CMV pneumonia in allogeneic BMT recipients undergoing early treatment or pre-emptive ganciclovir therapy

Transcription

1 (2000) 26, Macmillan Publishers Ltd All rights reserved /00 $ CMV pneumonia in allogeneic BMT recipients undergoing early treatment or pre-emptive ganciclovir therapy CM Machado 1,2, FL Dulley 1, LS Vilas Boas 2, JB Castelli 3, MCA Macedo 1, RL Silva 1, R Pallota 1, RS Saboya 1 and CS Pannuti 2 1 BMT Program, Fundação Pró Sangue Hemocentro de São Paulo, Discipline of Hematology; 2 Virology Laboratory of Instituto de Medicina Tropical de São Paulo; and 3 Pathology Department of INCOR-HC, University of São Paulo Medical School, São Paulo, Brazil Summary: The incidence, treatment and outcome of CMV interstitial pneumonia (CMV-IP) were reviewed in 139 consecutive allogeneic BMT patients undergoing extended CMV antigenemia surveillance and two different ganciclovir (GCV) strategies to control CMV infection. Nineteen cases of CMV-IP were reviewed, 16 of 63 patients (25.4%) who received early GCV treatment (ET) and three of 76 patients (3.9%) who received preemptive (PE) GCV therapy. In the ET group, the median time for occurrence of CMV-IP was 55 (range 36 to 311) days. Two patients had three episodes of CMV-IP recurrences after day CMV-IP-related death occurred in two patients (15.4%). In the PE group, 41 patients received pre-emptive GCV therapy prompted by the appearance of positive antigenemia 2 cells. The median time for the occurrence of CMV-IP was 92 (range 48 to 197) days. Response to therapy was observed when GCV was introduced within 6 days of antigenemia positivity. The use of IVIg in association with GCV did not play a major role in response to therapy. The median time for occurrence of CMV-IP was delayed during PE strategy and the cost-effectiveness of CMV surveillance after day +100 should be investigated in this population. (2000) 26, Keywords: CMV; pneumonia; BMT; antigenemia; extended surveillance; IVIg In the early eighties, cytomegalovirus interstitial pneumonia (CMV-IP) was considered a major infectious problem after allogeneic bone marrow transplantation (BMT) because mortality rates higher than 80% were observed even after the introduction of ganciclovir (GCV), the first antiviral with good in vitro and in vivo activity against cytomegalovirus. 1 More recently, the development of new techniques such as the CMV antigenemia (AG) assay and PCR have significantly contributed to the earlier diagnosis of active Correspondence: Dr CM Machado, Av Dr Eneas de Carvalho Aguiar, 470, Sao Paulo, SP Brazil Received 4 June 1999; accepted 6 May 2000 CMV infection, allowing the introduction of pre-emptive antiviral therapy. 2,3 Despite these advances, mortality rates due to CMV pneumonia of around 50% are still observed 4 and late CMV disease has been increasingly reported, particularly in patients who received ganciclovir prophylaxis after engraftment. 4,5 These data emphasise the need for a better understanding of the pathophysiology of the disease and of what can be expected according to different ganciclovir treatment strategies. In the present study, we reviewed the incidence, clinical and radiological findings, treatment and outcome of CMV- IP in 139 consecutive allogeneic BMT patients exposed to extended antigenemia surveillance and two different ganciclovir treatment strategies for the control of CMV infection: early GCV treatment and pre-emptive antigenemiaguided GCV therapy. Patients and methods From May 1993 to July 1997, 202 patients underwent bone marrow transplantation at the BMT unit of the Clinical Hospital, University of Sao Paulo Medical School. Thirteen patients who experienced premature death ( 30 days) and 52 patients who underwent autologous BMT were excluded from the present study. Two patients underwent a second allogeneic BMT. Therefore, the charts of 139 consecutive allogeneic bone marrow transplants were reviewed. Donors and recipients underwent CMV serology assessment prior to transplantation. CMV antigenemia was assessed weekly as part of the routine viral surveillance, from conditioning until day +120 and thereafter, once a month, until 1 year after BMT. During the above-mentioned period, two consecutive ganciclovir treatment strategies were adopted for the control of CMV infection: early treatment (ET) and pre-emptive therapy (PE). Early treatment Patients only received ganciclovir therapy (5 mg/kg every 12 h for 14 or 21 days) if CMV disease was diagnosed. To avoid a delay in the diagnosis, whenever an antigenemia greater than one positive cell was detected, patients underwent CMV screening that consisted of upper gastrointestinal endoscopy, arterial blood gas, XR and CT scan,

