WO1999055395A1

WO1999055395A1 - Method of determining osteogenic potential of human demineralized bone matrix powder

Info

Publication number: WO1999055395A1
Application number: PCT/US1999/009008
Authority: WO
Inventors: Wilson H. Burgess; William N. Drohan; S. Randolph May
Original assignee: American National Red Cross
Priority date: 1998-04-28
Filing date: 1999-04-26
Publication date: 1999-11-04
Also published as: AU5947299A

Abstract

A protocol for determining the osteogenic potential of demineralized bone matrix is disclosed. The protocol includes measurement of physicochemical properties and an in vitro assay. Optionally, the protocol further includes one or more in vivo assays. Criteria that reliably predict osteogenic potential of demineralized bone matrix are disclosed. The present invention provides a reliable source of osteogenically effective demineralized bone matrix powder. Further, the invention provides a means of increasing the osteogenic potential of demineralized bone matrix.

Description

METHOD OF DETERMINING OSTEOGENIC POTENTIAL OF HUMAN DEMINERALIZED BONE MATRIX POWDER

5 This invention pertains to a method or protocol for determining the osteogenic potential of human demineralized bone matrix (DBM) powder wherein the DBM powder is subjected to various combinations of assays. The assays of the present invention include measurement of physicochemical properties, one or more in vitro assays, and optionally, one or

10 more in vivo assays. The invention further identifies criteria indicative of osteogenic potential of a lot of human DBM. The present invention also avails means for selecting osteogenically effective DBM, and DBM formulations thus selected. The invention further presents a means of increasing the osteogenic potential of a lot of human DBM.

15 BACKGROUND

Autogenous bone is used in many procedures to repair bone defects resulting from trauma, tumors, infection and surgical reconstructive procedures. Donor sites for autografts include the iliac crest, ribs and the outer table of the cranium. The use of the grafts suffer from both limited

20 supply and potential post-operative complications associated with their procurement. An alternative to the autograft is allograft materials supplied by a variety of tissue banks. These include cortical and cancellous bone segments, blocks, chips and demineralized bone matrix, often in powder form. Demineralized bone is cadaver bone which is demineralized, sterilized, freeze-dried and packaged. Demineralized bone from a single cadaver constitutes a single lot. Demineralized bone comes in many forms, including whole and split rib, sectioned fibula, femoral head or trochanter, cancellous chips, crushed cortical chips and cortical powder. Demineralized bone is preferred over autogenous bone grafts because no harvesting is necessary, there is a potentially unlimited supply of material, and the bone growth is rapid because demineralized bone provides both osteoinductive proteins and growth factors. (Urist, M. R. and B. S. Strates. 1970, Clin. Orthop. 71 :271-278; Urist, M. R., et al. 1973. Proc. Natl. Acad. Sci. USA 70:3511-3515; EXTRACELLULAR MATRIX BIOCHEMISTRY, Eds. K. A. Piez and A. H. Reddi, Elsevier Science Publishing Co., Inc., New York, 1984). However, there are drawbacks to the use of demineralized bone. For example, demineralization, sterilization and storage procedures vary between suppliers, which may affect the quality of the demineralized bone product. Further, different lots of demineralized bone product exhibit different degrees of osteogenic potential, and some lots exhibit no osteogenic potential.

A demineralized bone implant requires adequate contact with the host tissue to realize its osteogenic potential because it functions by an inductive process. (Toriumi, D. M., et al. 1990, Arch Otolaryngol Head Neck Surg 116:676-680.) For this reason, the use of DBM powder is preferred in certain applications over the use of demineralized bone segments, blocks and chips. The powder provides more surface area, allowing more contact with the host tissue and promoting bone growth. Osteogenesis proceeds more quickly in response to powder than to blocks (Glowacki, J., et al., 1981 , Calcif. Tissue. Int., 33:71-76; Glowacki, J., et al. 1981 , Lancet 1 :959-962). DBM powder is particularly preferred over bone segments, blocks and chips for use in periodontal applications. (Quintero, G. et al., 1982, J. Periodontol. 53:726-730; Mellonig, J. T., 1984, J. Period. Res. 4:40-55; Sonis, S. T., et al., 1983, J. Oral Med. 38:117-122). In particular, DBM powder is used for repairing alveolar clefts and cystic jaw lesions (Kaban, L. B., et al., 1982, J. Oral. Maxillofac. Surg. 40:623-626), correcting small dorsal nasal defects (Toriumi, D. M., et al., 1990, Arch. Otolaryngol. Head. Neck. Surg., 116:676-680), and contouring bone-graft irregularities (Mulliken, J. B. and J. Glowacki, 1980, Plast. Reconstr. Surg., 65:553-560; Glowacki, J. and J. B. Mulliken, 1985, Clin. Plast. Surg., 12:233-241). However, clinical application of DBM powder is limited due to several drawbacks. Among the more significant drawbacks is a substantial variation in osteogenic potential or effectiveness of DBM powder as among various lots. Other limiting drawbacks include difficulties in handling, concerns about particle migration, lack of stability postoperatively and possible resorption into the body. (Lasa, C, et al., 1995, Plast. Reconstr. Surg., 96:1409-1417; Shen, W.J., et al., 1993, Clin. Orthop., 293:346-352).

Some of those drawbacks have been addressed. For example, the initial handling properties of DBM powder have been improved by formulations which allow formation of pliable sheets or moldable mixtures (Levine, S. S., et al., 1995, J. Oral Implant, 18:366-371 ; Gazdag, A. R., et al., 1995, J. Am. Acad. Orthop. Surgeons 3:1-8). These formulations include such materials as glycerol, fibrin, thrombin and like substances. The added material improves handling in the delivery of DBM to the required site and allows shaping of the material to more accurately fill the site (Lasa, C, et al., 1995, Plast. Reconstr. Surg. 96:1409-1417; Schwarz, N., et al., 1989, C//V7. Orthop. 238:282-287). Although fibrin is not known to increase the osteoinductive properties of DBM powder, and in large quantities actually retards them (Lasa, C.I. et al., 1993, J. Surg. Res., 54:202-206), fibrin allows the use of smaller particle size DBM powder (Schwarz, N., et al., 1989, Clin. Orthop., 238:282-287), thereby providing more surface area for osteoinductive growth.

