OIFE Meeting at Vingsted, Denmark, Sept 25-26, 1998
Summary - Bisphosphonates as a potential treatment for Osteogenesis Imperfecta - Kim Brixen, MD, Ph.D., Aarhus University Hospital, Tage-Hansensgade 2, 8000 Aarhus, Denmark.
Dr. Brixen began his lecture with a review of the pharmacological facts about bisphosphonates. They have their origin in a chemical, called a polyphosphate, which was used to remove mineral deposits from water pipes. A similar substance, termed pyrophosphate, occurs naturally in the body and helps regulate mineral deposition.
Bisphosphonates were developed and came into use in the 1960’s. The first published paper on the effect of bisphosphonates on the formation and dissolution of crystals came out in 1968 authored by Dr. Fleisch . Publications describing the use of bisphosphonates in the treatment of osteoporosis , Paget’s disease , and hypercalcemia  rapidly followed. The use of bisphosphonates in OI was first described in 1987 .
Dr. Brixen next provided a thorough description of how bisphosphonates work on bone. The first thing one must be aware of is that bone is constantly being remodelled – a perpetual process of tearing down and building up (a comprehensive review is given by Dempster ). The cells that are involved in the tearing down process are called osteoclasts and the cells rebuilding bone are called osteoblasts. During bone remodelling, the actions of these two cell types are closely coupled ensuring balance between breakdown and formation of bone. The coupling is mediated by large number of local growth factors and hormones. For example, osteoblasts secrete a number of growth factors that regulate osteoclastic activity and visa versa.
Bisphosphonates affect bone remodelling in two ways - directly on the "tearing down" process and indirectly by decreasing the signal from the osteoblasts to the osteoclast. The first mechanism seems to be the most important. Bisphosphonates are taken up by the osteoclasts during their work (i.e. during their digging in the bone), rendering them inactive by inhibiting some of the enzymes within the cell . This leads to lower bone turnover and to the formation of more shallow "holes" during bone remodelling .
After the age of 40-50 years and in a number of diseases, the "holes" in the bone made by the osteoclasts are only incompletely filled by the osteoblasts. This results in a net bone loss which is more pronounces when bone remodelling is high. Also, with deeper "holes" and higher bone remodelling the risk of inadvertent perforation of individual bone trabeculae increases. This leads to structural damage of the latticework of trabecular bone tissue, making the bone less able to resist stress. Thus, the bisphosphonates strengthen bone by reducing perforations in trabecular bone tissue and by decreasing the bone loss during each turn of the bone remodelling cycle.
When given in large dosages for prolonged periods, the first-generation bisphosphonates also inhibits bone formation. This undesirable effect, however, is not seen with the new types of bisphosphonates since they inhibit the breakdown 100 to 100.000 times more effectively than bone formation .
If taken orally, bisphosphonates are not absorbed through the stomach, but get into the blood via absorption through the intestines, however, only a few percent of the ingested dosage is absorbed. Put directly into the bloodstream, approximately 50% goes to the kidney and is excreted, and 50% goes to the bone.
Generally, bisphosphonates are well tolerated and side effects are seen only in a small minority of the patients treated. Adverse effects observed in human use of bisphosphonates include: bone pain, hypocalcemia, chorioretinitis, fever and flu-like symptoms (occurring only following the first dose and resolving within 1-2 days), and (when given in pill form) ulcers in the oesophagus. With first generation bisphosphonates (etidronate), impaired fracture healing and decreased calcification of bone (osteomalacia) have also been seen with large dosages, however, these effect are not seen with the newer drugs. In animal studies, altered growth has been seen. No negative effects on growth have been reported in human use, but the amount of data on this issue is very limited at the present time. In animal studies, some of the bisphosphonates also causes severe birth defects. No such defects have been described in humans since the drugs are not to be used during pregnancy.
Bisphosphonates cannot address the collagen defect at the root of OI – in that sense, it is not a cure. Bone turnover, however, is high in OI  and the drug may very well have a positive effect on the physical expression of the disease.
A number of studies have been published on the use of bisphosphonates in OI patients, but they have involved small numbers of patients, various duration of treatment, various OI types, and patients of various ages [5; 10-15]. It seems clear that bisphosphonates will increase bone mineral density (BMD), and in some patients will lead to a decrease in bone pain. Bisphosphonates may also possibly reduce the fracture rate (see note added in proof).
There are no data on the effect of bisphosphonates on hypercalciuria, hearing loss, or dentinogenesis imperfecta. Data are also missing concerning the effect of bisphosphonates in adult persons with OI.
Dr. Brixen cautioned against assigning too much importance to the increase in BMD with bisphosphonate treatment and reliance on uncontrolled trials. He gave two examples of medications which were enthusiastically embraced by the medical community, but which in the long run failed.
Fluoride was thought to be an effective treatment for osteoporosis, because bone mineral density (BMD) increases markedly in during treatment. But eventually it became clear that there was in fact no real decrease in the fracture rate . Although the dispute is unsettled, fluoride is no longer in routine use in osteoporosis. This experience underscores the importance of asking the right question (i.e. does treatment protect against fractures?) rather than relying on "surrogate" outcomes (i.e. does treatment increase bone mineral density?).
Also, estrogen was for many years thought to decrease the risk for coronary heart disease. This view was based on a large number of studies comparing women taking estrogen with women not taking estrogen as a matter of choice. However, in a randomized study comparing treatment with placebo drug on the effect of estrogen on coronary heart disease, no positive effect could be demonstrated . The discrepancy between these studies may be explainable if women choosing to take estrogen are more likely to be healthy and women choosing not to take estrogen are more likely to suffer from heart conditions. Indeed, a confusion of variables seems very likely and the experience underscores the necessity of placebo-controlled trials.
A recent study (Muratori 1998 unpublished data) suggests that bisphosphonates not only can increase bone mineral density, but also increase mobility and reduce fractures. However, it is not possible to conclude that these positive results are solely due to bisphosphonates. One must take into consideration that a patients condition can improve over time even in the absence of bisphosphonate treatment. At least two studies [18; 19] document that the fracture rate is not constant, but overall, decreases with increasing age. It is also documented that some improvement in mobility among OI Type III patients occurs with increasing age in the first decade of life .
In Dr. Brixen’s opinion, the only way to establish whether or not bisphosphonates are an effective treatment for OI is to make a placebo-controlled double-blind study with clinical end-points which are quantifiable: growth, fractures, bone mineral density, and motor performance. This will reduce the chance that future patients will unnecessarily take medication which has no real positive effect, and which may have serious negative side effects.
In terms of negative side effects, he stressed that one should not apply the experience of postmenopausal women with osteoporosis, who take bisphosphonates for the last two decades of life, to growing children with many more decades ahead of them or to adult women of childbearing age.
In conclusion, bisphosphonates seem very promising as a treatment for OI, however, large-scaled placebo-controlled trials are clearly needed and should be performed as rapidly as possible.
Note added in proof:
The largest study of bisphosphonate treatment in OI which has been published so far has just appeared in New England Journal of Medicine by Glorieux and co-workers . Thirty children aged 3 to 16 years were all treated for 1 to 5 years. There was a marked decrease in bone turnover and increase in bone mineral density. Also, fracture occurrence decreased and motor performance increased. No adverse effects on growth were seen. Indeed, the only adverse effect seen was the well known short-lived "flu-like" fewer. This study, although, very encouraging was not randomised or placebo-controlled.