We performed a retrospective analysis of all patients undergoing ALOTH in our program from 2012 to 2018. The transplant database and electronic medical records were reviewed to obtain epidemiological, clinical and laboratory characteristics, survival times, and complications. The main objective was to compare the results obtained with HAPLO and SIB modalities in terms of engraftment, transplant-related mortality, morbidity, the response obtained, relapse, the incidence of graft-vs-host disease (GVHD), and global and event-free survival. The ethical committee of our institution approved this retrospective analysis.

The patients who were candidates for ALOTH were evaluated with spirometry and pulmonary diffusion (DLCO), echocardiography and adrenal, thyroid, parathyroid, and gestational analyses. All patients were tested for herpes virus, hepatotropic virus, retrovirus, measles, rubella, mumps, toxoplasma, and parasitic stool. The disease status prior to ALOTH was determined with bone marrow sampling to classify leukemia cases into complete remission when there was an absence of immature cells by cytology and flow cytometry; partial remission, when there was the presence of immature cells, but to a lesser extent than at diagnosis; and refractory disease or relapse, when there were more than 5% blast immature cells. In lymphoma patients, complete remission was established by PET-CT, without evidence of active disease within a period not exceeding one month prior to transplantation. In patients with aplastic anemia or non-transformed myelodysplasia, ALOTH was performed with active disease. The donor selection was performed with HLA A, B, C, and DR genotyping in intermediate or high resolution, depending on the type of transplant; in the cases of haploidentical transplantation, donor selection was complemented with specific donor anti-HLA antibodies. In patients who have anti-HLA antibodies, our program performs desensitization, as reported, with intravenous immunoglobulin, rituximab, plasmapheresis, and autologous stem cell collection for the hematopoietic recovery, in the case of engraftment failure.6

The patients were treated with the myeloablative conditioning (MA) and reduced intensity (AIR) scheme, according to our previously reported data.7 In summary, our program has no chronological age limit for ALOTH with conditioning chemotherapy, adjusted according to the functional status of the patient. The myeloablative scheme consisted of intravenous (IV) cyclophosphamide 60mg/kg each day for 2 days and 1320cGy of total body irradiation fractionated in 3 or 4 days. Reduced intensity regimen for lymphoma patients consisted of cyclophosphamide 50mg/kg as a single dose, fludarabine IV 40mg/m2 every day for 5 days, and total body irradiation 200cg as a single dose. For leukemia or myelodysplasia, patients were given cyclophosphamide EV 50mg/kg as a single dose, fludarabine EV 40mg/m2 every day for 5 days, and busulfan EV 3.2mg/kg for 2 days. In patients with medullary aplasia, the same MA or AIR schemes were used along with thymoglobulin.

In patients who received SIB, the GVHD prophylaxis was methotrexate EV 15mg/m2 day +1; 10mg/m2 on days 3, 6, 9, and 12, and cyclosporin 3mg/kg every 12h IV or oral route (OR) for serum levels between 200 and 400ng/μL. In patients with HAPLO, the prophylaxis of GVHD was tacrolimus 0.03mg/kg EV or VO, mycophenolate 1g every 8h VO, and cyclophosphamide post-transplant 50mg/kg on days +3 and +4, with an equivalent dose of mesna. Immunosuppressive treatment with tacrolimus or cyclosporine was maintained for the necessary time, according to the evolution of GVHD.

Once the patient was on the day before ALOTH, the hematopoietic precursor collection from the donor was performed, after stimulation with 600μg of filgrastim subcutaneous administration (sc)/day for 5 days, followed by the apheresis process to obtain a dose of at least 2×106CD34/kg. In this analysis, there was no bone marrow collection from any donor and no collection product was cryopreserved.

During the aplasia, patients received transfusions to maintain platelets >20,000/μL, hemoglobin >8g/L%, and filgrastim 300μg/day until achieving neutrophil counts >500/μL. The preventive antimicrobial treatment consisted of levofloxacin 500mg daily, acyclovir 250mg EV, or 800mg orally every 12h, and fluconazole 400mg daily. Febrile neutropenia was treated according to the institutional guidelines. When fever presented, blood and urine culture and chest X-ray were performed immediately, and an antibiotic treatment was started with third-generation cephalosporins, aminoglycoside and vancomycin. With the persistence of fever for 72h, a computed tomography (CT) of paranasal cavities, thorax, abdomen, and pelvis was performed and blood fungal infection tests were performed. The adjustment of antimicrobial therapy was decided according to the radiological or blood test results. In the case of cytomegalovirus reactivation, preventive therapy with ganciclovir or foscarnet was initiated.

