Dilute human serum has been shown to stimulate proliferation

Transcription

1 BASIC INVESTIGATION Time- and Temperature-Dependent Stability of Growth Factor Peptides in Human Autologous Serum Eye Drops Jay C. Bradley, MD,* Jan Simoni, DVM, PhD, Rachael H. Bradley, PhD,* David L. McCartney, MD, and Sandra M. Brown, MD Purposes: To develop a step-by-step production method for human autologous serum (AS) eye drops that was broadly compliant with US Food and Drug Administration requirements for reinjection of processed biological substances. To determine optimum storage conditions for AS eye drops by measuring the concentration of growth factor peptides (GFP) as a function of storage temperature and storage duration. Methods: AS derived from the blood of 3 healthy male volunteers was produced using a closed, vacuum-driven, cascade-filtration system under sterile, low-pyrogen conditions. In-process controls included methods for monitoring protein electrophoretic mobility and degradation rate and the content of free hemoglobin and endotoxin. Stability of transforming growth factor b 1, substance P, nerve growth factor, calcitonin gene related peptide, insulin-like growth factor 1, and epidermal growth factor was evaluated at -15 C, +4 C, +25 C, +37 C, and +42 C at different time intervals (hours to weeks). The main outcome measures were the concentrations of GFP, endotoxin, and lipid peroxidation by-products (a proxy measure for protein degradation) in dilute AS. Results: The stability of GFP varies: transforming growth factor b 1, nerve growth factor, epidermal growth factor, and insulin-like growth factor 1 were more temperature and time resistant, but substance P and calcitonin gene related peptide significantly degraded at +4 C in 24 hours. Endotoxin and lipid peroxidation by-products were not significantly increased by processing. Conclusions: This pilot study developed a closed, cascade-filtration system that was an effective method for the production of highquality, low-pyrogen AS. The processing method broadly complied with Food and Drug Administration requirements for reinjection of biological substances. Variable GFP stability was observed at +4 C and above. For clinical use, AS should be packaged in daily-use containers, which should be stored frozen; the container in active use should be refrigerated between doses. Received for publication January 2, 2008; revision received June 17, 2008; accepted July 4, From the *Department of Ophthalmology and Visual Sciences, University of California at Davis, Sacramento, CA; Departments of Surgery and Ophthalmology and Visual Sciences, Texas Tech University Health Sciences Center, Lubbock, TX; and Cabarrus Eye Center, Concord, NC. Supported by a Texas Tech University School of Medicine Seed Grant Reprints: S. Brown, MD, Cabarrus Eye Center, 201 LePhillip Court NE, Concord, NC ( sbrownmd@cabarruseye.com). Copyright Ó 2009 by Lippincott Williams & Wilkins Key Words: autologous serum, neurotrophic keratopathy, growth factor peptides (Cornea 2009;28: ) Dilute human serum has been shown to stimulate proliferation of corneal epithelial cells in culture. 1,2 Clinically, diluted serum eye drops derived from the patient s own blood [autologous serum (AS)] promote healing in eyes with neurotrophic keratopathy. 3,4 This effect is ascribed to the rich concentration of numerous growth factor peptides 5,6 (GFP) in serum. According to current literature, transforming growth factor b 1 (TGFb 1 ), nerve growth factor (NGF), substance P (SubP), calcitonin gene related peptide (CGRP), insulinlike growth factor-1 (IGF 1 ), and epidermal growth factor (EGF) are among the most relevant proteins in anesthetic ocular surface disease treatment. Although several comprehensive laboratory protocols for harvesting AS have been published, 7 9 the effect of storage conditions on GFP concentration has not been specifically investigated. We developed a sterile protocol for AS eye drop production that is broadly in compliance with US Food and Drug Administration (FDA) requirements for processing biological substances for reinjection and measured the concentration of 6 major GFPs as a function of storage time and temperature. MATERIALS AND METHODS This human subject research protocol was approved by the Investigational Review Board of Texas Tech University Health Sciences Center. Three healthy male subjects with an average