The main deterrent to expanded use of DBM powder in surgical procedures, especially in orthopedic settings, is that about 50% of the lots of human DBM powder fail clinically (Toriumi, D. M., et al., 1990, Arch. Otolaryngol. Head. Neck. Surg. 116:676-680). Schwartz et al. recently reported on the ability of commercial demineralized freeze-dried bone allograft (DFDBA) obtained from 6 different bone banks to induce new bone formation. Based on the variability of the performance of the samples, it was concluded that the osteogenic potential of each DFDBA lot should be determined before it is sold to the practitioner for dental or other use (Schwartz, Z., et al., 1996, J. Periodontol. 67:918-926). The pre-clinical experience of the inventors herein with various lots of human DBM powder has been similar.

The variation in effectiveness of DBM powder lots has been speculated to be caused by multiple factors, including the age, sex and health of the donor, the part of the body from which the bone is harvested, the method of harvesting and preparing the DBM powder, and the storage of the DBM powder. It is not known in the art which of these factors, if any, is actually critical to the determination of osteogenic potential of any given lot of DBM powder.

It is possible to determine the osteogenic potential of a given DBM powder lot through examination of its osteoinductive and osteoconductive abilities. Osteoinduction is the phenotypic conversion of connective tissue into bone by an appropriate stimulus, and osteoconduction is the ingrowth of vessels over the allograft bone implant, followed by resorption of the implant and deposition of new bone. The sequence of bone induction resulting from the use of DBM powder was first described using rat demineralized cortical bone matrix (Urist, M. R. and B. S. Strates, 1970, Clin. Orthop., 71 :271-278; Urist, M. R., et al., 1973, Proc. Natl. Acad. Sci. USA, 70:3511-3515). DBM powder was implanted subcutaneously in recipient rats and resultant bone growth was observed. By day 5, mesenchymal cells differentiated, by day 7 chondroblasts developed, and osteoblasts appeared at day 9. Angiogenesis and chondrolysis were observed at day 11. New bone formation was detected between days 12 and 18. Ossicle formation and hematopoietic marrow were observed by day 21 (EXTRACELLULAR MATRIX BIOCHEMISTRY. Eds. K. A. Piez and A. H. Reddi, Elsevier Science Publishing Co., Inc., New York, 1984). The details of the temporal cascade of cells, enzymes and stromal products leading to DBM-induced heterotopic osteogenesis have been described in various articles. See, for example, Reddi, A. H. and W. A. Anderson, 1976, J. Cell Biol. 69:557-572; Triffitt, J. T., 1987, Ada Orthop. Scand., 58:673-684; Bernick, S., et al., 1989, J. Orthop. Res., 7:1-11 ; and Sawyer, J. R., et al., 1991 , Calcif. Tissue. Int. 49:208-215. DBM is known to provide both osteoinductive proteins (Sampath, T. K., et al., 1987, Proc. Natl. Acad. Sci. USA 84:7109-7113; Urist, M. R., et al. 1984, Proc. Natl. Acad. Sci. USA 81 :371-375) and growth factors (Hauschka, P. V., et al., 1986, J. Biol. Chem., 261 :12665-12674). Testing of the osteogenic potential of DBM powder has been done both in vitro and in vivo. In vitro testing has been done, for example, on rat osteosarcoma lines, wherein DBM powder has been shown to promote bone growth, as evidenced by a rise in alkaline phosphatase levels and increased osteoblast production (Shteyer, A., et al., 1990, Int. J. Oral Maxillofac. Surg. 19(6):370). An increase in calcium concentration is also indicative of bone growth promotion (U.S. Patent Nos. 5,171 ,574; 5,162,114 and 4,975,526). However, in vitro testing is not considered indicative of success in vivo. In vivo testing has been done in rats by inserting DBM into a muscle pouch in the rat and observing whether bone formation takes place over a period of time. The osteogenic potential of DBM has also been tested in rats by creating a muscle island flap, placing DBM next to it, and observing whether ossification of the muscle island flap occurs (Viljanen, V. V., et al., 1997, J. Recon. Microsurg. 13(3):207; and U.S. Patent Nos. 5,171 ,574; 5,162,114 and 4,975,526). Further, critical defects, such as 6 mm or larger holes in rat calvaria, have been used to determine the osteogenic potential of DBM materials, including powders. Such studies indicate that the size of the DBM powder particles does not appear relevant to the success or failure of DBM powder to promote bone growth (Schwartz, Z., et al., 1996, J. Periodontol. 67:918-926; Fucini, S. E.,et al., 1993, J. Periodontal. 64(9):844).

Although testing of the osteogenic potential of DBM powders has been performed as indicated above, the testing has been done almost exclusively as part of experiments to test other factors, such as the effect of particle size or use of fibrin with DBM powder, or to test the efficacy of DBM powder as compared to other forms of demineralized or autogenous bone grafts. Other testing for bioactivity of demineralized bone, including powder, has been done to confirm bioactivity before use in a specific protocol, or to examine consistency of the processing and storage of DBM. These tests have led to the determination that some simple means of determining the osteogenic potential of demineralized bone, particularly powder, needs to be formulated in order to better predict the efficacy of a given sample before use and, preferably, to formulate a standard of osteogenic potential for DBM powder sold by various suppliers. Glowacki et al. in particular have suggested the use of a rat bioassay as the most reliable indicator of bioactivity (1985, Clin. Plast. Surg. 12:233-241). 7

However, despite the recognition that various demineralized bone samples possess variable levels of bioactivity, no procedure for screening samples has yet been established. No in vitro test has been recognized as reliably predicting in vivo bioactivity, and no correlation between in vitro and in vivo results is currently known in the art. Therefore, animal testing (in vivo), which is time consuming, expensive and labor intensive, would seem to be required. Further, other criteria for predicting the osteogenic potential of a given lot of DBM have not been identified, nor have any correlations between such criteria and DBM effectiveness been determined. This invention provides means for combining in vitro and in vivo screening assays to reliably determine the osteogenic potential of different lots of DBM powder. These combination protocols minimize animal usage and provide predictive value for performance of the powder in clinical settings. Criteria for determining the osteogenic potential of a given DBM powder sample are disclosed, and a solution to the problem of variability in performance of different lots of DBM powder is proposed. Further, a means of increasing the osteogenic potential of DBM lots is provided.

