were sent for immunohistopathologic staining for CMV. Clinical indications for bronchoscopy and CMV testing included increase in dyspnea, cough, fever, new radiographic changes and/or decline in forced expiratory volume of 10% on pulmonary function testing. Treatment for acute rejection included IV solumedrol for 3 days, followed by a prednisone taper. If rejection was recurrent or high grade, patients were treated with anti-lymphocyte globulin, followed by a prednisone taper along with a 3-week course of ganciclovir.
Treatment All patients with resistant isolates were treated with a combination of ganciclovir and foscarnet, or foscarnet monotherapy. The treatment regimen provided a dose of 2.5 mg/kg ganciclovir divided once or twice daily in combination with foscarnet at 90 mg/kg divided twice daily or foscarnet at 180 mg/kg divided twice daily. Patients were treated with IV therapy until clinical and virologic improvement was seen. Selection of single vs combination therapy was made at the discretion of the physician based in part on the patient’s renal function at the time of treatment. CMV immunoglobulin was also administered at the discretion of the treating physicians. Treatment Monitoring Patients were considered to be treated effectively and clinically improved when two consecutive negative CMV DNA hybrid capture tests at least 6 weeks apart with no recurrence of CMV disease as documented by CMV DNA hybrid capture testing or BAL/histopathology. Testing was routinely first performed while still on therapy, but patients underwent continued serial testing (as shown in Figure 1) after the completion of therapy. CMV viral load and serum creatinine in patients with ganciclovir-resistant CMV infection were followed periodically and as clinically indicated from the time of diagnosis of initial CMV infection until the end of the study period. Glomerular filtration rate (GFR) calculations were based on the Modification of Diet in Renal Disease formula. Statistical Analyses Patients with ganciclovir-resistant CMV isolates were compared with patients with ganciclovir-sensitive CMV isolates using Fisher’s exact test for dichotomous variables (e.g., proportion with CMV mismatch) and 2-tailed t-test for continuous variables (e.g., age). Comparisons between CMV clearance and renal function before and after treatment in patients with ganciclovir resistance were compared by paired t-test using the SIGMASTAT v3.0 program (San Jose, CA). One patient died before complete clearance of CMV, and the most recent values for
1288
Reddy et al.
The Journal of Heart and Lung Transplantation December 2007
Figure 1. Trends in CMV viral loads in patients with ganciclovir-resistant isolates. Arrows indicate point of initiation of ganciclovir/foscarnet or foscarnet therapy. Bars indicate periods of resistant CMV treatment. Asterisk indicates episodes of acute rejection.
The Journal of Heart and Lung Transplantation Volume 26, Number 12
Reddy et al.
1289
fully treated with standard ganciclovir regimens and negative CMV hybrid capture assays were documented at the end of therapy.
CMV viral load and renal function prior to death were used for this analysis. All reported values represent median time with standard error measurements and interquartile ranges (IQRs).
