- Email: [email protected]
Hematuria Due to Adenoviral Infection in Bone Marrow Transplant Recipients
Hematuria Due to Adenoviral Infection in Bone Marrow Transplant Recipients I. Bil-Lula, M. Ussowicz, B. Rybka, D. Wendycz-Domalewska, R. Ryczan, E. Gorczyn´ska, K. Kałwak, and M. Woz´niak ABSTRACT Late-onset hemorrhagic cystitis (HC) caused by adenovirus (AdV) infection is a common complication in hematopoietic stem cell transplantation (HSCT) recipients. However, limited information exists regarding adenovirus-associated HC. We report a retrospective study of 84 hematopoietic stem cell transplant recipients that evaluated the incidence and risk factors for AdV-induced HC. The development of HC was strongly related to adenoviral infection (P ⫽ .004). Among 13 patients who developed late-onset HC, AdVs were identified as a causative agent in 10 cases. AdV preferentially affected younger (P ⫽ .013) and male patients. Affected subjects had been transplanted for either malignant (7/10) or nonmalignant disorders (3/10). Most cases of AdV-hematuria were self-limited single or recurrent mild hemorrhagic episodes (P ⫽ .000), occuring at a median of 41 days after transplantation and lasting an average of 4 days. Viral load in patients with AdV-induced HC was similar to infected subjects who did not develop HC (2.5 ⫻ 103 vs 3.4 ⫻ 103 copies/mL). We HC occurring before 200 days was associated with a greater risk of a fatal outcome (P ⫽ .002) but occurrence of AdV infection did not affect a patient’s survival. Our study confirmed the suggestion that non-AdV coinfections may worsen the course of AdV-HC. mmunodeficiency renders allogeneic hematopoietic stem cell transplantation (HSCT) recipients prone to opportunistic infections that usually occur during the early posttransplantation period.1 Adenovirus (AdV) infections are believed to be a frequent complication of HSCT. Although AdV infection is usually not life threatening, intensified immunosuppression or lack of effective antiviral therapy can favor, a devastating clinical course leading to dangerous complications of obstructive uropathy, hydronephrosis, and renal failure.1– 4 Late-onset hemorrhagic cystitis (HC) has been attributed predominantly to AdV infection with serotypes 3, 7, 11, 21, 34, 35.5–10 HC after bone marrow transplantation is among the leading clinical problems in posttransplant care. The etiology of hematuria may pose a diagnostic problem. Late-onset HC is more likely to result from a viral pathogenesis,5,11 but cyclophosphamide may also play a role.12 The HC etiology may not be clear also because some BK and JC viruses are commonly found in the urine of patients without HC. Patients with AdV-induced hemorrhagic cystitis (AdV-HC) exhibit renal dysfunction and fever, which are also common signs for HC due to BK virus
I
(BKV) infection. Fever, urodynia, and hematuria may also accompany acute graft-versus-host disease (aGVHD) or bacterial infections arise as or a complication of the conditioning regimen.3,6 Many reports have described risk factors for the development of HC caused by BKV as the main pathogen of HC.5,13 The aim of the present study was to focus on the role of risk factors for AdV replication in hematuria development. Because of limitations in culture methodology for AdV detection, we used highly specific
From the Department of Clinical Chemistry (I.B.-L., M.W.) and Department of Paediatric Bone Marrow Transplantation, Oncology, and Haematology (M.U., B.R., R.R., E.G., K.K.), University of Medicine, Wrocław, Poland; Central University Hospital (D.W.-D.), Wrocław, Poland; and Department of Pharmacology (M.W.), University of Saskatchewan, Saskatoon, Canada. Sponsored by the Polish Ministry of Science and Higher Education (grant no. NN-401-004-536). Address reprint requests to Iwona Bil-Lula. Department of Clinical Chemistry, University of Medicine, Wrocław, Poland. E-mail: [email protected]
© 2010 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/–see front matter doi:10.1016/j.transproceed.2010.08.049
Transplantation Proceedings, 42, 3729 –3734 (2010)
3729
3730
BIL-LULA, USSOWICZ, RYBKA ET AL
sensitive molecular methods based on real-time polymerase chain reaction (PCR).
Table 2. Conditioning Regimens of HC Patients (n ⴝ 13) Type of Conditioning
Myeloablative
MATERIALS AND METHODS Study Group We retrospectively examined 84 mostly pediatric patients who underwent allogeneic hematopoietic stem cell transplantation between spring 2007 and summer 2009. To exclude cyclophosphamide-induced HC, we considered only patients who presented with hematuria in the late period after HSCT. Urine samples from all HSCT recipients were collected weekly during the hopitalization period and once monthly thereafter. The recipient characteristics are summarized in Table 1. Owing the high variability of treated patients, we used various conditioning regimens (Table 2). Ex vivo T-cell depletion of the graft was performed for all haploidentical transplantations (n ⫽ 9).
