Infusion of cryopreserved autologous lymphocytes using a standard peripheral i.v. catheter

Cryopreserved hematopoietic progenitor cell (HPC) components are generally administered via a central venous catheter (CVC). As most patients receiving these products have an established central venous line, alternative administration routes have not been carefully explored. Here, we report the first series of patients receiving cryopreserved, thawed lymphocytes using peripheral i.v. lines. Newly diagnosed adult, high-grade glioma patients underwent lymphocyte collection before starting radiation with reinfusion after completing radiation. Reinfusion was performed using a temporary 20-gauge peripheral i.v. line. Patients were observed during and after infusion for toxicity. Ten patients underwent a 2-h lymphocyte collection and eight underwent lymphocyte reinfusion. The median cryopreserved product composition contained a volume of 60 cm³, 3 mL of DMSO, a median of 9.9 × 10⁷ nucleated cells/kg and a median of 5.9 × 10⁷ CD3⁺ cells/kg. No toxicities of any kind occurred during or following the peripheral infusion of thawed lymphocytes. This feasibility study in patients with high-grade gliomas provided a unique opportunity to assess the frequency and severity of patient reactions associated with peripheral venous infusions of small volumes of cryopreserved, thawed lymphocytes. The lack of any notable toxicity suggests that patients without central venous access could receive similar products via a peripheral i.v. line.

DMSO dissipates the osmotic stress that would otherwise result in leukocyte lysis during cryopreserving HPCs.^1–3 The presence of DMSO and granulocytes dominate the adverse events (AEs) associated with cryopreserved cellular therapy (CT) infusions.^2,4,5 The DMSO infusion-related AEs are minimalized by limiting receipts to < 1 g/kg of DMSO per day, acetaminophen, histamine blockade and CT infusion via a CVC.^1,2,6

Chemoradiation produces profound and prolonged lymphopenia that is associated with early death from tumor progression in patients with solid tumors.^7,8 This feasibility study (NCT01653834) explored the use of cryopreserved autologous DLIs via peripheral i.v. catheters after induction of chemoradiation.

Patients were eligible if they had newly diagnosed high-grade gliomas with plans to receive standard radiation and temozolomide, Karnofsky performance status ≥60%, hematocrit ≥30%, ANC >1000/μL, absolute lymphocyte count ≥ 1000/μL and a plt count ≥100 000/μL. Patients on anticoagulation or with recent central nervous system (CNS) bleeds were ineligible.

A targeted 2 × 10⁸ lymphocyte collection, by one apheresis procedure, using peripheral i.v. lines was performed 1–10 days before initiating radiation and temozolomide. The cells were transferred to conical tubes and centrifuged for 10 min at 2–8 °C at 1200 r.p.m. The plasma was removed and the cell pellets were resuspended in cryoprotectant (6% Hetastarch in 0.9% sodium chloride injection supplemented with 2% human serum albumin and 5% DMSO). The cells were cryopreserved in multiple cryopreservation bags at a maximum nucleated cell concentration of 2 × 10⁸ cells/mL. The lymphocyte products were frozen in a controlled-rate freezer to −80 °C. The products were stored in the vapor phase of a liquid nitrogen freezer at less than −135 °C and underwent sterility testing before release.

Lymphocyte infusion: All CT infusions occurred as outpatients within 5 days of radiation/temozolomide completion. Seven patients (patients: 2–7 and 10) had CT infused via a 20-gauge peripheral i.v. line. Patients were pre-hydrated with 200 cm³ of dextrose 1/2 normal saline for 1 h and 400 cm³ of dextrose 1/2 normal saline over 2 h after lymphocyte infusion and premedicated with diphenhydramine and acetaminophen. Vital signs were recorded and repeated after each thawed product infusion.

The cryobags were thawed in a 37 °C water bath, visually inspected for clumps and within 1 min transferred into a 60-cm³ syringe and infused via syringe i.v. push through a 3-way stopcock and tubing primed with NS over a 5-min period with planned interruptions for NS flushes, if the patient experienced discomfort or coolness from the infusion. This process was repeated until all of the cells were infused, and patients were hydrated before discharge.

Between 23 July 2012 and 15 May 2013, 26 patients were screened and signed an institutional review board-approved informed consent. Ten (38%) of the screened patients underwent lymphocyte collection. The reasons for not proceeding with harvesting for the remaing patients were: inadequate peripheral venous access (seven patients); total lymphocyte count < 1000/μL (four patients); worsening CNS disease (one patient); active infection at the time of planned collection (two patients); unobtainable insurance clearance (two patients).

