Double umbilical cord blood transplantation with reduced intensity conditioning and sirolimus-based...

Double Umbilical Cord Blood Transplantation with ReducedIntensity Conditioning and Sirolimus-Based GVHD Prophylaxis

Corey Cutler, MD, MPH, FRCP(C)1, Kristen Stevenson, MS2, Haesook T. Kim, PhD2, JuliaBrown, PhD3, Sean McDonough, MS1, Maria Herrera, MD, PhD1, Carol Reynolds, PhD1,Deborah Liney1, Grace Kao, MD1, Vincent Ho, MD1, Philippe Armand, MD, PhD1, JohnKoreth, MD, DPhil1, Edwin Alyea, MD1, Bimalangshu R. Dey, MD4, Eyal Attar, MD4, ThomasSpitzer, MD4, Vassiliki A. Boussiotis, MD, PhD3, Jerome Ritz, MD1, Robert Soiffer, MD1,Joseph H. Antin, MD1, and Karen Ballen, MD4

1 Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA2 Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston,MA3 Department of Medicine, Division of Hematology, Oncology and Cancer Biology, Beth IsraelDeaconess Medical Center, Boston, MA4 Department of Medicine, Division of Hematology/Oncology, Massachusetts General Hospital,Boston, MA

AbstractThe main limitations to umbilical cord blood transplantation (UCBT) in adults are delayedengraftment, poor immunological reconstitution and high rates of non-relapse mortality (NRM).Double UCBT (DUCBT) has been used to circumvent the issue of low cell dose, but acute graft-vs.-host disease (GVHD) remains a significant problem. We describe our experience in 32subjects who underwent DUCBT after reduced-intensity conditioning with fludarabine/melphalan/anti-thymocyte globulin and who received sirolimus and tacrolimus to prevent acute GVHD.Engraftment of neutrophils occurred in all patients at a median of 21 days, and plateletengraftment occurred at a median of 42 days. Three subjects had grade II-IV acute GVHD (9.4%)and chronic GVHD occurred in 4 subjects (cumulative incidence 12.5%). No deaths were causedby GVHD and NRM at 100 days was 12.5%. At two years, NRM, progression-free survival (PFS)and overall survival (OS) were 34.4%, 31.2% and 53.1%, respectively. As expected, immunologicreconstitution was slow, but PFS and OS were associated with reconstitution of CD4+ and CD8+

lymphocyte subsets, suggesting that recovery of adaptive immunity is required for prevention ofinfection and relapse after transplantation. In summary, sirolimus and tacrolimus provide excellentGVHD prophylaxis in DUCBT, and this regimen is associated with low NRM after DUCBT.

Corresponding Author: Corey Cutler, MD MPH FRCP(C), Department of Medical Oncology, Dana-Farber Cancer Institute, 44Binney St., Boston, MA 02115, (617) 632-5946, (617) 632-5168, [email protected] Disclosure Statement:The authors report no relevant financial relationshipsAuthorship Statement:Research Design: CC, KBPatient Care: CC, DL, VH, PA, JK, EA, BD, EA, TS, RS, JHA, KBLaboratory Analysis: JB, SM, MH, CR, VAB, JRData Collection: CC, KS, HTK, GKStatistical Analysis: CC, KS, HTKManuscript Preparation: CCManuscript Review: All authors

NIH Public AccessAuthor ManuscriptBone Marrow Transplant. Author manuscript; available in PMC 2011 November 1.

Published in final edited form as:Bone Marrow Transplant. 2011 May ; 46(5): 659–667. doi:10.1038/bmt.2010.192.

NIH

-PA Author Manuscript

NIH


NIH


IntroductionUmbilical cord blood (UCB) transplantation is a viable option for patients requiringallogeneic transplantation when a suitable adult donor is not available. The main limitationto the broader use of UCB transplantation is the small number of hematopoietic progenitorsfound in an UCB unit. As a result of this low number, engraftment is often delayed,immunologic reconstitution is less complete, and treatment-related mortality often exceedsthat seen with traditional adult donors.1,2 Several strategies have been employed toameliorate outcomes after UCB transplantation, including the use of reduced intensityconditioning (RIC), and the simultaneous or sequential transplantation of two UCB units(double UCB transplantation, DUCBT). The engraftment characteristics of DUCBT appearto be similar to that of traditional bone marrow transplantation, with a median time toneutrophil engraftment of approximately 12–23 days.3,4

While UCB transplantation is often associated with a reduction in the incidence and severityof acute GVHD,2 our prior experience and the experience of the Minnesota group hasdemonstrated that GVHD remains a challenging problem after DUCBT.3,4 In adults, the useof cyclosporine and mycophenolate mofetil is associated with a grade II-IV acute GVHDrate of 40%–58%.3,5 We hypothesized that the use of sirolimus and tacrolimus in theDUCBT setting would lead to a reduction in the rate of grade II-IV acute GVHD.

Sirolimus is a potent immunosuppressant that prevents T cell mediated alloimmunitythrough a variety of mechanisms.6 In addition, it may also be immunosuppressive throughits effects on antigen presenting cells.7 We have previously demonstrated that sirolimus isefficacious in preventing acute GVHD after related and unrelated adult donortransplantation.8 The use of sirolimus in UCB transplantation would be particularlyattractive, since Cytomegalovirus (CMV) reactivation after UCB transplantation is common,and sirolimus may have independent suppressive effects on CMV.9

This report details our phase II experience using the immunosuppressive regimen ofsirolimus and tacrolimus after reduced-intensity conditioning and DUCBT.

MethodsThis research protocol was reviewed and approved by the Institutional Review Board of theDana Farber/Harvard Cancer Center. Written informed consent was obtained from allpatients prior to enrollment and participation. The trial was prospectively registered atwww.clinicaltrials.gov (NCT00133367).

