If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Unrelated umbilical cord blood transplants in adults: Early recoveryof neutrophils by supportive co-transplantation of a low number of highlypurified peripheral blood CD34+ cells from an HLA-haploidentical donor
Offprint requests to: Manuel N. Fernández, M.D., Ph.D., Professor of Medicine, Universidad Autómoma de Madrid, Head of the Servicio de Hematologı́a y Hemoterapia, Hospital Universitario “Clı́nica Puerta de Hierro,” C/ San Martı́n de Porres 4, 28035-Madrid, Spain
Objective, Methods, and Results. To reduce the period of posttransplant neutropenia and related early morbidity and mortality of cord blood (CB) transplants, we assessed the feasibility of co-infusion of a low number of highly purified peripheral blood CD34+ cells from a related haploidentical donor with a CB graft. Between March 1999 and May 2002, 11 patients with high-risk hematologic malignancies were transplanted using this strategy. The seven patients who received a haploidentical peripheral blood graft and a CB graft from a sibling (6) or the father (1) had prompt recovery (9–17 days, median 10) of the absolute neutrophil count (ANC) to greater than 0.5×109/L. Analysis of DNA polymorphisms showed initial predominance of the haploidentical genotype both in granulocytes and in mononuclear cells, and subsequent progressive replacement by cells of CB genotype until final complete CB chimerism was achieved by patients who survived for sufficient periods of time. The four patients who received maternal haploidentical cells had no significant contribution of these to blood leukocytes, although complete CB chimerism was achieved by three of them and two reached engraftment of the CB on days +20 and +36. Morbidity due to early bacterial or fungal infections was remarkably low in patients with prompt ANC recovery.
Conclusion. Our data show that co-infusion of a CB unit and a low number of haploidentical CD34+ cells may result in a shortened period of posttransplant neutropenia. This is likely the result of prompt and transient engraftment of the haploidentical hematopoietic stem cells that may provide the patient antimicrobial protection until the later engraftment of the CBhematopoietic stem cells.
Unrelated cord blood transplant (UCBT) is now a well-established modality of allogeneic hematopoietic stem cell (HSC) transplantation. Compared to unrelated voluntary donor transplant (UVDT), it has the advantages of rapid availability, lower risk of transmission of viral diseases, and lower incidence and severity of graft-vs-host host disease (GVHD) despite HLA mismatches, but the disadvantage of higher early transplant-related mortality (TRM) [
Successful transplantation of HLA-matched and HLA-mismatched umbilical cord blood from unrelated donors: Analysis of engraftment and acute graft-vs-host disease.
Cell dose and speed of engraftment in placental/umbilical cord blood transplantation: graft progenitor cell content is a better predictor than nucleated cell quantity.
Survival after transplantation of unrelated donor umbilical cord blood is comparable to that of human leukocyte antigen-matched unrelated donor bone marrow.
Comparison of unrelated cord blood and unrelated bone marrow transplants for leukemia in children: A collaborative study of the New York Blood Center and the International Bone Marrow Transplant Registry.
]. This is in part related to late engraftment due to the relatively low number and primitive biological features of the transplanted HSC, which implies risk of serious early neutropenia-related infections that may result in death prior to engraftment. Toxic mucositis induced by the conditioning regimen is an added risk factor for these early infections, mainly in previously heavily treated patients. All of this represents a major limitation to the use of UCBT for patients of high body weight. Strategies that are being investigated to reduce duration of posttransplant neutropenia of UCBT include in vitro culture expansion of an aliquot of the transplanted cord blood (CB) unit and transplantation of multiple units, neither of which have proved clinically efficient to date [
Cord blood transplants: early recovery of neutrophils from co-transplanted sibling haploidentical progenitor cells and lack of engraftment of cultured cord blood cells, as ascertained by analysis of DNA polymorphisms.
Prompt and durable engraftment in two adult patients with high risk chronic myelogenous leukemia (CML) using ex-vivo expanded and unmanipulated unrelated umbilical cord blood.
]. With the same objective we are assaying co-transplantation of a single CB unit and a limited number of highly purified CD34+ peripheral blood (PB) mobilized cells from a haploidentical donor, together with conditioning regimens of attenuated toxicity. Here we report results obtained in 11 adults, three of whom were part of a previous study [
Cord blood transplants: early recovery of neutrophils from co-transplanted sibling haploidentical progenitor cells and lack of engraftment of cultured cord blood cells, as ascertained by analysis of DNA polymorphisms.
The dual transplants of an unrelated CB unit and PB CD34+ haploidentical cells were performed in accordance with a Phase II protocol that was approved by the Institution Ethics and Clinical Investigation Review Board.
