• Open Access

Collection of Peripheral Blood CD34+ Progenitor Cells from Healthy Dogs and Dogs Diagnosed with Lymphoproliferative Diseases Using a Baxter-Fenwal CS-3000 Plus Blood Cell Separator


  • S.E. Suter

    Corresponding author
    1. Center for Comparative Medicine and Translational Research, North Carolina State University, Raleigh, NC
    2. Lineberger Comprehensive Cancer Center, Chapel Hill, NC
    • Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC
    Search for more papers by this author

  • This work was performed at the North Carolina State College of Veterinary Medicine Veterinary Teaching Hospital.

Corresponding author: S.E. Suter, Department of Clinical Sciences College of Veterinary Medicine, North Carolina State University, 4700 Hillsborough Street, CVM Research #308, Raleigh, NC 27606; e-mail:steven_suter@ncsu.edu



Canine peripheral blood mononuclear cell (PBMC) apheresis using a Baxter-Fenwal CS-3000 Plus automated blood cell separator has not been reported.


To determine the feasibility and safety of using a CS-3000 Plus blood cell separator with a small volume separation container holder (SVSCH) and small volume collection chamber (SVCC) to harvest canine PBMCs from dogs weighing <50 kg.


Eight healthy mongrel dogs and 11 client-owned dogs in clinical remission for lymphoproliferative diseases (LPD).


In this prospective study, aphereses were performed using a Baxter-Fenwal CS-3000 Plus blood cell separator, with or without recombinant human granulocyte colony-stimulating factor (rhG-CSF) treatment.


Aphereses from 6 healthy dogs given rhG-CSF yielded an average of 1.1 × 107 ± 8.2 × 106CD34+ cells/kg. Aphereses from LPD dogs given rhG-CSF yielded an average of 5.4 × 106 ± 3.25 × 106CD34+ cells/kg (= .17). Higher hematocrit in both groups of dogs receiving rhG-CSF correlated with an increased number of CD34+ cells/kg harvested (healthy, = .04; LPD, = .05). Apheresis was well tolerated by all dogs.

Conclusions and Clinical Importance

Canine PBMC apheresis using the Baxter-Fenwal CS-3000 Plus cell separator with an SVSCH and SVCC is a feasible and safe option for harvesting an adequate number of CD34+ peripheral blood progenitor cells from dogs weighing ≥17 kg for hematopoietic cell transplantation.


acid-citrate-dextrose formula


acute lymphocytic leukemia


constant rate infusion




hematopoietic cell transplantation


high-dose cytoxan


ionized calcium


lymphoproliferative disease




peripheral blood mononuclear cell


recombinant canine granulocyte colony-stimulating factor


recombinant human granulocyte colony-stimulating factor


small volume collection chamber


small volume separation container holder


total blood volume


white blood cells

Lymphoma (LSA) in dogs is the most common hematopoietic tumor (83%),[1, 2] with an overall cure rate of <5%.[3] Dogs have been a valuable preclinical model for hematopoietic cell transplantation (HCT) for 5 decades,[4] and therefore, there is a large body of research literature supporting the notion that HCT can cure some malignant hematologic conditions in dogs.[5, 6]

Mononuclear hematopoietic progenitor cells are present at low concentrations in the peripheral blood of a variety of species.[7-9] Development of techniques to both increase peripheral blood mononuclear cell (PBMC) counts using cloned hematopoietic cytokines[10, 11] and collect large numbers of these cells via centrifugation using continuous-flow blood separators increased yields.[12-14] In dogs, both autologous[15, 16] and allogeneic[17] PBMC grafts were able to reconstitute normal hematopoiesis after lethal bone marrow ablation. Freshly isolated PBMC grafts were used to treat dogs in remission for LSA after high-dose total body irradiation, with approximately 25% becoming disease-free long-term survivors.[18]

