Apheresis instrument settings influence infused absolute ... - Nature

Bone Marrow Transplantation (2006) 37, 811–817 & 2006 Nature Publishing Group All rights reserved 0268-3369/06 $30.00

www.nature.com/bmt

ORIGINAL ARTICLE

Apheresis instrument settings influence infused absolute lymphocyte count affecting survival following autologous peripheral hematopoietic stem cell transplantation in non-Hodgkin’s lymphoma: the need to optimize instrument setting and define a lymphocyte collection target R Katipamula1, LF Porrata2, DA Gastineau1,2, SN Markovic2, SB Moore1, C Greiner1, EA Burgstaler1, DJ Padley1 and JL Winters1 1 Division of Transfusion Medicine, Division of Hematology, Mayo Clinic, Rochester, MN, USA and 2Division of Hematology/ Department of Internal Medicine, Mayo College of Medicine, Rochester, MN, USA

Autograft absolute lymphocyte count (A-ALC) is an independent prognostic factor for survival after autologous peripheral blood hematopoietic stem cell transplantation (APHSCT) for non-Hodgkin’s lymphoma (NHL). Factors enhancing A-ALC collections are unknown. We hypothesize that apheresis instrument settings could affect A-ALC. Data from 127 NHL patients collected from 15 January 1999 to 30 July 2004 using a single apheresis instrument (COBE Spectra (SP), Baxter Amicus (AM), and CS3000 Plus (CS)) were analyzed. The primary end point of the study was to assess the correlation between apheresis instrument settings and A-ALC. The secondary end point was to determine the effect of apheresis instrument on survival post-APHSCT. Patients collected using SP achieved higher A-ALC compared to AM (with modified settings) or CS (Po0.05) and demonstrated superior overall (OS) and progression-free survival (PFS) (Po0.03). Multivariate analysis demonstrated A-ALC and not the apheresis instrument as an independent prognostic factor for OS and PFS, cancelling the prognostic effect of the apheresis instruments observed in the univariate analysis. The survival advantage observed by SP was from the higher A-ALC collected compared to AM and CS. These data suggest that apheresis instrument settings should be optimized to collect CD34 þ cells as well as an A-ALC target, with direct impact on survival post-APHSCT. Bone Marrow Transplantation (2006) 37, 811–817. doi:10.1038/sj.bmt.1705338; published online 13 March 2006 Keywords: apheresis instruments; autograft absolute lymphocyte count; non-Hodgkin’s lymphoma; autologous peripheral blood hematopoietic stem cell transplantation; survival

Correspondence: Dr LF Porrata, Division of Hematology/Department of Internal Medicine, Mayo College of Medicine, 200 First St SW, Rochester, MN 55902, USA. E-mail: [email protected] Part of these data were presented at the 2005 American Society of Apheresis National meeting on April 28, 2005 – Plenary Session. Received 22 June 2005; revised 11 January 2006; accepted 14 February 2006; published online 13 March 2006

Introduction Absolute lymphocyte count at day 15 (ALC-15) has been reported to be an independent prognostic factor for patients undergoing autologous peripheral blood hematopoietic stem cell transplantation (APHSCT) for acute myelogenous leukemia,1 metastatic breast cancer,2,3 Hodgkin’s disease,4,5 non-Hodgkin’s lymphoma (NHL),5–7 multiple myeloma7 and primary systemic amyloidosis.8 ALC-15 is directly dependent on the collected and infused autograft absolute lymphocyte count (A-ALC).9,10 Identification of strategies to increase lymphocyte collection during autologous peripheral blood stem cell collection will translate into a faster immune recovery and better survival postAPHSCT. To assess whether apheresis instrument setting is associated with A-ALC yield, we analyzed the effect of apheresis instruments used in our facility on A-ALC.

Patients, materials and methods Patient population Approval for conduct of this retrospective study was obtained from the Mayo Clinic Institutional Review Board and was in accordance with US federal regulations and the Declaration of Helsinki. Written, informed consent was obtained from all the patients allowing the use of their medical records for medical research. We reviewed the medical records of 294 patients with NHL who underwent APHSCT between 15 January 1999 to 30 July 2004. A computer database containing the characteristics of the peripheral blood stem cell collections from these patients was examined to determine the apheresis instruments used for collection. Instrument assignment was random and based solely upon availability of the instrument during the time period studied. Of the 294 NHL patients, 127 (43%) were collected with a single type of apheresis instrument and only this group was included for our study. Patient data regarding the response to therapy, relapse and survival were maintained in a second clinical database, which was examined to obtain survival data.

