Comparison of two doses of intravenous immunoglobulin after - Nature

Bone Marrow Transplantation, (1999) 23, 929–932  1999 Stockton Press All rights reserved 0268–3369/99 $12.00 http://www.stockton-press.co.uk/bmt

Comparison of two doses of intravenous immunoglobulin after allogeneic bone marrow transplants A Abdel-Mageed1, J Graham-Pole1, MLU Del Rosario1, J Longmate2, S Ochoa3, M Amylon4, GJ Elfenbein5, J Janiec1, J Jansen6 and HM Lazarus7 1

University of Florida, Pediatric Hematology/Oncology, Gainesville, FL; 2City of Hope Medical Center, Duarte, CA; 3University of Florida, Department of Statistics, Gainesville, FL; 4Stanford University, Pediatric Hematology/Oncology, Palo Alto, CA; 5University of South Florida, Department of Internal Medicine, Tampa, FL; 6Indiana Blood and Marrow Transplantation, Indianapolis, IN; and 7 University Hospital of Cleveland, Department of Medicine, Cleveland, OH, USA

Summary:

Patients and methods

Intravenous immunoglobulin has been used after bone marrow transplants to prevent infections and acute graft-versus-host disease. However, the minimum dose required for protection is unknown. This may have significant economic implications. A multicenter randomized clinical trial compared the impact of two intravenous immunoglobulin doses on systemic infections and acute graft-versus-host disease in transplant recipients. Either 250 mg/kg or 500 mg/kg was given weekly from day −8 to day +111. Multivariate analysis was used to assess the effect of dose and other risk factors on event-free survival, systemic infection, and acute graftversus-host disease. The two-dose cohorts had similar event-free survival and infection frequencies. The higher dose was associated with less acute graft-versushost disease (P = 0.03). Keywords: intravenous immunoglobulin; infection; graft-versus-host disease; allogeneic bone marrow transplantation

Study design Three hundred and fifty patients transplanted in 10 centers entered the trial. Eligible patients were those without documented infection at entry who had an Institutional Review Board-approved consent form. Patients were stratified by age (⬍21 vs ⬎21), CMV serology (donor/recipient negative vs either positive), and HLA-match (6 of 6 antigen match vs other), and randomized to receive either 250 mg/kg or 500 mg/kg of IVIG weekly from day −8 to day +111 post-transplant (18 doses). All donors were related. Pre-study forms and study flow sheets provided weekly information for 24 weeks on the clinical status and development of any infection or acute GVHD requiring treatment. An off-study form was completed either at 24 weeks post-transplant, if they had completed the whole IVIG course, or earlier if they were taken off study. The reasons for removal from study were: (1) death; (2) relapse; (3) deviation from the protocol such as termination of IVIG use or use of a different dosing schedule. IVIG

Intravenous immunoglobulin (IVIG) is often given after allogeneic bone marrow transplants to decrease incidence of interstitial pneumonia, particularly that related to cytomegalovirus (CMV).1–4 Other studies of IVIG report less systemic bacterial infections and acute graft-versus-host disease (GVHD).4–7 Although a weekly IVIG dose of 500 mg/kg is often given, the optimum dose is unknown. We conducted a multicenter randomized study to compare incidence of systemic infections and acute GVHD after 250 and 500 mg/kg of IVIG.

Correspondence: Dr J Graham-Pole, University of Florida, Pediatrics Hematology/Oncology, PO Box 100296, Gainesville, FL 32610–0296, USA Received 27 October 1997; accepted 15 November 1998

Gammagard (Baxter Hyland, Glendale, CA, USA) was used in this study. This is a native monomeric IgG preparation produced by the combination of Cohn fractionation with DEAE-Sephadex ion exchange chromatography. This procedure avoids the use of low pH proteolytic enzymes or other chemical purification methods that tend to reduce the required biologic activity. Commercially available lots had minimal inter-lot variability.8 Definition of terms Infections were identified by organism and site of involvement. Severity was graded as follows: (1) grade 1: mild, consisting of either local or systemic infection with no reported symptoms and/or lasting ⭐3 days; (2) grade 2: moderate to severe symptomatic systemic infection and/or lasting ⬎3 days but non-fatal; (3) grade 3: fatal infections. Bone marrow engraftment was defined as neutrophils 0.5 × 109/l. Acute GVHD in this study was considered present if the overall grade was II or greater according to the system of Glucksberg et al.9

Immunoglobulin, infection, and GVHD post-transplant A Abdel-Mageed et al

930

Table 1

Patient characteristics

Table 2 IVIG regimen

Patients Age 0–21 22–52 Diagnosis ALL ANL/AML/AMOL CML AA variants Other Race Black White Other Gender of donor Missing Male Female HLA match Missing Yes No Laminar air flow Missing Yes No Oral antibiotics Missing Yes No Donor CMV status Missing Negative Positive Recipient CMV status Missing Negative Positive