The 2 most frequently utilized commercially available blood cell separators are the CaridianBCT (Formerly COBE Spectra) and the Baxter-Fenwal CS-3000 Plus. Both machines depend on buoyant separation of blood components via centrifugation, although they differ in physical setup. The CaridianBCT packs the mononuclear cell layer and collects by pushing them into a small collection channel. The CS-3000 Plus, conversely, aspirates plasma and mononuclear cells from a primary collection chamber and separates these via a second centrifugal procedure in a secondary chamber. Since one of the limiting factors for pediatric PBMC collection is the patient's whole blood volume,[19, 20] a potential advantage of the CS-3000 Plus is that it can be altered for the collection of PBMC from patients weighing ≤50 kg. The use of a small volume separation container holder (SVSCH) and a small volume collection chamber (SVCC) reduces the extracorporeal volume of the machine from approximately 453 to 240 mL, conserves patient's platelets, reduces the risk of hypovolemia-induced vasovagal reactions, and negates the need for machine blood priming in small patients.

Although PBMC apheresis[14, 21] and plateletpheresis[22] have been reported in dogs using a CaridianBCT machine, the use of a Baxter-Fenwal CS-3000 Plus cell separator for apheresis in dogs has not, to the author's knowledge, been reported. The purpose of this study was to determine if the CS-3000 Plus SVSCH and SVCC can be safely utilized to collect an adequate number of peripheral blood CD34+ mononuclear progenitor cells from dogs weighing <50 kg for HCT after bone marrow ablation.

Materials and Methods

Study Design

The clinical and hematologic effects of rhG-CSF administration in healthy mongrel dogs and dogs diagnosed with lymphoproliferative diseases (LPD) were evaluated. Peripheral blood CD34+ progenitor cells were collected from non-rhG-CSF treated healthy dogs, treated healthy dogs, and treated LPD dogs using a Baxter-Fenwal CS-3000 Plus cell separator. Apheresis product analysis was used to enumerate cell yields and determine the collection efficiency of the cell separator.


Eight adult, healthy mongrel dogs (3 spayed females, 5 neutered males) underwent 1 apheresis. The dogs weighed from 17 to 34 kg (mean = 26.4 kg) and were 24 to 60 months (mean = 42.3 months) old. The study protocol was approved by the Institutional Animal Care and Use Committee of North Carolina State University and the study was conducted according to the principles outlined in the Guide for Laboratory Animal Facilities and Care prepared by the National Academy of Sciences, National Research Council.

Eleven LPD dogs (6 spayed females, 5 neutered males) diagnosed with LSA (10) or acute lymphoblastic leukemia (ALL, 1) underwent apheresis. Dogs weighed from 19.6 to 48 kg (mean = 31 kg) and were 48 to 132 months (mean = 72.6 months) old. All dogs were in clinical remission, although the chemotherapy protocols and the length of time they were receiving chemotherapy varied (range 3–15 months, mean 8 months). Seven dogs had high-grade B-cell LSA, 2 dogs had high-grade T-cell LSA, 1 dog had an unknown phenotype high-grade LSA, and 1 dog had B-cell ALL. Dogs presented to the North Carolina State College of Veterinary Medicine Veterinary Teaching Hospital (NCSU-VTH) Bone Marrow Transplant Unit for PBMC apheresis in preparation for an autologous HCT.

Donor Preparation for PBMC Apheresis

Aphereses were performed after treatment with rhG-CSF1 (5 μg/kg SQ q12h) for 6 days in 6/8 healthy dogs with a double dose given 2 hours before apheresis commenced.[14] During treatment, CBCs were performed daily to monitor white blood cell elevation.

Ten of 11 privately owned dogs received high-dose IV cyclophosphamide2 (HDC) at 300 (1), 450 (1), or 500 (8) mg/m2 at their referring veterinarians. For the dogs receiving 450 and 500 mg/m2 cyclophosphamide, 2 doses of IV mesna3 were also given.[23] In addition, 2 mg/kg IV furosemide4 was given 30 minutes before and 6 hours after HDC. Dogs were also sent home on furosemide (PO) for 48 hours and prophylactic antibiotics for 2 weeks. HDC was followed with a 1-week rest before starting rhG-CSF mobilization 6 days prior to apheresis.[14]

Venous Access and Anesthesia

After propofol5 sedation, dogs were intubated and maintained on Sevoflurane.6 A 12-gauge, 20-cm central venous dual-lumen hemodialysis catheter7 was aseptically placed percutaneously in a jugular vein for venous access and return.[13] Three additional 18-gauge 2.9-cm catheters were placed in peripheral veins for (1) dexmedetomidine HCl8 constant rate infusion (CRI), (2) 10% calcium gluconate CRI, and (3) serum ionized calcium (iCa) monitoring. Sevoflurane was discontinued and a light plane of anesthesia was maintained using a dexmedetomidine HCl CRI (0.2–0.5 μg/kg/min) after the first approximately 30 minutes of aphereses.

