I prefer to give a holiday of 3 months for bisphosphonates after every 1 year of therapy. I have observed and treated many bisphosphonate induced atypical femoral fractures (BMJ case reports. 2013 Nov 28;2013(nov28_1). doi:pii: bcr2013201931. 10.1136/bcr-2013-201931).

Osteoporosis treatment with Bisphosphonates is usually has a "break" after two years of continuous treatment for about six months and it is started again, but some say for one more years others for two more years. The reason for the treatment to be stopped is the so called accumulation of the sclerotic brittle bone which leads to "pathological" fractures of the long bones (mainly in the sub-trochanderic region).

I use bisphosphonate for osteoportic patients and I follow up them by BMD if there is response I will continue till 4 years then I give holiday for 1year because there is reports that bisphonate more than 5years increase the incidence of femoral insufficiency fracture .

I have encountered many cases of BP-induced atypical femoral fractures, requiring surgical treatment also (Vaishya R, Vaish A, Nadeem A. Bisphosphonate induced atypical sub trochanteric femoral fracture. BMJ case reports. 2013 Nov 28;2013(nov28_1). doi:pii: bcr2013201931. 10.1136/bcr-2013-201931). Hence,  I give three months holiday period for BP therapy after each year of BP therapy.

I agree with Raju and almost follow the same regime although is 6 months every 2 years, but I am in doubt for how long I would continue the treatment with bisphosphonates and if I have to change it by administer another type of anti-osteoporotic medication.

There is no simple rule in stopping bisphosphonate therapy.

The recommendations of some international scientific societies are to evaluate the fracture risk of every single patient. If the fracture risk is low (e.g. < 20% with frax or similar algorithms) you can stop the treatment every 3-5 years. If the risk is elevated you can't stop the treatment. Maybe you can shift to teriparatide or strontium ranelate if possible.

In any case stops of less than 1 year are not useful, because the bisphosphonates persist in the bone for many months. In 3 months the bone metabolism remains depressed and there is no "holliday effect".

A little different is the case of an other antiriabsorptive drug, denosumab, that does not persist in the bone and has an effect of depression for only 6 months.

There are no strictly determined rules for duration of biphosphonates.You see the great deviation of the answers above ! In the past I kept to a 3 years-regime followed by 6 months pause. Currently, the problem with the BP-induced fractures causes some doubts about duration and efficacy as a whole. In my opinion there is eclectics about the prescription of biphosphonates. New medications are recommended, but unfortunately, I think this chaos may be "firma-oriented".

I think that different aspects have to be taken into account when you are considering to suspend an antiresorptive therapy with bisphosphonates, mainly the risk of a new fragility fracture, duration of theraphy and the presence of risk factors for the development of an adverse effect.

According to the Spanish osteoporosis guidelines, the following recommendations can be suggested:

1. Patients treated with bisphosphonates should be assessed after 3-5 years of treatment.

2. Treatment should be continued (with the same drug or another) in the following circumstances:

- absence of fractures but BMD at femoral neck lower than -2.5 Tscore.

- development of fractures in the 3-5 years prior to the assessment.

- presence of fractures prior to this period and BMD at femoral neck lower than -2 Tscore (it is likely that BMD values at lumbar spine can be used in a similar way).

3. If none of these circumstances are present, bisphosphonates can be removed.

4. If treatment is maintained, it should be re-evaluated periodically with the possibility of removing it at any time. It is not known how often re-evaluation should be done or if there is a maximum time of treatment. A limit of 10 years is often considered because of the absence of reports on the efficacy of bisphosphonates and the high risk of an atypical femoral fracture beyond this period of time.

5. If you decide to remove the antiresorptive treatment, patients should not be left unprotected if they are at high risk , and they should take another drug instead (e.g. teriparatide or strontium ranelate).

6. When treatment with bisphosphonates is removed, the suspension must be temporary ("therapeutic holiday"). We don't know the appropiate duration of suspension or when bisphosphonates should be reintroduced. A period from 1 to 3 years is usually considered (a year for Risedronate, two for Alendronate and three for Zoledronate?).