age of 28 years participated. Harvesting and Storage of AS Blood collection was performed using blood-banking standards. Serum processing was performed under sterile, low-pyrogen conditions in the Texas Tech Blood Substitute Production Facility (Lubbock, TX). Specific products used and product suppliers are given in Table 1. A 19-gauge winged infusion set connected to a doublemale Luer lock and then attached to an 18-gauge needle was used for the blood draw. A second 18-gauge needle was fitted with a 0.2-mm filter and was introduced into a 150-mL evacuated container to break the vacuum and prevent hemolysis. Under standard sterile conditions with povidone-iodine or chlorhexidine skin prep, 100 ml of whole blood was drawn 200 Cornea Volume 28, Number 2, February 2009

2 Cornea Volume 28, Number 2, February 2009 Stability of GFPs TABLE 1. Materials for Sterile Autologous Serum Production Blood Collection and Serum Transfer Purpose Manufacturer; Catalog Number Manufacturer Location; Contact Number Approved for Reinjection? Winged infusion set 19 gauge Blood draw Numerous suppliers Yes, if collection container is approved Luer lock double-male adapter Blood draw Numerous suppliers Yes Needle, 18 gauge, 1 inch Blood draw Numerous suppliers Yes Evacuated container 150 ml Blood storage Hospira Inc.; Product # Lake Forest, IL; Yes, approved for total parenteral nutrition Evacuated container 500 ml Serum dilution Hospira Inc.; Product # Lake Forest, IL; Yes, approved for total parenteral nutrition Needle, 18-gauge needle, 3 or 4 inches (can be any sterile needle of sufficient length to extract serum component after centrifugation in 150-mL evacuated container) Statlock IV select standard bore catheter extension set with preattached Clearlink system Serum filtration (in order of use) PharmAssure filter with Supor membrane 0.2-mm pore, 32-mm diameter PharmAssure filter with Supor membrane 1.2-mm pore, 32-mm diameter PharmAssure filter with Supor membrane 0.45-mm pore, 32-mm diameter Serum extraction Numerous suppliers Yes, confirm with manufacturer Serum extraction and filtration (all transfers) Release vacuum Serum filtration: remove residual blood clot Serum filtration: remove residual cellular elements Baxter International; Product # 2N9220K Pall Corporation; Product # HP4642 Pall Corporation; Product # HP4648 Pall Corporation; Product # HP4644 Dilution and storage BSS (balanced salt solution) Plus Diluent Alcon Inc.; Product # or Daily-dose dropper bottle 2 ml (bottles and dropper tips = low-density polyethylene; caps = polypropylene) Product storage Alcan Global Pharmaceutical Packaging Inc. Deerfield, IL; Baxter East Hills, NY; East Hills, NY; East Hills, NY; Fort Worth, TX; Millville, NJ; Yes, if collection container is approved Yes, extemporaneous solutions including ophthalmic drops (as stated on Pall Web site) Yes, extemporaneous solutions including ophthalmic drops (as stated on Pall Web site) Yes, extemporaneous solutions including ophthalmic drops (as stated on Pall Web site) Yes Bottle material type used for most ophthalmic eye drops; can be ethylene gas sterilized from the patient into the 150-mL container. Both needles were removed, and a sterile glove was placed over the top of the container for transport. The blood was allowed to clot for 90 minutes at room temperature (+25 C) and then centrifuged in a swinging bucket rotor at 3000g for 15 minutes; this force and duration were chosen based on prior studies. 7 An 18-gauge needle of sufficient length to reach the serum (we used a spinal needle) was attached to an intravenous (IV) infusion line tubing extender, which was attached to a 1.2-mm filter fitted with an 18-gauge needle. Under vacuum, serum was drawn from the collection bottle into a 150-mL evacuated container. Serum was transferred a second time through a 0.45-mm filter into a 500-mL evacuated container. For each transfer, the bottle access stopper was wiped with 100% ethanol before needle insertion. Serum volume was assessed using a balance, assuming 1 ml of serum is equivalent to 1 gm of weight. The empty bottle was weighed before collection for baseline weight. Serum was diluted to 20% concentration by adding 4 parts BSS (balanced salt solution) Plus to the 500-mL bottle using IV tubing. The dilute serum was filtered a final time through a 0.2-mm filter into a 500-mL evacuated container. The serum was then dripped