|Home | About | Journals | Submit | Contact Us | Français|
Allogeneic hematopoietic cell transplantation (HCT) effectively treats seeveral non-malignant disorders (NMD) such as selected lysosomal disorders, cerebral adrenoleukodystrophy, and hemoglobinopathies. However, rates of graft failure (GF) in non-malignant populations exceed those of patients with malignant indications for HCT. Salvage conditioning regimens and outcomes for second HCT for GF vary immensely in the literature. We report 17 consecutive pediatric patients with NMD who underwent a second allogenic HCT for GF using a non-myeloablative (NMA), low dose busulfan-based regimen. Graft sources for the second transplant included umbilical cord blood, unrelated bone marrow and unrelated peripheral blood stem cells. Median age at time of second HCT was 6.6 years (1.1 – 14.6 years). Fourteen of seventeen patients (82%) achieved engraftment, with a 3-year overall survival of 82% (95% CI, 54–94%). Day 100 transplant related mortality was 12% (95% CI, 0–27%). CMV and adenovirus reactivation occurred in 30% and fungal infections in 18%. The incidence of grade II-IV acute graft-versus-host disease was 35% (95% CI, 13–58%) with only 6% grade III-IV (95% CI, 0–17%). In summary, we illustrate excellent overall survival and acceptable toxicity using a NMA conditioning regimen for second HCT as salvage therapy for first graft failure in patients with non-malignant conditions.
Hematopoietic cell transplant (HCT) is uniquely therapeutic for several non-malignant disorders (NMD) including severe thalassemia syndromes, severe sickle cell disease, Hurler syndrome, and the cerebral form of adrenoleukodystrophy1–4. However, graft failure (GF) remains an important contributor to morbidity and mortality in these patients, occurring at an incidence of 5–30%5–8. GF has been defined in several ways. In general, “primary GF” is used to describe the scenario of insufficient donor derived hematopoiesis by post-transplant day +42. Primary GF can be further stratified into neutropenic GF (NGF, the absence of neutrophil recovery prior to day +42) or non-neutropenic GF (NNGF, predominant autologous hematopoietic recovery prior to day +42)9. In contrast, secondary GF is defined as the loss of donor cells following initial donor engraftment. Autologous recovery is common in patients with non-malignant disorders who develop secondary GF, however late marrow aplasia and pancytopenia may also develop resulting in either NNGF or NGF subtypes of secondary GF, respectively9.
Many factors, both extrinsic and intrinsic to the allograft, may contribute to GF. Those which show the strongest and most consistent correlations comprise of conditioning regimen intensity, the underlying disease, presence of recipient anti-HLA antibodies, drug toxicity, infections, (bacterial, fungal and viral; particularly CMV, human herpes virus type 6 and parvovirus), cell dose and the degree of donor-recipient HLA disparity10–14. Patients with NMD are at notable risk of NNGF. This is particularly true following reduced intensity conditioning regimens, which may be in part due to incomplete immunosuppression as well as incomplete eradication of marrow cells10. Untreated NGF is usually fatal, as patients cannot survive long-term without sufficient hematopoietic and immune recovery.
Currently there are no universally accepted conditioning regimens for subsequent HCT following GF in patients with NMD. Here we describe a non-myeloablative (NMA) conditioning regimen utilizing low dose total body irradiation, low dose busulfan, and the immunosuppressive agent fludarabine for the management of graft failure in patients with non-malignant disorders. In addition, we review the current relevant literature and highlight the outcomes of second HCT for graft failure.
We evaluated 17 consecutive patients with non-malignant disorders who underwent a second allogeneic transplant for graft failure at the University of Minnesota from January 2000 – July 2014 uniformly treated with the protocol described herein. Donor selection was based on the institutional donor selection algorithm, utilizing the best matched donor, and appropriate total nucleated cell dose/kg available. Unrelated donor selection was based on allele-level HLA matching at HLA-A, -B, -C, and -DRB1 based on high-resolution molecular testing. Sibling donors and umbilical cord blood donor HLA selection was based on antigen level typing at HLA-A and -B and allele-level typing at HLA-DRB1. All clinical and laboratory data were obtained from the prospectively maintained institutional Blood and Marrow Transplant Database and supplemented with review of the medical record. The study was approved by the University of Minnesota's Institutional Review Board. Informed consent was obtained from all subjects or their representatives prior to second transplantation.
