Cytotherapy, 2010; 12: 418–424

Correspondence: William E. Janssen, PhD, Blood and Marrow Transplant, H. Lee Moffi tt Cancer Center, 12902 Magnolia Drive, Tampa, FL 33612, USA. E-mail: William.Janssen@Moffi tt.org

(Received 29 June 2009; accepted 10 October 2009)

ISSN 1465-3249 print/ISSN 1477-2566 online © 2010 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)DOI: 10.3109/14653240903479663

Large-scale Ficoll gradient separations using a commercially available, effectively closed, system

WILLIAM E. JANSSEN1,2, ALBERT RIBICKAS2, LAURA V. MEYER2 & RENEE C. SMILEE1,2

1Department of Blood and Marrow Transplant, and 2Cell Therapies Core Facility, Moffi tt Cancer Center, Tampa, Florida, USA

Abstract Background aims. Multiple cell-therapy products require density separation as a part of manufacturing. The traditional method for Ficoll separation, layering cell suspensions over Ficoll in tubes, followed by centrifugation and collection of cells from the interface, is too cumbersome and poses too high a risk of contamination for clinical-scale use. Recently, a system for clinical-scale Ficoll gradient applications has been introduced (Sepax™) but this system has limited availability and is costly. Methods. For preparations of mononuclear cells (MNC) for dendritic cell (DC) production, we developed a Ficoll separation protocol that employs the Haemonetics™ Cell Saver5™ surgical blood salvage and wash instrument. This system uses standard blood bags and tubing, has single-use components, and is effectively closed. We analyzed 37 recent separation processes using this instrument and protocol. We measured depletion of red blood cells (RBC) and polymor-phonuclear leukocytes (PMN), and recovery of CD14� monocytes and MNC. Results. Starting cell counts were 14.6 � 8.0 (�109). Total cell recovery was 49.2 � 15.2%, RBC depletion was 88.4 � 2.8%, PMN depletion was 86.9 � 6.1%, MNC recovery was 63.6 � 5.0% and CD14� monocyte recovery was 75.3 � 9.9%. Conclusions. The Cell Saver5™ is relatively inexpensive to purchase and use. The instrument and its disposables are licensed by the United States Food and Drug Administration (FDA) for intra-operative blood salvage, and we have obtained approval for investigational use. Our method with this instrument has proven to be simple and effi cient for clinical-scale Ficoll separations.

Key Words: clinical-scale, closed system, Ficoll, gradient separation

Introduction

There are a number of cellular therapy products that are manufactured from starting materials collected as a heterogeneous mixture of blood cells. These products may require effectively complete depletion of erythro-cytes and enrichment of mononuclear cells (MNC) as part of the manufacturing process. In our own facil-ity we manufacture dendritic cells (DC), which are produced from plastic-adherent MNC in a short-term culture system (1). This method is widely used, and a common requirement for all investigators reporting this method is that the starting material for the cultur-ing of DC is depleted of erythrocytes and enriched for MNC (2–5). The starting cellular material for other cell-based therapeutic applications that have been described also requires nearly complete depletion of erythrocytes and enrichment of MNC, such as may be obtained using Ficoll density-gradient separation.

The published ‘traditional’ methods for Ficoll separation are clearly designed for small-scale appli-cations. With these methods, heterogeneous cell pop-ulations are layered over Ficoll in tubes, centrifuged, and the MNC carefully pipetted from the interface

between the Ficoll and non-density medium (6). While these methods have withstood the test of time, they are inadequate for clinical-scale applications. Even the largest of tubes suitable for such separa-tions are of such limited volume that it might require dozens of tubes to separate a large-volume aphere-sis product. Intra- and interoperator variation in the harvesting of the Ficoll interface leads to inconsistent collected products. Moreover, because the separation is conducted in open vessels, the risk of contamina-tion is continually present.

Recently, a device whereby Ficoll separation can be conducted in an effectively closed system has been introduced into the market place (Sepax™; BioSafe, Eysins, Switzerland) (7). This device and its disposable components are costly, and their dis-tribution is somewhat limited. An earlier instrument (SteriCell™; Terumo, Somerset, NJ, USA) (8) was also marketed for closed-system Ficoll separations. Although originally intended for use in the produc-tion of lymphokine-activated killer (LAK) cells and tumor-infi ltrating lymphocytes (TIL), this instrument was employed in hematopoietic progenitor cell-processing applications. The SteriCell™ instrument

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.

