`
`Bone Marrow Transplantation (2001) 28, 827–834
` 2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00
`
`www.nature.com/bmt
`
`T cell depletion
`
`Excessive T cell depletion of peripheral blood stem cells has an
`adverse effect upon outcome following allogeneic stem cell
`transplantation
`
`R Chakraverty1, S Robinson1, K Peggs1, PD Kottaridis1, MJ Watts1, SJ Ings1, G Hale2,
`H Waldmann2, DC Linch1, AH Goldstone1 and S Mackinnon1
`
`1Department of Haematology, University College London, London, UK; and 2Sir William Dunn School of Pathology, University of
`Oxford, UK
`
`Summary:
`
`We evaluated the outcome of two modes of T cell
`depletion for HLA-identical sibling stem cell transplants
`in 34 consecutive adult patients: group A (n = 11)
`received PBSC post CliniMACs
`immuno-magnetic
`enrichment of CD34+ cells and group B (n = 23) received
`bone marrow following in vitro incubation with CAM-
`PATH-1M and complement. All patients received an
`identical conditioning regimen which consisted of in vivo
`CAMPATH-1H 20 mg over 5 days, thiotepa 10 mg/kg,
`cyclophosphamide 120 mg/kg and 14.4 Gy TBI. No
`additional graft-versus-host disease prophylaxis was
`given. The mean T cell dose administered was
`0.02 ± 0.05 × 106/kg for group A and 2.8 ± 2.8 106/kg for
`group B (P ⬍ 0.001). With a median follow-up of 28
`months overall survival was 36.4% for group A at 12
`months compared to 78.3% for group B (P = 0.001).
`Transplant-related mortality in group A at 12 months
`was 63.6% as compared to 18.0% in group B
`(P = 0.003). Most of the procedure-related deaths in
`group A occurred secondary to infection. These results
`suggest that extensive in vitro T cell depletion of periph-
`eral blood stem cells in combination with in vivo T cell
`depletion may have profound effects upon the incidence
`of infections following allogeneic stem cell transplan-
`tation and this may adversely effect transplant-related
`mortality. Bone Marrow Transplantation (2001) 28,
`827–834.
`Keywords: CD34 selection; allogeneic transplantation; T
`cell depletion
`
`T cell depletion (TCD) is an effective means of reducing
`the incidence and severity of acute or chronic graft-versus-
`host disease (GVHD) following allogeneic stem cell trans-
`plantation (SCT).1–4 A number of different methods for
`TCD have been developed over the last two decades that
`
`Correspondence: Dr S Mackinnon, Department of Haematology, Univer-
`sity College Hospital, 98 Chenies Mews, London WC1E 6HX, UK
`Received 9 May 2001; accepted 27 July 2001
`
`lectin agglutination,4
`include counter-flow elutriation,2
`monoclonal antibodies directed at T cell antigens with nar-
`row5 or broad specificity1 or depletion through positive
`selection of CD34+ stem cells.6,7 The clinical benefit of
`TCD allogeneic SCT is controversial since this approach
`may be compromised by increased rates of graft rejection8
`and a higher rate of relapse.9 To overcome the risk of graft
`rejection additional approaches have been adopted that
`immunosuppression,10
`include increasing pre-transplant
`‘partial’ TCD11 or increasing the number of stem cells
`administered by using peripheral blood stem cells
`(PBSCs).12,13 The optimal T cell content of a graft that
`maintains a significant graft-versus-leukaemia (GVL) effect
`has not yet been defined. Although TCD is used widely by
`many different transplant centres, few studies have assessed
`directly how different modes of TCD affect clinical
`outcome.
`One area of potential concern is the significant delay in
`immune reconstitution that occurs following TCD allog-
`eneic SCT. This is particularly the case in adults, where
`the initial T cell repertoire is dependent upon peripheral
`expansion of mature T cells in the graft.14–16 Delayed recov-
`ery of CD4+, CD4+CD45RA+ and TCR␥␦+ T cells with a
`concomitant reduction in TCR diversity are typical features
`of the early post-transplant period following TCD allog-
`eneic SCT.16–19 T cell purging may thus exacerbate post-
`transplant immunodeficiency and be complicated by an
`increased incidence of opportunistic infections, particularly
`CMV re-activation.20
`In this report, we highlight the major impact on clinical
`outcome of two approaches to in vitro TCD, using either
`positive selection of CD34+ cells from PBSC or CAM-
`PATH 1M treatment of bone marrow, as part of a protocol
`that employed additional
`in vivo TCD. Patients who
`received donor PBSCs heavily depleted of T cells (5 log)
`by immuno-magnetic selection of CD34+ cells had a sub-
`stantially higher procedure-related mortality than recipients
`of CAMPATH 1M-treated bone marrow. The higher num-
`ber of procedure-related deaths was caused by a marked
`increase in the number of opportunistic infections within
`the TCD PBSC group. This study suggests that following
`HLA-identical sibling SCT in adults, extensive in vitro
`TCD of PBSC in combination with in vivo TCD is compli-
`
`
`
`Adverse outcome of T cell-depleted allogeneic PBSCT
`R Chakraverty et al
`
`cated by a profound immunodeficiency that outweighs any
`benefit in terms of reduction of GVHD.