SUMMARY OF THE INVENTION

The present invention provides a protocol for determining the osteogenic potential of DBM powder. The protocol includes analyzing a sample of DBM powder from a particular lot for physico-chemical properties such as calcium concentration, particle size, and/or osteocalcin concentration. The sample is subsequently tested for osteoinductive and osteoconductive potential by in vitro and/or in vivo testing. The results of those tests are reliable predictors of osteogenic potential. The present invention thus further affords a means for formulating DBM powder lots according to osteogenic potential and identifying the resulting DBM 8 formulations. The resulting DBM formulations possess consistent and/or quantifiable osteogenic potential.

One objective of the invention is to reduce reliance on in vivo animal tests for determining osteogenic potential. This objective is met by the present invention by identifying and specifying one or more physicochemical characteristics for DBM powder that correlate with improved performance, and further by performing in vitro testing of DBM powder samples. Upon receipt of positive results for the specified physico-chemical properties and positive in vitro assays, one might then perform in vivo animal testing to further confirm, and perhaps quantify, osteogenic potential. With the continued application of the present protocols, and use of the resulting DBM powder lots, reliance on, and resort to, in vivo animal testing will be still further reduced.

A further objective met by the present invention is the improved and more efficient use of DBM resources. By providing DBM lots of greater and more consistent osteogenic potential, use of DBM powder will be more reliable, and DBM powder will find more routine application in clinical procedures now considered high risk, e.g., spine reconstruction/restoration. Thus, the invention satisfies the clinician's unmet desire for a reliable means for predicting osteogenic potential of DBM powder from relatively small and widely varying lots, expands the clinical application of DBM, and thereby provides previously unavailable therapies.

Further, the present invention provides a means of increasing the osteogenic potential of some existing DBM lots. This will further aid in making the most effective use of DBM resources. 9

BRIEF DESCRIPTION OF THE DRAWINGS

The figures are intended to help illustrate the invention described herein. Other embodiments of the invention will be evident to those of skill in the art from the description herein and accompanying claims. Figure 1 : Photomicrographs of Murine 2T3 cells after 5 and 15 days of treatment with DBM powder in vitro. Figure 2: Alizarin stain of in vitro 2T3 cultures. Figure 3: Calcium content of DBM powder implants following 28 days implantation in the rat in vivo assay. Figure 4: Photomicrographs (magnification 400X) of sections of various implants of rat or human DBM powder following explant at 28 days. Figure 5: X-rays from rat craniotomy implant assay. Figure 6: Graph showing % calcium of human DBM powder versus % calcium dry weight of 30 mg ectopic implants of varying amounts of a good (G) lot of human DBM powder mixed with a bad lot, following 28 days in vivo. Figure 7: The effect of particle size on implant mineralization when delivered with fibrin sealant. Figure 8: Decision tree or flow chart of testing protocol.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

The present invention includes a series of assays for determining the relative osteogenic potential of a lot of human DBM powder. Those assays can be used in various combinations to better identify and, hence, screen for effective lots of DBM. The use of the term "relative osteogenic potential" is used to indicate that the protocols and assays of the present invention can be used to demonstrate the osteogenic potential of a DBM lot relative to other lots, mixed lots, or some other standard of osteogenic potential. 10

A preferred protocol minimizes the use of animals while enhancing the ability to detect osteogenically effective lots of DBM powder, as defined by the osteogenic potential. For purposes of the present disclosure, an osteogenically effective lot is that which demonstrates about 60% healing over about 28 days of a critical cranial defect study as defined below, and/or that which stimulates the proliferation and subsequent bone nodule formation by osteoblast cells in culture. The invention further provides various physico-chemical properties and other criteria and assay results predictive of effective lots of DBM. The invention further includes DBM powder selected by such protocols, and DBM powder having effective or high osteogenic potential. Further, the invention includes means of increasing the osteogenic potential of existing DBM lots.

Inconsistency in the effectiveness or clinical performance of DBM powder is a longstanding and widely acknowledged problem. Wide variation in osteogenic potential among lots of DBM powder in both animals and humans has been observed. Until now, a meaningful correlation of osteogenic potential with any of numerous variables among lots of DBM powder has not been reported.

We have identified several criteria useful in correlating, if not quantifying, osteogenic potential in a given sample of DBM powder.

Contrary to common belief and expectations, the osteogenic potential of DBM does not correlate with the age, sex or race of a donor. Rather, osteogenic potential correlates inversely with several physico-chemical properties, including particle size, calcium concentration, and osteocalcin concentration. For example, as particle size decreases, osteogenic potential increases. This is particularly so when DBM powder is delivered as a fibrin sealant putty, or when DBM powder comprises cortical bone material. DBM powder particle size is preferably from about 75 to about 180 μ , more preferably no more than about 150 μm, and still more preferably 11 no more than about 125 μm to maximize osteoinductive activity. Further, reducing a sample of DBM to a smaller particle size of < 180 μm, preferably < 150 μm, most preferably < 125 μm, greatly increases the osteogenic potential by expressing more surface area of the DBM, allowing for greater osteoinductive and osteoconductive growth.

Osteogenic potential of DBM also correlates inversely with calcium concentration (i.e., % Ca, dry weight). Preferably, calcium concentration is in the range of from about 0 to about 0.2 %; more preferably no more than about 0.05%; and still more preferably no more than about 0.02%. Preferred lots of DBM have both a small particle size and low calcium concentration. That is, preferred osteogenically effective DBM lots are those wherein the DBM is a powder having, for example, a particle size of no more than about 180 μm, and a calcium concentration of no more than about 0.2%. We have also discovered that the osteogenic potential of DBM correlates inversely with osteocalcin concentration. Lots of DBM having lower quantities of osteocalcin concentration have greater osteogenic potential. Accordingly, the osteocalcin concentration should be minimized; or lots having low osteocalcin concentrations should be preferentially selected.

