Volume 13, Issue 11 , Pages 1369-1379, November 2007
Central Nervous System Complications after Allogeneic Hematopoietic Stem Cell Transplantation: Incidence, Manifestations, and Clinical Significance
Article Outline
Abstract
The incidence of central nervous system (CNS) complications, their risk factors, and impact on outcome are not well defined in recipients of allogeneic hematopoietic stem cell transplant (aHSCT). We reviewed the medical records of 302 consecutive patients who underwent aHSCT for malignant and nonmalignant hematologic diseases at Princess Margaret Hospital between 2002 and 2005. The cumulative incidences of all CNS complications and posterior reversible encephalopathy syndrome (PRES) at days 100 and 365 were 18% (95% confidence interval [CI]: 14-22) and 23% (95% CI: 19-29), and 6% (95% CI: 4-9) and 7% (95% CI: 5-11), respectively. Seizures occurred in 37% of all CNS events in the first 100 days and 73% of PRES episodes. Female gender and high-dose total-body irridiation (TBI) were identified as independent risk factors for CNS complications in the first 100 days posttransplant. Survival at 1-year was significantly inferior in patients who developed CNS complications within 100 days of transplant (28% [95% CI: 17-41] versus 72% [95% CI: 66-77]) and PRES (27% [95% CI: 10-47] versus 67% [95% CI: 62-73]) compared to those who did not (P < .0001). Death from graft-versus-host disease (GVHD) was more common in patients with CNS complications in the first 100 days (P = .006) and PRES (P = .01) compared to patients without complications.
Key Words: Allogeneic, Stem cell transplant, Neurologic, PRES, Survival, Risk factors
Introduction
The manifestations, etiology, and time course of central nervous system (CNS) complications, especially posterior reversible encephalopathy syndrome (PRES), are not well characterized in adult allogeneic hematopoietic stem cell transplant (aHSCT) recipients. aHSCT patients are at risk for CNS complications due to various regimen-related toxicities (RRT), cytopenias, drug toxicities, infections, and graft-versus-host disease (GVHD). Previous studies describe an incidence of neurologic complications of 8% to 42% in bone marrow transplant recipients [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]. This wide variation may be related to type of transplant, definition of neurologic complications, populations examined, preparatory regimens, duration of follow-up, and study designs. Prospective studies in this area are inconclusive due to small sample size [1, 10]. Additionally, these studies do not reflect current clinical practices such as increasing use of reduced intensity transplants, transplantation in older patients, and widespread use of peripheral blood stem cells.
In most clinical studies [1, 2, 3, 8], the most common causes of CNS complications are cyclosporine neurotoxicity or metabolic disturbances, whereas in neuropathologic studies [12, 13] cerebrovascular disorders are the primary diagnosis. Various studies have identified type of transplant, pretransplant disease risk status, acute myeloid malignancy, greater than grade II GVHD, cyclosporine, and total-body irradiation (TBI) as risk factors for neurologic complications [2, 3, 11]. No studies have analyzed clinical risk factors for PRES in this population. Moreover, previous studies in this area are difficult to interpret due to different definitions of this complication [14, 15, 16, 17, 18, 19]. No modifiable risk factors prior to stem cell transplantation have been examined with regard to CNS complications or PRES. Intensity or type of treatment prior to transplant may be an important factor for CNS complications given that some chemotherapies such as high-dose ara-C (HiDAC) and fludarabine are associated with CNS complications especially in older patients and those with renal impairment [20, 21, 22, 23, 24]. Moreover, the impact of CNS complications on posttransplant survival is unclear due to conflicting results [1, 2, 7].
We retrospectively analyzed data on 302 patients treated consecutively with aHSCT for malignant and nonmalignant hematologic disorders at Princess Margaret Hospital (PMH). The main objectives of the study were to understand the spectrum of CNS complications, especially PRES, in aHSCT recipients, and to identify patient, disease, treatment prior to aHSCT, and transplant-related risk factors for these complications.
Patients and Methods
Patients
We analyzed the clinical course of 302 consecutive patients who underwent aHSCT for malignant and nonmalignant hematologic disorders at PMH (University Health Network [UHN], Toronto, Ontario, Canada) between January 2002 and November 2005. These patients were identified from a prospective transplant database updated regularly for various transplant-related complications and events. Data on patients, donors, disease, treatment prior to transplant, and transplant-related factors were collected from the transplant database and computerized hospital records. The study was approved by the Cancer Registry Data Access Committee (CRDAC) and UHN Research Ethics Board.
Conditioning Regimens, GVHD Prophylaxis, and Supportive Care
Patients received conditioning therapy according to ongoing institutional protocols. The choice of intensity of conditioning regimen was based on the patient’s age, comorbidities, and underlying disease according to transplant program policies. The intensity of conditioning therapy was defined as conventional (myeloablative), reduced, or minimal. For hematologic malignancies, the conventional intensity conditioning (CIC) protocols were: cytarabine (ara-C) (500 mg/m2)/high-dose cyclophosphamide (120 mg/kg) and TBI (500 cGy in a single fraction) [25], high-dose cyclophosphamide (120-180 mg/kg) and high-dose TBI (1200 cGy in 6 fractions) [26], and i.v./p.o. busulphan (16 mg/kg p.o. or 12.8 mg/kg i.v.) and cyclophosphamide (120 mg/kg) [27]. The reduced-intensity conditioning (RIC) protocol included fludarabine (120 mg/m2) and i.v. busulphan (6.4 mg/kg) and the minimal intensity conditioning (MIC) protocol consisted of fludarabine (125 mg/m2) and low-dose TBI (200 cGy) [28]. For patients with aplastic anemia, CIC consisted of a high-dose cyclophosphamide-based protocol (200 mg/kg) and MIC consisted of fludarabine (120 mg/m2) and low-dose cyclophosphamide (40 mg/kg). GVHD prophylaxis for CIC and RIC/MIC protocols were with cyclosporine and a short course of methotrexate [26], and cyclosporine and mycophenolate [29], respectively. Cyclosporine trough levels were maintained between 200 and 400 ng/mL. Supportive care was provided as described previously [28].
CNS Complications
Data on all clinically significant CNS complications in the first year following aHSCT were collected using a case report form specifically designed for this study. The following complications were defined as clinically significant: seizures, loss of vision in the absence of local pathology, altered level of consciousness (ALOC), involuntary movements, ataxia, paresis, cranial nerve palsies, speech impairment, and delirium. The etiology of CNS complications were determined by clinical, microbiologic, and radiologic characteristics, and were broadly classified as focal white matter abnormalities or PRES, metabolic encephalopathy (renal failure, liver failure, or electrolyte imbalance), cerebrovascular (hemorrhage, ischemia/infarct), infectious (viral, fungal, bacterial, protozoal), thrombotic thrombocytopenic pupura/hemolytic uremic syndrome-like (TTP/HUS) and vasculitis. CNS events that occurred following relapsed disease were not included. Cyclosporine trough levels and serum magnesium levels at the onset of CNS complications were obtained from computerized hospital records.
Management of Patients with CNS Complications
CNS complications in the posttransplant period such as seizures, loss of vision, and ALOC were managed by a defined strategy of the Blood and Marrow Transplant Program involving discontinuation of cyclosporine and urgent brain imaging studies (MRI and/or CT). Other CNS complications were managed according to clinical needs. If imaging studies revealed evidence of focal white matter abnormalities or PRES, cyclosporine was discontinued. Serial imaging studies were conducted once per week until resolution of PRES changes. Upon resolution of these changes, patients were switched to FK-506. If the complications recurred, calcineurin inhibitors were discontinued and patients were given GVHD prophylaxis with mycophenolate mofetil (MMF) or at the discretion of the attending physician. Some patients underwent additional investigations such as lumbar puncture and/or electroencephalogram (EEG) as clinically indicated. The time from onset of CNS complication to first imaging study was <12 hours in the majority (>90%) of patients with CNS complications in the first 100 days post-aHSCT.
Imaging Studies
All neuroimaging studies conducted following CNS events were reviewed by 2 neuroradiologists blinded to clinical histories, outcomes, and original imaging reports. All patients had magnetic resonance (MR) scans of the brain on a 1.5 Tesla scanner (GE Medical Systems, Milwaukee, WI) using an 8-channel head coil. The study included FLAIR, gradient echo, and diffusion sequences in addition to the usual spin echo sequences. Gadolinium was given to most patients to exclude infectious complications. Noncontrast CT scans were done to exclude hemorrhage in 17 patients. Criteria for PRES included the presence on FLAIR images of confluent subcortical white matter lesions with little or no adjacent cortical involvement [30, 31]. The lesions were found to be occipital (unilateral or bilateral), with or without frontal involvement (unilateral or bilateral). Note was made of any brainstem and cerebellar involvement. The following arbitrary criteria were used to grade severity of PRES: (1) mild: size <2 cm, unilateral or bilateral occipital location (Figure 1A); (2) moderate: size >2 cm with bilateral occipital involvement (Figure 1B); (3) severe: bilateral occipital lesions with unilateral/bilateral frontal lesions, brainstem or cerebellar involvement, or any hemorrhage (Figure 1C and D).
Figure 1.
FLAIR axial images of brain show mild (A), moderate (B), and severe (C and D) forms of PRES corresponding with severity grading of 1, 2, and 3, respectively.
Pretransplant Treatment Intensity and Definitions of Posttransplant Events
We arbitrarily defined pretransplant treatment intensity as: (1) mild: single-agent chemotherapy <6 months, or minimally transfused (<20 transfusions); (2) moderate: single-agent chemotherapy ≥6 months, up to 2 cycles of intensive chemotherapy for acute leukemia (AL), 1-2 lines of chemotherapy for lymphoproliferative disorders (LPD), or moderate transfusions (20-50); (3) heavy: 3-4 cycles of chemotherapy for AL, 3-4 lines of chemotherapy for LPD, or heavily transfused (>50); and (4) very heavy: ≥5 cycles of chemotherapy for AL, ≥5 lines of chemotherapy for LPD, or previous autologous or allogeneic transplant.
HiDAC was defined as ≥9 g/m2 of ara-C in 1 chemotherapy cycle. The majority of younger AML patients (<60 years) received HiDAC according to the Cancer and Leukemia Group B (CALGB) schedule [24]. Older AML patients received HiDAC as previously described [28]. Patients with relapsed or refractory acute myelogenous leukemia (AML) received HiDAC as part of the NOVE-HiDAC protocol [28]. Engraftment, acute and chronic GVHD (aGVHD, cGVHD), and severity of aGVDH and cGVHD were defined as previously described [28]. Overall survival was calculated from the date of transplant to death or last follow-up.
Statistical Analysis
Results were reported as proportions and 95% CI. The cumulative incidence of CNS complications and PRES was estimated with the Cmprsk package in the R software (The R Foundation, Vienna, Austria). Death from any cause and relapse were treated as competing risks [32]. The Kaplan-Meier method was used to estimate survival [33]. Multivariate analysis was used to explore risk factors for CNS complications within the first 100 days based on the logistic regression method. The likelihood ratio test was used to choose the multivariate model. Overall goodness of fit of the multivariate logistic regression model was assessed using the Hosmer and Lemeshow test. All analyses were performed with SAS 9.1 (SAS Institute, Cary, NC). A P-value <.05 was considered significant.
Results
Patient Characteristics
Patient, donor, disease, and treatment-related characteristics are shown in Table 1. Significant neurologic history prior to transplant was identified in 21 patients (7%) as history of seizures (n = 5), CVA/TIA (n = 4), and “other,” including CNS involvement by primary underlying disease (n = 12).
Table 1. Patient, Disease, and Treatment-Related Characteristics
| Characteristics | Number of Patients (n = 302) N (%) |
|---|---|
| Median age of recipients (range) | 46 (18-71) |
| Median age of donors (range) | 46(11-76) |
| Proportion of male patients (%) | 59 |
| Diagnosis | |
| AML/MDS | 125(42) |
| CML | 39(13) |
| ALL | 37(12) |
| NHL/CLL | 53(17) |
| AA | 15(5) |
| MF | 16(5) |
| Others | 17(6) |
| CNS involvement of primary disease | |
| Yes | 12(4) |
| No | 290(96) |
| Risk of disease | |
| Standard risk | 87(29) |
| High risk | 215(71) |
| Disease status at transplant | |
| CR1/CR2/CP1 | 158(53) |
| Chemosensitive LPD | 62(20) |
| CP2/active/accelerated | 32(11) |
| Untreated/other | 50(16) |
| HiDAC prior to transplant | |
| None | 176(58) |
| 1 course | 29(10) |
| 2 courses | 46(15) |
| ≥3 courses | 45(15) |
| Missing | 6(2) |
| Fludarabine anytime prior to transplant conditioning | |
| Yes | 41(14) |
| No | 256(85) |
| Missing | 5(1) |
| Cranial irradiation prior to transplant | |
| Yes | 18(6) |
| No | 282(93) |
| Missing | 2(1) |
| Conditioning regimen | |
| Conventional intensity | 243(81) |
| Reduced intensity | 22(7) |
| Minimal intensity | 37(12) |
| TBI in conditioning regimen | |
| None | 89(20) |
| Low dose (200-300 cGy) | 40(13) |
| Moderate dose (500 cGy) | 48(16) |
| High dose (1200 cGy) | 125(41) |
| GVHD prophylaxis | |
| CSA + MTX | 234(77) |
| CSA + MMF | 59(20) |
| Other | 9(3) |
| Type of graft | |
| Bone marrow | 89(29) |
| Peripheral blood | 213(71) |
| HLA match | |
| Matched family donor | 224(74) |
| Mismatched family donor | 26(9) |
| Matched unrelated donor | 52(17) |
Incidence and Manifestations of All CNS Complications and PRES
All CNS complicationsThe cumulative incidence of CNS complications at days 30, 100, 180, and 365 was 9% (95% CI: 7-13), 18% (95% CI: 14-22), 20% (95% CI: 16-25), and 23% (95% CI: 19-29), respectively (Figure 2). Within 100 days posttransplant, 52 patients developed 57 CNS events. Of the 250 patients surviving 100 days, 22 developed 25 CNS events.
Figure 2.
The cumulative incidences of all CNS complications and PRES at days 30, 100, 180, and 365 were 9%, 18%, 20%, and 23%, and 4%, 6%, 7%, and 7%, respectively.
The etiology of CNS events is described in Table 2. Within 100 days of transplant, manifestations were seizures (n = 21), visual disturbance (n = 14), ALOC (n = 7), delirium (n = 18), and other (n = 7). From 101 days onward, manifestations were seizures (n = 6), visual disturbance (n = 7), ALOC (n = 2), delirium (n = 17), and other (n = 8).
Table 2. Etiology of CNS Complications
| Etiology | Number of CNS Events 0-100 Days Posttransplant (n = 57) N (%) | Number of CNS Events 101 Days-1 Year Posttransplant (n = 25) N (%) |
|---|---|---|
| Focal white matter abnormalities | 29(51) | 6(24) |
| PRES | 18 | 4 |
| Other | 11 | 2 |
| Metabolic | 8(14) | 2(8) |
| Hepatic | 7 | 2 |
| Renal | 1 | 0 |
| Cerebrovascular | 1(2) | 0 |
| Infectious | 2(4) | 4(16) |
| Bacterial | 0 | 1 |
| Viral | 0 | 2 |
| Fungal | 2 | 0 |
| Brain abscess of unknown origin | 0 | 1 |
| TTP-like | 1(2) | 1(4) |
| Other | 5(9) | 9(36) |
| Undetermined | 11(19) | 3(12) |
There were a total of 22 episodes of PRES in 21 patients (1 patient had 2 episodes of PRES). The cumulative incidence of PRES at 30, 100, 180, and 365 days was 4% (95% CI: 2-7), 6% (95% CI: 4-9), 7% (95% CI: 4-10), and 7% (95% CI: 5-11), respectively (Figure 2). The median number of days to onset of PRES was 30 (range: 8-226) with 82% (n = 18) of cases occurring within 100 days, and 18% (n = 4) occurring between 101 and 365 days posttransplant.
The presenting manifestations of PRES were seizures (n = 16), visual disturbance (n = 6), delirium (n = 3), and ALOC (n = 2). Imaging findings associated with PRES were most apparent on the FLAIR sequence. Findings included diffusion abnormality (n = 1), confluent white matter subcortical FLAIR abnormality (n = 20), infarct (n = 1), and hemorrhage (n = 1). PRES severity was classified as mild (n = 5), moderate (n = 8), and severe (n = 9). Gadolinium was not useful in the diagnosis of PRES. None of the FLAIR lesions were enhanced with contrast.
Out of 5 patients with a history of seizure prior to transplant, 2 developed seizures post-aHSCT. One patient had PRES according to the described criteria and the other patient had other focal white matter abnormalities. Seizures due to PRES occurred in 1 patient with a history of a syncopal episode prior to transplant. Of 12 patients with CNS involvement of primary disease prior to transplant, 17% (n = 2) experienced CNS complications. One patient developed PRES and the other patient had delirium of undetermined origin.
Risk Factors
Risk factors for CNS complications in the first 100 days were analyzed by univariate analysis as shown in Table 3. Significant factors associated with development of CNS complications included imatinib anytime prior to transplant (P = .02), no fludarabine prior to transplant conditioning (P = .04), CIC regimen (P = .006), high-dose TBI in conditioning regimen (P = 003), and cyclosporine plus methotrexate as GVHD prophylaxis (P = .02). No-risk factors were identified for the development of CNS complications after 100 days posttransplant (data not shown). Univariate analysis did not identify any significant risk factors associated with the development of PRES (data not shown). Multivariate regression analysis identified female gender (P = .03) and high-dose TBI (P = .0008) as independent risk factors for CNS complications within the first 100 days posttransplant (Table 4).
Table 3. Risk Factors for Development of CNS Complications within 100 Days Posttransplant
| Characteristics | Patients with CNS Complications (n = 52) | Patients without CNS Complications (n = 250) | P-Value |
|---|---|---|---|
| Median age of recipients (range) | 46 (20-67) | 45 (18-71) | .94 |
| Age of recipients (years) | .42 | ||
| <40 | 22 | 83 | |
| 40-59 | 26 | 140 | |
| ≥60 | 4 | 27 | |
| Median age of donors (range) | 47 (12-76) | 45 (11-73) | .29 |
| Gender | .08 | ||
| Male | 25 | 153 | |
| Female | 27 | 97 | |
| Diagnosis | .20 | ||
| AML/MDS | 17 | 108 | |
| CML | 8 | 31 | |
| ALL | 8 | 29 | |
| NHL/CLL | 5 | 48 | |
| AA | 3 | 12 | |
| MF | 5 | 11 | |
| Others | 6 | 11 | |
| CNS involvement of primary disease | .96 | ||
| Yes | 2 | 10 | |
| No | 50 | 240 | |
| Risk of disease | .51 | ||
| Standard | 13 | 74 | |
| High | 39 | 176 | |
| Treatment intensity prior to transplant | .76 | ||
| None | 3 | 21 | |
| Minimal | 9 | 31 | |
| Moderate | 15 | 61 | |
| Heavy | 15 | 80 | |
| Very heavy | 10 | 57 | |
| HiDAC prior to transplant | .53 | ||
| None | 31 | 145 | |
| 1 course | 6 | 23 | |
| 2 course | 5 | 41 | |
| ≥3 courses | 10 | 35 | |
| Missing | 0 | 6 | |
| Fludarabine anytime prior to transplant conditioning | .04 | ||
| Yes | 2 | 39 | |
| No | 50 | 206 | |
| Missing | 0 | 5 | |
| Imatinib anytime prior to transplant | .02 | ||
| Yes⁎ | 12 | 28 | |
| No | 40 | 222 | |
| Cranial irradiation prior to transplant | .39 | ||
| Yes | 5 | 13 | |
| No | 47 | 235 | |
| Missing | 0 | 2 | |
| Conditioning regimen | .006 | ||
| Conventional intensity | 49 | 194 | |
| Reduced + minimal intensity | 3 | 56 | |
| TBI in conditioning regimen | .003 | ||
| None | 7 | 82 | |
| Low dose (200-300 cGy) | 4 | 36 | |
| Moderate dose (500 cGy) | 8 | 40 | |
| High dose (1200 cGy) | 33 | 92 | |
| GVHD prophylaxis | .02 | ||
| CSA + MTX | 48 | 186 | |
| CSA + MMF | 3 | 56 | |
| All others | 1 | 8 | |
| Type of graft | .12 | ||
| Peripheral blood | 32 | 181 | |
| Bone marrow | 20 | 69 | |
| HLA match | .28 | ||
| Identical family donor | 34 | 190 | |
| Mismatched family donor | 6 | 20 | |
| Matched unrelated donor | 12 | 40 | |
| CVA/TIA prior to transplant | .36 | ||
| Yes | 0 | 4 | |
| No | 52 | 579 | |
| Seizures prior to transplant | .17 | ||
| Yes | 2 | 3 | |
| No | 50 | 247 |
⁎CML (n = 23), Ph + ALL (n = 9), AML (n = 6), Ph + biphenotypic acute leukemia (n = 1), and CMML (n = 1). The AML patients received imatinib in combination with NOVE-HiDAC as part of a Phase 1 trial for relapsed/refractory AML. |
Table 4. Multivariate Regression Analysis for CNS Complications within 100 Days Posttransplant
| Variable | Comparison | Odds Ratio | 95% CI | P-Value |
|---|---|---|---|---|
| Gender | Female versus male | 2.04 | 1.08-3.82 | .03 |
| TBI in conditioning regimen | None versus high dose (1200 cGy) | 0.22 | 0.09-0.52 | .0008 |
| Low/mod (200-500 cGy) versus high dose (1200 cGy) | 0.38 | 0.18-0.81 |
At the onset of CNS events, 77% (n = 17) of patients with PRES and 75% (n = 45) of patients with other CNS complications were receiving cyclosporine. Nine percent (n = 2) of patients with PRES and 13% (n = 8) of patients with other CNS complications were receiving tacrolimus. Information on drug type and blood levels was unavailable for 2 patients with PRES and 5 patients with other CNS complications. One patient with PRES and 2 patients with other CNS complications were receiving neither cyclosporine nor tacrolimus at complication onset. Mean cyclosporine levels (±SD) were 299 ± 116 ng/mL and 243 ± 170 ng/mL at the onset of PRES and other CNS complications, respectively (P = .25). Twenty-three percent (n = 4) of patients with PRES and 9% (n = 4) of patients with other CNS complications had calcineurin inhibitor levels above the target therapeutic range. At the onset of CNS complications, hypomagnasemia (<0.6 mmol/L) was found in 14% of patients with PRES and 9% (n = 5) of patients with other CNS complications.
Survival
Survival at 1 year was significantly reduced in patients who developed CNS complications within 100 days of transplant compared to those who did not (28% [95% CI: 17-41] versus 72% [95% CI: 66-77], P < .0001) (Figure 3A). Of the 52 patients experiencing CNS complications in the first 100 days, 75% (n = 39) died. The causes of death were GVHD (n = 21), underlying disease (n = 5), infections (n = 3), RRT (n = 6), and others (n = 4). Ninety-one of 246 patients without CNS complications died of GVHD (n = 23), underlying disease (n = 34), infections (n = 12), RRT (n = 8), and others (n = 14). A significantly larger proportion of patients with CNS complications in the first 100 days posttransplant died of GVHD when compared to patients without CNS complications (P = .006).
Figure 3.
Kaplan-Meier plots of overall survival in patients with and without CNS complications (within 100 days posttransplant) and PRES. In comparison to their counterparts, survival at 1-year was significantly reduced in patients who developed CNS complications in the first 100 days posttransplant (28% versus 72%, P < .0001) (A), and PRES (27% versus 67%, P < .0001) (B).
There was a significantly inferior 1-year survival in patients with compared to those without PRES (27% [95% CI: 10-47] versus 67% [95% CI: 62-73], P < .0001) (Figure 3B). Causes of death were GVHD (n = 11), underlying disease (n = 1), RRT (n = 1), and others (n = 3). Among the 281 patients who did not develop PRES, 114 died. Causes of death were underlying disease (n = 38), GVHD (n = 33), infections (n = 15), RRT (n = 13), and others (n = 15). A significantly larger proportion of patients with PRES died as a result of GVHD than those without PRES (P = .01). Median time to death in PRES patients was 41 days (range: 3-377 days). For patients with mild, moderate, and severe PRES, the mean number of days to death (±SD) was 213 ± 191, 82 ± 97, and 68 ± 70, respectively.
Discussion
The cumulative incidence of CNS complications and PRES at days 100 and 365 was 18% and 23%, and 6% and 7%, respectively, indicating they represent a significant clinical problem following aHSCT. The risk was higher during the first 100 days posttransplant when 69% of CNS complications and 86% of PRES episodes occurred. Previous studies report a wide variation of incidences of CNS complications in aHSCT recipients [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]. Some of these studies are difficult to compare due to methodologic issues. Our results are similar to a prospective study describing the 90-day incidence of major CNS complications as 25% [1]. PRES has been reported in 7.2% and 5.5% of aHSCT patients receiving calcineurin inhibitors for GVHD prophylaxis [15, 18]. However, only the crude incidence of PRES was determined and competing risks were not considered. Some studies have used cyclosporine toxicity and CNS events as synonyms. Our results are consistent with studies describing the incidence of severe cyclosporine neurotoxicity ranging from 4.2% to 9% [9, 17, 19]. A previous retrospective study from the M.D. Anderson Hospital described the crude incidence of PRES as 1.6% in aHSCT recipients receiving tacrolimus following aHSCT [34]. There are no comparative data available on the impact of cyclosporine on CNS complications when compared to tacrolimus. The lower incidence of PRES in tacrolimus-treated aHSCT recipients raises a possibility that there may be differences between cyclosporine and tacrolimus with regards to PRES development.
Female gender and high-dose TBI were identified as independent risk factors for CNS complications within the first 100 days posttransplant. Gender has not been previously identified as a risk factor for CNS complications and we are unable to explain this finding. A previous study in pediatric patients also identified high-dose TBI as a risk factor for CNS complications [3]. In addition, TBI-containing regimens have been associated with a higher rate of cyclosporine neurotoxicity in adult aHSCT recipients [18]. No difference in CNS complications was observed between patients treated with reduced or MIC conditioning and those treated with chemotherapy or low-dose TBI-based myeloblative regimens. We did not identify any risk factors related to PRES development. Inadequate power may have affected the ability to detect statistically significant risk factors. Further studies with a larger number of patients with PRES are needed to identify risk factors.
There does not appear to be a correlation between supratherapeutic calcineurin inhibitor levels and CNS events or PRES as only 23% of patients with PRES and 9% of patients with other CNS complications had calcineurin inhibitor trough levels higher than the therapeutic range at the onset of complications. This is consistent with the lack of direct correlation between cyclosporine levels and seizures/cyclosporine neurotoxicity described previously [17, 35]. However, it is possible that differences in neurotoxicity may be related to genetic differences in drug metabolism and needs further exploration in future studies [36]. Hypomagnesemia has been suggested as a risk factor for cyclosporine toxicity [37]. However, in our study, it was present in only a small number of patients experiencing PRES or other CNS events.
No modifiable risk factors for CNS complications or PRES prior to AHSCT were identified in this study. It is noteworthy that of 5 patients with a history of seizures prior to transplant, 2 developed seizures. Although this is a limited number of patients, it may be reasonable to avoid high-dose TBI in such patients.
An important finding of this study was significantly reduced survival of patients with CNS complications and PRES compared to those without [28% versus 72% (P < .0001) and 27% versus 67% (P < .0001), respectively]. Reduced 90-day survival has been shown in aHSCT patients who develop neurologic complications compared with their counterparts [1]. A study of pediatric stem cell transplant recipients, however, showed no difference in overall survival between patients with and without neurologic complications [7]. Both autologous and allogeneic transplant recipients were included in this analysis. A lack of association between neurologic complications and treatment-related mortality has also been described in adult aHSCT recipients [2]. However, survival data and analysis were not presented in this report.
A significantly larger number of patients with CNS complications in the first 100 days and PRES died as a result of GVHD compared to patients without these complications. GVHD has been shown to represent the primary cause of death following calcineurin inhibitor-induced neurotoxicity, but survival was not compared to patients without neurotoxicity [15]. Significant mortality due to GVHD may be related to discontinuation of calcineurin inhibitors resulting in poor prevention of GVHD. Novel strategies are needed for GVHD prevention in patients whose posttransplant course is complicated by CNS events. Increased severity of PRES appears to be related to a shorter interval between PRES onset and death. However, these observations are derived from a small number of patients, and should be interpreted with caution. Further studies are needed to clarify the relationship between severity of PRES and clinical outcomes.
In conclusion, CNS complications are a significant clinical problem in the posttransplant period in aHSCT patients and are associated with increased risk of death primarily due to GVHD. MR imaging can provide early diagnosis and follow-up to monitor treatment changes. Novel strategies are needed in these patients for effective control of GVHD.
Acknowledgments
D.S. received a Leukemia Research Summer Studentship from The Leukemia and Lymphoma Society of Canada and a Trainee Research Award from The American Society of Hematology in support of this project.
References
- Early neurologic complications following allogeneic bone marrow transplant for leukemia: a prospective study. Neurology. 1998;50:1441–1445
- . Increased incidence of neurological complications in patients receiving an allogenic bone marrow transplantation from alternative donors. J Neurol Neurosurg Psychiatry. 2000;68:36–40
- Severe neurologic complications after hematopoietic stem cell transplantation in children. Neurology. 2002;59:1895–1904
- Neurologic complications after allogeneic bone marrow transplantation. Bone Marrow Transplant. 1996;18:1135–1139
- Neurologic complications of autologous and allogeneic bone marrow transplantation in patients with leukemia: a comparative study. Neurology. 1996;46:1004–1009
- Neurological complications after autologous stem cell transplantation. Eur Neurol. 1999;41:48–50
- Neurological complications after stem cell transplantation in childhood. Bone Marrow Transplant. 1999;24:647–652
- Early central nervous system complications after reduced-intensity stem cell transplantation. Biol Blood Marrow Transplant. 2004;10:561–568
- Neurologic complications in allogeneic bone marrow transplant patients receiving cyclosporin. Bone Marrow Transplant. 1991;8:393–401
- . Prospective evaluation of neurological complications after allogeneic bone marrow transplantation. Neurology. 2003;60:842–848
- Life-threatening neurological complications after bone marrow transplantation in children. Bone Marrow Transplant. 2005;35:71–76
- . Neuropathologic findings after bone marrow transplantation: an autopsy study. Hum Pathol. 1990;21:630–639
- Neuropathological findings after bone marrow transplantation: an autopsy study of 180 cases. Bone Marrow Transplant. 2000;25:301–307
- . Variable incidence of cyclosporine and FK-506 neurotoxicity in hematopoeitic malignancies and marrow conditions after allogeneic bone marrow transplantation. Neurocrit Care. 2005;3:33–45
- Long-term outcomes of allogeneic stem cell transplant recipients after calcineurin inhibitor-induced neurotoxicity. Br J Haematol. 2003;123:110–113
- . Etiology of cortical and white matter lesions in cyclosporin-A and FK-506 neurotoxicity. AJNR Am J Neuroradiol. 2001;22:1901–1914
- . Clinical findings and magnetic resonance imaging in severe cyclosporine-related neurotoxicity after allogeneic bone marrow transplantation. Eur J Haematol. 2001;67:94–99
- . Pretransplantation conditioning influence on the occurrence of cyclosporine or FK-506 neurotoxicity in allogeneic bone marrow transplantation. AJNR Am J Neuroradiol. 2004;25:261–269
- . The effect of human leukocyte antigen disparity on cyclosporine neurotoxicity after allogeneic bone marrow transplantation. AJNR Am J Neuroradiol. 1998;19:601–660[discussion 9-10]
- . Reversible neurologic toxicity in patients treated with standard-dose fludarabine phosphate for mycosis fungoides and chronic lymphocytic leukemia. Ann Intern Med. 1993;118:114–116
- . Fludarabine phosphate (NSC 312878) infusions for the treatment of acute leukemia: phase I and neuropathological study. Cancer Res. 1986;46:5953–5958
- . Progressive multifocal leukoencephalopathy after fludarabine therapy for low-grade lymphoproliferative disease. Am J Hematol. 2002;70:51–54
- . Phase I and II study of fludarabine phosphate in leukemia: therapeutic efficacy with delayed central nervous system toxicity. J Clin Oncol. 1986;4:74–79
- Intensive postremission chemotherapy in adults with acute myeloid leukemia (Cancer and Leukemia Group B). N Engl J Med. 1994;331:896–903
- Long-term results of bone marrow transplantation for patients with AML, ALL and CML prepared with single dose total body irradiation of 500 cGy delivered with a high dose rate. Bone Marrow Transplant. 1991;8:453–463
- The role of allogeneic bone marrow transplantation in adult patients below the age of 55 years with acute lymphoblastic leukemia in first complete remission: a donor vs no donor comparison. Bone Marrow Transplant. 2004;33:397–404
- Busulfan and cyclophosphamide as a preparative regimen for allogeneic blood and marrow transplantation in patients with non-Hodgkin’s lymphoma. Bone Marrow Transplant. 2003;31:73–78
- Disease biology rather than age is the most important determinant of survival of patients > or = 60 years with acute myeloid leukemia treated with uniform intensive therapy. Cancer. 2005;103:2082–2090
- Nonmyeloablative stem cell transplantation for myelodysplastic syndrome or acute myeloid leukemia in patients 60 years or older. Biol Blood Marrow Transplant. 2005;11:764–772
- . Posterior reversible encephalopathy syndrome: utility of fluid-attenuated inversion recovery MR imaging in the detection of cortical and subcortical lesions. AJNR Am J Neuroradiol. 2000;21:1199–1206
- A reversible posterior leukoencephalopathy syndrome. N Engl J Med. 1996;334:494–500
- . Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med. 1999;18:695–706
- . Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457–481
- Tacrolimus-associated posterior reversible encephalopathy syndrome after allogeneic haematopoietic stem cell transplantation. Br J Haematol. 2003;122:128–134
- CsA-associated neurotoxicity and ineffective prophylaxis with clonazepam in patients transplanted for thalassemia major: analysis of risk factors. Bone Marrow Transplant. 1996;18:157–162
- . Cyclosporin toxicity at therapeutic blood levels and cytochrome P-450 IIIA. Lancet. 1990;335:11–15
- . Association between cyclosporin neurotoxicity and hypomagnesaemia. Lancet. 1984;2:1116–1120
PII: S1083-8791(07)00370-9
doi:10.1016/j.bbmt.2007.07.013
© 2007 American Society for Blood and Marrow Transplantation. Published by Elsevier Inc. All rights reserved.
Volume 13, Issue 11 , Pages 1369-1379, November 2007
