Combination Prophylaxis with Ganciclovir and Cytomegalovirus (CMV) Immune Globulin After Lung Transplantation: Effective CMV Prevention Following Daclizumab Induction
Abstract
Despite the serious direct and indirect deleterious effects caused by cytomegalovirus (CMV), the optimal prophylactic strategy remains unknown. We sought to determine whether combination prophylaxis using intravenous ganciclovir (GCV) and CMV-IVIG reduced the incidence of CMV compared to GCV alone. Donor CMV positive/recipient negative (D+/R–) patients received GCV (6 weeks i.v. + 6 weeks oral) and CMV-IVIG (every 2 weeks for 7 doses), while R+ patients received GCV (2 weeks i.v. + 4 weeks oral) and CMV-IVIG (every 2 weeks for 3 doses). The group receiving combination prophylaxis (GpA) was compared to a historical, case-controlled group receiving GCV alone (GpB). Groups were matched by CMV donor/recipient serology, pretransplant diagnosis, age, and sex in reverse chronological order. Cyclosporine, azathioprine, and prednisone were used in both groups. Additionally, GpA received daclizumab induction therapy. Groups were compared as to the incidence of CMV disease, CMV infection, and acute rejection (AR). In GpA, 38 patients were evaluable and matched to 48 patients in GpB. Three GpA patients (8%) (2 D+/R–) developed CMV disease vs. 16 patients (33%) in GpB, p = 0.0077, Fisher's exact. There was also a trend toward a delay in CMV onset (148 days in GpA vs. 92 days in GpB, p = 0.07, Mann–Whitney). CMV infection did not occur in GpA, and one case occurred in GpB. There was no difference in the incidence of AR (66% in GpA vs. 79% in GpB, p = 0.22, Fisher's exact) or the need for cytolytic therapy between groups. Despite the use of daclizumab induction therapy, combination prophylaxis with GCV and CMV-IVIG reduced the incidence and probably delayed the onset of CMV infection compared to GCV alone. Longer follow-up will be needed to evaluate the impact of combination therapy on the incidence of bronchiolitis obliterans syndrome (BOS).
Introduction
Cytomegalovirus (CMV) infection causes significant direct morbidity and mortality in lung transplant recipients. In addition to its direct effects, the indirect consequences of CMV infection are profound and may include immunomodulatory effects that contribute to the development of bronchiolitis obliterans syndrome (BOS) (1), which is the major limiting factor in long-term survival following lung transplantation. Therefore, the prevention of infection could result in improved survival in lung transplant recipients.
Despite the importance of effectively preventing CMV disease, the optimal prophylactic strategy remains unknown. Most lung transplant centers have adopted prophylactic regimens unique to their particular program without evidence-based rationale. Despite a variety of strategies developed since the advent of ganciclovir (GCV), the reported incidence of CMV infection and disease after lung transplantation has remained high, ranging from 35% to 86% with an associated mortality rate of 2–12% (2). Clearly, the highest risk exists among the donor/recipient CMV mismatched patients (3) and in those patients in whom immunosuppression has been augmented (4).
Cytomegalovirus immune globulin intravenous (CMV-IGIV) contains anti-CMV IgG antibodies derived from pooled adult human plasma selected for its high titers of CMV antibodies. In some small series of lung transplant patients (5,6), improved outcomes have been attributed to the use of CMV-IGIV. However, no study to date has compared prophylactic regimens using GCV alone to combination therapy using GCV and CMV-IGIV in lung transplantation. Therefore, we evaluated the effectiveness of two different prophylactic strategies by comparing monotherapy prophylaxis with GCV vs. combination prophylaxis using GCV and CMV hyperimmune globulin (CMV-IGIV) in lung transplant patients during the first 180 days after transplant.
Materials and Methods
Study design
A retrospective review of medical records was performed on all patients undergoing lung transplantation between January 1, 1995 and December 1, 2001. Patients transplanted between July 1, 2000 and December 1, 2001 were included in the study group and designated as Group A (GpA). The study group differed from the historical comparison group, group B (GpB), in two ways. Firstly, our CMV prophylaxis regimen changed with the addition of CMV-IVIG to the previous use of GCV alone. Recipients received CMV prophylaxis based on donor/recipient (D/R) serologies, as depicted in Table 1(A and B). Secondly, the study group patients, GpA, received daclizumab induction immunosuppressive therapy. Daclizumab was administered preoperatively and every 2 weeks at a dose of 1 mg/kg for a total of 5 doses.
| R+ | R– | |
|---|---|---|
| D+ | 12 weeks GCV*(6 week i.v. & 6 week p.o.) | |
| CMV-IG+ 7 doses | ||
| 6 weeks GCV*(2 weeks i.v. & 4 week p.o.) | ||
| CMV-IG 3 doses(1 dose every 2 weeks) | ||
| D– | No prophylaxis used |
- * i.v. dose 5 mg/kg q12 h adjusted for creatinine clearance.