Large-scale Ficoll separation 419

the two instruments, however, is the time required to achieve collection targets (9). Following product collection and transport to the Moffi tt Cancer Cen-ter Cell Therapies Core Facility (Tampa, FL, USA), 4 mL were removed from the product for initial safety and quality testing. Testing included micro-biologic culture assay (BacT/Alert PF; Biomeriuex, Durham, NC, USA), total leukocyte count (Sysmex KX-21, Sysmex Corporation, Kobe, Japan), viability test using the method of trypan blue exclusion, and MNC/CD14� cell enumeration by fl ow cytometry.

Cells were prepared for fl ow cytometry by admix-ing 5 µL each of PerCP-tagged CD45 (Becton-Dickinson, San Jose, CA, USA) and phycoerythrin (PE)-tagged CD14 (Becton-Dickinson) with 4 � 105 cells suspended in 200 µL Dulbecco’s phosphate-buffered saline (DPBS; Mediatech, Manassas, VA, USA) supplemented with 5% fetal bovine serum (FBS; Mediatech). After 15 min incubation, 2 mL FACSLysing solution (Becton-Dickinson) were added to each tube, followed by an additional 10-min incubation. Two milliliters of 5% FBS in DPBS were added to each tube, followed by centrifugation at 800 g for 4 min. The supernatant was decanted from each tube, 4 mL additional DPBS with 5% FBS added, pellets suspended, and the tubes recentrifuged at 800 g. The supernatants were again decanted. The cells were resuspended in 200 µL 5% FBS in DPBS and acquired on a FacsCalibur™ fl ow cytometer (Becton-Dickinson) with open gating and a minimum of 10 000 events. The listmode fi le was analyzed by fi rst isolat-ing leukocytes by identifying medium to high CD45-expressing cells with any forward scatter. MNC were identifi ed on leukocyte-gated CD45 by side-scatter histogram. Low to medium side-scatter cells with high CD45 were identifi ed as MNC. In a fi nal MNC-gated scatter plot, CD14� cells were identifi ed as those with any level of CD14 expression above isotype control and low to medium side-scatter.

Ficoll separation on Haemonetics® Cell Saver5™

The Cell Saver5™ instrument, as set up for Ficoll separation, is shown in Figure 1, with all of the components referenced in the following paragraphs identifi ed. Prior to cellular separation, a ‘W’ tubing adapter obtained from the manufacturers’ ‘sequester-ing kit’ (Haemonetics) was inserted between the out-let line from the bowl (225 mL CellSaver bowl set) and the waste bag inlet. This was to allow the addi-tional connection, using a sterile connecting device (SCD) (SCD IIB; Terumo Medical), of a 600-mL collection bag (Charter Medical, Winston-Salem, NC, USA) to the second line on the ‘W’. The third line on the ‘W’ was not used. Five milliliters of hepa-rin sodium (1000 USP U/mL; APP Pharmaceuticals,

was a modifi cation of the CellSaver Plus™ surgical blood salvage system (Haemonetics Corporation, Braintree, MA, USA). Production of both the Cell-Saver Plus™ and the SteriCell™ was discontinued in the mid-1990s.

We have developed a Ficoll separation method using the successor instrument to the SteriCell™ and Cell Saver Plus platform, namely the Haemon-etics™ Cell Saver5™ surgical blood salvage instru-ment. We based this method on the one employed with the earlier SteriCell™ instrument (8), tak-ing advantage of the bell-shaped bowl design. This instrument and associated disposable compo-nents are widely available internationally, and cost substantially less than the Sepax™ instrument and associated disposables. The Cell Saver5™ has regu-latory approval for surgical blood salvage in every market within which it is distributed. This, in turn, simplifi es application for regulatory approval for the Ficoll separation application as the sterility docu-mentation and manufacturing quality of the con-sumables are already established to the satisfaction of regulatory organizations. All of the applications where we use this instrumentation are approved for investigational use within the scope of our own Inves-tigative New Drug (IND) fi lings with the United States Food and Drug Administration (FDA).

