Abstract
In a recently published and validated definition of fluid overload (FO), grade ≥ 2 FO was significantly associated with an increased risk of non-relapse mortality (NRM) after unmodified and haploidentical allogeneic hematopoietic cell transplantation (allo-HCT) using calcineurin inhibitor-based (CNI) graft-versus-host disease (GVHD) prophylaxis. We evaluated the effect of FO on outcomes in 169 patients undergoing myeloablative-conditioned ex-vivo CD34+ selected allo-HCT using the same grading scale. Thirty patients (17.8%) had grade ≥ 2 FO within the 30 days after ex-vivo CD34+ selected allo-HCT with a median onset at day 11 (range −8 to 28 days). Age ≥ 55 years [OR 3.43, p=0.005] and chemotherapy-based conditioning [OR 3.89, p=0.007] were associated with an increased risk of grade ≥ 2 FO. Patients with early grade ≥ 2 FO had a significantly higher NRM when compared to patients with grade < 2 FO [24.1% vs. 3.6% at day 100, p=0.01]. HCT-CI ≥ 3, FEV1 < 80, adjusted DLCO < 80, and HLA mismatch were associated with an increased risk of NRM, while total-body-irradiation-based conditioning was associated with a reduced risk of NRM. In a multivariate analysis, grade ≥ 2 FO was associated with increased NRM after adjusting for HCT-CI and HLA match [HR 2.3, p=0.014]. There was a trend toward inferior RFS in patients with grade ≥ 2 FO compared to patients with grade < 2 FO, 62% vs. 72% at 1 year (p=0.07), and a trend toward inferior OS, 69% vs. 79% at 1 year (p = 0.06), respectively. Our findings show that FO should be routinely assessed to identify patients at risk for NRM. Despite a CNI-free allo-HCT platform, regimen-related tissue and endothelial injury leads to FO in susceptible patients. FO is a highly relevant post-HCT toxicity that requires further inquiry.
INTRODUCTION:
A large body of observational data has shown that fluid overload (FO) is associated with prolonged intensive care unit (ICU) lengths of stay, organ injury, and increased mortality in critically ill patients.1–7 In a systematic review, patients that died in the ICU had significantly more FO than patients who survived. Moreover, ICU patients who achieved negative fluid balance had improved organ function and overall survival (OS).5 FO in critically ill patients exerts its effects through disruption of organ architecture, vascular drainage, and oxygen perfusion secondary to interstitial edema.6 FO is common among patients during the early post allogeneic hematopoietic cell transplantation (allo-HCT) period, even in the absence of sepsis or critical illness. Moreover, its influence on outcomes after allo-HCT has been poorly understood until recently. Rondon et al. developed and validated a stringent definition of FO to evaluate the incidence and prognostic implications of this complication in patients undergoing allo-HCT.8 When evaluating multiple HOT characteristics in multivariate analyses, grade ≥ 2 FO was the only factor associated with an increased risk of NRM at day +100 after unmodified and haploidentical allo-HCT using calcineurin inhibitor (CNI)-based graft-versus-host disease (GVHD) prophylaxis.8
In our recently published retrospective, comprehensive assessments of toxicities incurred by patients after ex-vivo CD34+ selected allo-HCT, FO was not commonly appreciated by CTCAE criteria.9–11 Considering the recent Rondon et al. findings demonstrating an association between FO and outcomes in the unmodified allo-HCT setting, we sought to determine the effect of FO in a cohort of patients undergoing myeloablative-conditioned allo- HCT with ex-vivo CD34+ selection as GVHD prophylaxis. This unique allo-HCT platform does not require CNIs which are known to cause endothelial cell damage through several mechanisms including the downregulation of thrombomodulin, reduction of endothelium-derived nitric oxide production, and the upregulation of endothelial pro-inflammatory mediators such as TNF-α, ICAM, VCAM, IL-6, that ultimately lead to vascular permeability.12–15 We hypothesized that despite a CNI-free GVHD prevention strategy, early grade ≥ 2 FO, as defined by a more specific criterion than the NCI Common Terminology Criteria for Adverse Events, would similarly be associated with an increased risk of NRM in this unique patient population.