After aphereses, the healthy dogs were reversed with Antipamezole HCl9 and returned to the research colony after removal of the hemodialysis catheter. For the client-owned dogs, the hemodialysis catheter was removed, and a 16-gauge, 20-cm central venous triple-lumen catheter10 was aseptically placed percutaneously in the contralateral jugular vein before reversal.

PBMC Apheresis

Aphereses were performed using a Baxter-Fenwal CS-3000 Plus with Access Management System continuous blood separator11 using the SVSCH and SVCC system and a disposable double needle closed system apheresis tubing kit12 (Baxter-Fenwal CS-3000 Mononuclear Cell Collection, CLE005 rev.2-1/1/01). The tubing kit and collection chambers were primed with acid-citrate-dextrose formula A13 anticoagulant (ACD-A), and 0.9% saline. Machine computer adjustments were made by modifying “locations”, which are sets of instructions that are required for microprocessor control. The machine uses 89 run and 59 reinfusion “locations”, which alter machine parameters such as centrifuge speed and pump flow rates to optimize collection efficiencies. The cell separator inlet pump speed (25 mL/min), centrifuge speed (1,600 rpm), and interface/offset detector setting (50, Location 71, optical density measurement of the cell suspension) were not changed throughout the procedure. ACD-A infusion was overseen by the Access Management System (WBC : ACD-A ratio of 10 : 1). This ratio was reduced to 8 : 1 after the 5th LPD dog apheresis, due to product clumping after overnight refrigeration. For all aphereses, the procedure length was based on body weight.[24, 25] Aphereses run lengths were based on a canine total blood volume (TBV) of 90 mL/kg.

Throughout aphereses, dogs were monitored using continuous electrocardiogram, indirect blood pressure measurement, and body temperature measurement. Serial blood iCa, sodium (Na), potassium (K), pH, oxygen partial pressure (PO2), oxygen saturation (SO2), and glucose concentrations were evaluated every half hour using a portable clinical analyzer.14 Hot water and air heating pads were used to maintain body temperature in the physiological range, while a CRI of 10% Ca gluconate solution at 20 mL/h was used to control citrate-induced hypocalcemia. If serum iCa levels dropped by >0.2 mmol/L below the pre-apheresis iCa measurement, the infusion rate was increased to 40–60 mL/h.

Apheresis Product Analysis

Two hours after apheresis initiation and upon completion, 1 mL of harvested product was used for CBC analysis using an automated hematology analyzer.15 Another 1 mL was assayed for CD34+ progenitor cell content using a commercially available mouse anti-canine CD34 monoclonal antibody[26] and flow cytometry16 using CELLQuest17 software.

Mononuclear Cell Product Yield and Collection Efficiency Calculations

Equations used were:

display math
display math
display math

where total blood volume processed = blood volume processed (mL) – anticoagulant volume, absolute donor precount = donor total WBC/μL × predifferential number (%), and absolute donor postcount = donor total WBC/μL × postdifferential (%).