PatientsPatients were eligible to participate in this research study if they had no 6/6 or 5/6 HLA-matched related donor or 10/10 matched unrelated donor, or if an unrelated donor was notavailable within the time frame necessary to perform a potentially curative stem celltransplant. Patients were between the ages of 18–65, had an ECOG performance status of 0–2, had adequate measures of hepatic and renal function and met standard transplanteligibility criteria including cardiac ejection fraction greater than 40% and a DLCO greaterthan 50% of predicted.

Malignant disease criteria for entry included acute leukemia in second or subsequentremission or in first remission with adverse cytogenetics or an antecedent hematologicdisorder. Patients with myelodysplastic syndrome were eligible with any WHO subtype.Patients with chronic myeloid leukemia were eligible if they had accelerated or secondstable phase disease, or were intolerant to tyrosine kinase inhibitors. Patients with

Cutler et al.

Bone Marrow Transplant. Author manuscript; available in PMC 2011 November 1.

NIH


NIH


NIH


lymphoma were eligible in second or subsequent complete remission or with chemotherapy-sensitive partial remission. Patients with CLL had Rai stage III or IV disease, a lymphocytedoubling time of <6 months, or with earlier stage disease after disease progression with ≥ 2chemotherapy regimens, while in partial remission.

HLA Typing and Umbilical Cord Unit SelectionUCB units were obtained from a variety of national and international registries. UCB unitshad to meet a minimum combined pre-cryopreservation cell dose of 3.7 × 107 totalnucleated cells (TNC)/kg and each individual unit was required to have a minimum of 1.5 ×107 TNC/kg prior to cryopreservation. Confirmatory HLA typing was performed on all UCBunits prior to transplantation using polymerase chain reaction and sequence-specific primertechnology (One Lambda, Canoga Park, CA). UCB units were required to be a 4/6 match orbetter at the allele level for HLA-A, -B, and -DRβ1 with each other and with the recipient.Typing at HLA-C and -DQ was performed but was not used in the search strategy. Thechoice of UCB units, when multiple units were available, was hierarchically based on ahigher cell dose, greater HLA compatibility and a younger age of the cord blood unit.

Treatment PlanPatients received pre-transplant conditioning therapy with fludarabine (30mg/m2/day) on 6consecutive days (days -8 through -3; total dose 180mg/m2), melphalan (100mg/m2) on day-2, and rabbit antithymocyte globulin (Thymoglobulin, 1.5 mg/kg/day) on 4 alternating days(days -7, -5, -3, -1; total dose 6.0 mg/kg). UCB stem cells were infused on day 0. UCB unitswere thawed according to the methods of Rubinstein et al,10 and administered sequentiallybetween one and six hours apart. The larger of the two units, based on pre-cryopreservationTNC dose, was administered first.

GVHD prophylaxis began on day -3 and consisted of continuous intravenous infusiontacrolimus (target serum level 5–10 ng/ml) and a 12 mg oral loading dose of sirolimus.Sirolimus was then dosed orally once daily to maintain a serum trough level of 3–12 ng/ml,as previously described.8 Tacrolimus was given orally prior to discharge. In the absence ofGVHD, both GVHD prophylaxis agents were tapered from day 100 through 180, or earlierat the discretion of the treating physician. GVHD was graded according to the consensuscriteria.11

After transplantation, patients received supportive care with transfusion support. Bacterialprophylaxis consisted of levofloxacin until neutrophil engraftment. Antifungal prophylaxisconsisted of fluconazole until day +100. All patients received acyclovir as antiviralprophylaxis. Prophylaxis against Pneumocystis jirovecii was started upon hospital discharge.Monitoring for reactivation of CMV was performed weekly, while monitoring for Epstein-Barr Virus (EBV) and Human Herpesvirus-6 (HHV-6) reactivation were performed everyother week. Pre-emptive therapy against CMV was administered upon viral reactivation.

Filgrastim (5 mcg/kg/day) was administered from day +5 until the absolute neutrophil countwas greater than 2.0 × 109 cells/L for two consecutive days. Donor chimerism wasmonitored according to previously described methodology.3 Chimerism was measuredroutinely at 30 and 100 days, and then regularly through the first two years aftertransplantation.

Immunologic Reconstitution MonitoringWhole blood was collected in EDTA containing lavender top tubes (Kendall Vacutainer,Mansfield, MA). The blood samples were obtained from patients immediately prior totransplantation and at 4, 8, 12, 26, and 52 weeks after transplantation. Leukocyte

Cutler et al.


NIH


NIH


NIH


populations and specific lymphocyte subpopulations were analyzed by flow cytometry usingfluorescence-conjugated monoclonal antibodies directed against lineage-specific cell surfacemarkers. Flow cytometry was performed using either a FC500 (Beckman Coulter) or a BDFACSCanto (BD Biosciences) flow cytometer and data was analyzed using either BeckmanCoulter CXP software or BD FACSDiva software.

Thymopoiesis was measured via analysis of T cell receptor excision circles (TRECs). DNAwas isolated from peripheral blood mononuclear cells (PBMC) using the QIAamp DNAMini Kit (Qiagen, Valencia, CA.), and the DNA concentration determined by standardspectrophotometry. The quantitation of signal-joint TCR excision circle (sjTREC) DNA wasperformed using Taqman real-time PCR following a previously described method using aRotor-Gene 6000 thermal cycler (Corbett Life Science, Sydney, Austrialia).12 Quantitationof the sjTREC copy number for each patient sample was performed using a standard curveprepared with 10-fold dilutions of a plasmid containing the sjTREC sequence (kindlyprovided by Dr. D. Douek, NIH, Bethesda, MD).

