BoneKEy-Osteovision | Commentary
Does clodronate treatment affect non-boney metastases in breast cancer?
DOI:10.1138/2002058
Commentary on: Powles T, Paterson S, Kanis JA, McCloskey E, Ashley S, Tidy A, Rosenqvist K, Smith I, Ottestad L, Legault S, Pajunen M, Nevantaus A, Mannisto E, Suovuori A, Atula S, Nevalainen J, Pylkkanen L. Randomized, placebo-controlled trial of clodronate in patients with primary operable breast cancer. J Clin Oncol. 2002 Aug 1;20(15):3219-24.
One of the hotly debated and still unresolved issues in the bone metastasis field is whether bisphosphonates have a beneficial effect on cancer cells in soft tissue and other organ structures, in addition to their well-known and well documented effect of reducing bone lesions. The long-anticipated report by Powles et al. () that was published recently in the Journal of Clinical Oncology examined the potential role of clodronate as an adjuvant therapy for primary breast cancer, and sheds some light on the issue, but still does not answer this question definitively. However, it is the largest clinical study published so far on this topic and does provide useful information that ensures that this debate will continue. The study shows that there are fewer bone metastases in patients treated with clodronate (although the treatment must be maintained), that clodronate treatment provides a likely survival advantage, and most importantly, that there is no deleterious effect of clodronate on soft tissue metastases.
The major debate is focused around a provocative earlier study of Diel et al. (), which showed that, in women with metastatic breast cancer, clodronate not only reduced tumor burden and skeletal-related events associated with bone metastases, it also reduced soft tissue tumor burden and suggested an increase in survival. This result was totally unexpected. There was, by that time, a growing body of preclinical data that suggested that bisphosphonates could reduce tumor burden at bone sites (), but no data had been reported to suggest that there were likely to be effects on tumor cells at non-bone sites. In fact, the consensus in the bone field, was that bisphosphonates were strictly bone-active drugs, lacking effects other than those related specifically to osteoclasts.
The study of Diel et al., when published, received a lot of attention-some of it, of course, directed to the agent clodronate itself. One question was whether this was an effect peculiar to clodronate, or could it be a class effect of the bisphosphonate family? This issue was clouded by two other emerging notions of the time, namely that 1) clodronate had a different mechanism of action on osteoclasts than the nitrogen-containing bisphosphonates, and 2) that bisphosphonates as a class may have direct cytotoxic effects on tumor cells. Several groups have shown that nitrogen-containing bisphosphonates inhibit osteoclast function by inhibition of the enzyme farnesyl diphosphate synthase, whereas clodronate and etidronate, the earlier bisphosphonates, do not share this mechanism (). To make things even more awkward for U.S.-based oncologists, clodronate was not, and is not, available in the U.S. For several reasons, it has never been approved or marketed here. Following publication of the Diel report, many patients clamored to know where they could receive this drug, but were denied access to it in the U.S. Thus, U.S.-based oncologists are often asked to extrapolate the findings with clodronate to available bisphosphonates such as pamidronate or zoledronate, but there is no similar clinical data available with which to draw a comparison.
To complicate matters further, Saarto et al. () reported that clodronate in fact caused an increase, not a decrease, in tumor burden, and this effect occurred not just in soft tissue sites, but also in bone. At the time the Saarto paper was under review and then subsequently published, the abstract describing the study of Powles et al. was presented at ASCO (), and the full manuscript was then finally published this year.
How could these study results be so discrepant? There are two likely factors: first, patient selection, and second, the power of the study. Diel et al. selected patients for their study using an eligibility criterion that is in common practice in Germany, namely bone marrow aspiration for detection of cancer cells. This is not a practice used in the U.S., Canada or England; therefore, the multi-center, multi-national studies by Powles and his collaborators could well represent different patients in a different phase of the disease. The patients of the Powles study were begun on study within six months of primary treatment, and they were almost certainly at a different stage of the disease. The patients in the Saarto study were not placebo controlled, and as pointed out by Powles et al., those treated with clodronate may have had a poorer prognosis because of hormone receptor status. However, the results of Diel et al. were so striking and so unexpected, this alone seems an unsatisfactory explanation. The more likely explanation for the discordance in results in my view is the element of chance. The reality is that probably none of these studies are sufficiently powered, although this is a tall task for such studies. The Powles study contained three times the number of patients as the others, and was multi-centered. For a definitive answer to the use and potential benefits of adjuvant therapy in this setting, it is possible that a study three times as large again may be required, and the treatment period may need to be longer than two years.
One persistent thought, however, is that bisphosphonates still may indeed have a direct effect on tumor cells. There is a steadily accumulating body of in vitro evidence that supports this notion (). The counter-argument is that the studies supporting such an effect have been performed in vitro, and large concentrations of bisphosphonates are required. However, no one knows what the concentration of bisphosphonate, when administered in these studies, is present locally in the bone microenvironment. It is conceivable that bisphosphonates could be present in the tumor microenvironment in bone in sufficient concentrations to cause the direct types of anti-tumor effects that have been observed in vitro.
Where then are we left in this debate? Bisphosphonates have been shown to cause a decrease in tumor burden in bone metastatic sites in both preclinical and most clinical studies. There is little argument on that point. Whether they have additional effects in vivo at non-bone sites remains possible, but has still not been satisfactorily proven. There does not seem to be a special advantage for the nitrogen-containing bisphosphonates over clodronate in this situation and, in fact, the most extensive experience so far is with clodronate. Finally, based on data from the study of Powles et al. (), it appears likely that bisphosphonates can be used safely in breast cancer in the adjuvant setting, although their benefits as far as long-term effects on visceral metastases are concerned remain unclear.
This work is licensed under a
Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.