Pre-antibiotic Treatments Spur Modern Fungal Infection Research

A look back at the pre-antibiotic era may uncover ways to dealwith immunosuppression and drug resistance in today's patients,Dr. Matthew Scharff said at a recent talk sponsored by the IrvingtonInstitute.

"If you have a historical bent, you think about the timewhere there were no antibiotics," said Dr. Scharff, directorof the Cancer Center, Albert Einstein College of Medicine, NewYork. "Virtually the only intervention that physicians hadwhen someone had an infection was to administer antibodies, andthe treatment worked."

Spurred by the example of early passive antibody treatment, researchersat Einstein under Dr. Arturo Casadevall are trying to make monoclonalantibodies that will protect patients from infection with Cryptococcusneoformans (see box), Dr. Scharff said. The antibodies seemto be protective in mouse studies, and he hopes that clinicaltrials will be possible in the future.

The research at Einstein is a new chapter in an old story thatbegan in the early 1900s, Dr. Scharff said, when investigatorsimmunized horses to make antibodies to pneumococcal and meningococcalorganisms.

With pneumococcus, the serum from the immunized horse was mixedwith the organism and injected into a mouse. If the mouse survived,the antiserum was deemed protective. With meningococcus, the antiserumwas mixed with the bacteria and injected into the cerebrospinalfluid of a monkey via a lumbar puncture, sending it to the brain.If the monkey did not develop meningitis, the antiserum was thoughtto be protective and could be used on human patients.

Data from the 1920s showed that such treatments were successfulin both diseases. Among pneumococcal pneumonia patients studied,those who had not received antibodies had a 30% mortality, comparedwith 18% of those treated with antibodies. The survival rate waseven higher among those who had earlier interventions, Dr. Scharffsaid.

Meningococcal meningitis patients had a 70% to 80% mortality ifnot treated with antibodies. The rate fell to 30% among the treated."There was no doubt that the antiserum saved lives,"he said.

'Sigh of Relief'

Physicians continued to use passive antibody treatment into the1930s and 1940s until antibiotics were discovered. Treatment withantibodies was abandoned because of toxicities associated withhorse serum and because of difficulties in producing commercialquantities of the treatment, he said.

Dr. Scharff explained that since the horse antibodies and theproteins in the horse serum are foreign substances, the treatmentproduced allergic responses when injected into humans. "Youcould inject it once, but if you came back a second time, youwould make the person very sick." Thus, he said, "everyonebreathed a sigh of relief when antibiotics appeared."

But these problems in the use of antibody therapy were not insurmountable.The solution to the problem of producing large quantities "inprinciple, was obvious," Dr. Scharff said. "If you couldcapture one of these antibody cells that was making a useful antibodyand propagate it, either in an animal or in a tissue culture tube,you could make a reproducible reagent."

In 1975 in Cambridge, England, researchers came up with the technologyfor making monoclonal antibodies, he said, but "you stillhad an antibody that was foreign to man, and you could still getan allergic reaction."

One solution suggested recently is the transgenic mouse, whichmakes human antibodies instead of mouse antibodies. In addition,gene splicing can be used to turn a mouse antibody into a nearlyhuman antibody that will not cause an allergic reaction and canbe used as a passive therapeutic agent, Dr. Scharff said.

Antibody Therapy Appears Promising for Cryptococcal Infections

Cryptococcal meningitis kills 8% to 10% of AIDS patients, Dr.Scharff said, highlighting the need for new therapies for thisfungal infection.

Yet antibody research in this area is hampered by several problems,including the view that antibodies do not play a role in fungalinfections, the expense and difficulty involved in producing antibodiesin the amounts needed for a clinical trial, and the fact thatpotential patients are likely to die of other complications.

At Einstein, researchers have shown that a lethal infection ofcryptococcus kills all of the control mice in 10 or 12 days. Butif the antibody against the cryptococcus is given, many of themice are alive at 20 and 30 days. Even at 250 days, half of theanimals are still alive, and two out of three have cleared theorganism, he said.

Dr. Scharff cited historical precedents for using antibodies inpatients with fungal infections. In 1925, a patient with a cryptococcalinfection received rabbit antiserum for the first time, and inthe early 1960s, Dr. Morris Gordon of the Albany Health Centerused rabbit antiserum to treat patients with cryptococcal meningitiswho had not responded to amphotericin.