All patients received opioid analgesic treatment in case of grade 2–4 severe mucositis with parenteral nutrition, as needed.

The GVHD was defined according to the time of appearance and development of the clinical characteristics. Thus, patients with GVHD in the first 100 days post-transplant were classified as acute GVHD and those with manifestations after that period were classified as chronic GVHD. Acute and chronic GVHD were graded using standard criteria.8,9 Overall survival was defined as the time elapsed between the day of the transplant and the date of death or the date of the last control. The progression-free survival was defined as the time elapsed between the day of the transplant and the date of relapse. Relapse was defined as evidence of recurrence of the underlying disease. Non-relapse mortality (NRM) was defined as any cause of death not associated with recurrence of the disease, including those associated with complications directly related to the transplant.

The Shapiro-Wilk test evaluated the normality of the quantitative variables. The demographic and baseline characteristics were presented using mean, percentage and ranges. The comparisons between the variables were made with the Chi-square method. The SG probabilities were estimated with the Kaplan–Meier method and the comparisons, with the log-rank test. The relapse, NRM, and GVHD analyses were performed in a competitive risk framework, using the non-parametric cumulative incidence estimator. The effect of the events that occurred during the follow-up and after transplantation, such as acute GVHD, was analyzed as a time-dependent covariate. The software SPSS.V.15 (IBM Software, USA) and Prisma Software V 6.0.1 (GraphPad Software, USA) were used for analysis. The differences were considered significant for values of p<0.05, with 95% confidence intervals (CIs).

Table 1 shows the characteristics of the patients. During the analysis period, 91 patients were transplanted, 49 with HAPLO and 42 with SIB, with a mean age of 34 and 32 years, respectively, and a mean follow-up of 36 months (range 2–72 months). Two patients had anti-HLA antibodies specific to the donor (4%). A total of 40% of the SIB patients and 48% of the HAPLO patients had reactivation of the cytomegalovirus (CMV) (p=0.8), but there was no disease or mortality associated with this viral infection. In 8 patients, 4 in each group, the presence of polyomavirus (BK virus) was detected in the urine, which required adjustment of immunosuppression and symptomatic treatment, but none of the patients required the use of Cidofovir or renal replacement therapy.

As shown in Table 2, there were no differences regarding the day of granulocyte or platelet graft arrest. The incidence at 30 days of hematopoietic recovery was 93% (95% CI 78–93) in HAPLO and 95% (95% CI 91–95) in SIB; p=0.9. The incidence of platelet recovery at 6 months was 88% (95% CI 79–93) and 93% (95% CI 91–94) in HAPLO and SIB (p=0.15), respectively.

No significant differences were detected in transplant-related mortality, as shown in Tables 2 and 3. Similarly, mortality due to relapse did not differ significantly (HR 0.5814, CI 95% 0.2800–1.207). In two patients with a specific anti-donor antibody, it was necessary to perform desensitization; one survived with a normal functional graft and the other died due to sepsis associated with prolonged neutropenia.

As shown in Tables 2 and 3, the incidence of mild (grades 1 and 2) and severe (grades 3 and 4) acute and chronic GVHD was not different in the two groups of patients. Cumulative incidence adjusted by competitive risks for global acute and chronic in HAPLO was 34% and 43%, respectively, and for SIB, 41% and 29%, respectively (p=0.9%).

As shown in Table 2 and Figure 1, the overall survival estimated at 3 years in HAPLO was 48%, 95% CI 42–67, and in SIB, 50%, 95% CI 43–72 (p=0.5), without significant differences in the two groups when adjusting the analysis by incidence of GVHD, relapse, or conditioning intensity. Table 3 shows global survival and disease-free survival analysis, adjusted by conditioning and GVHD, but no significant differences were observed.

Figure 1.

Overall survival.

(0.04MB).

As shown in Table 3, there were no significant differences among the patient causes of death. Thus, 13 patients in the HAPLO group died, 61% due to relapse and 39% due to mortality related to transplantation; 11 patients died in the SIB group, 60% due to relapse, 35% due to mortality related to transplantation and 5% due to other causes. Patients who died due to HAPLO transplant-related mortality were by sepsis3 and cerebral hemorrhage,1 and in the SIB group, by sepsis2 and multiorgan failure.1