Volume 14, Issue 5 , Pages 510-517, May 2008
Preengraftment Serum C-Reactive Protein (CRP) Value May Predict Acute Graft-versus-Host Disease and Nonrelapse Mortality after Allogeneic Hematopoietic Stem Cell Transplantation
Article Outline
Abstract
In a mouse model, inflammatory cytokines play a primary role in the development of acute graft-versus-host disease (aGVHD). Here, we retrospectively evaluated whether the preengraftment C-reactive protein (CRP) value, which is used as a surrogate marker of inflammation, could predict posttransplant complications including GVHD. Two hundred twenty-four adult patients (median age, 47 years; range: 18-68 years) underwent conventional stem cell transplantation (CST, n = 105) or reduced-intensity stem cell transplantation (RIST, n = 119). Patients were categorized according to the maximum CRP value during neutropenia: the “low-CRP” group (CRP < 15 mg/dL, n = 157) and the “high-CRP” group (CRP ≥ 15 mg/dL, n = 67). The incidence of documented infections during neutropenia was higher in the high-CRP group (34% versus 17%, P = .004). When patients with proven infections were excluded, the CRP value was significantly lower after RIST than after CST (P = .017) or after related than after unrelated transplantation (P < .001). A multivariate analysis showed that male sex, unrelated donor, and HLA-mismatched donor were associated with high CRP values. The high-CRP group developed significantly more grade II-IV aGVHD (P = .01) and nonrelapse mortality (NRM) (P < .001), but less relapse (P = .02). The present findings suggest that the CRP value may reflect the net degree of tissue damage because of the conditioning regimen, infection, and allogeneic immune reactions, all of which lead to subsequent aGVHD and NRM.
Key Words: C-reactive protein, Allogeneic transplantation, Acute graft-versus-host disease, Nonrelapse mortality
Introduction
Allogeneic hematopoietic stem cell transplantation (HSCT) is associated with high treatment-related mortality (TRM) because of acute graft-versus-host disease (aGVHD) and infections 1, 2. Inflammatory cytokines, for example, tumor necrosis factor-α (TNF-α), interleukin-1 (IL-1), and IL-6 3, 4, 5, 6, 7, 8, 9, 10, 11, are produced following conditioning and play a primary role in activating T cells, leading to GVHD and resultant target tissue destruction 12, 13. An acute-phase protein, C-reactive protein (CRP), is produced by hepatocytes downstream of IL-6 [14] and is widely used as a reliable surrogate marker of infectious diseases 15, 16, 17, 18, 19. This process is further stimulated by other cytokines including TNF-α 12, 13. After allogeneic HSCT, the elevation of CRP was observed with infectious complications, but not in uncomplicated aGVHD 8, 20. On the other hand, elevation of CRP has been shown to be associated with TRM 21, 22, 23, 24. Nevertheless, these previous studies adopted the sporadic measurement of CRP and mostly focused on patients undergoing conventional HSCT (CST) with a myeloablative regimen. It has been hypothesized that recently developed reduced-intensity HSCT (RIST) decreases regimen-related toxicities and, hence, may reduce inflammation that augments the subsequent allogeneic immune reaction to induce GVHD and nonrelapse mortality (NRM).
In this study, the correlation between the preengraftment CRP value and subsequent clinical events was analyzed to test whether high CRP reflected the degree of tissue damage because of the conditioning regimen, infections, and allogeneic immune reactions and/or inflammation, all of which could contribute to subsequent aGVHD and NRM.