Definitions We defined patients as developing AdV-HC as the presence of hematuria accompanied by AdV viruria. To eliminate other causes of hematuria, we excluded samples from female patients suspected of vaginal bleeding and those with possible bacteruria. We included only de novo hematuria that was manifested ⬎7 days after HSCT. Urinalysis and microscopic techniques were performed as screening tests on a regular basis. Hematuria was defined as mild if 3–10 erythrocytes were observed in the field of vision, moderate for 11–50 erythrocytes, and severe for erythrocytes densely covering the field or if there was macroscopic hematuria. Table 1. Characteristics of HSCT Recipients Suffering from Hemorrhagic Cystitis (HC), Based on Occurrence of AdV Infection (n ⴝ 13) Characteristic/Variable
HC of Adenoviral Infection
HC of Other Causes
No. of patients, n (%) 10 (76.9) 3 (23.1) Gender, n (%) Male 7 (53.8) 2 (15.4) Female 3 (23.1) 1 (7.7) Age, median/range (y) 6/1–19 26/10–42 Diagnosis, n (%) Malignancy 7 (53.8) 2 (15.4) Nonmalignant disease 3 (23.1) 1 (7.7) Donor type, n (%) Matched unrelated donor 7 (53.8) 2 (15.4) Relative donor HLA-identical sibling 1 (7.7) 0 donor Haploidentical donor 2 (15.4) 1 (7.7) T-cell depletion, yes/no 2 (15.4)/8 (61.5) 1 (7.7)/2 (15.4) [n (%)] Mieloablative conditioning 8 (61.5)/2 (15.4) 1 (7.7)/2 (15.4) yes/no [n (%)] Antithymocyte globulin, 6 (46.1)/4 (30.8) 3 (23.1)/0 yes/no [n (%)] Grade of hematuria Mild 9 (69.2) 1 (7.7) Moderate 1 (7.7) 0 Severe 0 2 (15.4) Duration of symptoms; 3.8/1–12 18.1/4–44 mean/range (d)
P Value
Reduced intensity
Conditioning Agent
n
TBI ⫹ VP ⫾ ATG Bu ⫹ Cy ⫾ ATG Bu ⫹ VP ⫹ Cy ⫹ ATG Bu ⫹ Cy ⫹ Mel ⫹ ATG Treo ⫹ Flu ⫹ Mel ⫹ ATG Flu ⫹ Mel Treo ⫹ Cy VP ⫹ ATG Treo ⫹ Cy ⫹ Mel ⫾ ATG
4 2 1 1 1 1 1 1 1
Abbreviations: TBI, total body irradiation; VP, etoposide; ATG, antithymocyte globulin; Bu, busulphan; Cy, cyclophosphamide; Mel, melphalan; Flu, fludarabine; Treo, treosulfan.
AdV Infection Therapy In patients with identified AdV replication developing hematuria and a worsening clinical course, we instituted therapy with cidofovir; 5 mg/kg in association with adequate hydration and probenecid 2 hours before as well as 3 and 8 hours after the infusion. The therapy was continued at weekly intervals until we achieved clearance of AdV-DNA.
Extraction of Viral DNA DNA from urine samples and control DNA from viral lysates were extracted using the QIAamp Viral RNA Mini Kit and QIAamp Blood DNA Mini Kit (Qiagen, Hilden, Germany), respectively, according to the manufacturer’s instructions. RNA carrier was added to urine samples and 5 L internal control (IC) was inserted at the initial extraction stage. From a starting amount of 140 L urine or 200 L lysate, we collected 50 L extracted DNA of 6 L which was used for standard PCR and 7 L for real-time PCR (RQ-PCR) amplification.
.000
Standards and Calibration Curve Generation
.54
To quantify virus load, we generated 2 external standard curves. The standard curves were based on serial dilutions of fluorometrically quantified in-home cloned plasmid standard (ACF groups) and an amplicon quantitative standard (BDE). Plasmid and amplicon were purified by using Qiagen Plasmid Midi and QIAquick Gel Extraction Kits (Qiagen), respectively. Concentration and purity were determined with NanoPhotometer (Implen, München, Germany). Viral copy number was calculated based on the known molecular weights of the plasmid and the amplicon.
.013 .54
.54
PCR for Screening Testing .58 .41 .55
.000
.061
The initial examination of urine samples was performed with the qualitative PCR method. In brief, 10 mL urine was sedimented by centrifugation at 3,000 rpm for 10 minutes. PCR was performed as a singleplex assay in a total volume of 20 L, containing 10 L HotStarTaq Master Mix Kit (Qiagen), 1 L 0.5 mol/L forward and reverse primers,14 2 L distilled water, and 6 L target viral nucleic acid. The amplification was performed according to following protocol: 95°C for 15 minutes; 34 cycles of 91°C for 1.30 minutes, 58°C for 1.30 minutes, and 72°C for 2.30 minutes; followed by 72°C for 10 minutes. PCR products stained with ethidium bromide were visualized after electrophoresis in 3% agarose gel. Reference strains of adenoviruses obtained from the American Type Culture Collection (Manassas, Va, USA) were used as
ADENOVIRUS-ASSOCIATED HEMATURIA IN BONE MARROW RECIPIENTS positive control samples in each run with distilled water serving as a negative control. Each sample that tested positive was quantified by RQ-PCR to determine the virus load in the clinical sample.
Real-Time PCR RQ-PCR was elaborated in our laboratory, based on the previously described specific primers and probes.15 Quantitative amplification was carried out with Real Time 7500 PCR Systems (Applied Biosystems, Foster City, USA) in a 96-well format. In brief, the optimized master mix for ACF reaction consisted of 13 L TaqMan Universal PCR Master Mix (2⫻) (Applied Biosystems), 1 L each primer (160 nmol/L), except second reverse (320 nmol/L), 1 L probes (80 nmol/L), 2.6 L IC Master Mix, 0.52 L water, and 7 L target DNA solution in a final total volume of 32 L. BDE stock consisted of 12 L TaqMan Universal PCR Master Mix (2⫻), 1 L primers (160 nmol/L), 1 L probes (80 nmol/L), 2.4 L IC MasterMix, 0.48 L water, and 7 L extracted DNA. Reactions were performed under the following conditions: 1 cycle at 50°C for 2 minutes for AmpErase UNG activation, then 95°C for 10 minutes, 55 cycles of 95°C for 15 seconds, and 60°C for 1 minutes. A TaqMan Exogenous Internal Positive Control (Applied Biosystems) was coamplified in the multiplex reaction for inhibition detection. Sensitivity was demonstrated at 2.5 (ACF) and 24 (BDE) copies of viral genome per reaction. Evaluation of BKV infection was performed using the Alert Q-PCR BKV Kit (Nanogen, Torino, Italy).
Statistical Evaluation Statistica v. 8.0 (StatSoft, Kraków, Poland) was used for the statistical analysis. A univariate analysis of risk factors was performed using chi-square test or Fisher exact test. The KaplanMeier product-limit method and Cox-Mantel tests were used to evaluate patient survival. An age differential between groups was confirmed with the Mann-Whitney test. A P value of ⬍.05 was considered to be significant.
