Editorials

Oral bisphosphonates and oesophageal cancer

BMJ 2010; 341 doi: http://dx.doi.org/10.1136/bmj.c4506 (Published 02 September 2010) Cite this as: BMJ 2010;341:c4506
  1. Diane K Wysowski, epidemiologist
  1. 1Division of Epidemiology, Office of Surveillance and Epidemiology, Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA
  1. diane.wysowski{at}fda.hhs.gov

    Evidence is inconclusive but cautious prescribing and active follow-up are advised

    Oral bisphosphonates have been marketed for the treatment and prevention of osteoporosis since the mid-1990s. In the linked nested case-control study (doi:10.1136/bmj.c4444), Green and colleagues find that the risk of oesophageal cancer increases with the number of prescriptions of oral bisphosphonates.1 Is the association likely to be true? And if so, what should clinicians tell their patients?

    In 1994, erosive oesophagitis was reported in five patients in a clinical trial of oral pamidronate.2 A year later, multiple ulcerative oesophagitis was described in a 59 year old woman who had taken one alendronate tablet a day for seven days with half a glass of water while lying down.3 She developed progressively severe retrosternal pain, odynophagia, and dysphagia, then stopped the drug and consulted her doctor, who prescribed an H2 receptor blocker without success. Endoscopy showed an ulcerated lesion in the lower oesophagus with spontaneously bleeding mucosa and a white exudate.

    White exudate was also reported in five patients who took alendronate 10 mg a day and developed oesophageal symptoms after taking no specific precautionary measures.4 Endoscopy showed similar erosions and ulcerations covered with a thick, white, loosely adherent exudate. Biopsies from four of the five patients contained actively inflamed granulation tissue and a leucofibrinous exudate. Two patients had crystalline material in the exudate that was similar to shavings from alendronate tablets. Follow-up endoscopy in two patients suggested that mucosal abnormalities may persist after ulceration heals.4

    Crystalline foreign material was also reported in biopsies from six of 10 patients who experienced erosive and ulcerative oesophagitis, an inflammatory exudate, and inflamed granulation tissue after taking alendronate.5 Other reports of oesophageal injury included a perforated oesophagus resulting in death,6 and an oesophageal stricture resistant to dilation in a woman who had taken alendronate correctly for 10 months.7

    All these reports prompted a review of case reports of oesophageal cancer in people taking oral bisphosphonates that had been voluntarily submitted to the US Food and Drug Administration, which was reported in a letter to the editor in the January 2009 issue of the New England Journal of Medicine.8 By September 2009, the case series included 34 people from the US (one with previously diagnosed Barrett’s oesophagus), and the same number from both Europe and Japan. Several oral bisphosphonates were implicated. Mean latency from drug use to diagnosis was four years or more in the United States and two years or more in Europe and Japan. The reports could not prove that bisphosphonates caused the cancers because of possible detection bias, the absence of controls, and missing information about potential confounders. Replies to the 2009 letter included two studies suggesting that oral bisphosphonates have no effect on cancers of the digestive tract, but the methods and data were sparse.1 9 10

    A recent epidemiological study conducted in the UK’s General Practice Research Database, which compared the incidence of oesophageal cancer and gastric cancer in patients who were exposed or not exposed to oral bisphosphonates, found no increase in the risk of either cancer.11 This is in contrast to Green and colleagues,1 who used the same database to compare the frequency of oral bisphosphonate exposure in cases versus matched non-cases. They found that one (or more) prescription for oral bisphosphonates increased the risk of oesophageal cancer by 30%; 10 or more prescriptions nearly doubled the risk. No increased risk was found for cancers of the stomach or colorectum.

    What could cause these differences? A major difference in the two studies was the length of follow-up, with the negative study having 4.5 years and the positive one 7.7 years. The positive study also had an adequate sample size, control group, and adjustment for covariates (age; sex; smoking status; alcohol drinking; body mass index; diagnosis of osteoporosis; previous fracture; upper gastrointestinal disease; and prescription of non-steroidal anti-inflammatory drugs, corticosteroids, or acid suppressants). However, neither study validated diagnoses by medical records nor provided information on whether drugs were taken according to directions.

    Even if oral bisphosphonates increase oesophageal cancer, the incidence in the population would be expected to remain relatively low. Nevertheless, the possibility of adverse effects on the oesophagus should prompt doctors who prescribe these drugs to consider risks versus benefits,12 to ask patients about digestive disorders before prescribing, and to reinforce directions on the basis of the individual product with each prescription (take on rising for the day with a full glass of plain water at least 30 minutes or one hour before the first food, beverage, or medication, and do not recline for at least 30 minutes or one hour and not until after the first food of the day) with each prescription. In addition, doctors should tell patients to report difficulty in swallowing and throat, chest, or digestive discomfort so that they can be promptly evaluated and possibly advised to discontinue the drug.

    Notes

    Cite this as: BMJ 2010;341:c4506

    Footnotes

    • Research, doi:10.1136/bmj.c4444
    • Competing interests: The author has completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; and no other relationships or activities that could appear to have influenced the submitted work.

    • Provenance and peer review: Commissioned; not externally peer reviewed.

    • The views expressed are those of the author and do not necessarily represent the official position of the US Food and Drug Administration.

    References