Eligibility
Patients were eligible for enrollment if they were older than 15 years, had high-risk acute leukemia, and did not have a readily available HLA-identical related or unrelated donor or CB unit providing more than 4.0×107 total nucleated cells (TNC) per kilogram of patient body weight, as determined before freezing. Written informed consent signed by the patient or legal guardian was required. Eleven patients were entered in the protocol from March 1999 until May 2002: 9 males and 2 females, ages 16 to 47 years (median 23), weighing 53 to 85 kg (median 66) (Table 1). All had poor-risk acute leukemia with one or more of the following adverse features: partial or unstable remission (PR, UR), second or third complete remission (CR2, CR3), mixed leukemia or Ph+ acute lymphoblastic leukemia.
Table 1Patients' pretransplant and conditioning regimens data
Figures in this column under codes of each component of the preparative regimen indicate total doses given: in cGy for TBI, mg/kg for cytoxan (CTX), busulfan (Bus), and ALG, and mg/m2 for fludarabine (FL).
S-1
M
40
70
ALL, CR2
TBI 12 + CTX 120 + ALG 75
S-2
M
28
66
ALL, PR
TBI 12 + CTX 120 + ALG75
S-3
M
47
80
AML(M7), PR
TBI 12 + CTX 120 + ALG 90
S-4
M
22
58
ALL-Ph1+, CR3
FL 120 + TBI 10 + CTX 120 + ALG 30
S-5
M
21
64
ALL/AML, CR2
FL 120 + BUS 8 + CTX 120 + ALG 90
S-6
F
33
60
ALL-Ph1+, CR1
FL 120 + TBI 10 + CTX 120 + ALG 90
M-1
M
29
70
ALL-Ph1+, CR1
TBI 12 + CTX 120 + ALG 75
M-2
M
18
60
ALL, CR2
FL 120 + TBI 10 + CTX 120
M-3
M
23
67
ALL-Ph+, CR3
FL 120 + TBI 10 + CTX 120
M-4
M
18
85
CML, CP2
FL 120 + TBI 10 + CTX 120 + ALG 30
F-1
F
16
53
ALL-Ph1+, UR
FL 120 + TBI 10 + CTX 120 + ALG 30
ALL and AML = acute lymphoblastic and acute myeloblastic leukemia; CML = chronic myeloid leukemia; CP = chronic phase.
Other abbreviations as in the text.
∗ median age, 23 years; range 16–47.
† median weight, 66 kg; range 53–85.
‡ Figures in this column under codes of each component of the preparative regimen indicate total doses given: in cGy for TBI, mg/kg for cytoxan (CTX), busulfan (Bus), and ALG, and mg/m2 for fludarabine (FL).
CB units were sought in accredited CB banks using the patient HLA phenotype as determined by serologic typing for Class I HLA-A and HLA-B antigens and by DNA typing for Class II HLA alleles. The same methods were used to select the HLA-haploidentical donors. High-resolution HLA-DRβ1 typing was used for confirmatory typing of the eligible CB units. The CB units were selected primarily on the basis of HLA match and TNC as determined before freezing. Minimum requirements were a match of 3/6 HLA loci and TNC of 1.2×107 per kilogram of recipient body weight. ABO compatibility and serology for cytomegalovirus (CMV) were used as secondary criteria. Cross-match of the CB cells with the recipient serum was not part of the selection criteria because of the usual unavailability of these cells for this purpose. The CB units used for the 11 patients were provided by the CB Banks of Dusseldorf, Milan, and London (two each), and New York, St. Louis, Besançon, Barcelona, and Madrid-12 Octubre (one each).
The donor of the haploidentical CD34+ cells was sought among the first-degree relatives of the patient. Candidates with ABO compatibility or minor incompatibility were preferred to those with major ABO incompatibility, and siblings were usually preferred to parents or offspring. Cytotoxic cross-match was done with serum of the recipient and the haploidentical cells; candidates with a negative test were preferred. The protocol design did not include evaluation of killer-cell immunoglobulin-like receptor (KIR) compatibility.