Materials and Methods
Patient Characteristics
The data from a cohort of 224 consecutive adult patients with hematologic malignancies, who were treated between January 2002 and July 2006 at the National Cancer Center Hospital (NCCH, Tokyo, Japan), were reviewed retrospectively. Patients who developed graft failure or who had previous allogeneic transplantation were excluded. Their characteristics are listed in Table 1. The median age of the patients was 47 years (range: 18-68 years), and their diagnosis included acute myeloid leukemia (AML, n = 94), acute lymphoblastic leukemia (ALL, n = 23), non-Hodgkin lymphoma (NHL, n = 62), myelodysplastic syndrome (MDS, n = 27) and chronic myeloid leukemia (CML, n = 12). Standard risk included acute leukemia in first complete remission, chronic leukemia in the first chronic phase, MDS in refractory anemia, and NHL in complete remission, with the rest of the patients categorized as a high-risk group. Stem cell sources used for transplantation included bone marrow (BM, n = 108), peripheral blood stem cells (PBSC, n = 98) and cord blood cells (CB, n = 18). One-hundred five patients received a CST regimen including total-body irridiation (TBI)-based (n = 50) and non-TBI-based busulfan-containing regimens (n = 55), whereas 119 patients received a RIST regimen including fludarabine or cladribine plus busulfan or melphalan (Table 1). CMV serostatus was positive in 157 patients and negative in 67 patients. The median age of the patients was 49 years in the high-CRP group (range: 19-67) and 47 years in the low-CRP group (range: 18- 68). Written informed consent was obtained according to the Declaration of Helsinki.
Table 1. Patients' Characteristics
| N (%)/ Median | |||
|---|---|---|---|
| Variable | Low CRP Group CRP < 15 mg/dL n = 157 | High CRP Group CRP ≥ 15 mg/dL n = 67 | P Value |
| Age (year) | 47 (18-68) | 49 (19-67) | .85 |
 <40 | 53 (34) | 26 (39) | |
| ≥40 | 104 (66) | 41 (61) | .47 |
| Patient sex | |||
| Male | 84 (54) | 48 (72) | |
| Female | 73 (46) | 19 (28) | .01 |
| Donor sex | |||
| Male | 81 (52) | 30 (45) | |
| Female | 76 (48) | 37 (55) | .35 |
| CMV serostatus | |||
| Positive | 140 (89) | 64 (96) | |
| Negative | 17 (11) | 3 (4) | .20 |
| Disease risk | |||
| Standard | 35 (22) | 17 (25) | |
| High | 122 (78) | 50 (75) | .62 |
| Conditioning | |||
| CST | 72 (47) | 33 (50) | |
| RIST | 85 (53) | 34 (50) | .64 |
| GVHD prophylaxis | |||
| Cyclosporin-based | 122 (78) | 52 (78) | |
| Tacrolimus-based | 35 (22) | 15 (22) | .99 |
| Short term MTX (+) | 107 (68) | 58 (87) | .004 |
| Relation to donor | |||
| Related | 94 (60) | 13 (19) | |
| Unrelated | 63 (40) | 54 (81) | <.001 |
| Stem cell source | |||
| Bone marrow | 63 (40) | 45 (67) | |
| PBSC | 87 (55) | 11 (16) | |
| Cord blood | 7 (5) | 11 (16) | <.001 |
Transplantation Procedures
GVHD prophylaxis included cyclosporine- (n = 174) and tacrolimus-based regimens (n = 50), with an additional short course of methotrexate (MTX) in 165 patients. Granulocyte colony-stimulating factor (G-CSF) was administered in all patients from day +6 of transplantation until engraftment was confirmed. Most patients received ciprofloxacin (200 mg orally 3 times daily) for bacterial prophylaxis until neutrophil engraftment. Fluconazole (100 mg once daily) was administered for fungal prophylaxis. Low-dose acyclovir was given for prophylaxis against herpes simplex virus and varicella zoster virus until the cessation of immunosuppressive agents. Prophylaxis against Pneumocystis jiroveci infection was provided with trimethoprim-sulfamethoxazole (400 mg of sulfamethoxazole once daily) from the first day of conditioning to day −3 of transplantation, and from day +28 until day +180 or the discontinuation of immunosuppressive agents. Patients with fever during the neutropenic period were treated with cefepime, and additional agents including vancomycin and aminoglycosides, and amphotericin B were given as clinically indicated. Neutrophil engraftment was defined as the first of 3 consecutive days after transplantation that the absolute neutrophil count exceeded 0.5 × 109/L. In our institute, the CRP level was serially measured as part of our routine checkup at least 3 times a week. Hence, all serially admitted patients were subjected to this analysis. Every patient had started CRP measurement before the initiation of the conditioning regimen, and the median pretransplant CRP level was 0.3 mg/dL (range: 0.0-20.5 mg/dL). The median maximum CRP value during neutropenia was 8.9 mg/dL (0.1-42.7, Table 2).
Table 2. Comparison of Preengrafment CRP Value Stratified According to the Conditioning Regimen (CST versus RIST) and the Relation to Donor (Related versus Unrelated)
| Patients' Characteristics | CRP Value Median (Range) |
|---|---|
| All patients | 8.9 (0.1-42.7) |
| CST | 10.5 (0.3-31.3)∗ |
| Related | 9.4 (0.6-30.0)† |
| Unrelated | 10.6 (0.3-31.3)† |
| RIST | 6.2 (0.1-42.7)∗ |
| Related | 1.6 (0.1-9.7)‡ |
| Unrelated | 16.2 (0.5-42.7)‡ |
∗P = .017. |
†P = .33. |
‡P < .001. |
The “maximum CRP level” was determined by measuring both the CRP level and the neutrophil count, as shown in the example in Figure 1A. The average number of levels assessed for each patient was 8 (range: 1-30). The median day of the maximum CRP level was day 10 of HSCT (range: 0-25), with 79% of patients developing this in later days (≥8 days). The patients were categorized according to the maximum CRP level after the threshold CRP level was determined following a preliminary analysis of the maximum CRP level after CST using an ROC curve analysis (data not shown). The “low-CRP” group (CRP <15 mg/dL) included 157 patients and the “high-CRP” group (CRP ≥15 mg/dL) included 67 patients.
Figure 1.
An example of how we measured CRP in a representative patient (A). Dot plot of the CRP level. All patients (B), CST versus RIST (C) and related versus unrelated (D).
Statistical Analyses
The primary endpoint of this study was the occurrence of grade II-IV and grade III-IV aGVHD, according to the Consensus Criteria [25]. The secondary endpoints were overall survival (OS) and nonrelapse mortality (NRM). Standard descriptive statistics were used. Student t, chi-square, Fisher's exact test, and Wilcoxon rank-sum tests were used to compare clinical and patient characteristics. To analyze the pretransplant risk factors for a high CRP level, logistic analysis was used. OS was estimated using Kaplan-Meier curves. The cumulative incidence of aGVHD and NRM was estimated based on a Cox regression model for cause-specific hazards by treating progressive disease or relapse as a competing event. Cox proportional hazard models were used for the multivariate analysis of variables in aGVHD, NRM, and OS after HSCT. Clinical factors that were assessed for their association with aGVHD included patient age, patient sex, donor sex, CMV serostatus, conditioning regimen (CST versus RIST), donor (human leukocyte antigen [HLA]-matched versus HLA-mismatched, related versus unrelated), GVHD prophylaxis (cyclosporine-based versus tacrolimus-based, short-term MTX versus no MTX) and disease risk (standard versus high risk). NRM and OS were also assessed for their association with these factors. Factors with P < .10 in the univariate analyses were subjected to a multivariate analysis using a multiple logistic analysis and Cox proportional hazard modeling. In Japan, only BM and CB are allowed for unrelated transplantation, and most transplantations with a related donor use PBSC as a stem cell source. Therefore, the stem cell source was not included as a factor in the multivariate analysis. A level of P < .05 was defined as statistically significant. All P values are 2-sided. All analyses were made with SPSS ver 10.0 statistical software (Chicago, IL). This analysis was approved by the institutional review board.
Results
Infections
The median duration of follow-up in surviving patients was 965 days (61 to 1432 days) in the high-CRP group and 915 days (76 to 1803 days) in the low-CRP group, and the incidence of total documented infections during neutropenia was, respectively, 23 cases in the high-CRP group (34%) and 27 cases in the low-CRP group (17%, P = .004). The incidence of bacteremia was, respectively, 20 cases (30%) and 20 cases (13%, P = .002), and the incidence of pneumonia was 7 cases (10%) and 4 cases (3%, P = .01). The incidence of central venous catheter infection was, respectively, 4 cases (6%) and 7 cases (4%, P = .63).
Serial changes in the CRP level are shown in Figure 1B; in most cases, the CRP level was elevated within 2 weeks of HSCT. Stratified data according to conditioning regimen (CST versus RIST) or relation to donor (related versus unrelated) are shown in Figure 1C and D, respectively.