Risk Factors Of the 6 patients who developed ganciclovir-resistant CMV, however, 3 (50%) recipients were D⫹/R⫹ for CMV. When the proportion of patients with CMV mismatch was compared among those who developed ganciclovir-resistant infection (n ⫽ 6, incidence 50%) to the other patients with ganciclovir-sensitive infection (n ⫽ 107, incidence 25%), a trend was observed toward an increased incidence of CMV mismatch among the patients with ganciclovir-resistant isolates (p ⫽ 0.18, Fisher’s exact test). Four of 6 patients with ganciclovirresistant CMV infection had episodes of acute rejection prior to detection of CMV. The majority of these episodes were low grade and were treated with corticosteroids, but 2 patients were treated with anti-lymphocyte globulin for persistent rejection. The rates of acute rejection, persistent rejection and use of anti-
RESULTS Demographics A total of 210 patients underwent lung transplantation between April 2002 and October 2005 at Duke University Medical Center. Among these patients, 113 patients were diagnosed with CMV viremia and/or disease (53.8%), and 6 of those with CMV disease (5.3%) exhibited ganciclovirresistant CMV infection (Table 1). The median time to detection of the resistant isolate after transplantation in all patients with a resistant isolate was 231 ⫾ 76 days (IQR 94 to 493). All patients received prior ganciclovir or valganciclovir for treatment or prophylaxis for CMV prior to detection of a resistant isolate. Three of 6 patients (50%) had episodes of recurrent CMV disease prior to diagnosis of ganciclovir-resistant disease (range of one to three episodes). These episodes were success-
Table 1. Demographic Data for Patients With Ganciclovir-resistant CMV Patient
Demographic data Age Gender Disease Transplantation CMV serostatus Clinical data Rejection episodes prior to Ganc R CMV Cytolytic therapy prior to Ganc R CMV diagnosis Therapy at time of initial CMV diagnosis Therapy at time of Ganc R CMV diagnosis Symptomatic CMV Episodes Time from transplant to Ganc R CMV (days) Ganc R CMV isolate UL97 UL54 Susceptibility Treatment Treatment for Ganc R CMV Time from CMV diagnosis to clear CMV (days)
1
2
3
4
5
6
28 F CF
48 M IPF
57 M COPD
60 F COPD
60 M COPD
61 F COPD
D⫹/R⫺
D⫹/R⫹
D⫹/R⫹
D⫹/R⫺
D⫹/R⫹
D⫹/R⫺
1
3
3
4
2
0
None
None
1
1
None
None
Valganc
Valganc
Ganc
Ganc
None
Valganc
Ganc, fosc
None
Ganc
Fosc
Ganc
Ganc
0
3
0
1
2
0
94
284
93
502
178
493
G598D/G L501L/I R-ganc, cido, S-fosc
C607F
N510S
N510S
R-ganc, S-fosc, cido
R-ganc, S-fosc, cido
R570Q A834P R-ganc (high), fosc, cido (low)
E596G/E P522L R-ganc, cido, S-fosc
Ganc, fosc
Fosc, CMV IgG
46
158
Ganc, fosc, CMV IgG 575
Ganc, fosc, CMV IgG 324
N/A, not applicable; R, resistant; S, sensitive; ganc, ganciclovir; fosc, foscarnet; cido, cidofovir; valganc, valganciclovir.
R-ganc, S-fosc, ido Ganc, fosc 64
Ganc, fosc, CMV IgG N/A
1290
Reddy et al.
lymphocyte globulin were similar in those patients with ganciclovir-sensitive CMV. Chronology of Infection The median time to CMV diagnosis was 91 days in the resistant CMV group, with a median of 2.5 prior rejection episodes prior to diagnosis of CMV disease (Table 1). The median time to detection of a resistant isolate after lung transplantation in the D⫹/R⫺ sub-group was longer than in the D⫹/R⫹ sub-group (493 days vs 178 days). Treatment Five patients received combination therapy with ganciclovir and foscarnet, whereas 1 patient received foscarnet alone for treatment of ganciclovir-resistant CMV. The average duration of therapy for these patients was 189 days. Four of 6 patients had two episodes of acute rejection, each prior to isolation of a resistant CMV isolate. Of these patients, only 2 received cytolytic therapy (Patients 3 and 4). Outcomes All patients with clinically resistant CMV infection were genotyped for the presence of UL97 and UL54 mutations. Interestingly, all patients were found to have mutations in the UL97 gene. All 3 patients who were CMV mismatched also had the UL54 mutation (Table 1). There did not seem to be a pattern associated with a particular nucleotide mutation within each gene and outcomes. CMV viral loads before and after treatment for ganciclovir-resistant CMV were also determined (Figure 1). All patients treated with ganciclovir/foscarnet or foscarnet alone had significant declines in CMV viral loads compared with the beginning to the end of therapy (p ⫽ 0.03, paired t-test), demonstrating that either single or combination therapy with foscarnet was an effective treatment in the management of lung transplant patients with documented ganciclovir-resistant CMV. Large increases in viral load were apparent in 2 patients between treatment periods (Patients 