Preparative treatment and GVHD prophylaxis
Pretransplant preparative treatment is adjusted to patient condition and previous treatments. Conditioning regimens included combinations of either a) cyclophosphamide (CTX) with total-body irradiation (TBI), b) busulfan and equine antithymocyte globulin (ALG), c) fludarabine, CTX, ALG, and lower doses of TBI, or d) busulfan. For prophylaxis against GVHD, patients received cyclosporine (CsA), first intravenously and then orally, aiming at a serum concentration of 180 to 240 μg/L, and low-dose (1 mg/kg) prednisone, both initiated 3 to 5 days prior to the transplant. Unless the patient developed manifestations of GVHD that advised otherwise, tapering of prednisone was initiated once sustained CB engraftment was documented and CsA was tapered from day +100. Four of the 11 patients in the study (S-1, S-2, S-3, and M-1) received a preparative regimen consisting of TBI, 12 cGy given in six 2-cGy fractions over three days (days −7 to −5) with lungs shielded at 8 cGy, CTX 120 mg/kg for two days (−4 to −3), and 30 to 75 mg ALG (linfoglobulina, IMTIX-SANGSTAT, Barcelona, Spain) in the last three days of the preparative program. The other patients received a regimen consisting of fludarabine 90 mg/m2/day given over three days, CTX 120 mg/kg for two days, TBI 10 cGy given in five 2-cGy fractions (lungs shielded at 8 cGy), and 30 to 90 mg/kg ALG in the final 1 to 2 days. Busulfan 8 mg/kg was substituted for TBI in one patient (S-5) previously irradiated. ALG was omitted in two other patients (M-2 and M-3) for whom the administration of this long-acting immunosuppressive agent was deemed undesirable because of high risk of recurrence of the leukemic disease or previous infectious problems or both.
Transplantation procedure and supportive care
The selected CB units were shipped from the collecting bank in liquid nitrogen. The haploidentical CD34+ cells were obtainedfrom the selected donor by granulocyte colony-stimulating factor(G-CSF) mobilization, apheresis, and positive selection, and frozen as previously described [
Cord blood transplants: early recovery of neutrophils from co-transplanted sibling haploidentical progenitor cells and lack of engraftment of cultured cord blood cells, as ascertained by analysis of DNA polymorphisms.
]. The fraction to be infused was prepared to contain less than 104 CD3+ cells and no more than 2.6×106 CD34+ cells per kilogram of recipient body weight. In both cases, the frozen cells were maintained in liquid nitrogen in our bank until the time of infusion. Also in both cases, thawing was done in a 37°C water bath.
For the CB units, prior to infusion the thawed cells were diluted in an equal volume of a sterile solution of 10% dextran 40 (Baxter, Glendale, CA, USA) supplemented with human albumin up to a concentration of 5%. The CD34+ haploidentical cells were immediately infused without previous dilution. In both cases, a small aliquot of the resuspended cells was used for tests for TNC, CD34+ and CD3+ cell quantification, cell viability (exclusion of trypan blue dye), and assays both for colony-forming units (CFUs) and for bacterial and fungal cultures.
Patients were nursed in single air-filtered rooms in the Hematology Transplantation Unit under the usual strict hygienic conditions. They received irradiated and leukocyte-depleted transfusions of platelets and red cells as required to maintain hemoglobin and platelets at appropriate levels to prevent and treat symptoms due to anemia or thrombocytopenia. When oral alimentation was insufficient, they also received parenteral nutrition. Filgrastim (Amgen, Thousand Oaks, CA, USA) was given at a dose of 5 to 10 μg per kilogram of patient body weight per day, from day +1 until sustained recovery of absolute neutrophil counts (ANC) was achieved. As prophylaxis for infections, patients received sterile food, oral cyprofloxacin, and intravenous fluconazol. A microbiological surveillance program that included regular cultures and tests for CMV (pp65 antigenemia and PCR) was continued until hospital discharge. To treat infections, other antimicrobials were given intravenously or orally as required for documented agents, or according to our usual protocol for empirical antimicrobial therapy for fever of undetermined origin in neutropenic or immunocompromisedpatients.
Characteristics of the transplants
Table 2 shows the ABO groups and HLA typing data of patients, the CB units, and the haploidentical donors. Recipient and CB were of the same ABO group in four cases and seven patients had ABO incompatibilities: major in five, minor in four, both major and minor in two. HLA broad mismatches, rejection-wise, between recipients and the transplanted CB occurred at 0 to 2 (median 1) of the 6 tested loci: four patients had no mismatches, three had one, and four had two. In terms of GVHD, mismatches between CB units and recipients were 0 to 3 (median 1): two patients had no mismatches, five had one, three had two, and one had three. Donors of the haploidentical cells were the father for one patient (F-1), the mother for four (M-1 to M-4), and a sibling (brother or sister) for six (S-1 to S-6). All of these carried the maternal HLA antigens not inherited by the recipient (NIMA), which was the rule for maternal donors. Between recipients and the haploidentical donors there were no cases of major ABO incompatibility, and HLA broad mismatches were 2 to 3, both for rejection and for GVHD. In cytotoxicity cross-match tests, there were no positive cases. The haploidentical cells were mismatched with the CB units received by the patient at 1 to 4 loci (median 3), both regarding rejection of the haploidentical cells by CB and vice versa.
