Osteoporosis International

, Volume 21, Issue 6, pp 1041–1045

Of mice and men: divergent risks of teriparatide-induced osteosarcoma


  • V. Subbiah
    • Department of Sarcoma Medical OncologyThe University of Texas M.D. Anderson Cancer Center
  • V. S. Madsen
    • Department of Sarcoma Medical OncologyThe University of Texas M.D. Anderson Cancer Center
  • A. K. Raymond
    • Department of PathologyThe University of Texas M.D. Anderson Cancer Center
  • R. S. Benjamin
    • Department of Sarcoma Medical OncologyThe University of Texas M.D. Anderson Cancer Center
    • Department of Sarcoma Medical OncologyThe University of Texas M.D. Anderson Cancer Center
Short Communication

DOI: 10.1007/s00198-009-1004-0

Cite this article as:
Subbiah, V., Madsen, V.S., Raymond, A.K. et al. Osteoporos Int (2010) 21: 1041. doi:10.1007/s00198-009-1004-0



Since approval by the U.S. Food and Drug Administration (FDA) in December 2002, teriparatide (recombinant 1-34 PTH; Forteo®) has been safely used by more than 430,000 patients. Prior to FDA approval, however, there was concern that teriparatide might increase the risk for patients to develop osteosarcoma, as almost 45% of the rats treated with this drug at the highest-tested dose level developed this aggressive form of bone cancer. Balancing the proven benefits of teriparatide shown by clinical trials with the theoretical risk for teriparatide-induced human osteosarcoma, the FDA mandated both a ‘black-box’ warning of this potential side-effect and a company-sponsored postmarketing surveillance program. As a participating institute of that surveillance program, we report upon the second person with potential teriparatide-induced osteosarcoma, in this case, complicated by a history of pelvic radiation.


Given the theoretic risk of the drug teriparatide and the known risk of radiation in inducing osteosarcoma, we raise the issue of whether teriparatide magnified the risk of radiation-induced osteosarcoma in our patient and try to determine which factor played the predominant role in the development of his disease.


We analyzed preclinical rat data, human clinical experience with teriparatide, and our patient’s clinical history to assess the human risk of teriparatide and radiation exposure.


After the first case of suspected osteosarcoma was reported in December 2005, we encountered a second possible teriparatide-induced osteosarcoma less than a year later. Review of the preclinical animal data would suggest that teriparatide is safe for human use when used as recommended by the manufacturer. Given the location of the sarcoma within the field of radiation and the limited exposure to teriparatide before diagnosis, it is unlikely that teriparatide played the predominant role in the emergence of this patient’s osteosarcoma. We cannot, however, exclude the possibility that teriparatide magnified the carcinogenic effect of radiation therapy to induce the osteosarcoma.


Of more than 430,000 persons who have received teriparatide for treatment of severe osteoporosis, we report the second patient to develop osteosarcoma. Although teriparatide reduces osteoporosis-related fractures in select patient populations, important contraindications, such as prior radiation exposure, should be considered before use.


Bone cancerOsteoporosisOsteosarcomaRecombinant 1-34 PTHSarcomaTeriparatide


The best-laid plans of mice and men often go awry Robert Burns, To a Mouse, 1785

As noted in the U.S. Surgeon General’s Report on bone health and osteoporosis, the increased incidence of osteoporosis among the aging U.S. population has the potential to heavily burden our health care system unless we, as physicians, act to preserve the bone mineral density (BMD) of our patients [1]. Currently, more than 10 million Americans aged 50 years or older have osteoporosis (T-score for BMD less than −2.5; the T-score indicates the number of standard deviations below the mean bone mass of a young adult). Of the 1.5 million persons each year who experience an osteoporosis-related fracture, a large percentage (including 20% of the patients with hip fractures) die within 1 year.

As the baby-boom generation ages, the number of osteoporosis-related fractures may double or triple [1]. An aggressive approach to osteoporosis management that advocates both pharmacologic and nonpharmacologic interventions to preserve, and in some case even bolster, BMD is paramount if we are to curb the incidence of bone fractures.

Fortunately, physicians have numerous therapeutic options to effectively treat osteoporosis. Nonpharmacologic measures to preserve bone health include adequate dietary consumption of calcium and vitamin D, age-appropriate exercise, and avoidance of tobacco and alcohol. Pharmacologic interventions include the bisphosphonates (alendronate, etidronate, ibandronate, and risedronate), strontium ranelate, raloxifene (a selective estrogen-receptor modulator), and either full or partial parathyroid hormone (PTH). The latter, recombinant PTH (1–34) (teriparatide), was of particular concern for the patient we report (see Fig. 1) because there is a theoretical possibility that teriparatide may have contributed to the development of his osteosarcoma.
Fig. 1

Case report: (in a box) osteosarcoma in a patient taking teriparatide: A 6 × 5.7 × 5.5-cm high-grade chondroblastic osteosarcoma of the left pubic ramus was diagnosed in a 67-year-old man in August 2006. Findings from a complete staging workup were negative except for a 2-mm pulmonary nodule that was too small to further characterize. The patient’s medical history was notable for prostate cancer diagnosed in 1990 and treated by radical prostatectomy. On localized disease recurrence in 1999, the patient received proton therapy to the bed of the prostate and had no evidence of disease recurrence. He also received teriparatide (20 µg/day subcutaneously) for treatment of osteoporosis since June 2006, approximately 2 months before his diagnosis of osteosarcoma. After the patient was evaluated at our institution, teriparatide was discontinued because preclinical models had demonstrated a link between teriparatide and osteosarcoma in rats. Our patient is currently receiving neoadjuvant chemotherapy for his osteosarcoma

Materials and methods

We analyzed preclinical rat data, human clinical experience with teriparatide, and our patient’s clinical history to assess the human risk of teriparatide and radiation exposure.


