C ytomegalovirus (CMV) infections contribute significantly

Transcription

1 The Effect Of Polyimmune Gammaglobulin For Prophylaxis Against Reactivation Cytomegalovirus Infection In Kidney And Kidney/Pancreas Transplant Recipients13 Thomas R. McCune, H. Keith Johnson, Robert C. William A. Nylander, David H. Van Buren, and J. MacDonell, Jr., Robert E. Richie, Harold Helderman4 T.R. McCune, H.K. Johnson, P.C. MacDonell, Jr., RE. Richie, WA. Nylander. D.H. Van Buren, J.H. Helderman, Nephrology Division, Department of Surgery, Vanderbilt Transplant Center, Vanderbilt University Medical Center, Nashville, TN (J. Am. Soc. Nephrol. 1992; 2: ) ABSTRACT Cytomegalovirus (CMV) remains the most important infection in the renal transplant recipient. Few data are available that provide guidance for approaches that seek to reduce the reactivation of latent disease after transplantation. To test the efficacy of polyimmune gammaglobulin in kidney and kidney/pancreas transplantation, consenting recipients with serologic evidence of previous CMV disease were randomized to receive i.v. polyimmune gammaglobulin (500 mg/kg) within 3 days of transplant with 250 mg/ kg at weeks 1, 2, 4, and 6 or no prophylaxis. Both groups received identical induction and rejection immunosuppressive therapy. Polyimmune gammaglobulin prophylaxis reduced CMV reactivation infections. The incidence of reactivation infections was half in patients receiving Nashville/rabbit antithymocyte serum (N/R-ATS) compared with those receiving monoclonal anti-cd-3 therapy. Patients receiving polyimmune gammaglobulin along with N/R-ATS had an incidence of infection of only 10%. Reactivation infections were twice as common in patients who Received July Accepted December The abstract was presented at the American Society of Transplant Physicians, May 28, 1991, In Chicago, IL. The manuscript was sent to Dr. william Bennett, Guest Editor for review by expert referees, for editorial decision, and for final disposition. 4Correspondence to Dr. J.H. Helderman, Vanderbilt Transplant Center, Vanderbilt University Medical Center, S-3223 MCN, Nashville. TN /021 0-f 469$03.O0/O Journal of the American Society of Nephrology Copyright C 1992 by the American Society of Nephrology had primary nonfunction and nearly three times as common in patients with acute rejection. Both risk factors were associated with longer anti-t-cell therapy. Polyimmune gammaglobulin prophylaxis should be considered in transplant patients with previous CMV exposure who will be receiving prolonged anti-t-ceil therapy because of acute rejection or primary nonf unction. Key Words: Renal transplantation, cytomegalovirus infection, polyimmune gammaglobulin, primary nonfunction, acute rejection C ytomegalovirus (CMV) infections contribute significantly to morbidity after kidney transplantation. Patients can develop reactivation of this infection after transplantation in spite of previous exposure and the development of protective antibodies against CMV. The greatest risk factor for the recurrence of CMV infection is the intensity of immunosuppression during induction and antirejection therapy (1-5). Various prophylaxis strategies against CMV infections have been developed for these periods of immunosuppression to prevent primary disease (5-8). Few studies guide one s approach at mmimizing the number of and severity of reactivation events. One such study has suggested that the infusion of polyimmune gammaglobulin may decrease the severity of CMV infections in renal allograft recipients who develop reactivation in spite of previously mounted anti-cmv antibodies (5). Three distinct clinical syndromes are described in the renal transplant recipient: (1) primary infection in which a patient who has not been previously infected by CMV (is serologically anti-cmv antibody negative) develops this viral infection); (2) reactivation in which a patient who has had CMV infection before the transplant (is serologically anti-cmv antibody positive at the time of surgery) becomes clinically ill secondary to endogenous virus; and (3) superlnfection in which a seropositive patient becomes infected with a virus of donor origin. Our purpose was to study the effects of our immu- Journal of the American Society of Nephrology 1469