Apheresis instrument settings on lymphocyte yield R Katipamula et al

812

End points The primary end point was to correlate the A-ALC yield with the different types of apheresis instrumentation used for collection. The secondary end point of the study was to evaluate the effect of apheresis instrumentation on survival of NHL patients who underwent APHSCT. Pre-mobilization ALC was calculated from a complete blood cell count (CBC) performed before the start of growth factor for stem cell mobilization. The absolute white blood cell count (WBC) from the autograft was calculated as follows: autograft bag volume (ml) autograft WBC cells/ml (million cells/ml) 0.001. The autograft WBC was obtained using the ACT 10 Series Analyzer Coulter, Miami, FL, USA and % lymphocytes was determined microscopically using Wright stain. The ALC15 was calculated from a CBC performed on day 15 following transplant. The A-ALC for each collection was calculated as follows: (total absolute WBC)(% lymphocytes). ALC/kg for each collection was calculated as follows: (total absolute WBC)(% lymphocytes)/kilogram. A-ALC/l for each collection was calculated as follows: (total absolute WBC)(% lymphocytes)/liter of blood processed. Pre-apheresis absolute lymphocyte count (PAALC) was obtained form the CBC before the start of each collection. The collection efficiency for lymphocytes was calculated as follows: lymphocytes collected divided by lymphocytes that passed through the machine. Lymphocytes that passed through the machine was calculated as pre-lymphocyte count þ post-lymphocyte count/2 volume of blood processed (corrected for anticoagulant). We did not include the collection efficiency for the Fenwal CS3000 (CS) because we only had data on post-procedure CBC counts for five collections. Of the 55 procedures performed with the Spectra (SP), we did not have data on seven (13%), and of the 129 procedures performed with the Amicus (AM), we did not have data on 19 (15%). Prognostic factors Prognostic factors for post transplant overall survival (OS) and progression-free survival (PFS) evaluated in this study included age at transplantation, A-ALC/kg, apheresis instrument used for collection, extra-nodal disease status, pre-transplant lactate dehydrogenase (LDH), number of prior chemotherapy treatment regimens before transplant, performance status, pre-transplant clinical response as a result of salvage chemotherapy (complete remission vs partial remission), post transplant cytokine therapy (granulocyte-colony stimulating factor (G-CSF) vs granulocyte– macrophage colony stimulating factor (GM-CSF)) and stage of disease. Additional factors evaluated in association with A-ALC included gender, CD34 þ cell dose and premobilization ALC. Further factors evaluated in association with ALC-15 included conditioning regimens, gender, CD34 þ cell dose and pre-mobilization ALC. CD34 enumeration Fifty microliter of fresh hematopoietic progenitor cell product (obtained by reverse pipetting) was stained with 20 ml of the control reagent (nucleic acid dye/IgG1 PE/ CD45 PerCP, ProCount Progenitor Cell Enumeration Kit, Bone Marrow Transplantation

Becton Dickison Immunocytometry System, San Jose, CA, USA) and 20 ml of CD34 regent (nucleic acid dye/CD34 PE/ CD45 PerCP, ProCount Progenitor Cell Enumeration Kit, Becton Dickison Immunocytometry System, San Jose, CA, USA). Both tubes (control and CD34-labeled tubes) were incubated for 15 min at room temperature in the dark. A measure of 450 ml of 1 FACS Lysing Solution was added to each tube and incubated for 30 min at room temperature in the dark. The CD34 counts were obtained using a threecolor assay on a FACSCalibur flow cytometry and analyzed using ProCount software (BD Bioscience, San Jose, CA, USA).

Peripheral blood stem cell (autograft) collections The three different types of instruments used at our facility during the period examined were the COBE Spectra (SP) (Gambro BCT, Lakewood, CO, USA), Baxter Amicus (Baxter Healthcare, Deerfield, IL, USA) and CS (Baxter Healthcare, Deerfield, IL, USA). All patients were collected using a single instrument type based on availability of the instrument on the day of collection. Patients received G-CSF for mobilization at a dose of 10 mg/kg daily for 5–7 consecutive days by subcutaneous injection. None of the patients received chemotherapy or other mobilizing agents. Once their peripheral blood CD34 þ cell count was X10 cells/ml, patients began daily apheresis until a minimum target of 2.0 106 CD34 þ cells/kg was reached. Multiple CS instruments were used with 4 and 5 h blood processing times. The modified platelet collection procedure for mononuclear cell (MNC) collection was used (interface offset setting of 100). The anticoagulant ACD-A (Baxter Healthcare Corp., Deerfield, IL, USA) ratio was 11:1. Draw rates were set at 50 ml/min, but were changed to 40 ml/min when required owing to poor access. The actual blood processing time (when pumps were running) was used as the end point. Multiple SP instruments Software Version 7 was used with 4 h (SP4) and 5 h (SP5) blood processing times. Fivehour large volume leukapheresis procedures (SPLVL) were also performed. For the standard SP4 and SP5 procedures, the draw rates were determined by the instrument according to patient blood volume. Draw rates of p100 ml/min were used for the SPLVL depending on adequacy of the access. The MNC collect rate for all procedures was 1.0 ml/min and the color goal was 1–3% hematocrit (HCT) on the SP color chart. The anticoagulant infusion rate for all SP procedures was 1.0 ml/min/l total blood volume. The anticoagulant ratio was 12:1 for the SP4 and SP5 procedures and 26:1 for the SPLVL procedures. ACD-A was used as the anticoagulant for most of the SP4 and SP5 procedures. However, if citrate toxicity was encountered, a mixture of ACD-A, 0.9% sodium chloride and heparin was used. In the first bag, 500 ml ACD-A, 500 ml 0.9% sodium chloride and 7000 U of heparin were combined. If a second bag was needed, 500 ml ACD-A, 500 ml 0.9% sodium chloride and 3500 U of heparin were combined. An ACDA-heparin solution was used in the SPLVL. To a 1 l bag of ACD-A, 10 000 U of heparin was added. Additionally, 30 ml of the ACD-A-heparin solution was added to the collection bag at the beginning of the procedure to help anticoagulate the collection.