Statistical Design and AnalysisThe study was designed as a single stage phase II trial. A sample size of 32 patients wasplanned with the hypothesis that the rate of GVHD would be less than 50%. With thisdesign, the probability of concluding the rate is acceptable is 0.84 if the true underlying rateof developing grade II-IV acute GVHD was 40% and 0,09 if the true rate was 60%, using anexact binomial distribution.

Patients were enrolled between September 2005 and November 2007. The dataset foranalysis was closed on March 1, 2009 and analyzed thereafter. Survival data were updatedto March 1, 2010. Patient baseline characteristics were reported descriptively. Neutrophilengraftment was defined as the first of three consecutive days with neutrophil recovery to atleast 0.5 × 109 cells/L. Platelet engraftment was defined as the first day of a platelet count ofat least 20 × 109 cells/L, without supporting transfusion in the prior 3 days.

Cumulative incidence curves for acute GVHD and chronic GVHD were constructedreflecting early death and death or relapse as competing risks, respectively. Cumulativeincidence curves for non-relapse mortality (NRM) and relapse with or without death wereconstructed reflecting time to relapse and time to NRM as competing risks. Time to relapseand time to NRM were measured from the date of stem cell infusion. Overall survival (OS)was defined as the time from transplant to death from any cause, while progression-freesurvival (PFS) was defined as the time from transplant to progression or death from anycause. Surviving patients were censored at their date of last known follow-up. OS and PFSestimates were calculated using the method of Kaplan and Meier.13 Univariable andmultivariable Cox regression analyses were performed for OS and PFS. Cellular phenotypedata were included as log10 transformed time-varying covariates in the Cox model. All p-values are based on two-sided tests, and were computed using SAS v9.2 software (SASInstitute, Cary, North Carolina).

ResultsThirty-two sequential eligible subjects underwent DUCBT and are all included in thisanalysis. Patient and umbilical cord blood characteristics are shown in Table 1. The medianage was 53 years, while the median weight was 75.9 kg. The majority of patients hadadvanced lymphoid malignancies.

Cutler et al.


NIH


NIH


NIH


Umbilical Cord Blood Units and EngraftmentFull cellular characteristics of the UCB units are available for 30 of the 32 recipients. Themedian single UCB size was 2.43 × 107 TNC/kg prior to cryopreservation (range 1.51–3.94). The first infused unit had a median cell count of 2.67 × 107 TNC/kg while the smallersecond infused unit had a median cell count of 2.33 × 107 TNC/kg prior to cryopreservation.When combined, the median total cell dose administered to recipients was 5.16 × 107 TNC/kg prior to cryopreservation (range 3.66–7.58). The median number of CD34+ progenitorsprior to cryopreservation in the individual UCB units was 0.9 × 105 CD34+/kg (n = 29 and25 for the two units, range 0.2–3.4). When combined, the median dose of CD34+ progenitorswas 1.9 × 105 CD34+/kg (range 0.5–4.1).

The cumulative incidence of neutrophil engraftment was 100%. Recipients attained anabsolute neutrophil count of 0.5 × 109/L at a median of 21 days from transplantation (range13–70 days) (Figure 1). There were no cases of primary graft failure, however, 3 subjectshad late graft loss. All three of these patients eventually had a recurrence of theirmalignancy, although graft rejection occurred in the absence of obvious malignant relapse.Three subjects did not attain platelet engraftment and transfusion independence. Overall, themedian time to platelet engraftment was 42 days (range 25–162 days) (Figure 1). Twenty-three subjects attained a platelet count of 100 × 109/L and the median time to attain thisplatelet count was 89 days (range 26–209 days).

When stratified into two groups based on the combined TNC dose above or below a cutoffof 5.0 × 107 TNC/kg, selected based on the median, there was no relationship between TNCand the time to neutrophil or platelet engraftment (p=0.32 and p=0.25 respectively).Similarly, there was no relationship with the combined CD34+ cell dose (cutoff of 2.0 × 105

CD34+/kg) administered and neutrophil or platelet engraftment (p=0.55 and p=0.91,respectively).

Graft-vs.-Host DiseaseThree subjects experienced grade II-IV acute GVHD at a median of 21 days fromtransplantation. Two subjects had isolated skin stage 3 GVHD (overall grade II GVHD)while the third subject had both skin and hepatic stage 3 disease (overall grade III GVHD).The cumulative incidence of acute GVHD at 100 days was 9.4% (Standard Error, SE =5.2%). There were no deaths attributable to acute GVHD.

Four subjects were diagnosed with chronic GVHD at a median of 212.5 days fromtransplantation. The cumulative incidence of chronic GVHD at 1 year was 12.5% (SE6.0%). All 4 subjects had cutaneous involvement, and one subject had additional oralinvolvement and bronchiolitis obliterans.

Treatment-Related Toxicity and MortalityThe conditioning regimen was well tolerated. There were no cases of idiopathic pneumoniasyndrome, thrombotic microangiopathy, or veno-occlusive disease of the liver. One subjectdeveloped massive proteinuria requiring discontinuation of sirolimus, and two othersdeveloped an acneiform rash related to sirolimus not requiring discontinuation. CMVdisease occurred in 1 subject despite a DNA-based monitoring and pre-emptive treatmentstrategy. Four patients developed meningoencephalitis, with the causative agent beingHHV-6 and EBV in 2 subjects each. EBV reactivation led to the occurrence of post-transplantation lymphoproliferative disease (PTLD) in 5 individuals; all of these caseseventually were fatal.

Cutler et al.


NIH


NIH


NIH


Non-relapse mortality at 100 days from transplantation was 12.5% (SE 5.9%). At 2 years,NRM was 34.4% (SE 8.6%) (Figure 2).