METHODS:
All data were retrospectively queried from the electronic medical record. We included adult patients age ≥ 18 with adequate pre-transplantation organ function who underwent granulocyte colony stimulating factor (G-CSF)-mobilized peripheral blood stem cell allo-HCT at Memorial Sloan Kettering Cancer Center (MSKCC) between July 2006 and December 2012 using the CliniMACS CD34 Reagent System (Miltenyi Biotech, Gladbach, Germany) as GVHD prophylaxis. No further pharmacologic immunosuppression was used after transplantation. Eligible patients included all consecutive adult recipients of first allografts for the treatment of acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), myelodysplastic syndrome (MDS) or myeloproliferative neoplasms (MDN). Patients with multiple myeloma were excluded from the analysis. HLA typing was done using high-resolution DNA sequence-specific oligonucleotide typing for the HLA-A, -B, -C, -DRB1, and -DQB1 loci. All patients were treated with one of the following previously described myeloablative conditioning regimens at the discretion of the treating physician: busulfan, melphalan, and fludarabine (Bu/Mel/Flu); clofarabine, melphalan, and thiotepa (Clo/Mel/Thio); total body irradiation (1375 cGy), thiotepa, and cyclophosphamide (TBI/Thio/Cy); or TBI (1375 cGy), thiotepa, and fludarabine (TBI/Thio/Flu).16–18 All patients received rabbit anti-thymocyte globulin 2.5 mg/kg/day intravenously (IV) on days −3, −2, and on −1 for mismatched allografts.19,20 Patients who received a TBI-based preparative regimen received keratinocyte growth factor 60 μg/kg IV on days −13, −12, −11, 0, +1, and +2.21 G-CSF was initiated on day +7 and continued until absolute neutrophil count recovery. All patients received standard supportive care for prevention of sinusoidal obstruction syndrome and opportunistic antimicrobial prophylaxis according to standard MSKCC institutional guidelines. Disease risk was assessed using the Disease Risk Index for patients undergoing allo-HCT.22 Comorbidities were assessed using the Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI).23
Fluid Overload Assessment and Biostatistics
FO was scored using the following criteria: grade 1 - asymptomatic or mild edema, weight gain (WG) <10% from baseline; might require decreasing IV fluids or occasional diuretics; grade 2 - symptomatic and/or WG 10 to <20% from baseline; might require continuing diuretics; grade 3 - WG ≥20% from baseline or not responding to diuretics; and grade 4 - major organ dysfunction (renal, pulmonary, or cardiac dysfunction or requiring further treatment such as dialysis).8 FO was assessed between the time of admission (day −10) and day 30 post allo-HCT. The cumulative incidence of grade ≥ 2 FO was estimated using the cumulative incidence method, with day −10 considered to be time zero. Prior to day 30, 1 patient who had grade ≥ 2 FO died, and none were lost to follow-up. As such, risk factors associated with the development of grade ≥ 2 FO by day 30 were assessed using a logistic regression model. Risk factors of interest included patient gender, age, body-mass index (BMI), HCT-CI, conditioning regimen, disease, and pre-HCT creatinine, estimated glomerular filtration rate (eGFR) by Cockcroft-Gault, albumin, left ventricular ejection fraction, adjusted diffusing capacity of carbon monoxide (DLCO), HLA, and forced expiratory volume in 1 second (FEV1). A multivariate model for risk factors associated with FO was determined using a forward selection procedure, with all univariate variables with p-value < 0.05 considered as candidates for the multivariate model. The cumulative incidence of NRM and acute GVHD (aGVHD) were estimated using the cumulative incidence method for competing risks, with disease relapse considered to be a competing risk for NRM, and death or disease relapse for aGVHD. OS and relapse-free survival (RFS) were estimated using the Kaplan-Meier method. The effect of grade ≥ 2 FO on NRM, aGVHD, RFS, and OS was estimated using a landmark analysis at day 30, with differences between groups assessed using Gray’s test or the log-rank test for NRM/aGVHD and RFS/OS, respectively. A multivariate cause-specific Cox proportional hazards regression model for NRM was also determined using a forward selection procedure as above. All analyses were completed using R 3.5.0.
RESULTS:
We evaluated 169 patients in this analysis with a median follow-up of 48 months (range 12–113 months) among survivors. Table 1 summarizes all patient and HCT characteristics. The median age at transplantation was 57 years with a similar gender distribution. Most patients underwent allo-HCT for acute leukemia (57%), had a low risk DRI (69%), were CMV seropositive (60%), had received a 10/10 HLA-matched related or unrelated donor (78%), and had received chemotherapy-based myeloablative conditioning (67%). Thirty-seven patients received HLA mismatched allografts. Of the mismatched unrelated donor allografts, 36 patients received a one-allele mismatched allograft, and 1 patient received a two-allele mismatched allograft. The most common HCT-CI score was ≥3 (43%). The median infused cell doses were 8.52 × 106 CD34+ cells/kg and 3.2 × 103 CD3+ cells/kg.