RESULTS
Among 84 patients, 13 developed hematuria after HSCT, yielding a cumulative incidence of 15.5%. Only 3 cases of hematuria were not caused by AdVs; most patients (10/13, 76.9%) developed hematuria due to AdV infection, as confirmed with the chi-square test (P ⫽ .0000). Most patients suffering from AdV-HC were male (7/10) or recipients of matched unrelated donors (7/10), but these differ-
3731
ences were not statistically significant (P ⫽ .54). HC was observed among groups transplanted for malignant (n ⫽ 7) or nonmalignant (n ⫽ 3) disorders. The type of conditioning regimen (myeloablative vs nonmyeloablative) did not affect HC occurrence. Most AdV-HC cases in this study were not life-threatening; 90% of them were classified as mild hematuria with only minor single or recurrent bleeding episodes. Only 1 patient developed moderate hematuria (P ⫽ .0000). We did not observe massive hematuria caused by adenoviruses. Mean duration from day of transplantation to the onset of hematuria was 41.2 days (range 9 –144 days) and the mean symptomatic period was 3.8 days (range 1–12 days). It occurred later and persisted shorter than hematuria caused by non-AdV agents: an average of 12.6 and 18.2 days, respectively. Because the P value (.06) was nearly significant, it might be assumed that HC of non-AdV etiology was linked to a worse clinical course. Characteristics of patients with AdV-induced hematuria are summarized in Table 3. We examined HC occurring in 2 time periods; in 7 patients, AdV-related HC appeared before 60 days after transplantation, and in only 3 patients after day 60. In contrast, all cases of non–AdV-related hematuria occurred only in the early posttransplantation period. The median age of patients with AdV-HC was 6 years (range 1–19 years). The Mann-Whitney test showed that hematuria due to AdV infection occurred more often in young than in adult recipients (P ⫽ .013), confirming this as a significant risk factor (Fig 1). Viral load in AdV-HC patients was similar to patients infected with AdV who did not develop HC (2.5 ⫻ 103 copies/mL and 3.4 ⫻ 103 copies/mL, respectively). Seven of 10 patients with AdV-HC were suffering from aGVHD grades I–III. All cases of non–AdV-HC were observed in patients with aGVHD grade IV (P ⫽ .000). We also observed a seasonal fluctuation in AdV-HC; there were higher incidences in winter, spring, and autumn (P ⫽ .026; Fig 2). Because of the leading role attributed to BK in HC pathogenesis, we analyzed rare cases of AdV/BKV coinfection among our study group. The development of HC was strongly related to AdV infection (P ⫽ .0004), but cases with BK/AdV coinfection were near the limit of significance (P ⫽ .057), which warrants further studies on this subject.
Table 3. Characteristics of Patients with AdV-Related Hematuria Patient No.
Age (y)
Gender
Donor Type
Underlying Disease
HC Grade
HC Onset (Days After HSCT)
Patient’s Survival
1 2 3 4 5 6 7 8 9 10
19 2 6 6 7 2 9 1 9 3
M F M M F M M M M M
Related Related MUD MUD MUD MUD MUD Related MUD MUD
AML M4 CML NHL-T WAS ALL-T ALL MDS SCID ALL ALL
Moderate Mild Mild Mild Mild Mild Mild Mild Mild Mild
9 144 16 17 38 11 69 18 69 21
No No Yes No Yes No Yes Yes Yes Yes
Abbreviations: M, male; F, female; MUD, matched unrelated donor; ALL, acute lymphoblastic leukemia; AML, acute myeloblastic leukemia; NHL, non-Hodgkin lymphoma; WAS, Wiskott-Aldrich syndrome; MDS, myelodysplastic syndrome; SCID, severe combined immunodeficiency.
3732
Fig 1. Median age in groups of patients with hemorrhagic cystitis. HC/AdV⫹, hemorrhagic cystitis due to adenoviral infection; HC/AdV⫺, hemorrhagic cystitis due to other causes. MannWhitney test: P ⫽ .013.
Patients with AdV infections seem to be much more predisposed to HC than those with other viral infections, such as rarely pathogenic BKV. Upon univariate analysis, donor type, gender, underlying disease, use of ex vivo T-cell depletion, and antithymocyte globulin treatment did not correlate with the incidence of hematuria due to AdV infection. Survival analysis showed that patients suffering from hemorrhagic cystitis had a lower probability of survival, particularly HC occurring before 200 days, which was associated with a greater risk of a fatal outcome (P ⫽ .002; Fig 3A). The etiology of hematuria (AdV infection vs other factors) was not significant for patient survival (P ⫽ .27; Fig 3B). That is, although hemorrhagic cystitis reduced the patient’s chance for survival, adenoviral infections did not affect survival after HSCT.
BIL-LULA, USSOWICZ, RYBKA ET AL
the high incidence of AdV infections among patients with HC suggested an important role of AdVs in the development of hemorrhagic cystitis. The clinical course of generalized adenoviral disease is generally believed to be more severe than that of BKV infection.1,9 It is more likely to be associated with nephropathy and multiple organ necrosis. Many earlier reports underlined the significant role of BKV infections in the development of HC.12 Arthur et al16 showed that ⬎50% of HC cases were associated with BKV presence in urine samples. Chen et al9 documented a correlation between clinical improvement and clearance of AdV after gancyclovir treatment, suggesting a greater role of AdV than BKV as an etiologic agent. Our study concentrated on the role of adenoviral infections and coinfection with BKV on the development of HC among graft recipients. We noticed that patients with AdV infections were much more predisposed to HC than those with other viral infections, including BKV. However, the P value of coinfection cases was nearly significant which suggests that coinfection with BKV increases the risk of HC but that our data need to be broadened. Despite the majority of patients (70%) having undergone transplantation from matched unrelated donors, they were not more prone to AdV-HC than other groups. The source of AdV infection is not unambiguous. In contrast to Akiyama et al5 and Kondo et al,11 we noticed that younger age was a risk factor for the incidence of AdV-related HC. AdV often stays dormant in the renal parenchyma until it is reactivated in immunocompromised patients. The possibility of latent AdV infection is greater in children than in adults. The reactivation of latent AdV rather than de novo infection can explain this phenomenon.7,18,11 However, we need to mention that we have also documented seasonal fluctuations of AdV-HC, in winter, spring, and autumn, suggesting that patients undergoing HSCT are also more prone to community infections at these times of the year, although this may reflect only a local phenomenon.
