Prognostic Value of the Pretransplant Fibrosis-4 Index on Non-Relapse and Overall Mortality following Unrelated Single-Unit Cord Blood Transplantation in Adults

Cord blood transplantation (CBT) from an unrelated donor is an alternative allogeneic hematopoietic cell transplantation (HCT) option in the absence of appropriate human leukocyte antigen-matched related or unrelated donors [1‒5]. In contrast to allogeneic HCT from adult donors, CBT has several distinct clinical characteristics, including delayed hematopoietic recovery, a lower incidence of graft-versus-host disease (GVHD), and a higher incidence of infectious complications. Liver injury is also common in patients undergoing CBT [6, 7]; its causes are multifactorial and include GVHD, infection, veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS), and drug toxicities.

The fibrosis-4 (FIB-4) index, based on age and routine laboratory parameters, including aspartate aminotransferase (AST), alanine aminotransferase (ALT), and platelet count, was initially established as a noninvasive marker of liver fibrosis in patients with hepatitis virus infection and nonalcoholic fatty liver disease [8‒10]. Recent studies demonstrate that the FIB-4 index predicts poor outcomes in patients with hepatic diseases [11‒13]. Furthermore, the FIB-4 index is associated with all causes of mortality in non-hepatic diseases, including cancer [14, 15], collagen disease [16], heart diseases [17, 18], diabetes [19], and coronavirus disease 2019 (COVID-19) [20‒22]. However, the clinical impact of the pretransplant FIB-4 index on patient outcomes after CBT remains unexplored. To investigate this, we conducted a retrospective analysis assessing whether the pretransplant FIB-4 index influences mortality and liver-related and unrelated posttransplant complications in adults who underwent CBT in our institution.

Between August 1998 and September 2023, we performed single-unit CBT as the first allogeneic HCT for 336 adult patients at the Institute of Medical Science, the University of Tokyo. All cord blood units were obtained from the cord blood bank in Japan. As a standard practice in our institution, ursodeoxycholic acid (600 mg/day) and a continuous intravenous infusion of unfractionated heparin (100 U/kg/day) were used as VOD/SOS prophylaxis, which started on the day before the conditioning regimen [7]. This retrospective study was approved by the Institutional Review Board of the Institute of Medical Science at the University of Tokyo (2023-83-0131).

Blood samples were obtained within 3 days before the initiation of the conditioning regimen. The FIB-4 index was calculated using the following formula: age (years) × AST (U/L)/(platelet [10⁹/L] × ALT^1/2 [U/L]) [8]. Based on previous studies [10‒12], the FIB-4 index was divided into 3 groups as follows: FIB-4 index <1.3, 1.3 ≤ FIB-4 index ≤2.67, and FIB-4 index >2.67.

Neutrophil recovery was characterized as an absolute neutrophil count exceeding 0.5 × 10⁹/L on the first of three consecutive days. Platelet recovery was characterized as a platelet count exceeding 20 × 10⁹/L on the first of seven consecutive days from the last platelet transfusion. The diagnosis and severity of GVHD were determined using previously established standard criteria [23, 24]. The treating physician diagnosed VOD/SOS using standard criteria [7, 25]. Relapse was defined as morphologic evidence of hematological disease. Non-relapse mortality (NRM) was defined as death during remission. Overall survival (OS) was defined as the duration between the date of CBT and the date of death or most recent contact. The hematopoietic cell transplantation-specific comorbidity index (HCT-CI) [26] and refined disease risk index (DRI) [27] were classified according to published criteria. The number of HLA disparities was defined as a low resolution for HLA-A, -B, and -DR in the graft-versus-host direction.