Table 1:
Patient and HCT Characteristics
| Characteristics | N (%) |
|---|---|
| Number of Patients | 169 |
| Age, years (Median, Range) | 57 (20–73) |
| Gender | |
| Female | 87 (51) |
| Male | 82 (49) |
| Diagnosis | |
| Acute Leukemia | 96 (57) |
| MDS/MPD | 56 (33) |
| Other | 17 (10) |
| Disease Risk | |
| Low | 111 (66) |
| Intermediate | 22 (13) |
| High | 29(17) |
| Unevaluable | 7(4) |
| HCT-CI | |
| 0 | 37 (22) |
| 1–2 | 60 (36) |
| ≥ 3 | 72 (43) |
| Patient CMV serostatus | |
| Seropositive | 101 (60) |
| Seronegative | 4 68 (40) |
| HLA | |
| MUD | 64 (38) |
| MRD | 68 (40) |
| MM | 37 (22) |
| Conditioning Regimen | |
| Bu/Mel/Flu | 100 (59) |
| Cy/Thio/TBI (1375 cGy) | 55 (33) |
| Clo/Mel/Thioi | 13(8) |
| Flu/Thio/TBI (1375 cGy) | 1(1) |
| Infused Cell Dose (Median, Range) | |
| CD34+ × 106 | 8.5 (1.2–31.2) |
| CD3+ × 103 | 3.2 (0.7–63) |
Abbreviations: MDS, Myelodysplastic Syndrome; MPD, Myeloproliferative Disorders; HCT-CI; Hematopoietic Cell Transplantation Comorbidity Index; CMV, cytomegalovirus; MUD, matched unrelated donor; MRD, matched related donor; MM, mismatched donor; Bu, Busulfan; Mel, Melphalan; Flu, Fludarabine; Cy, Cyclophosphamide; Clo, Clofarabine; Thio, Thiotepa; TBI, Total Body Irradiation.
Incidence of Fluid Overload and Risk Factors
Thirty patients had grade ≥ 2 FO during the first 30 days post allo-HCT [cumulative incidence 17.8% (95% CI: 12.4–23.9)]. Of those, 18 (60%) had grade 2, 6 (20%) grade 3 and 6 (20%) grade 4. The median onset of any grade FO was the day of the allo-HCT [day 0] (range - 9 to 28 days), or 10 days after admission. The median onset of grade ≥ 2 FO was day 11 (range −8 to 28 days) after allo-HCT, or 21 days after admission. Patient and HCT characteristics associated with grade ≥ 2 FO by day 30 post-HCT are shown in Table 2. The univariate analyses showed that age ≥ 55 years [Odds Ratio (OR) 3.43 95%CI (1.44–9.12), p=0.005] and chemotherapy-based conditioning [OR 3.89 95%CI (1.42–13.71), p=0.007] were associated with an increased risk of grade ≥ 2 FO. In contrast, pre-HCT albumin ≥ 4 mg/dL [OR 0.43 (0.18–0.96), p=0.039] and increased BMI [25 – 29.9, OR 0.61 (0.25–1.43); ≥30 OR 0.23 (0.05–0.74), Overall p=0.043] were associated with a reduced risk of grade ≥ 2 FO (Table 2). Patient gender, disease, pre-transplantation HCT-CI score, and baseline pre-HCT levels of creatinine, eGFR, LVEF, DLCO, FEV1, and HLA match were not associated with the development of grade > 2 FO. In the multivariable model, age ≥ 55 years [OR 3.93 (1.60–10.84), p=0.005] and BMI [25– 29.9, OR 0.47 (0.18–1.14); ≥ 30, OR 0.20 (0.04–0.67), overall p=0.024] remained significantly associated with the development of grade ≥ 2 FO.