BOA guidlines put patients who are high risk (multiple fragility fractures for example) blindly under Aledronate treatment and refer the follow up to the Rheumatologists in case that BMD is not improving. 

Hallo,

my vast experience in treating osteoporotic patients shows, that treatment episodes of less than 3 years followed by "drug holidays" results in relatively rapid loss of formerly gained BMD as measured by DXA, so You should treat for 3 years at least.

Data about "atypical femoral fractures" (AF) seem to show, that with increasing duration of therapy the number of patients with AF increase, but data from different regions of the world show some discrepancies. However, a "mode of action" for AF due to BIS is still not clear. Besides local stress due to accented varisation of the femur probably genetic variances in the TNSALP-genes (not formerly diagnosed hypophosphatasia) may influence the risk for AF. Clearly, there is no "regional over-suppression" of bone turnover or osteoclasts. And also osteoblasts are present - otherwise one would not see lokal spots in bone scintigraphy.

There are no randomised studies on this fact, due to only very few patients with AF (or ONJ) in controlled trials. Health economic studies show, that at least in the first 5 years You prevent about 50 "normal" osteoporotic femoral fractures by getting one additional AF. And old heroes in osteology also report, that "AF" existed long before the BIS era, for instance in patients treated with fluorides....

Some lang lasting studies with BIS showed benefit of therapy for more than 5 years only in "high risk for fracture" patients. So it might be okay, to use FRAX or other tools in patients after 3-5 years to decide about the remainig fracture risk and further therapy.

The BIS persist many months (risedronate), several years (alendronate) or even 30-100 half-time (zoledronic acid) in bone, so any holiday less than 1-2 years is rubbish in my opinion. 

We have NO data about a benefit of change of therapy to teriparatide or strontium ranelate, while this seems a reasonable idea.

I by myself deal pragmatic, not really evidence based (and know, that change of BMD reflexts only small parts of fracture reducing effect of BIS): treat for 3-4-5 years, measure BMD every 1.5-2 years until reaching a "plateau", than stop BIS and follow every 1.5-2 years BMD (one could also use bone turnover markers instead) - restart therapy if relevant BMD decline (5% or more) or BTM increase (>30-50%) is seen. Maybe it is the correct idea, probably future will show. I was one of the first osteologists in Germany regularly pausing BIS after 3-5 years in 2002-2003, some years before discussions abount ONJ, AF and drug holidays :)

Dr Burkhard .thank you for this valuable information I benefit alot .

Dr Burkhard .thank you for this valuable information I benefit alot .

Dr Burkhard .thank you for this valuable information I benefit alot .

Dr Burkhard's entry is very detailed and clear. Excellent.

Dr. Muche's posting is a pretty good one. It says a lot. Especially, the last paragraph is very informative.

Dear colleagues - Thank you all for your contribution so far. Although , as Panayot says , the issue is  - eclectic - , it  is both interesting and instructive to learn from you. Please continue posting .

Dear Dr. Muche

Your answer is very informative and detailed one but need your expert opinion regarding following article.

Alendronate is well tolerated and effective for at least 10 years. In an extension study of 247 postmenopausal women with osteoporosis, alendronate (5 or 10 mg/day) resulted in continued increases in bone mineral density over a 10-year period.

Bone HG, Hosking D, Devogelaer JP, et al. Ten years' experience with alendronate for osteoporosis in postmenopausal women. N Engl J Med 2004; 350:1189.

Hallo Muhammad,

yes, there ist some evidence about further (relatively small) increases of BMD after 5 yrs of BIS. But, those data come from small subgroups of the initially included patients and the information about fractures is - if ever - very limited. Effects differ a bit between studies with alendronate and zoledronic acid, once you see significant effects of prolonged therapy longer than 5 (or 3 resp.) years only in CLINICAL vertebral fractures,the other time only in MORPHOMETRIC vertebral fractures.