All patients were categorized according to their type of GF following their first, unsuccessful HCT. Patients were assigned as having NGF if after the first HCT they demonstrated persistent neutropenia (ANC< 0.5×109/L) for ≥ 42 days, or if following initial neutrophil recovery, they experienced subsequent prolonged, persistent aplastic neutropenia. NNGF following the first HCT was defined as autologous hematopoietic recovery, evidenced by donor myeloid chimerism ≤ 10% as measured in CD15 positive cells (after June 2009) or mononuclear fractioned cells (prior to June 2009) anytime. Neutrophilic recovery was defined as the first of three consecutive days following this second transplant regimen with an ANC ≥ 0.5×109/L. Transplant-related mortality was defined as death attributable to complications of HCT, but not from progression of the underlying disease. Acute and chronic graft-versus-host disease (aGVHD) was graded according to standard consensus criteria15–17.
In preparation for their second transplantation, all patients received a uniform NMA conditioning regimen consisting of low dose busulfan, fludarabine and total body irradiation (TBI) as diagrammed in Figure 1. The total dose of busulfan was either 3.2 mg/kg via intravenous (IV) route (patients ≥4 years old), or 4 mg/kg (patients < 4 years old) administered on day -8 and -7, as 8 doses on a 6-hour schedule. Busulfan therapeutic drug monitoring and dose-adjustment was not used on this protocol as there were only 2 days of busulfan. Fludarabine total dose was 200mg/m2 (40 mg/m2/day) given daily on days -6 to day -2; followed by a single fraction of 200 cGy total body irradiation (TBI) on day -1. Anti-seizure prophylaxis was co-administered with busulfan as per the standard institutional protocol. Patients received blood product support and infectious disease prophylaxis (both bacterial and fungal) and viral reactivation monitoring per institutional guidelines.
Patient characteristics and first HCT characteristics are summarized in Table 1. The median age at time of second HCT was 6.6 years (range 1.1 – 14.6 years) and 77% of the patients were male, largely due to the group of 9 patients with X-linked adrenoleukodystrophy (53%). Four patients (23.5%) had Hurler syndrome and the remaining 4 (23.5%) had other disorders, including aplastic anemia, Wolman disease, Diamond-Blackfan anemia, and recessive dystrophic epidermolysis bullosa. Among these patients, 13 (76%) received reduced intensity conditioning with their first HCT, and 4 (24%) received a myeloablative conditioning regimen. Fifty-three percent of the patients developed NGF with their first HCT, while 47% developed NNGF. Characteristics of second HCT for graft failure are summarized in Table 2. Karnofsky performance scores (KPS) were available on 12 patients at the time of second HCT with a median score of 95 (range 30 – 100). The median time from first to second HCT was 85 days with a range of 43 to 334 days.
Umbilical cord blood (UCB) was the graft source in 11 patients (65%), while bone marrow was used in 4 patients, and peripheral blood stem cells (PBSC) was utilized as the cell source in 2 patients. Fifteen patients (88%) received a second HCT from a different donor while 2 patients received the same allograft donor at the discretion of the treating transplant physician. Donors were selected based on the institutional donor selection algorithm, utilizing the best matched donor, and appropriate total nucleated cell dose/kg available. The median time to follow-up from second HCT was 24 months (range 1 – 124 months).
Patient and transplant characteristics were summarized using descriptive statistics. All patients were followed longitudinally until death or last follow up. Kaplan-Meier analysis was used to estimate overall survival (OS). Cumulative incidence functions were calculated for estimates of neutrophil engraftment, grade II–IV acute GVHD, grade III–IV acute GVHD and transplant related mortality (TRM). All statistical analyses used Statistical Analysis System statistical software version 9.3 (SAS Institute Inc., Cary, NC).