Statistical Analysis

All data were normally distributed as determined by a D'Agostino and Pearson omnibus K2 normality test and Kolmogorov–Smirnov normality test in the LPD dogs (n = 11) and normal mobilized dogs (n = 6), respectively. A Welch's unpaired t test was used to calculate significant differences between the dog populations. The relationship between pre-apheresis parameters and the collected CD34+ cells was estimated by linear regression and correlation analysis. A P value of ≤ .05 was considered statistically significant. All statistics were calculated using GraphPad.18


Effects of rhG-CSF Treatment

No signs of adverse effects reported to occur in human patients, including bone pain, allergic reactions, nausea, or vomiting, were seen in any dogs. In healthy dogs, WBC (mean 32,190 ± 6,068 cells/μL) and mononuclear cell counts (mean 1,870 ± 452 cells/μL) were much higher than those in our hospital's normal mean cell counts (WBC reference interval 4,390–11,610 cells/μL, mean 8,000 cells/μL; monocyte reference interval 75–850 cells/μL, mean 462 cells/μL; Pre Aph, Table 1) after rhG-CSF treatment. There was not a statistically significant difference when comparing the mean WBC (41,693 ± 8,730 cells/μL, = .02) and mononuclear cell counts (2,648 ± 1,075 cells/μL, = .056) of dogs diagnosed with LPD after 6 days of rhG-CSF to the treated healthy dogs (Pre Aph, Table 2). When compared to treated healthy dogs' mean WBC and mononuclear cell counts measured approximately 8 hours after the double dose of rhG-CSF (Post Aph, Table 2), higher mean WBC values (51,014 ± 15,426 cells/μL, = .04) were documented in the dogs' diagnosed with LPD, whereas mean mononuclear cell counts were not significantly different (3,345 ± 2,181 cells/μL, = .07; Post Aph, Table 2). Based on a WBC count in all rhG-CSF treated dogs of very near to or >30,000/μL, all dogs underwent apheresis.

Table 1. Healthy dog apheresis characteristics using the Baxter-Fenwal CS-3000 Plus, with or without rhG-CSF treatment for 6 days.
 Weight (kg)Coll Length (m)Blood Volume (mL)TBVACD-A (mL)Citrate (g)Pre rhG-CSF WBC/μLPre rhG-CSF Mono/μLPre Aph HctPre Aph WBC/μLPre Aph Mono/μLPost Aph HctPost Aph WBC/μLPost Aph Mono/μL
  1. TBV, total blood volume; ACD-A, acid-citrate-dextrose formula; Pre-rhG-CSF, pre-rhG-CSF mobilization; Pre Aph, pre-apheresis; Post Aph, postapheresis; Mono, mononuclear cells; Hct, hematocrit; WBC, white blood cells.

Nonmobilized Healthy (n = 2)
Mobilized healthy (n = 6)
Mean±SD26.4 ± 6.1273 ± 356,175 ± 8382.63 ± 0.30563 ± 8212.0 ± 1.77,922 ± 1,921437 ± 37047.2 ± 7.632,190 ± 6,0681,870 ± 45343.4 ± 6.037,405 ± 9,1081,977 ± 354
All healthy dogs (n = 8)
Mean±SD27.0 ± 6.0265 ± 356,138 ± 8112.60 ± 0.40542 ± 8411.5 ± 1.87,922 ± 1,921437 ± 37046.7 ± 6.625,358 ± 13,6731,491 ± 80242.8 ± 5.328,930 ± 17,4841,520 ± 879
Table 2. LPD dog apheresis characteristics using the Baxter-Fenwal CS-3000 Plus, with rhG-CSF treatment for 6 days.
LPD Dogs (n = 11)aWeight (kg)Coll Length (m)Blood Volume (mL)TBVACD-A (mL)Citrate (g)Pre rhG-CSF WBC/μLPre rhG-CSF Mono/μLPre Aph HctPre Aph WBC/μLPre AphMono/μLPost Aph HctPost Aph WBC/μLPost Mono/μL
  1. LPD, lymphoproliferative diseases; TBV, total blood volume; ACD-A, acid-citrate-dextrose formula; Pre rhG-CSF, pre-rhG-CSF mobilization; Pre Aph, pre-apheresis; Post Aph, post-apheresis; Mono, mononuclear cells; Hct, hematocrit; WBC, white blood cells.

  2. a

    One dog underwent apheresis twice.