Table 2ABO groups and HLA typing of patients and grafts
Cellularity of the grafts is shown in Table 3. The infused TNC of the transplanted CB units ranged from 1.31 to 3.7×107/kg (median 2.01) and the CD34+ cell counts prior to cryopreservation, available for seven units, ranged from 0.035 to 0.37×106/kg (median 0.11). Total CFU counts (CFU-GM + BFU-E + CFU-Mix) in the infused CB products ranged from 0.35 to 6.49×104/kg (median 1.81). Infused haploidentical products contained 1.05 to 2.54×106/kg CD34+ cells (median 2.28), all with CD3+ cells less than 104/kg (range 0.05–0.7×104, median 0.22).
Posttransplant tracing of the grafts and engraftment
Aside from karyotype and in situ hybridization for the Y chromosome when these procedures were applicable, chimerism was evaluated by polymerase chain reaction (PCR) analysis of microsatellite DNA markers and by reference strand conformation analysis (RSCA) analysis of HLA differentiating polymorphisms as previously described [
Cord blood transplants: early recovery of neutrophils from co-transplanted sibling haploidentical progenitor cells and lack of engraftment of cultured cord blood cells, as ascertained by analysis of DNA polymorphisms.
]. The RSCA method was used also to evaluate quantitatively the relative contribution of cells of the CB and the haploidentical donor to bone marrow (BM) cellularity and to the circulating leukocytes, both as a whole and in separated fractions of mononuclear cells and neutrophils. Samples for these evaluations were collected 2 to 3 times per week from PB and weekly from BM from day +7 until engraftment and full chimerism of CB cells was achieved. Thereafter, samples were collected only when deemed convenient according to clinical course. PCR analysis of microsatellite and HLA polymorphisms also were used to detect the presence of CB or haploidentical donor cells in biopsy materials from GVHD lesions of the patients.
ANC greater than 0.5×109/L for three consecutive days was seen as evidence of myeloid engraftment, and the number of days elapsed from the transplant to the first of these three days was defined as time to engraftment. The number of days until ANC of CB derivation greater than 0.5×109/L, calculated on the basis of data of ANC and the proportion of CB DNA in the circulating granulocytes, was defined as time to CB engraftment. Times to red cell and platelet recovery were defined as the number of days elapsed from transplant to the first day of a period of seven with hemoglobin greater than 8 g/dL and platelet count greater than 20,000×109/L without transfusion support.
Posttransplant complications
Until hospital discharge, patients were evaluated daily for development of toxic, infectious, or other complications and for manifestations of GVHD, and after this as outpatients as clinically required. Standard criteria were used for the diagnosis of infections, veno-occlusive disease (VOD), and acute and chronic GVHD.
Results
Posttransplant course, engraftment, and tracing of the grafts
Data for posttransplant tracing of the grafts and engraftment are shown in Table 4. All patients had pancytopenia with an ANC nadir of less than 0.2×109/L between days −4 and +3 (median 0). In all cases, DNA of CB genotype first was de-tected in BM and/or PB on days +9 to +14 (median +10). DNA of haploidentical genotype also first was detected on days +9 to +14 (median +9) in all but two of the four patients who received haploidentical cells from their mothers. Nine patients reached a level of ANC greater than 0.5×109/L after intervals of 9 to 36 days (median 12).
Table 4Posttransplant tracing of grafts and engraftment
Case ID
S-1
S-2
S-3
S-4
S-5
S-6
M-1
M-2
M-3
M-4
F-1
Day of induced neutropenia <0.2×109/L
+3
−1
+3
0
+3
−1
0
+2
−4
−3
0
DNA of haploidentical genotype first detected in BM and/or PB
+10
+14
+10
+9
+11
+10
ND
+9
ND
+9
+10
DNA of CB genotype first detected in BM and/or PB
+10
+14
+10
+9
+10
+10
+15
+9
+11
+9
+10
Initiation of ANC > 0.5×109/L
+12
+17
+10
+9
+11
+10
+20
+36
NA
NA
+11
Initiation of ANC of CB genotype > 0.5×109/L
+21
+24
+19
+51
+53
+54
+20
+36
NA
NA
+14
Days from ANC > 0.5×109/L to CB-ANC > 0.5×109/L
9
7
9
42
42
42
44
0
0
—
—
Complete CB chimerism without traceable haploidentical cells
+29
+26
NA
+56
+94
+96
+21
+21
+11
NA
+30
Red cells transfusion independence
+21
+38
NA
+30
+31
+14
NA
+67
NA
NA
+16
Platelet transfusion independence
+54
+51
NA
+43
+20
+32
NA
+60
NA
NA
+17
Data preceded by − or + indicate days before or after the day of the transplant (day 0).