Eighty-eight percent of patients (15 of 17) achieved stable donor hematopoiesis (neutrophil engraftment and/or appropriate donor chimerism >10%), with a 3-year overall survival of 82% (95% CI, 54 – 94%) (Figure 2). The cumulative incidence of neutrophil recovery by 42 days after second HCT was 82% (95% CI, 62 – 96%) (Figure 3), with NGF occurring in 3 patients (18%). Median time to neutrophil recovery was 16 days (Interquartile range (IQR), 11 – 21 days). A third allogeneic HCT was performed in 2 of the 3 individuals with failed donor engraftment. Day 100 transplant related mortality (TRM) was 12% (95% CI, 0 – 27%). Two of the 3 deceased patients did not achieve donor engraftment with the 2nd HCT, and the deaths were attributable to bilateral acute organizing pneumonitis with the 2nd HCT and sepsis during a third HCT. Infectious complications following second HCT included cytomegalovirus (CMV) and/or adenovirus viral reactivation in 5 patients (30%) and invasive fungal infections in 3 patients (18%). The cumulative incidence of grade II–IV acute graft-versus-host disease (GvHD) at day+180 was 35% (95% CI, 13 – 58%) with only 6% (95% CI, 0 – 17%) developed grade III–IV GvHD. The incidence of chronic GvHD was 19% (95% CI, 1 – 44%).
Despite appropriate donor selection and intensive conditioning regimens in patients with non-malignant disorders (NMD), primary and secondary GF continue to be a concern. HCT for non-malignant disease is associated with up to 3-fold higher risks of GF, primarily NNGF10. Many patients with NMD are chemotherapy naïve and thus more immunocompetent at initial HCT, possibly leading to immune based GF. While there is a notable risk of NNGF in patients with NMD following RIC regimens, an attempt to reduce TRM, late effects and increase OS has led to rising use of RIC in these patients. In particular, 13 of 17 patients in this cohort initially received RIC conditioning regimens during their first HCT. The majority of these patients had cerebral ALD or other storage disorders which affect the CNS, and in an attempt to reduce the neurotoxicity associated with previous MA regimens, a local RIC regimen was utilized.
Several case series of successful salvage therapy after graft failure employing various chemotherapy regimens and donor sources describe variable survival rates ranging from 31 – 80%18–34 reported in Table 3. In the largest published data set, the CIBMTR reported 122 patients with NGF between the ages of 2 – 61 years (median 30 years), undergoing second HCT after primary graft failure, reporting a low overall survival (OS) rate of 11%, a high 1-year post HCT TRM of 86% (95% CI, 79 – 92), and 1-year relapse rate of 7% (95% CI, 2 – 12)23. Ahmed at el22, described 12 consecutive children with GF after initial myeloablative HCT, undergoing a second HCT using a non-myeloablative conditioning regimen consisting of alemtuzumab (Campath) and fludarabine. Fifty percent of the patients had non-malignant disorders, and the same stem cell donor was used as for the primary graft. Their RIC regimen resulted in achieving neutrophil engraftment and donor chimerism by day 28 in all 12 patients. However, 6 patients received donor lymphocyte infusions (DLI) for progressive loss of donor cell chimerism after initial engraftment, of which 2 developed secondary graft failure (24%). Four of the twelve patients (33%) reactivated CMV and adenovirus, but none were reported to have clinical CMV disease. They concluded that a RIC conditioning regimen with Campath and fludarabine allowed for successful donor stem cell engraftment with a good overall survival and low TRM, although maintenance of donor chimerism required DLI in half of the patients treated22. Similarly, Ayas et al conducted a retrospective analysis of a cohort of 30 pediatric patients with non-malignant disorders who underwent a second HCT for GF24. Myeloablative regimens were used for re-conditioning in 18 patients (10 of whom received irradiation based regimens), 3 patients received RIC, and 9 patients received only serotherapy with ATG. They reported a 5-year OS and event free survival (EFS) of 53% and 47% respectively with a TRM of 30%. An event was defined as graft failure or death of any cause24.
One of the chief concerns when contemplating a second HCT is the choice of the most appropriate preparative regimen. Many regimens used for second HCTs have resulted in poor engraftment and high mortality. Repeat conditioning with fully myeloablative preparative regimens produces unacceptable organ toxicity and risk for overwhelming infections, especially in patients with NGF9. Thus, use of a RIC regimen for second HCT represents an attractive and less toxic means of rescue. By permitting mixed or progressive donor engraftment, RIC HCTs avoid the anticipated prolonged period of aplasia associated with more intensive myeloablative regimens. A major concern with using RIC regimens for salvage HCT has been an increased rate of second GF, which is of particular concern in a patient who has previously experienced GF following first transplantation9. Our NMA conditioning regimen utilizes fludarabine to provide appropriate immunosuppressive conditioning; while combination of low dose busulfan and low dose TBI allows for adequate myeloablation to ensure engraftment. Additionally, lower dose busulfan was used in the hopes to decrease the risk of hepatic sinusoidal obstructive syndrome in this high-risk group of patients. Serotherapy was not included in this regimen for concerns of high rates of viral reactivation and delayed immune reconstitution which contributes to higher TRM especially patients with NGF. Ultimately, 82% (95% CI, 62 – 96%) of our patients achieved stable donor hematopoiesis by 42 days after second transplant and the 3-year overall survival was 82% (95% CI, 54 – 94%) which appears superior to outcomes previously published.
Second HCT has historically been associated with high treatment-related morbidity and mortality at least in part because of the toxicity of the conditioning used. We report TRM of 12% in our cohort, which was attributable to sepsis, and pneumonitis. Of the 3 deceased patients, two patients received MA conditioning regimens with their first HCT and developed NGF. Unfortunately, these 2 individuals also failed to achieve engraftment with the 2nd HCT, thus significantly increasing the period of neutropenia and susceptibility to infections. The remaining 2 patients who received MA conditioning regimens with their first HCT, tolerated our NMA regimen, and achieved successful engraftment and donor chimerism.
In addition to the conditioning regimen, several other variables including the underlying disease, co-morbid conditions, age, timing of second HCT, donor type as well as graft source must be taken into account at the time of considering second HCT. Previous studies23, 26 found no significant difference in outcomes when the same or different donors were utilized as graft sources during second HCT. In our cohort, 14 out of 17 patients received UCB donors during their first HCT, thus resulting in the use of a different donor during the second HCT. Of the remaining 3 patients, the initial donor was available for utilization during the second HCT in 2 patients, while the third patient received a different donor. The optimal timing of the second HCT must also be considered in these patients since a longer period between the two transplants might result in better recovery from the toxic effects of the first transplant and better outcome. Remberger et al26 and Ayas et al24 reported a trend for better survival in patients with more than six months between first and second HCT compared to those with shorter intervals, suggesting there may be greater toxicity when there is a short interval between the two conditioning therapies. However, patients with NGF are likely to proceed to the second HCT sooner due to the risk of infections and transfusion dependence. Comparatively, we report lower TRM and greater OS in our patients even though the median time to 2nd HCT in our cohort was early at 85 days (43 – 334 days), suggesting that use of a NMA conditioning regimen is reasonable for patients early after first HCT. The incidence of grade II–IV acute graft-versus-host disease was 35% with only 6% grade III–IV, which is similar to that reported in the literature23. Three of seventeen patients (18%) in our case series experience a second episode of GF. Two of the 3 patients proceeded to a third HCT, of which one survived with successful donor engraftment. Rates of CMV and adenovirus reactivation in our cohort appears to be similar to that reported in literature22. No patients in this study developed hepatic sinusoidal obstructive syndrome or seizures attributable to busulfan. While the dosing of busulfan was every 6 hours in this study, it is not clear that frequent dosing is necessary. It may be possible to use less frequent dosing to achieve similar outcomes35.
We report the use of our NMA regimen as a feasible choice for patients with non-malignant disorders undergoing a second HCT for prior graft failure that results in very good neutrophil recovery and overall survival while having relatively low TRM.
This work was supported in part by grants from the Children’s Cancer Research Fund and National Cancer Institute P01 CA65493.
Conflicts of Interest
No Author has any conflict of Interest.