Mean ± SD31 ± 7.5364 ± 718,630 ± 1,6833.2 ± 1.11,021 ± 18622.4 ± 4.12,421 ± 2,650250 ± 33437 ± 6.041,693 ± 8,7302,648 ± 1,07536 ± 6.451,014 ± 15,4263,345 ± 2,181
Average difference (95% CI, P value) for rhG-CSF treated(−11.843 to 2.788, = .202)(−148.25 to −36.63, = .003)(−3,760.34 to −1,220.57, = .0009)(−1.3293 to 0.1571, = .112)(−598.67 to −318.39, = .0001)(−13.535 to −7.331, = .0001)(2,967.33 to 8,282.84, = .0007)(36.54 to 815.46, = .035)(−18.689 to −2.051, = .021)(−17,311.66 to −1,693.79, = .02(−1,578.50 to 22.95, = .0560.427 to 14.364, = .04(−26,381.57 to −837.52, = .04(−2,869.24 to 132.14, = .07
normal versus rhG-CSF treated LPD       .0282 (rank sum).0347 (rank sum)     

Apheresis Procedure

All dogs completed the apheresis (Table 1). The mean collection length for all healthy dogs was 265 ± 35 min, while the mean TBV processed for all dogs was 6,138 ± 811 mL (or 2.6 ± 0.40 × TBV). Characteristics of the aphereses using the mobilized LPD dogs are summarized in Table 2. One 11-year-old dog underwent apheresis once on 2 successive days since an inadequate amount of CD34+ cells were harvested on day 1. The mean collection length and TBV processed for these dogs were 364 ± 71 minutes and 8,630 ± 1,683 mL (or 3.2 ± 1.1 TBV, respectively).

Effects of Apheresis

No signs of hypotension were seen in any dogs. All dogs exhibited varying degrees of hypothermia that necessitated temperature management using either fluid warming or forced-air warming blankets. The mean starting rectal temperature for all dogs was 100.5°F, while the mean lowest rectal temperature recorded was 98.8°F (range 98.3–101.6°F). In all dogs, when appropriate measures were taken, rectal temperatures remained at or near starting values.

In all dogs, the serum iCa levels were kept within 0.2 mmol/L of the pre-apheresis measurement throughout the procedure by adjusting the 10% Ca gluconate solution infusion rate. Harvest product clumping was seen with LPD dog 5; therefore, the WBC/ACD-A ratio was decreased from 10 : 1 to 8 : 1 in the remaining 6 dogs. Harvest product clumping was not observed thereafter. No statistically significant difference in the amount of citrate administered between the dogs harvested at a WBC/ACD-A ratio of 10 : 1 and 8 : 1 (data not shown) was seen. No signs of hypocalcemia, such as liplicking, agitation, tremors, or hypertension were noted in any dogs.

The apheresis was also associated with changes in blood pH, bicarbonate, pCO2, chloride, lactate, sodium, potassium, and glucose, although all these values typically remained within reference range and no changes were clinically significant.

Apheresis Cell Yields

The final collection volume of the small volume CS-3000 Plus cell separator collection system was approximately 57 mL. Table 3 calculations are based on a final collection volume of 50 mL, since approximately 5–7 mL of product was used for harvest evaluation.

Table 3. Apheresis product counts in nontreated and rhG-CSF treated healthy dogs and rhG-CSF treated LPD dogs.
 WBCs (Cells/mL)Plts (Cells/mL)Grans (%)Hct (%)Mono (Cells/mL)CD34+%CD34+ Cells/kgMCCE (%)
  1. LPD, lymphoproliferative diseases; WBC, white blood cells; Plts, platelets; Grans, granuloctyes; Hct, hematocrit; Mono, mononuclear cells; MCCE, mononuclear cell collection efficiency.

  2. a

    One dog underwent apheresis twice.