ChimerismChimerism measurements at day 100 (+/-30 days) were assessed to determine the relativecontribution to hematopoiesis in 29 surviving patients. At day 100, single umbilical corddominance (defined as > 80% contribution to hematopoiesis) was noted in 18 of 29 patients(62%), with both cords contributing to hematopoiesis in 8 subjects. Three subjects had noevidence of cord engraftment at day 100 despite having evidence of donor chimerism atearlier time points. Of the 18 subjects with single cord predominance at day 100, 12 subjectshad single cord hematopoiesis only, with no evidence of contribution by the other unit. Fiveof these 12 subjects had evidence of single unit early graft rejection or loss, as there wasnever any evidence of dual chimerism. Of the 18 subjects with single unit predominance atday 100, the first infused cord was responsible for the majority of hematopoiesis in 11(61%), while the second infused cord blood unit was responsible in the remaining 7 (39%).Thus, there was no statistical evidence that the initial cord infused was more likely tocontribute to hematopoiesis in this limited sample (90%CI for the likelihood of equalcontribution of two cord units 29–71%).

Progression-Free and Overall SurvivalThe median follow-up of survivors is 48 months (range 30–62). To date, 12 subjects haverelapsed. Seven of these patients had advanced lymphoid malignancies at the time oftransplantation; however 5 of these subjects remain alive. The median time to relapse aftertransplantation was 12.6 months (range 2.7–30.0), and the cumulative incidence of relapsewas 34.4% (SE 8.6%) at two years (Figure 2). Progression-free survival (PFS) at 2 years is31.2% (SE 8.2%), and overall survival (OS) at 2 years is 53.1% (SE 8.8%) (Figure 3).Causes of death are summarized in Table 2.

In a univariate analysis of factors related to outcome, age ≥ 50 years at the time oftransplantation was associated with inferior overall survival, but not progression-freesurvival (HR 3.37, 95%CI 1.14–9.88, p=0.03 for OS, HR 1.76, 95% CI 0.75–4.16, p=0.19for PFS). In a multivariable model adjusted for age, sex, risk group, and HLA matching, ageretained significance for overall survival (HR 7.42, 95%CI 2.13–25.83, p=0.001) and nowattained statistical significance for progression-free survival (HR 2.87, 95%CI 1.11–7.40,p=0.03). The combination of two 4/6 HLA-matches was associated with inferior overallsurvival (HR 4.05, 95%CI 1.37–12.05, p=0.01), but was not significant in the adjustedmodel for progression-free survival when compared to other HLA combinations (HR 1.94,95%CI 0.79–4.74, p=0.15) (Table 3). Disease risk was unassociated with overall andprogression-free survival outcomes in univariate analysis. Adding total nucleated cell dose(dichotomized by the cutoff of 5.0 × 107 TNC/kg) or CD34+ cell dose (dichotomized by thecutoff of 2.0 × 105 CD34+/kg) did not meaningfully change the hazard ratios for overallsurvival, and themselves were not associated with outcome (N=30, HR 1.03, 95%CI 0.31–3.46 p=0.96 for TNC; N=24, 1.74, 95%CI 0.56–5.43 p=0.34 for CD34+ cell dose).

Immunologic ReconstitutionThe reconstitution of lymphocyte and monocyte populations was measured by flowcytometry in 27 subjects (Figure 4). The recovering CD4+ T cell population had apredominantly memory phenotype (CD45RO+) and the naïve CD4+ population (CD45RA+)did not begin to recover until 6 months after transplantation (Figure 4b). Restoration ofnormal median values for both CD4+CD45RA+ and CD4+CD45RO+ T cells was notobserved until we examined 6 patients who survived 2 years from transplantation (data notshown). Similarly, CD4+ regulatory T cells (CD4+CD25+) recovered very slowly and

Cutler et al.


NIH


NIH


NIH


remained at low levels in peripheral blood throughout the first year after transplantation.CD8+ T cells also recovered gradually and were also predominately of memory phenotype(CD45RO+) in the first 3 months after transplantation (Figure 4c). CD8+ T cells with a naïvephenotype (CD45RA+) began to recover 3–6 months after transplantation. In contrast toCD4+ T cells, the median values of CD8+CD45RA+ and CD8+CD45RO+ T cells reached thelower limits of normal values by 1 year after transplantation. Commencing at 6 months andcontinuing through 1 year after transplantation, there was a dramatic increase in CD20+ Bcells resulting in the restoration of normal cell numbers. CD56+16+ NK cells maintainedtheir pre-transplantation level throughout the course of the study, at values slightly belowthe normal range, until 1 year after transplantation when they reached normal levels (Figure4d). The median value for CD14+ monocytes doubled their pre-transplantation level at 1month after transplantation before falling to near pre-transplantation values at 8 weeks andremained steady up to 12 months after transplantation. The CD14+ monocyte populationremained at or above the upper limit of the normal control range throughout our study(Figure 4e).

In accordance with our analysis of CD4+ and CD8+ T cell numbers, median TREC valueswere below the limit of detection until 12 weeks after transplantation. We saw a substantialincrease in TREC copy number at 6 months (median number 117 copies/μg DNA), one year(median number 2136 copies/μg DNA) and two years after transplantation (median number6749 copies/μg DNA). This increase in TREC copy number paralleled the increase inCD4+CD45RA+ naïve T cell numbers from 6 months to 12 months after transplantation andthe concordant plateau in CD4+CD45RO+ memory cells.

In a time-varying covariate analysis examining reconstitution of cellular immune parametersand outcomes, several parameters were associated with both PFS and OS. Total lymphocytecount and lymphocyte subsets (CD3+, CD4+, CD4+CD45RO+, CD8+CD45RA+, andCD8+CD45RO+) were all associated with improved OS. The maximal TREC value at 2years (≥2,000 vs. <2,000 copies/μg DNA) achieved per patient (N=22), was also associatedwith improved OS, in a non-time varying covariate model. Only recovery of CD56+CD16+

NK cells and maximal TREC copy number were associated with improved PFS in univariatetime-varying and non-time varying univariate models, respectively.