2 414 and ophthalmologic evaluation, irrespective of the presence of symptoms. If radiographic changes were seen on XR or CT, the patients underwent lung biopsy or bronchoalveolar lavage (BAL). Interstitial CMV pneumonia (CMV-IP) was defined as detection of CMV in lung biopsy or BAL fluid in the presence of radiographic changes as seen on XR or CT scan. 6 Methods used for CMV documentation were classical isolation, pathological findings and immunocyto- (BAL) or histochemical detection with monoclonal antibodies. CMV- IP was treated with intravenous ganciclovir administered at a dose of 5 mg/kg every 12 h for 14 to 21 days. When used in combination with ganciclovir, IVIg was administered at a dose of 500 mg/kg every other day for a maximum of 10 doses. In contrast to ganciclovir that is supplied by the Ministry of Health of Brazil and is available to all patients with CMV disease, the government does not subsidize intravenous immunoglobulin and its high cost has limited its use to a restricted number of patients. Pre-emptive therapy Whenever antigenemia 2 positive cells was detected, patients received intravenous ganciclovir (5 mg/kg every 12 h for 14 or 21 days). No diagnostic investigation was performed in this group of patients before the introduction of GCV. The antigenemia assay performed in the present study has been described elsewhere. 7 Positive antigenemia was defined as one or more peroxidase-positive (dark-brown nuclear staining) polymorphonuclear cells in Results Only three of the 139 BMT recipients (2.16%) were CMV seronegative before transplant and did not develop CMV disease during follow-up. Sixty-three patients were included in the early GCV treatment strategy and 76 in the pre-emptive strategy. In the ET group, positive antigenemia was detected in 41 patients (65.1%). The median time for the appearance of the first positive antigenemia was 47 (range 7 to 261) days. The incidence of CMV pneumonia was 20.6%. Thirteen patients developed 16 episodes of CMV-IP and the median time for its occurrence was 55 (range 36 to 311) days. Two patients had three episodes of CMV-IP recurrences on days +177, +246 and Positive antigenemia preceded CMV-IP by a median of 6 (range 1 to 17) days. Radiological changes on CT, carried out in accordance with the appearance of positive antigenemia, were seen in 11 of the 13 patients (84.6%) who developed CMV-IP. Only four of the 11 patients with CT changes (36.4%) had visible changes on XR. CMV screening failed to demonstrate established CMV pneumonia in two patients. In these cases, GCV therapy was delayed until respiratory symptoms were present and/or radiological patterns showing interstitial pneumonia (9 and 17 days after the appearance of positive antigenemia, respectively). Arterial blood gas results changed in these patients only after the appearance of respiratory symptoms. In the ET group, only three patients received GCV and IVIg for the treatment of CMV-IP. The remaining 10 received ganciclovir alone. CMV-related deaths were observed in the two patients in whom GCV therapy was delayed. These data are summarised in Table 1. In the PE group, positive antigenemia was detected in 50 patients (65.7%) and the median time for its appearance was 40.5 (range 5 to 128) days. Nine patients had only one positive cell on the antigenemia test and were not treated. These patients did not develop CMV disease. The remaining 41 received pre-emptive GCV therapy. CMV-IP was diagnosed in three patients (3.9%) on days +48, +92 and Two patients were already receiving pre-emptive antigenemia-guided GCV for 5 and 6 days, respectively, when CMV-IP was diagnosed. The remaining one had received pre-emptive therapy on day +46 for 14 days, and developed CMV-IP on day +92 without concurrent positive antigenemia. These patients developed fever of unknown origin and underwent a CT scan that revealed interstitial infiltrates. CMV-IP was confirmed in BAL fluid in one patient and on lung biopsy in the remaining two. The patients were treated with GCV for 21 days and no IVIg was added. CMV related deaths