- + 150 mg/kg within 72 h post transplant, then q2 weeks × 4 doses, then 100 mg/kg q4 weeks × 2 additional doses.
| R+ | R– | |
|---|---|---|
| D+ | 6 weeks GCV* (2 weeks i.v. & 4 weeks p.o.) | 12 weeks GCV* (6 weeks i.v. & 6 weeks p.o.) |
| D– | No prophylaxis used |
- * i.v. dose 5 mg/kg q12 h adjusted for creatinine clearance.
To be included in the study, patients in both groups had to have survived at least through the CMV prophylaxis period and were followed for 180 days post-transplant. All potential study patients who died from any cause other than CMV disease during the study period were excluded. Patients who had heart-lung transplants and lung transplant recipients who had D–/R– serologies were excluded from the study. GpB was matched to GpA transplant recipients by the following characteristics: indication for lung transplant, CMV donor/recipient serologies, sex, and age. Patients were matched consecutively using the above criteria in reverse chronological order. As many patients were included in the comparison group as could be found who met the matching criteria during the specified period.
Ganciclovir dosing adjustments were made according to renal function. The CMV antigen test was measured monthly for surveillance purposes from 2 to 6 months postoperative. During periods of cytolytic use, intravenous (i.v.) ganciclovir was used while the patient was hospitalized and orally as an outpatient for a total period of 2 weeks after the last dose of cytolytic therapy.
Definition of CMV disease and CMV syndrome
CMV disease was diagnosed if tissue invasion was present, reflected by histopathologic evidence of inclusion bodies in transbronchial biopsy tissue (or in other organ biopsy specimens), or if patients had compatible symptoms, such as fever, malaise, nausea, and vomiting together with a clinical response to GCV. Patients were also considered to have CMV disease if they had a positive BAL without evidence of tissue invasion but who responded to GCV treatment. CMV infection was diagnosed if blood specimens revealed positive CMV antigenemia or a positive CMV buffy coat.
Acute rejection
In our program, fiberoptic bronchoscopy with bronchoalveolar lavage (BAL) and transbronchial biopsy (TBB) are scheduled either for surveillance purposes (2 weeks, 6 weeks, 3, 6, 12, 18, and 24 months, and every 6 months thereafter) or if rejection or infection is suspected (new respiratory symptoms, oxygen desaturation, new chest radiograph infiltration, or >10% decline in FEV1). Acute rejection was diagnosed in one of two ways: 1) if surveillance transbronchial biopsy revealed grade A2 or B2 or greater by The International Society for Heart and Lung Transplantation (ISHLT) scale (7); or 2) if a transbronchial biopsy was not performed but a high clinical suspicion prompted treatment with high dose corticosteroids.
Immunosuppression protocol
Methylprednisolone 500 mg was given intravenously to all recipients intraoperatively followed by 125 mg every 8 h for three doses following surgery. Oral prednisone was then administered in a tapering fashion starting at 0.6 mg/kg/day and ultimately reduced to 0.1 mg/kg/day. Cyclosporine was initiated at a dose of 5 mg/kg/day and adjusted to maintain a trough level of 350–400 ng/mL during the first 6 postoperative months and then decreased according to the length of time after transplantation and renal function. Azathioprine (AZA) was begun at a dose of 2 mg/kg/day and then adjusted to target a white blood cell count of 5000–7000. All GpA patients received daclizumab in the manner described above.
Standard initial treatment for patients with acute cellular rejection (defined as Grade 2 or more) was methylprednisolone 15 mg/kg/day for 3 days followed by a return to the baseline prednisone dose.