The effects of denosumab and odanacatib (if ever available for use in our patients) differ from BIS with continuing increases of BMD (no plateauing) also beyond 5 years,but no one can say today (and probably will never have an robust answer) if it relates to better anti-fracture efficiacy.

Best regards, Burkhard

Bone strength does not depend directly on bone mass as assessed by DEXA. A bone's resistance to fracture results from a combinaiton of only two factors, namely, bone tissue "quality" (intrinsic stiffness and toughness) and bone tissue distribution (bone geometry, bone macro-architecture, bone design). What we know as "osteoporoses" are bone mass losses (with normal "quality" of the remaining tissue) that are severe enough as to impair bone design so much as to induce bone fragility (whatever the BMD value!, sorry about WHO's previous conceptions). That is to say, "osteoporoses" are just diseases of bone design derived from bone mass losses.

Accordingly, if you wish to know (at least partially) what are you doing to different parts of the skeleton of a patient when you give him/her bisphosphonates, then you should assess somehow bone material stiffnes and toughness and bone design regarding some given (directional) way of bone deformation as related to the predictable method of production of the fracture you wish to prevent in the selected bone. If you improve only DEXA-assessed bone mass (BMC, BMD) somewhere in the skeleton, you can by no means assure that you have also improved bone strength here or there, and vice-versa.

Furthermore, when you give a bisphosphonate, you are doing at the least two different things, namely, 1. blocking osteoclastic resorption (and hence bone remodeling, which is a healthy mechanism within the scheme of bone metabolism and bone design control) and 2, contaminating apatite crystals, thus modifying their cristallinity, which leads to enhanced bone tissue stiffness and reduced bone tissue toughness. That is to say, you are disorganizing the mechanically (i.e. vectorially)-oriented bone design control, and also impairing the two exclusive bone tissue properties that are relevant to bone strength. And this may be a pretty long, long effect.

On the good side, at the same time, you may be getting some increase in bone mass (DEXA) by perhaps a few units percent. This is too little when compared with the huge percent bone losses (more than 20%) involved in "osteoporosis" development, i.e. in determining an "osteopenic-derived bone fragility" (what is "osteoporosis" as per the NIH's - not WHO's - conception).

In addition, perhaps you are also desirably improving the osteocyte ability to govern the spatial distribution of bone tissue through modeling-remoding (by preventing osteocyte apoptosis), yet this has been proven currently only in small rodents.

Hence, it is impossible for anybody to know what kind of whole effects (favorable-disfavorable) is he-she inducing in the skeleton when givingh a bisphosphonate (not to tell if you are doing it without having measured bone remodeling previously!). Therefore, much less than that could you know (or even guess) about during how long could you maintain that treatment for skeleton goodness.

This is the truth. Everything else, based on epidemiology, statistics and sophisticated, noninvassive assessments of bone strength and Fx risk is just that, i.e. numbers. You are free to take these numbers the way you regard the best for the patient (not an easy task, indeed - "prima non nocere" could be a good point to start with).

@Dr. Ferretti:  This is an impressive post. Thank you. However, I did not understand if you use BP at all and how long if you do so ? This is the question.

Regards

Dear Dr Ferretti. This is a very informative entry and very detailed, although it is not clear which treatment do you prefer at the end. Is Bisphosphonates the treatment of choice for you or not and if it is what is your regime protocol?

I am not a clinician. I am just an investigator who prefer to think about what one is doing in every paticular case ("one is one plus his/her circumstances" - Ortega) rather than follow "general rules" to treatment or monitor patients concerning somehting so much relevant to QOL as bone strength or Fx risk. In case of BPs, one has to seriously remember that BPs just block bone resorption; they do not add any further bone mass than that resulting from the refilling of the previously ongoing resorption pits.