Table 2:
Univariate Associations between Patient/HCT Characteristics and Development of Grade ≥ 2 FO
| Variables | N | Odds Ratio (95% Cl) | P-value |
|---|---|---|---|
| Age (Years) | 0.005 | ||
| < 55 | 78 | Reference | |
| ≥ 55 | 91 | 3.43 (1.44–9.12) | |
| Gender | 0.064 | ||
| Female | 87 | Reference | |
| Male | 82 | 0.47 (0.20–1.04) | |
| BMI | 0.043 | ||
| ≤ 24.9 | 63 | Reference | |
| 25–29.9 | 64 | 0.61 (0.25–1.43) | |
| ≥ 30 | 42 | 0.23 (0.05–0.74) | |
| Creatinine | 0.814 | ||
| < 1.3 mg/dL | 154 | Reference | |
| ≥ 1.3 mg/dL | 15 | 1.18 (0.26–4.01) | |
| eGFR | 0.110 | ||
| < 60 mL/min | 122 | Reference | |
| ≥ 60 mL/min | 47 | 1.98 (0.85–4.49) | |
| Albumin | 0.039 | ||
| < 4 mg/dL | 84 | Reference | |
| ≥ 4 mg/dL | 85 | 0.43 (0.18–0.96) | |
| LVEF | |||
| < 63% | 70 | Reference | 0.862 |
| ≥ 63% | 99 | 1.07 (0.48–2.45) | |
| DLCO Adjusted | 0.422 | ||
| < 80% | 53 | Reference | |
| ≥ 80% | 115 | 0.71 (0.31–1.67) | |
| FEV1 | 0.264 | ||
| < 80% | 18 | Reference | |
| ≥ 80% | 151 | 0.52 (0.18–1.72) | |
| Conditioning Type | 0.007 | ||
| TBI-Based | 56 | Reference | |
| Chemo-Based | 113 | 3.89 (1.42, 13.71) | |
| Disease | 0.670 | ||
| Acute Leukemia | 96 | Reference | |
| MDS/MPD | 56 | 1.32 (0.55–3.10) | |
| Other | 17 | 1.66 (0.42–5.47) | |
| HCT-CI | 0.419 | ||
| 0 | 37 | Reference | |
| 1–2 | 60 | 1.13 (0.36–3.95) | |
| ≥ 3 | 72 | 1.83 (0.65–6.01) | |
| HLA | 0.479 | ||
| MRD | 68 | Reference | |
| MUD | 64 | 1.20 (0.47, 3.11) | |
| MM | 37 | 1.86 (0.67, 5.14) |
Fluid Overload and Outcomes
Among all patients, the cumulative incidence of NRM at day 100, 6 months, and 1 year was 7.7% (95% Cl: 4.3–12.4), 10.7% (6.6–15.9%), and 15.4% (10.4–21.3%), respectively. In univariate analysis, patients with early post allo-HCT grade ≥ 2 FO had a significantly higher cumulative incidence of NRM when compared to patients with grade < 2 FO [24.1% vs. 3.6% at day 100, p=0.01] as shown in Figure 1. HCT-CI ≥ 3, FEV1 < 80, adjusted DLCO < 80, and HLA mismatch were associated with an increased risk of NRM, while TBI-based conditioning was associated with a reduced risk of NRM as shown in Table 3. Age, gender, BMI, renal function (creatinine and eGFR), albumin, and disease were not associated with an increased risk of NRM. In a multivariate analysis, grade ≥ 2 FO was significantly associated with NRM [HR 2.3 (1.2–4.4), p = 0.014)] after adjusting for HCT-CI [1–2, HR 1.2 (0.4–3.6), ≥ 3, HR 4.4 (1.7–11.4), overall p<0.001] and HLA [MUD, HR 1.7 (0.8–3.7), MM, HR 2.7 (1.3 −5.7), overall p = 0.028].
Figure 1:
Cumulative Incidence of Non-Relapse Mortality Based on FO Grade by Day 30
Table 3:
Univariate Associations between Patient/HCT Characteristics and NRM
| Variables | N | 100 days (95% Cl) | 180 days (95% Cl) | 1 Year (95% Cl) | P-value |
|---|---|---|---|---|---|
| Day 30 FO Grade | 0.010 | ||||
| < 2 | 139 | 3.6 (1.3–7.7) | 6.5 (3.2–11.4) | 11.5 (6.9–17.5) | |
| ≥ 2 | 29 | 24 (10.5–40.9) | 27.6 (12.8–44.6) | 31 (15.3–48.3) | |
| Age (Years) | 0.439 | ||||
| < 55 | 78 | 5.1 (1.6–11.7) | 7.7 (3.1–15.0) | 11.5 (5.6–19.8) | |