813

Multiple Fenwal Amicus separators with software version 2.50 were used with 4 h (AM4) and 5 h regular (AM5) and large volume leukapheresis (AMLVL). The AMLVL procedure has been described in detail.11 Whole blood flow rates of 90 ml/min (maximum recommended by the manufacturer) were used when possible for the LVL procedures. The instrument determined the whole blood flow rates for the AM4 and AM5 procedures according to patient blood volume. Cycle volumes of 1000 ml/cycle were used for patients with pre-procedure WBC counts 435 109/l for the AM4 and AM5 procedures and 1400 ml/cycle for pre-procedure WBC counts p35 109/l. Cycle volumes of AMLVL were 1400 ml/cycle for preprocedure WBC counts p35 109/l and 1000, 1400 or 3000 ml for pre-procedure WBC counts 435 109/l. The 4 h procedures used the manufacturer default settings of MNC sense ¼ 0.45, MNC offset ¼ 2.3, RBC offset ¼ 6.0 and plasma flush ¼ 10. Owing to concern about total white blood cell content in the collection (number of freezer bags required), the default settings were changed to reduce white blood cell content in the regular and LVL 5 h procedures. Previous studies indicated that the new settings could be used without significant loss of CD34 þ cells.11 The settings used were MNC sense ¼ 0.45, MNC offset ¼ 1.5, RBC offset ¼ 5.0 and plasma flush ¼ 8. The citrate infusion rate (CIR) for the AM4 and AM5 procedures was 1.25 mg/kg/ min and the anticoagulant was ACD-A except in cases of citrate toxicity. When citrate toxicity occurred, the ACDA-heparin solutions detailed in the SP procedure were used. The CIR for AMLVL was 2.50 mg/kg/min and the anticoagulant was the ACD-A-heparin solutions used for citrate toxicity. In addition, for AMLVL only, 35, 30 and 25 ml of undiluted ACD-A were added to the collection bag for cycle volumes of 1000, 1400 and 3000 ml, respectively. Anticoagulant ratio of 12:1 was used for AM4 and AM5 and 13:1 for AMLVL.

Conditioning regimens Of the cohort of 127 patients, 121 patients received BEAM (BCNU (300 mg/m2), etoposide (100 mg/m2), ARA-C (100 mg/m2) and melphalan (140 mg/m2)), five patients received BEAC (BCNU (300 mg/m2), etoposide (100 mg/m2), ARA-C (100 mg/m2), and cyclophosphamide (35 mg/kg)) and one patient received cyclophosphamide (60 mg/m2) and total body irradiation (12 Gy). Response and survival Response criteria were based on the guidelines from the NHL International Workshop.12 Complete response (CR) was defined as complete regression of all measurable or evaluable disease including radiologically demonstrable disease, BM involvement or peripheral blood involvement. Partial response (PR) was defined as a reduction in the sum of the products of measurable lesions’ longest diameter and perpendicular diameters of 50% or greater, with a 50% or greater decrease in hepatomegaly or splenomegaly (measured from the costal margin), if there was previous known liver or spleen involvement. Stable disease was defined as less than PR, but was not progressive disease. Disease progression was defined as a 50% or more increase in the

sum of the products of the longest diameter and the perpendicular diameter of measurable lesion(s) from the pre-study measurement, the appearance of new lesions or a 2-cm increase in spleen or liver size owing to lymphoma. Relapsed disease was defined as the appearance of any new lesion or increase by 50% or more in the size of previously involved sites. Overall survival was measured from the date of transplantation to the date of death or last follow-up. Progression-free survival was defined as time from transplantation to disease progression, relapse, death or last follow-up. Those who died were considered to have had disease progression unless documented evidence clearly indicated that no progression had occurred.