Baseline characteristics between the three FIB-4 index groups were compared using a χ² or Fisher’s exact test for categorical variables and the Kruskal-Wallis test for continuous variables. The cumulative incidence method, considering competing risks, and Gray’s test were used to analyze hematopoietic recovery, GVHD, VOD/SOS, relapse, and NRM. The Kaplan-Meier method and log-rank test were used to analyze OS. In multivariate analyses, the Fine and Gray proportional hazards model was utilized to estimate hazard ratios (HRs) with a 95% confidence interval (CI) for hematopoietic recovery, GVHD, VOD/SOS, relapse, and NRM, while the Cox proportional hazards regression model was used for OS. Multivariate analyses took the following variables into account: FIB-4 index (<1.3 vs. 1.3–2.67 vs. >2.67), recipient sex (male vs. female), HCT-CI (0–2 vs. ≥3), refined DRI (low/intermediate vs. high/very high), conditioning regimen (TBI 10–12 Gy-based vs. TBI 2–4 Gy-based), cryopreserved cord blood total nucleated cell (TNC) count (<2.5 × 10⁷/kg vs. ≥2.5 × 10⁷/kg), and HLA disparities (0, 1 vs. 2).

The statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for the R 4.2.3 software program (R Foundation for Statistical Computing in Vienna, Austria) [28]. All p values were calculated using a two-sided test, and a p value less than 0.05 was judged statistically significant.

Among the entire cohort, the median pretransplant FIB-4 index was 2.06 (range, 0.02–24.17). The pretransplant FIB-4 index was <1.3 in 116 patients (34.5%), 1.3–2.67 in 86 patients (25.6%), and >2.67 in 134 patients (39.8%). Patient and CBT characteristics between the pretransplant FIB-4 index groups are shown in Table 1. The median age at HCT was 35.5 years, 47 years, and 49.5 years in patients with a FIB-4 index <1.3, 1.3–2.67, and >2.67 (p < 0.001), respectively. The proportion of female recipients was lower in patients with a FIB-4 index <1.3 (p = 0.031). Diagnosis (p < 0.001), refined DRI (p < 0.001), conditioning regimen (p < 0.001), and GVHD prophylaxis (p < 0.001) proportions differed among the pretransplant FIB-4 index groups. The patients with a FIB-4 index >2.7 were more likely to have myelodysplastic syndrome and high or very high risk by refined DRI. A higher proportion of patients with a FIB-4 index >2.7 received a TBI 4 Gy-based regimen, as well as cyclosporine and mycophenolate mofetil-based GVHD prophylaxis.

In univariate analyses, the cumulative incidence of neutrophil recovery significantly differed between the pretransplant FIB-4 index groups (90.9% for FIB-4 index <1.3, 94.2% for FIB-4 index 1.3–2.67, and 83.4% for FIB-4 index >2.67 at 30 days, p = 0.033) (Fig. 1a). The cumulative incidence of platelet recovery was comparable between the pretransplant FIB-4 index groups (p = 0.301) (Fig. 1b). In multivariate analyses, the FIB-4 index was not associated with neutrophil or platelet recovery (Table 2).

In univariate analyses, the cumulative incidence of grades III to IV acute GVHD (p = 0.264), stage ≥1 acute GVHD of the liver (p = 0.384), and VOD/SOS (p = 0.269) was comparable between the pretransplant FIB-4 index groups (Fig. 2a–c). In multivariate analyses, the FIB-4 index was not associated with grades III to IV acute GVHD, stage ≥1 acute GVHD of the liver, or VOD/SOS (Table 3).

In univariate analyses, the cumulative incidence of extensive chronic GVHD (p = 0.685) and chronic GVHD of the liver (p = 0.128) was also comparable between the pretransplant FIB-4 index groups (Fig. 2d, e). In multivariate analyses, the FIB-4 index was not associated with extensive chronic GVHD or chronic GVHD of the liver (Table 3).

In univariate analysis, the cumulative incidence of relapse was comparable between the pretransplant FIB-4 index groups (p = 0.222) (Fig. 3a). In multivariate analysis, the FIB-4 index was not associated with relapse (Table 4).