ANC indicates absolute neutrophil count; BM, bone marrow; CB, cord blood; NA, not achieved; ND, not detected; PB, peripheral blood.
Of the four patients receiving maternal haploidentical cells (1.05–2.33×105/kg, median 2.14), none had significant numbers of circulating leukocytes of maternal genotype at any time. Two of these patients (M-3 and M-4) died before reaching ANC greater than 0.5×109/L: one (M-3), who did not receive ALG, died on day +30 of acute GVHD-IV, with PB and BM cells only showing CB DNA (full CB chimerism); the other (M-4) was conditioned with fludarabine, ALG, and reduced doses of both TBI and CTX and died on day +24 due to toxic multiorgan failure syndrome (MOF). The other two (M-1 and M-2) achieved engraftment with full chimerism of CB cells on days +20 and +36.
The seven patients who received haploidentical cells from a sibling (either gender) or the father had early increases of ANC that achieved the threshold 0.5×109/L on days +9 to +17 (median +10). In these patients, both granulocytes and mononuclear cells were initially primarily of haploidentical derivation, as ascertained by analysis of the HLA and microsatellite genotypes, with subsequent progressive shift to the CB genotype in both populations, both in BM and in PB (Figure 1, Figure 2, Figure 3). Contribution of cells of haploidentical derivation to circulating leukocytes was noticeably higher and longer-lasting in patients S-3 to S-6 and F-1, all of whom received doses of haploidentical cells in the range of 2.28 to 2.54×106/kg (median 2.42) than in S-1 and S-2, who received 1.21 and 1.55×106/kg. ANC of CB derivation greater than 0.5×109/L, calculated on the basis of data of ANC and proportion of CB DNA in the circulating granulocytes, was reached between days +19 and +54 (median +24). Intervals to this day from the day of granulocyte engraftment ranged from 3 to 44 days (mean 22, median 9). Six of these seven patients achieved a final full CB chimerism between days +26 and +96 (median +56) as ascertained in both PB and BM cells. The only case in which this was not achieved was patient S-3, who was transplanted in PR and had persistent residual leukemia; at the time of death on day +56 his bone marrow was chimeric including contributions from CB, haploidentical and endogenous cells although there was a predominance of CB DNA (81% in granulocytes and 96% in mononuclear cells from the PB). In total, CB engraftment was achieved by 9 of the 11 patients (82%), with final CR and complete chimerism exclusively of CB cells reached by 8 (73%) (Table 5). One patient (M-4) died because of acute GVHD-IV on day +30 in complete chimerism of only CB cells, although without having reached the ANC criteria for engraftment.
Figure 1Examples of RSCA chimerism analysis. Profiles of samples of patients S-1 and S-5, who respectively received relative low and high number of haploidentical CD34+ cells (1.28 and 2.54×106/kg). (A): Patient S-1, Lane 1: Patient pretransplant; Lane 2: Sibling donor pretransplant; Lane 3: Cord blood unit pretransplant; Lane 4: BM mononucelar cells 11 days posttransplant (+11); Lane 5: PB mononuclear cells (+13); Lane 6: PB granulocytes (+13); Lane 7: PB total leukocytes (+21); and Lane 8: PB mononuclear cells (+34). Arbitrary values attributed to the lower and upper markers of 1000 and 2000 allow the evaluation of heteroduplex mobilities as shown in the “Mobility Scale” at the bottom of the figure. Note: In the example haploidentical donor marker B45 is not seen in lanes 7 and 8 and patient marker B51011 is not seen in these same lanes; CB markers B51 variant and B58 are the only markers seen in lanes 7 and 8. (B) Patient S-5, Lane 1: Patient pretransplant; Lanes 2–5 are RSCA profiles of BM granulocyte DNA at 18, 39, 53, and 70 days posttransplant, respectively. The figures under the duplex peaks, exhibiting HLA mismatching, are their assigned heights and the percentages are the relative cell concentrations in the mixed chimerisms. Note: The selected lanes of this example clearly show that the pretransplant sample of the patient lacks markers A30 (that progressively rises after the transplant) and A24 (that progressively declines after the transplant). Table 2 shows that these markers belong respectively to the CB and the haploidentical donor. The last lane is a sample of day +70 and Table 1 indicates that the patient was in full CB chimerism more than 8 months after the transplant. Tracings of CB and haploidentical donor have been omitted for simplicity.