Statistical analysis Overall survival and PFS were analyzed for each instrument using the method described by Kaplan and Meier.13 The differences between survival curves were tested for statistical significance using the two-tailed log-rank test. The Cox proportional hazard model14 was used to assess apheresis instruments as a prognostic factor for OS and PFS and to adjust for other known risk factors. Risk ratios reported are for risks associated with SP vs Baxter AM and CS. Other prognostic factors tested included age at transplant (X60 vs o60 years), A-ALC (X0.5 vs o0.5 109 lymphocytes/kg based upon our previous publication10), extra-nodal sites (X2 vs o2), LDH (abnormal vs normal), number of prior treatments before transplant, performance status (X2 vs o2), post transplant cytokines (G-CSF vs GM-CSF), pre-transplant clinical response (CR vs PR) and stage (III/IV vs I/II). Multivariate analysis using Cox regression model tested all variables with a Po0.2 in the univariate analysis. Prediction of AALC collection was explored in logistic regression models, univariately assessing continuous and dichotomized values of the prognostic factors described above. w2-test was used to determine relationships between categorical variables; the Wilcoxon rank-sum tests or the Kruskal–Wallis test was used to assess whether continuous variables differed significantly among categories and Spearman correlation coefficients were used to evaluate associations for continuous variables. All P-values represented were two-sided, and a P-value of o0.05 was considered significant.

Results Patient characteristics A total of 127 NHL patients were identified who had their peripheral blood stem cell (PBSC) collections using a single type of apheresis instrument. Median age for the cohort at the time of the transplant was 55 years (range, 22–76). Baseline characteristics of the patients are listed in Table 1. Except for the conditioning regimens and post transplant cytokines administered, no differences were noted with regard to patient characteristics or prognostic factors among the patients who underwent stem cell collections on the different apheresis instruments. None of the patients received purged or CD34-selected stem cells. None of the Bone Marrow Transplantation


814 Table 1

Baseline characteristics Fenwal CS3000 Plus (N ¼ 48)

COBE Spectra (N ¼ 29)

Characteristics

Baxter Amicus (N ¼ 50)

Median age at time of transplant (range) Gender Female

56 (22–73) 53 (23–76) 53 (22–76)

13

13

Histology DLC Mantle Follicular Ana-T Ana-B

34 7 5 1 1

37 4 3 2 0

20 3 5 0 0

Age X60

18

15

10

5

2

1

21

19

12

0

3

0

11 39

14 34

9 20

LDH Abnormal Performance status X2 Stage I/II III/IV Number of pre-transplant treatments 1 2 3 4 5 Pre-transplant clinical response CR PR Conditioning regimens BEAC BEAM CTX/TBI Post transplant cytokines G-CSF GM-CSF

0.59 0.141

12

Extranodal sites X2

P-value

0.736

0.881

0.39

0.969

0.08

0.605

0.213 8 28 9 4 1

6 26 12 3 1

4 23 2 0 0 0.155

7 43

8 40

1 28

0 50 0

5 42 1

0 29 0

1 49

9 39

1 28

o0.016

o0.0067

Abbreviations: Ana-B ¼ anaplastic B cell; Ana-T ¼ anaplastic T cell; BEAC ¼ BCNU, etoposide, ARA-C, cyclophosphamide; BEAM ¼ BCNU, etoposide, ARA-C, melphalan; CR ¼ complete response; CTX ¼ cyclophosphamide; DLC ¼ diffuse large cell; G-CSF ¼ granulocyte-colony stimulating factor; GM-CSF ¼ granulocyte–macrophage colony stimulating factor; IPI ¼ International prognostic index; LDH ¼ lactate dehydrogenase; PR ¼ partial response; TBI ¼ total body irradiation.

patients developed clinically evident autologous graft vs host disease.

Influence of apheresis instrument on autograft absolute lymphocyte count To identify the apheresis instrument factors influencing AALC collections, we compared the collection characteristics among the three different instruments listed in Table 2. We identified that patients collected with SP harvested higher Bone Marrow Transplantation