In univariate analysis, the cumulative incidence of NRM significantly differed between the pretransplant FIB-4 index groups (7.2% for FIB-4 index <1.3, 22.0% for FIB-4 index 1.3–2.67, and 17.1% for FIB-4 index >2.67 at 5 years, p = 0.031) (Fig. 3b). In multivariate analysis, when the FIB-4 index <1.3 group was used as the reference, NRM was significantly higher in the FIB-4 index 1.3–2.67 (HR, 2.51; 95% CI, 1.19–5.30; p = 0.015) and the FIB-4 index >2.67 (HR, 2.34; 95% CI, 1.12–4.90; p = 0.023) groups (Table 4).

In univariate analysis, the probability of OS significantly differed between the pretransplant FIB-4 index groups (77.7% for FIB-4 index <1.3, 62.1% for FIB-4 index 1.3–2.67, and 59.0% for FIB-4 index >2.67 at 5 years, p = 0.007) (Fig. 3c). In multivariate analysis, when the FIB-4 <1.3 group was used as the reference, overall mortality was significantly higher in the FIB-4 index >2.67 group (HR, 1.66; 95% CI, 1.00–2.73; p = 0.046), but with only marginal significance in the FIB-4 index 1.3–2.67 group (HR, 1.59; 95% CI, 0.96–2.64; p = 0.067) (Table 4).

Among 336 patients, 132 died. The patients’ causes of death grouped by the pretransplant FIB-4 index are shown in Table 5. Although the most common cause of death among the pretransplant FIB-4 index groups was relapsed, the proportion of infection and pulmonary complications were more common causes of death in the FIB-4 index 1.3–2.67 and >2.67 groups compared to the FIB-4 index <1.3 group. Hemorrhage was observed only in the FIB-4 index >2.67 group (Table 5).

Several studies suggest the FIB-4 index is associated with inflammatory response [20‒22]. Therefore, we investigated the effect of the pretransplant FIB-4 index on interleukin-6 (IL-6) levels at 2, 4, and 8 weeks after CBT, which we previously evaluated [29]. Among 39 evaluable patients, serum IL-6 levels at 2 weeks were significantly different between the pretransplant FIB-4 index groups (p = 0.045), but not those at 4 weeks (p = 0.436) or 8 weeks (p = 0.371) after CBT (Fig. 4).

This retrospective study evaluated the impacts of the pretransplant FIB-4 index on posttransplantation outcomes in adult recipients of a single CBT. Several previous studies have shown associations between the FIB-4 index and all causes of mortality in non-hepatic diseases, including cancer [14, 15], collagen disease [16], heart diseases [17, 18], diabetes [19], and COVID-19 [20‒22]. However, the clinical impact of the pretransplant FIB-4 index on patient outcomes after allogeneic HCT remains unexplored. In the present study, the pretransplant FIB-4 index significantly affected NRM and OS in adults after a single CBT, but it was not associated with liver-related outcomes, such as GVHD of the liver and VOD/SOS.

Liver-related complications are often encountered after allogeneic HCT [30, 31]. Although it is unclear whether pretransplant liver fibrosis increases the risk of posttransplant liver-related complications, elevated aminotransferase before the conditioning therapy is a risk factor for developing liver-related complications, such as VOD/SOS, and acute GVHD of the liver [32‒34]. Hyperbilirubinemia and liver cirrhosis are also risk factors for VOD/SOS [34]. Liver fibrosis increases liver-related morbidity in patients with chronic liver disease in general population [35]. Therefore, we hypothesized that pretransplant FIB-4 might affect liver-related posttransplant complications, such as GVHD of the liver and VOD/SOS. However, our data did not demonstrate an association between the pretransplant FIB-4 index and liver-related complications, including VOD/SOS and GVHD of the liver. These negative results might be partly due to a small number of occurrences of liver-related complications, which is too small to evaluate the impact of the pretransplant FIB-4 index on these outcomes. Furthermore, in the case of HCT, the platelet count, which is part of the FIB-4 index, is usually affected by the type and stage of the hematological disease. On the other hand, thrombocytopenia is affected by liver fibrosis in people who have chronic liver disease. This difference could also contribute to the different impact of the FIB-4 index on liver-related complications. Therefore, further study is needed to explore a possible correlation between the pretransplant FIB-4 index and liver-related complications after allogeneic HCT.