Methods

Patients and samples

The results presented here were derived from the routine processing, under IND and approval of the University of South Florida (FL, USA) Institutional Review Board (IRB), of 37 autologous, non-mobi-lized cytapheresis products collected for production of DC for use in vaccine protocols. The procedure had been validated using purchased buffy coats for elimination of erythrocytes. Data on MNC enrich-ment and CD14� cell recovery were not retained at that time. The data reported here were retained as part of the DC production effort associated with multiple IRB-approved clinical studies. Separate approval of the University of South Florida IRB was obtained for this ‘chart review’ study.

All starting products were collected with two-donor blood volume apheresis. Collection was performed using either a Haemonetics MCS�™ (Haemonetics Corporation) or a Cobe Spectra™ (Caridian Corpo-ration, Lakewood, CO, USA) apheresis instrument using the respective instrument’s MNC collection protocol. In general, products from the MCS� were smaller in volume but contained more eryth-rocytes and granulocytes than those collected using the Spectra. The only signifi cant difference between

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.

420 W. E. Janssen et al.

the line to the saline was again shut and the pump restarted. When all possible product was in the bowl, the pump was again paused. The line from the saline bag was detached using a tube sealer (Sebra, Tuscon, AZ, USA) and the Ficoll source was attached using a SCD. The centrifuge speed was increased to 4800 r.p.m. and the pump speed adjusted to 20 mL/min. As the Ficoll was pumped into the bowl, a band of Ficoll-buoyant cells was observed to form at the Ficoll interface and was pushed to the top (Figure 2). When this band moved to approximately 1 cm from the top of the bowl, the line to the waste bag was closed and the line to the 600-mL collection bag opened. The Ficoll-buoyant cell fraction was pumped into the collection bag, and the line to the collection bag remained open until the fl uid running into it appeared by visual inspection to be clear, indicating that all cells remaining in the bowl were of greater density than the Ficoll medium. The pump and centrifuge were then stopped and the collection bag removed. This pro-cedure, including pump and centrifuge speeds, is an adaptation of the protocol that was hard-wired into the earlier SteriCell™ instrument (8).

Following cell separation, the collection bag was weighed to fi nd the volume of product collected. A 0.5-mL sample was removed from the collection bag for a cell count. The collection bag was then sterile docked to a 1-L bag of Plasma-Lyte-A (Baxter) and fi lled to a volume of 400 mL. The collected product was then centrifuged at 1300 g for 8 min at room temperature. Following centrifugation, the centri-fuge bucket was removed from the centrifuge to the biologic safety cabinet with the product bag remain-ing in it, to avoid risk of disrupting pelleted cells. The product bag was spiked with a 4-inch plasma transfer set (Charter Medical, Winston-Salem, NC, USA) for the removal of the supernatant using a 60-cc syringe. Supernatant was removed so that 207 mL of cells and medium remained. The bag was then removed from the centrifuge bucket and the cell pellet resus-pended by agitation of the bag. Sampling was per-formed for follow-up microbiologic assay, cell counts and cell type enumeration.

Statistical methods

Post-processing counts are dependent on the pre-processing counts. That is, if there are 100 granulo-cytes pre-processing, and the process removes 90% of them, there will be 10 remaining post-processing. If, however, there are 1000 granulocytes pre-processing, then 100 will be expected post-processing. Because of this dependent relationship, the t-test for depen-dent variables was employed, and the null hypothesis tested was that no change in cell numbers would occur between pre- and post-processing.

LLC, Schaumburg, IL, USA) were added to the collection bag to prevent clotting. Haemonetics has recently discontinued production of the sequester-ing kit with the ‘W’ adapter. It has been replaced with a new sequestering kit that has all bags already attached to a ‘W’ tubing confi guration that is simply attached to the outlet from the bowl in the general purpose kit, simplifying the method described here.

A 1-L bag of saline (Baxter, Deerfi eld, IL, USA) and the product were each spiked with non-vented spike adapters (American Fluoroseal, Gaithers-berg, MD, USA), which in turn were connected by SCD to spike port adapters (American Fluoroseal). The saline and product were then each attached to the two yellow input lines. A 500-mL bottle of Ficoll-paque plus (GE Healthcare BioSciences AB, Uppsala, Sweden) was spiked with a non-vented spike adapter (Baxter) and the bottle was vented with an 18-gauge needle to which a luer-lock syringe fi lter had been attached.

