The Importance of Age, Fludarabine and Total Body Irradiation in the Incidence and Severity of Chronic Renal Failure Following Allogeneic Hematopoietic Stem Cell Transplantation.

Nonmalignant late effects significantly impair the quality of life of long-term survivors following hematopoietic stem cell transplantation (HSCT), the major risk factors being chronic graft-versus-host disease (GVHD) and the use of total body irradiation (TBI) in the preparative regime. There has been a shift towards the use of fractionated, as opposed to single-dose TBI, in an attempt to reduce toxicity. Few reports exist on the renal consequences of HSCT, although many warn about the potential toxicity that eventually leads to chronic renal failure (CRF). However, the exact incidence of HSCT nephropathy is unknown, as there is a lack of well-designed analyses to provide cumulative incidence rates and descriptions of risk factors. The aim of this study was to assess retrospectively the risk and severity of CRF following TBI-based preparative regimes and the contribution of other patient, disease and post-transplant factors. From February 1996 to April 2004, 274 patients had TBI-based preparative regimes for allogeneic HSCT at our institution. Chemotherapy mainly consisted of cyclophosphamide (120 mg/kg) alone or in combination with fludarabine (90 mg/m²). TBI was started the day after completion of chemotherapy and was delivered as a single dose of 7.5 Gy (7.5S), 12 Gy in 6 fractions (12F), or 14.4 Gy in 8 fractions (14.4F). Fludarabine and high-dose TBI (14.4 Gy) were administered preferentially to those patients receiving T-cell depleted (TCD) grafts. GVHD prophylaxis was with a combination of cyclosporin (CSA) + methotrexate in unmanipulated transplants and with CSA alone or nothing in case of TCD grafts. CRF was defined as a persistent increase of serum creatinine (SCr) to greater than 120 μmol/l in males and 97 μmol/l in females (upper limits at our institution) over a period of at least six months. The cumulative incidence of CRF was 11% for the whole group. Renal tissue was available for review in 9 patients. All except one had histologic features suggestive of thrombotic microangiopathy ± radiation nephropathy. Kaplan-Meier analysis revealed a significant association between the incidence of CRF and older age at transplantation (>30 years; P = 0.0043), administration of fludarabine (P = 0.0014), high-dose/unfractionated TBI (14.4F > 7.5S > 12F; P = 0.0005) and TCD (P = 0.0031). Since fludarabine was usually associated to TCD grafts and high-dose TBI, we investigated whether the use of fludarabine was masking the effect of these other factors. Only older age at transplant (HR = 2.97; 95% CI, 1.32–6.68; P = 0.008) and fludarabine administration (HR = 3.22; 95% CI, 1.47–7.04; P = 0.003) remained significant in a multivariate Cox regression analysis. Of note, although TBI dose/fractionation did not fit in the multivariate model, CRF severity was significantly different as 38% of the patients in the 7.5S group currently require or have required permanent dialysis as opposed to 7% in both 12F and 14.4F groups combined. In conclusion, the incidence and severity of CRF post-transplant is associated with age, fludarabine administration and TBI dose/fractionation and, therefore, these factors should be kept in mind in order to prevent this late effect and devise new preparative regimens for HSCT.