2 Prophylaxis Against CMV nosuppression protocol specifically on reactivation CMV infections in patients undergoing kidney or combined kidney/pancreas transplantation. Additionally, we studied in a prospective, randomized manner the effects of prophylactic polyimmune gammaglobulmn CMV infections in relation to the intensity of the immunosuppression. MATERIALS AND METHODS During the months of March through August 1990, 32 patients with previous CMV exposure confirmed by elevated immunoglobulmn titers were admitted to our center for kidney or kidney/pancreas transplant. Enrollment in this study was offered to all of these patients. Thirty patients who gave consent for the study were randomized to receive either polyimmune gammaglobulmn (Sandoglobulmn, Sandoz, Basle) prophylaxis infusions after transplant or no therapy. Those who were randomized to the treatment group received the first dose of 500 mg/kg of Sandoglobulin within the first 3 days posttransplantation. Subsequent doses of 250 mg/kg were administered i.v. at wk 1, 2, 4, and 6. Infusions were prepared as a 6% concentration with normal saline. The infusion rate was started at 30 ml/h and increased as tolerated to a maximum rate of 1 50 ml/h. During each study, drug infusion, patients were monitored for any changes in blood pressure, pulse, temperature, or symptoms of anaphylaxis. Those who were randomized to the control group received no prophylaxis therapy. Routine immunosuppression for all patients consisted of 250 mg of methylprednisolone given intraoperatively followed by 30 mg/day of oral prednisone for 2 months with a subsequent taper to 1 0 mg/day of maintenance immunosuppression. Azathioprine was administered to all patients intraoperatively at a dose of 3 mg/kg tapered to 2 mg/kg for the next 3 postoperative days and finally tapered to 1 mg/kg per day at a maintenance dose. Polyclonal rabbit antithymocyte serum (Nashville/rabbit-antithymocyte serum [N/R-ATSJ) was administered to patients at a dose of 0.2 ml/kg starting on the first postoperative day and continuing until adequate cyclosporinc levels were obtained. Preparation of, testing of, and efficacy of N/R-ATS is described in detail in the 1990 annual report to the FDA (9). Doses of N/R-ATS serum were administered to obtain a total CD3 lymphocyte level of less than 150 cells/mm3 by flow cytometry. Patients in whom the immunosuppressive therapy with N/R-ATS was unable to reduce the CD3 count to target levels or antibodies against the N/R- ATS developed because of previous exposure to the preparation were treated with a monoclonal anti-cd3 antibody (Orthoclone; Ortho Biotech, New Brunswick, NJ) at a dose of 5 mg/day and maintained until cyclospormn A levels were therapeutic. Cyclosporin A therapy was started when serum creatinine was below 265 mol/l (3.0 mg/dl) at a dose of 13 mg/kg. After 3 days of this dose, cyclosporin A was decreased to 1 0 mg/kg and subsequently individualized to maintain a level of 1 75 to 225 ng/ml whole blood level as measured by HPLC. This level was subsequently tapered by 6 months to obtain a level of 125 to 200 ng/ml and was finally tapered at 1 yr toward a target level of 75 to 1 25 ng/ml for the remainder of the life of the graft. Immunosuppression therapy for acute rejection consisted of methyl prednisolone (250 mg) i.v. on days 1, 2, 4, and 6. N/R-ATS or anti- CD3 were administered for 1 0 to 1 4 days to treat rejection in the face of the lack of response to steroids alone. All patients were monitored for at least 6 months after transplantation for evidence of primary nonfunction (at least one dialysis required within the first 3 days of transplantation), acute rejection, graft failure, and evidence of CMV infections. For the purposes of this study, CMV infections were defined as the presence of organ dysfunction with confirmatory serologic or culture specimens positive for CMV. Bone marrow suppression due to the CMV infections was defined as a white blood cell count below 4,000 celis/ mm3, in the face of a fever of or above 38#{176}Cnot associated with other drugs or the presence of other