Patient characteristics are given in Table 1. Patients had a median peripheral WBC of 6.1 × 10⁶/mm (range 4.77–8.45 × 10⁶) with a median absolute lymphocyte count of 1.4 × 10⁶/mm (range 0.11–2.550 × 10⁶). The median apheresis time was 120 min. No major complications occurred during the apheresis.

Table 1. Baseline patient characteristics.

	Age (years)	Gender	Karnofsky performance status	Histology	Previous debulking surgery	Active use of steroids at time of collection	Peripheral blood counts at time of apheresis
							WBC count (× 106/mm)	Absolute lymphocyte count (× 106/mm)
1	59	F	100	Grade IV glioblastoma	+	No	4.8	1.5
2	40	M	100	Grade III anaplastic astrocytoma	+	Yes	5.7	1.1
3	63	M	80	Grade III anaplastic astrocytoma	Stereotactic biopsy only	No	5.6	1.2
4	42	M	80	Grade IV glioblastoma	+	No	6.4	2.3
5	52	M	80	Grade IV glioblastoma	+	Yes	6.7	0.5
6	41	M	90	Grade IV glioblastoma	+	Yes	8.0	0.1
7	60	M	90	Grade IV glioblastoma	+	Yes	5.0	1.5
8	63	F	90	Grade IV glioblastoma	+	Yes	4.8	0.8
9	67	M	90	Grade IV glioblastoma	+	Yes	8.7	2.0
10	43	M	100	Grade III anaplastic astrocytoma	+	Yes	8.5	2.6
Median	55.5						6.1	1.4

The median volume of the infused cryopreserved product was 60 cm³ (range: 40–100 cm³), containing a median 3 cm³ (range: 2–5 cm³) of DMSO, a median of 9.9 × 10⁷ (range: 5.0–16.7 × 10⁷) nucleated cells/kg, a median of 5.9 × 10⁷ (range: 2.0–9.6 × 10⁷) CD3⁺ cells/kg and a median of 4.2% granulocytes (range: 0.2–32.8%) or 6.2 × 10⁶ (range: 0.2–51.4 × 10⁶) granulocytes/kg (Table 2).

Table 2. Composition of the eight infused cryopreserved products.

	Total volume (cm3)	DMSO volume (cm3)	Nucleated cells × 107/kg	CD3+ cells × 107/kg	Granulocytes × 106/kg
1	40	2	5.0	2.0	13.8
2	60	3	10.8	7.1	–
3	40	2	7.0	3.7	0.7
4	80	4	10.6	6.0	10.4
5	100	5	16.6	6.4	51.4
6	60	3	8.5	5.1	0.2
7	60	3	9.2	5.8	1.5
10	80	4	14.7	9.6	6.2
Median	60	3	9.9	5.9	6.2

After lymphocyte collection, two patients (patients 8 and 9) entered another clinical trial and did not receive their products. Seven patients had their CT delivered within the prescribed time frame via peripheral i.v. infusion. No infusional pain was experienced, no change in vital sign measurements occurred, no infiltrations or other complications of the peripheral i.v. insertion occurred and no hospitalizations were required for any patient.

This feasibility/toxicity study of autologous lymphocyte collection and infusion was undertaken to restore immune function in patients receiving radiation therapy. This study provided an opportunity to determine whether CT requires a CVC. Currently rigorous clinical studies demonstrating that CVCs are required for the infusion of cryopreserved CT products in general or lymphocytes in particular are not available.

Autologous HPCs are cryopreserved in 10% DMSO without hydroxyethylstarch (HES) or in 5% DMSO with 6% HES. 10% DMSO cryoprotectants yield high-osmotic solutions (1400 mOsm) which are associated with peripheral i.v. infusional pain. In a randomized phase III study of autologous HPC comparing cryopreservation using 10% DMSO versus 5% DMSO/6% HES, similar engraftment kinetics were observed suggesting that lower DMSO concentrations are advantageous since DMSO infusion toxicity is dose related.⁴ Recently, Martin-Henao et al.⁴ determined that the granulocyte number in cryopreserved HPCs was the most predictive for infusion-related AEs, occurring when the granulocyte content was >5 × 10⁸/kg.⁴ They surmised that the increased osmotic fragility of granulocytes as compared with mononuclear cells results in their destruction during the freezing process and hence the infusion-related AEs. In comparison, the autologous lymphocytes infused in this trial contained 2 log fewer granulocyte cells/kg.

The lack of AE in our study of patients receiving peripheral i.v. infusions of cryopreserved products may be explained by the use of a 5% DMSO cryoprotectant and the low granulocyte numbers in the products. In an era where the use of CT autologous lymphocytes is likely to expand, practice policies requiring the use of CVC for the infusion of cryopreserved cells should be reviewed and potentially altered depending on product characteristics.