through IVextension tubing with stopcock into individual 2-mL daily-use storage bottles, which had undergone ethylene gas sterilization, placing 1 gm in each bottle. Bottles were stored in a sterile sealed plastic bag at 280 C until analysis. Storage Conditions Samples stored at 215 C were thawed and assayed at baseline and 1, 2, 3, and 6 weeks; samples stored at +4 C and +25 C were assayed at baseline and 3, 6, 12, and 24 hours. Samples stored at +37 C (normal body temperature) and +42 C (febrile body temperature) were assayed at baseline; 30 minutes; and 1, 3, and 9 hours. Baseline levels were measured once samples were completely thawed at room temperature, which required approximately 15 minutes at +25 C. Measurement of Serum Peptide Concentrations The proteins and other substances assayed, including the method of measurement, are summarized in Table 2. q 2009 Lippincott Williams & Wilkins 201

3 Bradley et al Cornea Volume 28, Number 2, February 2009 TABLE 2. Serum Peptides Analyzed Peptide Abbreviation Molecular Weight Assay Method Units per ml Function(s) in Cornea 5,6 Transforming growth factor b 1 TGFb 1 25 kda ELISA ng Inhibits corneal epithelial and endothelial cell proliferation promoted by other factors; increases keratinocyte migration and proliferation; possibly involved in corneal neovascularization Nerve growth factor NGF 14 kda ELISA* pg Stimulates epithelial cell proliferation and differentiation; stimulates stromal fibroblast growth; stimulates nerve regrowth; induces corneal neovascularization Substance P SubP 1.5 kda ELISA pg Increases corneal epithelial migration and proliferation; stimulates nerve regrowth; stimulates corneal keratinocyte migration and proliferation; stimulates endothelial proliferation; stimulates fibroblast proliferation; involved in corneal neovascularization Calcitonin gene related peptide CGRP 4 kda ELISA ng Stimulates corneal epithelial migration and proliferation Insulin-like growth factor-1 IGF kda ELISA pg Increases corneal epithelial proliferation, migration, and differentiation; stimulates keratinocyte migration and proliferation Epidermal growth factor EGF 6.4 kda ELISA pg Stimulates proliferation of corneal epithelial and endothelial cells; stimulates migration of corneal epithelial cells; mildly increases keratinocyte migration and proliferation; may inhibit corneal epithelial terminal differentiation; stimulates angiogenesis (in conjunction with other factors) during wound healing Lipopolysaccharide (ie, endotoxin) LPS 1 EU = 0.1 ng LPS QCL-1000 EU Component of gram-negative bacterial outer membrane; marker of contamination HGF, hepatocyte growth factor; kda, kilo-daltons; KGF, keratinocyte growth factor; ng, nanograms; pg, picograms; QCL-1000, Limulus Amebocyte Lysate quantitative chromogenic limulus endpoint assay. Available from Cambrex Inc., Walkersville, MD. *For NGF, premeasurement acidification required. For SubP and CGRP, purification required. FDA approved for clinical use. Levels of TGFb 1, NGF, SubP, CGRP, IGF 1, and EGF were quantified by enzyme-linked immunosorbent assay (ELISA) according to the manufacturers instructions. ELISA kits for human TGFb 1,IGF 1, and EGF were purchased from R&D Systems (Minneapolis, MN); for human SubP and CGRP were purchased from Cayman Chemical (Ann Arbor, MI); and for human NGF was purchased from Promega (Madison, WI). Serum samples tested for NGF were acidified by incubation with 1 M hydrochloric acid before the ELISA. Serum samples tested for SubP and CGRP were first purified by C-18 column chromatography as suggested by the manufacturer. Serum endotoxin levels were measured using the Limulus Amebocyte Lysate QCL-1000 assay from Cambrex Inc. (Walkersville, MD). The rate of lipid peroxidation was used as a proxy for the rate of protein degradation. This was assessed by measuring the concentration of thiobarbituric reactive substances (TBARS) via a method developed and described 10 by one of us (J.S.). Three in-process control analyses were performed to assess (1) the effect of filtration on serum peptide profile, using isoelectric focusing on PhastGel IEF 3-9 with silver staining done with Pharmacia PhastSystem high-speed electrophoresis (GE Healthcare Bio-Science AB, Uppsala, Sweden) and digital densitometry (Scan Analysis, Biosoft, Cambridge, UK); (2) the effect of filtration on peptide concentration using the concentration of the heaviest tested molecule, TGFb 1 ; and (3) the effect of production on lipid peroxidation, measured by TBARS concentration. The effect of clotting time was investigated by assaying the concentrations of free hemoglobin and TGFb 1 after clotting times of 1, 2, 4, and 6 hours at room temperature (+25 C) to determine the optimal GFP yield with minimal hemolysis. Free hemoglobin was determined using the benzidine method. 