Healthy (n = 2)
 45,8795,18712141,3450.403 × 10511.2
 56,3408,52033246,8790.355.3 × 10519
rhG-CSF treated healthy (n = 6)
 Range99,020–297,4403,726–8,5201–5.024.5–3892,089–282,5681–4.34 × 106–2.5 × 10727.5–77
 Mean±SD214,858 ± 78,7415,763 ± 1,6242.3 ± 1.530.4 ± 6195,384 ± 72,5162.7 ± 1.41.1 × 107 ± 8.2 × 10644 ± 17.8
rhG-CSF treated LPD (n = 11)
 Rangea18,855–280,1602,899–11,9420–7.019–51.318,690–252,1441.67–4.831.9 × 106–13.7 × 1064–63.6
 Mean±SD140,748 ± 89,3965,806 ± 3,3465.4 ± 7.035 ± 10.0135,869 ± 68,5352.40 ± 1.105.4 × 106 ± 3.2 × 10625 ± 19.3
Average difference (95% CI, P value) for mobilized normal versus mobilized LPD(−25,126.42 to 153,274.90, P = .14)(−2,632.85 to 2,546.54, = .97)(−7.92 to 1.86, = .20)(−14.237 to 5.055, = .33)(−22,155.98 to 141,187.31, = .13)(−1.2392 to 1.7877, = .69)(−3,374,773.98 to 14,594,167.92, = .17)(−39.3826 to 1.6742, = .07)

In all cases the mononuclear cell population comprised >80% of the total harvested WBC. In rhG-CSF treated healthy dogs, the apheresis product had a much higher number of WBC and monocytes when compared to untreated healthy dogs (Table 3). Importantly, rhG-CSF treatment significantly increased the CD34+ percentage of the product, which facilitated harvesting a larger amount of CD34+ cells/kg. rhG-CSF treatment resulted in 6/6 harvests having >2 × 106 CD34+ cells/kg (mean 1.1 × 107 ± 8.2 × 106/kg), while this dose was not achieved in the 2 untreated healthy dogs.

The apheresis yields in rhG-CSF treated LPD dogs were very similar to treated healthy dogs. The mean WBCs/μL in LPD and healthy dogs were 140,748 ± 89,396 and 214,858 ± 78,741, respectively (= .14), while the mean monocytes/μL in LPD and healthy dogs were 135,869 ± 68,535 and 195,384 ± 72,516, respectively (= .13). The mean CD34+ percentage in LPD and healthy dogs was also similar (= .69). rhG-CSF treatment in LPD dogs resulted in 10/11 harvests having >2 × 106 CD34+ cells/kg (mean 5.4 × 106 ± 3.2 × 106/kg).

The cell separator MCCE was calculated for all aphereses. There was no statistically significant difference (= .23) when comparing the mean MCCE in rhG-CSF treated LPD dogs (25 ± 19.3%) to the mean MCCE in all healthy dogs (36.8 ± 20.3). The MCCE range in LPD dogs was broad (4–63.6%).

Factors Affecting Harvest Yield

In both rhG-CSF treated healthy and LPD dogs, no correlation was found between the harvested CD34+ cells dose/kg and the donor age (= .64 and .45, respectively), weight (= .58 and .39, respectively), or sex (= .51 and .52, respectively). There was also no correlation between pre-apheresis WBC or monocyte counts and the number of CD34+ cells/kg harvested or MCCE in both groups of dogs. However, pre-apheresis hematocrit (Hct) affected the number of CD34+ cells/kg harvested in both groups, although this effect was less pronounced in LPD dogs (Fig 1). Pre-apheresis Hct did not affect the MCCE in both mobilized healthy (R2 = 0.06, = .64) and LPD (R2 = 0.26, = .11) dogs.

Figure 1.

Correlation between pre-apheresis hematocrit (Hct) and the number of CD34+/kg harvested. (A) rhG-CSF treated healthy dogs (n = 6; R2 = 0.70, = .04), (B) rhG-CSF treated lymphoproliferative diseases (LPD) dogs (n = 11; R2 = 0.36, = .05). Linear regression analysis showed a correlation between harvested CD34+ cells/kg and the pre-apheresis Hct in both cohorts of dogs.

Hematologic Changes before and after Apheresis

In rhG-CSF treated healthy dogs, there was no statistically significant difference when comparing the total WBC count (= .28) and Hct (= .37) before and after apheresis, although the peripheral blood platelet count was decreased after apheresis (= .015; Table 4). No clinical signs relating to the postapheresis thrombocytopenia were observed. There was not a statistically significant difference in the total WBC count (= .10), Hct (= .76), or platelet count (= .09) before and after apheresis in rhG-CSF treated LPD dogs.