`
`the lack of patients with lymphoproliferative disorders in
`group A. Three patients in group B had received a previous
`SCT and none in group A.
`
`Patients and methods
`
`Antibodies
`
`A total of 34 consecutive adult patients who received a
`TCD HLA-identical sibling donor stem cell transplant at
`University College Hospital London, UK between January
`1997 and October 1999, were included in the analysis. Dur-
`ing this period, the method of in vitro TCD was changed
`due to an insufficient supply of CAMPATH 1M. The proto-
`cols were approved by the local institutional review com-
`mittees and all patients gave informed consent. Two groups
`of patients were defined according to the mode of T cell
`depletion employed: ‘group A’ consisting of 11 patients
`who received PBSC following CliniMACs immuno-mag-
`netic enrichment of CD34+ cells and ‘group B’ comprising
`23 who received bone marrow following in vitro incubation
`with CAMPATH 1M and complement.
`
`Patient characteristics
`
`The patient characteristics for each group are shown in
`Table 1. Patients were considered to be ‘standard risk’ in
`the case of acute myeloid leukaemia (AML) in first com-
`plete remission (CR) or chronic myeloid leukaemia (CML)
`in first chronic phase.21 All other diagnoses were con-
`sidered high risk. There were no significant differences
`between the groups in terms of age, sex, performance status
`at the time of transplant, time to transplant or those at risk
`of CMV disease (Table 1). However, only one patient in
`group A was considered to be high risk as compared to 11
`of 23 patients in group B (P = 0.05). This reflected, in part,
`
`Table 1
`
`Comparison of groups A and B
`
`Group A
`(n = 11)
`
`Group B
`(n = 25)
`
`P value
`
`Sex, male/female (n)
`Median, age years (range)
`Diagnosis (n)
`AML CR1
`AML CR2
`MDS
`MM
`HD
`HG-NHL
`CML 1st cp
`CML ap
`Risk status, standard/high
`Median time to transplant,
`months (range)
`Previous transplant
`CMV serology (n)
`P−/D−
`P+/D−
`P−/D+
`P+/D+
`
`5/6
`41 (26–51)
`
`4/19
`41 (17–59)
`
`8
`1
`0
`0
`0
`0
`2
`0
`10/1
`6.5 (2–17)
`
`7
`2
`3
`2
`2
`1
`5
`1
`12/11
`7.5 (4–57)
`
`0
`
`5
`0
`4
`2
`
`3
`
`4
`3
`8
`8
`
`0.05
`
`MDS = myelodysplastic syndrome; MM = multiple myeloma; HD =
`Hodgkin’s disease; HG-NHL = non-Hodgkin’s lymphoma; 1st cp = first
`chronic phase; ap = accelerated phase; P = patient; D = donor.
`
`CAMPATH-1H is a humanized IgG1 monoclonal antibody
`against the CD52 antigen.22 It was prepared from the cul-
`ture supernatant of Chinese hamster ovary cell transfectants
`cultured in a hollow fibre fermentor. It was purified by
`affinity chromatography on Protein A sepharose and size
`exclusion chromatography on Superdex 200 and formulated
`in phosphate-buffered saline. The half-life of CAMPATH-
`1H in humans is dependent on the amount of target CD52
`antigen in the patient. Based on work in progress, there is
`persistence of CAMPATH-1H in vivo post day 0 sufficient
`to cause T cell lysis by ADCC. CAMPATH-1M is a rat
`IgM antibody that recognises the same antigen. It was pre-
`pared from hybrid myeloma cells using stirred fermentors,
`purified by fractionation with ammonium sulphate and
`reformulated in phosphate-buffered saline.
`
`Conditioning regimen
`
`All patients received the same conditioning regimen which
`consisted of in vivo CAMPATH-1H 20 mg on days −9 to
`day −5, thiotepa 5 mg/kg on days −8 and −7, cyclophos-
`phamide 60 mg/kg on days −6 and −5 and 14.4 Gy total
`body irradiation, with partial
`lung shielding,
`in eight
`fractions over 4 days.
`For stem cell collection for group A patients, sibling
`donors received G-CSF at 10 g/kg subcutaneously once
`daily on day −4 to day 0. Leukaphereses were performed
`on day 0 ± day +1 using conventional techniques for PBSC.
`TCD was performed by positive immuno-magnetic selec-
`tion of CD34+ cells using a CliniMACs, (Miltenyi Biotec,
`Bergisch Gladbach, Germany) cell separation system.23
`For group B patients, donor bone marrow was aspirated
`under general anaesthesia and TCD performed in vitro upon
`the derived buffy coats by incubation with 25 mg CAM-
`PATH 1M at room temperature for 10 min followed by
`incubation with 10–30% autologous plasma at 37°C for
`45 min.
`For both protocols, the level of TCD was monitored by
`flow cytometric analysis of CD3 staining.
`
`Supportive care
`
`Patients were managed in reverse isolation in conventional
`or laminar air flow rooms. All patients received prophylaxis
`with cotrimoxazole or pentamadine against Pneumocystis
`carinii infection. Aciclovir and triazole prophylaxis were
`routinely used. Blood products were irradiated to 25 Gy.
`Red cell and platelet transfusions were given to maintain
`the Hb ⬎9 g/dl and platelet count ⬎10 × 109/l. Patients
`who were CMV seronegative received only blood products
`from CMV seronegative donors; seropositive patients
`received blood products from donors unscreened for CMV.
`Febrile neutropenic patients received intravenous piptazob-
`actam and gentamicin as first
`line antibiotic therapy.
`Patients received G-CSF at 5 g/kg/day from day +6 until
`
`9
`
`828
`
`Bone Marrow Transplantation
`
`
`
`9
`
`829
`
`Adverse outcome of T cell-depleted allogeneic PBSCT
`R Chakraverty et al
`
`between the curves was estimated by the log rank test.
`Patient characteristics in the two groups were compared by
`Fisher’s exact test or the Mann–Whitney test, whichever
`was appropriate.
`
`Results
`
`T cell depletion
`
`Patients in group A received approximately 2 log less T
`cells than group B patients (Table 2). The mean T cell dose
`was 0.02 ± 0.05 × 106/kg for group A recipients and
`2.8 ± 2.8 106/kg for group B patients (P ⬍ 0.001). Three
`patients in group A had ‘add-back’ of donor T cells follow-
`ing the transplant (Table 2). Two of these patients, UPN6
`and UPN8, received 1 × 106/kg T cells at 4 and 5 months,
`
`Table 2
`
`Patient and graft characteristics
`
`UPN
`
`Age
`
`Sex
`
`Diagnosis
`
`CD34+ dosea
`
`T cell dosea
`
`Group A
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`
`Group B
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`29
`30
`31
`32
`33
`34
`
`51
`37
`49
`26
`44
`38
`44
`43
`27
`35
`41
`
`44
`17
`28
`41
`41
`40
`46
`27
`46
`36
`36
`37
`45
`59
`26
`57
`36
`57
`44
`56
`38
`37
`42
`
`F
`F
`M
`M
`M
`M
`F
`F
`M
`F
`F
`
`M
`F
`M
`F
`M
`F
`M
`M
`F
`M
`M
`M
`F
`M
`F
`M
`M
`M
`M
`F
`F
`F
`M
`
`AML CR1
`AML CR1
`CML 1st cp
`AML CR1
`AML CR1
`AML CR1
`AML CR1
`AML CR1
`AML CR2
`AML CR1
`CML 1st cp
`
`MM PR
`AML CR1
`HD Rel 2
`HD Rel 2
`AML CR1
`AML CR1
`AML CR1
`MDS RAEB
`MM PR
`CML 1st cp
`HG-NHL CR3
`MDS RA
`AML CR2
`CML ap
`AML CR1
`CML 1st cp
`AML CR1
`MDS RAEB
`CML 1st cp
`CML 1st cp
`AML CR1
`CML 1st cp
`AML CR2
`
`4.6
`5.1
`2.1
`3
`6.8
`2.4
`4.5
`6.7
`4.8
`6.4
`2
`
`ND
`1.6
`2
`3.1
`1.2
`4.2
`2.9
`0.7
`0.7
`1.6
`1
`3.2
`1.4
`1.87
`1.4
`ND
`0.4
`0.7
`1
`2
`1
`1.3
`1.7
`
`0.0005
`0.02
`0.004b
`0.0006
`0.005
`0.2b
`0.0005
`0.00007b
`0.003
`0.0007
`0.0007
`
`ND
`0.5
`ND
`6.6
`1
`3b
`1.7
`2.8
`0.6b
`1.1
`0.3
`10.5
`0.6
`2.4
`2.1
`7.4
`0.07
`2.9
`4.7
`0.2
`6.4
`4.2
`3.9
`
`the ANC was at least 1.0 × 109/l for 2 consecutive days.
`CMV seropositive patients were monitored weekly from
`transplantation until at least day 120 by qualitative PCR of
`CMV DNA from peripheral blood. Pre-emptive ganciclovir
`therapy (5 mg/kg twice daily intravenously or adjusted
`according to renal function) was given following two con-
`secutive positive PCR results and discontinued after 2
`weeks if a negative PCR was obtained.24 In the event of
`continued PCR positivity, foscarnet was substituted for
`ganciclovir and the drugs alternated every 2 weeks accord-
`ing to the PCR results. A single patient (UPN5) received
`cidofovir as initial pre-emptive therapy for PCR detection
`of CMV re-activation.
`
`GVHD prophylaxis and grading
`
`No additional GVHD prophylaxis was given. Patients who
`survived 100 days or longer were evaluable for chronic
`GVHD. Both acute GVHD and chronic GVHD were graded
`according to the consensus criteria.25,26
`
`Evaluation of infective complications
`
`An infective complication was defined as any infection
`occurring post day 21 which required continued or new
`hospital admission/referral. CMV re-activation was defined
`as two consecutive peripheral blood PCR positive results.
`CMV disease was diagnosed on the basis of an inflamma-
`tory process due to CMV confirmed by the presence of
`typical cytopathic and immuno-fluorescent features in his-
`tological preparations or positive detection of early antigen
`fluorescent foci (DEAFF) and/or CMV culture from rel-
`evant material such as washings from broncho-alveolar lav-
`age (BAL). Pulmonary fungal
`infection was diagnosed
`either by histological confirmation or characteristic high
`resolution CT appearances (halo sign) plus positive cultures
`from a BAL. Fungal infection at other sites was identified
`from post-mortem histological analysis of affected organs.
`RSV, parainfluenza I and III or influenza B infection were
`defined as pulmonary signs plus direct immunofluorescence
`and/or culture for the relevant viruses from naso-pharyn-
`geal aspirate or BAL. Confirmation of Legionaire’s disease
`was made by the presence of urinary Legionella antigen on
`two consecutive occasions and confirmed by post mortem
`histological examination of lung. Adenoviral, RSV and
`measles infection were confirmed by histological examin-
`ation of the relevant organs postmortem. Pneumonia was
`defined as fever, associated with signs of lung consolidation
`and new infiltrates identified on chest X-ray or high
`resolution CT.
`
`Statistical analysis
`
`Overall survival (OS) was measured from transplantation
`until death from any cause. Patients still alive at the time
`of the analysis were censored at the last follow-up date.
`Transplant-related mortality (TRM) was determined from
`the date of transplantation until death related to transplan-
`tation. Patients who died from other causes were censored
`at the time of death. OS and TRM were estimated by the
`Kaplan–Meier method and the significance of differences
`
`a×106 cells/kg recipient weight.
`bPatients had donor lymphocyte infusions at following doses × 106 CD3
`cells/kg recipient weight (time post transplant in months): UPN3 (10+);
`UPN6, 1 (4+); UPN8, 1 (5+); UPN17, 3 (11+); and UPN20, 202 (12+).
`MDS = myelodysplastic syndrome; RA = refractory anaemia; RAEB =
`refractory anaemia with excess blasts; MM = multiple myeloma; HD =
`Hodgkin’s disease; HG-NHL = non-Hodgkin’s lymphoma; 1st cp = first
`chronic phase; ap = accelerated phase; Rel = relapse; PR = partial
`response; ND = not done.
`
`Bone Marrow Transplantation
`
`
`
`Adverse outcome of T cell-depleted allogeneic PBSCT
`R Chakraverty et al
`
`respectively, because of progressive pulmonary infections.
`UPN3 was given 3 × 106/kg T cells for cytogenetic relapse
`of CML at 10 months. Two patients in group B received
`donor T cells. UPN17 received 3 × 106/kg T cells at 11
`months for relapsed AML and UPN20 received 2 × 108/kg
`T cells for treatment of relapsed multiple myeloma.
`
`Engraftment
`
`The mean CD34+ cell dose per patient was 4.4 ± 1.8 ×
`106/kg in group A and 1.7 ± 1.0 × 106/kg in group B (P
`⬍ 0.001). Neutrophil engraftment occurred more rapidly in
`group A, with neutrophils ⬎0.5 × 109/l on day 12.0 ± 2.1
`compared to day 17.2 ± 4.0 for group B (P ⬍ 0.001). One
`patient in group B failed to engraft and died on day +74.
`All patients in group A engrafted. No cases of secondary
`graft failure occurred in either group. No differences were
`observed between the groups in terms of transfusional inde-
`pendence by day 100 (four of seven evaluable patients in
`group A as compared to 15 of 22 evaluable patients in
`group B were transfusion independent).
`
`GVHD
`
`The incidence of acute GVHD was low in both groups
`(Table 3). No patients in group A developed acute GVHD
`of greater than grade I as compared to two of 23 patients
`in group B. For patients who survived day 100 no differ-
`ence in either the incidence or extent of chronic GVHD
`was observed. Thus, three of seven evaluable patients in
`group A developed chronic GVHD (extensive in all cases)
`as compared to 11 of 22 patients in group B (extensive in
`three patients and limited in the remaining eight patients).
`In both groups, patients who received T cell add-back
`post SCT were at high risk for the development of exten-
`sive chronic GVHD. Thus, four of five patients who
`received a donor T cell infusion following the transplant
`subsequently developed extensive chronic GVHD. One
`patient (UPN3) died of progressive GVHD at 12 months
`following SCT and 2 months following infusion of
`3 × 106/kg donor T cells for relapsed CML. The remaining
`patients have all
`required prolonged treatment with
`prednisolone and cyclosporine.
`
`Infective complications
`
`Most of the deaths in group A occurred secondary to infec-
`tion (Table 3). Thus, at 12 months six of 11 patients in
`group A died secondary to infection (CMV disease n = 2,
`invasive pulmonary aspergillosis (IPA) n = 1, RSV/measles
`pneumonits n = 1, Legionella n = 1, E. coli sepsis n = 1)
`compared to only one of 23 patients (adenovirus) in group
`B (P = 0.002).
`The propensity to infection in group A is also highlighted
`by the greater number of documented serious infections
`post day 21 in group A than in group B (Table 4). Thus,
`with a median follow-up of 28 months, four of 11 patients
`in group A had three or more significant infections as com-
`pared to one of 23 in group B (P = 0.03). Furthermore,
`patients in group A were more likely to have co-existent
`infections, with six of 11 patients in group A having two
`
`or more simultaneous infections as compared to only two
`of 23 patients in group B (P = 0.005). In a number of
`patients from group A, infection progressed despite the
`appropriate anti-microbial
`therapy. Patients UPN2 and
`UPN6 both developed IPA which failed to respond to con-
`ventional, and then liposomal amphotericin therapy. Patient
`UPN2 also developed a large, deep skin ulcer secondary to
`HSV II (proven on culture and biopsy), which failed to
`respond to aciclovir, foscarnet or cidofovir. Patient UPN5
`developed CMV colitis and pneumonitis despite pre-
`emptive
`therapy with foscarnet
`and cidofovir
`and
`subsequent
`therapy with ganciclovir and intravenous
`immunoglobulin.
`There were no significant differences between the two
`groups in terms of the frequency of CMV reactivation in
`that five of six CMV seropositive individuals in group A
`and 11 of 19 in group B met the criteria for pre-emptive
`treatment for CMV reactivation or CMV disease. The
`median total duration of anti-CMV therapy administered
`was 98 days (range 8–127 days) in group A and 33 days
`in group B (range 13–129 days). Two patients from group
`A died from CMV disease in group A and none from
`group B.
`
`Immune reconstitution
`
`Group A patients showed a significant delay in the recovery
`of the absolute lymphocyte count following transplant.
`Thus, at 2 months post transplant the absolute lymphocyte
`count was ⬎1.0 × 109/l in none of nine evaluable patients
`in group A but 10 of 21 patients in group B (P = 0.01).
`Absolute lymphocyte numbers for the first 5 months for
`both groups are shown in Figure 1. Analysis of T cell sub-
`sets at our centre is usually performed at 3 monthly inter-
`vals post transplant. The poor outcome of patients in group
`A meant that the majority were not evaluated and thus an
`evaluation of T cell subset reconstitution was not possible.
`
`Overall survival and transplant-related mortality
`
`With a median follow-up of 28 months the overall survival
`(OS) and TRM for all 36 patients at 12 months were 64.7%
`and 33.0%, respectively. However, there was a major dif-
`ference in OS between the groups (Figure 2). Thus, at 12
`months OS was 36.4% for group A and 78.8% for group
`B (P = 0.001). This difference was accounted for to a great
`extent by the high number of procedure-related deaths in
`group A (Figure 3). At 12 months the TRM was 63.6% in
`group A and 18.0% in group B (P = 0.003). The causes of
`procedure-related deaths for the whole group are shown in
`Table 3. The median time for procedure-related death in
`group A was 87 days (range 28–221 days) and in group B
`was 184.5 days (range 74–287 days). All but one of the
`procedure-related deaths in group A were secondary to
`infection.
`
`Discussion
`
`This report highlights important differences in the outcome
`of HLA-identical sibling SCT following two approaches to
`
`9
`
`830
`
`Bone Marrow Transplantation
`
`
`
`9
`
`831
`
`Adverse outcome of T cell-depleted allogeneic PBSCT
`R Chakraverty et al
`
`Table 3
`
`Patient outcome
`
`UPN
`
`Group A
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`
`Group B
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`27
`28
`29
`30
`31
`32
`33
`34
`
`Neutrophils
`⬎0.5 × 109/l
`
`aGVHD
`
`cGVHD
`
`Infectionsa
`
`Current status (months)
`
`Cause of death
`
`10
`10
`17
`13
`11
`11
`13
`10
`11
`14
`12
`
`21
`19
`14
`14
`16
`17
`17
`23
`14
`30
`15
`19
`11
`16
`15
`13
`21
`16
`19
`17
`NA
`17
`15
`
`0
`I
`I
`0
`0
`I
`0
`0
`0
`0
`0
`
`0
`0
`0
`0
`0
`0
`0
`0
`III
`0
`0
`0
`0
`0
`I
`0
`0
`0
`0
`0
`0
`II
`0
`
`0
`E
`E
`0
`0
`E
`0
`0
`NE
`NE
`NE
`
`0
`0
`0
`0
`0
`E
`0
`0
`E
`0
`0
`0
`0
`L
`0
`L
`0
`0
`E
`L
`NE
`E
`0
`
`1
`4
`1
`0
`2
`5
`1
`4
`0
`3
`2
`
`1
`0
`1
`3
`0
`1
`0
`1
`1
`2
`0
`1
`2
`2
`1
`2
`0
`2
`1
`0
`1
`1
`1
`
`Alive in CR 19+
`Dead
`Dead
`Alive in CR 13
`Dead
`Alive in CR 12+
`Dead
`Dead
`Dead
`Dead
`Dead
`
`Dead
`Alive in CR 37+
`Alive in CR 37+
`Dead
`Alive in CR 35+
`Alive-relapse 11+
`Alive in CR 35+
`Alive in CR 35+
`Dead
`Alive in CR 34+
`Alive in CR 34+
`Alive in CR 34+
`Dead
`Alive in CR 29+
`Alive in CR 27+
`Alive in CR 25+
`Alive in CR 24+
`Alive in CR 23+
`Alive in CR 19+
`Dead
`Dead
`Alive in CR 16+
`Alive in CR 11+
`
`IPA
`Relapse
`
`CMV Pn
`
`IP
`Measles/RSV
`E. coli sepsis
`Legionella Pn
`CMV Pn
`
`Relapse
`
`Adenovirus
`
`Relapse
`
`ARDS
`
`EBV-LPD
`Graft failure
`
`aInfections post day 21.
`IPA = invasive pulmonary aspergillosis; CMV = cytomegalovirus;
`IP = idiopathic pneumonitis;
`L = limited; E = extensive; Pn = pneumonitis;
`RSV = respiratory syncitial virus; ARDS = adult respiratory distress syndrome; EBV-LPD = Epstein–Barr virus-associated lymphoproliferative disorder;
`NE = not evaluable.
`
`in vitro TCD, using either positive selection of CD34+ cells
`from PBSC or CAMPATH 1M treatment of bone marrow.
`Patients in both treatment groups received additional in vivo
`TCD,
`identical
`conditioning and no post-transplant
`immunosuppression. We found that allogeneic SCT using
`PBSC heavily depleted of T cells by immuno-magnetic
`selection of CD34+ cells, was associated with prolonged
`lymphocytopenia and an extremely high number of oppor-
`tunistic infections leading to a high rate of procedure-
`related deaths. In contrast, allogeneic SCT using in vitro
`CAMPATH 1M treatment of bone marrow was associated
`with a lesser depletion of T cells, fewer infections and more
`rapid immune reconstitution. Both strategies were effective
`at preventing acute GVHD, and excluding those patients
`who received donor lymphocytes post transplant, at pre-
`venting extensive chronic GVHD. There was a low risk of
`graft rejection in both groups. Since the follow-up is rela-
`tively short and there are too few long-term survivors in
`group A, no conclusions can be made regarding the risk
`of relapse.
`
`It seems likely that the overall determinant of clinical
`outcome in this report was degree of T cell depletion. There
`are five important qualifications to this statement. First, the
`type of graft differed in each group, such that patients in
`group A received PBSC and patients in group B received
`bone marrow. Thus, there may have been both quantitative
`(eg the number of CD34+ cells or CD34+CD38+ lymphocyte
`progenitors)27 and qualitative differences (eg the balance
`between Th1 or Th2 T cells)28 that could have conceivably
`affected outcome. However, to our knowledge, there are no
`substantive clinical data to support the contention that these
`differences could account for such a poor outcome in the
`PBSC arm. Second, exact comparison of the degree of TCD
`for the two methods is difficult since following CD34+ cell
`selection nearly all the remaining T cells will be viable
`whereas this may not be the case following in vitro treat-
`ment with CAMPATH 1M.29 Indeed, antibody coated T
`cells may undergo further lysis when they encounter fresh
`complement following infusion. Furthermore, pre-trans-
`plant in vivo CAMPATH 1-H treatment could have resulted
`
`Bone Marrow Transplantation
`
`
`
`Group A
`
`Group B
`
`~ ••• • ••••• 1. •• , • •••••• • ••••• ,
`
`r;
`.. r
`
`!"····i
`
`100 200 300 400 500 600 700 800 900 1000 1100 1200
`Time (days)
`
`100
`
`75
`
`50
`
`25
`
`Transplant-related mortality
`
`0
`
`0
`
`Figure 3 Transplant-related mortality. Group A shown as dotted line and
`group B as solid line.