A reliable means for identifying DBM lots having osteogenic potential in the effective range can include measuring or detecting any one or more of the above factors. An assay for determining the osteogenic potential of a DBM powder sample therefore determines relative quantities or measures such physico-chemical properties as particle size, calcium concentration and/or osteocalcin concentration.

While means exist to reliably measure such properties as particle size and calcium concentration in materials such as DBM powder, the correlation of those measurements with effectiveness of a DBM lot has not heretofore 12 been reported. Thus, for example, particle size and calcium concentration can be measured by conventional, facile means. The information gained from the assay can then be used to determine effectiveness or osteogenic potential of a sample by resort to the information contained herein and/or by comparison to other samples of known osteogenic potential.

Osteocalcin is the most abundant non-collagenous protein in bone extracellular matrix. Although the precise function of osteocalcin in not known, recent studies suggest that endogenous osteocalcin might be a negative regulator of de novo bone formation. (Ducy, P. et al. 1996, Nature 382:448-452). Measurement of osteocalcin is a laborious, but worthwhile process. Osteocalcin can be extracted from DBM powder using 4M guanidine HCI at neutral pH. The extracted proteins can be purified by reverse-phase HPLC and the identity of individual fractions (osteocalcin) can be confirmed by direct protein sequencing. Other methods of extracting and detecting osteocalcin are known to those skilled in the art, and include protein sequencing, PCR, and/or quantitative analyses such as HPLC. Alternatively, an ELISA assay can be used for bone extracts to detect the presence and relative quantity of osteocalcin. Thus, osteocalcin assays can be facilitated, and the effectiveness of given lots of DBM powder more easily determined.

Heretofore, osteocalcin has not been reported to be a reliable indicator of the osteogenic potential of DBM powder, nor has it been shown to be retained in DBM powder samples. However, osteocalcin can be found in DBM powder samples and its concentration appears to be a valuable and reliable criterium for measuring osteogenic potential. As osteocalcin concentration decreases, the likely effectiveness of a DBM powder sample increases. Consequently, by screening for DBM powder lots that have low concentrations of osteocalcin, one can reliably produce DBM powder having demonstrable effectiveness. 13

The present invention further provides protocols for new assays, including both in vitro and in vivo testing, that also correlate reliably with osteogenic potential. These assays are discussed more fully below. A screening protocol that uses quantitative chemical assays as described above in combination with one or more of in vitro or in vivo assays for determining osteogenic potential of DBM powder enhances the reliability and confidence in the assessment of osteogenic potential of any given DBM sample. Such a protocol has not heretofore been developed or used. However, the enhanced reliability of the protocols described herein will eliminate ineffective lots of DBM powder from distribution, enhance consumer confidence, and enhance the range of applications for DBM powder, thus providing a wider range of available, reliable therapies for the public.

Protocols of the present invention include sampling a DBM lot, assaying the sample for one or more of the physico-chemical properties identified above, and conducting in vitro testing of that sample. Optionally, and if indicated, in vivo testing might be performed on the sample. As will be described more fully below, there is a reliable, predictive relationship between the outcome of one or more of the assays described herein and the osteogenic potential of a DBM lot. In particular, a negative in vitro test is indicative of little or no osteogenic potential, and no further testing is warranted of such a sample. A positive in vitro test is indicative of moderate to excellent osteogenic potential of such a sample. A positive in vivo test, in combination with a positive in vitro test, is indicative of excellent osteogenic potential. Therefore, upon receipt of a positive in vitro test, one might conduct an in vivo test to further confirm and quantify the relative osteogenic potential.

Several in vitro assays predictive of osteoinductive and osteoconductive capacity are known in the art, and can be used to 14 determine the relative osteogenic potential of DBM without resort to animal testing. For example, an osteoblastic cell assay, which can use 2T3 cells, can identify osteogenically effective lots of human DBM, i.e., those that stimulate the proliferation and subsequent bone nodule formation by osteoblast cells in culture. Other useful in vitro assays include Western blot analysis for expression of specific proteins involved in new bone formation and analysis of cDNA expression assays (Clontech) wherein analysis of hundreds of cDNA's can be evaluated in a single hybridization. Other tests may be known to those skilled in the art. Positive results of any of the above in vitro assays conclusively indicates moderate to excellent osteogenic potential of the tested sample. Thus, a positive result in an in vitro assay indicates that the DBM powder sample will have at least a moderate level of osteogenic activity. A positive result is defined as hyperproliferation of cells around the DBM powder and relatively large nodules positive for alizarin staining (see Fig. 2).

In contrast, a negative result in an in vitro assay reliably indicates failure in subsequent in vivo assays. A negative result is indicated by a lack of hyperproliferation of cells adjacent the DBM powder and alizarin staining, similar to controls. Thus, the number of experimental animals required to assess the osteogenic potential of a DBM powder sample can be reduced because a positive in vitro result is indicative of at least moderate osteogenic activity, while a negative in vitro result indicates a definite lack of osteogenic activity.

