Technical Updates

USAID/W Technical Assessment of Recent Evidence on Depo-Provera and Bone Density - November 2004

Summary

The Pfizer company has been conducting 2 long-term studies to assess an ongoing concern about the effects of the injectable contraceptive Depo-Provera (DMPA) on bone density. The company recently released new data from the 2 studies in its drug labeling and strengthened its warning about the effect. While USAID/W takes the new findings seriously, we believe Pfizer's position is overly cautious. We await the scientific assessment by WHO's Medical Eligibility Criteria (MEC) before recommending any major changes.

Introduction

For a number of years the potential effect of DMPA on bone density has been a concern. Minor reductions in bone density during women's reproductive age are problematic if they carry forward into the later years and decades of the menopause when osteoporosis becomes an important health issue. This effect from DMPA appears to be related to its suppression of endogenous ovarian estrogen production. New evidence from Pfizer has heightened attention to this issue.

New evidence and labeling from Pfizer

A new long-term study, of women aged 25-35 using DMPA for about 4½ years (and then followed for almost 2 years after discontinuation) found decreases in bone density of about 5-6% at various bone sites by the end of DMPA use. Loss in density was most rapid the first year or two, but continued to some extent for the entire 4½ years. When DMPA was discontinued, bone density fairly rapidly increased toward original baseline levels, but were still 1-5% below the baseline values almost 2 years after discontinuation. It is possible that had the women been followed longer, their bone density might have returned to baseline. No women actually developed osteoporosis.

Pfizer also conducted a similar study on adolescents with an average age of 15-16 at enrollment. At about 4½ years of DMPA use bone density had declined by about 4-7% at various bone sites. However, at about 3 years after discontinuation, bone density was actually a small amount above the baseline bone density from before DMPA initiation. It is difficult to interpret the significance of these declines and subsequent increases, because bone density is expected to increase gradually in the years following puberty. (Pfizer had a control group of adolescents, but their characteristics were rather different from the DMPA users and comparisons are difficult.) The adolescent study is ongoing.

As a result of these findings Pfizer with the agreement of the USFDA has added the following "black box" at the front of their physician's label:

"Women who use DEPO-PROVERA Contraceptive Injection may lose significant bone mineral density. Bone loss is greater with increasing duration and may not be completely reversible.

It is unknown if use of DEPO-PROVERA Contraceptive Injection during adolescence or early adulthood, a critical period of bone accretion, will reduce peak bone mass and increase the risk of osteoporotic fracture in later life.

DEPO-PROVERA Contraceptive Injection should be used as a long-term birth control method (e.g. longer than 2 years) only if other birth control methods are inadequate. (See WARNINGS.)"

USAID/W assessment of the evidence

These findings are of concern with respect to the older women because the decreases in bone density had not entirely reversed by 2 years. The findings in adolescents are of concern because they may have impact on ultimate bone formation. Even though it is fair to say that the period leading up to and including puberty is most critical, the few years after puberty may also be important. It will be crucial to get a better assessment of the expected bone development over time in such adolescents for comparison with the actual increase seen after discontinuation among the DMPA users.

On the other hand a number of considerations mitigate the Pfizer findings:

*This concern about DMPA and bone loss is not new. The previous label contained a warning about bone effects. Indeed the Pfizer findings in older women are consistent with previous evidence.

*While important, concerns are theoretical and relate to potential health problems in menopause some decades away, when bone thinning becomes problematic. No one actually developed osteoporosis in the studies.

*Notably, other studies on bone density in postmenopausal women have been fairly reassuring with respect to previous DMPA use.

*The actual number of women followed to the end date cut-off was quite small.

*It is important to consider the trade-offs with unwanted pregnancy and specifically lactation in this context. Lactation is associated with a roughly 4-5% decrease in bone density, though the evidence indicates it is reversible once lactation ends.

*Bone density is but one (albeit important) measure of bone functionality and health. Bone is actually a very complicated and dynamic organ. Additional research is needed to address issues of potential effects on configuration and overall mineral content.

Significance for other progestin-only methods

The concern about loss of bone density is not, in general, directed toward certain other progestin-only methods such as Norplant or progestin-only pills. These, unlike DMPA, allow for a considerable amount of ovarian activity and production of natural estrogen. The effects are not clear (as far as we are aware) of the 2-month injectable contraceptive norethindrone enanthate (Noristerat, Norigest) or the single rod implant, Implanon, which do inhibit ovulation.

Prevalence of long-term use of DMPA

One-year first-segment DMPA continuation rates are on the order of 50% in the developing world. We also know that women often switch back and forth from one method to another. Thus most women who use DMPA do not use it continuously for 2 or more years. However, a substantial number of women are highly satisfied with DMPA, and many use it for prolonged periods of time.

USAID/W technical position

USAID/W takes these concerns about bone density seriously. But in view of the discussion above and the full context of needs of women, and balance of benefits and risks of all the methods, believes the position on longer-term use (e.g. 2 years) taken by Pfizer is overly cautious. Bear in mind that the medico-legal climate in the U.S. motivates pharmaceutical companies to make labeling as cautionary as possible. Moreover, it is not clear from the data that 3 years of continuous use is of appreciably more concern than 2. (Indeed the wording of the new label itself is accordingly somewhat indistinct in its use of the phrase "e.g. 2 years" to describe long-term.)

While most women will not use DMPA for even 2 years continuously, we are concerned about the very women who are satisfied enough with the method to use it on a longer-term basis. Family planning has many benefits. Contraceptive choice and successful use are a result of a host of individual preference factors, and we are especially reluctant to interfere with a client's ongoing successful use of a method, since it could well lead to unintended pregnancy and its consequences.

We await the final results of the ongoing adolescent study. In the meantime it is our understanding that WHO has been conducting a systematic review of all the evidence and will be taking a position on this issue as part of its Medical Eligibility Criteria (MEC) process, in the relatively near future. We intend to follow their proceedings and adjust our recommendation accordingly.

For additional comments/questions contact G/PRH Jim Shelton - <mailto:JShelton@USAID.gov> or G/PRH Jeff Spieler - <mailto:Jspieler@USAID.gov>.

See also previous USAID Guidance on Depo Study - September 2004.