Skip to main content
NCBI home page
Therapeutic Advances in Musculoskeletal Disease logoLink to Therapeutic Advances in Musculoskeletal Disease
. 2015 Aug;7(4):160–168. doi: 10.1177/1759720X15582144

Oral bisphosphonates and colon cancer: an update

Pia Eiken 1,✉, Peter Vestergaard 2
PMCID: PMC4530382  PMID: 26288666

Abstract

Bisphosphonates (BPs) are widely used as the main treatment for osteoporosis. In vitro and animal studies suggest that use of BPs may have a potential for colorectal cancer (CRC) prevention. Safety and efficacy in terms of osteoporosis prevention have only been evaluated in randomized controlled trials (RCTs) of relatively short duration (3–5 years), with smaller extension studies. The evidence for a benefit beyond 5 years is limited and intake of BPs has not shown any relationship with CRC in intervention studies. Observational studies and meta-analysis have shown unchanged or decreased risk of CRC. BPs used for treatment and prevention of osteoporosis should not be applied for prevention of CRC in clinical practice.

Keywords: bisphosphonates, colorectal cancer, risk

Introduction

Colorectal cancer (CRC) is the third most commonly diagnosed cancer in men and the second in women worldwide [Jemal et al. 2011]. It is often found in people aged 50 or older with a median age at diagnosis around 70 years. The use of bisphosphonates (BPs) for the treatment of osteoporosis is well established and generally well tolerated, and makes up by far the most widely used drug class for the treatment of osteoporosis. BPs have a relatively low cost and the ability to reduce the risk of osteoporotic fractures by 30–70 % depending on fracture site and setting [Reginster, 2011].

BPs are chemically stable analogues of inorganic pyrophosphate. Etidronate, clodronate and tiludronate are non-nitrogen containing BPs and are used less in osteoporosis treatment today. Most BPs in routine use today contain a nitrogen atom in the R2 side chain and are known as the ‘nitrogen-containing BPs’ (N-BPs) (e.g. alendronate, risedronate, ibandronate, pamidronate and zoledronate). This configuration increases their potency and enables them to inhibit the effects of farnesyl pyrophosphate synthase (FPPS). FPPS is key enzyme of the mevalonate/cholesterol biosynthetic pathway [Russell, 2011]. Inhibition of FPPS reduces post-translational expression of small guanidine triphosphate (GTPase) proteins and thus disturbs the function of osteoclasts [Russell et al. 2008]. There is evidence from animal studies that N-BPs may modify intestinal inflammatory diseases. Thus, local administration of BPs to animals with chemically induced ulcerative colitis and local exposure to a procarcinogen ameliorates inflammation and significantly reduces the development of tumors in the large bowel [Ballester et al. 2007; Sassa et al. 2009].

In vitro studies have shown that several N-BPs inhibit the growth and induce apoptosis of human CaCo-2 colon carcinoma cells [Suri et al. 2001], inhibit the proliferation and induce apoptosis of HTC-116 colon carcinoma cells [Sewing et al. 2008]. Furthermore, N-BPs are known to stimulate gamma delta (γδ) T cells, which may induce antitumor properties [Todaro et al. 2009]. BPs have been shown to exert powerful antitumoral properties in a variety of human cancers and are efficacious in preventing and delaying cancer-related bone disease [Rizzoli et al. 2013]. When BPs are given orally, usually administered weekly (previously daily) or monthly, less than 1% is absorbed [Ezra and Golomb, 2000; Pazianas and Russell, 2011]; this leaves 99% to pass unchanged through the small and large intestine, where it potentially could affect the function of the enteric epithelium. Most of patients treated for osteoporosis use oral formulations, whereas intravenous (I.V.) formulations are most often used for malignant diseases. For these reasons, orally administered BPs might have local antineoplastic actions in the colon in patients treated for osteoporosis and therefore the prospect of using them as prophylaxis for CRC is becoming a matter of debate.

Randomized controlled trials (RCTs) have provided relatively little information on potential long term safety issues. This is especially due to the short duration of the trials (mainly less than 3 years), the small number of patients who completed long-term extensions and the lack of a long-term placebo group for comparison of rates. For some unexpected safety concerns such as osteonecrosis of the jaw, atypical femur fractures, adjudication procedures were not part of the study plan for the initial RCTs. With regard to other studies of BPs in relation to gastrointestinal cancers, none of the many large prospective double-blind randomized trial registration studies for BPs in osteoporosis have revealed any relationship [Bone et al. 2004; Mellstrom et al. 2004; Black et al. 2006, 2012], the longest with 10 years follow up. However, the observation period in these studies was probably too short and the number of exposed patients too small to show any effects.