To clarify the pretransplant risk factors for high CRP values during neutropenia, we performed a logistic regression analysis, which showed that male, unrelated donor, stem cell source with BM or CB transplantation (versus PBSCT), HLA-mismatched donor, and immunosuppression with MTX were associated with high CRP values during neutropenia (Table 1). Factors that showed significant associations (P < .1) were subjected to a multiple logistic regression analysis, and the results showed that unrelated donor, HLA mismatch and male sex were associated with high CRP (P < .001, P = .005, P = .028, respectively), as shown in Table 3. The median CRP levels after CST and RIST were 10.5 (0.3-31.3) and 6.2 (0.1-42.7), respectively, with a significant difference (P = .017) (Table 2). Notably, within the RIST group, the median CRP level was significantly lower in related than in unrelated transplantation (1.6 mg/dL [0.1-9.7] veruss 16.2 mg/dL [0.5-42.7]: P < .001). However, the logistic analysis failed to disclose any overall significant difference between CST and RIST.
Table 3. Multiple Logistic Regression Analysis of Risk Factors for High CRP during NeutropeniaFactors with P < .10 in a Multivariate Analysis Was Shown∗
| Multiple Logistic Regression Analysis | |||
|---|---|---|---|
| Outcomes and Variables | Odds | 95% CI | P Value |
| Unrelated donor | 4.6 | 2.2-9.6 | <.001 |
| HLA mismatch | 2.6 | 1.3-5.0 | .005 |
| Patient sex (male) | 2.1 | 1.1-4.2 | .0028 |
∗Factors included in univariate analysis: patient sex, donor sex, CMV serostatus, use of short-term MTX, relation to donor, HLA mismatch, conditioning, GVHD prophylaxis, stem cell source. |
Primary Outcomes
The cumulative incidences of aGVHD grade II-IV and grade III-IV are shown, respectively, in Figure 2A and B. Grade II-IV and grade III-IV aGVHD were both more frequent in the high-CRP group than in the low-CRP group (P = .001 and P = .04, respectively). A Cox proportional hazard model showed that a high CRP level and CMV serostatus were associated with an increased risk of grade II-IV aGVHD (Table 4). Similar results were obtained when we included only the patients who received a myeloablative conditioning regimen (grade II-IV aGVHD 25% in the low-CRP group and 58% in the high-CRP group, P < .001, grade III-IV aGVHD 7% in the low-CRP group and 21% in the high-CRP group, P = .047).
Figure 2.
Cumulative incidence of grade II-IV aGVHD (A) and grade III-IV aGVHD (B) stratified according to the maximal CRP level during neutropenia.
Table 4. Multiple Variate Analysis for aGVHD, NRM, and OS∗
| Outcomes and Variables | Hazard Ratio | 95% CI | P value |
|---|---|---|---|
| Grade II-IV aGVHD | |||
| High CRP | 1.7 | 1.1-2.6 | .02 |
| CMV positivity | 3.1 | 1.0-9.8 | .5 |
| Disease risk (high) | 1.6 | 0.9-2.7 | .10 |
| NRM | |||
| High CRP | 4.0 | 2.0-8.0 | <.001 |
| Age (≥40 years old) | 1.9 | 0.9-3.9 | .07 |
| Disease risk (high) | 2.6 | 1.1-6.2 | .03 |
| OS | |||
| High CRP | 2.0 | 1.3-3.1 | .002 |
| Disease risk (high) | 2.2 | 1.2-4.0 | .02 |
∗Factors included in univariate analysis: patient sex, donor sex, CMV serostatus, use of short-term MTX, relation to donor, HLA mismatch, conditioning, GVHD prophylaxis, stem cell source |
Secondary Outcomes
OS and NRM are shown, respectively, in Figure 3A and B. OS was significantly worse in the high-CRP group than in the low-CRP group (1-year OS 47% versus 75%, P = .001). NRM was significantly higher in the high-CRP group than in the low-CRP group (1-year NRM 47% versus 13%, P < .001). Similar results were obtained when we included only patients who received a myeloablative conditioning regimen (1-year NRM 8% in the low-CRP group and 38% in the high-CRP group, P = .007). A Cox proportional hazard model showed that the risk factors for poor OS were high CRP (P = .002, hazard ratio [HR] 2.0, 95% confidence interval [CI] 1.3-3.1) and high-risk disease (P = .015, HR 2.2, 95% CI 1.2-4.0), whereas those for high NRM were high CRP (P < .001, HR 4.0, 95% CI 2.0-8.0) and high-risk disease (P = .029, HR 2.6, 95% CI 1.1-6.2), as shown in Table 4. When the threshold was set at 15 mg/dL, the sensitivity and specificity of the CRP level for prediction of grade II-IV aGVHD, NRM, or OS were 37% and 75%, 59% and 79%, and 40% and 78%, respectively. The relapse rate was significantly lower in the high-CRP group than in the low-CRP group (1-year relapse 21% versus 33%, P = .02).