The FIB-4 index predicts mortality in patients with or without hepatic disease [11‒22], consistent with our results showing that a higher pretransplant FIB-4 index predicted higher mortality in adult patients following CBT. Although the proportion of high-risk and very high-risk ratings by refined DRI was significantly higher in patients with a higher pretransplant FIB-4 index in our cohort (p < 0.001), multivariate analysis did not show the higher pretransplant FIB-4 index was associated with increased risk of relapse. However, a higher pretransplant FIB-4 index was associated with higher NRM, although the pretransplant FIB-4 index was not associated with HCT-CI (p = 0.159). Given that infection and pulmonary complications were more common causes of death in patients with a higher pretransplant FIB-4 index, a higher pretransplant FIB-4 index predicted non-liver-related mortality following CBT. Furthermore, IL-6 at 2 weeks after CBT was higher in patients with a higher pretransplant FIB-4 index, resulting in an association between the FIB-4 index and subsequent inflammatory response [20‒22]. Furthermore, recipient age as a continuous variable is included in the FIB-4 index. Therefore, higher recipient age was significantly associated with a higher pretransplant FIB-4 index in our cohort, which could also affect NRM and OS in this study. These data suggest that the pretransplant FIB-4 index may also reflect the role of the inflammatory response, which could lead to higher NRM in patients following CBT.

Our study demonstrates that the FIB-4 index predicts mortality after a single-unit CBT for adults. However, our study has several limitations. First, we conducted a retrospective analysis at a single institution in Japan with a limited number of patients, the validation cohort was absent, and the study period was long. Therefore, our cohort included heterogeneous patients and transplant backgrounds. Additional research is required to confirm the effect of the FIB-4 index on mortality in adult recipients of CBT. Second, the cutoff value for the FIB-4 index was based on previous studies [10‒12]. The ideal cutoff values may differ depending on the study population, so additional research is required to determine the most effective cutoff values for allogeneic HCT. Third, liver fibrosis was not confirmed by liver biopsy. Recently, liver stiffness measurement (LSM) can be noninvasively assessed by elastography for early detection of VOD/SOS [36]. Unfortunately, there have been no previous data about LSM in CBT, and we could not evaluate the association between LSM and FIB-4 index. Therefore, the extent to which the predictive significance of the FIB-4 index is limited to the liver and can be affected by non-hepatic illnesses remains uncertain.

In summary, our findings indicate that adult patients receiving single-unit unrelated CBT have a significantly increased risk of NRM and shorter OS if their pretransplant FIB-4 score is greater than 2.67. This association might be less pronounced in patients with underlying liver disorders. Further research should assess the correlation between the FIB-4 index and posttransplantation outcomes in adult patients after CBT.

The authors thank all of the physicians and staff at our hospital for their help in this study.

The Institutional Review Board of the Institute of Medical Science, the University of Tokyo, approved this retrospective study (2023-83-0131), and an opt-out consent mechanism is used in this retrospective study.

The authors declare no competing financial interests.

No funding was provided.

T.K. designed the research, collected and analyzed the data, performed the statistical analysis, and wrote the manuscript. M.M.-O. collected the data. S.K., M.I., and S.T. participated in the treatment of the patients and acquired the clinical data. Y.N. contributed to the manuscript’s critical review. All authors approved the final version.

The data that support the findings of this study are not publicly available due to ethical and legal reasons but are available from the corresponding author upon reasonable request.