| ≥ 55 | 91 | 9.9 (4.8–17.1) | 13.2 (7.2–21.0) | 18.7 (11.4–27.3) | |
| Gender | 0.820 | ||||
| Female | 87 | 8.0 (3.5–15.0) | 11.5 (5.9–19.2) | 16.1 (9.3–24.6) | |
| Male | 82 | 7.3 (3.0–14.3) | 9.8 (4.5–17.4) | 14.6 (8.0–23.2) | |
| BMI | 0.226 | ||||
| ≤24.9 | 63 | 7.9 (2.9–16.3) | 11.1 (4.8–20.3) | 14.3 (7.0–24.1) | |
| 25–29.9 | 63 | 6.2 (2.0–14.0) | 9.4 (3.8–18.1) | 15.6 (8.0–25.6) | |
| ≥30 | 42 | 9.5 (3.0–20.7) | 11.9 (4.3–23.7) | 16.7 (7.2–29.5) | |
| Creatinine | 0.329 | ||||
| < 1.3 mg/dL | 154 | 8.4 (4.7–13.5) | 11.0 (6.7–16.6) | 14.9 (9.8–21.0) | |
| ≥ 1.3 mg/dL | 15 | 0.0 (NA) | 6.7 (0.4–26.9) | 20.0 (4.5–43.3) | |
| eGFR | 0.277 | ||||
| < 60 mL/min | 122 | 8.2 (4.2–13.9) | 10.7 (6.0–16.9) | 15.6 (9.8–22.6) | |
| > 60 mL/min | 47 | 6.4 (1.6–15.9) | 10.6 (3.8–21.4) | 14.9 (6.5–26.6) | |
| Albumin | 0.110 | ||||
| < 4 mg/dL | 84 | 9.5 (4.4–17.0) | 13.1 (6.9–21.3) | 19.0 (11.4–28.1) | |
| ≥ 4 mg/dL | 85 | 5.9 (2.2–12.3) | 8.2 (3.6–15.3) | 11.8 (6.0–19.6) | |
| LVEF | 0.750 | ||||
| < 63% | 70 | 8.6 (3.5–16.6) | 8.6 (3.5–16.6) | 11.4 (5.3–20.2) | |
| ≥ 63% | 99 | 7.1 (3.1–13.3) | 12.1 (6.6–19.4) | 18.2 (11.3–26.4) | |
| DLCO Adjusted | 0.002 | ||||
| < 80% | 53 | 13.2 (5.7–23.8) | 17.0 (8.3–28.3) | 26.4 (15.4–38.8) | |
| ≥ 80% | 115 | 5.2 (2.1–10.4) | 17.0 (8.3–28.3) | 26.4 (15.4–38.8) | |
| FEV1 | 0.003 | ||||
| < 80% | 18 | 22.2 (6.6–43.6) | 22.2 (6.6–43.6) | 38.9 (16.6–60.9) | |
| ≥ 80% | 151 | 6.0 (2.9–10.5) | 9.3 (5.3–14.6) | 12.6 (7.9–18.4) | |
| Conditioning Type | 0.018 | ||||
| TBI-Based | 56 | 3.6 (0.7–11.0) | 5.4 (1.4–13.5) | 8.9 (3.2–18.2) | |
| Chemo-Based | 113 | 9.7 (5.1–16.1) | 13.3 (7.8–20.2) | 18.6 (12.0–26.3) | |
| Disease | 0.718 | ||||
| Acute Leukemia | 96 | 8.3 (3.9–15.0) | 10.4 (5.3–17.5) | 16.7 (10.0–24.8) | |
| MDS/MPD | 56 | 7.1 (2.3–15.9) | 12.5 (5.4–22.6) | 14.3 (6.6–24.8) | |
| Other | 17 | 5.9 (0.3–24.2) | 5.9 (0.3–24.2) | 11.8 (1.8–32.0) | |
| HCT-CI | <0.001 | ||||
| 0 | 37 | 2.7 (0.2–12.3) | 5.4 (0.9–16.1) | 8.1 (2.0–19.8) | |
| 1–2 | 60 | 3.3 (0.6–10.3) | 5.0 (1.3–12.7) | 10.0 (4.0–19.2) | |
| ≥3 | 72 | 13.9 (7.1–22.9) | 18.1 (10.2–27.8) | 23.6 (14.5–34.0) | |
| HLA | 0.012 | ||||
| MRD | 68 | 4.4 (1.2, 11.3) | 5.9 (1.9, 13.3) | 10.3 (4.5, 18.9) | |
| MUD | 64 | 9.4 (3.8, 18.1) | 10.9 (4.8, 20.0) | 12.5 (5.8, 21.9) | |
| MM | 37 | 10.8 (3.4, 23.3) | 18.9 (8.2, 33.0) | 29.7 (15.9, 44.9) |
There was no significant difference in the incidence of grade 2–4 acute GVHD between patients with grade < 2 FO and grade ≥ 2 FO (16% and 14% respectively at day 180, p=0.55), or grade 3–4 acute GVHD (7% and 3% respectively at day 180, p=0.79). There was a trend toward inferior RFS, 62% vs. 72% at 1 year (p=0.07), and inferior OS as shown in Figure 2, 69% vs. 79% at 1 year (p = 0.06), in patients with grade ≥ 2 FO compared to patients with grade < 2 FO, respectively. Among patients that developed grade ≥ 2 FO there were 18 deaths, with causes of death including: infection (n=7, 39%), relapse (n=4, 22%), GVHD (n=3, 17%), toxicity/organ failure (n=3, 17%), and unknown (n=1, 6%).
Figure 2.