250 mg/kg

Reasons for coming off study

Reason

No. of patients (%)

500 mg/kg

165 167

80 87

85 80

87 94 87 21 43

41 45 44 10 25

46 49 43 11 18

25 301 6

11 149 5

14 152 1

12 174 146

5 83 77

7 91 69

2 292 36

0 149 14

2 143 22

2 264 66

0 131 34

2 133 32

2 163 167

0 91 74

2 72 93

16 165 151

10 84 71

6 81 80

7 162 163

5 78 82

2 84 81

Completed follow-up without adverse event Died of GVHD and/or other complication Died of infection Relapse Total

201 26 77 28 332

(60.5) (7.8) (23.2) (8.5) (100.0)

Statistics Comparisons of proportions and survivorship curves were analyzed. The former included overall deaths, death from infections and deaths from acute GVHD with no evidence of infection by cultures and/or post-mortem examinations. Those with evidence of both were classified as infectionrelated deaths but were included also in acute GVHD incidence. Survivorship data included event-free survival (EFS), incidence of fatal infections, of non-fatal infections, of engraftment and of acute GVHD. For survival analyses, follow-up was truncated to 160 days to avoid the bias caused by selective follow-up in estimating the survival curves. Pearson chi-square tests were used to compare proportions. Survival curves were estimated using the Kaplan– Meier method and compared using logrank tests and multiproportional hazards regression models. Dependent variables analyzed were: (1) numbers of fatal and non-fatal infections; (2) numbers of fatal and non-fatal bacterial, viral, fungal and protozoal infections; (3) numbers of patients with multiple infections; (4) distribution of infections by severity (grade 3 vs grade 1 plus 2); (5) mean time of onset of infection post-BMT; (6) mean duration of infections; (7) number of cases of acute GVHD requiring treatment; (8) time of onset of acute GVHD. Results

Supportive care

Patients

Patients were housed in laminar air flow or HEPA-filtered rooms. Hand washing was carried out before every patient contact. Patients also received low-bacterial diets, and prophylaxis with cotrimoxazole or pentamidine against Pneumocystis carinii infections. Many received acyclovir against herpes virus reactivation and fluconazole against fungal infections. Almost half of the patients received nonabsorbable prophylactic oral antibiotics. The CMVseronegative patients received only CMV-negative blood products; seropositive patients received CMV-unscreened blood products. All blood products were irradiated to 15 Gy. Blood transfusions were given to keep hematocrits ⭓25% and platelets ⭓20 × 109/l. Febrile (T ⭓38.5°C) patients with neutropenia received broad-spectrum intravenous antibiotics according to each investigator’s discretion, after cultures were obtained. Amphotericin B was added if a patient continued febrile after 3–5 days with no identified source.

Eighteen patients were excluded because of inadequate data or protocol deviation. The remaining 332 patients form the basis of this report. Median age was 21 years (mean, 22 years). Distribution of risk factors was balanced in the two arms (Table 1). Thirty-six subjects were lost to follow-up without an event before 160 days. Of these, 19 were followed for less than 140 days. The reasons for early study termination were unrelated to the dose of IVIG given (Table 2). Event-free survival There was no significant relationship between IVIG dose and EFS (Figure 1). With a 90% probability of detecting a 57% change in the relative hazard, there was also no significant difference in EFS after adjusting for age, CMV status and HLA matching in a multivariate proportional hazards model.


931

1.0

1.0

Cumulative incidence

Probability of survival

250 mg/kg 0.8 500 mg/kg 0.6 0.4 0.2

0.8 250 mg/kg 0.6 0.4

500 mg/kg

P = 0.03

0.2 0.0

0.0 0

50

100

150

200

0

5

10

15

20

25

30

Weeks post-BMT

Days after BMT Figure 1 Event-free survival and IVIG dose in BMT recipients: P = 0.4.

Figure 3 Incidence of GVHD and its relationship with IVIG dose: P = 0.03.

Fatal infections and infection-specific survival Infection accounted for, or played a major role in 77 (59%) of the 131 deaths. There was no significant difference in the infection risk between the two IVIG arms (P = 0.8). The study had 90% power to detect a 54% difference in relative hazard of infection and the estimated hazard of infection on the low-dose arm, relative to the high-dose arm, was between 0.60 and 1.48, adjusted for age, CMV status and HLA matching, using a proportional hazards regression. Age (P = 0.0001) was the only risk factor significantly associated with fatal infection (Figure 2). IVIG dose did not interact with age to explain this difference in incidence of fatal infections (P ⬎ 0.7). GVHD Of the 332 patients, 164 (49.4%) developed grade II or higher acute GVHD. Median time of onset was 2.5 weeks. The higher dose of IVIG was associated with less acute GVHD (approximately 58% vs 44%: P = 0.03 log rank: Figure 3). Discussion