In multivariable models adjusted for age, HLA matching and risk group examiningreconstitution of time-varying cellular immune parameters and outcome, several parameterswere associated with both PFS and OS. Total lymphocyte count and lymphocyte subsets(CD3+, CD4+, CD4+CD45RA+, CD4+CD45RO+, CD8+ and CD8+CD45RA+) were allassociated with improved OS. The maximal TREC value at 2 years (≥2,000 vs. <2,000copies/μg DNA) achieved per patient (N=22), was also associated with improved OS, in anon-time varying covariate model. Only recovery of CD56+CD16+ NK cells and maximalTREC copy number at 2 years were associated with improved PFS in univariate time-varying and non-time varying univariate models, respectively (Table 4).

DiscussionHere we report favorable GVHD outcomes after reduced-intensity double umbilical cordblood stem cell transplantation when sirolimus and tacrolimus are used as GVHDprophylaxis. GVHD rates were very low, and in addition, we demonstrate early engraftmentand low 100 day NRM. At two years, overall survival was over 50% in a cohort of patientswith high-risk malignancies, including a high proportion of patients with resistant lymphoidmalignancies.

Cutler et al.


NIH


NIH


NIH


The most striking difference in this experience in comparison with prior DUCBT studies isthe rate of acute GVHD. Using an identical reduced-intensity conditioning regimen andumbilical cord blood selection algorithm, we previously reported a 40% incidence of gradeII-IV acute GVHD when cyclosporine and mycophenolate mofetil were used as GVHDprophylaxis.3 When comparing cohorts, this reduction was statistically significant (p=0.035)and not due to an excess in the former series, since the rate of 40% was similar to the largerMinnesota series, in which the rate of grade II-IV acute GVHD was reported to be as high as58% in a sample size of 185 indivduals,5 although HLA matching was only at the antigenlevel. In that analysis, when compared with single UCB transplantation, the use of two UCBunits, omission of ATG in the preparative regimen and reduced-intensity conditioning wererisk factors for the development of acute GVHD. Similar risk factors for acute GVHD werefound in a retrospective database review by Eurocord-Netcord focusing on patients withlymphoid malignancies, where univariate analysis demonstrated an association betweenrecipient age, total body irradiation use, lack of ATG use and reduced intensity conditioningas risks for acute GVHD.14 While the risk of acute GVHD was increased in patientsreceiving double in comparison to single UCB transplantation (32 vs. 22%), this result wasnot statistically significant.

The reduction in acute GVHD is likely related to the introduction of sirolimus andtacrolimus as GVHD prophylaxis. Used extensively at our center since 2000, our previousresults in adult peripheral blood stem cell transplantation have demonstrated impressivereductions in grade II-IV and grade III-IV acute GVHD in the related and unrelated donorsettings.8 While associated with an increase in the relative risk of treatment-relatedmorbidity associated with thrombotic microangiopathy15 and veno-occlusive disease,16

these apparent increases were not noted in this experience, as much of this increased risk isthought to be related to the interplay between myeloablative conditioning regimens andendothelial injury associated with sirolimus.

Twelve subjects relapsed, 7 of whom had lymphoid malignancies, leading to a cumulativeincidence of relapse at two years of 34.4%. This is similar to a recently described series of65 patients with low or intermediate grade lymphoid malignancies who received reducedintensity conditioning followed by single (14%) or double (86%) UCB transplants, in whichthe cumulative incidence of relapse at 3 years was 42%, with longer follow-up than in ourseries.17 We have previously described an effect in relapse prevention when sirolimus isused following reduced-intensity conditioning and transplantation of adult stem cells inpatients with lymphoid malignancies.18 It is well known that the mTOR inhibitors possessindependent anti-tumor activity in lymphoid malignancies,19 and therefore the postulate thatmaintenance immunotherapy with these agents after transplantation can both prevent GVHDand relapse is attractive, since other attempts to prevent GVHD (such as T cell depletion)generally increase the risk of relapse.20 The utility of sirolimus to prevent relapse in the cordblood setting remains unknown as small samples sizes and differences in tumorcharacteristics preclude direct comparison of relapse rates between cohorts.

In this experience, we noted a higher than expected occurrence of reactivation of EBV withsubsequent EBV-related PTLD (crude incidence 15.6%). This rate of EBV-related disease isin line with other large retrospective reviews of the literature, where the incidence of EBV-PTLD was noted to be the highest in patients who underwent non-myeloablativeconditioning with ATG as part of the preparative regimen.21 Since sirolimus use isassociated with a reduced incidence of reactivation of Cytomegalovirus in adult stem cell9and less clearly in renal transplantation,22,23 we hypothesized that the administration ofsirolimus as primary GVHD prophylaxis would result in a reduced incidence of EBVreactivation and PTLD. Some of the mechanisms behind sirolimus’ inhibitory function inEBV disease have been elucidated, and include the inhibition of IL-10 secretion by

Cutler et al.


NIH


NIH


NIH


transformed tumor cells, a necessary autocrine signal for tumor growth.24,25 Presumably, inthe absence of a memory T cell response against EBV antigens, and in the presence of otherimmunosuppressants, a permissive environment for EBV-transformed lymphocytes wasestablished and PTLD ensued. This underscores the need for the development of noveltherapeutics for the treatment of established PTLD, since our high incidence oftransformation to aggressive PTLD occurred despite alternate weekly DNA-based EBVmonitoring, in contrast to other reports.26 In addition, others have reported favorableoutcomes with the early administration of rituximab,26 whereas all patients treated with thisagent in our series had progressive disease, and some individuals had a disease course thatwas too fulminant to permit the administration of this drug. Moving forward, we plan onreducing the dose of ATG in the preparative regimen in an attempt to reduce the rate ofPTLD. An alternative strategy would be to administer rituximab to all transplant recipientsearly after transplantation.