1 and 3). We also found that CMV viral load increased slightly shortly after the initiation of therapy in 3 of 6 patients (Patients 1, 2 and 4) prior to improvement in viral load. In these 3 patients, it took longer to clear CMV (mean: 315 days vs 111 days). All patients with ganciclovir-resistant infection were treated with combination ganciclovir/foscarnet or foscarnet alone. GFR was calculated for patients before and after treatment for ganciclovir-resistant CMV. There was no significant difference in GFR in patients before and after treatment (p ⫽ 0.07, paired t-test). Two patients were noted to have transient marked worsening of renal function during treatment, with some subsequent improvement of GFR. The remaining 4 patients did have dimin-
The Journal of Heart and Lung Transplantation December 2007
ished GFR after therapy for CMV. Worsening of GFR was not associated with monotherapy vs combination therapy. Five of 6 patients with resistant isolates were treated effectively and showed clinical improvement. Median days to resolution of CMV in all patients was 223 ⫾ 220 days (IQR 64 to 575 days). Four of 6 patients with ganciclovir-resistant isolates were treated with CMV intravenous immunoglobulin (IVIg), and there was no difference in time to clear CMV in patients treated with immunoglobulin. Mismatch patients appeared to have delayed clearance of CMV (89 ⫾ 60 days in the D⫹/R⫹ sub-group vs 450 ⫾ 177 days in the D⫹/R⫺ sub-group). In the D⫹/R⫺ sub-group, 1 patient died unexpectedly from unknown causes prior to documented clearing of infection, although virologic response to treatment was noted. Mortality from all causes in the resistant group was 16.7%, which was not significantly different from the 22.4% in the sensitive group ( p ⫽ 0.939; log-rank test). In this study we have reported our experience with ganciclovir-resistant CMV in lung transplant recipients. In prior reports, ganciclovir resistance was associated with a greater number of prior CMV episodes, increased exposure to cumulative intravenous ganciclovir therapy, use of induction anti-thymocyte globulin, as well as decreased survival.12,15 Other important factors for anti-viral resistance include D⫹/R⫺ status, the intensity of immunosuppression, and sub-optimal anti-viral prophylaxis.13–16 In our analysis, similar rates of acute rejection, persistent rejection and anti-thymocyte globulin were observed among patients with ganciclovir-resistant CMV and ganciclovir-sensitive CMV disease. All patients who received anti-thymocyte globulin did have 3 weeks of prophylaxis with ganciclovir. Based on these findings, it is difficult to conclude that resistant isolates occur as a result of treatment for acute rejection. In contrast, it appears that all lung transplant recipients have high rates of acute rejection and that other factors likely contribute to the development of ganciclovir resistance. CMV-mismatch patients represented 50% of the ganciclovir-resistant population as compared with only 25% of the ganciclovir-sensitive patients. Although not statistically significant, these findings are consistent with an increased risk for resistance in CMV in mismatch patients. Interestingly, we also found that mismatch patients were diagnosed with a resistant isolate at a later time than non-mismatch patients, which could be attributed to longer periods of prophylaxis therapy in this subset of patients. Thus, duration of ganciclovir prophylaxis might also independently, or in conjunction with CMV donor–recipient serology, contribute to the risk of development of resistant isolates. In addition, we found that all mismatch patients had mutations in both UL97 phosphotransferase and UL54
The Journal of Heart and Lung Transplantation Volume 26, Number 12
viral DNA polymerase genes, whereas all non-mismatch patients only had UL97 mutations. Mismatch patients were also noted to clear CMV infection more slowly than non-mismatch patients, possibly due to higher levels of ganciclovir resistance associated with the UL54 mutation. This may reflect the decreased efficacy of ganciclovir in comparison to foscarnet due to the presence of the UL54 gene in these patients, which is associated with increased levels of ganciclovir resistance.11,17 Another potential mechanism could be related to lack of prior immunologic memory to CMV in these patients, which appears necessary to efficiently clear CMV infection.18,19 The development of higher levels of ganciclovir resistance due to UL54 mutations in this sub-group could also be attributed to longer periods of prophylaxis in mismatch patients. These results have direct clinical implications and suggest that CMV mismatch patients with