infections associated with leukopenia. Hepatic infection due to CMV was defined as an elevation of serum glutamic-oxaloacetate transaminase or serum glutamic-pyruvic transaminase in the face of the use of no other hepatotoxic drugs or other infections. Gastrointestinal complications due to CMV included biopsy-proven colitis, gastritis, esophagitis, and/or duodenitis with no other evidence of secondary causes. Pneumonitis due to CMV infections was defined as an interstitial pneumonitis on chest x-ray coupled to hypoxemia and biopsy proof of the presence of CMV In the lung. CMV-associated renal dysfunction was defined as a transient renal dysfunction and biopsy showing no evidence of acute or chronic rejection. Confirmatory studies for all CMV infections included a fourfold increase in immunoglobulin G (IgG) titers and/or a twofold increase in 1gM against CMV temporarily related to organ dysfunction. Culture isolation of CMV obtained from the infected organ, from urine, or from buffy-coat blood cultures obtained within 2 wk of organ dysfunction was also used to confirm evidence for reactivation disease. Results of all but the length of N/R-ATS/anti-CD3 therapy were analyzed by chi-square statistical analysis. The variable lengths of N/R-ATS/anti-CD3 therapy were analyzed by t test. RESULTS Of the 30 patients who consented to participate in this study, 29 were able to complete the entire obser Volume 2 Number 10#{149} 1992

3 McCune et al TABLE 1. Demographics of patient population Polyimmune Gammaglobulin Treatment Group (N= 13) Control Group (N= 16) Sex (Male:Female) 2:8 6:10 NS#{176} Age 38±10 41±10 NS Donor Status (LRD:CAD) 2:11 3:13 NS Transplant Organ Kidney: (Kidney/pancreas) 12:1 14:2 NS Donor CMV Status (Positive/Negative) 10:3 15:1 NS HLAb Match > NS <1 2 3 Transplant History (First:regraft) 6:7 13:3 P< 0.05 CMV Serology (Dilution. IgG) 1:856 ± 277 1:508 ± 222 NS o NS, not significant. b HLA, human leukocyte antigen. vation period. The one patient who dropped out because of the travel requirements had been randomized to the treatment group. This patient received only one infusion of the drug and was therefore excluded from further analysis. Characteristics of the remaining 29 patients are described in Table 1. There were 1 3 patients in the treatment group and 1 6 patients in the control. There was no difference between the groups with respect to age or sex of the recipient or human leukocyte antigen (HLA) match. There were equal proportions of cadaveric to live donor organs within each group. There was no difference in the groups in respect to the CMV antibody status of the donor. Within the treatment group, there was a higher proportion of retransplants than in the control group. There were no complications related to study drug. There was a total of 64 infusions of Sandoglobulmn, and all were tolerated well. One patient did not receive the final infusion of Sandoglobulin because he had developed a severe CMV infection that required i.v. ganciclovir. Although he did not complete the infusion protocol, he was included in the analysis as a treatment failure. There were no factors associated with the recipient, donor, or cross-match that predisposed the subsequent development of CMV disease. Although there was a higher incidence of retranspianted patients who developed CMV disease (Table 2), this difference was not statistically significant. Of the 10 retransplanted patients from both the control and treatment groups, 5 subsequently developed CMV infections. Additionally, the retransplant population accounted for 7 of the 10 cases of primary nonfunction encountered in this study (P < 0.05). The characteristics of the two groups during the 6- month follow-up period are represented in Table 3. TABLE 2. Association of previous transplant with primary nonfunction and CMV disease First Transplant Regraft Total Primary Nonfunction 3 7 (P< 0.05) CMV Disease 3 5 There was no difference between the two groups as to the average total length of N/R-ATS or anti-cd3 induction therapy. Twenty-seven patients received N/R-ATS initially during induction therapy. Two patients received anti-cd3 during induction therapy because of high