Plasma Citrulline Level As a Biochemical Marker to Predict and Diagnose Graft-Versus-Host Disease

Background:

Graft-versus-host disease (GVHD), more especially manifestations involving the gastro-intestinal (GI) tract, remains the major cause of morbidity and mortality after allogeneic stem cell transplantation (allo-SCT). Biochemical tools are needed both to predict and prevent acute GVHD development. In addition, deciphering infectious or toxic diarrhea from early manifestations of GI GVHD can be challenging. Citrulline is an amino acid produced by enterocytes of the small bowel. Plasma citrulline level (PCL) has been reported as a reliable quantitative marker of enterocyte mass and function. In an attempt to assess the impact of intestinal damage on the development of GVHD and further outcomes after allo-SCT, we prospectively monitored PCL in 146 consecutive patients admitted in our unit between January 2013 and May 2014.

Patients and methods:

The study included 93 males and 53 females with a median age of 52 years (range, 18-69), who received an allo-SCT for a hematological malignancy (acute leukemia: n=78; lymphoma: n=20; myelodysplastic syndrome: n=17; myeloproliferative neoplasm: n=12; multiple myeloma: n=10; and others: n=9). Fifty-one patients had a progressive disease. Donors were HLA-identical sibling (n=50) and unrelated (n=96). Source of stem cells was bone marrow (BM; n=76), peripheral blood stem cell (PBSC; n=63) and cord blood (CB; n=7). Conditioning was myeloablative in 88 (60%) patients. Enteral nutrition (EN) was systematically offered and started shortly after conditioning, although parenteral nutrition was provided when EN had been refused or was poorly tolerated. PCL were measured before conditioning (BC), monitored until day 30 after transplantation (d30), and assessed the first day of admission in case of hospitalization for diarrhea.

Results:

After a median follow-up of 220 days (6-564), grade II-IV acute GVHD was observed 31% patients (n=45) (GI-GVHD: n=17; grade III-IV: n=15; corticoresistant GVHD: n=10) and chronic GVHD in 27% patients (n=39) (extensive forms: n=13). The incidences of relapse and non-relapse mortality (NRM) were 18% and 21%, respectively. Mean PCL BC was 28 µmol/l (10-59). After a significant fall at d1, PCL increased slowly until a median of 16 µmol/l (0-52) at d15 and 17 µmol/l (1-66) at d30. PCL <15 µmol/l BC was associated with a higher risk of grade III-IV (p=0.008), GI (p=0.013), corticoresistant (p<0.001) acute GVHD and even chronic GVHD (p=0.003). Although recipient age seemed protective of PCL (p=0.014), progressive disease BC was the only factor predictive of low citrulline BC in multivariate analysis [HR: 1.14; 95%CI 1.06-1.22] (p=0.001). Patients with PCL <15 µmol/l at d15 had more often grade II-IV than the others (49% vs 10%, p=0.002). Similar results were observed at d30 (p<0.001). It is noteworthy that in patients with PCL >15 µmol/l at d15 and/or d30, no grade III-IV, GI or corticoresistant acute GVHD were observed. In addition, the incidence of NRM in these patients was only 2% at d15 (vs 24%, p=0.002) and 3% at d30 (vs 25%, p=0.024). In multivariate analysis, only 2 factors influenced low d15 citrulline: EN refused or poorly tolerated [HR: 1.50; 95%CI 1.20-1.88] (p=0.001) and antithymocyte globulin incorporated in conditioning [HR: 0.65; 95%CI 0.47-0.88] (p=0.001). D15 PCL <15 µmol/l was the only factor associated with grade II-IV acute GVHD [HR: 0.21; 95%CI 0.07-0.62] (p=0.005). Three factors were associated with a better overall survival: d30 PCL >15 µmol/l [HR: 0.20; 95%CI 0.05-0.74] (p=0.016), BM graft vs PBSC/CB [HR: 0.19; 95%CI 0.05-0.79] (p=0.023) and good tolerance of EN [HR: 1.07; 95%CI 1.01-1.14] (p=0.026). Finally, 30 of the 146 patients were re-admitted for diarrhea post-transplant. The final diagnoses were GI-GVHD (n=20), infection (n=9) or functional diarrhea (n=1). In case of confirmed GI-GVHD, mean PCL at admission were lower (5, range 1-9) than for other diagnoses (22, range 8-47). PCL <15 µmol/l was significantly associated with GI-GVHD (p<0.001).

Conclusion:

Monitoring PCL until day 30 post-transplant seems simple and useful in order to identify patients at higher risk of acute GVHD and NRM. Since d15 PCL are correlated to EN tolerance and duration, low PCL could be an indication for starting/prolonging EN in patients who receive allo-SCT. In addition, PCL could be an interesting diagnostic and decision-making tool in case of post-transplant diarrhea.