ganciclovir-resistant CMV require prolonged courses of therapy to fully eradicate the infection. Treatment regimens for patients confirmed to have ganciclovir-resistant CMV infection in our study were given a combination of ganciclovir and foscarnet or foscarnet alone. Combination therapy has been shown to be synergistic in patients with low-level resistant isolates.20 Both of these treatment regimens were effective and resulted in a significant clinical and virologic improvement. Conclusions regarding utility of CMV IVIg and foscarnet monotherapy are limited by the small number of patients in this group. Furthermore, during initial treatment for resistant disease, CMV viral loads were noted to increase in some patients, but this did not alter the clinical outcome of these patients. Boeckh suggested that the continued increase of viral load during therapy in the setting of profound underlying immunosuppression leads to very short in vivo replication times of CMV, and the overwhelming replication is not immediately and completely blocked by anti-viral treatment.21 Our results suggest that checking the viral load at a time early after initiation of treatment may not be beneficial and an early increase in viral load is not necessarily an indication of long-term treatment failure or success. Concern over selecting the appropriate treatment for resistant isolates is warranted due to potentially serious side effects, including decline in renal function. We demonstrated that combination therapy led to a virologic response in every patient with non-significant reductions in GFR, although we recognize that the absence of a difference between pre- and post-treatment values in GFR could represent a lack of power due to our small sample. This suggests that adequate treatment should be initiated with careful monitoring of renal function, rather than avoiding the use of the appropriate regimen for fear of developing renal insufficiency.
Reddy et al.
1291
Efficacy of the chosen regimen is also important. At our institution, both regimens of combination ganciclovir/foscarnet and foscarnet alone significantly reduce CMV viral loads, as well as provide a significant clinical improvement, although the small numbers make any direct comparison of these two regimens impossible in our current analysis. A possible approach to treatment could be to choose therapy based on the gene mutation present. Patients with UL97 mutations could be treated with the ganciclovir/foscarnet combination, whereas patients with UL54 mutations could be treated with foscarnet or cidofovir due to higher levels of ganciclovir resistance found to be associated with this gene. Alternatively, one could begin treatment with combination therapy in all patients given that our results suggest that this treatment is effective in patients with both UL97 and UL54 mutations. There are several limitations to our study, including the small number of patients with proven ganciclovirresistant CMV disease at our center. It is also possible that cases of ganciclovir resistance were undetected due to lack of genotype testing, which was not performed in all patients with CMV disease. Finally, only 1 patient was treated with foscarnet alone for ganciclovirresistant CMV, thus not allowing us to compare combination treatment with ganciclovir/foscarnet to foscarnet alone. Despite these limitations, our study represents one of the largest case series addressing outcome data in lung transplant patients who have undergone a uniform periand post-transplant regimen. We have shown that mismatch patients develop resistant disease at a later time and also have mutations in genes that confer high-level ganciclovir resistance compared with non-mismatch patients. Furthermore, our study is the first to quantify the effect of treatment of ganciclovir-resistant CMV upon a patient’s virologic response and renal function. Ganciclovir-resistant CMV is increasing in prevalence in solid-organ transplant patients. Additional studies are needed to define the optimal treatment in patients who have ganciclovir-resistant isolates, although our results suggest that combination ganciclovir and foscarnet is a reasonable treatment option in patients with ganciclovir-resistant CMV. The overall approach to CMV prophylaxis remains controversial in lung transplantation. Additional multicenter studies are needed to define the relative risks and benefits of prolonged ganciclovir prophylaxis as compared with pre-emptive approaches, particularly with regard to the development of CMV resistance. REFERENCES 1. Fishman JA, Rubin RH. Infection in organ-transplant recipients. N Engl J Med 1998;338:1741–51.
1292
Reddy et al.