levels of N/R-ATS antibodies developed during previous renal transplantation. Four patients required the conversion to anti-cd3 because of the failure to lower CD3 counts below target during N/R-ATS therapy. There was no difference between the groups in the incidence of primary nonfunction in spite of the higher incidence of retransplants in the treatment group and the higher incidence of primary nonfunction in that subpopulation. None of the pancreas transplants had immediate function, and one kidney failed. There were no patient deaths during the observation period. There were eight CMV infections that met the criterla for significant CMV disease. Three CMV infections occurred in the treatment group, and five occurred in the control group (Table 4). There was one case of pneumonitis in each group that required ganciclovir therapy. There was one episode of self-limiting CMV hepatitis in each group. In the treatment group, there was one case of transient renal dysfunction without evidence of rejection identified on renal biopsy or any associated nephrotoxic drugs. This case Journal of the American Society of Nephrology 1471

4 Prophylaxis Against CMV TABLE 3. Characteristics of patient population during 6-months follow-up Polyimmune Gammaglobulin Treatment Group Contr Group (N=13) ( - ) Days of N/R-ATS/anti-CD3 I 7 ± 7 19 ± I 0 NS#{176} Delayed Graft Function 5 4 NS Failed Grafts (Kidney:Pancreas) 1:1 0:2 NS 0 NS, not significant. TABLE 4. CMV disease Patient No. Organ Dysfunction Confirmatory Study Therapy Polyimmune Globulin Treatment I Pneumonititis with respiratory Sputum cytology positive for CMV Ganciclovir failure 2 Hepatitis Fourfold increase in anti-cmv lgg Self-limiting titer 3 Renal dysfunction with bone Urine and buffy-coat culture grew Self-limiting marrow suppression CMV Renal biopsy negative for rejection Control I Pneumonititis and bone mar- Buffy coat grew CMV Ganciclovir row suppression 2 Esophagogastroduodenitis Urine culture grew CMV Self-limiting 3 Hepatitis with colitis and Fourfold increase in anti-cmv lgg Self-limiting bone marrow suppression titer 4 Bone marrow suppression Fourfold increase in anti-cmv lgg Self-limiting titer Urine grew CMV 5 Bone marrow suppression Fourfold increase in anti-cmv lgg Self-limiting titer TABLE 5. Demographics of risk factors associated with CMV disease Induction Therapy Which Total Delayed Acute With N/R-ATS Included Anti- Graft Function Rejection Only CD3 Treatment CMV Disease No CMV Disease Control CMV Disease No CMV Disease of renal dysfunction was associated with CMV cultured from the serum and urine and was diagnosed as CMV nephritis. There was one case of CMV esophogoduodenitis and two cases of CMV bone marrow suppression in the control group with white blood cell counts below 2,500 cells/mm3 in each. The immunosuppressive drugs used and the postoperative course of those patients who developed CMV infections as compared with that of those who remained illness free is revealed in Table 5. Within the treatment group, there was only one patient who received N/R-ATS alone as induction therapy who 1472 Volume 2 Number

5 McCune ef al developed CMV infection, whereas four patients in the control group developed CMV infection. Of those patients who received anti-cd3 either alone or in addition to ATS, three developed CMV infections; two of these cases resulted in the face of Sandoglobulin therapy. Within the treatment group, all patients who developed CMV infections had primary nonfunction of the graft, whereas two instances of primary nonfunction were encountered in those who remained illness free. Acute rejection was associated with subsequent CMV infection in four patients in the control group and in two patients in the treatment group. No patient in either group died from CMV infections. DISCUSSION Polyimmune gammaglobulin infusion prophylaxis therapy resulted in a trend of decrease in the mcidence of reactivation CMV (Table 6). In the treatment group, there was a 23% (3 of 1 3) incidence of CMV infections, whereas the control group had an mcidence of 3 1 % (5 of 1 6). Although the sample size was insufficient to allow statistical significance, this decrease does