Patients and methods

Patients

In total, 13 adult patients with SAA, six males and seven females, underwent allogeneic BMT using an alternative donor between May 1999 and February 2004 at the Asan Medical Center (Table 1). Six of these patients were reported previously.³⁰ Median age of the patients was 22 years (range, 15–34). One patient had SAA/PNH syndrome and all other patients had acquired idiopathic SAA. SAA was defined as having any two of absolute neutrophil count (ANC) less than 500/l, platelet count less than 20 000/l, and corrected reticulocyte count less than 1%. Bone marrow biopsy should reveal hypocellularity.³¹ The median interval from diagnosis of SAA to BMT was 10.1 months (range, 1.6–180.1). Nine patients had received systemic immunosuppressive treatment with one cycle of ATGcyclosporine before BMT, while four had not. Three patients had pretransplant neutrophil counts below 200/l. All patients had been transfused before BMT with a median of 12 U (range, 6–41) of red blood cells and a median of 66 U (range, 6–353) of platelets. All patients had 80 or more of Karnofsky performance scale at the time of HCT.

Table 1 - Pre-transplant patient characteristics.

Full table

The selection of donors was based on serological typing for HLA A, B, and DR according to standard techniques.³² Allogeneic BMT was considered if HLA-identical or HLA one-locus mismatched related or unrelated donor was available. If a patient had two or more alternative donor candidates, an HLA C matched donor was preferred. Molecular typing by sequence-specific oligonucleotide probes was performed for the DRB1 loci in three patients (UPNs 312, 330, and 339) and for the A, B, and C loci in two patients (UPNs 330 and 339). One donor was an HLA one-locus mismatched sibling and 12 donors were unrelated volunteers, among whom five were HLA ABCDR-matched, four were HLA ABDR-matched, and three were HLA ABDR-one-locus mismatched with respective recipients (Table 2). Three sex pairs were female-to-male and eight recipient–donor pairs were ABO-incompatible. Patients received a median of 0.68 10⁸ mononuclear cells/kg (range, 0.37–1.46) and a median of 4.31 10⁶ CD34+ cells/kg (range, 1.45–10.90).

Table 2 - Histocompatibility data, preparative regimen, and bone marrow cell dose.

Full table

Preparative regimens

Preparative regimen was Cy-ATG (cyclophosphamide (50 mg/kg/day on days -5 to -2) plus ATG (Atgam^® 30 mg/kg/day on days -4 to -2)) until June 2003, but two (UPNs 120 and 241) of 10 patients during the period had experienced an anaphylactic reaction to ATG prior to BMT and received fludarabine (30 mg/m²/day on days -4 to -2) in place of ATG. Since July 2003, three patients were included into a randomized trial, which was intended to reduce the cyclophosphamide dose of preparative regimen in allogeneic BMT for SAA. The patients received one of two preparative regimens, which were Cy-ATG (cyclophosphamide (50 mg/kg/day on days -5 to -2) plus ATG (Atgam^® 30 mg/kg/day on days -4 to -2)) in one patient (UPN 312) and Cy-Flu-ATG (cyclophosphamide (50 mg/kg/day on days -3 to -2), fludarabine (30 mg/m²/day on days -6 to -2) plus ATG (Atgam^® 30 mg/kg/day on days -4 to -2 or Thymoglobuline^® 3 mg/kg/day on days -4 to -2)) in two patients (UPNs 330 and 339) (Table 2).

Transplantation procedure

All patients were nursed in laminar air flow rooms. Ciprofloxacin and acyclovir were administered for gut decontamination and viral prophylaxis, respectively. Hyper-hydration and mesna were given for the prevention of cyclophosphamide-induced hemorrhagic cystitis. All cellular blood products were leukocyte-depleted and irradiated prior to transfusion. Immunoglobulin (500 mg/kg) was administered intravenously on day -7, every other week until day 120, and monthly until day 180. Patients were treated for prophylaxis of GVHD by administration of cyclosporine (1.5 mg/kg intravenously every 12 h starting on day -1) plus a short course of methotrexate (15 mg/m² intravenously on day 1, and 10 mg/m² intravenously on days 3, 6, and 11). No measure was performed for the prevention of hepatic veno-occlusive disease (VOD). On day 0, non-T-cell depleted marrow from the donor was infused over 3–4 h. Recombinant human granulocyte colony-stimulating factor (rhG-CSF) (450 g) was given intravenously once daily starting on day 0 or day 5. All female patients received oral contraceptives until platelet counts increased over 100 10³/l. Total parenteral nutrition was given if indicated.

Monitoring of the patients

All patients were prospectively monitored for the occurrence of post transplant toxicities, including GVHD, hepatic VOD, infections, and other transplantation-related toxicities. Blood was drawn daily for complete blood counts, including reticulocyte counts. Blood chemistry and electrolytes, including magnesium level, were determined twice weekly, or more frequently if necessary, whereas prothrombin time (PT) and activated partial thromboplastin time (PTT) were measured weekly. Acute and chronic GVHD was diagnosed on the basis of clinical symptoms, laboratory tests, and whenever possible, histopathological findings of the skin, oral mucosa, liver, or gastrointestinal tract,^{33, 34} and was classified according to clinical criteria.^{35, 36} Hepatic VOD was diagnosed in patients having at least two of the following before day 30: (1) hyperbilirubinemia (bilirubin 2.0 mg/dl), (2) painful hepatomegaly, and (3) unexplained weight gain (>2% from baseline), with no other explanation for these signs and symptoms present at the time of diagnosis.³⁷ Severity of VOD was classified as mild, moderate or severe.³⁸ Cytomegalovirus (CMV) infection was monitored weekly using shell vial culture³⁹ until July 1997, thereafter, by both shell vial culture and CMV antigenemia assay;^{40, 41} ganciclovir 5 mg/kg every 12 h was initiated when CMV infection or disease was documented.

Toxicities within 100 days after allogeneic BMT were graded according to NCI Common Terminology Criteria for Adverse Events (CTCAE) v3.0, which classifies each toxicity as grades I through V.

Evaluation of bone marrow engraftment and hematopoietic chimerism

The first day of ANC of 500/l or more for 2 consecutive days was recorded for bone marrow engraftment. The first day of unsupported platelet count of 20 000/l or more for 7 consecutive days and that of reticulocyte count of 1% or more for 3 consecutive days were also recorded.

Hematopoietic chimerism was evaluated in all patients using peripheral blood samples from the donor and the recipient by PCR amplification of short tandem repeats (STRs) or amelogenin loci.⁴² After BMT, recipient peripheral blood samples were drawn monthly for the first 3 months and then every 3 months for additional 1–2 years or until death. A panel of nine paired primers for STR loci including CSF1PO, F13A01, FESFPS, LPL, TPOX, TH01, HPRTB, vWA, and F13B was used in the initial screening process to identify the most informative STR locus in a given donor–recipient pair. Amelogenin displays a 212 base X-specific band and a 218 base Y-specific band, thus allowing discrimination between X and Y-chromosomes. Complete donor chimerism was defined as the presence of only donor type hematopoietic cells after allogeneic BMT. Mixed chimerism was defined as coexistence of both recipient and donor hematopoietic cells after allogeneic BMT. The degree of mixed chimerism was defined as the proportion of recipient cells in a given sample and determined by the proportion of the peak areas corresponding to recipient signals as compared to the sum of peak areas of the donor and recipient signals.

Results

Engraftment data and hematopoietic chimerism

All patients were engrafted on a median of day 21 (range, 15–27). Except one patient (UPN 330) who died of CNS bleeding on day 53, all patients also achieved unsupported platelet count over 20 000/l on a median of day 27 (range, 21–90) and reticulocyte count over 1% on a median of day 32 (range, 22–77) (Table 3). The patients required a median of 12 U of red blood cells (range, 4–24) and a median of 110 U of platelets (range, 70–379). All patients attained stable complete donor chimerism although three patients experienced transient mixed chimerism, which was converted into a complete donor chimerism without any intervention.

Table 3 - Engraftment data and hematopoietic chimerism.