DISCUSSION
In this study, we have reported the incidence, clinical course, and risk factors for AdV-induced hematuria among 84 patients grafted in a local transplantation unit between 2007 and 2009. The overall moderate (15.5%) incidence of HC was similar to other reports.3,8 Many studies have reported BKV to be the predominant viral cause of HC,13,16,17 but in our trial almost 77% of patients presenting with hematuria were AdV positive, emphasizing its pathogenic role in HC development and the strong relation between AdV viruria and clinical symptoms. Among 3 patients with HC without AdV viruria, only 1 case had a viral pathogenesis of BKV documented; the two remaining cases seemed to result from chemotherapy. Moreover, we noticed that BKV viruria had a low predictive value for the development of HC after HSCT (unpublished data). Thus,
Fig 2. Correspondence analysis of seasonal factors and AdVrelated hematuria. HC⫹AdV⫹, patients with hematuria and AdV in urine; HC⫹AdV⫺, patients with hematuria without adenovirus (P ⫽ .026).
ADENOVIRUS-ASSOCIATED HEMATURIA IN BONE MARROW RECIPIENTS
3733
Fig 3. Survival analysis of (A) patients with and without hemorrhagic cystitis after HSCT (P ⫽ .002) and (B) patients with HC due to different etiologies (P ⫽ .27). Cox-Mantel test.
Most cases of AdV-HC in our study were short-lasting, resolving within 1–2 weeks on cidofovir therapy with immunosuppression tapering. However, there was no clear evidence that patients with virus-related HC responded successfully to specific antiviral therapy, although in some cases the treatment improved the clinical course.10,19,20 There was no significant difference between symptomatic and asymptomatic patients based upon AdV loads (2.5 ⫻ 103 copies/mL vs 3.4 ⫻ 103 copies/mL), which confirms that other agents participate in the pathogenesis of HC, as we have already mentioned. The results of the present study showed an important impact of HC on patient survival: There was a shorter survival among HC patients, but we also noticed that the etiology of HC was not of predictive value. In other words, HC must be treated effectively regardless of etiologic factor. In contrast to some studies3,9 and in agreement with others4, we were unable to confirm a relation between AdV-HC and risk factors such as donor type, gender, and underlying disease. HC risk associated with T-cell depletion and antithymocyte globulin as reported in previous studies,3 was not observed in the present analysis. Similarly to Hale et al,3 we did not find data to support the hypothesis that aGVHD was a risk factor for the development of AdVHC.8 However, we identified aGVHD stage IV as a risk factor for HC development from a non-AdV cause. This observation confirms earlier suggestions that patients with HC due to causes other than AdVs display worse clinical courses. The practical use of molecular methods to detect asymptomatic AdV infections was confirmed in this study. Their sensitivity and pathogen specificity were helpful to rapidly detect AdV and reduce the cost of diagnostics. We also report that microscopic hematuria may be an early warning of AdV infection in the urinary tract. Early detection of AdV-HC enables reduction of immunosuppression and anti-AdV treatment. Monitoring of AdV copy number is also helpful to evaluate infection severity and to evaluate the results of therapy.
REFERENCES 1. Ho Lim AK, Parsons S, Ierino F: Adenovirus tubulointerstitial nephritis presenting as a renal allograft space occupying lesion. Am J Transplant 5:2062, 2005 2. Mori K, Yoshihara T, Nishimura Y, et al: Acute renal failure to adenovirus-associated obstructive uropathy and necrotizing tubulointerstinal nephritis in a bone marrow transplant recipient. Bone Marrow Transplant 31:1173, 2003 3. Hale GA, Rochester RJ, Heslop HE, et al: Hemorrhagic cystitis after allogeneic bone marrow transplantation in children: clinical characteristics and outcome. Biol Blood Marrow Transplant 9:689, 2003 4. Childs R, Sanchez C, Engler H, et al: High incidence of adeno- and polyomavirus-induced hemorrhagic cystitis in bone marrow allotransplantation for hematological malignancy following T cell depletion and cyclosporine. Bone Marrow Transplant 22:889, 1998 5. Akiyama H, Kurosu T, Sakashita C, et al: Adenovirus is a key pathogen in hemorrhagic cystitis associated with bone marrow transplantation. Clin Infect Dis 32:1325, 2001 6. Teramura T, Naya M, Yoshihara T, et al: Adenoviral infection in hematopoietic stem cell transplantation: early diagnosis with quantitative detection of viral genome in serum and urine. Bone Marrow Transplant 33:87, 2004 7. Gorczyn ´ ska E, Turkiewicz D, Rybka K, et al: Incidence, clinical outcome, and management of virus-induced hemorrhagic cystitis in children and adolescents after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 11:797, 2005 8. Tsuboi K, Kishi K, Ohmachi K, et al: Multivariate analysis of risk factors for hemorrhagic cystitis after hematopoietic stem cell transplantation. Bone Marrow Transplant 32:903, 2003 9. Chen FE, Liang RHS, Lo JY, et al: Treatment of adenovirusassociated haemorrhagic cystitis with ganciclovir. Bone Marrow Transplant 20:997, 1997 10. Miyamura K, Hamaguchi M, Taji H, et al: Successful ribavirin therapy for severe adenovirus hemorrhagic cystitis after allogeneic marrow transplant from close HLA donors rather than distant donors. Bone Marrow Transplant 25:545, 2000 11. Kondo M, Kojima S, Kato, Matsuyama T: Late-onset hemorrhagic cystitis after hematopoietic stem cell transplantation in children. Bone Marrow Transplant 22:995, 1998 12. Baldwin A, Kingman H, Darville M, et al: Outcome and clinical course of 100 patients with adenovirus infection following bone marrow transplantation. Bone Marrow Transplant 26:1333, 2000 13. Giraud G, Priftakis P, Bogdanovic G, et al: BK-viruria and haemorrhagic cystitis are more frequent in allogeneic haematopoietic stem cell transplant patients receiving full conditioning and