Table 4. Comparison of blood cell counts of rhG-CSF treated healthy and LPD dogs before and after apheresis.
 Pre AphPost AphP Value
  1. Pre Aph, pre-apheresis; Post Aph, postapheresis; LPD, lymphoproliferative diseases; WBC, white blood cells; Hct, hematocrit; Plts, platelets.

rhG-CSF treated healthy (n = 6)
WBC (cells/μl)32,190 ± 6,06837,405 ± 9,108.28
Hct47.2 ± 7.643.4 ± 6.0.37
Plts/μl217,500 ± 54,010137,000 ± 34,450.015
rhG-CSF treated LPD (n = 11)
WBC (cells/μl)41,693 ± 8,73051,014 ± 15,426.10
Hct36.9 ± 6.036 ± 6.4.76
Plts/μl197,030 ± 28,000116,000 ± 24,035.09


This report documents the use of a Fenwal CS-3000 Plus SVSCH and SVCC blood cell separator for PBMC apheresis in dogs. We show the CS-3000 Plus small volume collection system can be safely utilized to perform PBMC apheresis in dogs weighing ≥17 kg.

G-CSF is a hematopoietic cytokine produced primarily by bone marrow stromal cells that acts, at higher doses, by stimulating the differentiation of several hematopoietic progenitors, including CD34+ cells. As reported previously[4, 14] 6 days of rcG-CSF treatment increased total white blood cell, neutrophil, and monocyte counts in healthy and LPD dogs with no apparent clinical side effects. The rhG-CSF treatment enabled us to harvest >2 × 106 CD34+ cells/kg with 1 apheresis in 10/11 mobilized LPD dogs, which is the target dose in humans needed to ensure appropriate hematologic reconstitution in most transplant patients.[25] In addition, all dogs also received a double dose of rhG-CSF approximately 2 hours before apheresis to maximize CD34+ cell yield.[27]

Although rhG-CSF is able to adequately stimulate healthy dogs’ bone marrow,[14] its use in the setting of LPD dogs who had received many months of systemic chemotherapy has not been documented. In a previous report, 28 dogs were given 7 days of rcG-CSF after 8 weeks of a modified VELCAP-S protocol and bone marrow was collected via humeral head aspiration.[23] One dog with T-cell lymphoma underwent rcG-CSF treatment prior to allogeneic HCT, although the dog also received rhG-CSF during a 1st relapse and then did not receive any chemotherapy for 8 months before relapsing a 2nd time.[21] The LPD dogs here all received chemotherapy for an extended period and had obvious decreases of both total WBC and monocyte counts before rhG-CSF administration. In spite of this, the LPD dogs did not have statistically significant differences in their total WBC and monocyte counts after 6 days of rhG-CSF treatment when compared to healthy dogs. Most importantly for autologous HCT, we also found no statistically significant differences in the CD34% of the apheresis product or the number of CD34+ cells/kg harvested between treated healthy and LPD dogs, although LPD case 1 did require 2 aphereses to harvest enough CD34+ cells/kg for HCT. In combination, these results suggest that long-term systemic chemotherapy does not affect either rhG-CSF treatment of LPD dogs or decrease the ability of the bone marrow to produce adequate numbers of peripheral blood CD34+ cells for collection and HCT.

MCCE is used to calculate the number of mononuclear cells harvested compared to the number of mononuclear cells entering the cell separator. No statistically significant difference in MCCE was seen between rhG-CSF treated healthy and LPD dogs. Although the LPD dogs MCCE range was broad (4–63.6%), this is also seen in human patients.[28-30] Factors affecting MCCE in people include pre-apheresis WBC counts,[29, 31] cell separator machine settings, differences in donor groups, donor Hct, and patient status at collection,[31] although in one study only 29% of the variability in MCCE could be attributed to patient characteristics such as WBC count, Hct, and albumin concentration.[32] In this study, we found no statistically significant factors that affected both rhG-CSF treated healthy and LPD dogs MCCE. We also did not find, as described in a previous report documenting canine aphereses using a CaridianBCT cell separator,[14] a correlation between higher pre-apheresis WBC counts and higher apheresis WBC yields. Interestingly, in this study, increased donor Hct was correlated with increased numbers of CD34+ cells/kg harvested in both groups, which is in contrast to human data.[32] The reason for this is not apparent, although canine PBMCs appear to have a different density than their human couterparts.[33]

