Aim: to value the safety and efficacy of local intramuscular administration of immunoselected autologous endothelial progenitor cells in the treatment of critical limb ischemia in patients without revascularization options.

Primary goal: to value the feasibility of mobilization, harvesting, immunoselection and auto transplantation of endothelial progenitor cells.

Secondary goal: to value the efficacy of local administration of autologous endothelial progenitor cells in the treatment of critical limb ischemia

Type of the study. Prospective single centre not randomized. Aim of the study. To assess the safety, feasibility and efficacy of local intramuscular administration of autologous selected CD 133+ cells in patients suffering from CLI.

All the patients enrolled were suffering from CLI according to the TASC 2 definitions and had no revascularization option, on the basis of contrast CT or angiography imaging. A detailed informed consent, approved by our EC, had been required. An age lower than 18 years and upper than 70 years because of a poor marrow responsiveness to drug simulation in elderly people Clinical unsteadiness of the CLI, such as gangrene requiring major amputation and a poor life expectancy were introduced as exclusion criteria for supposed latency of the EPCs action. Severe systemic illness was judged to increase the risk of the marrow stimulation. Patients with allergic diathesis, child-bearing age, previous EPCs muscular implant, previous medical experimental protocol and any conflict of interest with the study were excluded.

Patients: we enroll patients with a history of Rutherford stage 4 (rest pain) or 5 (small ischemic lesions) PAD. All the patients have previous (contrast CT, MR or angiography) vascular imaging excluding revascularization options, both endovascular and open and encountered the enrolment and exclusion criteria. Every patient undergo routine physical and instrumental examination including electrocardiogram, chest X-ray and blood sample analysis. Younger patients met the Buerger disease criteria, whereas others had pure atherosclerotic lesions.

CEUS Imaging Protocol. Two operators (F.C. and A.G., with 20 and 3 years experience, respectively) who are blinded to treatment perform contrast-enhanced US for all patients. Two US scanner (Philips iU22 (Bothell, WA, USA) with linear array L9-3 transducer and GE Logiq e9 (GE Healthcare, Milwaukee, WI, USA) with linear array L6-9 transducer are used.

The imaging parameters are: reduced transmit power (mechanical index <0.06), at approximately 7-10 frames per second and one focus well below the level of the target to ensure a more uniform pressure field. Dual-mode presentation of a grayscale image side-by-side with the contrast image facilitated real-time identification of anatomic structures and ROI selection. Image loops of approximately 60 seconds were acquired. Effort was made to have a uniform gain across the image and to avoid gain saturation. The TGC (time gain compensation) was set such that before contrast arrival a uniform black image was shown. A vial of contrast agent (SonoVue BR1; Bracco, Milan, Italy) is prepared at a concentration of about 2x108 sulfur hexafluoride-filled microbubbles per milliliter, according to the manufacturer's recommendations. Prior to each injection, microbubbles are resuspended by shaking the vial. The position of the probe is recorded for each patient in order to maintain the same position during follow up. The contrast injection consists of an intravenous bolus of 4.8 ml of contrast agent injected in antecubital vein in 2 seconds followed by a saline flush of 5 ml. The injection is made with the patient supine and after 10 minute of rest to avoid exercise related micro-vascular dilatation. The radiologist maintains a constant image plane with the aid of the tissue (fundamental image) of the "Contrast Side/Side" imaging mode.

Image analysis. The image loops are transferred to a personal computer for further analysis. The main image analysis tasks are: a) identification of tibialis anterior artery (TAA) area, b) selection of a representative region of normal tibialis anterior muscle (TAM) and c) formulation of time-intensity curves. Two manually defined region of interest (ROI), 2 cm and 4 cm sided-squares, are placed, respectively, over the tibialis anterior muscle with no evidence of arterial branches, over tibialis anterior artery and over a small tibialis arterial branch. The ROIs are placed in the same anatomical position for each patient to avoid unwanted differences during follow up examinations. A time/intensity curve (TIC) is obtained for each ROI. From an analysis of CEUS time-intensity curves, we compute regional blood flow (RBF) and volume (RBV). Time-intensity curves are extracted using commercial quantification software (QontraXt v.3.60, AMID, Rome, Italy). This software allows manual region of interest (ROI) selection, measurement of the selected ROI area and provides linear data for the time-intensity curves. For the ROI in the normal ATM effort is made to place the region in an area without large vessels. The ROI of the ATA is a 2 cm square area and the ATM ROI is a 4 cm square area. Time-intensity curves are obtained by computing the mean intensity of pixels comprised within the ROI at each time point. For each image loop are calculated:

• RBV which consists in the total amount of contrast media within the selected ROI, in a period of time. Due to the characteristics of US contrast media, it reflects the quantity of blood in a defined region. It is directly related to the area under the curve (AUC).
• RBF that is in direct ratio to perfusion in a given ROI. It consist in the Contrast Media flow (related to the Blood Flow) in a selected ROI. It is related to the Mean Transit Time.

Marrow stimulation. Human recombinant granulocyte colony-stimulating factor (rhGCSF) is administered subcutaneously for 4-5 consecutive days at a dosage of 10 µg/kg daily, split in 2 doses. Starting from the third day of mobilization, the CD133+ cell count is monitored daily by cytofluorimetric analysis on a 2 ml sample obtained from the patients' peripheral blood. The minimum CD133+ cell count acceptable for leukapheresis (LKF) collection was 10/µl. Patients are monitored for any G-CSF related side effects.