In this analysis we were unable to demonstrate a correlation between pre-transplantationgraft characteristics and overall and progression-free survival outcomes. However, severalparameters associated with post-transplantation graft recovery were significantly associatedwith improved outcome. For example, improved CD3+ lymphocyte recovery (of both CD4+

and CD8+ subsets) was associated with an improvement in overall survival, while measuresof T cell neogenesis and thymopoiesis such as TREC measurements were also correlatedwith an improvement in progression-free survival.

Recovery of thymopoiesis as determined by TREC analysis was delayed and incompletecompared to recipients of adult stem cell grafts, who reach normal TREC numbers within 3–12 months after transplantation.27,28 Quantitative TREC recovery was also inferior to thatobserved in pediatric recipients of cord blood allografts.29,30 The differences in the kineticsand the quantitative recovery of thymopoiesis in these patient groups may be related to theregenerative properties of umbilical cord blood, residual thymic activity in pediatric patients,the use of two umbilical cord blood units, or the immunomodulatory effects of our GVHDprophylaxis regimen. There is reason to suspect that either or both of the latter twoexplanations may be responsible, since our patient group displayed an improved thymicregeneration profile compared to a previously reported group of adult recipients of singleumbilical cord blood grafts, who had a nearly universal lack of thymic recovery for 12months after transplantation31. Regardless of the reasons contributing to this difference, ourdata provides evidence that functional thymic regeneration can successfully recover aftercord blood transplantation in adults and represents a critical parameter that governsprogression-free survival. The recovery of innate immunity after transplant may also havebeen improved, as the recovering NK cell population remained relatively constant aftertransplantation, and these cells may have anti-tumor activity after umbilical cord bloodtransplantation.32

In summary, we demonstrate that double umbilical cord blood transplantation in adultrecipients using sirolimus and tacrolimus as GVHD prophylaxis is effective therapy inadvanced hematologic malignancies. The rates of acute and chronic GVHD using thisGVHD prophylaxis regimen are low, and the treatment-related morbidity and mortality isalso acceptably low. Immunologic reconstitution after double umbilical cord bloodtransplantation is related to overall outcome, related to either fewer opportunistic infections,or potentially even to enhanced graft-vs.-tumor effects.

AcknowledgmentsSupported by NCI P01 CA142106, NCI R01 CA123855 and the Ted and Eileen Pasquarello Research Fund.Funded in part by unrestricted research grants from Astellas, Inc and Genzyme, Inc. CC is supported by the StemCell Cyclists of the Pan Mass Challenge.

Cutler et al.


NIH


NIH


NIH


Reference List1. Laughlin MJ, Eapen M, Rubinstein P, et al. Outcomes after Transplantation of Cord Blood or Bone

Marrow from Unrelated Donors in Adults with Leukemia. N Engl J Med. 2004; 351:2265–2275.[PubMed: 15564543]

2. Rocha V, Labopin M, Sanz G, et al. Transplants of Umbilical-Cord Blood or Bone Marrow fromUnrelated Donors in Adults with Acute Leukemia. N Engl J Med. 2004; 351:2276–2285. [PubMed:15564544]

3. Ballen KK, Spitzer TR, Yeap BY, et al. Double unrelated reduced-intensity umbilical cord bloodtransplantation in adults. Biol Blood Marrow Transplant. 2007; 13:82–89. [PubMed: 17222756]

4. Brunstein CG, Barker JN, Weisdorf DJ, et al. Umbilical cord blood transplantation afternonmyeloablative conditioning: impact on transplantation outcomes in 110 adults with hematologicdisease. Blood. 2007; 110:3064–3070. [PubMed: 17569820]

5. MacMillan ML, Weisdorf DJ, Brunstein CG, et al. Acute graft-versus-host disease after unrelateddonor umbilical cord blood transplantation: analysis of risk factors. Blood. 2009; 113:2410–2415.[PubMed: 18997171]

6. Sehgal SN. Sirolimus: its discovery, biological properties, and mechanism of action. TransplantProc. 2003:S7–S14.

7. Wang GY, Chen GH, Li H, et al. Rapamycin-treated mature dendritic cells have a unique cytokinesecretion profile and impaired allostimulatory capacity. Transpl Int. 2009

8. Cutler C, Li S, Ho VT, et al. Extended follow-up of methotrexate-free immunosuppression usingsirolimus and tacrolimus in related and unrelated donor peripheral blood stem cell transplantation.Blood. 2007; 109:3108–3114. [PubMed: 17138818]

9. Marty FM, Bryar J, Browne SK, et al. Sirolimus-based graft-versus-host disease prophylaxisprotects against cytomegalovirus reactivation after allogeneic hematopoietic stem celltransplantation: a cohort analysis. Blood. 2007; 110:490–500. [PubMed: 17392502]

10. Rubinstein P, Dobrila L, Rosenfield RE, et al. Processing and cryopreservation of placental/umbilical cord blood for unrelated bone marrow reconstitution. Proc Natl Acad Sci U S A. 1995;92:10119–10122. [PubMed: 7479737]

11. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on Acute GVHD Grading.Bone Marrow Transplant. 1995; 15:825–828. [PubMed: 7581076]

12. Douek DC, McFarland RD, Keiser PH, et al. Changes in thymic function with age and during thetreatment of HIV infection. Nature. 1998; 396:690–695. [PubMed: 9872319]

13. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc.1958; 53:457–481.

14. Rodrigues CA, Sanz G, Brunstein CG, et al. Analysis of Risk Factors for Outcomes AfterUnrelated Cord Blood Transplantation in Adults With Lymphoid Malignancies: A Study by theEurocord-Netcord and Lymphoma Working Party of the European Group for Blood and MarrowTransplantation. J Clin Oncol. 2009; 27:256–263. [PubMed: 19064984]

15. Cutler C, Henry N, Li S, et al. Sirolimus and thrombotic microangiopathy after allogeneic stem celltransplantation. Biol Blood Marrow Transplant. 2005; 11s1:4. Abstract.