correlate with a similarly conducted study by Steinmuller et al. (5). These investigators were able to show a statistically significant decrease in reactivation CMV with polyimmune gammaglobulin prophylaxis with a study population similar in size to our own (34 patients) (5). The method of immunosuppressive induction therapy may have affected the incidence of CMV reactivation we encountered. The type of induction therapy employed had an effect on subsequent CMV disease (Table 6). Patients who received N/R-ATS as induction therapy had a CMV disease rate of 26% (6 of 27). Patients who received anti-cd3 as part of their induction therapy had a disease rate of 50% (3 of 6) (P < 0.0 1, chi square). The latter group included four patients who required conversion to anti-cd3 because of the failure of N/R-ATS to lower the total CD3 count to the predetermined level. Relatively high rates of CMV infections have been associated with anti-cd3 (3, 10-12). Induction therapy with CD3 was associated with evidence of CMV reactivation by both a fourfold increase in IgG titer and positive CMV cultures from biopsies in 7 of 16 patients with 5 TABLE 6. Incidence of CMV disease as a function of drug therapy Prophylaxis Group 23% (3/13) Control Group 31% (5/16) N/R-ATS (Prophylaxis + Control) 26% (6/27) Anti-CD3 (Prophylaxis + Control) 50% (3/6) N/R-ATS/Prophylaxis Alone I0% (1/10) N/R-ATS/Control Group Alone 31% (4/13) subsequently developing CMV-associated organ dysfunction (10). This suggests that the use of N/R-ATS is associated with a lower intrinsic rate of CMV reactivation disease than is anti-cd3. The lower intrinsic rate of CMV reactivation with N/R-ATS was additionally reduced to 1 0% (1 of 1 0) when polylmmune gammaglobulmn was employed as compared with 31% (4 of 1 3) in N/R-ATS-treated controls who did not receive prophylaxis. The decreased risk of CMV reactivation with N/R- ATS was not at a cost of less efficacy. Only 1 4% (4 of 27) patients initiated on N/R-ATS required conversion to anti-cd3, and none ultimately lost the graft. The only failed kidney occurred in a patient induced with anti-cd3 and was probably less related to the induction drug than to the patient s high-risk status. We can conclude that N/R-ATS is associated with fewer CMV reactivations while being equally efficacious than is anti-cd3 antibody therapy for allograft induction. In patients who experienced acute rejection of the kidney, there was a significant increase in the mcidence of CMV disease regardless of background immunosuppression (Table 7). The incidence of CMV disease with acute rejection was 75% (6 of 8), whereas only 33% (7 of 2 1) of patients without CMV disease had acute rejection (P < 0.05). The association of acute rejection and CMV disease corresponds to that of other studies, which show a similar relationship (1,2,4,5). There are two potential causes for this relationship. The first is related to the increased immunosuppression that reversal of rejection requires. The increased rate of CMV shedding and infections when steroid therapy is used for other diseases is well described (2). The longer length of administration of anti-t-cell preparations has also been associated with increased CMV infections (3, 1 2). In our study, patients who had acute rejection had a longer length of N/R-ATS or anti-cd3 of 23 ± 7 days as opposed to 1 2 ± 6 days in patients who did not have acute rejection (mean ± 1 SE; P < 0.05). CMV has been shown to increase the immunogenicity of graft cells and may be the primary cause itself for the relationship between acute rejection and CMV infections (1 3). It would seem likely that prophylaxis against CMV infections may secondarily decrease the incidence of acute rejection. Our study found that only 38% (5 of 13) of the prophylaxis group developed acute rejection as compared with 50% (8 TABLE 7. Incidence of acute rejection CMV Disease 75% No CMV Disease 33% Prophylaxis Group 38% Control Group 50% Journal of the American Society of Nephrology 1473