total absolute numbers of A-ALC compared to either the AM or the CS. Using liters, kilograms or CD34 þ cell count as denominators to standardize the comparison between the apheresis instruments, SP achieved a higher A-ALC/l, A-ALC/kg or A-ALC/CD34 þ compared to AM and CS. Of note, there was no difference between the instruments for CD34 þ cell yield. To rule out the possibility that there was any difference in the number of lymphocytes at the time of collection (PA-ALC), we analyzed the total PAALC between the machines and found no difference. The discrepancy between the absolute number of lymphocytes and collection efficiency may be due to the following reasons: (1) collection efficiency is dependent on the volume of blood processed; (2) SP usually processes more blood than AM; (3) less lymphocytes will probably pass through the AM during a procedure; and (4) it is difficult to compare absolute number of lymphocytes collected and collection efficiencies with different volumes processed. We previously published a strong positive correlation between A-ALC/kg and ALC-15.11 We continue to observe the same strong correlation between A-ALC/kg and ALC15 in the study (rs ¼ 0.60, Po0.0001). Therefore, not surprisingly, the higher A-ALC/kg collected by SP led to a higher ALC-15 recovery. There was no difference between the patients in the three groups with regard to pre-mobilization ALC, whole blood processed and collected/infused CD34 þ cells. We identified no association between A-ALC and age at transplant (P ¼ 0.12), CD34 product yield (P ¼ 0.12), extranodal disease (P ¼ 0.11), gender (P ¼ 0.42), histology (P ¼ 0.89), LDH (P ¼ 0.50), number of prior treatments before transplantation (P ¼ 0.10), pre-transplant clinical response (P ¼ 0.16), performance status (P ¼ 0.12) and stage (P ¼ 0.14). No association was also identified between ALC-15 and age at transplant (P ¼ 0.37), CD34 þ cell product yield (P ¼ 0.51), conditioning regimens (P ¼ 0.14), extra-nodal disease (P ¼ 0.36), gender (P ¼ 0.25), histology (P ¼ 0.85), LDH (P ¼ 0.59), number of prior treatments before transplantation (P ¼ 0.38), pre-transplant clinical response (P ¼ 0.11), performance status (P ¼ 0.15), posttransplant cytokines (P ¼ 0.10) and stage (P ¼ 0.33).

Survival based on apheresis instrument Patients were followed for a median time of 13 months (range, 1–62). By May 2005, 45 deaths had occurred among the 127 patients in this study. All of the 45 deaths were due to the patients’ NHL. The median 3-year OS (Figure 1) was significantly better for patients collected using SP compared with patients collected using the AM (modified settings) or CS (Po0.0286). As with OS, the 3-year PFS rate was superior in patients who underwent autograft collections with SP (74%) vs AM (47%) or CS (34%) (Figure 2). Patients infused with an A-ALCX0.5 109 lymphocytes/ kg experienced a higher 3-year OS rate compared with patients infused with an A-ALCo0.5 109 lymphocytes/kg (71 vs 49%, respectively, Po0.0127). Three-year PFS rate was also higher in patients infused with an A-ALC X0.5 109 lymphocytes/kg vs patients infused with an A-ALCo0.5 109 lymphocytes/kg (62 vs 40%, respectively, Po0.014).


815 Collection comparison between apheresis instruments

Table 2

Characteristics

Baxter Amicus

Pre-mobilization ALC, median (range) PA-ALC of total collection, median (range) Whole blood processed per collection median (range) CD34+ cells collected, median (range) A-ALC of total collection, median (range) A-ALC/l of total collection, median (range) ((A-ALC/l)/PA-ALC), median (range) A-ALC/kg of total collection, median (range) A-ALC/CD34+ of total collection ALC-15, median (range)

1.3 4.1 15 4.79 32.4 0.021 5.1 0.37 80.2 0.39

Fenwal CS3000 Plus

(0.46–12.07) (2.02–8.6) (9–23) (2.0–9.95) (7.8–100.9) (0.004–0.054) (2–6.2) (0.11–1.09) (13.3–290.8) (0.02–1.09)

1.45 4.46 12 4.12 35.6 0.026 6 0.41 106 0.29

(0.33–8.14) (1.62–9.8) (12–13) (2.01–36.07) (3.78–151.53) (0.006–0.149) (3.7–15.2) (0.055–1.3) (5.26–494.7) (0.06–1.1)

COBE Spectra 1.3 4.3 18 4.01 44.6 0.038 8.8 0.52 130.2 0.57

(0.21–4.3) (1.56–8.4) (8–28) (2.46–7.11) (9.63–127.2) (0.01–0.144) (6.4–17.1) (0.136–1.51) (31.1–469.4) (0.1–1.81)

P-value* 0.422 0.858 0.375 0.244 o0.05 o0.0002 o0.0001 o0.05 o0.043 o0.0001

Abbreviations: A-ALC ¼ absolute, autograft-absolute lymphocyte count harvested 109 for the total collection; A-ALC/l ¼ autograft absolute lymphocyte divided by whole blood processed in liters 106; A-ALC/kg ¼ autograft absolute lymphocyte count divided by patients’ weight in kilograms 109; A-ALC/ CD34+ ¼ absolute, autograft-absolute lymphocyte count divided by the stem cells/kg 103; ALC-15 ¼ absolute lymphocyte count at day 15 post transplant 109/l; CD34+ cells for total collection ¼ number of CD34+ cells harvested 106/kg; pre-mobilization ALC ¼ absolute lymphocyte count before growth factor injection for stem cell mobilization 109/l; whole blood processed ¼ the total amount of whole blood processed per collection per machine in liters; PA-ALC ¼ pre-apheresis absolute lymphocyte count 109; ((A-ALC/l)/PA-ALC) ¼ ratio showing the normalization between A-ALC/l to the PA-ALC 106. *P-values shown above are for COBE Spectra vs Baxter Amicus vs CS3000 Plus using the Kruskal–Wallis test.