While it is widely held that a DBM lot producing a negative in vivo test in a rat model indicates little or no osteogenic potential, our results show that the more reliable negative indicator as to whether a lot is osteogenically effective or ineffective is an in vitro test such as disclosed herein. Depending upon the results of the in vitro test, a negative in vivo test might signify either: (1 ) little or no osteogenic potential; or (2) at least moderate 15 osteogenic potential. That is, a DBM lot producing a positive in vitro test, despite a negative in vivo test, might still have at least moderate osteogenic potential. Moreover, a positive in vitro test in combination with a positive in vivo test indicates excellent osteogenic potential. A DBM lot producing a negative in vitro test, however, most likely has little or no osteogenic potential, and the DBM lot can be discarded before proceeding to the in vivo test. Thus, the in vitro test alone is a valuable predictor and can be used as a reliable pre-screen to reduce reliance on, and hence the number of, in vivo tests. An in vivo test predictive of an osteogenically effective lot of DBM might be one of many known in the art, such as insertion of DBM powder into a muscle pouch or by a muscle island flap (also referred to herein, collectively, as an "ectopic implant assay"), as described previously herein and known to those skilled in the art. Most desirable is a test that is simple, relatively non-invasive, and that requires a short evaluation period. A positive ectopic implant assay is one wherein the resulting explant possesses greater than about 5% dry weight calcium and demonstrates new bone formation by histological assay (See Figures 3 and 4). This simple ectopic bone formation assay reflects the ability of different lots of human DBM powder to facilitate the differentiation of mesenchymal stem cells that subsequently remodel the matrix into new, autologous bone.

If desired, yet another in vivo assay might be added to, or substituted into, the protocol to confirm the results of the first two assays. An in vivo assay known as a fatal or critical cranial defect might be employed. This test involves the removal of a circular section of the skull of a rat, insertion of DBM into the void created in the skull, and a determination or measurement of the restoration of bone. An example of such a critical defect assay is the calvaria defect model. Our results with the calvaria defect model demonstrate that failure of the DBM powder in both the in vitro and ectopic 16 in vivo assays reliably predicts failure in a critical defect in vivo model, showing a lack of any obvious osteoconductive activity (see Figs. 4 and 5, lot 9). The calvaria critical defect model further demonstrates that a human DBM powder lot that performs well in an in vitro assay, but poorly in an ectopic implant assay, produces substantial healing (see Figs. 4 and 5, lot 14). However, a human DBM powder lot that performs well in both in vitro and ectopic in vivo assays promotes highly efficient and reliable healing (see Figs. 4 and 5, lot 15). Therefore, the critical defect assay may be used to demonstrate the utility of a negative result in the in vitro screen, which accurately predicts failure in in vivo studies, and to verify the outcome of positive results, as indicated in the table below.

IN VITRO TEST IN VIVO TEST RESULT: CRIT. DEFECT TEST

- - negative growth

+ - moderate growth

+ + excellent growth

It is surprising that a human DBM powder lot that performs poorly in both the in vitro and ectopic in vivo assays does not exhibit osteoconductive activity because prior studies in the field have been unable to determine any correlation between in vitro and in vivo assays that would reliably predict the osteogenic potential of a DBM powder lot.

In contrast, the inventors herein have discovered an in vitro assay is sufficient to conclude whether or not a sample will possess osteogenic activity. However, a further in vivo test, or consideration of other indicative factors of osteogenic potential, is desirable to determine the level of relative osteogenic activity of a DBM powder sample. The determination of the amount of osteogenic potential of a sample offers a solution to the problem of false positives in the screening of DBM powder and the questionable 17 utility of lots which perform well in vitro but not in vivo. Thus, for example, lots of DBM powder with a positive in vitro test can be reliably characterized as having at least moderate or effective osteogenic potential, and might be pooled. Testing of mixtures of good and bad lots indicate that the presence of a small fraction of "bad" DBM powder (provides little to no osteogenic potential) in a pool of "good" material (having moderate to excellent osteogenic potential) does not affect the osteogenic potential of the good lots in the pool, as shown in Figure 6. Although the American Association of Tissue Banks currently does not support the pooling of DBM powder derived from different donors, the results presented here indicate that pooling of screened lots of the material would insure that every sample of DBM powder would provide at least an effective amount of osteogenic potential. The specific activity of the pool would only be diluted by the percent of the bad or less than optimal lots that passed the screen. The concerns over pooling of human DBM powder from different donors is also driven by issues of viral safety. However, viral reduction occurs during the processing of human DBM powder, and is documented (Scarborough, N. L., et al. 1995, Cont. Orthep. 31 :257-261 ; Prewett, A. B., et al. 1992, ORS Trans. 17:436). Further, current studies are in progress to provide additional viral reduction to the human DBM powder in its source vials. Therefore, combining marginal and good DBM powder lots can be done with relative safety, providing samples with osteogenic potential to all consumers. This will extend the amount of osteogenic material available, while ensuring that all such material has osteogenic potential. Excellent and marginal lots can be combined in a ratio of from about 3 to about 1 of excellent to moderate DBM powder lots, preferably in a ratio of from about 9 to about 1 excellent to moderate DBM powder lots.

As can be seen from the foregoing, particular combinations of assays can vary. Preferred assay protocols will include assessment of one or more 18 of particle size, calcium concentration and osteocalcin concentration, in vitro testing and, if necessary, the in vivo muscle pouch or muscle island flap test. Alternatively, if indicated by in vitro results, the fatal rat defect (or critical cranial defect) test can be used as a reliable means for confirming the osteogenic potential of a given lot of DBM powder. The fatal rat defect test can be used as a substitute for, or in addition to, the muscle pouch or muscle island flap test. It will be appreciated by one of skill in the art that the various protocols (i.e., combinations of assays) can be varied according to the intended indication for the given lot of DBM powder being tested. For example, more rigorous tests (e.g., fatal rat defect test, osteocalcin concentration) can be implemented where the particular lot of DBM powder is to be used for clinical applications of a more complicated or intrusive nature, such as spinal repair or osteoporosis therapies. Less rigorous protocols might be used for DBM lots indicated for use in less intrusive therapies such as periodontal applications.

It is noted that the use of relatively small particle sizes of DBM is enabled by the use of fibrin sealant in the delivery of DBM powder. This allows the use of a relatively small particle size of DBM in a format which is moldable and adherent to desired implantation sites. Further, the use of smaller DBM particles is encouraged because the inventors herein have determined that the osteogenic potential of a sample of DBM is increased upon pulverizing or reducing the sample to smaller particle sizes. The advantage to smaller particle size is thus twofold. The smaller particle size of DBM increases the osteogenic potential per particle, thus lowering the amount of DBM needed in any particular application, and this enables the existing stores of DBM powder to be spread over more patients in need, thereby reducing the need for DBM material.