Overall Survival Probability Stratified by FO Grade by Day 30
DISCUSSION:
In this study, we determined that early FO after ex-vivo CD34-selected allo-HCT is associated with a significantly increased risk of NRM, and that the grading criteria developed by Rondon et al. is a useful tool in FO assessment.8 Patients with grade ≥ 2 FO had a trend toward poorer OS, though not achieving statistical significance, which is likely being driven by an increased incidence of NRM. Our study was not designed to address the precise etiology of FO in these patients; however, the increased risk of NRM with high-grade FO appears to be multifactorial as indicated by the variable causes of death. Despite the absence of CNI-based GVHD prophylaxis, endothelial disruption due to multiple stressors, including direct regimen-related effects, may contribute to excessive extravascular fluid accumulation in at-risk patients. This endothelial vulnerability has been mechanistically associated with multiple serious post- HCT complications, and clinical FO may represent another such toxicity.24–26
FO occurring early after ex-vivo CD34-selected allo-HCT is associated with an increased risk of NRM even when adjusting for important predictors such as HCT-CI and HLA match. We found that age is an important risk factor in the development of FO. This is relevant given that increasing age is associated with greater baseline endothelial dysfunction, which has been shown to be a predictor of organ injury and NRM after HOT.12,27,28 Although not significant in MV analysis, patients receiving TBI-based myeloablative regimens had a decreased risk of FO and NRM, which may seem unanticipated, given the marked tissue injury associated with high-dose radiation.21,29 However, this finding may be explained by our institutional standard of practice, which is avoidance of myeloablative TBI-based regimens in patients above the age of 60.
The median onset of any grade FO occurred at the completion of the conditioning regimen and before allograft infusion, as was also noted in the previously published FO analysis.8 This highlights the cytotoxic damage caused by myeloablative conditioning on the vascular endothelial cell layer even in the absence of CNI-based conditioning.12–15,27 The endothelium has many essential physiologic roles including: hemostasis, maintenance of normal vascular tone, regenerative capacity after tissue injury, and fluid homeostasis. Vascular capillary bed fluid permeability is largely mediated through the endothelial cell layer, specifically the glycocalyx. Loss of normal glycocalyx and endothelial integrity secondary to cytotoxic injury causes a markedly disrupted systemic fluid balance leading to a deleterious cycle of intravascular fluid loss, interstitial edema, impaired vascular tone, and plasma protein loss.4,6 This is congruent with our finding that patients with normal albumin levels prior to allo-HCT had a reduced risk of FO. These events likely influence the use and overuse of IV fluids that may worsen excessive FO.
Our recently published comprehent of toxicities after ex-vivo CD34+ selection did not demonstrate many FO events, likely in part because the CTCAE does not include adequate criteria to grade this clinically relevant toxicity.9–11 Also, our toxicities were collected retrospectively, whereas the Rondon et al. FO grading system was evaluated in realtime which allows for clinicians to more accurately assess weight gain and edema. Despite this, the FO grading score was associated with NRM in our patients, demonstrating that it should be routinely used to identify patients at risk for NRM after allo-HCT. Moreover, our results highlight the need for more precise predictors of endothelial dysfunction as they relate to FO, and for routine, accurate assessments of fluid status in the early post-HCT period. With this, it will be possible to develop a more intricate understanding of optimal fluid management that may reduce NRM after allo-HCT.
Table 4.
Multivariate Associations between Patient/HCT Characteristics and NRM
| Variable | HR (95 % CI) | P-value |
|---|---|---|
|
Day 30 FO Grade |
||
| < 2 | Reference | 0.014 |
| ≥ 2 | 2.3 (1.2–4.4) | |
| HCT-CI | ||
| 0 | Reference | <0.001 |
| 1–2 | 1.2 (0.4–3.6) | |
| ≥3 | 4.4 (1.7–11.4) | |
| HLA | ||
| MRD | Reference | 0.028 |
| MUD | 1.73 (0.8–3.7) | |
| MM | 2.7 (1.3–5.7) |
HIGHLIGHTS:
We applied a novel grading system to assess the effects of early fluid overload (FO) on outcomes in patients after ex-vivo CD34+ selected allo-HCT.
Patients with grade > 2 FO had significantly higher NRM than those with grade < 2 FO.
Even in the absence of CNI-based GVHD prophylaxis, FO remains a highly relevant early transplant toxicity that should be routinely assessed.
ACKNOWLEDGMENTS:
This research was supported in part by the National Institutes of Health (NIH) Grant P01 CA23766 and NIH/NCI Cancer Center Support Grant P30 CA008748. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The Matlin Fund and Bergstein Family Fund provided philanthropic research support.