Concretely, in this case I think as follows 

Treatments (BPs included) should be always addressed to the variable(s) you have found to be out of control (what ever else?). In this case, unfortunately, the only measurable variables you can assess in daily clinic to deduce that a BP treatment would be of some help (provided that the well-knosn risk factors have been dealt with) are just two, namely,

1. the DEXA BMD of the bone region you suppose to be on danger (this will show you, though quite roughly, whether there has or has not already taken place some really relevant bone loss to neutralize) and

2. one suitable indicator of osteoclastic resorption (to be sure that an over-stimulation of that process is the real cause of "1").

Only in case you have found both these indicators affected you will really be on right at all to indicate a BP treatment.

In that "necessary" case you should monitor the effect of BP treatment strictly guided by the values of the variables you had selected to control, namely, DEXA BMD and the resorption indicator. Usually, the resorption indicator under BP treatment goes either right or reasonably controlled in a question of months, regardless of the BMD value. You should look at the indicator (etiological treatment), not at the BMD (circumstantial bone feature). When the indicator is normalized, you have just to take note of the (eventual) BMD improvement as just a side effect.

Now you have two problems to face, namely,

1. To follow the resorption indicator values until they reach a real plateau. It would be nonsense to continue the BP treatment thereafter, unless you get surprised by a later reversion of this situation (bone resorption going up again), regardless of the BMD value.

2.  To follow the (possible) improvement of the BMD value in terms of the T-score (as an expression of having no better thing to do at all, rather badly). Then, you may fall into one of the following two opposite situations:

     a. A reasonable good result is achieved, as shown by obtaining normal or acceptable values of both the resorption indicator and the T-score value within the established term of BP treatment (God bless). Then, you should follow the untreated patient every 6 months (indicator + DEXA) during the next first year, and every year thereafter in the happy case of a maintainance of reasonable good values of both vartiables, or

    b. Things look to go bad, and bone loss starts again. Then you should update your patient evaluation as a whole and decide whether to reinstall BP treatment following the above principles or just to change it to another type of drug.

        If the predominant source of concern is now a re-acceleration of bone remodeling rate with bone mass being reasonably maintained within acceptable values, then you could give Denosumab during 1-2 years (perhaps just to find that after retrieval things could go bad again). If, opposedly, bone mass is being lost at an undesirable rate and/or the T-score reaches totally unwanted levels, you should think seriously on a PTH treatment (please consult a dedicated Osteologist, if you yourself are not one, before taking this decision). Strontium ranelate could be a further option, but going into this type of discussion would fall outside this matter.

Now, when could you decide to maintain BP treatment? I have already expressed my concerns about the lack of suitable monitoring control elements to assess the actual result of any BP treatment on what really matters, i.e. bone strength and Fx risk. From my point of view, you should only give the BP again and again (provided that everything else is OK to take that decision) when you face a reiteration of the "surprise" referred to in "1" above. To maintain this decision for long will face you to some known risks and perhaps also a lot of unknown ones, as spontaneous fractures because of an excessive bone tissue stiffness and an impaired bone tissue toughness, ONJ, etcetera. 

I have nothing to tell about when, in these cases, you should stop BP treatment. If I was required to settle an opinion, I would prefer to do it rather earlier than later (obviously, this is an empyrical proposition, yet I guess you couldnot get much more than that from the many excellent clinicians who have lots more expertise than myself in treating patiens.

It is difficult to tell "I don't know". I hope to have been successful in doing that after writing a not too large number of words.

@ Dr. Ferretti:

Thank You for Your contributions. I agree in many parts, but some of Your ideas are hypotheses too and not confirmed by clinical data in humans.

1. Osteoporosis is not only induced by increased osteoclastic activity, but can be due to reduced osteoblast function (especially in higher age oder after somme weeks /onths of glucocorticoid therapy).

2. Clinical data did show in the large patient cohorts of Phase III, that higher values of bone turnover markers (BTM) result in higher fracture risk compared to lower ones. But, RELATIVE fracture risk reduction is independent of BTM at therapy start. This is valid for BIS as well as for teriparatide.

3. To my knowledge, the structural reasons of atypical femoral fractures after long term BIS therapy are still not clear, supposed "over-mineralisation" seams not to occur. And clinical data comparing "normal" osteoporotic patients with atypical fracture patients did not show clear differences in bone material properties (see publication by Güerri-Fernandez and colleagues from Barcelona:

http://onlinelibrary.wiley.com/doi/10.1002/jbmr.1731/abstract;jsessionid=6E3EBA4E40D3A077BAED51D758E2A79E.f02t02

Dear Dr Muche,

                               Thank you for your authorized comments, with which I totally agree. In fact, your observations complement some aspects of the arguments I expressed in my previous answers, rather than showing a different conception of the matter, in a way that most clinicians surely will find interesting to analyze.

1.  My allusion to increased osteoclast activity as a determinant of osteoporosis by no means implies putting aside osteoblasts; on the contrary, when I regarded osteoporosis as a disease of bone design (i.e. comprising bone modeling as a relevant factor) I did imply a combined role of osteoblasts and clasts in the mechanism of this disease.

2. Just intending to argue concerning the length a BP treatment could be given, I have proposed that what should be monitored to assess the effect of BPs in an individual patient is the effect over the resorption markers, which must obviously have to be higher than normal at baseline to have justified such therapy. The statistical independence of Fx risk from the baseline BTM values in cohort studies reflects a different finding that does not contradict that assumption. On the other side, for sure, abnormal values of either or both, bone formation and/or resorption markers could add some concern about the future of any kind of osteoporosis. However, BTM are by no means direct determinants of neither bone strength (which always results from a combination of a deterioration of bone material and/or geometric properties) nor fracture risk (which is an epidemiological indicator that depends on the combined occurence of many, skeletal and extra-skeletal factors).

3. My allusion to the facilitation of atypical femoral fractures as a risk derived from BP treatment (regardless of the involvement or not of bone material properfties in their pathogenesis) was rather addressed to the importance of the length of BP treatment than to the intervention of one or more types of BP-evoked factors in the mechanism of fracture. Nevertheless, BP treatment does decrease bone tissue toughness, and the intensity of this effect could be crucial to bone fragility determination in any site of the skeleton, and this looks as a valid argument to incline specialists to shorten BP treatments as a general rule.

Dear Collegues   - Thank you all for your valuable and stimulating contribution to this discussion. On this opportunity may I wish you all :  A Happy  New  Year  2016  !

And now a branching issue : To what extent do these understandings relate equally to women and men  ? 

Thank you, Dr. Leibovitz ! 

In return, I also wish you and all the colleagues a Happy New year ! 

Regards, 

Panayot Tanchev

@ Dr Leibovitz:

"To what extent do these understandings relate equally to women and men ?"

In my opinion/knowledge/experience, there is no significant difference between men and women regarding treatment effects of BIS, denosumab, strontium ranelate or teriparatide. I am not sure about effects of testosteron (supplements) in women, while Estrogens and SERM do not have relevant positive effects in male osteoporosis.

So, also negative effects of long term BIS should not differ between men&women (AFF, ONJ).

Hopefully, we will leave the concepts of "postmenopausal" vs. "male" osteoporosis behind and propable future clinical trials with fracture endpoint include both genders (which makes trial recruitment easier and improves availability of therapies to every patient in need).

Thanks for the interesting contributions in this website & all the best for 2016!

Should the treatment for osteoporosis with bisphosphonates be limited in time ? For how long ?

Hard QUESTION

My experience in the treatment of postmenopousal osteoporosis  is more than 20 years. I have many patients requiring therapy with bisphosphonates has exceeded 5 years with modest increases sometimes BMD-DXA, or better results,but which do not have reduced the T-score under -2.5. Was not warranted interruption therapy for side effects. And yet last year I had the first major comlpicati of long-time therapy (ten years) with bisphosphonates =atypical femoral fractures(AFF).