There are no RCTs that addressed colon cancer prevention among patients on BPs. There are six cohort studies: two from the same Danish pharmacy prescription database [Pazianas et al. 2012; Vestergaard, 2011], two from the US [Passarelli et al. 2013; Khalili et al. 2012], one from Taiwan [Chiang et al. 2012] and one from the UK [Cardwell et al. 2012]. There are four case-control studies: one from Israel [Rennert et al. 2011], two from the UK [Green et al. 2010; Vinogradova et al. 2013] and one from Canada [Singh et al. 2012] (Table 1), all addressing the question of the possible association between use of oral BPs against osteoporosis and risk of developing colon cancer and on survival. Several meta-analyses in the field have also been performed [Ma et al. 2013; Bonovas et al. 2013; Thosani et al. 2013; Singh et al. 2013a; Yang et al. 2013; Oh et al. 2012]. The purpose of this review was to give an update on these observational studies.

Table 1.

Characteristics of studies on exposure to oral bisphosphonates and risk of colorectal cancer.

All On BPs Not on BPs RR, OR or HR adjusted (95% CI) Confounding variables adjusted* Used BPs
Study Design Sex Country/setting Period CRC Total CRC Controls Cases Controls
Green et al. [2010] CC F (46%) UK, GPRD 1995–2005 10,641 63,663 276 1555 10,365 51,467 RR 0.87 (0.77–1.00) 1–5 E, A, R
Rennert et al. [2011] CC F (100%) Israel, Clalit Health Service Database 2000–2006 933 1866 97 138 836 795 RR 0.41 (0.35–0.71) 1–3, 6–13 A (94.7%)
Singh et al. [2012] CC F (46%) Canada, Manitoba Health database 2000–2009 5425 59,667 293 3392 5132 50,850 OR 2–13 BP prescription >5 years, 0.84 (0.71–1.00) 1, 2, 10, 13–18 A (79.2%), E, R
⩾14 prescription >5 years, 0.78 (0.65–0.94)
Vinogradova et al. [2013] CC QRD: F (40.1%) UK, QRD and CPRD 1997–2011 QRD: 20,106 QRD: 114,060 QRD: 929 QRD: 4345 QRD: 19,177 QRD: 89,609 QRD: OR 1.03 (0.94–1.14) 1–7, 10, 12, 14, 15, 19–22 A, E, R, I
CPRD: F (41.9%) CRPD: 19,035 CRPD: 108,146 CRPD 902 CRPD: 4231 CRPD: 18,133 CRPD: 84,880 CPRD: OR 1.10 (1.00–1.22)
Vestergaard [2011] Cohort F (84.7%) DK, National register Database 1996–2006 1682 371,432 411 92,518 1271 277,232 A: HR: 1.16 (0.93–1.45) 1, 2, 4, 5, 10, 18, 21, 22, 23 A, E, C
C: HR 1.31 (0.28–6.10)
E: HR 1.05 (0.92–1.12)
Cardwell et al. [2012] Cohort F (81%) UK,GPRD 1996–2006 608 92,072 264 45,772 344 45,692 HT 0.74 (0.60–0.91) 3–5,10, 12, 13, 20 A, R, I, E
Pazianas et al. [2012] Cohort F (100%) DK, National register Database 1996–2005 1683 155,030 262 30,344 1421 123,003 HR 0.69 (0.60–0.72) 1, 2, 10, 13, 15, 20, 22 A
Khalili et al. [2012] Cohort F (100%) US, US female registered nurses 1998–2008 801 86,277 95 4147 706 81,329 HR 1.04 (0.82–1.33) 1–13, 16, 19, 22, 24 A, E, R
Chiang et al. [2012] Cohort F (100%) Taiwan, National Health Insurance Research Database 1998–2009 518 27,603 119 6787 399 20,298 HR 1.02 (0.83–1.26) 1, 2 A
Passarelli et al. [2013] Cohort F (100%) US, Women’s Health Initiative 1993–1998 1931 156,826 126 3044 1805 153,782 HR 0.88 (0.72–1.07) 1–8, 10, 12, 13, 16, 18, 25, 26 A (>90%), R, E
Open in a new tab
*