Results
Patient demographics
Study group. Thirty-eight patients were included in GpA. GpA (19 female, 19 male) had a mean age of 52.1, range 21.4–65.7. The serologic statuses of donors and recipients and diagnosis requiring transplantation are listed in Table 2.
| Study group (GpA) No. of patients (%) n = 38 | Comparison group (GpB) No. of patients (%) n = 48 | |
|---|---|---|
| Age (mean, range) | 52.1, 21.4–65.7 | 54.4, 29.7–65.4 |
| Gender | 19F, 19M | 22F, 26M |
| Diagnosis | ||
| COPD | 22 (58) | 33 (69) |
| IPF | 7 (18) | 7 (15) |
| Cystic fibrosis | 3 (8) | 4 (8) |
| α-1 antitrypsin def. | 1 (3) | 2 (4) |
| Sarcoidosis | 3 (8) | 2 (4) |
| Other* | 2 (5) | 0 (0) |
| Donor/recipient (D/R) | ||
| CMV serologies | ||
| D+/R– | 6 (16) | 10 (21) |
| D+/R+ | 20 (52) | 25 (52) |
| D–/R+ | 12 (32) | 13 (27) |
- * Includes one bronchiectasis patient and one primary pulmonary hypertension patient.
Comparison group. Using the previously described matching protocol, 48 patients were included in the comparison group, GpB (22 female, 26 male) had a mean age of 54.4, range 29.76–65.42. The serologic statuses of donors and recipients and diagnosis requiring transplant are listed in Table 2.
CMV incidence
The incidence of CMV disease in GpA was 3/38 (7.9%) vs. GpB 16/48 (33.3%), [p = 0.0077, Fisher's Exact]. Of the three patients in GpA diagnosed with CMV disease, two had CMV pneumonia and one had CMV syndrome. Of the 16 patients in GpB diagnosed with CMV disease, 11 had CMV pneumonia and five had CMV syndrome (Table 3). The onset of disease in GpA occurred at 148 days ± 14.9 days (SD) vs. GpB, 92 days ± 26.4 days (SD), [p = 0.07, Mann–Whitney]. There were no additional CMV infections in GpA and one case of a CMV infection (as diagnosed by positive CMV antigenemia) in GpB. Furthermore, in our cohort, there were no instances where a patient had asymptomatic CMV BAL positivity, either in the study or control groups.
| Study group (GpA) No. of patients (%) n = 38 | Comparison group (GpB) No. of patients (%) n = 48 | |
|---|---|---|
| CMV disease | ||
| Pneumonia | 2 (5) | 11 (23) |
| Syndrome | 1 (3) | 5 (10) |
| Total | 3 (8) | 16 (33) p = 0.008 |
| CMV infection | 0 | 1 |
| Acute rejection | 25 (66) | 34 (79) p = 0.22 |
CMV surveillance was more rigorous in the study group (CMV antigen checked monthly for the first year) than in the control group. No pre-emptive antiviral therapy was used in either group. The difference in CMV surveillance would be unlikely to have impacted the results of the study and, if any effect was seen, would have been likely to have found more infection in the study group.
In both the study and control groups, all patients who were diagnosed with CMV pneumonia had an abnormal chest radiograph. None of the patients in either group with either CMV syndrome or CMV infection had chest radiograph abnormalities.
Acute rejection
The incidence of AR in the 6 months following transplantation was determined for both groups. Acute rejection occurred in 25/38 patients (66%) in GpA vs. 38/48 patients (79%) in GpB (p = 0.22, Fisher's Exact). There was no difference in AR rates between GpB patients with or without CMV disease (Table 3). Of the GpB patients who had frequent (> 4) episodes of AR requiring bolus steroid therapy (n = 17), six (35%) developed CMV disease. Therefore, the disease rate in the subset of patients experiencing multiple episodes of AR was similar to patients in the control group who did not have frequent AR. Furthermore, in the three patients with CMV disease in GpA, two had no episodes of AR. Finally, the use of lympholytic immunosuppressive medication did not differ between the two groups.
Discussion
The direct and indirect consequences of CMV infection continue to have an important role in limiting long-term survival following lung transplantation. Although effective antiviral medications have been available for much of the lung transplant era, the optimal strategy for preventing CMV infection remains unknown and largely untested. Furthermore, as immunosuppressive strategies have continued to evolve, comparative studies analyzing various anti-CMV regimens have been difficult to perform and even more difficult to interpret. However, as newer immunosuppressive agents become available, a major concern remains the impact of these agents on infectious complications, particularly those from CMV disease. In the current study, we found that combination prophylaxis using GCV and CMV-IVIG significantly reduced the incidence of CMV disease, even with the use of daclizumab induction therapy, despite no significant differences in AR rates or total immunosuppression between the study and control groups.