11 All experiments and chemical analyses were run in duplicate. Data were expressed as mean 6 SD. Statistical analysis was performing using the Student 2-tailed t test; P, 0.05 was considered significant. RESULTS Summary results are given in Table 3. SubP showed significant degradation at $4 C, TGFb 1 at 37 C and 42 C, NGF at $25 C, and CGRP at all temperatures except 4 C. At 4 C the CGRP concentration showed substantial decline, although not to the level of statistical significance (P, 0.091). The TBARS concentration did not significantly increase at any temperature or time; endotoxin levels had a single significant increase after 3 hours at 37 C(P, 0.029), however the 9-hour value was not significantly different from baseline; this discrepancy is likely due to the small number of test subjects and the large SDs and requires a larger study to assess fully. The concentration of TGFb 1 was not affected by clotting time (Fig. 1). Filtration had a minimal effect on the qualitative 202 q 2009 Lippincott Williams & Wilkins

4 Cornea Volume 28, Number 2, February 2009 Stability of GFPs TABLE 3. Time to Statistically Significant Decline in GFP Concentration as a Function of Storage Temperature 215 C +4 C +25 C +37 C +42 C TGFb 1 9 h 3 h NGF 24 h 3 h, 9 h 3 h SubP 24 h 24 h 3 h, 9 h 9 h CGRP 6 w 6 h, 24 h 3 h, 9 h 3 h, 9 h IGF 1 EGF LPS 3 h not 9 h TBARS Cells with dashes indicate decline in concentration not statistically significant at P, 0.05 (Student 2-tailed t test); 215 C tested at baseline and 1, 2, 3, and 6 weeks; +4 C tested at baseline and 3, 6, 12, and 24 hours; +25 C tested at baseline and 1, 3, 6, and 24 hours; +37 C tested at baseline, 30 minutes and 1, 3, and 9 hours; +42 C tested at baseline, 30 minutes, and 1, 3, and 9 hours; LPS, lipopolysaccharide. peptide profile as demonstrated by isoelectric focusing and densitometry (Fig. 2). No decrease in TGFb 1 concentration (Fig. 3) and no increase in TBARS (Fig. 4) or endotoxin (Fig. 5) concentrations were observed during in-process controls. DISCUSSION The relative clinical importance of the major GFP in AS topical ophthalmic therapy is unknown, 5,6 and it is likely that synergies between peptides exist. The threshold concentrations on the ocular surface and the duration of contact needed for clinical effect are also undefined. Because AS is a biological substance unique to each patient, it is impossible to know individual peptide concentrations with certainty unless a comprehensive ELISA-based analysis is performed, a process which is inaccessible and cost-prohibitive under clinical circumstances. However, storage conditions are controllable, and they can be designed to minimize protein denaturation from baseline (fresh serum) levels. Because heat accelerates peptide hydrolysis, most investigators use frozen storage of packaged AS with refrigerated storage of 1 active bottle. Some authors 12,13 ascribe therapeutic benefit to AS which has been kept refrigerated for several days or weeks. Although the small number of test subjects limits the strength FIGURE 2. In-process control: effect of filtration on the serum peptide profile. Isoelectric focusing and densitometry showed no qualitative change in the peptide profile as serum traveled through the filtration process. pi = isoelectric point of the protein; when the ph in the gel equals the protein s pi, it no longer migrates within the gel. of statistical analysis, our results do not support this manner of storage. SubP, NGF, and CGRP all showed significant declines in concentration when held at 25 C for 24 hours, with the mean concentration of SubP falling by.50% (P, ). SubP was 1 of the first GFP with proven clinical benefit in the treatment of neurotrophic keratopathy. 14 Having developed the processing method, a study with a larger number of subjects, and testing of additional serum substances such as vitamin A, could be conducted. It is important to remember that an individual patient will not behave like an average and that storage conditions should be conservatively selected to FIGURE 1. Free hemoglobin and TFGb 1 levels in serum as a function of clotting time before centrifugation. There was no significant change in either protein. FIGURE 3. In-process control: effect of filtration on the concentration of TFGb 1, the largest molecular weight protein assayed. No significant reduction occurred either before or after dilution. q 2009 Lippincott Williams & Wilkins 203