Adjusted variables. 1 Sex; 2 Age; 3 Body mass index; 4 Smoking (use of inhaled corticosteroids and beta-agonists as a proxy for smoking); 5 Alcohol; 6 Ethnicity; 7 Family history of CRC; 8 Physical activity; 9 Food consumption (e.g. vegetables, meat, fruit, fat); 10 Aspirin/ nonsteroidal anti-inflammatory drugs/ cyclo-oxygenase-2 inhibitors; 11 Statin use; 12. Use of calcium and vitamin D; 13. Hormone replacement therapy; 14 Diabetes mellitus;15 Comorbidities (e.g. coronary heart disease, inflammatory bowel disease, rheumatoid arthritis, any cancer; Charlson’s index);16 Lower gastrointestinal procedures (endoscopies, colonic surgery) or screening; 17 Total number of ambulatory care visits preceding the index date; 18 Socioeconomic status, education; 19 History of osteoporosis (including diagnosis of osteoporosis, osteopenia or previous fracture); 20 Systemic corticosteroids; 21 Acid suppressive drugs; 22 Gastrointestinal disorders (dysphagia, oesophagitis, gastro-oesophageal reflux disease, hiatus hernia, oesophageal ulcers, Barrett’s oesophagus, gastritis, duodenitis, peptic ulcers, dyspepsia, polyps, Crohn’s disease or ulcerative colitis; 23 Gastric surgery 24 Folate intake; 25 Mammography; 26 5-year hip fracture probability (measured at baseline).

A, alendronate, BPs, bisphosphonates; C, clodronate; CC, case-control; CI, confidence interval; CRC, colorectal cancer; CRPD, Clinical Practice Research Datalink, previously known as GPRD, General Practice Research Database; DK, Denmark; E, etidronate, F, female; I, ibandronate; OR, odds ratio; QED, QResearch Database; R, risedronate; RR, relative risk; HR, hazard ratio; UK, United Kingdom; US, United States.

Methods

A PubMed search using the search terms “bisphosphonate” or trade names of the drug AND [“colorectal cancer” OR “colorectal carcinoma” OR “colorectal neoplasms”] was conducted in August 2014. Details of the search string and results can be obtained from the authors. A total of 139 articles were retrieved from the search. All references cited in the identified articles were also reviewed. A total of 10 studies were included in the final analysis and 129 studies were excluded as they did not report original data. All the searches were restricted to human studies. No formal meta-analysis of the retrieved articles was carried out because of the heterogeneity of the populations, interventions and lengths of follow up, but we performed a limited meta-analysis using a random effects model (using a DerSimonian and Laird estimator [Böhning, 2000]) on the estimates from the observational studies (Table 1). The random effects model was chosen due to the heterogeneity of the studies.

Using Centre For Evidence based medicine grading of evidence for treatment/prevention (www.cebm.net), the level of evidence was 1 for the randomized controlled trials, 3 for the cohort trials and 4 for the case-control trials.

Exposure to BPs

As seen from Table 1, the number of cases with CRC among those treated with BPs is low. Alendronate was the most commonly prescribed type, followed by etidronate in most studies. In some of the studies, all exposed used alendronate [Pazianas et al. 2012; Chiang et al. 2012]. In the studies covering a time period before year 2000, the most commonly prescribed BP was etidronate and the use declined markedly after introduction of weekly administered alendronate.

In all studies, the overall use of BPs was low: 2–3 % [Green et al. 2010; Pazianas et al. 2012; Passarelli et al. 2013]; 4.5% [Vinogradova et al. 2013); 6% [Singh et al. 2012]; 13% [Rennert et al. 2011]; and up to 25% [Chiang et al. 2012]. This may have an impact on the strength of the association between BPs and CRC. The exposure of men to BPs use was low; for example, Singh and colleagues reported that only 1.3 % of men were on BPs while 12% of the women were exposed to BPs [Singh et al. 2012]. The Women’s Health Initiative (WHI) study began around the time when BPs first started to be commonly prescribed [Passarelli et al. 2013]. Use of BPs was low at baseline as expected (2%) and half of the users stayed on the medicine for less than 1 year, but BPs became increasingly common over the course of follow up [Passarelli et al. 2013]. Likewise in the Nurses’ Health Study (NHS) [Khalili et al. 2012], at baseline in 1998, only 4.9 % of women regularly used BPs, increasing to 19.8% by 2008. When the WHI and NHS studies were initiated hormone therapy (HT) was commonly considered for prevention of osteoporosis, but after the WHI study in 2003 reported increased risk of breast cancer, use of HT declined significantly.