did not occur in this group. These data are summarised in Table 1. Among the 14 patients with CMV-IP who responded to treatment, 13 (92.8%) had concurrent positive antigenemia. Four of them were already receiving GCV therapy when CMV pneumonia was diagnosed: patients 12 and 13 were receiving GCV for gastrointestinal CMV disease diagnosed on days +82 and +38, respectively, and patients 14 and 16 were receiving pre-emptive GCV therapy. The remaining nine patients were started on GCV within 6 days of antigenemia positivity. Patient 15 did not develop positive antigenemia at the time of diagnosis of CMV-IP, but the CT scan allowed early identification of the interstitial infiltrate and GCV was introduced as soon as CMV pneumonia was confirmed (BAL fluid). IVIg was added to GCV in three episodes of CMV-IP and did not seem to play a major role in improving response to ganciclovir therapy. During the present study, the overall mortality rate of CMV-IP was 12.5%. Discussion As expected, we observed a reduction in the incidence of CMV-IP from 20.6% (when no intervention with ganciclovir was done until diagnosis was established), to 3.9% after the introduction of pre-emptive antigenemia-guided GCV therapy. This was a non-randomized prospective study and patients were switched from early treatment to pre-emptive therapy as soon as the former was shown to be ineffective in treating all cases of CMV-IP. CMV screening during the ET strategy failed to demonstrate CMV-IP in two of the 13 patients (15.4%) who developed pneumonia and the preemptive strategy did not prevent development of CMV-IP in three of the 41 patients (7.3%) who received antigenemia-guided ganciclovir therapy. In the present study we observed that patients with

3 Table 1 Patient characteristics acccording to ganciclovir policy after allogeneic BMT Early treatment (63) Pre-emptive (76) P value 415 Median age (range) (years) 30 (4 to 47) 25 (3 to 51) NS Serological status pre-bmt (%) Positive 62 (98.4%) 74 (97.3%) NS HLA graft type (%) Full match 61 (96.8) 74 (97.3) NS Mismatch/Unrelated 2/0 1/1 Underlying disease CML NS SAA NS AML 8 6 NS ALL 7 7 NS Other 4 6 NS Conditioning Busulphan + melphalan NS Busulphan + cyclophosphamide NS TBI + cyclophosphamide TBI + cyclophosphamide + CS 4 3 NS Other 1 1 NS Acute GVHD Grade II 44 (69.8%) 48 (63.2%) NS Chronic GVHD Extensive 29 (60.4%) a 29 (50.8%) a NS a Evaluated in 48 (ET) and 57 (PE) patients with survival greater than 100 days. TBI = total body irradiation; CS = corticosteroids. Table 2 Time of occurrence, treatment and outcome of CMV-IP in allogeneic BMT recipients during the first year after BMT No. Day of Ag CMV-IP CMV-IP rec XR/CT changes a Therapy Response no Yes/Yes ET-GCV Yes no No/Yes ET-GCV Yes /+311 Yes/Yes ET-GCV + IVIg Yes No/Yes ET-GCV Yes no No/Yes ET-GCV Yes no No/Yes ET-GCV + IVIg Yes no No/No ET-GCV No no Yes/Yes ET-GCV Yes no No/Yes ET-GCV Yes no No/Yes ET-GCV Yes no No/No ET-GCV No no Yes/Yes ET-GCV + IVIg Yes no No/Yes ET-GCV Yes no Not done PE-GCV Yes no Not done PE-GCV Yes no Not done PE-GCV Yes a Changes concurrent or soon after the appearance of positive antigenemia. CMV-IP rec = CMV pneumonia recurrence. CMV-IP who received pre-emptive therapy or were diagnosed within 6 days of developing antigenemia, responded to GCV treatment. This observation suggests that the earlier the introduction of GCV, the better the response to therapy, and the addition of IVIg seemed to play only a minor role in the outcome of CMV-IP. The combination of high-dose intravenous immunoglobulin (IVIg) and ganciclovir to treat CMV-IP has been adopted by the majority of BMT centers 8 since 1988, when two historic controlled studies showed a decrease in mortality rates due to pneumonia from 100% to 20% and from 85% to 48%, respectively. 9,10 In both studies, results were evaluated based on comparison with historical controls and no randomized, placebo-controlled studies are available to validate these observations. The next step should have been the inclusion of controls in the studies, before establishing combined therapy as the optimum treatment of CMV-IP in BMT recipients. 11 However, such studies were never conducted and the use of combined treatment was adopted at most transplant centers. Indeed, at some of these centers this combination was extended to the treatment of CMV gastrointestinal disease (CMV-GI) for which the need for combined treatment is even more questionable. 8,12 According to murine models of CMV-IP pathogenesis, the rationale for combination IVIg is modulation of the immunological changes caused by expression of viral antigens by