5 Bradley et al Cornea Volume 28, Number 2, February 2009 FIGURE 4. In-process control: effect of filtration on lipid peroxidation. Filtration does not increase the concentration of TBARS, a marker for lipid peroxidation and proxy indicator of protein degradation; the decline after filtration at 0.2 mm is due to dilution. encompass the outlier patients showing more rapid degradation. We believe that for greatest stability, dilute AS should be stored in multiple single-day bottles, which are kept frozen until the day of use at approximately 215 C (a typical home freezer is 220 C to 215 C, 24 F to +5 F). Between administrations the thawed bottle should be refrigerated at no greater than +4 C (a typical home refrigerator is +1.7 C to +4 C, +30 F to +35 F). Patients should be specifically instructed not to place a bottle in their pocket or on a table in a warm location for more than a few moments and to discard bottles that have been left unrefrigerated for more than 2 3 hours. Even under frozen storage conditions, CGRP may FIGURE 5. In-process control: effect of filtration on endotoxin lipopolysaccharide (LPS) concentration in dilute serum. There was no significant increase in endotoxin concentration after passage through the 0.2-mm filter. The FDA upper limit for endotoxin contamination of sterile water for irrigation or injection is 0.25 EU/mL. After dilution = AS diluted to 20% concentration with BSS (balanced salt solution) Plus; after filtration = dilute serum passed through filter with 0.2-mm pore sizes. show a statistically significant decline in concentration; the clinical importance of this change is unknown. Human research on the use of topical ophthalmic AS is regulated by the Division of Hematology, Office of Blood Research and Review, Center for Biologics Evaluation and Research of the FDA. This office requires that every step in production utilize only materials that are approved for reinjection of substances into the human body (B. Golding, MD, official correspondence, May 2004). Standard laboratory blood-draw tubes used for serum separation are not approved, for example. One major obstacle was finding an approved container for blood collection that could be centrifuged. The only evacuated container of adequate volume approved for reinjection at the time of study design was a total parenteral nutrition bottle. The manufacturer stated that 3000g would not harm the bottle (J. C. Bradley, personal communication with Hospira Inc.), which we confirmed by a trial centrifugation. The filters utilized in this protocol were selected based on their reported optimal membrane for body fluid filtration (JCB, personal correspondence with Pall Corporation); the manufacturer s Web site specifically describes their use for filtration of ophthalmic drops. The 1.2-, 0.45-, and 0.2-mm filters acted to remove blood clot debris, residual corpuscular elements, and bacteria, respectively. These filters showed no major qualitative effect on the serum peptide profile (Fig. 2) and did not significantly reduce the concentration of the heaviest peptide (Fig. 3). The filtration process did not cause lipid peroxidation (Fig. 4). The FDA limit for endotoxin contamination of sterile water for injection or irrigation is #0.25 endotoxin units/ml (EU/mL) ( With the exception of the spinal needle, all fluids and disposable medical supplies (other needles, tubing, connectors, evacuated containers, and storage bottles) used were United States Pharmacopeia grade and approved for reinjection. We specifically investigated the effect of filtration on endotoxin levels; because all 3 filters were manufactured, packaged, and sterilized identically with the exception of pore size, we verified the manufacturer s claim that each filter contains,0.25 EU/mL of lipopolysaccharide (endotoxin) by measuring lipopolysaccharide before and after passage through the final 0.2-mm filter (Fig. 5). Mean endotoxin level