Figure 3.
OS stratified according to the maximal CRP level during neutropenia (A). Cumulative incidence of TRM stratified according to the maximal CRP level during neutropenia (B).
Causes of death are summarized in Table 5. A total of 57 patients (36%) in the low-CRP group and 39 patients (58%) in the high-CRP group died (P = .002, OR 2.4 [1.4-4.4]). Six patients (4%) in the low- and 5 (7%) in the high-CRP group died because of aGVHD, for example, death because of infectious diseases associated with aGVHD and its treatment. Seven patients (4%) in the low- and 11 (16%) in the high-CRP group (P = .003, OR 4.2 [1.6-11.4]) died because of chronic GVHD (cGVHD), including death because of infectious diseases associated with cGVHD and its treatment. No patient (0%) in the low- and 5 (7%) in the high-CRP group (P = .002) died because of infectious diseases excluding infectious disease concomitant with GVHD. No patient in the low-CRP group and 4 (6%) in the high-CRP group (P = .008) died because of multiple-organ failure (MOF) excluding MOF because of GVHD and infectious disease.
Table 5. Causes of Death Stratified According to CRP Value during Neutropenia
| Causes of death | Low CRP Group CRP < 15 mg/dL n = 157 | High CRP Group CRP ≥ 15 mg/dL n = 67 | P Value |
|---|---|---|---|
| Total | 57 (36%) | 39 (58%) | .002 |
| Relapse/progressive disease | 34 (22%) | 8 (12%) | .09 |
| acute GVHD (total) | 6 (4%) | 5 (7%) | .25 |
| acute GVHD | 5 (3%) | 3 (5%) | .63 |
| acute GVHD + infection | 1 (1%) | 2 (3%) | .16 |
| chronic GVHD (total) | 7 (4%) | 11 (16%) | .003 |
| chronic GVHD | 3 (2%) | 7 (10%) | .005 |
| chronic GVHD + infection | 4 (3%) | 4 (6%) | .21 |
| Infection∗ | 0 (0%) | 5 (7%) | .002 |
| MOF† | 0 (0%) | 4 (6%) | .008 |
| Respiratory failure‡ | 3 (2%) | 4 (6%) | .11 |
| Others | Stroke 2 | VOD 1 | |
| VOD 2 | Myocardial infarction 1 | ||
| Secondary cancer 1 | |||
| Unknown 2 |
∗Excluding infection during GVHD or GVHD treatment. |
†Excluding MOF due to GVHD, infection. |
‡Excluding respiratory failure because of GVHD, infection, and MOF. |
Discussion
The results of this retrospective study suggested that higher CRP values during the neutropenic period may reflect net inflammation secondary to tissue damage because of the conditioning regimen, infection, and subsequent allogeneic immune reactions, all of which lead to aGVHD/cGVHD and ultimate NRM. In a mouse model, the concept that the production of inflammatory cytokines plays an important role in the development of aGVHD, by affecting the afferent and effector phase 12, 13, has been accepted. Cooke et al. [26] showed that LPS antagonism reduced aGVHD in a mouse model, as indicated by Ferrara et al. [4]. However, in human studies, the value of determining individual levels of cytokines to monitor aGVHD has not been fully explored, because this approach is very costly and requires sophisticated techniques, which impedes its universal applicability. On the other hand, CRP is already being widely used worldwide, especially in Japan, to distinguish bacterial infections from other causes of fever 15, 16, 17, 18, 19. Based on this practice, we reviewed the value of the CRP level after HSCT, and our data suggest that it might be useful to monitor the CRP value as a net surrogate marker for produced cytokines, and for predicting the subsequent development of aGVHD and NRM.
Our patients had various interacting backgrounds, and it is still difficult to predict whether a patient with a high CRP level is destined to suffer from GVHD or major infectious complications. Infectious diseases were previously reported to be a primary cause of elevated CRP 8, 20, which might, in turn, affect the severity of aGVHD. In this study, we made every effort, including intense culture studies, to exclude infection as a primary cause of increased CRP, and showed that there were significantly more documented infections in the high-CRP group than in the low-CRP group. Current practice for the prevention of infection mostly focuses on the effective control of Gram-negative bacteria, considering the potent immediate pathologic effect of the organisms. However, if the hypothesis that decreasing the net production of cytokines is important for the prevention of subsequent GVHD is correct, more effort should be paid to broadly cover other types of organisms or even clinically less significant infection, that is, stomatitis, at least during the early period of neutropenia, particularly in patients carrying risk factors for high CRP, which included unrelated donor, HLA mismatch, BM, and CB transplantation in this study. The addition of other markers, such as procalcitonin, may be useful for identifying the risk of major infectious complications [24].
Tissue damage caused by the conditioning regimen, complicated infections, and allogeneic immune reactions are the primary factors that are associated with the initial elevation of CRP early in the course of allogeneic HSCT. Consequently, it can be speculated that a reduced-intensity conditioning regimen results in decreased cytokine release and a resultant lower CRP value, which may lead to less chance of developing GVHD. Although the RIST regimens we used were relatively dose-intense, in this retrospective review we still found that CRP levels tended to be decreased after RIST compared to conventional myeloablative transplantation, particularly in a related compared to an unrelated transplantation setting. Because augmentation of allogeneic immune and inflammation reactions may induce a higher CRP value, we speculate that the benefit of RIST is diminished when a strong allogeneic reaction is induced, as in cases of unrelated transplantation.
To further evaluate the relationship between a higher CRP value during neutropenia and common risk factors associated with transplantation, we performed a multivariate analysis and showed that unrelated donor, HLA mismatch, and male sex were associated with higher CRP values. Additionally, from the finding in the multivariate analysis that unrelated donor and HLA mismatch were independently associated with high CRP, we surmised that the degree of genetic disparity might be associated with higher CRP during neutropenia. Based on a consideration of these findings together, we think that a higher CRP value may reflect the degree of tissue damage because of the transplant regimen and the subsequent magnitude of allogeneic immune reactions. Nevertheless, our analysis was hampered, because in Japan only BM and CB are allowed for unrelated transplantations, and most transplantations with a related donor use PBSC as a stem cell source. In these settings, a theoretically longer neutropenic period after unrelated BM or CB transplantation might be associated with a higher risk of infection, which could lead to higher CRP, as shown in this study.
In this study, the primary causes of death in the low-CRP group were mainly relapse and progression, whereas in the high-CRP group this was NRM. Notably, the observation that the relapse rate was higher in the low-CRP group than in the high-CRP group, as previously suggested by Min et al. [23], may further support our hypothesis that serum CRP values represent overall inflammation and cytokine production, which paves the way to GVHD and related graft-versus-leukemia (GVL) effects. A possible reason for this finding is that a low CRP level resulted in a lower incidence of GVHD and a resultant decrease in the GVL effect, or the high-CRP group developed earlier and more-frequent death from NRM compared to the low-CRP group, which left fewer patients for evaluation of the later occurrence of relapse.
In conclusion, our results suggest that the CRP value in the neutropenic period before engraftment in patients undergoing allogeneic HSCT may be a net surrogate marker of early inflammation that leads to the development of aGVHD/cGVHD and subsequent NRM, as has been proposed in mouse models. The intensity of the conditioning regimen, infectious diseases, and degree of allogeneic immune response attributed to HLA compatibility and the stem cell source may be the major factors that predict higher CRP values. Based on the results of this retrospective study, future clinical studies to evaluate the feasibility of earlier intervention and adjustment of the procedure for preventing GVHD and NRM based on monitoring of the early CRP value are warranted.
Acknowledgments
This work was presented in part as a poster presentation at the annual Meeting of EBMT, Lyon, March 2007. This study was supported in part by grants from the Ministry of Health, Labor and Welfare, and Advanced Clinical Research Organization, Japan. There is no potential conflict of interest to declare.