Figure 1Examples of RSCA chimerism analysis. Profiles of samples of patients S-1 and S-5, who respectively received relative low and high number of haploidentical CD34+ cells (1.28 and 2.54×106/kg). (A): Patient S-1, Lane 1: Patient pretransplant; Lane 2: Sibling donor pretransplant; Lane 3: Cord blood unit pretransplant; Lane 4: BM mononucelar cells 11 days posttransplant (+11); Lane 5: PB mononuclear cells (+13); Lane 6: PB granulocytes (+13); Lane 7: PB total leukocytes (+21); and Lane 8: PB mononuclear cells (+34). Arbitrary values attributed to the lower and upper markers of 1000 and 2000 allow the evaluation of heteroduplex mobilities as shown in the “Mobility Scale” at the bottom of the figure. Note: In the example haploidentical donor marker B45 is not seen in lanes 7 and 8 and patient marker B51011 is not seen in these same lanes; CB markers B51 variant and B58 are the only markers seen in lanes 7 and 8. (B) Patient S-5, Lane 1: Patient pretransplant; Lanes 2–5 are RSCA profiles of BM granulocyte DNA at 18, 39, 53, and 70 days posttransplant, respectively. The figures under the duplex peaks, exhibiting HLA mismatching, are their assigned heights and the percentages are the relative cell concentrations in the mixed chimerisms. Note: The selected lanes of this example clearly show that the pretransplant sample of the patient lacks markers A30 (that progressively rises after the transplant) and A24 (that progressively declines after the transplant). Table 2 shows that these markers belong respectively to the CB and the haploidentical donor. The last lane is a sample of day +70 and Table 1 indicates that the patient was in full CB chimerism more than 8 months after the transplant. Tracings of CB and haploidentical donor have been omitted for simplicity.
Figure 2Profiles of ANC recovery of four patients: S1 and S2, who received a relatively low number of haploidentical CD34+ cells (1.28 and 1.55×106/kg), and S5 and S6, who received a relatively high number of these (2.54 and 2.28×106/kg). Bars represent percentage of haploidentical and CB DNA recovered from peripheral blood granulocytes. Parallel values were observed in DNA from PB mononuclear cells and in DNA recovered from bone marrow granulocytes and mononuclear cells seen in the example shown in Figure 3.
Figure 3Example of evolution of the haploidentical and CB grafts (patient S-5). Early engraftment of haploidentical cells and progressive replacement by CB graft. Bars represent percentage of haploidentical and CB DNA in peripheral blood and marrow cells as indicated in panel legends.
After the transplant, red cell transfusion independence was reached by six patients within 16 to 67 days (median 30), and platelet transfusion independence with sustained counts higher than 20,000/dL was achieved by the same patients after 17 to 60 days (median 43).
Posttransplant complications
Observed complications and patients' final status are shown in Table 5.
Extrahematological toxicity
VOD and MOF were the cause of two early deaths (patients M-1 and M-4, respectively conditioned with regimens of full doses and reduced doses of TBI and CTX). One patient (S-1), treated with full-dose TBI, had an intracraneal hygroma that regressed spontaneously. There were no other major toxic manifestations. Mucositis was noticeably mild in the patients receiving the preparative regimen with lower doses of TBI and CTX.
Acute GVHD
Manifestations of acute GVHD grade II or higher occurred in four cases. In two patients (S-5 and M-3) it was higher than grade II and did not respond to treatment. PCR analysis of microsatellite or HLA polymorphisms on biopsy material from skin and intestinal GVHD lesions of the patients showed DNA of CB cells in several instances, but DNA of the haploidentical donor never was detected.
Infections
The only significant infections during the period of neutropenia were a gram-negative urinary tract infection in patient S-2 and persistent candidemia observed in patient S-4 from day +2, that cleared immediately after ANC recovery on day +9. Recognized etiologies of major infections subsequent to engraftment were CMV, BK virus, and toxoplasmosis. Treatment of these required the use of drugs with myelotoxic side effects (such as gancyclovir, sulfadiazine, and trimetoprim-sulfamethoxaol) that were tolerated, with or without supportive treatment with filgrastim, without prohibitive granulocytopenia.
Survival and causes of death
Figure 4 shows the actuarial disease-free survival (DFS). There have been six deaths (on days +240, +71, +56, +38, +30, and +24), respectively caused by long-lasting acute GVHD, relapsing focal pneumonia of undetermined etiology, fulminating CMV pneumonia, VOD, MOF, and acute GVHD-IV (one of the two patients who did not receive ALG as part of the conditioning regimen). Five patients survived at least until November 1, 2002, in CR and in good general condition more than 43, 29, 15, 8, and 6 months posttransplant; all are in full CB chimerism without detectable cells of haploidentical genotype.