Footnotes
Conflicts of Interest Statement: There are no relevant conflicts of interest to report.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES:
- 1.Alsous F Negative Fluid Balance Predicts Survival in Patients With Septic Shock *. CHEST J. 2000; 117(6): 1748–1749. doi: 10.1378/chest.117.6.1749. [DOI] [PubMed] [Google Scholar]
- 2.Ozkahya M, Ok E, Toz H, et al. Long-term survival rates in haemodialysis patients treated with strict volume control. Nephrol Dial Transplant. 2006;21(12):3506–3513. doi: 10.1093/ndt/gfl487. [DOI] [PubMed] [Google Scholar]
- 3.Kalantar-Zadeh K, Regidor DL, Kovesdy CP, et al. Fluid retention is associated with cardiovascular mortality in patients undergoing long-term hemodialysis. Circulation. 2009;119(5):671–679. doi:10.1161/CIRCULATIONAHA.108.807362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Prowle JR, Echeverri JE, Ligabo EV, Ronco C, Bellomo R. Fluid balance and acute kidney injury. Nat Rev Nephrol. 2010;6(2):107–115. doi:10.1038/nrneph.2009.213. [DOI] [PubMed] [Google Scholar]
- 5.Boyd JH, Forbes J, Nakada T, Walley KR, Russell JA. Fluid resuscitation in septic shock: A positive fluid balance and elevated central venous pressure are associated with increased mortality*. Crit Care Med. 2011;39(2):259–265. doi:10.1097/CCM.0b013e3181feeb15. [DOI] [PubMed] [Google Scholar]
- 6.O’Connor ME, Prowle JR. Fluid Overload. Crit Care Clin. 2015;31(4):803–821. doi:10.1016/j.ccc.2015.06.013. [DOI] [PubMed] [Google Scholar]
- 7.Salahuddin N, Sammani M, Hamdan A, et al. Fluid overload is an independent risk factor for acute kidney injury in critically III patients: results of a cohort study. BMC Nephrol. 2017; 18(1 ):45. doi: 10.1186/s12882-017-0460-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rondon G, Saliba RM, Chen J, et al. Impact of Fluid Overload as New Toxicity Category on Hematopoietic Stem-Cell Transplant Outcomes. Biol Blood Marrow Transplant. 2017;0(0). doi: 10.1016/j.bbmt.2017.08.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kosuri S, Adrianzen Herrera D, Scordo M, et al. The Impact of Toxicities on First-Year Outcomes after Ex Vivo CD34 + -Selected Allogeneic Hematopoietic Cell Transplantation in Adults with Hematologic Malignancies. Biol Blood Marrow Transplant July 2017. doi: 10.1016/j.bbmt.2017.07.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Scordo M, Shah GL, Kosuri S, et al. Effects of Late Toxicities on Outcomes in Long-Term Survivors of Ex-Vivo CD34+-Selected Allogeneic Hematopoietic Cell Transplantation. Biol Blood Marrow Transplant. 2017. doi:10.1016/j.bbmt.2017.08.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shah GL, Scordo M, Kosuri S, et al. Impact of Toxicity on Survival for Older Adult Patients after CD34+ Selected Allogeneic Hematopoietic Stem Cell Transplantation. Biol Blood Marrow Transplant. 2017. doi:10.1016/j.bbmt.2017.08.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cooke KR, Jannin A, Ho V. The contribution of endothelial activation and injury to endorgan toxicity following allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2008; 14(1 Suppl 1):23–32. doi: 10.1016/j.bbmt.2007.10.008. [DOI] [PubMed] [Google Scholar]
- 13.Palomo M, Diaz-Ricart M, Carbo C, et al. The Release of Soluble Factors Contributing to Endothelial Activation and Damage after Hematopoietic Stem Cell Transplantation Is Not Limited to the Allogeneic Setting and Involves Several Pathogenic Mechanisms. Biol Blood Marrow Transplant. 2009;15(5):537–546. doi:10.1016/j.bbmt.2009.01.013. [DOI] [PubMed] [Google Scholar]
- 14.Ikezoe T, Yang J, Nishioka C, Honda G, Furihata M, Yokoyama A. Thrombomodulin protects endothelial cells from a calcineurin inhibitor-induced cytotoxicity by upregulation of extracellular signal-regulated kinase/myeloid leukemia cell-1 signaling. Arterioscler Thromb Vase Biol. 2012;32(9):2259–2270. doi:10.1161/ATVBAHA.112.251157. [DOI] [PubMed] [Google Scholar]