increased by a nonsignificant 0.02 EU/mL and was,0.2 EU/mL for the final diluted AS product. The use of non United States Pharmacopeia medical supplies or other filters with different membranes might reduce the final product concentration of GFP and/or increase the endotoxin level to a clinically important degree, and we do not recommend substitution without in-process control testing. We used a spinal needle (for lumbar puncture) to withdraw serum from the centrifuged total parenteral nutrition (TPN) container. This needle was inexpensive and readily available to our laboratory as it was stocked for hospital use; however, it is not approved for reinjection. An 18-gauge needle 3 or 4 inches in length that is approved for reinjection can be specially ordered; the only difference is the design of the needle tip and hub. The same company often supplies both needles. We produced AS in the Texas Tech Blood Substitute Production Facility (Lubbock, TX), which is Good 204 q 2009 Lippincott Williams & Wilkins

6 Cornea Volume 28, Number 2, February 2009 Stability of GFPs Manufacturing Practice certified to manufacture sterile, lowpyrogen, protein-based IV biopharmaceuticals for experimental use (note, however, that we did not perform formal sterility testing, which would be required by the FDA for approval for clinical research). A standard blood-banking facility could use our process, with the purchase of a few disposable medical supplies and filters (Table 1). Because all steps in manufacture before final decanting occur in a closed system, if aseptic technique is used for blood draw, the risk of contamination is minimal unless there is an environmental breach. Ideally, the blood bank would have a laminar flow hood for sterility during decanting into the daily-use dropper bottles. Because the cascade filtration is driven by the intrinsic vacuum of the containers, no specialized equipment is needed. ACKNOWLEDGEMENTS During the course of this research, Dr. Jay Bradley was a Heed Ophthalmic Foundation Fellow ( ) and would like to acknowledge the foundation s support. Drs. J. Bradley and Simoni had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. REFERENCES 1. Garcia-Hirschfeld J, Lopez-Briones LG, Belmonte C. Neurotrophic influences on corneal epithelial cells. Exp Eye Res. 1994;59: Nakamura T, Inatomi T, Sotozono C, et al. Transplantation of autologous serum-derived cultivated corneal epithelial equivalents for the treatment of severe ocular surface disease. Ophthalmology. 2006;113: Matsumoto Y, Dogru M, Goto E, et al. Autologous serum application in the treatment of neurotrophic keratopathy. Ophthalmology. 2004;111: Bonini S, Lambiase A, Rama P, et al. Topical treatment with nerve growth factor for neurotrophic keratitis. Ophthalmology. 2000;107: ; discussion Imanishi J, Kamiyama K, Iguchi I, et al. Growth factors: importance in wound healing and maintenance of transparency of the cornea. Prog Retin Eye Res. 2000;19: Klenkler B, Sheardown H. Growth factors in the anterior segment: role in tissue maintenance, wound healing and ocular pathology. Exp Eye Res. 2004;79: Geerling G, MacLennan S, Hartwig D. Autologous serum eye drops for ocular surface disorders. Br J Ophthalmol. 2004;88: Lagnado R, King AJ, Donald F, et al. A protocol for low contamination risk of autologous serum drops in the management of ocular surface disorders. Br J Ophthalmol. 2004;88: Tsubota K, Goto E, Fujita H, et al. Treatment of dry eye by autologous serum application in Sjogren s syndrome. Br J Ophthalmol. 1999;83: Simoni J, Simoni G, Garcia EL, et al. Protective effect of selenium on hemoglobin mediated lipid peroxidation in vivo. Artif Cells Blood Substit Immobil Biotechnol. 1995;23: Crosby WH, Furth FW. A modification of the benzidine method for measurement of hemoglobin in plasma and urine. Blood. 1956;11: Yoon KC, Heo H, Im SK, et al. Comparison of autologous serum and umbilical cord serum eye drops for dry eye syndrome. Am J Ophthalmol. 2007;144: Holzer MP, Auffarth GU, Specht H, et al. Combination of transepithelial phototherapeutic keratectomy and autologous serum eyedrops for treatment of recurrent corneal erosions. J Cataract Refract Surg. 2005; 31: Brown SM, Lamberts DW, Reid TW, et al. Neurotrophic and anhidrotic keratopathy treated with substance P and insulinlike growth factor 1. Arch Ophthalmol. 1997;115: q 2009 Lippincott Williams & Wilkins 205