Full table

Post-transplant toxicities

Eight patients (67%) developed acute GVHD: four had grade 1, one had grade 2, and 3 had grade 3 (Table 4). Six patients responded to methylprednisolone therapy and two patients progressed to chronic GVHD. Skin was involved in six patients, gastrointestine in four, and liver in one. Chronic GVHD occurred in eight (67%) of 12 evaluable patients on a median of day 154 (range, 96–206); limited disease in four and extensive disease in four. Hepatic VOD was diagnosed only in one patient (8%) and severity was mild. CMV infection was documented in eight patients (62%) and preemptive ganciclovir treatment was given for 9–105 days (median, 35.5). No patient developed CMV disease. Hemorrhagic cystitis occurred in three patients (23%) on days 25, 58, and 474, respectively. When toxicities within 100 days after BMT were graded by CTCAE v3.0, toxicities of grade III or more developed as follows: pulmonary toxicities in 0 (0%), hepatic in 10 (77%), renal in 1 (8%), coagulation in 1 (8%), cardiac in 0 (0%), gastrointestinal in 5 (39%), infection in 10 (77%), bleeding in 1 (8%), and metabolic in 13 (100%).

Table 4 - Post transplant complications and outcomes.

Full table

Survival

The median follow up duration of surviving patients was 1138 days (range, 118–1553). Overall survival rate was 74.6% (95% confidence interval, 49.5–99.7%) (Figure 1). Three patients died and cause of the deaths was CNS bleeding in one (UPN 330) and chronic GVHD in two (UPNs 272 and 145). In these three patients, there was no evidence of secondary graft failure. Other 10 patients are alive with stable engraftment. They do not require transfusion and have 90 or more of Karnofsky performance scale on last follow-up. Three patients are under treatment for chronic GVHD.

Figure 1.

Overall survival curve.

Full figure and legend (7K)

Discussion

Our results suggest that non-TBI preparative regimen may be sufficient to ensure durable engraftment in alternative donor BMT for SAA. This finding is in contrast to earlier reports, which showed that nonirradiation-containing regimens provided insufficient immunosuppression to prevent rejection among patients receiving unrelated or HLA-mismatched related marrow graft.^{8, 43} Although graft rejection was prevented with TBI doses of 10–14 Gy, GVHD and infections are still frequent barriers to successful BMT.^{6, 9, 44} Promising results have been reported from recent studies to achieve engraftment with less intensive conditioning regimens.^{23, 26, 27, 28, 29} An NMDP-sponsored study showed that a TBI dose of 2 Gy in combination with cyclophosphamide and ATG was sufficient to allow for engraftment.²³ In two small studies, Campath-1H or fludarabine based conditioning regimens enabled alternative donor hematopoietic cells to engraft in pediatric SAA patients.^{27, 29} Successful engraftment and better outcomes with less intensive conditioning regimen in recent studies including ours may be explained in several ways. Advances in molecular HLA-typing method have led to better donor selection. Two large registry data demonstrated that HLA genotypic matching resulted in significantly favorable effects on survival and GVHD.^{22, 45} Recently, most patients received irradiated leukocyte-poor blood products before BMT. As with the HLA-identical sibling BMT, this change in transfusion practice might be critical to decrease the risk of graft rejection in alternative donor BMT for SAA. Interval from diagnosis to BMT is an important prognostic factor for survival after alternative donor BMT for SAA.^{22, 23} In earlier studies, most patients had long duration of disease and failed one or more courses of systemic immunosuppressive therapy before BMT. In contrast, the patients in our study had received no or only one cycle of systemic immunosuppressive therapy. Durable engraftment is likely with alternative donor BMT carried out earlier in the course of SAA. In addition, genetic differences between ethnic groups may explain, at least in part, the durable engraftment of our patients. Lower incidence of GVHD in the Japanese patients was assumed to reflect a lower degree of diversity of HLA and minor histocompatibility antigens among the Japanese compared to the patients of the Western countries.⁴⁶ The frequency of IL-10 592A allele, which is associated with lower incidence of GVHD, is reported to be higher among Asians than the whites.⁴⁷

In our study, incidence of GVHD was relatively high and chronic GVHD was a major factor contributing to two deaths. Major attributable factor to high incidence of GVHD might be higher degree of HLA disparity. The donor selection was based on serological typing of HLA antigens and four patient–donor pairs were phenotypically HLA-one-locus mismatched. With improved donor selection through molecular HLA-typing, patients may have superior outcome after unrelated donor BMT.^{22, 23} The occurrence of GVHD is related to the intensity of preparative regimen.⁴⁸ Although the preparative regimens in our study were less intensive and nonirradiation containing, it is necessary to investigate further refined preparative regimen, such as addition of Campath-1H and omission of cyclophosphamide.

In conclusion, our study suggests that non-TBI containing preparative regimen may be able to ensure durable engraftment in alternative donor BMT for adult patients with SAA.

References

References