- 15.Rodrigues-Diez R, Gonzalez-Guerrero C, Ocana-Salceda C, et al. Calcineurin inhibitors cyclosporine A and tacrolimus induce vascular inflammation and endothelial activation through TLR4 signaling. Sci Rep. 2016;6. doi:10.1038/srep27915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Papadopoulos EB, Carabasi MH, Castro-Malaspina H, et al. T-cell-depleted allogeneic bone marrow transplantation as postremission therapy for acute myelogenous leukemia: freedom from relapse in the absence of graft-versus-host disease. Blood. 1998;91 (3):1083–1090. http://www.ncbi.nlm.nih.gov/pubmed/9446672. [PubMed] [Google Scholar]
- 17.Jakubowski AA, Small TN, Kernan NA, et al. T cell-depleted unrelated donor stem cell transplantation provides favorable disease-free survival for adults with hematologic malignancies. Biol Blood Marrow Transplant. 2011;17(9):1335–1342. doi:10.1016/j.bbmt.2011.01.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Castro-Malaspina H, Jabubowski AA, Papadopoulos EB, et al. Transplantation in Remission Improves the Disease-Free Survival of Patients with Advanced Myelodysplastic Syndromes Treated with Myeloablative T Cell-Depleted Stem Cell Transplants from HLA-Identical Siblings. Biol Blood Marrow Transplant. 2008;14(4):458–468. doi: 10.1016/j.bbmt.2008.02.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jakubowski AA, Small TN, Young JW, et al. T cell-depleted stem-cell transplantation for adults with hematologic malignancies: Sustained engraftment of HLA-matched related donor grafts without the use of antithymocyte globulin. Blood. 2007;110(13):4552–4559. doi: 10.1182/blood-2007-06-093880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Small TN, Papadopoulos EB, Boulad F, et al. Comparison of immune reconstitution after unrelated and related T-cell-depleted bone marrow transplantation: effect of patient age and donor leukocyte infusions. Blood. 1999;93(2):467–480. [PubMed] [Google Scholar]
- 21.Goldberg JD, Zheng J, Castro-Malaspina H, et al. Palifermin is efficacious in recipients of TBI-based but not chemotherapy-based allogeneic hematopoietic stem cell transplants. Bone Marrow Transplant. 2013;48(1):99–104. doi:10.1038/bmt.2012.115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Armand P, Kim HT, Logan BR, et al. Validation and refinement of the Disease Risk Index for allogeneic stem cell transplantation. Blood. 2014;123(23):3664–3671. doi: 10.1182/blood-2014-01-552984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sorror ML, Logan BR, Zhu X, et al. Prospective Validation of the Predictive Power of the Hematopoietic Cell Transplantation Comorbidity Index: A Center for International Blood and Marrow Transplant Research Study. Biol Blood Marrow Transplant. 2015;21(8):1479–1487. doi:10.1016/j.bbmt.2015.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Dietrich S, Falk CS, Benner A, et al. Endothelial Vulnerability and Endothelial Damage Are Associated with Risk of Graft-versus-Host Disease and Response to Steroid Treatment. Biol Blood Marrow Transplant. 2013;19(1):22–27. doi:10.1016/j.bbmt.2012.09.018. [DOI] [PubMed] [Google Scholar]
- 25.Zeisbrich M, Becker N, Benner A, et al. Transplant-associated thrombotic microangiopathy is an endothelial complication associated with refractoriness of acute GvHD. Bone Marrow Transplant. 2017;52(10):1399–1405. doi:10.1038/bmt.2017.119. [DOI] [PubMed] [Google Scholar]
- 26.Luft T, Benner A, Jodele S, et al. EASIX in patients with acute graft-versus-host disease: a retrospective cohort analysis. Lancet Haematol. 2017;4(9):e414–e423. doi:10.1016/S2352-3026(17)30108-4. [DOI] [PubMed] [Google Scholar]
- 27.Palomo M, Diaz-Ricart M, Carbo C, et al. Endothelial dysfunction after hematopoietic stem cell transplantation: role of the conditioning regimen and the type of transplantation. Biol Blood Marrow Transplant. 2010;16(7):985–993. doi:10.1016/j.bbmt.2010.02.008. [DOI] [PubMed] [Google Scholar]
- 28.Carreras E, Diaz-Ricart M. The role of the endothelium in the short-term complications of hematopoietic SCT. Bone Marrow Transplant. 2011;46(12):1495–1502. doi:10.1038/bmt.2011.65. [DOI] [PubMed] [Google Scholar]
- 29.Takatsuka H, Wakae T, Mori A, Okada M, Okamoto T, Kakishita E. Effects of total body irradiation on the vascular endothelium. Clin Transplant. 2002;16(5):374–377. doi:10.1034/j.1399-0012.2002.02035.x. [DOI] [PubMed] [Google Scholar]