References
- Acute graft-versus-host disease. The incidence and risk factors. Ann Transplant. 2006;11:16–23
- Risk factors for acute graft-versus-host disease in histocompatible donor bone marrow transplantation. Transplantation. 1991;51:1197–1203
- . Cytokine cascades in acute graft-versus-host disease. Transplantation. 1997;64:553–558
- . The cytokine modulation of acute graft-versus-host disease. Bone Marrow Transplant. 1998;21(Suppl 3):S13–S15
- . The role of endotoxin and the innate immune response in the pathophysiology of acute graft versus host disease. J Endotoxin Res. 2002;8:441–448
- . Soluble interleukin-6 receptors in hematology patients undergoing bone marrow transplantation. Transplantation. 1996;62:138–142
- . Serum cytokine levels after HLA-identical bone marrow transplantation. Transplantation. 1998;66:863–871
- Serum levels of interleukin 6, interleukin 8, and C-reactive protein after human allogeneic bone marrow transplantation. Transplantation. 1994;58:430–436
- . Serum cytokine levels in patients undergoing bone marrow transplantation. Bone Marrow Transplant. 1993;12:331–336
- Serum interferon-gamma and C-reactive protein levels as predictors of acute graft-vs-host disease in allogeneic hematopoietic precursor cell (marrow or peripheral blood progenitor cells) recipients. Transplant Proc. 1996;28:3522–3525
- . The relationship of serum IL-6 levels to acute graft-versus-host disease and hepatorenal disease after human bone marrow transplantation. Transplantation. 1992;54:457–462
- Clinical “cytokine storm” as revealed by monocyte intracellular flow cytometry: correlation of tumor necrosis factor alpha with severe gut graft-versus-host disease. Clin Gastroenterol Hepatol. 2004;2:237–245
- . Cytokine dysregulation and acute graft-versus-host disease. Blood. 1992;80:2964–2968
- . Interleukin-6 and the acute phase response. Biochem J. 1990;265:621–636
- . Acute-phase proteins and other systemic responses to inflammation. N Engl J Med. 1999;340:448–454
- . C-reactive protein: a valuable aid for the management of febrile children with cancer and neutropenia. Clin Infect Dis. 1994;18:589–595
- . A prospective study of daily measurement of C-reactive protein in serum of adults with neutropenia. Clin Infect Dis. 1995;21:114–121
- Use of inflammatory markers for early detection of bacteraemia in patients with febrile neutropenia. Scand J Infect Dis. 2004;36:365–371
- Markers of bacteremia in febrile neutropenic patients with hematological malignancies: procalcitonin and IL-6 are more reliable than C-reactive protein. Eur J Clin Microbiol Infect Dis. 2004;23:539–544
- . The effects of pretransplant conditioning, graft-versus-host disease and sepsis on the CRP levels in bone marrow transplantation. Infection. 1997;25:335–338
- Monitoring of C-reactive protein after allogeneic bone marrow transplantation identifies patients at risk of severe transplant-related complications and mortality. Bone Marrow Transplant. 1998;22:79–85
- An early increase in serum levels of C-reactive protein is an independent risk factor for the occurrence of major complications and 100-day transplant-related mortality after allogeneic bone marrow transplantation. Bone Marrow Transplant. 2002;30:441–446
- Patterns of C-reactive protein release following allogeneic stem cell transplantation are correlated with leukemic relapse. Bone Marrow Transplant. 2006;37:493–498
- Evaluation of C-reactive protein, interleukin-6, and procalcitonin levels in allogeneic hematopoietic stem cell recipients. Eur J Haematol. 2006;76:93–101
- 1994 consensus conference on acute GVHD grading. Bone Marrow Transplant. 1995;15:825–828
- LPS antagonism reduces graft-versus-host disease and preserves graft-versus-leukemia activity after experimental bone marrow transplantation. J Clin Invest. 2001;107:1581–1589
PII: S1083-8791(08)00087-6
doi:10.1016/j.bbmt.2008.02.008
© 2008 American Society for Blood and Marrow Transplantation. Published by Elsevier Inc. All rights reserved.
Volume 14, Issue 5 , Pages 510-517, May 2008