`
`in further destruction of donor T cells in the recipient and
`it is conceivable that there are differences between the two
`modes of TCD in terms of the sensitivity of the remaining
`T cells to in vivo lysis. It seems unlikely, however, that
`these factors could account for the ⬎2 log difference in T
`cell dose observed. Third, although CD34+ cell selection
`removes most other cell types, CAMPATH 1M may be less
`efficient at removing certain cells such as NK cells (our
`unpublished observations) that could be of benefit to the
`patient post transplant. However, there are few clinical data
`available that clearly define the effect of changing the num-
`ber of NK cells or other non-T cells in the graft upon the
`post-transplant course. Fourth, this was not a randomised
`study and ‘clusters’ can arise due to local or seasonal fac-
`tors, which make an assessment of the reasons for differ-
`ences between the two groups problematic. Finally, the
`patients in the two groups may have differed in certain
`characteristics that biased our evaluation of outcome.
`Although there were some differences between the groups
`in terms of the spectrum of diseases treated, group B had
`more high risk patients and included three patients who had
`received a previous
`transplant,
`suggesting any bias
`was likely to favour the null hypothesis that there was no
`difference in outcome between the groups.
`All of the above factors may have had the potential to
`influence outcome, but one of the key differences between
`the groups was the measured number of T cells adminis-
`tered with the graft. A ‘safe’ threshold could be defined as
`the minimum T cell dose that permits high rates of
`engraftment, low rates of GVHD and low rates of serious
`infection. Any such level will be heavily influenced by
`patient factors such as age or status, donor characteristics
`including the degree of HLA disparity or CD34+ cell dose,
`and by other transplant factors such as the use of in vivo
`TCD or post-transplant immunosuppression. For example,
`there is abundant evidence that ‘safe’ T cell doses as
`defined for TCD HLA-identical sibling SCTs are less safe
`in the context of unrelated or mismatched transplants with
`resulting high rates of graft failure or infection.12,30,31
`Using this protocol, our data suggest that extensive TCD
`of PBSC was associated with a very low risk of GVHD,
`high rates of engraftment but severely compromised
`immune reconstitution. There is little consistent
`infor-
`mation regarding the ‘safe’ T cell threshold in terms of
`immune competence of the recipient following HLA-ident-
`ical sibling SCT. Recent studies indicate that infections
`
`Adverse outcome of T cell-depleted allogeneic PBSCT
`R Chakraverty et al
`
`832
`
`Table 4
`
`Infectious complicationsa
`
`Bacterial
`
`bacteraemia
`other
`
`Fungal
`
`Viral
`
`IPA
`other
`CMV re-
`activation
`CMV disease
`VZV
`HSV I or II
`PF I or III
`RSV
`other
`
`Miscellaneous
`
`Pneumonia
`
`Group A (n = 23 Group B (n = 24
`events)
`events)
`
`1
`4 (MRSA,
`Legionella, C.
`difficile, E. coli)
`2
`0
`3
`
`1
`0
`
`0
`2
`11
`
`2
`0
`2
`4
`1
`3 (Influenza B
`n = 2, measles)
`1
`
`0
`4
`1
`2
`0
`1 (adenovirus)
`
`2
`
`aInfections post day 21.
`staphylococcus aureus; VZV = varicella
`MRSA = methicillin-resistant
`zoster virus; HSV = herpes simplex virus; PF = para-influenza; IPA =
`invasive pulmonary aspergillosis.
`
`1.0
`
`2.0
`
`3.0
`Time (months)
`
`4.0
`
`5.0
`
`Figure 1 Absolute lymphocyte counts post-transplant. Mean ± s.e.m.
`absolute lymphocyte counts (×109/l) post transplant are given for group
`A (clear bars) and B (shaded bars). The number of patients in each group
`for each time point are as follows: 1 month – A 11, B 22; 2 months – A
`8, B 20; 3 months – A 6, B 19; 4 months – A 6, B 19; 5 months – A 5,
`B 15. Significant differences between group A and B were observed at 2
`months (P = 0.005), 3 months (P = 0.007) and 4 months (P ⬍ 0.05).
`
`Group A
`
`Group B
`
`.