3734 unrelated–HLA-mismatched grafts. Bone Marrow Transplant 41:737, 2008 14. McDonough M, Kew O, Hierholzer JC: PCR detection of human adenoviruses. In Persing DH, Smith TF, Tenover FC, White TJ (eds): Diagnostic molecular microbiology: principles and applications. Washington: American Society for Microbiology; p 389, 1993 15. Ebner K, Suda M, Watzinger F, Lion T: Molecular detection and quantitative analysis of the entire spectrum of human adenoviruses by two-reaction real-time PCR assay. J Clin Microbiol 43:3049, 2005 16. Arthur RR, Shah KV, Charache P, Saral R: BK and JC virus infections in recipients of bone marrow transplants. J Infect Dis 158:563, 1988
BIL-LULA, USSOWICZ, RYBKA ET AL 17. Giraud G, Bogdanovic G, Priftakis P, et al: The incidence of hemorrhagic cystitis and BK-viruria in allogeneic hematopoietic stem cell recipients according to intensity of the conditioning regimen. Stem Cell Transplant 91:401, 2006 18. de Mezerville MH, Tellier R, Richardson S, et al: Adenoviral infections in pediatric transplant recipients: a hospital-based study. Pediatr Infect Dis J 25:815, 2006 19. Keswani M, Moudgil A: Adenovirus-associated hemorrhagic cystitis in a pediatric renal transplant recipient. Pediatr Transplant 11:568, 2007 20. Hoffman JA, Shah AJ, Ross LA, Kapoor N: Adenoviral infections and a prospective trial of cidofovir in pediatric hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 7:388, 2001
I
(BKV) infection. Fever, urodynia, and hematuria may also accompany acute graft-versus-host disease (aGVHD) or bacterial infections arise as or a complication of the conditioning regimen.3,6 Many reports have described risk factors for the development of HC caused by BKV as the main pathogen of HC.5,13 The aim of the present study was to focus on the role of risk factors for AdV replication in hematuria development. Because of limitations in culture methodology for AdV detection, we used highly specific
From the Department of Clinical Chemistry (I.B.-L., M.W.) and Department of Paediatric Bone Marrow Transplantation, Oncology, and Haematology (M.U., B.R., R.R., E.G., K.K.), University of Medicine, Wrocław, Poland; Central University Hospital (D.W.-D.), Wrocław, Poland; and Department of Pharmacology (M.W.), University of Saskatchewan, Saskatoon, Canada. Sponsored by the Polish Ministry of Science and Higher Education (grant no. NN-401-004-536). Address reprint requests to Iwona Bil-Lula. Department of Clinical Chemistry, University of Medicine, Wrocław, Poland. E-mail: [email protected]
© 2010 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/–see front matter doi:10.1016/j.transproceed.2010.08.049
Transplantation Proceedings, 42, 3729 –3734 (2010)
3729
3730
BIL-LULA, USSOWICZ, RYBKA ET AL
sensitive molecular methods based on real-time polymerase chain reaction (PCR).
Table 2. Conditioning Regimens of HC Patients (n ⴝ 13) Type of Conditioning
Myeloablative
MATERIALS AND METHODS Study Group We retrospectively examined 84 mostly pediatric patients who underwent allogeneic hematopoietic stem cell transplantation between spring 2007 and summer 2009. To exclude cyclophosphamide-induced HC, we considered only patients who presented with hematuria in the late period after HSCT. Urine samples from all HSCT recipients were collected weekly during the hopitalization period and once monthly thereafter. The recipient characteristics are summarized in Table 1. Owing the high variability of treated patients, we used various conditioning regimens (Table 2). Ex vivo T-cell depletion of the graft was performed for all haploidentical transplantations (n ⫽ 9).
Definitions We defined patients as developing AdV-HC as the presence of hematuria accompanied by AdV viruria. To eliminate other causes of hematuria, we excluded samples from female patients suspected of vaginal bleeding and those with possible bacteruria. We included only de novo hematuria that was manifested ⬎7 days after HSCT. Urinalysis and microscopic techniques were performed as screening tests on a regular basis. Hematuria was defined as mild if 3–10 erythrocytes were observed in the field of vision, moderate for 11–50 erythrocytes, and severe for erythrocytes densely covering the field or if there was macroscopic hematuria. Table 1. Characteristics of HSCT Recipients Suffering from Hemorrhagic Cystitis (HC), Based on Occurrence of AdV Infection (n ⴝ 13) Characteristic/Variable
HC of Adenoviral Infection
HC of Other Causes
No. of patients, n (%) 10 (76.9) 3 (23.1) Gender, n (%) Male 7 (53.8) 2 (15.4) Female 3 (23.1) 1 (7.7) Age, median/range (y) 6/1–19 26/10–42 Diagnosis, n (%) Malignancy 7 (53.8) 2 (15.4) Nonmalignant disease 3 (23.1) 1 (7.7) Donor type, n (%) Matched unrelated donor 7 (53.8) 2 (15.4) Relative donor HLA-identical sibling 1 (7.7) 0 donor Haploidentical donor 2 (15.4) 1 (7.7) T-cell depletion, yes/no 2 (15.4)/8 (61.5) 1 (7.7)/2 (15.4) [n (%)] Mieloablative conditioning 8 (61.5)/2 (15.4) 1 (7.7)/2 (15.4) yes/no [n (%)] Antithymocyte globulin, 6 (46.1)/4 (30.8) 3 (23.1)/0 yes/no [n (%)] Grade of hematuria Mild 9 (69.2) 1 (7.7) Moderate 1 (7.7) 0 Severe 0 2 (15.4) Duration of symptoms; 3.8/1–12 18.1/4–44 mean/range (d)
P Value
Reduced intensity
Conditioning Agent
n
TBI ⫹ VP ⫾ ATG Bu ⫹ Cy ⫾ ATG Bu ⫹ VP ⫹ Cy ⫹ ATG Bu ⫹ Cy ⫹ Mel ⫹ ATG Treo ⫹ Flu ⫹ Mel ⫹ ATG Flu ⫹ Mel Treo ⫹ Cy VP ⫹ ATG Treo ⫹ Cy ⫹ Mel ⫾ ATG
4 2 1 1 1 1 1 1 1
Abbreviations: TBI, total body irradiation; VP, etoposide; ATG, antithymocyte globulin; Bu, busulphan; Cy, cyclophosphamide; Mel, melphalan; Flu, fludarabine; Treo, treosulfan.