Thus, combining lots of human DBM powder that perform well in the relatively simple protocol of assays, in vitro testing and ectopic in vivo 19 testing will improve the utility of human DBM powder in clinical applications and provide more DBM powder for use. This process insures the inherent osteogenic potential for all lots of human DBM powder. Further, following the protocol will greatly reduce reliance on animal testing. A means of ascertaining the osteogenic potential of DBM powder is demonstrated in the following examples, meant to be illustrative only and not limiting of the scope of the invention. Other embodiments of the invention will be obvious to one of ordinary skill in the art based on the foregoing disclosure. All references cited herein are incorporated in their entirety by reference.

EXAMPLES Materials:

Sterile human DBM was obtained from the American National Red Cross (ANRC) Tissue Services. Human fibrin sealant and thrombin preparations were obtained from the Plasma Derivatives Department at the Holland Laboratory (Rockville, MD). Male athymic rats were from Harlan Sprague Dawley, Inc. (Indianapolis, IN). The mouse 2T3 osteoblastic cell line was established as described in Ghosh-Choudhury, N., et al. 1996, Endocrinology 137: 331-339. All cell culture reagents were from Biofluids. Other chemicals were reagent grade or better.

ANRC DBM (size 1-3 mm) was ground in a Micro-Mill (Scienceware, Bel-Art Products) and sieved through a serial sieve. The opening micrometer of the sieves used were: 75, 180, 500, 850 or 1000 μms. The appropriate size of DBM powder was collected according to the experimental design. 20

Methods:

1. In vitro bioassay: Calvarial osteoblasts were isolated from BMP-2 T- Ag-3 transgenic founder mice as described by Ghosh-Choudhury, et al. 1996, Endocrinology 137 at 331-339. The cells were plated at -10,000 cells/well in 6-well tissue culture plates in αMEM containing 7% fetal calf serum (FCS). They were grown to confluence (day 0), and the media was changed to differentiation media (7% FCS in αMEM containing 100 μg/ml ascorbic acid and 5 mM β-glycerophosphate). Approximately 0.5 cc of human DBM powder (ANRC, Tissue Services) from various lots were added to the plates. The media was changed every 2-3 days with no further addition of DBM powder. At various times, the cells were examined by light microscopy and photographed or fixed with 10% formulin prior to Alizarin staining for mineralized nodules.

2. In vivo bioassay: Four-week old, male athymic rats (rnv/rnv, Frederick Cancer Research and Development Center, Frederick, MD) were anesthetized with a mixture of 9 ml ketamine hydrochloride (Ketaject, Fort Dodge Laboratories, Inc., Fort Dodge, Iowa) USP, 100 mg/ml and 1.0 ml xylazine (Rompun, Miles Inc., Shawnee, KA) USP, 100 mg/ml. The dose was 0.1 ml/100 g body weight, administered intraperitoneally with a 27.5 gauge needle. After the anesthesia, the back of each rat was disinfected with 70% isopropyl alcohol and draped for sterile surgery. A 1 cm incision was made along the dorsal midline through the skin and subcutaneous tissue near the gluteal region. The fascia of the longissimus dorsi muscle was incised over the implantation site. The muscular pouch was created between fascia and muscle by blunt dissection and 30 mg DBM powder with 20 mg/ml fibrin sealant (Plasma Derivatives Department, Holland Laboratory) was packed into the muscle pouch using a 1 ml syringe. The pouch was closed with sutures of 6-0 polypropylene and the skin was closed 21 with 9 mm Michael wound clips. Rats were euthanized by CO₂ 28 days after implantation. Each rat received 2 DBM implants, and at least 3 rats were used for each data point.

The implants were retrieved and excess tissue was removed. One half of each explant was fixed in 10% neutral phosphate buffered formalin for 24 hours, decalcified for 48 hours in 10% formic acid, embedded in paraffin, sectioned and stained with hematoxylin and eosin prior to examination by light microscopy. The second half of each explant was dried at 95°C, weighed, ashed at 600°C for 18 hours, dissolved in 1 N HCI and the calcium content was determined with arsenazo III (Sigma, St. Louis,

MO). The calcium content of each explant was expressed as weight percent calcium of explant ashed weight.

In conducting research on the athymic rats, the investigators adhered to the "Guide for the Care and Use of Laboratory Animals" (NIH publication 85-23). All protocols were approved by the Institutional Animal Care and Use Committee. The studies were conducted in an AALAC accredited facility.

3. Craniotomy Implant Assay: Thirty milligrams of human DBM powder was placed into a plastic mold that produced 1 x 8 mm disks to fit the 8 mm cranial trepan defect. The DBM powder was mixed with human fibrinogen (20 mg/ml final concentration) and human thrombin (2.5 U/ml final concentration) and then added to the mold to make the disks. Disks containing fibrin sealant alone and rat demineralized bone plus fibrin sealant were also made to serve as controls. The rats were anaesthetized with a solution containing 10 ml ketamine hydrochloride (Ketaject, Fort Dodge Laboratories, Inc., Fort Dodge, Iowa), 5 ml xylazine (Rompun, Miles Inc., Shawnee, KA) and 1 ml physiologic saline (0.9% NaCI, Abbott Laboratories, North Chicago, IL). Each rat received 0.14 ml of this mixture per 100 g body 22 weight intramuscularly. All surgical procedures on athymic rats were performed under sterile conditions in a laminar flow hood. Once the operative site was treated with 70% alcohol and 1 % Chloramphenicol ointment (Pharmaderm, Melville, NY), a midsagittal incision was made from the posterior occipital protuberance to the nasal bone. Soft tissues were removed. The periosteum was removed from the operative site as well (occipital, frontal, parietal bones). An 8 mm trepanation defect was created using a trephine in a low speed rotary handpiece. Copious irrigation with physiologic saline was used to keep the site moist and cool. The calvarial disk was carefully removed avoiding dural perforations and superior sagittal sinus intrusion. The fibrin sealant-human DBM disk (1 x 8 mm) was then added to the defect. Separate defects were filled with fibrin sealant or rat demineralized bone. Soft tissues were closed with absorbable sutures and the skin was closed with staples. Post-operative care included keeping the animals warm to minimize heat loss and housing the athymic rats in individual, sterile cages. After 28 days, rats were euthanized in CO₂ chambers. Using a bur and a handpiece, the fronto-occipito-parietal complexes plus the craniotomy sites were retrieved. The implants were x- rayed with Dupont Microvision C mammography film in a MinXray X750G (MinXray, Inc., Northbrook, IL) benchtop x-ray system at 55-60 kVp, 14-13 mA, for 4-8 seconds.