The efficacy of combination prophylaxis has been demonstrated previously in human solid organ transplants. In renal (9) and liver transplantation (10), prophylaxis using GCV plus CMV-IVIG resulted in a reduced incidence of CMV disease, an effect that was modulated somewhat by the type of immunosuppression used. In lung transplantation, Maurer reported a decreased incidence of CMV pneumonitis and increased survival after prophylaxis with CMV-IVIG and either ganciclovir or oral acyclovir (41% vs. 67%) (5). Furthermore, Zamora et al. (6) found that, after using combination CMV-IVIG and ganciclovir, the incidence and severity of CMV infection was decreased and time to onset of CMV infection was increased, as compared to a small historical control group. Prior to our study, no large study has compared prophylactic regimens using GCV alone to combination therapy using GCV and CMV-IGIV in lung transplantation.
One of the potential confounders in this study was the fact that we changed our immunosuppression protocol to include daclizumab induction at the time we began to use combination prophylaxis. These changes occurred concomitantly because of our concerns regarding the potential increase in CMV disease with induction therapy. While not yet tested in a randomized fashion, daclizumab, a monoclonal antibody directed against the interleukin-2 receptor, has shown promise in reducing acute rejection following lung transplantation (8). Daclizumab has been proven to be efficacious in decreasing acute rejection in kidney, liver, and heart transplantation (11–14).
However, given the high incidence of CMV infection in transplant recipients, concerns still remain regarding the increase in the total immunosuppression with the use of induction agents. The other side of the argument, however, is that less acute rejection would result in less administration of steroid boluses (or other augmentation of immunosuppression) and therefore a reduced incidence of opportunistic infections such as CMV. In a study in the kidney transplant literature by Hengster et al. (15), daclizumab administration added to standard triple or dual immunosuppression therapy was associated with a reduction in the incidence of acute rejection without increasing the risk of infections in general and CMV in particular. In lung transplantation, the impact of induction therapy on CMV disease is largely unknown, particularly in the current era where effective combination prophylactic therapy is available.
In the current study, the use of daclizumab was associated with a trend toward a reduced incidence of acute rejection in the first 6 months post transplant. Also, the administration of steroid boluses (or other enhancements in immunosuppression) was similar in both the study and comparison groups. Even in a subset of patients experiencing AR in the group given monotherapy prophylaxis, there did not seem to be an increase in the incidence of CMV disease as compared to the group with fewer episodes of AR and therefore less exposure to bolus steroid treatments. Conversely, even in the combination prophylaxis group, the few patients who did develop CMV disease did not have a predilection toward the development of AR. Therefore, in our cohort, neither episodes of AR nor enhancement of immunosuppression either with steroids or cytolytics was associated with the development of CMV disease. Instead, the method of prophylaxis, namely with monotherapy alone, was strongly associated with subsequent CMV disease during the first 6 months post transplant.
Our rejection findings differ from those of Garrity et al. who reported a significant decrease in AR in those patients receiving daclizumab induction therapy. The difference in our experience, where there was only a trend toward a reduction in AR in the group given daclizumab, is difficult to explain. However, given the relatively high incidence of AR seen in both our comparison and study groups, the lower incidence of AR in the Loyola study may be reflective of the difference in baseline immunosuppression between the two programs at the time, specifically the use of tacrolimus at Loyola vs. the use of cyclosporine at UAB. Although untested, one could speculate that the use of tacrolimus in combination with daclizumab induction is perhaps more efficacious in preventing AR than an induction regimen involving cyclosporine.
In the Garrity study, despite the reduction in the incidence of AR using daclizumab (and presumably a reduction in the administration of steroid boluses or cytolytics), no difference in CMV disease was observed between the induction and the noninduction groups. These findings were similar to ours in that the frequency of AR and the use of daclizumab was not a predictor of subsequent CMV disease development, and support our contention that only combination prophylactic therapy positively impacts the disease.
Even with a rejection pattern that was similar in both our comparison and study groups, combination CMV prophylaxis resulted in marked, statistically significant reduction in the incidence of CMV disease. Whether daclizumab therapy would ultimately be associated with a reduction in AR over a longer follow-up is unknown. However, given the very low incidence of CMV disease during the first 6 months post transplant in the group receiving combination prophylaxis, it is likely that this effect would persist during the period associated with the highest risk of CMV disease. We conclude, therefore, that in an era where the use of induction agents and other more efficacious immunosuppressive agents will become more common, combination prophylaxis against CMV disease is vital and should be routinely employed in high- and intermediate-risk patients.
Acknowledgments
The authors would like to thank Dr Martin R. Zamora for his assistance with this manuscript.Presented at the American Transplant Congress, April 26 – May 1, 2002 Washington, D.C.