2. Alexander BD, Tapson VF. Infectious complications of lung transplantation. Transplant Infect Dis 2001;3:128 –37. 3. Calhoon JH, Nichols L, Davis R, et al. Single lung transplantation: factors in postoperative cytomegalovirus infection. J Thorac Cardiovasc Surg 1992;103:21– 6. 4. Pollard RB. Cytomegalovirus infections in renal, heart, heart– lung, and liver transplantation. Pediatr Infect Dis J. 1988:7(suppl): S97–102. 5. Weinberg A, Hodges TN, Li S, Cai G, Zamora MR. Comparison of PCR, antigenemia assay and rapid blood culture for detection and prevention of cytomegalovirus disease after lung transplantation. J Clin Microbiol 2000;38:768 –72. 6. Snydman DR. Infection in solid organ transplantation. Transplant Infect Dis 1999;1:21– 8. 7. Falagas ME, Arbo M, Ruthazer R, et al. Cytomegalovirus disease is associated with increased cost and hospital length of stay among orthotopic liver transplant recipients. Transplantation 1997;63:1595–601. 8. McCarthy JM, Karim MA, Krueger H, Keown PA. The cost impact of cytomegalovirus disease in renal transplant recipients. Transplantation 1993;55:1277– 82. 9. Duncan SR, Paradis IL, Yousem SA, et al. Sequelae of cytomegalovirus pulmonary infections in lung allograft recipients. Am Rev Respir Dis 1992;146:1419 –25. 10. Heng D, Sharples LD, McNeil K, et al. Bronchiolitis obliterans syndrome: incidence, natural history, prognosis and risk factors. J Heart Lung Transplant 1998;17:1255– 63. 11. Erice A. Resistance of human cytomegalovirus to antiviral drugs. Clin Microbiol Rev 1999;12:286 –97. 12. Kruger RM, Shannon WD, Arens MQ, et al. The impact of ganciclovir-resistant cytomegalovirus infection after lung transplantation. Transplantation 1999;68:1272–9.
The Journal of Heart and Lung Transplantation December 2007
13. Limaye AP, Corey L, Koelle DM, David CL, Boeckh M. Emergence of ganciclovir-resistant cytomegalovirus disease among recipients of solid-organ transplants. Lancet 2000;356:645–9. 14. Mylonakis E, Kallas WM, Fishman JA. Combination antiviral therapy for ganciclovir-resistant cytomegalovirus infections in solid-organ transplant patients. Clin Infect Dis 2002;34: 1337– 41. 15. Bhorade SM, Lurain NS, Jordan A, et al. Emergence of ganciclovirresistant cytomegalovirus in lung transplant recipients. J Heart Lung Transplant 2002;21:1274 – 82. 16. Limaye AP, Raghu G, Koelle DM, et al. High incidence of ganciclovir-resistant cytomegalovirus infection among lung transplant recipients receiving preemptive therapy. J Infect Dis 2002; 185:20 –7. 17. Mousavi-Jazi M, Hokeberg I, Schloss L, et al. Sequence analysis of UL54 and UL97 genes and evaluation of antiviral susceptibility of human cytomegalovirus isolates obtained from kidney allograft recipients before and after treatment. Transplant Infect Dis 2001;3:195–202. 18. Ganepola S, Gentilini C, Hilbers U, et al. Patients at high risk for CMV infection and disease show delayed CD8⫹ T-cell immune recovery after allogeneic stem cell transplantation. Bone Marrow Transplant 2007;39:293–9. 19. La Rosa C, Limaye AP, Krishnan A, Longmate J, Diamond DJ. Longitudinal Assessment of cytomegalovirus (CMV)-specific immune responses in liver transplant recipients at high risk for late CMV disease. J Infect Dis 2007;195:633– 44. 20. Drew WL. Is combination antiviral therapy for CMV superior to monotherapy? J Clin Virol 2006;35:483– 8. 21. Boeckh M. Rising CMV PP65 antigenemia and DNA levels during preemptive antiviral therapy. Haematologica 2005;90:439.