Cumulative incidence

Similar rates of fatal and non-fatal infections and EFS were seen in both arms. This is consistent with the findings of other studies using different IVIG doses.10 Although this trial was not designed primarily to address whether IVIG

logrank P = 0.0001

0.8

0.4

over 31 21–30 11–20 0–11

0.0 0

50

100

150

200

Days post BMT Figure 2 Incidence of fatal infections in relation to age: P = 0.0001.

reduces acute GVHD, other controlled trials report significantly less acute GVHD in people receiving IVIG (500– 1000 mg/kg) weekly for 3–4 months post-transplant.11,12 Meta-analysis of controlled trials on 379 patients found less acute GVHD (RR = 0.68) in IVIG recipients relative to untreated controls.11 One study showed less acute GVHD and suggested it may depend on immunoglobulin trough levels.12 Several studies have reported the protective effect of IVIG against CMV infection in CMV-seropositive patients.13–16 Among CMV-positive recipients/donors, we found no significant difference in CMV infections between the two doses. There were more fatal CMV infections amongst CMV-positive patients in both arms. The toxicity attributable to IVIG was minimal. However, it is expensive and transmission of infection has been reported.17,18 Cost considerations and advances in infection prevention explain the trend in using IVIG selectively. References 1 Wingard JR. Advances in the management of infectious complication after marrow transplant. Bone Marrow Transplant 1990; 6: 371–383. 2 Winston DJ, Ho WG, Champlin RE. Cytomegalovirus infections after allogeneic marrow transplant. Rev Inf Dis 1990; 12: S776–S792. 3 Bowden RA, Sayers M, Fluornoy N et al. Cytomegalovirus immune globulin and seronegative blood products to prevent primary cytomegalovirus infections after marrow transplant. New Engl J Med 1986; 314: 1006–1010. 4 Graham-Pole J, Camitta P, Casper J et al. Intravenous immunoglobulin may lessen all forms of infection in patients receiving allogeneic marrow transplantation for acute lymphoblastic leukemia: a Pediatric Oncology Group study. Bone Marrow Transplant 1988; 3: 559–566. 5 Peterson FB, Bowden RA, Thornquist M et al. The effect of prophylactic intravenous immune globulin in the incidence of infections in marrow transplant recipients. Bone Marrow Transplant 1987; 2: 141–148. 6 Stiehm ER. Human gammaglobulin as therapeutic agents. Adv Pediatr 1988; 35: 1–72. 7 Pirofsky B. Intravenous immune globulin therapy in hypogammaglobulinemia. Am J Med 1984; 76: 53–60.


932

8 Peltier MKH, Filipovich AH, Bechtel M et al. Randomized double-blinded comparison of three intravenous immunoglobulin products in bone marrow transplantation. Semin Hematol 1992; 29: 112–115. 9 Glucksberg H, Storb R, Fefer A et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors. Transplant 1974; 18: 295–304. 10 Wolff S, Winston DJ, Antin J et al. Multi-center randomized double-blind dose response trial of IVIG for the prevention of infection after allogeneic bone marrow transplantation (AlloBMT). Blood 1996; 88 (Suppl. 1): 303A (Abstr. 1199). 11 Sullivan KM. Immunomodulation in allogeneic marrow transplantation: use of intravenous immunoglobulin to suppress acute graft-versus-host disease. Clin Exp Hematol 1996; 104 (Suppl. 1): 43–48. 12 Cottler-Fox M, Lynch M, Pickle LW et al. Some but not all benefits of intravenous immunoglobulin therapy after marrow transplantation appear to correlate with IgG trough levels. Bone Marrow Transplant 1991; 8: 27–33. 13 Wolff SN, Fay JW, Herzig RH et al. High-dose weekly intra-

14

15

16

17

18

venous immunoglobulin to prevent infections in patients undergoing autologous bone marrow transplantation or severe myelosuppressive therapy. Ann Intern Med 1993; 118: 937– 942. Kapoor N, Copelan EA, Tutschka PJ. Cytomegalovirus infection in bone marrow transplantation recipients: use of intravenous gammaglobulin as a prophylactic and therapeutic agent. Transplant Proc 1989; 21: 3095–3096. Winston DJ, Ho WG, Lin CH et al. Intravenous immunoglobulin for prevention of cytomegalovirus infection and interstitial pneumonia after bone marrow transplantation. Ann Intern Med 1987; 106: 12–18. Casper JT, Sedmak G, Harris RE et al. Intravenous immunoglobulin: use in pediatric bone marrow transplantation. Semin Hematol 1992; 29: 100–105. Duhem C, Dieato M, Ries F. Side-effects of intravenous immune globulin. Clin Exp Immunol 1994; 97 (Suppl. 1): 79–83. Mirbak SA, Chapel HM. Adverse effects of intravenous immunoglobulins. Drug Safety 1993; 9: 254–262.