AdV Infection Therapy In patients with identified AdV replication developing hematuria and a worsening clinical course, we instituted therapy with cidofovir; 5 mg/kg in association with adequate hydration and probenecid 2 hours before as well as 3 and 8 hours after the infusion. The therapy was continued at weekly intervals until we achieved clearance of AdV-DNA.
Extraction of Viral DNA DNA from urine samples and control DNA from viral lysates were extracted using the QIAamp Viral RNA Mini Kit and QIAamp Blood DNA Mini Kit (Qiagen, Hilden, Germany), respectively, according to the manufacturer’s instructions. RNA carrier was added to urine samples and 5 L internal control (IC) was inserted at the initial extraction stage. From a starting amount of 140 L urine or 200 L lysate, we collected 50 L extracted DNA of 6 L which was used for standard PCR and 7 L for real-time PCR (RQ-PCR) amplification.
.000
Standards and Calibration Curve Generation
.54
To quantify virus load, we generated 2 external standard curves. The standard curves were based on serial dilutions of fluorometrically quantified in-home cloned plasmid standard (ACF groups) and an amplicon quantitative standard (BDE). Plasmid and amplicon were purified by using Qiagen Plasmid Midi and QIAquick Gel Extraction Kits (Qiagen), respectively. Concentration and purity were determined with NanoPhotometer (Implen, München, Germany). Viral copy number was calculated based on the known molecular weights of the plasmid and the amplicon.
.013 .54
.54
PCR for Screening Testing .58 .41 .55
.000
.061
The initial examination of urine samples was performed with the qualitative PCR method. In brief, 10 mL urine was sedimented by centrifugation at 3,000 rpm for 10 minutes. PCR was performed as a singleplex assay in a total volume of 20 L, containing 10 L HotStarTaq Master Mix Kit (Qiagen), 1 L 0.5 mol/L forward and reverse primers,14 2 L distilled water, and 6 L target viral nucleic acid. The amplification was performed according to following protocol: 95°C for 15 minutes; 34 cycles of 91°C for 1.30 minutes, 58°C for 1.30 minutes, and 72°C for 2.30 minutes; followed by 72°C for 10 minutes. PCR products stained with ethidium bromide were visualized after electrophoresis in 3% agarose gel. Reference strains of adenoviruses obtained from the American Type Culture Collection (Manassas, Va, USA) were used as
ADENOVIRUS-ASSOCIATED HEMATURIA IN BONE MARROW RECIPIENTS positive control samples in each run with distilled water serving as a negative control. Each sample that tested positive was quantified by RQ-PCR to determine the virus load in the clinical sample.
Real-Time PCR RQ-PCR was elaborated in our laboratory, based on the previously described specific primers and probes.15 Quantitative amplification was carried out with Real Time 7500 PCR Systems (Applied Biosystems, Foster City, USA) in a 96-well format. In brief, the optimized master mix for ACF reaction consisted of 13 L TaqMan Universal PCR Master Mix (2⫻) (Applied Biosystems), 1 L each primer (160 nmol/L), except second reverse (320 nmol/L), 1 L probes (80 nmol/L), 2.6 L IC Master Mix, 0.52 L water, and 7 L target DNA solution in a final total volume of 32 L. BDE stock consisted of 12 L TaqMan Universal PCR Master Mix (2⫻), 1 L primers (160 nmol/L), 1 L probes (80 nmol/L), 2.4 L IC MasterMix, 0.48 L water, and 7 L extracted DNA. Reactions were performed under the following conditions: 1 cycle at 50°C for 2 minutes for AmpErase UNG activation, then 95°C for 10 minutes, 55 cycles of 95°C for 15 seconds, and 60°C for 1 minutes. A TaqMan Exogenous Internal Positive Control (Applied Biosystems) was coamplified in the multiplex reaction for inhibition detection. Sensitivity was demonstrated at 2.5 (ACF) and 24 (BDE) copies of viral genome per reaction. Evaluation of BKV infection was performed using the Alert Q-PCR BKV Kit (Nanogen, Torino, Italy).
Statistical Evaluation Statistica v. 8.0 (StatSoft, Kraków, Poland) was used for the statistical analysis. A univariate analysis of risk factors was performed using chi-square test or Fisher exact test. The KaplanMeier product-limit method and Cox-Mantel tests were used to evaluate patient survival. An age differential between groups was confirmed with the Mann-Whitney test. A P value of ⬍.05 was considered to be significant.
RESULTS
Among 84 patients, 13 developed hematuria after HSCT, yielding a cumulative incidence of 15.5%. Only 3 cases of hematuria were not caused by AdVs; most patients (10/13, 76.9%) developed hematuria due to AdV infection, as confirmed with the chi-square test (P ⫽ .0000). Most patients suffering from AdV-HC were male (7/10) or recipients of matched unrelated donors (7/10), but these differ-
3731
ences were not statistically significant (P ⫽ .54). HC was observed among groups transplanted for malignant (n ⫽ 7) or nonmalignant (n ⫽ 3) disorders. The type of conditioning regimen (myeloablative vs nonmyeloablative) did not affect HC occurrence. Most AdV-HC cases in this study were not life-threatening; 90% of them were classified as mild hematuria with only minor single or recurrent bleeding episodes. Only 1 patient developed moderate hematuria (P ⫽ .0000). We did not observe massive hematuria caused by adenoviruses. Mean duration from day of transplantation to the onset of hematuria was 41.2 days (range 9 –144 days) and the mean symptomatic period was 3.8 days (range 1–12 days). It occurred later and persisted shorter than hematuria caused by non-AdV agents: an average of 12.6 and 18.2 days, respectively. Because the P value (.06) was nearly significant, it might be assumed that HC of non-AdV etiology was linked to a worse clinical course. Characteristics of patients with AdV-induced hematuria are summarized in Table 3. We examined HC occurring in 2 time periods; in 7 patients, AdV-related HC appeared before 60 days after transplantation, and in only 3 patients after day 60. In contrast, all cases of non–AdV-related hematuria occurred only in the early posttransplantation period. The median age of patients with AdV-HC was 6 years (range 1–19 years). The Mann-Whitney test showed that hematuria due to AdV infection occurred more often in young than in adult recipients (P ⫽ .013), confirming this as a significant risk factor (Fig 1). Viral load in AdV-HC patients was similar to patients infected with AdV who did not develop HC (2.5 ⫻ 103 copies/mL and 3.4 ⫻ 103 copies/mL, respectively). Seven of 10 patients with AdV-HC were suffering from aGVHD grades I–III. All cases of non–AdV-HC were observed in patients with aGVHD grade IV (P ⫽ .000). We also observed a seasonal fluctuation in AdV-HC; there were higher incidences in winter, spring, and autumn (P ⫽ .026; Fig 2). Because of the leading role attributed to BK in HC pathogenesis, we analyzed rare cases of AdV/BKV coinfection among our study group. The development of HC was strongly related to AdV infection (P ⫽ .0004), but cases with BK/AdV coinfection were near the limit of significance (P ⫽ .057), which warrants further studies on this subject.