Use of BPs varied considerably in all the included studies and the quantity of medication per prescription also differed between the studies. Some studies [Rennert et al. 2011; Singh et al. 2012; Pazianas et al. 2012; Chiang et al. 2012; Vestergaard, 2011] considered the pharmacy records of filling a prescription as a measure of exposure, while some considered the number of written prescriptions as a measure [Green et al. 2010; Vinogradova et al. 2013; Cardwell et al. 2012]. Khalili and colleagues considered use of BPs as based on an individual’s response to the survey of questions, putting them at risk of recall bias [Khalili et al. 2012], whereas Passarelli and colleagues confirmed use of BPs, type and intravenous preparation via assessments of medication bottles [Passarelli et al. 2013].

The risk of CRC among individuals who were exposed to only one prescription could be expected to be very similar to that of nonexposed individuals. For drugs prescribed in cycles (e.g. etidronate), a prescription for 2 weeks was assessed as 90 days’ prescription. Exposure to BPs was therefore defined differently among the studies. Singh and colleagues defined exposure to BPs as the dispensing of 2 or more prescriptions [Singh et al. 2012], while Rennert and colleagues defined exposure to BPs as dispensation of 3 or more prescriptions (1 prescription = median 30 days) [Rennert et al. 2011]. Green and colleagues [Green et al. 2010] and Vinogradova and colleagues [Vinogradova et al. 2013] defined exposure as 1 prescription (1 prescription = 1–3 months of treatment). Chiang and colleagues, using only alendronate, defined exposure of at least 1 month [Chiang et al. 2012]. Vestergaard reported exposure from the first day of prescription of a drug against osteoporosis [Vestergaard, 2011]. Pazianas and colleagues used the medication possession ratio as exposure and calculated the number of defined daily doses (e.g. 10 mg per day for alendronate) issued to the patient from the pharmacy divided by the number of days in the time interval [Pazianas et al. 2012]. Passarelli and colleagues handled the use of BPs as a time-varying never/ever use by updating baseline use at years 1, 3 and 6 [Passarelli et al. 2013].

Duration of use of BPs was also defined differently. Short-term use was defined as use: <210 days [Singh et al. 2012]; <1 year [Rennert et al. 2011; Green et al. 2010]; >1 year [Cardwell et al. 2012]; 1–2 years [Khalili et al. 2012]; and <2 years [Vestergaard, 2011; Chiang et al. 2012]. Long-term use was defined as: >2 years [Chiang et al. 2012]; >3 years [Rennert et al. 2011; Green et al. 2010]; >1250 days [Singh et al. 2012]; >4 years [Cardwell et al. 2012]; and >5 years [Khalili et al. 2012; Vestergaard, 2011].

Duration of BPs use and risk of CRC

Most studies found no association between BPs use and risk of CRC [Khalili et al. 2012; Passarelli et al. 2013; Vinogradova et al. 2013; Green et al. 2010; Pazianas et al. 2012; Vestergaard, 2011; Chiang et al. 2012] even among women with longer duration of use [Khalili et al. 2012; Passarelli et al. 2013; Vinogradova et al. 2013; Green et al. 2010; Pazianas et al. 2012; Chiang et al. 2012]. A cohort study found a reduced crude relative risk (RR) for colon cancer with use of etidronate for less than 5 years, but not for alendronate [Vestergaard, 2011] and further found a significant decrease in hazard ratio (HR) for colon cancer with alendronate with increasing dose, so that high doses (⩾1 defined daily dose) had a protective effect [HR = 0.29, 95% confidence interval (CI) 0.14–0.62] [Vestergaard, 2011]. Singh and colleagues found BPs was associated with a reduced risk of CRC only for risedronate [odds ratio (OR) 0.50, 95% CI 0.30–0.85] and not for alendronate [Singh et al. 2012]. However, very few were patients exposed to risedronate. Weekly alendronate was associated with a greater risk reduction than daily alendronate [Pazianas et al. 2012]. It could well be that the higher concentration of alendronate (seven-fold) achieved in the gut on the day of the administration of the weekly regimen is more effective than the lower daily dose [Pazianas et al. 2012]. Other studies found no association BPs and risk of CRR with daily, weekly or monthly use [Vinogradova et al. 2013].

Performing a meta-analysis using a random effects model (DerSimonian and Laird estimator) on the estimates in Table 1 yielded a pooled risk estimate of CRC with bisphosphonate use of 0.89 (95% CI 0.79–1.00; borderline significant with p = 0.04). The studies were highly heterogeneous with respect to results (p < 0.01 by χ2 test for heterogeneity).