Figure 4Actuarial DFS of the 11 patients subjected to dual transplant of CB and haploidentical CD34+ cells: 54% DFS at 6 months and 40% at 42 months.
Contributing factors to the high TRM of UCBT performed with a single just-thawed CB unit are long time to en-graftment both for granulocytes and platelets (medians of 25–30 and 60–90 days, respectively), high rate of engraftment failures (10% or higher), and slow recovery of protective immunity. Toxicities related to the preparative regimen and the treatments required for infectious complications also may be added factors of morbidity and mortality [
Successful transplantation of HLA-matched and HLA-mismatched umbilical cord blood from unrelated donors: Analysis of engraftment and acute graft-vs-host disease.
Cell dose and speed of engraftment in placental/umbilical cord blood transplantation: graft progenitor cell content is a better predictor than nucleated cell quantity.
Survival after transplantation of unrelated donor umbilical cord blood is comparable to that of human leukocyte antigen-matched unrelated donor bone marrow.
Comparison of unrelated cord blood and unrelated bone marrow transplants for leukemia in children: A collaborative study of the New York Blood Center and the International Bone Marrow Transplant Registry.
]. It thus is conceivable that strategies that may result in shorter posttransplant neutropenia also may result in lower risk of early bacterial and fungal infections and in reduction of early TRM of the UCBT. The use of preparative regimens of reduced extrahematological toxicity could be another contributing factor.
Late engraftment of UCBT relates to the low number of HSC and the very primitive features of these [
Transplantation of unrelated donor umbilical cord blood in 102 patients with malignant and nonmalignant disease: influence of CD34+ cell dose and HLA disparity on treatment-related mortality and survival.
]. One of the strategies being investigated to obtain a faster recovery of ANC is the previous culture expansion of an aliquot of the transplanted CB unit to increase numbers of primitive and partially differentiated HSC. High expansion of TNC, CD34+ cells, and CFUs thus can be obtained, but shortening of posttransplant neutropenia has not been achieved with this strategy in clinical assays. Moreover, there are experimental and clinical data that suggest that “expansion cultures” may involve depletion of the CB repopulating HSC pool or loss of homing and/or proliferation capacity [
Cord blood transplants: early recovery of neutrophils from co-transplanted sibling haploidentical progenitor cells and lack of engraftment of cultured cord blood cells, as ascertained by analysis of DNA polymorphisms.
Prompt and durable engraftment in two adult patients with high risk chronic myelogenous leukemia (CML) using ex-vivo expanded and unmanipulated unrelated umbilical cord blood.
]. Similarly, faster ANC recovery has not been obtained by the simultaneous transplantation of several uncultured CB units in order to provide high numbers of TNC and CD34+ cells. These assays consistently have shown final engraftment exclusively, or predominantly, of one of the transplanted units [
Based on the typical rapid engraftment of transplants of PB HSC and on the also rapid engraftment and lack of GVHD reported for transplants of large numbers of highly purified haploidentical PB HSC [
], we theorized that co-transplantation of a limited number of the latter plus a CB unit could result in rapid recovery of the ANC, presumably without risk of rejection of the UCBT or added risk of GVHD. Induction of immunotolerance also could be conceivable. We further theorized that the eventually engrafted haploidentical cells either could persist without unfavorable effects or fade away, either as a result of rejection or because of exhaustion of their HSC. Their expected main role in this strategy would be to allow the patient to gain life time for engraftment of the primitive HSC of the CB. The final objective of the strategy is a persistent CB engraftment that, providing cells of lymphoid lineage, may result in a more efficient immune and graft-vs-leukemia (GVL) effect than by the exclusive transplantation of lymphocyte-depleted haploidentical HSC. In experiments of competitive engraftment of human PB and CB HSC in SCID/NOD mice we have observed (unpublished results) a pattern of initial transient engraftment of PB HSC and later grafting advantage for the CB cells when the proportion of haploidentical CD34+ to CB CD34+-infused cells is relatively low (5/1).
The results we report refer to the use of this dual transplant strategy in a group of 11 adults with high-risk heavily treated acute leukemia, who were transplanted with a CB unit of relatively low TNC/kg together with haploidentical cells from a sibling or a parent. Molecular analysis of chimerism showed clearly distinct patterns of engraftment for the two components of the dual transplants in the seven patients receiving haploidentical cells from the father or a sibling: relatively rapid expansion-maturation but transient engraftment of the haploidentical cells resulting in early recovery of the ANC primarily from the haploidentical cells; slower expansion and differentiation but sustained and final exclusive or highly predominant engraftment of the CB progenitors.
In the four patients receiving haploidentical stem cell grafts from their mothers there was no significant contribution of these to the circulating granulocytes. These engraftment failures cannot be related to the number of infused cells, to ABO incompatibility, or to host rejection targeting the HLA NIMA per se, since NIMA also were carried by the six sibling donors and, moreover, cytotoxic cross-matches were negative. The failures could be dependent on unrecognized factors not necessarily related to mothers as donors. Based on available data, the mechanism responsible for the final effacement of the engrafted haploidentical cells cannot be recognized either. Rejection by the CB graft or exhaustion of the HSC pool contained in the infused haploidentical product under heavy proliferative stress are possible mechanisms that cannot be disregarded. Rejection exerted by the patients' own residual immune system seems less likely.
The early recovery of the ANC primarily from the haploidentical engraftment observed in patients receiving cells from the father or a sibling may have provided support, allowing time for the CB engraftment to occur. In fact, these patients had limited morbidity and no deaths due to neutropenia-related infections. They also had good capacity to withstand, without prohibitive neutropenia, treatments with myelosuppressive drugs that were required to treat infections due to other opportunistic agents (CMV, toxoplasma, BK virus) that occurred after recovery of the ANC and that appeared to be related basically to deficient protective immunity and GVHD.
The relatively mild mucosal toxicity observed in our patients also may have contributed to the low early morbidity by bacterial and fungal infections. The resulting low requirements of antimicrobials in the early posttransplant period also may have been a factor contributing to the low early toxicity of the procedure. The main causes of the six registered deaths were infections by the opportunistic agents, mainly CMV, in four cases and toxic complications in two.
Unfavorable effects related to the co-transplantation of the haploidentical cells have not been observed. GHVD has not been more frequent or severe than in single UCBT and DNA only from the own patient and the CB have been detected in the biopsy specimens from the GVHD cutaneous and gastrointestinal lesions. As it is well known that GVHD may be facilitated by infections, a favorable impact on GVHD might occur as a consequence of fewer infections, among other possible factors. Hypothetically the exposure of the engrafting CB progenitors to the haploidentical antigens also could be a factor that could induce immunotolerance and less GVHD. The number of cases we have transplanted to date is not sufficient to properly evaluate the impact of this strategy on GVHD.
Our strategy of dual transplants of a CB unit and a limited number of haploidentical PB CD34+ cells resulted in an initial growth advantage for the haploidentical cells, contributing to early recovery of the ANC. However, CB HSC appeared to have a competitive advantage for long-term engraftment. This strategy thus may reduce risks of early bacterial and fungal infections that are due to the long posttransplant period of neutropenia and allow sufficient time for the engraftment of the CB HSC. Our approach appears to be a promising strategy to improve outcome for UCBT in patients with higher body weights.
Acknowledgements
Research was carried out as part of the Eurocord II Program, supported by EC Quality of Life and Management of Living Resources, Contract Number QLRT 1999-00380. We are indebted to Dr. Pablo Rubinstein for reviewing our results and for his comments and helpful suggestions.
References
Wagner J.E.
Kernan N.A.
Steinbuch R.
et al.
Allogeneic sibling umbilical-cord-blood transplantation in children with malignant and non-malignant disease.
Successful transplantation of HLA-matched and HLA-mismatched umbilical cord blood from unrelated donors: Analysis of engraftment and acute graft-vs-host disease.
Cell dose and speed of engraftment in placental/umbilical cord blood transplantation: graft progenitor cell content is a better predictor than nucleated cell quantity.
Survival after transplantation of unrelated donor umbilical cord blood is comparable to that of human leukocyte antigen-matched unrelated donor bone marrow.
Comparison of unrelated cord blood and unrelated bone marrow transplants for leukemia in children: A collaborative study of the New York Blood Center and the International Bone Marrow Transplant Registry.
Cord blood transplants: early recovery of neutrophils from co-transplanted sibling haploidentical progenitor cells and lack of engraftment of cultured cord blood cells, as ascertained by analysis of DNA polymorphisms.
Prompt and durable engraftment in two adult patients with high risk chronic myelogenous leukemia (CML) using ex-vivo expanded and unmanipulated unrelated umbilical cord blood.
Transplantation of unrelated donor umbilical cord blood in 102 patients with malignant and nonmalignant disease: influence of CD34+ cell dose and HLA disparity on treatment-related mortality and survival.
00067-5&id=gr1a.jpg){kind=link}
00067-5&id=gr1b.jpg){kind=link}
00067-5&id=gr2.jpg){kind=link}
00067-5&id=gr3.jpg){kind=link}
00067-5&id=gr4.jpg){kind=link}