Table 3. Characteristics of Patients with AdV-Related Hematuria Patient No.
Age (y)
Gender
Donor Type
Underlying Disease
HC Grade
HC Onset (Days After HSCT)
Patient’s Survival
1 2 3 4 5 6 7 8 9 10
19 2 6 6 7 2 9 1 9 3
M F M M F M M M M M
Related Related MUD MUD MUD MUD MUD Related MUD MUD
AML M4 CML NHL-T WAS ALL-T ALL MDS SCID ALL ALL
Moderate Mild Mild Mild Mild Mild Mild Mild Mild Mild
9 144 16 17 38 11 69 18 69 21
No No Yes No Yes No Yes Yes Yes Yes
Abbreviations: M, male; F, female; MUD, matched unrelated donor; ALL, acute lymphoblastic leukemia; AML, acute myeloblastic leukemia; NHL, non-Hodgkin lymphoma; WAS, Wiskott-Aldrich syndrome; MDS, myelodysplastic syndrome; SCID, severe combined immunodeficiency.
3732
Fig 1. Median age in groups of patients with hemorrhagic cystitis. HC/AdV⫹, hemorrhagic cystitis due to adenoviral infection; HC/AdV⫺, hemorrhagic cystitis due to other causes. MannWhitney test: P ⫽ .013.
Patients with AdV infections seem to be much more predisposed to HC than those with other viral infections, such as rarely pathogenic BKV. Upon univariate analysis, donor type, gender, underlying disease, use of ex vivo T-cell depletion, and antithymocyte globulin treatment did not correlate with the incidence of hematuria due to AdV infection. Survival analysis showed that patients suffering from hemorrhagic cystitis had a lower probability of survival, particularly HC occurring before 200 days, which was associated with a greater risk of a fatal outcome (P ⫽ .002; Fig 3A). The etiology of hematuria (AdV infection vs other factors) was not significant for patient survival (P ⫽ .27; Fig 3B). That is, although hemorrhagic cystitis reduced the patient’s chance for survival, adenoviral infections did not affect survival after HSCT.
BIL-LULA, USSOWICZ, RYBKA ET AL
the high incidence of AdV infections among patients with HC suggested an important role of AdVs in the development of hemorrhagic cystitis. The clinical course of generalized adenoviral disease is generally believed to be more severe than that of BKV infection.1,9 It is more likely to be associated with nephropathy and multiple organ necrosis. Many earlier reports underlined the significant role of BKV infections in the development of HC.12 Arthur et al16 showed that ⬎50% of HC cases were associated with BKV presence in urine samples. Chen et al9 documented a correlation between clinical improvement and clearance of AdV after gancyclovir treatment, suggesting a greater role of AdV than BKV as an etiologic agent. Our study concentrated on the role of adenoviral infections and coinfection with BKV on the development of HC among graft recipients. We noticed that patients with AdV infections were much more predisposed to HC than those with other viral infections, including BKV. However, the P value of coinfection cases was nearly significant which suggests that coinfection with BKV increases the risk of HC but that our data need to be broadened. Despite the majority of patients (70%) having undergone transplantation from matched unrelated donors, they were not more prone to AdV-HC than other groups. The source of AdV infection is not unambiguous. In contrast to Akiyama et al5 and Kondo et al,11 we noticed that younger age was a risk factor for the incidence of AdV-related HC. AdV often stays dormant in the renal parenchyma until it is reactivated in immunocompromised patients. The possibility of latent AdV infection is greater in children than in adults. The reactivation of latent AdV rather than de novo infection can explain this phenomenon.7,18,11 However, we need to mention that we have also documented seasonal fluctuations of AdV-HC, in winter, spring, and autumn, suggesting that patients undergoing HSCT are also more prone to community infections at these times of the year, although this may reflect only a local phenomenon.
DISCUSSION
In this study, we have reported the incidence, clinical course, and risk factors for AdV-induced hematuria among 84 patients grafted in a local transplantation unit between 2007 and 2009. The overall moderate (15.5%) incidence of HC was similar to other reports.3,8 Many studies have reported BKV to be the predominant viral cause of HC,13,16,17 but in our trial almost 77% of patients presenting with hematuria were AdV positive, emphasizing its pathogenic role in HC development and the strong relation between AdV viruria and clinical symptoms. Among 3 patients with HC without AdV viruria, only 1 case had a viral pathogenesis of BKV documented; the two remaining cases seemed to result from chemotherapy. Moreover, we noticed that BKV viruria had a low predictive value for the development of HC after HSCT (unpublished data). Thus,
Fig 2. Correspondence analysis of seasonal factors and AdVrelated hematuria. HC⫹AdV⫹, patients with hematuria and AdV in urine; HC⫹AdV⫺, patients with hematuria without adenovirus (P ⫽ .026).
ADENOVIRUS-ASSOCIATED HEMATURIA IN BONE MARROW RECIPIENTS
3733
Fig 3. Survival analysis of (A) patients with and without hemorrhagic cystitis after HSCT (P ⫽ .002) and (B) patients with HC due to different etiologies (P ⫽ .27). Cox-Mantel test.