4. Particle Size: Lots of cortical DBM powder that had been processed as a 1-3 mm particle size were fractionated to a series of decreasing particle sizes as described above, combined with fibrin sealant, clotted and assayed in the simple in vivo implant model. The results of the analysis are shown in Figure 7. The implants exhibited a near linear increase in mineralization as the particle size was decreased over the range of 1-3 mm to 75-180 μm. This increase in osteoinductive activity with decreasing particle size appears 23 to be limited to cortical bone, as increasing the particle size of cancellous preparations generally reduces the osteoinductive activity of the preparation.

Assay Results

The above protocol was designed to assess the ability of DBM powder to induce new bone formation while reducing the amount of animal testing needed. The limitations of osteosarcoma and primary fetal rat calvarial osteoblast models have been described by Ghosh-Choudhary et al., 1996, Endocrinology 137: 331-339, whose 2T3 cell system was used for the in vitro studies. The 2T3 cell system is an immortalized osteoblastic cell line that maintains differentiation capabilities including the ability to form bone nodules in vitro. Typical positive and negative results of this assay are shown in Figures 1 and 2. The cells are cultured to confluence, then particles of DBM are added to the dish and the media is changed to a differentiation inducing formula. Microscopic examination of the cultures 5 days later reveals hyper-proliferation of the osteoblastic cells around the positive (+) DBM particles. This proliferation is not evident in the negative (-) cultures. By day 15, newly formed bone nodules can be seen in the positive cultures (Figure 1 ). The new bone formation is readily apparent following Alizarin staining for mineralized (calcified) nodules (Figure 2). The 2T3 cells alone will begin to form mineralized deposits but the presence of lots of DBM having osteogenic potential accelerates the process.

The general utility of the in vitro assay in assessing the osteogenic potential of different lots of DBM powder was examined in a simple in vivo model of ectopic bone formation. Discs consisting of 30 mgs of various lots of DBM powder formed with 20 mg/ml fibrin sealant were implanted into an intramuscular pouch in the gluteal region of athymic rats. The discs were excised after 28 days and analyzed for new bone formation using calcium assays, as a measure of mineralization, and histological examination. The 24 percent calcium values of explant dry weight of various lots of DBM powder are shown in Figure 3. The variability of these values is typical of that observed for lots of DBM powder analyzed in the past. These studies also included analysis of rat DBM as the standard for allograft performance. The results of the mineralization analysis were encouraging in that none of the lots of DBM powder that scored poorly in the in vitro assay performed well in the preliminary analysis of the simple in vivo assay. That is, they did not exhibit good mineralization. There were however, lots of DBM powder that performed well in the in vitro assay that did not produce significant mineralized bone as judged by the calcium measurements and subsequent histological examinations.

An example of these observations is provided in the histological analysis of implants shown in Figure 4. Hematoxylin and eosin stained decalcified sections of implants of rat DBM powder, a lot of human DBM powder that scored poorly in the in vitro assay (lot 9) and two lots of human DBM powder that scored well in the in vitro assay (lots 14 and 15), are shown. A clear osteoinductive response to the rat and human DBM (lot 15) can be seen in the figure. The rat DBM serves as a standard positive control. The H and E stained sections from the rat DBM and human lot 15 exhibit fatty marrow, robust angiogenesis and cellular bone with mature osteocytes occupying the lacunae. Clearly, a more modest response is seen with lot 14. The lot 9 explant consists almost solely of residual demineralized bone, and exhibits acellular lacunae and very little marrow formation. In order to evaluate further the differences in the osteoinductive properties of the different samples, and to determine the effectiveness of combining the above in vitro and in vivo assays for determining osteogenic potential, the ability of the DBM samples to promote healing of 8 mm diameter critical defects in the calvaria of athymic rats was examined. 25

Figure 5 shows the x-rays of the isolated calvaria of these animals. The calvaria that received rat DBM powder in fibrin sealant served as the positive control and exhibited -80% opacity to x-ray analysis. A disc of fibrin sealant alone did not promote any healing of the defects. Human lot 9 DBM which performed poorly in both the in vitro and the simple ectopic in vivo assay described above also did not provide any healing of the defects. In contrast, lot 14 which performed well in the in vitro, but poorly in the previous in vivo studies, stimulated significant new bone formation in this assay. Lot 15, which performed well in both the simple in vitro and in vivo assays described above, provided nearly complete healing of the calvaria defects in the three animals assayed.

Combination of Lots

The results of the in vitro and in vivo studies described above identified lots 9 and 15 of human DBM powder as bad and good, or ineffective and effective, with respect to their osteoinductive and osteoconductive activities. In order to determine whether the poor performance of lot 9 was likely to be due to the presence of negative effectors of osteoid formation or to the lack of positive factors, the osteoinductive activity of mixtures of different ratios of good and bad lots of DBM powder in the ectopic bone formation assay was evaluated. The results of one such study are shown in Figure 6. The 30 mg implants that were used contained 0, 7.5, 15, 22.5 or 30 mg of a good lot of human DBM with the appropriate amount of a bad lot. It is clear from the figure that the osteoinductive activity of the DBM, or the lack thereof, are both relatively passive properties. That is, there does not appear to be any component of the DBM lacking osteogenic activity (bad DBM) that adversely affects the adjacent DBM material having osteogenic ability (good DBM). Similarly, the presence of good DBM does not overcome the lack of bone promoting 26 activity in the bad DBM lot. The total mineralized bone appears to be the average of the two components.