4 416 CMV replication in the lung. 13 We may assume that if ganciclovir is introduced early and if control of viral replication is achieved, the immunological dysfunction triggered will be smaller. Consequently, a lower impact of the use of intravenous immunoglobulin on the response to treatment of pneumonia may be expected when the antiviral agent is introduced early in the course of the disease. We also observed a delay in the median time of occurrence of CMV-IP from 55 to 92 days when we moved from early treatment to pre-emptive therapy. This observation may reflect interference of ganciclovir on natural CMVspecific T cell recovery, which is expected to occur between days +40 and +90. The prolonged use of ganciclovir induces a delay in recovery of these responses and contributes to the appearance of late CMV disease, as demonstrated in patients receiving ganciclovir prophylaxis until day ,5 The reactivation of CMV infection after BMT is a powerful stimulus for the development of the CMVspecific T cell response, and introduction of ganciclovir therapy soon after the appearance of antigenemia may abort this response. Concerning the delay observed in the median time of occurrence of CMV-IP, the question arises whether CMV surveillance should be extended after day Most BMT centers perform CMV surveillance until day +100 and use ganciclovir prophylaxis or pre-emptive therapy during this period. However, antigenemia recurrences are common after allogeneic BMT if no secondary GCV prophylaxis is used. Patients who develop grade II acute GVHD or extensive chronic GVHD showed an 82.7% and 100% probability of antigenemia recurrence during the first year after transplant, respectively. 14 It seems reasonable to preemptively treat these episodes since their occurrence suggests a risk of developing CMV disease. These patients should benefit most from antigenemia surveillance after day +100, as should those at increased risk of developing severe GVHD, such as unrelated or mismatched donor BMT recipients. The low CMV-IP mortality rate observed in the present series may be a consequence of extending viral surveillance until 1 year after BMT. Although pre-emptive therapy has been widely adopted, some questions remain. What doses of GCV should be used and for how long? Which method is more cost-effective, antigenemia assessment or PCR to guide pre-emptive therapy? Should CMV surveillance be extended in patients receiving GCV pre-emptive therapy or prophylaxis? During the early treatment strategy period, when the CMV screening was performed, we observed that 11 of the 13 patients (84.6%) who developed CMV pneumonia had tomographic findings compatible with interstitial pneumonia suggesting established disease concurrently or soon after the appearance of positive antigenemia, even when the patient did not yet have any symptoms. Generally, the doses used for pre-emptive therapy are lower than those recommended for the treatment of CMV disease, based on the assumption that intervention occurs before the disease is established. Thus, Goodrich et al 15 used doses of 5 mg/kg every 12 h for only 7 days and reduced them to 5 mg/kg/day up to day In this study, shell-vial guided pre-emptive therapy was adopted which led to a later introduction of ganciclovir when compared to PCR or antigenemia-guided pre-emptive studies. In fact, 12% of the patients had already developed CMV disease at the time of intervention in the study by Goodrich et al. In another study, ganciclovir was introduced when positive antigenemia in three or more cells was detected, at a dose of 5 mg/kg every 12 h for 7 days followed by 5 mg/kg once a day for 3 weeks. 