1.0

Spectra, N = 29

0.9

Surviving

0.8

Surviving

0.7 0.6

Amicus, N = 50

0.5 0.4

Fenwal CS3000+, N = 48

0.3

P < 0.0286

0.2

Spectra, N = 13 Amicus, N = 14 Fenwal CS3000+, N = 19 P = 0.1434

0

0.1 0.0 0

10

20

30

40

50

60

Survival time in months

Figure 1 Kaplan–Meier estimates of overall survival (OS) of patients collected on COBE Spectra (SP), Baxter Amicus (AM) and CS3000 Plus (CS). The median OS was not reached for patients collected on SP and AM, and was 32 months for CS. The OS rates at 3 years were 84, 54 and 39%, respectively (Po0.0286).

1.0 0.9

Spectra, N = 29

0.8 0.7 Surviving

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

0.6 0.5

Fenwal CS3000+, N = 48

0.4

Amicus, N = 50

0.3 0.2

P < 0.03

0.1 0.0 0

10

20

30

40

50

60

Progression-free survival in months

10

20 30 40 50 Survival time in months

60

Figure 3 Kaplan–Meier estimates of overall survival (OS) comparing the COBE Spectra (SP), Baxter Amicus (AM) and CS3000 Plus in the subgroup of patients that collected A-ALCX0.5 109 lymphocytes/kg. No statistical difference in OS was observed between the apheresis instruments (P ¼ 0.1434).

To determine if the survival advantage post-APHSCT was owing to the apheresis instruments or solely owing to the number of A-ALC collected and infused, we compared survival between the apheresis instruments using the common denominator of A-ALC/kg. There was no difference between the instruments in OS (P ¼ 0.1434) (Figure 3) and PFS (P ¼ 0.231) for patients infused with an A-ALCX0.5 109 lymphocytes/kg. The same was true for patients infused with an A-ALCo0.5 109 lymphocytes/kg (OS (P ¼ 0.257), PFS (P ¼ 0.148)). Using the apheresis instruments as the common denominator, we identified superior OS and PFS for patients infused with an A-ALCX0.5 109 lymphocytes/kg red to patients infused with an A-ALCo0.5 109 lymphocytes/kg in the AM group (OS Po0.05; Po0.043), SP group (OS Po0.04; PFS Po0.045) and the CS group (OS Po0.022; PFS Po0.017).

Figure 2 Kaplan–Meier estimates of progression-free survival (PFS) of patients collected on COBE Spectra (SP), Baxter Amicus (AM) and CS3000 Plus (CS). The median PFS was not reached for patients collected on SP, and was 14 months for AM, and 21 months for CS. The PFS rates at 3 years were 74, 47 and 34%, respectively (Po0.03).

Univariate/multivariate analysis Applying univariate analysis, A-ALC/kg, stage of the disease and the apheresis instruments with different settings Bone Marrow Transplantation


816 Table 3

Univariate analysis for overall and progression-free survival

Factors

Overall survival

Age X60 vs o60 years A-ALC X0.5 vs o0.5 109 lymphocytes/kg Extra-nodal sites X2 vs o2 LDH 4normal vs normal Number of treatments before transplant Performance status X2 vs o2 Post transplant cytokines: G-CSF vs GM-CSF Pre-transplant clinical response: CR vs PR Stage III/IV vs I/II COBE Spectra vs Baxter Amicus/Fenwal CS3000 Plus

Table 4

Progression-free survival

RR

95% CI

P-value

RR

95% CI

P-value

1.001 0.708 1.078 0.960 0.910 0.881 0.859 0.947 1.421 0.524

0.725–1.354 0.492–0.979 0.530–1.790 0.703–1.290 0.611–1.315 0.209–1.887 0.580–1.394 0.584–1.402 0.995–2.163 0.287–0.823

0.994 o0.0363 0.806 0.789 0.628 0.794 0.505 0.803 o0.05 o0.0034

0.994 0.753 0.873 0.956 0.998 1.072 0.815 1.044 1.306 0.608

0.748–1.292 0.556–0.993 0.431–1.437 0.728–1.240 0.717–1.357 0.433–1.927 0.577–1.233 0.707–1.453 0.967–1.836 0.391–0.873

0.964 o0.0445 0.633 0.738 0.989 0.851 0.309 0.814 0.084 o0.0054

Multivariate analysis for overall and progression-free survival

Factors

Overall survival

A-ALC X0.5 vs o0.5 10 lymphocytes/kg Stage III/IV vs I/II COBE Spectra vs Baxter Amicus/Fenwal CS3000 plus 9

RR

95% CI

P-value

RR

95% CI

P-value

0.555 1.342 0.761

0.303–0.877 0.938–2.045 0.528–1.056

o0.0092 0.1116 0.1044

0.640 1.237 0.794

0.411–0.923 0.914–1.741 0.586–1.050

o0.0156 0.1251 0.1092

used for collection were significant predictors for OS and PFS (Table 3). In multivariate analysis, A-ALC/kg was an independent predictor for OS (RR ¼ 0.55; Po0.0092) and PFS (RR ¼ 0.64; Po0.0156) when compared to the significant predictors identified in the univariate analysis, including stage of the disease and apheresis instrument used for collection (Table 4).

Discussion In this study, we set out to determine if the apheresis instrument settings influence A-ALC collection and clinical outcomes of NHL patients treated with APHSCT. The stratification of our cohorts based on the apheresis instruments was balanced for all patients’ characteristics with the exception of conditioning regimens and choice of post transplant cytokine use. The reason for this difference is based on our evolving clinical practice, as BEAM and GM-CSF have become our standard conditioning regimen and post transplant cytokine, respectively. Despite a statistical difference, the majority of patients received BEAM conditioning (100% in the AM group, 88% in the CS group and 100% in the SP group) and GM-CSF (98% in the AM group, 81% in the CS group and 97% in the SP group). We identified no association between patients’ characteristics or prognostic factors and A-ALC or ALC15. In addition, all patients were mobilized with the same stem cell mobilization regimen (G-CSF without chemotherapy or other mobilization agents) and there was no difference between the groups examined regarding the baseline ALC before mobilization. This suggests that the main factor affecting A-ALC occurs during the stem cell collection (choice of apheresis instrument). The presented data support the conclusion that the choice of apheresis instrumentation and instrument settings affects A-ALC Bone Marrow Transplantation

Progression-free survival

(and ALC-15), thereby directly impacting the clinical end points (OS/PFS). Our study shows a strong association between the apheresis instrument settings and A-ALC. Patients who were collected by the SP achieved higher A-ALC compared to the AM or CS. This difference is explained by the fact that the settings on SP are set to harvest a wider mononuclear cell band, whereas the settings on AM are set to reduce white blood cell content for the same number of CD34 þ cells. This may explain the higher A-ALC to CD34 ratio in the SP cohort. Thus, modification of stem cell collections with optimum instrument settings may directly impact A-ALC and result in improved OS/PFS. The AM collections at our institution were 5 h collections modified to reduce total white blood cell content in the collection. Our goal during the time period examined was to collect as many CD34 þ cells as possible with the least amount of contaminating red blood cells, granulocytes and lymphocytes. The modified settings started and stopped collecting cells earlier. The default settings or more aggressive settings would collect additional lymphocytes but also more granulocytes and other white blood cells requiring more freezer bags for storage and a larger product volume at the time of infusion.11 The more aggressive settings were used by Ikeda et al. when comparing the AM to the SP.15 They found equivalent CD34 þ cell yields, white blood cell content, mononuclear cell content and colony-forming units granulocyte/macrophage (CFU-GM). They found similar collection efficiencies for CD34 þ cells, white blood cells and mononuclear cells but significantly lower collection efficiencies for platelets.16 Significantly less whole blood was processed in our CS collection compared to the SP collections. This was because of the lower flow rates used for CS. Previous work showed that the slower flow rates actually yielded more MNC than


817

high flow rates, likely owing to more dwell time in the centrifuge.17 The smaller amount of whole blood processed probably explains the lower numbers of lymphocytes in CS vs SP collections. The CS was the machine of choice for the first 4 years of the study and then was replaced by SP and AM owing to higher CD34-positive cell yields and pending obsolescence of the CS by Baxter. In the univariate analysis, we identified a superior OS and PFS for patients collected by SP compared with AM or CS. However, in the multivariate analysis, the prognostic effect of the apheresis instruments was canceled by the AALC/kg. This is important to demonstrate that the survival benefit is not owing to the specific type of apheresis instrument but owing to the ability of the apheresis instrument to harvest more lymphocytes leading to a faster immune recovery post-APHSCT. As superior OS and PFS was achieved in patients with X0.5 109 lymphocytes/kg,10 it would be prudent to establish a minimum lymphocyte target in addition to the CD34-positive cell target. This would equalize all instruments in the A-ALC dose. We are currently conducting a prospective study directed at identifying optimal apheresis instrument settings that will allow maximal A-ALC and CD34 þ cell collections while minimizing platelet loss using AM. Once optimal conditions are defined, we will evaluate their impact on clinical end points. In summary, the current report expands our prior work underscoring the importance of lymphocyte recovery postAPHSCT. This is the first study showing the impact of apheresis instrumentation and their settings on A-ALC and survival. We hope that these data will support further investigations directed at maximizing post-APHSCT lymphocyte (immune) recovery, as the impact of such interventions appears to have direct consequences on clinical outcomes in patients with multiple hematological malignancies.

5

6

7

8

9

10

11

12

References 1 Porrata LF, Litzow MR, Tefferi A, Letendre L, Kumar S, Geyer SM et al. Early lymphocyte recovery is predictive factor for prolonged survival after autologous hematopoietic stem cell transplantation for acute myelogenous leukemia. Leukemia 2002; 16: 1311–1318. 2 Porrata LF, Ingle JN, Litzow MR, Geyer S, Markovic SN. Prolonged survival associated with early lymphocyte recovery after autologous hematopoietic stem cell transplantation for patients with metastatic breast cancer. Bone Marrow Transplant 2001; 28: 865–871. 3 Nieto Y, Shpall EJ, Mcniece IK, Nawz S, Beaudet J, Rosinski S et al. Prognostic analysis of the early lymphocyte recovery in patients with advanced breast cancer receiving high-dose chemotherapy with an autologous hematopoietic progenitor cell transplant. Clin Cancer Res 2004; 10: 5076–5086. 4 Porrata LF, Inwards DJ, Micallef IN, Ansell SM, Geyer SM, Markovic SN. Early lymphocyte recovery post autologous

13 14 15

16

17

haematopoietic stem cell transplantation is associated with better survival in Hodgkin’s lymphoma. Br J Haematol 2002; 117: 629–633. Gordan LN, Sugrue MW, Lynch JW, Williams KD, Khan SA, Mored JS. Correlation of early lymphocyte recovery and progression-free survival after autologous stem-cell transplant in patients with Hodgkin’s and non-Hodgkin’s lymphoma. Bone Marrow Transplant 2003; 31: 1009–1013. Kim H, Sohn HJ, Kim SE, Kang HJ, Park S, Kim S et al. Lymphocyte recovery as a positive predictor of prolonged survival after autologous peripheral blood stem cell transplantation in T-cell non-Hodgkin’s lymphoma. Bone Marrow Transplant 2004; 34: 43–49. Porrata LF, Gertz MA, Inwards DJ, Litzow MR, Lacy MQ, Tefferi A et al. Early lymphocyte recovery predicts superior survival after autologous hematopoietic stem cell transplantation in multiple myeloma or non-Hodgkin’s lymphoma. Blood 2001; 98: 579–585. Porrata LF, Gertz MA, Litzow MR, Lacy MQ, Dispenzieri A, Inwards DJ et al. Early lymphocyte recovery predicts superior survival after autologous hematopoietic stem cell transplantation for patients with primary systemic amyloidosis. Clin Cancer Res 2005; 11: 1210–1218. Porrata LF, Gertz MA, Geyer SM, Litzow MR, Gastineau DA, Moore SB et al. The dose of infused lymphocytes in the autograft directly correlates with clinical outcomes after autologous peripheral blood hematopoietic stem cell transplantation in multiple myeloma. Leukemia 2004; 18: 1085–1092. Porrata LF, Litzow MR, Inwards DJ, Gastineau DA, Moore SB, Pineda AA et al. Infused peripheral blood autograft absolute lymphocyte count correlates with day 15 absolute lymphocyte count and clinical outcome after autologous peripheral hematopoietic stem cell transplantation in non-Hodgkin’s lymphoma. Bone Marrow Transplant 2004; 33: 291–298. Burgstaler EA, Pineda AA, Winters JL. Hematopoietic progenitor cell large volume leukapheresis (LVL) on the Fenwal amicus blood separator. J Clin Apheresis 2004; 19: 103–111. Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. J Clin Oncol 1999; 17: 1244–1253. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481. Cox DR. Regression models and life tables. J R Stat Soc B 1972; 34: 187–202. Snyder EL, Baril L, Cooper DL, Min K, Mechanic S, Stoddart L et al. In vitro collection and post transfusion engraftment characteristics of MNCs obtained by using a new separator for autologous PBPC transplantation. J Clin Apheresis 2000; 40: 961–967. Ikeda K, Ohto H, Nemoto K, Yamamoto G, Kato K, Ogata T et al. Collection of MNCs and progenitor cells by two separators for PBPC transplantation: a randomized crossover trial. J Clin Apheresis 2003; 43: 814–819. Lin J, Burgstaler EA, Pineda AA, Gertz MA. Effects of whole blood flow rates on mononuclear cell yields during peripheral blood stem cell collection using Fenwal CS3000 plus. J Clin Apheresis 1995; 10: 7–11.

Bone Marrow Transplantation