A challenge of HCT transplantation is in predicting the timing in and optimizing the efficiency of peripheral blood CD34+ harvests. Although expensive and time consuming, serial quantification of circulating peripheral blood CD34+ cells during mobilization is a reliable guide to predicting harvest yields in human medicine.[34, 35] Other data suggest that absolute monocyte counts are a useful indicator to predict the peak of circulating CD34+ progenitor cells,[36, 37] in addition to performing point-of-care CD34+ enumeration[38] approximately 2 hours into the procedure. We chose to rely on peripheral blood monocyte counts and point-of-care CD34+ enumeration to help reliably predict adequate CD34+ progenitor harvests and suggest that this strategy will most often reliably predict the optimal time for harvesting PBMN cells from dogs with LPD. Cell separator machine adjustments can also be employed to increase CD34+ harvests. These adjustments include increasing the interface/offset detector setting to 120–150 to maximize mononuclear cell collection, and decreasing the centrifuge speed to 1,400 rpm to reduce the number of platelets collected.[29-32, 39] Future studies in dogs ≤50 kg utilizing different cell separator settings may lead to increased collection efficiencies.

In summary, we present data showing that dogs weighing ≥17 kg can be treated with rhG-CSF and undergo PBMC apheresis using a Baxter-Fenwal CS-3000 Plus blood cell separator. Importantly, all dogs tolerated the procedure well, with the target dose of 2 × 106 CD34+ cells/kg achieved in 16/17 dogs in a single apheresis. With the introduction of Baxter-Fenwals next generation cell separator (AMICUS™) in the late 1990s,[40] used/refurbished CS-3000 Plus blood cell separators are becoming available at greatly reduced prices. Our data show that aphereses containing >80% mononuclear cells can be safely harvested from client-owned dogs with virtually no modifications to the cell separator itself or to the machine computer software. The aphereses also contain an adequate number of CD34+ progenitor cells to ensure hematologic recovery after bone marrow ablation in the setting of autologous HCT.


The author thanks Dr Nicholas Bandarenko (Duke University, Raleigh, NC) and Dr Jeffrey Winters (Mayo Clinic, Rochester, MN) for assistance in obtaining the blood cell separator machines. The author also thanks Donna Hardin and the Central Procedures Laboratory (NC State University) staff for assistance with the apheresis procedures, Linda English (Clinical Immunology, NC State University) for CD34+ cell enumeration, and Mr Al Schmidt (Fenwal, Inc) for CS-3000 Plus training.

This work was not supported by any grant.


  1. 1

    Neupogen, filgrastim, Amgen, Thousand Oaks, CA

  2. 2

    Cytoxan, cyclophosphamide for injection, USP, Baxter Healthcare Corporation, Deerfield, IL

  3. 3

    Mesnex, Sagent Pharmaceuticals, Schaumburg, IL

  4. 4

    Lasix, Sanofi-Aventis, Bridgewater, NJ

  5. 5

    Propofol Injectable Emulsion, AAP Pharmaceuticals, Schaumbur, IL

  6. 6

    Sevoflurane, USP, Piramal Critical Care Inc, Bethlehem, PA

  7. 7

    Arrow International, Reading, PA

  8. 8

    Dormitor, Pfizer, NY

  9. 9

    Antisedan, Pfizer

  10. 10

    Arrow International

  11. 11

    Fenwal-Baxter Healthcare Corporation, Deerfield, IL

  12. 12

    Access Open System Apheresis Kit, #4R2224, Baxter-Fenwal Healthcare Corporation

  13. 13

    ACD-A, Baxter-Fenwal Healthcare Corporation

  14. 14

    iSTAT, Abbott Point of Care, Inc., Princeton, NJ

  15. 15

    Advia 2120, Siemens, Deerfield, IL

  16. 16

    FASCan, Becton-Dickinson, Franklin Lakes, NJ

  17. 17


  18. 18

    GraphPad Software, San Diego, CA