16. Cutler C, Stevenson K, Kim HT, et al. Sirolimus is associated with veno-occlusive disease of theliver after myeloablative allogeneic stem cell transplantation. Blood. 2008; 112:4425–4431.[PubMed: 18776081]

17. Brunstein CG, Cantero S, Cao Q, et al. Promising progression-free survival for patients low andintermediate grade lymphoid malignancies after nonmyeloablative umbilical cord bloodtransplantation. Biol Blood Marrow Transplant. 2009; 15:214–222. [PubMed: 19167681]

18. Armand P, Gannamaneni S, Kim HT, et al. Improved survival in lymphoma patients receivingsirolimus for graft-versus-host disease prophylaxis after allogeneic hematopoietic stem-celltransplantation with reduced-intensity conditioning. J Clin Oncol. 2008; 26:5767–5774. [PubMed:19001324]

19. Costa LJ. Aspects of mTOR biology and the use of mTOR inhibitors in non-Hodgkin’s lymphoma.Cancer Treat Rev. 2007; 33:78–84. [PubMed: 17161912]

Cutler et al.


NIH


NIH


NIH


20. Ho VT, Soiffer RJ. The history and future of T-cell depletion as graft-versus-host diseaseprophylaxis for allogeneic hematopoietic stem cell transplantation. Blood. 2001; 98:3192–3204.[PubMed: 11719354]

21. Brunstein CG, Weisdorf DJ, DeFor T, et al. Marked increased risk of Epstein-Barr virus-relatedcomplications with the addition of antithymocyte globulin to a nonmyeloablative conditioningprior to unrelated umbilical cord blood transplantation. Blood. 2006; 108:2874–2880. [PubMed:16804113]

22. Mathew T, Kreis H, Friend P. Two-year incidence of malignancy in sirolimus-treated renaltransplant recipients: results from five multicenter studies. Clin Transplant. 2004; 18:446–449.[PubMed: 15233824]

23. Kirk AD, Cherikh WS, Ring M, et al. Dissociation of depletional induction and posttransplantlymphoproliferative disease in kidney recipients treated with alemtuzumab. Am J Transplant.2007; 7:2619–2625. [PubMed: 17868060]

24. Krams SM, Martinez OM. Epstein-Barr virus, rapamycin, and host immune responses. CurrentOpinion in Organ Transplantation. 2008:13.

25. Nepomuceno RR, Balatoni CE, Natkunam Y, et al. Rapamycin Inhibits the Interleukin 10 SignalTransduction Pathway and the Growth of Epstein Barr Virus B-cell Lymphomas. Cancer Res.2003; 63:4472–4480. [PubMed: 12907620]

26. Kinch A, Öberg G, Arvidson J, et al. Post-transplant lymphoproliferative disease and otherEpstein-Barr virus diseases in allogeneic haematopoietic stem cell transplantation afterintroduction of monitoring of viral load by polymerase chain reaction. Scandinavian Journal ofInfectious Diseases. 2007; 39:235–244. [PubMed: 17366054]

27. Hochberg EP, Chillemi AC, Wu CJ, et al. Quantitation of T-cell neogenesis in vivo after allogeneicbone marrow transplantation in adults. Blood. 2001; 98:1116–1121. [PubMed: 11493459]

28. Lewin SR, Heller G, Zhang L, et al. Direct evidence for new T-cell generation by patients aftereither T-cell-depleted or unmodified allogeneic hematopoietic stem cell transplantations. Blood.2002; 100:2235–2242. [PubMed: 12200390]

29. Talvensaari K, Clave E, Douay C, et al. A broad T-cell repertoire diversity and an efficient thymicfunction indicate a favorable long-term immune reconstitution after cord blood stem celltransplantation. Blood. 2002; 99:1458–1464. [PubMed: 11830500]

30. Weinberg K, Blazar BR, Wagner JE, et al. Factors affecting thymic function after allogeneichematopoietic stem cell transplantation. Blood. 2001; 97:1458–1466. [PubMed: 11222394]

31. Komanduri KV, St John LS, de Lima M, et al. Delayed immune reconstitution after cord bloodtransplantation is characterized by impaired thymopoiesis and late memory T-cell skewing. Blood.2007; 110:4543–4551. [PubMed: 17671230]

32. Beziat V, Nguyen S, Lapusan S, et al. Fully functional NK cells after unrelated cord bloodtransplantation. Leukemia. 2009; 23:721–728. [PubMed: 19151772]

Cutler et al.


NIH


NIH


NIH


Figure 1.Cumulative Incidence of Neutrophil and Platelet Engraftment.

Cutler et al.


NIH


NIH


NIH


Figure 2.Cumulative Incidence of Treatment-Related Mortality and Relapse.

Cutler et al.


NIH


NIH


NIH


Figure 3.Kaplan-Meier Estimates of Progression-Free and Overall Survival

Cutler et al.


NIH


NIH


NIH


Figure 4.a–e. Immunologic Reconstitution After Transplantationa) Lymphocyte Subsets b) CD4+ Lymphocyte Subsets c) CD8+ Lymphocyte Subsets d) Blymphocyte and NK Subsets d) Monocyte Subsets. Median values are plotted on a log10scale and zero values are set equal to one. Error bars extend to the 25th and 75th percentiles.

Cutler et al.


NIH


NIH


NIH


NIH


NIH


NIH


Cutler et al.