Stratifying the analysis by study design yielded the following risk estimate for the case-control studies: 0.83 (95% CI 0.66–1.05). This estimate was highly heterogeneous (p < 0.01 by χ2 test for heterogeneity). For the cohort studies, the combined risk estimate was 0.93 (95% CI 0.76–1.12). In addition, this estimate was highly heterogeneous (p < 0.01 by χ2 test for heterogeneity). The two estimates did not differ significantly (p = 0.46 for direct comparison of estimates). It thus did not seem that case-control studies, which by design are immune to say lead time bias, yielded different estimates from cohort studies. Study design and evidence level thus did not seem to bias the estimates. The combines estimate from this analysis was in the range found by others (Table 2).

Table 2.

Meta-analyses: risk for colorectal cancer and bisphosphonate use.

Study Cases of CRC Fixed effect model: any use RR (95% CI) Random effect model: any use RR/OR (95% CI) Duration of use RR/OR (95% CI) CC and cohort studies used in model
Oh et al. [2012] 11,574 RR 0.62 (0.30–1.29) NS* Green et al. [2010], Rennert et al. [2011]
Ma et al. [2013] 18,670 RR 0.80 (0.74–0.85) RR 0.80 (0.71–0.90) NS* Singh et al. [2012], Cardwell et al. [2012], Pazianas et al. [2012], Rennert et al. [2011], Green et al. [2010], Khalili et al. [2012]
Bonovas et al. [2013] 21,839 RR 0.85 (0.80–0.90) RR 0.85 (0.75–0.96) Short-term use: NS* Rennert et al. [2011], Green et al. [2010], Chiang et al. [2012], Pazianas et al. [2012], Cardwell et al. [2012], Singh et al. [2012], Vestergaard [2011], Khalili et al. [2012]
Long-term use$: 0.73 (0.57–.93)
Thosani et al. [2013] 18,666 OR 0.87 (0.78–0.97) ⩾ 10 prescription OR 0.71 (0.58–0.87) Vestergaard [2011], Singh et al. [2012], Rennert et al. [2011], Green et al. [2010]
<1 year: NS*
1–3 year use OR 0.76 (0.68–0.85)
>3 years’ use OR 0.78 (0.61–0.99)
Singh et al. [2013a] 20,001 OR 0.83 (0.76–0.90) OR 0.85 (0.74–0.98) NS* Green et al. [2010], Rennert et al. [2011], Singh et al. [2012], Pazianas et al. [2012], Khalili et al. [2012], Chiang et al. [2012]
Women only: NS
Yang et al. [2013] 22,291 OR 0.89 (0.80–0.99) Short-term: NS*Long-term: OR 0.76 (0.59–0.97) Rennert et al. [2011], Singh et al. [2012], Green et al. [2010], Chiang et al. [2012], Khalili et al. [2012], Vestergaard [2011], Cardwell et al. [2012], Pazianas et al. [2012]
Open in a new tab
*

NS: no significant evidence with duration of use of bisphosphonates and risk of CRC.

$

Long-term used defined as more than 3–5 years depending on the included study.

CC, case control; CI, confidence interval; CRC, colorectal cancer; OR, odds ratio; RR, relative risk.

All meta-analyses (Table 2) except one [Oh et al. 2012] found an overall significant reduction in the risk of CRC. However, none of the analysis had initially included the large WHI study [Passarelli et al. 2013], which did not find sufficient evidence to conclude that there was an association between use of BPs and CRC risk. A letter by Singh and colleagues [Singh et al. 2013b] reported an updated meta-analysis performed using studies from Table 1 apart from two [Vestergaard, 2011; Cardwell et al. 2012] which showed an inverse association (adjusted OR 0.88; 95% CI 0.79–0.99) between use of BPs and CRC. The meta-analysis suggests that uncontrolled confounding may explain why previous studies have observed decreased risk of CRC among BP users.

Effect of BPs on surviving colon cancer

A large Danish cohort study showed that administration of BPs to patients treated for osteoporosis with no previous history of cancer reduced the overall risk of dying from CRC significantly with an adjusted HR of 0.62 (95% CI 0.52–0.72) and a significantly longer survival after the diagnosis of CRC (adjusted HR 0.82, 95% CI 0.70–0.97) [Pazianas et al. 2012].

Confounders and limitations

The evidence we have so far for use of BPs and risk of CRC is based primarily on observational studies. Subjects were not randomized to receive medicine. Patients on BPs may be more likely to have a healthy lifestyle and the observed benefits in a few of the studies may be overestimated. Many studies did not have the ability to adequate control for confounders associated with CRC (Table 1) and the number of variables adjusted for varied from 2 [Chiang et al. 2012] to 17 [Khalili et al. 2012]. Some studies have indicated that supplemental calcium may have a protective effect against CRC [Carroll et al. 2010]. Some of the studies did not control for over-the-counter use of calcium and vitamin D [Green et al. 2010; Singh et al. 2012; Vestergaard, 2011; Pazianas et al. 2012; Chiang et al. 2012], which are usually administered alongside BPs. Only a few studies adjusted for statin use [Rennert et al. 2011; Khalili et al. 2012], family history of CRC [Rennert et al. 2011; Vinogradova et al. 2013; Khalili et al. 2012; Passarelli et al. 2013] and some did not adjust for body mass index (BMI) [Singh et al. 2012; Vestergaard, 2011; Pazianas et al. 2012; Chiang et al. 2012]. Confounding by low bone density may be less of a problem because CRC is not generally considered to be a hormone related malignancy and high bone mass has been associated with reduced rather than increased risk of CRC [Ganry et al. 2008; Zhang et al. 2001; Nelson et al. 2002].

The UK [Cardwell et al. 2012; Green et al. 2010; Vinogradova et al. 2013] and Danish [Vestergaard, 2011; Pazianas et al. 2012] studies used data from different but overlapping periods and there are overlap in the cases and potentially controls in the studies from each country. Using data from the large UK General Practice Research Database (GPRD) [now known as Clinical Practice Research Datalink (CRPD)], the two case-control studies reported no association [Green et al. 2010; Vinogradova et al. 2013], while the cohort study found a reduced adjusted HR 0.74 (95%CI 0.60–0.91) [Cardwell et al. 2012]. One explanation for the reduced risk of CRC in the cohort study(Cardwell et al. 2012) could be use of an earlier version of the same data source which had a higher proportion of missing data for BMI, smoking and alcohol consumption [Vinogradova et al. 2013], all of which are factors associated with risk of CRC and prescribing BPs. Registered nurses [Khalili et al. 2012] might not be representative of the general population. However, there are no data to suggest that the biological mechanisms by which BPs inhibit carcinogenesis would be expected to differ by occupation or socioeconomic status.

Limitations in the results of the meta-analysis include that research is restricted to indexed journals [Ma et al. 2013] and that no search for unpublished studies or original data was performed [Ma et al. 2013; Bonovas et al. 2013). Furthermore, the included studies were different in terms of study design and duration of exposure to BPs, and they could be biased due to confounding factors. Advantages of some of the meta-analyses were that they contacted the primary author for additional information [Singh et al. 2013a], re-examined their data according to definition of exposure [Thosani et al. 2013] and pooled the data for the meta-analysis and subgroup analyses. It is possible that the use of BPs for CRC prevention requires higher doses or longer use than seen in the studies.

The findings from the observational studies need to be confirmed in a RCT using a standardized definition of exposure to oral BPs. It will be vital that any such RCT includes use of BPs of a sufficient cumulative dose and duration as the incubation period for CRC is many years. The results of the meta-analyses could suggest a minimum of 1 year of use and/or more than 3 years of monitoring.

Few studies allowed a separation between non-nitrogen and N-BPs, and many reported an estimate based on a mix of BPs, mainly alendronate. In the only study [Vestergaard, 2011] that separated the BP types, neither alendronate nor the two non-nitrogen BPs etidronate and clodronate were associated with any significant change in risk of CRC.

Conclusion

Despite promising experimental evidence of anti-proliferative effects of BPs, case-control and cohort studies found either a slight reduction in developing CRC or no effect, and no increased risk of CRC was seen in users of BPs. BPs used for the treatment and prevention of osteoporosis should not be applied for the prevention of CRC in clinical practice.

Footnotes

Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement: P.E. is an advisory board member with Eli Lilly, Amgen and MSD, and on the speakers bureau with Amgen and Eli Lilly. P.V. has received unrestricted grants from MSD and Servier, and travel grants from Amgen, Eli Lilly, Novartis, Sanofi-Aventis and Servier.

Contributor Information

Pia Eiken, Department of Cardiology, Nephrology and Endocrinology, NOH Hillerød Hospital, Dyrehavevej 29, DK-3400 Hillerød, Denmark and Faculty of Health Sciences, University of Copenhagen, Denmark.

Peter Vestergaard, Department of Endocrinology and Clinical Institute, Aalborg University Hospital, Aalborg, Denmark and Clinical Institute, Aalborg University Hospital, Denmark.

References

  1. Ballester I., Daddaoua A., Lopez-Posadas R., Nieto A., Suarez M., Zarzuelo A., et al. (2007) The bisphosphonate alendronate improves the damage associated with trinitrobenzenesulfonic acid-induced colitis in rats, Br J Pharmacol 151: 206–215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Black D., Reid I., Boonen S., Bucci-Rechtweg C., Cauley J., Cosman F., et al. (2012) The effect of 3 versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-Pivotal Fracture Trial (PFT), J Bone Miner Res 27: 243–254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Black D., Schwartz A., Ensrud K., Cauley J., Levis S., Quandt S., et al. (2006) Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial, JAMA, 296: 2927–2938. [DOI] [PubMed] [Google Scholar]
  4. Böhning D. (2000) Computer-assisted Analysis of Mixtures and Applications: Meta-analysis, Disease Mapping and Others. Boca Raton, FL: Chapman & Hall/CRC Press. [Google Scholar]
  5. Bone H., Hosking D., Devogelaer J., Tucci J., Emkey R., Tonino R., et al. (2004) Ten years’ experience with alendronate for osteoporosis in postmenopausal women, New Engl J Med 350: 1189–1199. [DOI] [PubMed] [Google Scholar]
  6. Bonovas S., Nikolopoulos G., Bagos P. (2013) Bisphosphonate use and risk of colorectal cancer: a systematic review and meta-analysis, Br J Clin Pharmacol 76: 329–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cardwell C., Abnet C., Veal P., Hughes C., Cantwell M., Murray L. (2012) Exposure to oral bisphosphonates and risk of cancer, Int J Cancer 131: E717–E25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Carroll C., Cooper K., Papaioannou D., Hind D., Pilgrim H., Tappenden P. (2010) Supplemental calcium in the chemoprevention of colorectal cancer: a systematic review and meta-analysis, Clin Ther 32: 789-803. [DOI] [PubMed] [Google Scholar]
  9. Chiang C., Huang C., Chan W., Huang P., Chen T., Chung C., et al. (2012) Oral alendronate use and risk of cancer in postmenopausal women with osteoporosis: a nationwide study, J Bone Miner Res 27: 1951–1958. [DOI] [PubMed] [Google Scholar]
  10. Ezra A., Golomb G. (2000) Administration routes and delivery systems of bisphosphonates for the treatment of bone resorption, Adv Drug Deliv Rev 42: 175–195. [DOI] [PubMed] [Google Scholar]
  11. Ganry O., Lapotre-Ledoux B., Fardellone P., Dubreuil A. (2008) Bone mass density, subsequent risk of colon cancer and survival in postmenopausal women, Eur J Epidemiol 23: 467–473. [DOI] [PubMed] [Google Scholar]
  12. Green J., Czanner G., Reeves G., Watson J., Wise L., Beral V. (2010) Oral bisphosphonates and risk of cancer of oesophagus, stomach, and colorectum: case-control analysis within a UK primary care cohort, BMJ 341: c4444. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Jemal A., Bray F., Center M., Ferlay J., Ward E., Forman D. (2011) Global cancer statistics, CA-Cancer J. Clin 61: 69–90. [DOI] [PubMed] [Google Scholar]
  14. Khalili H., Huang E., Ogino S., Fuchs C., Chan A. (2012) A prospective study of bisphosphonate use and risk of colorectal cancer, J Clin Oncol 30: 3229–3233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ma J., Gao S., Ni X., Chen F., Liu X., Xie H., et al. (2013) Exposure to bisphosphonates and risk of colorectal cancer, Br J Clin Pharmacol 76: 320–328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Mellstrom D., Sorensen O., Goemaere S., Roux C., Johnson T., Chines A. (2004) Seven years of treatment with risedronate in women with postmenopausal osteoporosis, Calcif Tissue Int 75: 462–468. [DOI] [PubMed] [Google Scholar]
  17. Nelson R., Turyk M., Kim J., Persky V. (2002) Bone mineral density and the subsequent risk of cancer in the NHANES I follow-up cohort, BMC Cancer 2: 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Oh Y., Yoon C., Park S. (2012) Bisphosphonate use and gastrointestinal tract cancer risk: meta-analysis of observational studies, World J Gastroenterol 18: 5779–5788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Passarelli M., Newcomb P., LaCroix A., Lane D., Ho G., Chlebowski R. (2013) Oral bisphosphonate use and colorectal cancer incidence in the Women’s Health Initiative, J Bone Min Res 28: 2043–2048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Pazianas M., Abrahamsen B., Eiken P., Eastell R., Russell R. (2012) Reduced colon cancer incidence and mortality in postmenopausal women treated with an oral bisphosphonate –Danish National Register Based Cohort Study, Osteoporosis Int 23: 2693–2701. [DOI] [PubMed] [Google Scholar]