The protocol used for the Cell Saver5™ was changed to ‘sequestering’. The pump speed was set to 100 mL/min and the centrifuge speed was set to 3000 r.p.m. Saline was then pumped to fi ll the bowl to the point of running off into the waste bag. The pump was paused, the line to the saline closed and the line to the product opened. The pump was restarted, allowing the entire product to enter the bowl. After all the product had run from the bag, the pump was paused and the saline line re-opened to rinse the product bag. After a thorough rinse was completed,

Figure 1. Cell Saver5™ instrument set up for Ficoll separation of apheresis products. This is the appearance of the instrument after the saline bag has been removed and the Ficoll source attached using the SCD sterile tubing welder. Note the placement in the outlet line of the ‘W’ connector, also attached using the SCD.

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.

Large-scale Ficoll separation 421

RBC in the completed products. This translated to 88.4 � 2.8% of RBC depleted from products that were Ficoll separated using the Cell Saver5™. These data are refl ected in Figure 3. These deple-tions were not as exhaustive as reported for Sepax™ (99.6 � 0.5%) and conventional separation (99.9 � 0.1%) methods (7). However, our data were derived from separation of apheresis products where the starting material was already 90% depleted of RBC relative to whole blood or bone marrow. This was not directly comparable to results obtained from separating whole blood or marrow. Additionally, the use of apheresis products with relatively high white blood cells (WBC) may cause artifi cially high RBC counts because of small WBC being counted in the RBC channel.

MNC and CD14� cell enrichment and recoveries, and polymorphonuclear leukocyte (granulocyte) depletion

The pre- and post-processing fractions of MNC, CD14� cells and polymorphonuclear leukocytes (PMN; granulocytes) were measured by fl ow cytometric analysis with CD45, side-scatter and CD14. An example of the fl ow cytometry is shown in Figure 6. The results of this analysis are refl ected in Figure 5.

Results

Starting material accommodated

All starting products were apheresis collections, col-lected based on the donor’s computed total blood volume, and collected using two different apheresis instruments. Accordingly, there was substantial het-erogeneity among the starting products with respect to volume and cell count. There was also heteroge-neity with respect to the distribution of cell types within the product. These are outlined in Table I, and are also refl ected in Figures 3–5. The average nucle-ated cell viability of starting products, as measured by trypan blue exclusion and manually counted, was 99.75% (range 98–100%).

Erythrocyte depletion

There were (200. � 61.0) � 109 total red blood cells (RBC; erythrocytes) in the starting products, which was reduced to (16.6 � 4.8) � 109 total

Figure 2. Ficoll interface band in the latham bowl in the Cell Saver5™ instrument. (A) Early band lifting up from the packed cell mass that is not buoyant in Ficoll medium. (B) Fully formed band just prior to entering the neck of bowl and being collected.

Figure 3. RBC depletion from the Ficoll procedure with the Cell Saver5™. Samples of pre- and post-processing products were analyzed on an automated blood cell analyzer (Sysmex) and the RBC count recorded. Total RBC in the product was computed as RBC/mL multiplied by product volume. Pre- and post-RBC contents were compared by t-test for dependent variables, P� 0.001.

Starting Products

Median Minimum Maximum

VOLUME (mL) 195.0000 111.0000 490.0000TNCxIO9 12.7893 4.5066 37.9730RBCxIO9 150.4800 28.8600 955.1800Percent MNC 73.4550 34.3800 97.4100Percent CD 14+ 17.1250 2.3300 45.4300Percent PMN 26.0400 0.7700 63.8900

Table I. Starting apheresis products for the Ficoll process on Cell Saver5™: products collected by apheresis were heterogeneous in volume, nucleated cell content, RBC content and relative fractions of MNC, CD14� and PMN.

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.

422 W. E. Janssen et al.

Discussion

The Haemonetics Cell Saver5™ is the next genera-tion of cell-processing instrument after the CellSaver Plus™, which was the platform for the SteriCell™ instrument marketed through the mid-1990s for density-based cell separation, including Ficoll den-sity-separation processing (8). When the SteriCell™ instrument that our facility had been using since late 1989 became non-supportable because of lack of parts availability, the purchase of the newer Cell Saver5™ was identifi ed as the logical replacement choice, albeit with the necessity of developing new protocols to match the newer software that came with this platform.