16 If CMV-IP is already established as was observed in the majority of cases in the present study, the pre-emptive ganciclovir doses adopted in many BMT centers might not be sufficient to treat established disease and the use of lower doses might lead to a worse therapeutic response. The present data suggest that, in pre-emptive interventions dictated by antigenemia, full therapeutic doses of ganciclovir (5 mg/kg every 12 h for 21 days) should be adopted since some patients will present with established disease even though they are asymptomatic. In the preemptive approach dictated by PCR positivity, some authors have demonstrated that lower doses of GCV administered for less than 21 days are sufficient, due to the higher sensitivity of this technique in detecting active infection when compared to antigenemia assessment. A lower incidence of CMV disease, shorter duration of treatment, and a decrease in the side-effects of GCV have been observed with the use of PCR in the surveillance of CMV viremia after BMT. 3 Although the cost effectiveness of CMV antigenemia surveillance after day +100 needs to be established before this is widely accepted, we believe that a defined group of patients, such as those with grade II acute GVHD or extensive chronic GVHD, may benefit most from such an approach. The data presented in this study support the concept that several steps (effective CMV surveillance, diagnosis and treatment), are involved in the control of CMV-IP after BMT and a better comprehension of their roles is essential to decrease mortality rates from CMV pneumonia. References 1 Shepp DH, Dandliker PS, Miranda P et al. Activity of 9-[2- hydroxy-1-(hydroxymethyl) ethoxymethyl] guanine in the treatment of cytomegalovirus pneumonia. Ann Intern Med 1985; 103: Boeckh M, Bowden RA, Goodrich JM et al. Cytomegalovirus antigen detection in peripheral blood leukocytes after allogeneic marrow transplantation. Blood 1992; 80: Einsele H, Ehninger G, Hebart H et al. Polymerase chain reaction monitoring reduces the incidence of cytomegalovirus disease and the duration and side-effects of antiviral therapy after bone marrow transplantation. Blood 1995; 86: Boeckh M, Riddell S, Cunningham T et al. Increased risk of late CMV infection and disease in seropositive allogeneic marrow transplant recipients after ganciclovir prophylaxis is due to a lack of CMV-specific T-cell responses. Orange Beach, Alabama: 6th International CMV Workshop 1997 (Abstr. 70). 5 Li CR, Greenberg PD, Gilbert MJ, Riddell SR. Recovery of HLA-restricted cytomegalovirus (CMV)-specific T-cell responses after allogeneic bone marrow transplant: correlation with CMV disease and effect of ganciclovir prophylaxis. Blood 1994; 83: Ljungman P, Plotkin SA. Workshop on CMV disease: definitions, clinical severity scores and new syndromes. Scand J Infect Dis 1995; 99 (Suppl.):

5 7 Halwachs G, Zach R, Pogglitsch H et al. A rapid immunocytochemical assay for CMV detection in peripheral blood of organ-transplanted patients in clinical practice. Transplantation 1993; 56: Ljungman P, Bock RD, Cordonnier C et al. Practices for cytomegalovirus diagnosis, prophylaxis and treatment in allogeneic bone marrow transplant recipients: a report from the Working Party for Infectious Diseases of the EBMT. Bone Marrow Transplant 1993; 12: Emanuel D, Cunningham I, Jules-Elysee K et al. Cytomegalovirus pneumonia after bone marrow transplantation succcesfully treaeted with the combination of ganciclovir and highdose intravenous immune globulin. Ann Intern Med 1988; 109: Reed EC, Bowden RA, Dandliker PS et al. Treatment of cytomegalovirus pneumonia with ganciclovir and intravenous cytomegalovirus immunoglobulin in patients with bone marrow transplants. Ann Intern Med 1988; 109: Frank I, Friedman HM. Progress in the treatment of cytomegalovirus pneumonia. Ann Intern Med 1988; 109: Ljungman P, Cordonnier C, Einsele H et al. Use of intra- venous immune globulin in addition to antiviral therapy in the treatment of CMV gastrointestinal disease in allogeneic bone marrow transplant patients: a report from the European Group for Blood and Marrow Transplantation (EBMT). Infectious Diseases Working Party of EBMT. Bone Marrow Transplant 1998; 21: Zaia JA. Pathogenesis of CMV-associated diseases in Transplant Proc 1991; 23 (Suppl. 3): Machado CM, Vilas Boas LS, Mendes AV et al. Extended antigenemia surveillance and late CMV infection after allogeneic BMT. San Francisco, California: 39th ICACC, 1999, Abstr Goodrich JM, Mori M, Gleaves CA et al. Early tratment with ganciclovir to prevent cytomegalovirus disease after allogeneic bone marrow transplantation. New Engl J Med 1991; 325: Boeckh M, Gooley TA, Myerson D et al. Cytomegalovirus pp-65 antigenemia-guided early treatment with ganciclovir vs ganciclovir at engraftment after allogeneic marrow transplantation: a randomized double-blind study. Blood 1996; 88: