Proceedings of the Production of Monoclonal Antibodies Workshop

August 29, 1999 | Bologna, Italy

This document has been adapted with permission from a publication created by the Alternatives Research & Development Foundation (ARDF)

A Comparison of Three Documents

Peter Theran, VMD
Director, Center for Laboratory Animal Welfare & Vice President of the Hospital Division of the Massachusetts Society for the Prevention of Cruelty to Animals
350 South Huntington Avenue
Boston, MA 02130



The Mission of the Sponsoring Organizations

Each one of these three documents represent a resource for readers interested in monoclonal antibody production and specifically for institutional committees charged with the responsibility of monitoring the appropriate use of animals in research, testing and teaching. It is clear, and not surprising, that the authors of these reports have examined the same facts, but have chosen to provide differing points of emphasis in some cases, and, in others, have come to different conclusions. Before discussing details of the reports, I believe it is helpful to understand the missions of the organizations that sponsored these reports.

Technical Report Number 8, Alternatives in Monoclonal Antibody Production, is produced by the Center for Alternatives to Animal Testing (CAAT) at The Johns Hopkins University in Baltimore, Maryland, USA. The vision of this center is "to be a leading force in the development and use of reduction, refinement, and replacement alternatives in research, testing, and education to protect and enhance the health of the public." The mission of this center is "to promote and support research in the development of in vitro and other alternative techniques, serve as a forum to foster discussion among diverse groups leading to creative approaches to facilitate acceptance and implementation of alternatives, provide reliable information on the science, philosophy, and public policy of alternatives to academia, government, industry and the general public, and to educate and train in the application of alternatives." Alternatives are defined as new methods that refine existing tests by minimizing animal distress, reduce animal usage, or replace whole animal tests.

ECVAM Workshop Report 23, Monoclonal Antibody Production is, as the name suggests, the 23rd workshop report published by the European Centre for the Validation of Alternative Methods (ECVAM) at the JRC Environment Institute in Ispra, Italy. The mission of this organization is clear from its name: "validating methods and strategies to reduce or replace the use of live animals in laboratory studies". The main goal associated with this mission "is to promote the scientific and regulatory acceptance of alternative methods which are of importance to the biosciences and which reduce, refine or replace the use of laboratory animals."

The National Research Council (NRC) was organized by the National Academy of Sciences in Washington DC, USA. It is "a private, nonprofit, self-perpetuating society of distinguished scholars engaged in scientific and engineering research, dedicated to the furtherance of science and technology and to their use for the general welfare." This organization is much older than CAAT and ECVAM; the United States Congress under President Abraham Lincoln granted its charter in 1863. The purpose of the NRC is to "associate the broad community of science and technology with the Academy's purposes of furthering knowledge and advising the federal government."

All three of these groups are involved with science. CAAT and ECVAM, however, are both narrowly focused on the development, validation or acceptance of alternatives, while the NRC is dedicated to furthering science and technology and advising the United States government on scientific matters. Understanding this difference in missions will give the reader some appreciation of the potential for different interpretations of the same data related to the production of monoclonal antibodies.

A Summary of the Reports

The Johns Hopkins CAAT Technical Report No. 8: Alternatives in Monoclonal Antibody Production

This report emanates from a workshop entitled Alternatives in Monoclonal Antibody Production, which was sponsored by NIH, The Office for Protection from Research Risks (OPRR), and CAAT and held in Baltimore, MD in September of 1997. This workshop was developed as a way to respond to petitions from the American Anti-Vivisection Society (AAVS) to the NIH and the USDA requesting a ban on in vivo production of monoclonal antibodies.

The first portion of this workshop focused on the basic science and application of MAb technology and the various in vitro methods being used or, at least, available. In addition, the welfare of animals used for in vivo monoclonal antibody production was addressed, as well as, the IACUC's responsibility to inform investigators about available in vitro alternatives. The final portion of the workshop provided a public opportunity for NIH and the USDA to respond to the AAVS petitions. There was general consensus at the end of the workshop by both the faculty and the participants that there were acceptable in vitro methods available for the production of most monoclonal antibodies and the in vivo methods should be replaced by in vitro in most cases.

Pertinent points made by presenters follow:

Dr. Nelson Garnett (OPRR) and Dr. John McArdle (AAVS) presented the issues for the workshop. Dr. Garnett identified the following issues:

Dr. McArdle (AAVS) indicated that in vitro alternatives were practical and available. He summarized the clinical/pathological changes associated with the mouse ascites method; and concluded "simple common sense dictates that it is not possible to humanely produce MAbs using rodent ascites." In his opinion, the statutory and moral responsibilities of IACUC prohibit approval of in vivo methods since acceptable in vitro methods are already available. Dr. McArdle estimates that there were 500,000 to 1,000,000 mice involved with ascites monoclonal antibody production in the US.

Dr. Mark Soloski from Johns Hopkins University School of Medicine began the discussion of the science and uses of monoclonal antibodies. His comments provide a very good background that should be useful to IACUC members unfamiliar with the subject. He concludes with the statement, "it is clear that monoclonal antibodies will continue to have an important and positive impact on scientific endeavors in the near future."

Dr. Neil Lipman from Memorial Sloan-Kettering Cancer Center outlined negatives associated with in vivo methods and related his experience with hollow fiber bioreactors (HFBs) in comparison with the ascites method for the production of monoclonal antibodies. He came to the following conclusions:

Dr. Howard Petrie discussed his experience with Memorial Sloan-Kettering's Monoclonal Antibody Core Facility with Hollow Fiber Bioreactors and Minifermentors. Important points were:

Dr. Jerry Heidel discussed his experience at the College of Veterinary Medicine, Oregon State University with antibody production in gas-permeable (G-P) tissue culture bags using serum-free media. He favored this in vitro technique for several reasons:

Dr. Norman Peterson of the University of Pennsylvania Medical School compared in vivo production with tissue culture in standard plastic flasks, hollow fiber cartridge bioreactors, and gas permeable bags in a study limited to three hybridoma cell-lines. He drew several specific conclusions from the literature and his own experience:

Dr. Peterson summarized his presentation with two conclusions:

Dr. Marcia Etheridge (Johns Hopkins University School Of Medicine), Dr. Lou DeTolla (University of Maryland School Of Medicine) and Dr. DeWayne Walker (3M Corporation) discussed the responsibilities of Institutional Animal Care and Use Committees (IACUCs) and the need to advise investigators.

Dr. Etheridge indicated that the IACUC has the responsibility for safeguarding the welfare of animals in a way that does not unduly interfere with the progress of research. Her IACUC, relying on published data and feedback from monoclonal antibody users, recommends the use of in vitro methods. Investigators requesting approval for in vivo mono-clonal antibody production are asked to justify this method over in vitro methods. Her IACUC provides guidelines for refinement techniques for immunization and for the ascites method of monoclonal antibody production that, in their opinion, make it possible to produce monoclonal antibodies in vivo with minimal pain and discomfort to the animal. She expects that the recommendations of her IACUC will evolve as new data is available. Dr. Etheridge considers the recommendations of her IACUC to be realistic and humane.

The most current recommendations from Dr. DeTolla's IACUC stipulate that all hybridomas must be Mouse Antibody Production (MAP) tested and that all investigators should use in vitro methods. His facility provides MAP testing and tissue culture support for in vitro monoclonal antibody production. If the IACUC approves the ascites methods because in vitro production is not deemed possible, there are recommendations on ascites production, which include a limitation of the amount of pristane and a restriction to one abdominal tap for fluid removal which must be done under anesthesia.

Dr. Walker compared industrial (pharmaceutical, biotechnology, and contract service laboratories) versus academic production of monoclonal antibodies. While acknowledging a lack of consistency, Dr. WalkerÕs impression is as follows:

Dr. Margaret Penno (Johns Hopkins University School of Medicine) discussed the needs and cost effectiveness of core laboratories. She pointed out that her institution added monoclonal antibody production to its core laboratory services in the past year. The number of investigators using this facility (at the time of this publication) was 40 and growing. The experience of this facility allows for an ongoing comparison of in vitro and in vivo techniques. The concerns with both methods are quality, yield, and cost-effectiveness. Dr. Penno concludes that the first two concerns can be satisfied by either method, but that the increased cost of in vitro at her facility remains an issue. She makes the following specific observations:

Dr. Louis Sibal (NIH, USA), Dr. Jerry DePoyster (USDA) and Dr. William Stokes (NIEHS, ICCVAM) presented their thoughts on regulatory issues in the United States.

Dr. Sibal discussed the agency's denial of AAVS' request for a ban on the use of animals in monoclonal antibodies. He indicated that is was a very unusual request that was given serious consideration by NIH, but it was concluded that NIH does not have the authority to prescribe or proscribe methods of research. Dr. Sibal pointed out that he had surveyed many U.S. scientists and found considerable support of in vitro methods, but learned that in 4-5% of cases, in vitro methods failed to grow monoclonal antibodies or grew antibodies with inadequate antigen-binding ability. On the other hand, he felt that NIH was already accomplishing some of AAVSÕ goals in the following ways:

Dr. DePoyster pointed out that Animal Care's position is that the use of alternatives in monoclonal antibody production is already required by law. Investigators are already required to justify the use of animals. He pointed out, however, that the USDA is not authorized to promulgate rules regarding the conduct of the research itself. Dr. DePoyster's conclusions were:

Dr. DePoyster also presented Dr. Ron DeHavenÕs official response to two AAVS petitions: that the USDA modify the definition of "animal" so that rats, mice and birds be included in their inspections under the Animal Welfare Act (AWA), and that the USDA prohibit the use of monoclonal antibodies produced by ascites method. Dr. DeHaven indicated in his letter that including these additional species would add 2324 facilities to the 1735 that the USDA inspects. Given the fact that the Animal Care unit of APHIS had been level funded for 5 years, their resources were inadequate to expand their inspections and therefore, he concluded it would not be prudent to extend AWA coverage to these additional species.

When addressing the banning of the mouse ascites method, Dr. DeHaven's letter expressed the opinion that such a ban would constitute interference with the design or performance of research and that the USDA does not have such legal authority.

Dr. Stokes discussed criteria for the validation and acceptance of new scientific methods.

He drew several conclusions from the conferences discussions:

In the final presentation of this workshop, Dr. Coenraad Hendriksen (RIVM, The Netherlands) presented the European perspective. Dr. Hendriksen indicated that the Veterinary Health Inspectorate (VHI) in The Netherlands issued a code of practice in 1989 limiting the use of the ascites method and encouraging the use of in vitro technology. He classified the ascites method of monoclonal antibody production as "one of the most distressing procedures carried out in laboratory animals". The publication of this code caused a major decrease in the use of the in vivo method. In addition, the Dutch Ministry for Education, Cultural Affairs, and Science has supported the development of in vitro facilities in some of the universities. Dr. Hendricksen pointed out that while embryonated chicken eggs are not an effective culture system for mouse monoclonal hybridomas, the newly developed Phage Display technique might lead to the total replacement of laboratory animals in monoclonal antibody production. The VHI now considers in vivo ascites production to be an obsolete technique and is taking steps to ban its use. It was pointed out that some scientists might attempt to circumvent a ban by purchasing monoclonal antibodies from outside sources.

Conclusions and Recommendations of the Johns Hopkins Workshop
The consensus of the faculty and participants was that the ascites method is painful and causes suffering, and that there are in vitro alternatives for most applications. Use of the ascites method should be the exception that is only allowed by IACUCs after "rigorous and well documented justification." If it is found that the ascites method must be utilized, the 3 R's should be applied to the degree possible.

The group also recommended that IACUC's assume a greater role in promoting the use of alternatives.

The ECVAM Workshop Report 23: Monoclonal Antibody Production

This report was generated by the 23rd in a series of workshops organized by ECVAM and was held in Angera, Italy in November of 1996. The goal of the workshop was to evaluate the present status of in vitro techniques and to compare them with in vivo methods.

There are perspectives in the text of this report that will lead to important points for comparison. In the introduction of the report, the history of monoclonal antibody science was presented. It was concluded that expert information and recommendations, which take animal welfare issues into consideration, must be publicized.

A discussion of current demand for monoclonal antibodies is divided by amount of antibody required. In Europe, ninety percent of the scientists use 0.5 gm of less of antibody. Most of the investigators in this group use the ascites method. Their reluctance to switch to in vitro methods is attributed to a perceived economic advantage and a hesitancy to adopt new methods. Furthermore, it is felt that the ascites users in this group lack awareness of the disadvantages of the ascites method. Scientists who utilize large amounts of antibody are predominately in vitro users.

In a discussion of the in vivo method, advantages listed are high yields of antibodies with less labor. The first disadvantage listed is extreme pain caused by peritonitis, abdominal distension and an invasive malignancy. In addition, the presence of contaminating immunoglobulins reduces the immunoreactivity of ascites produced antibodies to 60-70%. This is compared to 90-95% for in vitro antibodies. Contamination with viruses and bioreactive cytokines and quality variation are also listed as disadvantages. It is concluded that these disadvantages outweigh the advantages.

Several types of in vitro methods are discussed. Static and agitated suspension cultures produce antibody yields in the 100 to 200 mg range. They are easy to utilize in standard tissue culture laboratories, investment costs are low, and operating costs can be reduced by utilizing serum-free media. Conventional stirred tank bioreactors can produce more antibodies but require more staff training. In addition, the antibody concentration is low, usually requiring concentration of the supernatant.

Membrane-based culture systems produce up to 100 mg of antibody, generate more concentrated product (500 µg/ml), are easy to handle and allow different cultures to be run simultaneously. Matrix-based systems, on the other hand, produce 0.1 to 10 gm of antibody with a concentration of 500 µg/ml. In most cases, antibody concentration is needed, and staff training must be more sophisticated.

Finally, in high cell density bioreactors, such as the hollow fiber bioreactor, the same quantity of antibody can be produced in concentrations of 0.5-5 mg/ml. These systems can be run in conventional tissue culture laboratories, and different cell lines can be run simultaneously. Training is required and usually provided by the vendor.

Problems associated with media containing fetal calf serum are identified as follows:

There are, however, serum replacements for bovine plasma, serum substitutes and serum free media available. Hybridomas vary in their ability to adapt to serum free media. There is a process to wean the hybridoma off of serum enriched media, but it must be confirmed that the immunoreactivity of the culture is preserved. The advantages of using an optimized serum-free formulation are that more antibodies are obtained with less cost and subsequent processing is facilitated.

A rather detailed discussion about the glycosylation of monoclonal antibodies concludes that there are "no reasonable arguments based on antibody glycosylation which support the use of in vivo methods." Cost is also discussed. It is pointed out that many conclude it is not a factor, but comparisons are often based on different assumptions. A full cost analysis reveals that in vivo costs are increasing and in vitro costs are decreasing. This trend will favor the adoption of in vitro methods, which at the present time are 1.5-3 times more expensive for the large quantity users. Centers of excellence to provide training for in vitro techniques are recommended.

In a discussion of advanced technologies, it is pointed out that the techniques of phage display libraries and direct cloning into plasmids are novel approaches that may replace the need for animals altogether.

After regulatory developments of several European countries are discussed, a number of workshop conclusions and recommendations are made. It was first concluded that there is a lack of important information. Several European Union countries do not have a system of project review and animal use justification and do not have statistics on the extent of in vivo use. It was suggested that supplier catalogues should indicate how their monoclonal antibodies are produced. It was also concluded that importation of ascites produced antibodies couldnÕt be justified.

The main conclusions of the workshop included the following:

In the appendix of the report, the workshop developed a proposed European guideline for antibody production. Included in this document are recommendations for the refinement of ascites production.

The Committee on Methods of Producing Monoclonal Antibodies, ILAR, NRC Report: Monoclonal Antibody Production

In the preface of this report it is explained that after the NIH refused the initial AAVS petition (April 1997) that the use of animals in monoclonal antibody production be banned, the AAVS submitted a second petition (March 1998). In response to this second petition, which claimed that NIH provided insufficient reason to reject the ban, NIH asked the NRC to conduct a study of antibody production methods.

Introductory comments emphasize the importance and wide spread application of monoclonal antibodies in research and cite the NRC Guide and US government principles and policies that speak to using appropriate numbers and species of animals, using non-animal alternatives where possible, and the application of the 3 R's in research. In the conclusion of this section, it is stated that IACUCs should only approve the ascites method when it is scientifically justified, when the minimization of pain and distress have been considered, and when in vitro methods have been found unsuitable.

A detailed section follows describing the production of cloned hybridoma cell lines. At the end of this discussion, it is pointed out that mouse ascites growth of hybridoma cell lines can avoid the possible loss of a cell line when it is first selected as a result of "inadequate in vitro growth at high dilution". Furthermore, it is pointed out that many investigators have experienced enhanced hardiness and optimal antibody production if a cell line is first grown by the mouse ascites method. In a final comment in this section on the techniques of hybridoma production, it is pointed out that there are published guidelines to assist the investigators who wish to utilize the mouse ascites method.

In a discussion of in vitro antibody production the authors state that these methods should be used when it is reasonable and practical. Batch tissue-culture methods are identified as the simplest method. It is stated that fetal bovine serum, containing about 50 µg/ml of bovine immunoglobulin, is utilized in most tissue culture media. These bovine immunoglobulins are reported to account for a significant amount of the immunoglobulins present in the culture fluid. It was pointed out that fortified serum-free media is now available, and it was estimated that 95-97% of hybridomas will grow in such media. These batch methods produce concentrations less than 20 µg/ml. Roller bottles, spinner flasks, and gas permeable bags increase MAb production and concentration from 0 to 100%. Concentration techniques can allow these methods to meet the needs of the majority of investigators. It was concluded that batch methods have low start-up costs, a start-to-finish time similar to the ascites method, and can produce quantities of antibody similar to the ascites method. Disadvantages listed are that large volumes of media are required, antibody concentration is low, and some antibodies are denatured by concentration or purification activities (by as much as 42% in one screen). Two membrane-based systems were discussed. In these systems, antibodies can be produced in large quantities in high concentration. This discussion ends with the comment that recent workshops and publications indicate that in vitro methods can provide most of the antibodies needed, and that it is the researcher's responsibility to justify the need for the mouse ascites method. Furthermore, the authors stipulate that it is the IACUC's responsibility to evaluate the justification and to approve or disapprove the use of ascites methods.

The authors devote a significant portion of the report to a discussion of the reasons that the mouse ascites method cannot be abandoned, but start the discussion with a recognition that in vitro methods can be used for over 90% of antibody production. Circumstances that, in the authors' opinions, could support the selection of the mouse ascites method over in vitro methods are as follows:

One section lists advantages and disadvantages of in vitroand in vivo methods. Once again, in the prologue, it is pointed out that in vitro methods will meet more than 90% of the needs for monoclonal antibodies. In the discussion, there are 5 advantages and 10 disadvantages listed for in vitro methods and 3 advantages with 4 disadvantages for the in vivo method. The issues are related to economics and quality of the product. Animal welfare issues introduced in this section include the fact that in vivo methods "can produce significant pain and distress"; in vitro methods reduce the use of mice, but can use mice for feeder cells, and the fetal bovine serum often used in tissue culture could have been collected inhumanely.

Commercial production of monoclonal antibodies is discussed. The competition-minded diagnostic industry produces small to medium amounts of antibody and is primarily concerned with cost, turnaround time and regulatory requirements; contaminating antigens are less important. They frequently choose in vivo technology. The therapeutic industry produces on a medium to large scale, is less concerned about cost and turn-around time, but very sensitive to the regulatory cost of procedural change. The concern of allergic responses to contaminating antigens usually leads to the use of serum free in vitro technology. Finally, the biotechnology industry, which develops new therapeutic antibodies on a small to medium scale, is focused on time-to-market and turnaround time. In the authors' opinion, when fully loaded costs of commercial production are weighed, the economic facts usually favor in vivo production. In large-scale production, however, in vitro methods are selected because they decrease animal use, are economically competitive, and have less contaminating proteins (when serum free media is used).

Commercial laboratories often optimize production in both in vitro and in vivo procedures. When cost is the driving consideration, optimized in vivo methods are favored, because optimization of in vitro methods requires "much labor by highly paid, highly trained employees." In discussion of in vivo optimization procedures, it is pointed out that, once the technique is learned, this method is simple, quick, more forgiving, more economical for small and medium scale production, produces higher concentration of antibody, and is easier to scale up to production. Disadvantages of the in vivo method are the use of animals, the possibility of harming an animal with improper technique, and the possibility of contamination with mouse immunoglobulins or pathogens.

Commercial in vitro antibody production advantages and disadvantages are also discussed. The benefits stated are absence of live animal use, antigen free product (when low serum or serum free media are used), and the absence of host immunoglobulins or pathogens. Disadvantages include higher material, labor, and equipment costs; 3-5% of all clones cannot be maintained in vitro; and "the great potential for microbial contamination, poor growth, and mechanical failure of the system or supporting systems requires constant monitoring and attention every day." It is pointed out, however, that one of the most common causes of failure of an in vitro system is poor tissue culture technique. It is the opinion of the authors that, as costs of in vitro methods continue to decrease, they should become the preferred methods for commercial antibody production.

US Food and Drug regulations are discussed. It is pointed out that, of the 13 monoclonal antibodies approved for clinical use, 2 must be produced by the ascites method. Most new applications utilize in vitro methods, which are encouraged by the FDA. Changes in the manufacturing process of require extensive new data collection; it could take 3-8 years to acquire approval for a switch from ascites to in vitro derived antibody.

The final section of this report deals with animal welfare concerns. Throughout this 11-page section there is a tendency to focus on scientifically documented, published data related to the presence or absence of pain with the ascites method of monoclonal antibody production. The ECVAM Workshop report 23 statement that the ascites method "is extremely painful" is characterized by NRC authors as "generally not supported in the literature" and based on reviews and papers that "contains only opinion and no data". A general conclusion is offered that there has been little recorded data to support the presence of pain with the ascites method, but that the possibility of distress (particularly around the removal of ascites fluid) should limit the use of in vivo techniques as a routine method.

It is concluded that pristane injection causes neither pain nor distress. Abdominal distension with ascites was discussed. The work of Jackson and others, where 98% of mice with ascites had rough hair coat and hunched posture, was discussed. These signs were characterized as those that "...can be observed in any sick mouse, regardless of the cause." It was concluded that it "...seems reasonable to assume that mice with large accumulations of ascitic fluid experience discomfort and distress."

The number of taps was discussed. An animal welfare conflict was identified with this subject, because fewer taps means more animals will be required. The report suggests that limiting distress (as opposed to numbers) should be the overriding concern, and that the number of taps should be determined by a combination of animal welfare concerns and the characteristics of the hybridoma.

It is also recommended that IACUCs should evaluate the reaction of animals to IFA before allowing routine use, and should pay special attention to mice after ascites develops. It is pointed out that most IACUCs limit abdominal taps to 2 or 3 (earlier in this chapter, however, it states that it is probably preferable to perform needle taps frequently to avoid the distress of abdominal distention). The authors also recommended that animals be euthanized if signs of distress occur. It is concluded, however, that no recommendations can be made regarding the use of anesthesia or saline replacement because of lack of data, which IACUC's are urged to collect.

The report includes a brief discussion of the animal welfare concerns regarding the collection of fetal bovine serum and feeder cells from mouse peritoneal cavities and embryoblast serum. It is pointed out that both procedures are used to some degree with in vitro methods.

The summary of animal welfare issues concludes, on the one hand, that there is a clear lack of data on this subject as is evident from the lack of best practice guidelines; and, on the other hand, there are issues related to the use of fetal bovine serum (although not connected to the ascites method). Finally readers are cautioned, because of the fetal bovine serum question, not to assume that in vitro methods are inherently more humane than the ascites method.

Final recommendations of the NRC report are as follows:

Finally, the committee gives of examples of scientific criteria that IACUCs can use in the consideration of a protocol using the ascites method for the production of monoclonal antibodies. Also, citing a lack of criteria to define a useful in vitro method, the committee feels that it is the IACUC's responsibility to decide if the ascites method should be used.

A Comparison of the Three Reports

When comparing the reports, one should keep in mind that they come from organizations with differing missions (as stated earlier) and from discussions that took place at different points in time: the ECVAM report in late 1996, the CAAT workshop in late 1997, and the NRC report in late 1998/early 1999.

In Vivo Versus In Vitro Methods

Application:
The three reports all agree that the monoclonal antibody is a very important research tool that has application across many areas of science.

All three publications acknowledge that in vitro techniques can be used for the production of over 90% of monoclonal antibodies; the NRC report concludes that the ascites method should not be banned because of the remaining percentage of hybridomas that need the ascites method. The CAAT workshop concludes that in vitro methods should be the accepted method, and that rare exceptions should have rigorous well-documented justification. ECVAM feels that the ascites method should be banned, but would allow a transition period and rare exceptions.

In this area, the three documents agree on the facts, but the NRC chooses to emphasize the need for the ascites method while the other two focus on the wide spread application of in vitro methods.

Cost:
Experience of authors seems to be somewhat different in this area. The NRC report indicates that the in vitro methods are generally more expensive for small and medium scale production, whereas they are less expensive in high volume. The CAAT report presented several different perspectives. Dr. Lipman demonstrated that HFB have a higher materials and capital cost, but can produce antibodies at a price comparable to the ascites method. Dr. Petrie felt that the higher costs of HFB and minifermentors (2-6 times that of ascites method) was more than offset by the higher purity of product without processing (70- 75%) and the absence of bioactive contaminants. Dr. Heidel found that antibodies producedin G-P bags at his institution were considerably cheaper than ascites produced antibodies. Dr. Peterson considered in vitro methods to be practical, and Dr. Walker pointed out that industry would be mostly guided by the bottom line costs. Eventually he felt that in vitro methods would win out on this basis. ECVAM reported that in vitro methods cost 1.5 to 3 times more than in vivo methods a ban on the cheaper method is still advocated because they consider the mouse ascites method to be a very painful methodology that has an acceptable alternative.

It is apparent that authors of these reports calculate the costs of antibody production with a somewhat different set of assumptions in some cases and, therefore, develop different facts. In addition, there seems to be a difference of opinion regarding purity and concentration of antibodies and how this affects cost. Finally, there is a significant difference of opinion between NRC, on the one hand, and CAAT and ECVAM, on the other, regarding the amount of pain and distress caused by the ascites method. This difference appears to influence authors in their view of the significance of cost differences.

Pain and Distress:
More than in any other aspect of these discussions, opinions regarding the degree of pain and distress in the mouse ascites method are most divergent. In the CAAT and ECVAM reports, the consensus is that the ascites method is painful and causes suffering. The ECVAM report authors label the ascites method as extremely painful. The NRC report, on the other hand, takes a painstakingly scientific approach. They point out that the ECVAM report provides no data to support their claim of pain, and that earlier publications by McGuill and Rowan (1989) and Amyx (1987) contain opinion rather than data. The NRC authors' approach is that very limited conclusions can be made without scientific data. There are inconsistencies in the text and the conclusions in the NRC report on this subject. It is acknowledged that ascites accumulation can be painful in humans and that tapping is not painful, but more of a relief in these cases (it is not mentioned if these human had peritonitis with an inflamed abdominal wall or benign peritoneal effusion with no inflammation. The presence of inflammation could certainly affect the level of pain associated with the penetration of the abdominal wall with a needle.) It is surmised then that frequent needle taps in mice are probably advisable. In the conclusion, however, limited numbers of taps are advocated.

Nowhere in the animal welfare section is there the conclusion that the ascites method causes pain. In fact, it is pointed out that there are no good methods to assess mild to moderate pain and, further, that although data are required to determine if pain is present, it was not clear to the committee what that data should be. On the other hand, in another section, pain and distress are listed as disadvantages of the ascites method and conclusion #3 of the NRC report indicates that every reasonable effort should be made to minimize pain and distress.

In the end, this reader concluded that CAAT and ECVAM authors took the approach that if the clinical and pathological findings were consistent with the presence of pain, they should conclude that the ascites method is painful. The NRC committee, on the other hand, appeared to apply the philosophy that unless it can be scientifically proven that pain is present, they are not willing to draw such a conclusion.

Purity, Concentration and Activity of Antibodies Produced by In Vivo and In Vitro Methods:
All three reports appear to agree that newer in vitro techniques can produce antibodies at a concentration that matches the ascites method, and that purity of product is superior with in vitro techniques if low serum or serum free media can be utilized.

When addressing biologic activity of monoclonal antibodies the CAAT report states that biologic activity depends more on the specific antibody than the method of production.

ECVAM indicates that mouse immunoglobulins reduce ascites-produced monoclonal antibodies' immunoreactivity to 60-70%, while in vitro-produced antibodies have an immunoreactivity of 90-95%. The NRC report concludes that the mouse ascites method offers the highest affinity and biologic activity if purity is not an issue. It points out that in vitro purification steps denature some immunoglobulins that are important in research.

The NRC report also points out that in vitro techniques can sometimes alter glycosylation patterns, which can have a significant effect on biologic activity, and that this is not a problem in the ascites method. The ECVAM report states that an advantage of the in vitro methods is that the pattern of glycosylation can be manipulated to suit the needs of research. The ECVAM authors conclude that glycosylation issues cannot be utilized to support the use of in vivo methodology. The NRC committee challenges this conclusion, however, stating that they can find no published data to support ECVAM's conclusion, and that they consider manipulation of glycosylation processes to be complex and sometimes impossible.

IACUC Responsibilities:
All three reports supported the concept that IACUCs or review committees should have the responsibility of judging the appropriateness of monoclonal antibody production methods. Three speakers in the CAAT workshop addressed this subject directly stating that IACUC's have the responsibility to inform investigators regarding valid in vitro alternatives, to inform themselves about monoclonal production science, and to set standards regarding the refinement of in vivo techniques. It was pointed out that the IACUC in industry might lack sufficient internal information resources to develop or insist on the application of appropriate in vitro alternatives. The recommendations of this workshop concluded that IACUCs should advise investigators about alternatives, require that the alternatives be used unless there is adequate justification for an exception, and, furthermore, such justification cannot be based solely on cost or convenience.

The ECVAM report took a slightly different approach. Acknowledging that some countries have no review system, it is recommended that they develop such a system. In those countries that have a review system, it is recommended that the system should require justification of in vivo methods. ECVAM recommends that review committees seek advice from scientists who are familiar with in vitro methods. But ECVAM also recommends that a ban of in vivo methods be put in place, and, further, that an inspection system be established to ensure that in vivo methods are only performed with proper justification.

The NRC committee states that in vitro methods should be used when scientifically appropriate, provided the cost of the method is consistent with the realities of funding. The report states that it is the IACUC's responsibility to approve in vivo use. The committee urges IACUC to focus more on preventing distress than reducing numbers if these two principles are in conflict. The IACUC system in the US is urged by this committee to collect data on animal welfare aspects of monoclonal antibody production and to share such data with others. In addition, it is pointed out that it is the IACUC's responsibility to ensure proper training of research staff using the in vivo method.

Centers of Excellence:
The CAAT report strongly endorses the concept of institutional and NIH supported core facilities for in vitro monoclonal antibody production. ECVAM recommends that centers of excellence should be developed to provide advice and assistance to laboratories converting to in vitro methods. The NRC committee points out that in vitro production facilities within institutions could help prevent cost from being a barrier to the acceptance of in vitro methods.

Future Developments:
The ECVAM report discusses newer recombinant DNA-based technologies such as phage display libraries and direct cloning of plasmids. It is speculated that the need for living animals could be eliminated by this science. Dr. Hendriksen mentions this in the CAAT workshop, but NRC discussions did not address this point.