Most cases of AdV-HC in our study were short-lasting, resolving within 1–2 weeks on cidofovir therapy with immunosuppression tapering. However, there was no clear evidence that patients with virus-related HC responded successfully to specific antiviral therapy, although in some cases the treatment improved the clinical course.10,19,20 There was no significant difference between symptomatic and asymptomatic patients based upon AdV loads (2.5 ⫻ 103 copies/mL vs 3.4 ⫻ 103 copies/mL), which confirms that other agents participate in the pathogenesis of HC, as we have already mentioned. The results of the present study showed an important impact of HC on patient survival: There was a shorter survival among HC patients, but we also noticed that the etiology of HC was not of predictive value. In other words, HC must be treated effectively regardless of etiologic factor. In contrast to some studies3,9 and in agreement with others4, we were unable to confirm a relation between AdV-HC and risk factors such as donor type, gender, and underlying disease. HC risk associated with T-cell depletion and antithymocyte globulin as reported in previous studies,3 was not observed in the present analysis. Similarly to Hale et al,3 we did not find data to support the hypothesis that aGVHD was a risk factor for the development of AdVHC.8 However, we identified aGVHD stage IV as a risk factor for HC development from a non-AdV cause. This observation confirms earlier suggestions that patients with HC due to causes other than AdVs display worse clinical courses. The practical use of molecular methods to detect asymptomatic AdV infections was confirmed in this study. Their sensitivity and pathogen specificity were helpful to rapidly detect AdV and reduce the cost of diagnostics. We also report that microscopic hematuria may be an early warning of AdV infection in the urinary tract. Early detection of AdV-HC enables reduction of immunosuppression and anti-AdV treatment. Monitoring of AdV copy number is also helpful to evaluate infection severity and to evaluate the results of therapy.
REFERENCES 1. Ho Lim AK, Parsons S, Ierino F: Adenovirus tubulointerstitial nephritis presenting as a renal allograft space occupying lesion. Am J Transplant 5:2062, 2005 2. Mori K, Yoshihara T, Nishimura Y, et al: Acute renal failure to adenovirus-associated obstructive uropathy and necrotizing tubulointerstinal nephritis in a bone marrow transplant recipient. Bone Marrow Transplant 31:1173, 2003 3. Hale GA, Rochester RJ, Heslop HE, et al: Hemorrhagic cystitis after allogeneic bone marrow transplantation in children: clinical characteristics and outcome. Biol Blood Marrow Transplant 9:689, 2003 4. Childs R, Sanchez C, Engler H, et al: High incidence of adeno- and polyomavirus-induced hemorrhagic cystitis in bone marrow allotransplantation for hematological malignancy following T cell depletion and cyclosporine. Bone Marrow Transplant 22:889, 1998 5. Akiyama H, Kurosu T, Sakashita C, et al: Adenovirus is a key pathogen in hemorrhagic cystitis associated with bone marrow transplantation. Clin Infect Dis 32:1325, 2001 6. Teramura T, Naya M, Yoshihara T, et al: Adenoviral infection in hematopoietic stem cell transplantation: early diagnosis with quantitative detection of viral genome in serum and urine. Bone Marrow Transplant 33:87, 2004 7. Gorczyn ´ ska E, Turkiewicz D, Rybka K, et al: Incidence, clinical outcome, and management of virus-induced hemorrhagic cystitis in children and adolescents after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 11:797, 2005 8. Tsuboi K, Kishi K, Ohmachi K, et al: Multivariate analysis of risk factors for hemorrhagic cystitis after hematopoietic stem cell transplantation. Bone Marrow Transplant 32:903, 2003 9. Chen FE, Liang RHS, Lo JY, et al: Treatment of adenovirusassociated haemorrhagic cystitis with ganciclovir. Bone Marrow Transplant 20:997, 1997 10. Miyamura K, Hamaguchi M, Taji H, et al: Successful ribavirin therapy for severe adenovirus hemorrhagic cystitis after allogeneic marrow transplant from close HLA donors rather than distant donors. Bone Marrow Transplant 25:545, 2000 11. Kondo M, Kojima S, Kato, Matsuyama T: Late-onset hemorrhagic cystitis after hematopoietic stem cell transplantation in children. Bone Marrow Transplant 22:995, 1998 12. Baldwin A, Kingman H, Darville M, et al: Outcome and clinical course of 100 patients with adenovirus infection following bone marrow transplantation. Bone Marrow Transplant 26:1333, 2000 13. Giraud G, Priftakis P, Bogdanovic G, et al: BK-viruria and haemorrhagic cystitis are more frequent in allogeneic haematopoietic stem cell transplant patients receiving full conditioning and
3734 unrelated–HLA-mismatched grafts. Bone Marrow Transplant 41:737, 2008 14. McDonough M, Kew O, Hierholzer JC: PCR detection of human adenoviruses. In Persing DH, Smith TF, Tenover FC, White TJ (eds): Diagnostic molecular microbiology: principles and applications. Washington: American Society for Microbiology; p 389, 1993 15. Ebner K, Suda M, Watzinger F, Lion T: Molecular detection and quantitative analysis of the entire spectrum of human adenoviruses by two-reaction real-time PCR assay. J Clin Microbiol 43:3049, 2005 16. Arthur RR, Shah KV, Charache P, Saral R: BK and JC virus infections in recipients of bone marrow transplants. J Infect Dis 158:563, 1988
BIL-LULA, USSOWICZ, RYBKA ET AL 17. Giraud G, Bogdanovic G, Priftakis P, et al: The incidence of hemorrhagic cystitis and BK-viruria in allogeneic hematopoietic stem cell recipients according to intensity of the conditioning regimen. Stem Cell Transplant 91:401, 2006 18. de Mezerville MH, Tellier R, Richardson S, et al: Adenoviral infections in pediatric transplant recipients: a hospital-based study. Pediatr Infect Dis J 25:815, 2006 19. Keswani M, Moudgil A: Adenovirus-associated hemorrhagic cystitis in a pediatric renal transplant recipient. Pediatr Transplant 11:568, 2007 20. Hoffman JA, Shah AJ, Ross LA, Kapoor N: Adenoviral infections and a prospective trial of cidofovir in pediatric hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 7:388, 2001