We have adapted the sequestering program that is included in the Cell Saver5™ instrument to perform Ficoll density separations of MNC from heterogeneous leukocyte populations. All the pump and centrifuge speeds were taken directly from the earlier SteriCell™ instrument (8), so the primary adaptation was in the specifi cs of connection of starting product, saline, Ficoll, waste and fi nal collection containers. Over the 6-year period dur-ing which we have been using this instrument and methodology, we have performed more than 80

Figure 4. TNC (leukocyte) recovery from the Ficoll procedure with the Cell Saver5™. Samples of pre- and post-processing products were analyzed on an automated blood cell analyzer (Sysmex) and the WBC count recorded. TNC in the product was computed as WBC/mL multiplied by product volume. Pre- and post-TNC contents were compared by t-test for dependent variables, P � 0.001.

The average enrichment of MNC was 1.4-fold, with a 95% confi dence interval range of 1.2–1.5-fold. The MNC concentrations as a percentage of total nucleated cells (TNC) were signifi cantly differ-ent, as refl ected by the t-test for dependent samples (P � 0.001). The average recovery of MNC was 63.6%, with a 95% confi dence interval range of 58.6–68.6%. This compared favorably with reported recoveries using Sepax™ (49.5 � 17.3%) and standard Ficoll separation methodology (26.8 � 11.7%) (7).

CD14� cells were enriched 1.5-fold on average (95% confi dence interval range 1.4–1.7 fold). CD14 concentrations as a percentage of TNC were signifi -cantly different, as refl ected by the t-test for depen-dent samples (P � 0.001). The average recovery of CD14� cells was 75.3%, with a 95% confi dence interval range of 65.4–85.2%.

PMN were depleted, on average, by 86.9% (95% confi dence interval range 80.8–90.0%). This was comparable to reported PMN depletion using Sepax™ (92.3 � 3.6%) and standard Ficoll separa-tion methods (94.0 � 4.9%) (7).

The viability of nucleated cells was not remea-sured following separation and prior to freezing. All products processed using this method had viabilities within an acceptable range, namely greater than 70%, with most greater than 80%, following liquid nitro-gen cryostorage and thawing for DC production.

Figure 5. Pre- and post-Ficoll processing fractions of MNC, CD14� cells and PMN. Samples were taken from pre- and post-processing products and stained with PE-labeled CD14 and PerCP-labeled CD45 monoclonal antibodies. Ten-thousand events were acquired by fl ow cytometer, then analyzed for relative fractions of PMN, MNC and CD14. Pre- and post-processing fractions for each cell type were compared using the t-test for dependent variables. MNC, P � 0.001; CD14�, P � 0.001; PMN, P � 0.001.

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.

Large-scale Ficoll separation 423

based on the relative time required to set up the Cell Saver5™ for washing compared with that required to remove the fi nal product bag and place it into a centrifuge.

The full capacity of the Haemonetics bowl is 225 mL. This has translated, in our hands, to an upper packed cell volume limit of about 175 mL. In other words, Ficoll separations are possible for products up to a very large size, including, in addition to the apheresis products that we have reported here, bone marrow, cord blood and whole blood.

The method for performing Ficoll separation as we have described here does require, in addition to the Cell Saver5™ instrument, the additional purchase of an SCD tube welder (Terumo). Even so, based on recent quotes we have obtained, the combined cost of acquisition of both Cell Saver5™ and SCD is less than 2/3 of the purchase price of the Sepax™. More-over, based on the information we have obtained, the consumables used in the operation of the Cell Saver5™ using our methodology are 35% less costly compared with the Sepax™.

The greatest advantage of the use of this instru-ment and methodology, however, is the wide international availability of the instrument and its components. According to their web site http://www.haemonetics.com, accessed on 6 November 2009, Haemo netics Corporation has offi ces throughout North America, Europe and the eastern pacifi c rim countries. Their instruments are also available in Africa and South and Central America.

Ficoll density separations. We selected a subset of 37 such processes for the analysis reported here based on having the complete dataset analyzed in our database.

The goal of Ficoll density separation is to reduce, as far as possible, PMN and RBC while losing as few MNC as possible. Our Ficoll-processing proce-dure on the Cell Saver5™ meets these criteria, with a PMN reduction of 86.9 � 3.1%, RBC reduction of 88.4 � 2.8% and MNC recovery of 63.6 � 5.0%. These values compare quite favorably with those reported for the Sepax™ system, where the RBC and PMN depletion reported were comparable, but the MNC recovery was only 49.5 � 12.3% before post-separation washing (7).