Fetal Calf Serum:
CAAT workshop participant, Dr. John McArdle, points out that there are serious animal welfare concerns related to the collection of fetal calf serum and that alternatives should be sought. The ECVAM authors agree that there are a number of problems with fetal calf serum, including animal welfare concerns, protein contamination, cost, quality variations, and the possibility of contaminating infectious agents. It is pointed out that there are options that avoid the use of this serum. The NRC committee indicates that in vitro methods generally require the use of fetal calf serum. In the conclusion of the animal welfare section, the committee advises readers that it cannot be assumed that in vitro methods are more humane because of the animal welfare concerns about fetal calf serum collections.

Endpoints:
In those circumstances where the ascites method is deemed necessary, this reader is struck by the fact that learned scientists in the United States and Europe are trying to make decisions regarding appropriate endpoints with crude clinical observations. In a time when imaging techniques are available and blood/serum measurements can be readily performed on microliter quantities of blood, scientists are attempting to draw important conclusions regarding animal suffering and experimental design based on an interpretation of clinical signs. More in-depth clinical assessments could hopefully lead to endpoints that were scientifically based and would effectively limit or even prevent suffering. The NRC report suggested that the IACUC needed to collect data regarding pain and distress. This reader's perspective is that investigators share this responsibility; they need to apply their scientific skills to help develop the clinical data that could lead to more precise health evaluation and more humane endpoints.

Relief of Pain:
All three reports, to one degree or another, agree that pain is involved with aspects of the mouse ascites method. Furthermore, the reports uniformly acknowledge that there is a need for the ascites method in a small percentage of circumstances. This reader acknowledges that it is not really feasible to eliminate all pain and distress with a rapidly growing abdominal malignancy, peritonitis, and ascites; but it is still striking that there is no discussion of the use of analgesics or anesthetics in an attempt to at least reduce the inevitable pain and distress. Dr. Lou DeTolla (CAAT report) does recommend that only one abdominal tap for fluid collection be made, and that euthanasia be performed prior to this tap. In this way, his protocol addresses the need to avoid pain during fluid collection. In general, however, the lack of discussion of analgesia by any of the 36 authors of the three reports exposes a need for greater focus on the important area of relief of pain and distress in laboratory animals.

The Banning of In Vivo Methods:
The CAAT workshop served as a public forum for Dr. John McArdle to support the AAVS petition for a ban on the ascites method of antibody production and for NIH and the USDA to announce their rejection of this petition. In the conclusions and recommendations of this workshop, banning of the method was not mentioned. However, the workshop did conclude that in vitro methods should be the norm, and that the ascites method should be the rare exception. The ECVAM workshop recommended a ban after a transition period of up to two years. The NRC committee, which was specifically commissioned to answer the question as to whether there was a scientific need for the mouse ascites method. Their conclusion was that the ascites method should not be banned; it acknowledged that in vitro methodology can produce monoclonal antibodies from over 90% of hybridomas, but there were still circumstances that justified the use of the ascites method.

To this reader, the difference of opinion regarding the banning of the ascites method has more to do with perspective and posture than substance. The NRC says that in vitro methods should be the standard except in the small percentage of situations where science or economic factors prohibit their use. The CAAT workshop recommends that in vitro methods should be the accepted method and that exceptions should be rigorously justified. They go further and state that cost and/or convenience should not be the sole basis for an exception. The ECVAM report recommends a ban that allows for some exceptions. One has to be struck by the fact that all three reports could have recommended a ban with exceptions without changing their conclusions. On the other hand, it would seem that the three reports would have been essentially unchanged if they had advocated that there be no ban, but that in vitro methods should be the standard method with rare, scientifically justified exceptions.

It is not entirely clear whether the positive animal welfare impact of the recommendations of the NRC and CAAT groups, reinforced both by the influence of OPRR and FDA recommending the use of alternative approaches to monoclonal antibody production and by a widespread, mature IACUC review system, will be lesser, greater, or the same as the impact of an ECVAM ban that makes allowances for necessary exceptions, but does not have the benefit of a uniform review system throughout all of Europe. This reader feels that it was unfortunate that the NRC chose not to recommend a ban that would have allowed for exceptions. Changing from a well-entrenched research method does not happen without some level of discomfort for the practicing scientist; this factor often delays any action. A ban would probably have provided greater impetus for scientists to make the change from in vivo technology where there is an appropriate in vitro method available. Furthermore, a ban (with exceptions) would have provided clearer direction for IACUCs on this issue, and a more prompt reduction in pain and suffering of animals would probably have resulted.

References

  1. On November 17, 1997, Dr. Garnett and Dr. Ellis issued a "Dear Colleague letter" in which they stated that there is evidence that the mouse ascites method of monoclonal antibody production causes discomfort, distress, or pain.