  21. Pazianas M., Russell R. (2011) Potential therapeutic effects of oral bisphosphonates on the intestine, Ann NY Acad Sci 1240: E19–E25. [DOI] [PubMed] [Google Scholar]
  22. Reginster J. (2011) Antifracture efficacy of currently available therapies for postmenopausal osteoporosis, Drugs 71: 65–78. [DOI] [PubMed] [Google Scholar]
  23. Rennert G., Pinchev M., Rennert H., Gruber S. (2011) Use of bisphosphonates and reduced risk of colorectal cancer, J Clin Oncol 29: 1146–1150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rizzoli R., Body J., Brandi M., Cannata-Andia J., Chappard D., El Maghraoui A., et al. (2013) Cancer-associated bone disease, Osteoporosis Int 24: 2929–2953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Russell R. (2011) Bisphosphonates: the first 40 years, Bone 49: 2–19. [DOI] [PubMed] [Google Scholar]
  26. Russell R., Watts N., Ebetino F., Rogers M. (2008) Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on clinical efficacy, Osteoporosis Int 19: 733–759. [DOI] [PubMed] [Google Scholar]
  27. Sassa S., Okabe H., Nemoto N., Kikuchi H., Kudo H., Sakamoto S. (2009) Ibadronate May prevent colorectal carcinogenesis in mice with ulcerative colitis, Anticancer Res 29: 4615–4619. [PubMed] [Google Scholar]
  28. Sewing L., Steinberg F., Schmidt H., Goke R. (2008) The bisphosphonate zoledronic acid inhibits the growth of HCT-116 colon carcinoma cells and induces tumor cell apoptosis, Apoptosis 13: 782–789. [DOI] [PubMed] [Google Scholar]
  29. Singh H., Nugent Z., Demers A., Mahmud S., Bernstein C. (2012) Exposure to bisphosphonates and risk of colorectal cancer: a population-based nested case-control study, Cancer 118: 1236–1243. [DOI] [PubMed] [Google Scholar]
  30. Singh S., Singh A., Murad M., Limburg P. (2013a) Bisphosphonates are associated with reduced risk of colorectal cancer: a systematic review and meta-analysis, Clin Gastroenterol Hepatol 11: 232–239.e1. [DOI] [PubMed] [Google Scholar]
  31. Singh S., Singh A., Murad M., Limburg P. (2013b) Reply: To PMID 23220165, Clin Gastroenterol Hepatol 11: 1209–1210. [DOI] [PubMed] [Google Scholar]
  32. Suri S., Monkkonen J., Taskinen M., Pesonen J., Blank M., Phipps R., et al. (2001) Nitrogen-containing bisphosphonates induce apoptosis of Caco-2 cells in vitro by inhibiting the mevalonate pathway: a model of bisphosphonate-induced gastrointestinal toxicity, Bone 29: 336–343. [DOI] [PubMed] [Google Scholar]
  33. Thosani N., Thosani S., Kumar S., Nugent Z., Jimenez C., Singh H., et al. (2013) Reduced risk of colorectal cancer with use of oral bisphosphonates: a systematic review and meta-analysis, J Clin Oncol 31: 623–630. [DOI] [PubMed] [Google Scholar]
  34. Todaro M., D’Asaro M., Caccamo N., Iovino F., Francipane M., Meraviglia S., et al. (2009) Efficient killing of human colon cancer stem cells by gammadelta T lymphocytes, J Immunol 182: 7287–7296. [DOI] [PubMed] [Google Scholar]
  35. Vestergaard P. (2011) Occurrence of gastrointestinal cancer in users of bisphosphonates and other antiresorptive drugs against osteoporosis, Calcif Tissue Int 89: 434–441. [DOI] [PubMed] [Google Scholar]
  36. Vinogradova Y., Coupland C., Hippisley-Cox J. (2013) Exposure to bisphosphonates and risk of gastrointestinal cancers: series of nested case-control studies with QResearch and CPRD data, BMJ 346: f114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Yang G., Hu H., Zeng R., Huang J. (2013) Oral bisphosphonates and the risk of colorectal cancer: a meta-analysis, J Clin Gastroenterol 47: 741–748. [DOI] [PubMed] [Google Scholar]
  38. Zhang Y., Felson D., Ellison R., Kreger B., Schatzkin A., Dorgan J., et al. (2001) Bone mass and the risk of colon cancer among postmenopausal women: the Framingham study, Am J Epidemiol 153: 31–37. [DOI] [PubMed] [Google Scholar]

Articles from Therapeutic Advances in Musculoskeletal Disease are provided here courtesy of SAGE Publications

RESOURCES