The duration of bisphosphonate treatment and the length of the drug holiday are based on fracture risk and the pharmacokinetics of the bisphosphonate used.Park-Wyllie et al. performed a nested case-control study to explore the association between bisphosphonate use and femoral fractures, and they reported that bisphosphonates treatment of more than five years was associated with an increased risk of atypical subtrochanteric or femoral shaft fractures . Therefore, it may be appropriate to consider a drug holiday in patients with a cumulative duration of bisphosphonate treatment of more than five years. For patients at low risk of fracture, bisphosphonates can be discontinued and the patients placed on a drug holiday.  For those at high risk of fracture, it may be beneficial to continue bisphosphonate treatment beyond five years. Patients who are at moderate risk of fracture and in a low turnover state can be managed in a fashion that is similar to those at low risk of fracture. However, patients who are at moderate risk but in a high-turnover state should be managed as if they have high risk of fracture. Fracture risk can be estimated with use of the World Health Organization’s fracture risk assessment tool (FRAX) as well as bone turnover markers.

Consequently, for most people with osteoporosis, the proven fragility-fracture risk-reduction benefits of bisphosphonates outweigh the risks of AFF .

1.Park-Wyllie LY, Mamdani MM, Juurlink DN, et al. Bisphosphonate use and the risk of subtrochanteric or femoral shaft fractures in older women. JAMA. 2011, 23; 305(8):7  2.T. Blood, R. J. Feller, E. Cohen, et al. Atypical Fractures of the Femur: Evaluation and Treatment. JBJS Reviews, 2015; 3 (3): e1 DOI: 10.2106/JBJS.RVW.N.00062. 

@ Dr. Galesanu:

Thank You for Your answer. I am aware of these papers.

One of the tricky questions is, how to define low-moderate-high risk? The percentage of fracture risk expected by FRAX maybe quite different in treated vs. untreated patients - even if some papers suggest to use FRAX also in treated patients.

No one really knows the answer.

Next is the individual succes of BIS therapy: What, if the patient started the therapy with BIS at a T-score of -4.0 and is now at -2.7 --- is it the same risk like one who started at -3.0? or even with no apparent change of BMD during 5 years of BIS?

In a patient with "high-turnover state" despite previous BIS (how exactly to evaluate given the problems of actually available BTMs), I would think about incompliance, incorrect tablet intake, missing intestinal resorption of BIS or (those patients shall exist sometimes) resistancy of the osteoclast to BIS.

And there is of course the option, that a new bone disease occured meanwhile (like bone metastasis or malignant gammopathy). I personally would not proceed in such a patient with a BIS (probably change from oral to i.v. or to denosumab).

These are questions we will never solve due to missing long-term clinical data with fracture endpoints in real world patients (and real world doctors as well).

But, compared to many other diseases and therapies (like high blood pressure, diabetes) we should be happy: we have at least strong data on the final therapeutic aim, the fracture over at least 3-5 years of therapy.

Once bisphosphonate therapy is started, it should continue for at least 3 years (in the absence of any adverse events. For alendronate, there is benefit in improving bone density and reducing vertebral fractures when treatment is continued for 10 years, and similarly for IV zoledronate for 6 years. The other bisphosphonates do not have such long-term data, both risedronate and ibnadronate do not suppress bone turnover markers for long after stopping treatment, and they can start to lose bone by 6 months of stopping. So, it is important to take into consideration which bisphosphonate the patient is taking.

Thus, who should continue long-term? Patients with previous hip or vertebral fractures or those who fracture during treatment should definitely continue for 5-10 years. Also those with hip BMD by DXA of less than T -2.5, and finally those at high risk (older patients, secondary osteoporosis etc) should also take treatment for longer.

There is a recent review of this question by the American Society of Bone and Mineral Research which you may find helpful.

http://onlinelibrary.wiley.com/doi/10.1002/jbmr.2708/abstract;jsessionid=05E0251ED4942A924D3949655BEDAFE1.f03t02