Claims

27We claim:

1. A protocol for determining relative osteogenic potential of demineralized bone matrix powder, comprising:

(a) measuring a calcium concentration of the demineralized bone matrix powder; and

(b) conducting an in vitro assay predictive of osteogenic potential.

2. The protocol of claim 1 , wherein the in vitro assay comprises: (a) culturing osteoblastic cells to confluence; (b) adding demineralized bone matrix powder;

(c) introducing a differentiation inducing media; and

(d) examining the subculture for new bone formation.

The protocol of claim 2, wherein said osteoblastic cells are 2T3 cells.

4. The protocol of claim 2, wherein examining the subculture comprises staining the subculture with alizarin.

5. The protocol of claim 1 , further comprising conducting an in vivo assay selected from the group consisting of an ectopic implant assay and a critical cranial defect assay.

6. The protocol of claim 5, wherein the in vivo assay is an ectopic implant assay, comprising:

(a) forming a disk comprising demineralized bone matrix powder; 28 (b) implanting the disk into a muscle pouch or flap of a live animal;

(c) removing the disk from the animal after a period of time; and (d) conducting histological examination for signs of new bone formation.

7. The protocol of claim 6, wherein the period of time is 28 days.

8. The protocol of claim 6, wherein the in vivo assay further comprises analyzing the removed disk for calcium concentration.

9. The protocol of claim 6, wherein the disk implanted into the animal further comprises one or more of fibrin and thrombin.

10. The protocol of claim 5, wherein the in vivo assay is the critical defect assay, comprising:

(a) forming a disk comprising demineralized bone matrix powder;

(b) inserting the disk into a critical defect;

(c) removing the disk after a period of time; and

(d) analyzing the disk for new bone formation.

11. The protocol of claim 10, wherein the critical defect is in the calvaria of a rat.

12. The protocol of claim 10, wherein the period of time is 28 days. 29

13. The protocol of claim 10, wherein the disk further comprises one or more of fibrin and thrombin.

14. A method for selecting osteogenically effective lots of demineralized bone matrix powder comprising: (a) collecting representative samples of one or more lots of demineralized bone matrix powder to be analyzed;

(b) selecting samples from step (a) having a measured calcium concentration less than about 0.2%;

(c) combining the respective samples of step (b) with osteoblast cells in vitro; and

(d) selecting those samples from step (c) that stimulate the proliferation and subsequent bone nodule formation by osteoblast cells.

15. The method of claim 14, wherein the demineralized bone matrix powder selected has a maximum particle size of about 180 ╬╝m.

16. The method of claim 14, wherein the demineralized bone matrix powder has a calcium concentration of less than about 0.05% by weight.

17. A method for selecting highly osteogenically effective demineralized bone matrix powder comprising: (a) collecting representative samples from demineralized bone matrix lots to be tested;

(c) combining the respective selected samples from step (b) with osteoblast cells in vitro; 30

(d) selecting those samples from step (c) that stimulate the proliferation and subsequent bone nodule formation by osteoblast cells;

(e) conducting an in vivo ectopic implant assay on the samples selected from step (d); and (f) selecting samples from step (e) corresponding with ectopic implant assay explants having greater than 5% calcium dry weight and exhibiting new bone formation.

18. The method of claim 17, wherein the demineralized bone matrix powder particle size is less than about 180 ╬╝m.

19. The method of claim 17, wherein the demineralized bone matrix powder calcium concentration is less than about 0.05% dry weight.

20. A demineralized bone matrix powder lot selected according to the method of claim 17.

21. A demineralized bone matrix powder pool comprising a mixture of demineralized bone matrix powder lots selected according to the method of claim 17.

22. The demineralized bone matrix powder pool of claim 21 , having a particle size of less than about 180 ╬╝m.

23. The demineralized bone matrix powder pool of claim 21 , having a calcium concentration of less than about 0.05%.

24. Osteogenically effective demineralized bone matrix powder comprising demineralized bone matrix powder having a particle size of no 31 more than about 180 ╬╝m, calcium concentration of no more than about 0.2%, and exhibiting a positive in vitro assay that is predictive of osteogenic potential.

25. The osteogenically effective demineralized bone matrix of claim 24, wherein the in vitro assay is an osteoblastic cell assay.

26. The osteogenically effective demineralized bone matrix powder of claim 24, wherein the demineralized bone matrix further exhibits a positive in vivo assay indicative of osteogenic potential.

27. The osteogenically effective demineralized bone matrix powder of claim 26, wherein the in vivo assay is an ectopic implant assay.

28. The osteogenically effective demineralized bone matrix of claim 26, wherein the in vivo assay is a critical defect assay.

29. Osteogenically effective demineralized bone matrix powder comprising demineralized bone matrix powder having a particle size of no more than about 150 ╬╝m, calcium concentration of no more than about 0.05%, and exhibiting a positive in vitro osteoblastic cell assay.

30. A method of improving osteogenic potential of a demineralized bone matrix sample, comprising processing the sample to achieve a particle size of less than about 180 ╬╝m.

31. The method of claim 30, wherein the particle size is less than about 150 ╬╝m. 32

32. The method of claim 30, wherein the particle size is less than about 125 ╬╝m.

33. A method of improving osteogenic potential of a demineralized bone matrix sample, comprising: (a) selecting a sample having osteogenic potential;

(b) processing the sample of step (a) to achieve a maximum particle size of less than about 180 ╬╝m.

34. The method of claim 33, wherein the step of selecting a sample comprises: (a) selecting a sample having a calcium concentration less than about 0.2%; and

(b) conducting an in vitro assay on the sample of step (a) to determine osteogenic potential.

35. The method of claim 34, further comprising conducting an in vivo assay to determine osteogenic potential demonstrated by new bone formation.

36. The method of claim 34, wherein the calcium concentration is less than about 0.05%.

37. The method of claim 33, wherein the particle size is less than about 150 ╬╝m.

38. The method of claim 33, wherein the particle size is less than about 125 ╬╝m.