Leukapheresis collection. A single LKF collection is planned for each patient using a third generation cell separator device (Spectra Cobe, Lakewood, CO), processing at least 2.5 blood volumes according to our internal protocol for stem cell collection. Patients are monitored for blood pressure and heart rate during the entire collection procedure. Immediately after LKF a sample from patient's peripheral blood is taken for haemocytometric analysis to evaluate platelet count and Hb levels. Each leukapheresis collection is diluted with 10% acide citrate dextrose (ACD-A) and maintains overnight at 4°C before immunomagnetic cell selection. A sample of 2 ml from each LKF bag is taken for cytofluorimetric analysis. Four hours after LKF collection coagulation parameters, electrolyte levels and hemocytometric analysis is evaluated.

Immunomagnetic cell selection. CD133 immunomagnetic cell selections are performed the day after LKF collection using the Clini-MACS (Miltenyi Biotec) device according to the manufacturer's standard protocol. Briefly, CliniMACS CD133 reagent (formed by super paramagnetic particles composed of iron oxide and dextran conjugated to murine monoclonal antibodies) is added to the cells for incubation. The product is subsequently washed by dilution with buffer (CliniMACS PBS-EDTA buffer supplemented with 0.5% human serum albumin) and the cell bag is hung on the device for the automated selection of CD133 labeled cells. A sample of 2 ml from the positive fraction is taken for cytofluorimetric analysis.

Quality controls. Clonogenic assays. A sample taken from the CD133 cell positive fraction after each immunomagnetic cell selection procedure is seeded for short term (14 days) clonogenic assays. A standard mixture of methylcellulose plus recombinant human erythropoietin, rh stem cell factor (SCF), rhGM-CSF, and rh interleukin-3 (IL-3) is employed (Stem Cell Technologies, Vancouver, BC, Canada; MACS Media, Miltenyi Biotec GmbH, Bergisch Gladbach,Germany). Microbial cultures. Microbial cultures on the immunoselection waste bag containing the negative fraction are carried out to detect aerobic-anaerobic bacteria and fungal contamination. A sample of 10ml is inoculated in the culture medium (Bact/Alert FA and BacT/Alert FN, Organon Teknika Corp., Durham, NC) and incubated for 10 days at 37°C.

Cytofluorimetric analysis. Samples obtained from peripheral blood before mobilization with G-CSF, at time of leukapheresis and after immunomagnetic cell selection are analyzed by flow cytometry to evaluate the expression of specific stem cell and endothelial antigens. Becton Dickinson FACSCanto was employed for all flow cytometric assays with a lyse no-wash technique, using the following monoclonal antibodies: anti-CD45 fluorescein isothiocyanate (FITC) (Becton Dickinson, San Jose, CA), anti-CD34 Peridinin-chlorophyll-protein complex (PerCP) (8G12 clone, Becton Dickinson), anti-CD133 phycoerythrin (PE) (AC133 clone, Miltenyi Biotec) and anti-VEGF-R2 allophycocyanin (APC) (R&D systems). Whole blood is processed following the instructions for VEGF-R2 (KDR) determination. Each blood sample is transferred to a 50 ml falcon tube and brought to a total volume of 30 ml by adding ammonium phosphate containing lysis buffer (Becton Dickinson). After a lysis period of 5 minutes PB is centrifuged at 500g for 5 minutes and washed two times with PBS containing 0,5% bovine serum albumin (BSA). 100 µl of each sample are then transferred in a BD tube for cytometric analysis and incubated for 15 minutes with 1µg/105 cells IgG in order to block all nonspecific sites. 50 µl of the IgG blocked sample are incubated with 20 µl of anti-VEGF-R2 antibody for 30 minutes at 4°C in the dark and then washed twice with PBS. At the end of the last wash step, 10 µl of each of the other antibodies (anti-CD45, anti-CD34, anti-CD133) are added and incubated 10 minutes at room temperature in the dark. Each sample is acquired with BD FACSCanto recording 100.000 events inside the lymphocyte plus monocyte gate. Data files are analyzed with FACS Diva 6.1 software. Viability is assessed using 7-amino-actinomycin D (7-AAD) (Molecular Probes, Eugene, OR). A sample for Hill and EBM2 clonogenic assays is taken from patient's peripheral blood before and after mobilization (at time of LKF collection).

Implant procedure. After loco-regional anesthesia and below the knee cutaneous disinfection, 45-48 ml of autologous CD133+ saline solution suspension is administered intramuscularly with 1 ml deep injection through 18G needle. The injections were so allocated: 10 ml in the anterior compartment of leg, 10 ml in the superficial posterior compartment, 10 ml in the deep posterior compartment, 10 ml in the lateral compartment and the remaining part in the foot.

Baseline assessment and follow up. Pain assessment is carried out with a personal scale of 3 degrees (mild, moderate and severe) and the pain killing drugs use monitoring. Ischemic lesions are treated weekly by a wound management skilled nurse. Preoperative and at 3, 6 and 12 months after the implant are assessed CEUS, lesion evolution and pain management.

Inclusion criteria:

1. age > 18 years
2. written informed consent
3. critical limb ischemia (TASC 2 criteria)
4. No surgical nor endovascular options

Exclusion criteria:

1. clinical instability
2. extensive gangrene
3. every serious systemic disease
4. life expectancy < 24 months
5. previous similar studies
6. previous experimental drug studies within 3 months
7. allergy
8. child bearing age
9. conflict of interest in the study