`.. --,
`' ·,
`' --~
`! - - - - •'- -~
`' ~-----~
`
`100
`
`75
`
`50
`
`25
`
`Percent survival
`
`0
`
`0
`
`100 200 300 400 500 600 700 800 900 1000 1100 1200
`
`Time (days)
`
`Figure 2 Overall survival. Group A shown as dotted line and group B
`as solid line.
`
`Bone Marrow Transplantation
`
`-D
`
`Group A
`
`Group B
`
`2.0
`
`1.5
`
`1.0
`
`0.5
`
`0.0
`
`Absolute lymphocytes ¥ 109/1
`
`
`
`9
`
`833
`
`Adverse outcome of T cell-depleted allogeneic PBSCT
`R Chakraverty et al
`
`ing CD34+ selected cells from peripheral blood and a mean
`T cell dose of 0.02 × 106 cells/kg which resulted in a con-
`siderable morbidity and which was responsible for 85% of
`procedure-related deaths. It
`is possible that
`the use of
`additional
`in vivo TCD with CAMPATH 1H further
`reduced the effective T cell dose. These results suggest that
`consideration should be given to fixed dose donor T cell
`add-back in patients who have received PBSC grafts which
`are extensively (⬎5 log) T cell depleted or that measures
`taken to achieve in vivo TCD are omitted.
`
`Acknowledgements
`
`We would like to thank Nikki McKaeg, Judith Stuart and Brenda
`Miller for help in the collection and analysis of data. We would
`also like to thank Professor Paul Moss and Dr Charlie Craddock
`for their helpful criticisms regarding the manuscript.
`
`References
`
`1 Waldmann H, Polliak A, Hale G et al. Elimination of graft-
`versus-host disease by in vitro depletion of alloreactive
`lymphocytes with a monoclonal rat anti-human lymphocyte
`antibody (CAMPATH-1). Lancet 1984; 2: 483–6.
`2 Wagner JE, Donnenberg AD, Noga SJ et al. Lymphocyte
`depletion of donor bone marrow by counterflow centrifugal
`elutriation: results of a phase I clinical trial. Blood 1988; 72:
`1168–1176.
`3 Marmont AM, Horowitz MM, Gale RP et al. T cell depletion
`of HLA-identical transplants in leukemia. Blood 1991; 78:
`2120–2130.
`4 Young JW, Papadopoulos EB, Cunningham I et al. T-cell-
`depleted allogeneic bone marrow transplantation in adults with
`acute nonlymphocytic leukemia in first remission. Blood 1992;
`79: 3380–3387.
`5 Soiffer RJ, Fairclough D, Robertson M et al. CD6-depleted
`allogeneic bone marrow transplantation for acute leukemia in
`first complete remission. Blood 1997; 89: 3039–3047.
`6 Urbano-Ispizua A, Rozman C, Martinez C et al. Rapid
`engraftment without significant graft-versus-host disease after
`allogeneic transplantation of CD34+ selected cells from per-
`ipheral blood. Blood 1997; 89: 3967–3973.
`7 Cornetta K, Gharpure V, Mills B et al. Rapid engraftment
`after allogeneic transplantation using CD34-enriched marrow
`cells. Bone Marrow Transplant 1998; 21: 65–71.
`8 Kernan NA, Bordignon C, Heller G et al. Graft failure after
`T-cell-depleted human leukocyte antigen identical marrow
`transplants for leukemia: I. Analysis of risk factors and results
`of secondary transplants. Blood 1989; 74: 2227–2236.
`9 Apperley JF, Jones L, Hale G et al. Bone marrow transplan-
`tation for patients with chronic myeloid leukaemia: T cell
`depletion with Campath-1 reduces the incidence of graft-ver-
`sus-host disease but may increase the risk of leukaemic
`relapse. Bone Marrow Transplant 1986; 1: 53–66.
`10 Schattenberg A, De Witte T, Preijers F et al. Allogeneic bone
`marrow transplantation for
`leukemia with marrow grafts
`depleted of lymphocytes by counterflow centrifugation. Blood
`1990; 75: 1356–1363.
`11 Verdonck LF, de Gast GC, van Heugten HG, Dekker AW. A
`fixed low number of T cells in HLA-identical allogeneic bone
`marrow transplantation. Blood 1990; 75: 776–780.
`12 Aversa F, Tabilio A, Terenzi A et al. Successful engraftment
`of T-cell-depleted haploidentical
`‘three-loci’
`incompatible
`
`Bone Marrow Transplantation
`
`cause death in less than 10% of patients following standard
`T cell-replete SCT, with no difference between recipients
`of peripheral blood stem cells or bone marrow.32,33 In an
`early study of 31 patients, intensive TCD of bone marrow
`by an E-rosette technique and administration of a fixed T
`cell dose of 0.1 × 106 T cells/kg was associated with only
`three i