Table 1

Subject and Graft Characteristics

Sample Size 32

Age (years) 53 (19–67)

Weight (kg) 75.9 (52.3–112.9)

Gender–Male 16 (50%)

Diagnosis

Non-Hodgkin Lymphoma 10

Acute Myelogenous Leukemia 9

Hodgkin Disease 5

Myelodysplastic Syndrome 4

Chronic Lymphocytic Leukemia 1

Acute Lymphoblastic Leukemia 1

Chronic Myelogenous Leukemia 1

Myeloproliferative Disorder 1

Prior Transplantation 9

HLA Match (A, B, DR)

4/6, 4/6 16

4/6, 5/6 13

5/6, 5/6 2

5/6, 6/6 1

HLA Match (A, B, C, DR)

4/8, 4/8 1

4/8, 5/8 2

4/8, 6/8 1

5/8, 5/8 10

5/8, 6/8 9

6/8, 6/8 1

6/8, 7/8 1

6/8, 8/8 1

7/8, 7/8 2

Missing data 4

Umbilical Cord Blood Cell Doses Infused

Cord Unit 1, TNC/kg, pre-thaw, median × 107 (min, max) 2.67 (1.87–3.91)

Cord Unit 2, TNC/kg, pre-thaw, median × 107 (min, max) 2.33 (1.51–3.94)

Combined Cord units TNC/kg, pre-thaw, median × 107 (min, max) 5.16 (3.66–7.58)

Cord Unit 1, CD34+/kg, pre-thaw, median × 106 (min, max) 0.09 (0.02–0.34)

Cord Unit 2, CD34+/kg, pre-thaw, median × 106 (min, max) 0.06 (0.03–0.20)

Combined Cord units CD34+/kg, pre-thaw, median × 106 (min, max) 0.16 (0.02–0.41)


NIH


NIH


NIH


Cutler et al.

HLA = Human Leukocyte Antigen

TNC = Total Nucleated Cell


NIH


NIH


NIH


Cutler et al.

Table 2

Causes of Death

Relapse 6

PTLD 5

Sepsis/MSOF 5

Donor-Derived Leukemia 1

Trauma 1

PTLD = Post-transplant Lymphoprolferative Disorder

MSOF = Multisystem organ failure


NIH


NIH


NIH


Cutler et al.

Table 3

Multivariable Analysis of Clinical Factors Related to Overall Survival and Progression-Free Survival

Variable OS HR (95% CI) p-value PFS HR (95% CI) p-value

Age (≥50 vs. <50) 7.42 (2.13–25.83) p = 0.001 2.87 (1.11–7.40) p = 0.03

Sex (Female vs. Male) 2.29 (0.74–7.10) p = 0.15 1.79 (0.73–4.39) p = 0.21

Disease Risk Group (High vs. Low) 3.08 (0.83–11.39) p = 0.09 2.73 (0.94–8.00) p = 0.07

HLA Match (4/6, 4/6 vs. Others) 4.05 (1.37–12.05) p = 0.01 1.94 (0.79–4.74) p = 0.15

Disease Risk Groups: Low risk disease includes patients with AML or ALL in 1st remission, CML in chronic phase and MDS RA or RARSsubtypes. All others are considered High Risk.

OS = Overall Survival

PFS = Progression-Free Survival

aGVHD = Acute Graft-vs.-Host Disease

TNC = Total Nucleated Cell

N= Sample Size


NIH


NIH


NIH


Cutler et al.

Table 4

Multivariable Modeling of Outcomes with a Time Varying Covariate Adjusted for Age, HLA match, andDisease Risk Group.

Variable N Overall Survival Progression-Free Survival

HR (95% CI) p-value HR (95% CI) p-value

White Blood Cell Count 27 1.16 (0.02–61.09) p = 0.94 3.31 (0.17–63.30) p = 0.43

Absolute Lymphocyte Count 27 0.18 (0.02–1.35) p = 0.10 1.04 (0.22–5.05) p = 0.96

CD45+ 27 0.68 (0.06–8.09) p = 0.76 2.12 (0.28–16.02) p = 0.47

CD3+ 27 0.22 (0.06–0.87) p = 0.03 0.81 (0.34–1.92) p = 0.63

CD4+ 27 0.20 (0.05–0.81) p = 0.02 0.56 (0.22–1.46) p = 0.23

CD4+25+ 27 0.22 (0.04–1.33) p = 0.10 0.31 (0.08–1.15) p = 0.08

CD4+45RA+ 27 0.26 (0.07–0.94) p = 0.04 0.66 (0.30–1.45) p = 0.30

CD4+45RO+ 27 0.15 (0.04–0.68) p = 0.01 0.51 (0.19–1.41) p = 0.19

CD8+ 27 0.38 (0.14–0.99) p = 0.05 0.94 (0.48–1.84) p = 0.85

CD8+45RA+ 27 0.29 (0.10–0.84) p = 0.02 0.87 (0.45–1.70) p = 0.68

CD8+45RO+ 27 0.44 (0.17–1.17) p = 0.10 1.15 (0.57–2.32) p = 0.69

CD19+ 27 0.99 (0.41–2.40) p = 0.98 1.41 (0.64–3.13) p = 0.40

CD20+ 27 0.95 (0.43–2.13) p = 0.90 1.35 (0.64–2.87) p = 0.44

CD56+16+ 27 0.73 (0.22–2.45) p = 0.61 0.43 (0.18–0.99) p = 0.05

CD14+ 27 1.13 (0.12–10.60) p = 0.92 1.82 (0.32–10.29) p = 0.50

Max TREC 2 yr (≥2000 vs. <2000)† 22 0.10 (0.01–1.48) p = 0.09 0.08 (0.01–0.53) p = 0.01

Note: Each row represents a separate multivariable analysis with adjustments for age, HLA match and disease risk group.

TREC = T cell Receptor Excision Circle

†Not time-varying.


Double umbilical cord blood transplantation with reduced intensity conditioning and sirolimus-based...

Documents

Transcript of Double umbilical cord blood transplantation with reduced intensity conditioning and sirolimus-based...