Possible concerns relative to our specifi c applica-tion of the cells separated using this method, namely production of DC ex vivo, are that cell viability might be adversely affected and monocytes in the collected cell product might be activated. That we have suc-cessfully generated more than 150 DC vaccines from the 37 products described in this report suggests that cell viability is adequate and that monocyte activa-tion, if any, is minimal.

In this report, we have described post-Ficoll washing of the bagged product in a conventional cen-trifuge. The Cell Saver5™ is, in fact, capable of doing cell washing with comparable cell recoveries. The choice not to use the instrument for the washing step was predicated by the preference of the technologists performing the processing. This preference was

Figure 6. Representative fl ow cytometry showing depletion of PMN with the Ficoll procedure. Samples were taken from pre- and post-processing products and stained with PE-labeled CD14 and PerCP-labeled CD45 monoclonal antibodies. Ten-thousand events were acquired by fl ow cytometer, then analyzed for relative fractions of PMN, MNC and CD14 (colored blue in these graphics).

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.

424 W. E. Janssen et al.

chemotherapy in patients with extensive stage small cell lung cancer. Clin Cancer Res. 2006;12:878–87.Dohnal AM, Graffi S, Witt V, Eichstill C, Wagner D, 2. Ul-Haq S, et al. Comparative evaluation of techniques for the manufacturing of dendritic cell-based cancer vaccines. J Cell Mol Med. 2009;13:125–35.Elkord E, Williams PE, Kynaston H, Rowbottom AW. Human 3. monocyte isolation methods infl uence cytokine production from in vitro generated dendritic cells. Immunology. 2005;114:204–12.Felzmann T, Witt V, Wimmer D, Ressmann G, Wagner D, 4. Paul P, et al. Monocyte enrichment from leukapharesis products for the generation of DCs by plastic adherence, or by positive or negative selection. Cytotherapy. 2003;5:391–8.Meyer-Wentrup F, Burdach S. Effi cacy of dendritic cell gen-5. eration for clinical use: recovery and purity of monocytes and mature dendritic cells after immunomagnetic sorting or adherence selection of CD14� starting populations. J Hema-tother Stem Cell Res. 2003;12:289–99.Noble PB, Cutts JH, Carroll KK. Ficoll fl otation for the 6. separation of blood leukocyte types. Blood. 1968;31:66–73.Aktas M, Radke TF, Strauer BE, Wernet P, Kogler G. Sepa-7. ration of adult bone marrow mononuclear cells using the automated closed separation system Sepax. Cytotherapy. 2008;10:203–11.Janssen WE, Lee C, Smilee R, Carter R. Use of the Terumo 8. SteriCell for the processing of bone marrow and peripheral blood stem cells. J Hematother. 1992;1:349–59.Abdelkefi A, Maamar M, Torjman L, Ladeb S, Lakhal A, 9. Othman TB, et al. Prospective randomised comparison of the COBE Spectra version 6 and Haemonetics MCS� cell separators for hematopoietic progenitor cell leucapheresis in patients with multiple myeloma. J Clin Apher. 2006;21:111–5.

While the Cell Saver5™ has not received regu-latory approval for this application, we have had no diffi culty with obtaining such approval on an investi-gational basis. We have described this application in all of our relevant IND fi lings and in our Type V master fi le. The Cell Saver5™ is approved in all of its mar-kets for intra-operative blood salvage. Thus it may be assumed that all of the major regulatory agencies have familiarity with this instrument, easing the pathway for obtaining approval for this additional application.

We submit, therefore, that the Cell Saver5™ instrument, coupled with the methodology described in this report, provides a cost-effective approach to clinical-scale Ficoll separation of heterogeneous blood and marrow products. This approach provides simplicity, speed, effective closure to environmental contamination and ready availability for any facility location in the world.

Confl ict of interest: The authors certify that they have no affi liation with or fi nancial involvement in any organization or entity with a direct fi nancial interest in the subject matter or materials discussed in this manuscript.

References

Antonia SJ, Mirza N, Fricke I, Chiappori A, Thompson P, 1. Williams N, et al. Combination of p53 cancer vaccine with

Cyt

othe

rapy

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

CD

L-U

C S

anta

Bar

bara

on

06/1

5/13

For

pers

onal

use

onl

y.