6 Prophylaxis Against CMV of 1 6) in the control group, although the study size was insufficient to allow statistical significance. The failure of the graft to function immediately was associated with the higher incidence of subsequent CMV disease. Primary nonfunction was associated with CMV disease in 50% (4 of 8) but occurred in only 24% (5 of 2 1 ) of those patients who developed no subsequent disease. The increase in CMV infection may be related to the increased length of N/R- ATS/anti-CD3 therapy that primary nonfunction demanded. The average length of N/R-ATS or anti-cd3 therapy was 22 ± 9 days in patients with primary nonfunction as opposed to 1 5 ± 7 days in patients with immediate function (P < 0.03). Polyimmune gammaglobulin prophylaxis therapy is safe and tends to reduce CMV reactivation after renal transplantation. It is possible that in patients with previous exposure to CMV, the titers required to suppress infection need not be high or that the polyimmune gammaglobulin infusion may boost the immune system through a mechanism other than provision of passively administered anti-cmv antibodies because these antibodies are found at a low titer in this preparation (14,15). There were no risk factors pretransplant that were predictive for the subsequent development of CMV reactivation. Although there was a trend toward retransplant in patients developing CMV infections, we feel that this was related to the fact that these patients had a higher incidence of primary nonfunction and thus longer anti-t-cell therapy. Although not needed in all patients after renal transplantation, polyimmune gammaglobulin prophylaxis therapy should be considered when induction therapy is to be prolonged because of primary nonfunction or acute rejection. REFERENCES 1. Peterson PK, Balfour HH, Marker SC, Fryd DS, Howard RJ, Simmons RL: Cytomegalovirus disease in renal allograft recipients: A prospective study of the clinical features, risk factors and impact on renal transplantation. Medicine 1 980;59: Glenn J: Cytomegalovirus infections following renal transplantation. Rev Infect Dis 1981 ;3: Thistlewaite RJ, Stuart JK, Mayes JT, et al.: Complications and monitoring of OKT3 therapy. Am J Kidney Dis 1988;6: Lewis RM, Johnson PC, Golden D, Van Buren CT, Kerman RH, Kahan BD: The adverse impact of cytomegalovirus infection on clinical outcome in cyclosporine-prednisone treated renal allograft recipients. Transplantation 1 988;45: Steinmuller DR. Novick AC, Streem SB, Graneto D, Swift C: Intravenous immunoglobulin infusions for the prophylaxis of secondary cytomegalovirusinfection. Transplantation 1 990;49: Snydman DR, Werner BG, Heinze-Lacey B, et al : Use of cytomegalovirus immunoglobulin to prevent cytomegalovirus disease in renal transplant recipients. N Engl J Med 1 987;3 17: Balfour BH, Chace BA, Stapleton JT, Simmons RL, Fryd DS: A randomized, placebo-controlled trial of oral acyclovir for the prevention of cytomegalovirus disease in recipients of renal allografts. N Engl J Med 1987;320: Kramer PB, Bijnen AB, Tenkate FWJ, Jeckel J, Weimar W: Recombinant leukocyte interferon A in doses steroid-resistant acute vascular rejection episodes in renal transplant recipients. Lancet 1984;2: Johnson HK, Miller HC: Annual Report-IND Nashville Rabbit Anti-thymocyte Serum (N/R-ATS). Applied Medical Research, Inc., Nashville, TN, Chou S, Norman DJ: Effect of OKT3 antibody therapy or cytomegalovirus reactivation in renal transplant recipients. Transplant Proc 1 985; 17: Mayes JT, Thistlewaite JR. Stuart JK, Buckingham MR. Stuart FP: Re-exposure to OKT3 in renal allograft recipients. Transplantation 1988; 45: Chiba S, Ikeda S, Agatsuma Y, Nakao T: Active infection with cytomegalovirus and herpes group viruses in children receiving corticosteroid therapy. J Exp Med 1972;106: von Willebrand E, Petterson E, Ahomen J, Hayry P: Cytomegalovirus infection, class II antigen expression and human kidney allograft rejection. Transplantation 1 986;42: Chehimi J, Peppard JR, Emanuel D: Comparative efficacy of different immune globulin intravenous preparations (IGIV) for in vitro human cytomegalovirus (H-CMV) antibody activity. Abstract 9th International Congress of Infections and Parasitic Diseases. Munich, Germany, July 20-26, Berkman SA, Lee ML, Gaale RP: Clinical use of intravenous immunoglobulins. Ann Intern Med l990;1 12: Volume 2 Number