Preclinical effects of teriparatide

Osteosarcoma, the most common nonhematologic primary bone malignancy, is relatively rare, accounting for less than 0.2% of all cancers diagnosed annually in the United States [2]. Conventional osteosarcoma has a slight gender bias (3:2 male-to-female), occurs most commonly in adolescents and young adults during growth spurts, and is most frequently located in the appendicular skeleton, particularly the distal femoral, proximal tibial, and proximal humeral metaphyses. The etiology of conventional osteosarcoma remains enigmatic; despite frequent chromosomal translocations, no single aberration is consistently observed. In contrast, osteosarcomas that occur later in life (>45 years) are more often secondary to well-defined risk factors such as Paget’s disease (0.95% risk of secondary osteosarcoma) [3] or prior radiation therapy (0.8% risk) [4].

In addition to these risk factors, a lesser known but potentially much more common risk factor is teriparatide (if this agent is widely adapted in the future for treatment of osteoporosis). Teriparatide is a 4.1-kDa recombinant Escherichia coli-derived protein identical to the biologically active 1-34 N-terminal amino acid region of human PTH. Before Food and Drug Administration (FDA) approval, the three largest clinical trials of teriparatide were temporarily halted in 2002 after the discovery that a significant percentage of Fisher 344 rats treated with this drug developed osteosarcoma or other bone-related neoplasms [5]. Exposure to even the lowest dose of teriparatide (5 µg/kg, which is 3.4 times the estimated human dose equivalent when used at the FDA-approved dose of 20 µg per day subcutaneously) for a 2-year period resulted in an increased risk of neoplasms. A subsequent study evaluating the effects of dose, treatment duration, and bone maturity on tumor risk indicated that risk depended on dose and duration, with the greatest risk noted in rats exposed to 30 µg/kg (20 times the estimated human dose equivalent) for 20 months or more (70–80% of their lifespan) [6]. Although the 5-µg/kg dose appeared to be safe, the number of animals per treatment group was not powered to detect a subtle increase in cancer risk from an estimated spontaneous rate of sarcoma of 0.4%.

Does teriparatide cause osteosarcoma in humans?

Whether teriparatide causes osteosarcoma in humans has not been proven for many reasons. First, almost 430,000 patients have received teriparatide with only one reported case of osteosarcoma [7, 8]. Given an expected incidence of 4 to 5 per million, our report of a second possible teriparatide-induced osteosarcoma is not significantly greater than the 1.7 patients expected to develop osteosarcoma by chance alone. This assumes that other cases of potential teriparatide-related osteosarcoma are reported, an unlikely scenario because the physicians most likely to prescribe teriparatide rarely encounter patients with sarcoma or are aware of the risk factors for this malignancy.

Further obfuscating the potential role of teriparatide in the development of human osteosarcoma was the fact that our patient had also received radiation therapy for prostate cancer (see Fig. 1). Although one cannot determine with certainty the extent to which radiation or teriparatide caused our patient’s osteosarcoma, we suspect that his exposure to external beam radiation therapy played the predominant role for several reasons. First, undergoing radiation therapy is a well-recognized risk factor for osteosarcoma; in fact, approximately 5% of patients with osteosarcoma have undergone this therapy. Second, our patient’s cancer arose within the radiation field instead of the more common locations of the distal femur, proximal tibia, and proximal humerus, recognizing, of course, that the incidence of pelvic osteosarcomas increases slightly with age. Finally, the time interval (or latency period) between the patient’s exposure to radiation and his diagnosis of cancer (7 years) is more typical of radiation-induced osteosarcoma.

Those points aside, we cannot exclude the possibility that teriparatide caused osteosarcoma in our patient or exacerbated the oncogenic effects of his radiation therapy. However, since neither a pathology evaluation nor a review of molecular markers can delineate the causative factor, we have only previous animal studies and nascent human clinical experience to draw upon in assessing the human risk of teriparatide exposure. Clearly, teriparatide markedly increased the dose- and duration-dependent incidence of osteogenic bone neoplasms in rats, thus leading to the temporary halt of human clinical trials of this agent. However, in subsequent data presented to the FDA before this drug’s approval, the relationship between the rat model and human cancer risk was questioned because [1] unlike humans and nonhuman primates, in which PTH stimulates bone turnover and remodeling, rats respond to PTH primarily by new bone formation [2, 9], longitudinal bone growth continues (albeit at a decreased rate) in rats throughout their adult life, and [3] rats were exposed to teriparatide for nearly their full lifespan, unlike humans who likely will receive teriparatide for a short duration much later in life. Additionally, although osteosarcomas can have a notoriously fast growth rate, the 2-month interval between drug exposure and diagnosis in our patient conflicted with the animal model, which required a lag phase of 524 to 639 days (72–82% of lifespan) before tumorigenesis was documented [6].


Considering the preclinical rat data, the community experience with teriparatide, and our patient’s clinical history, it seems unlikely that teriparatide alone directly caused our patient’s osteosarcoma. It remains possible, however, that teriparatide may have enhanced the tumorigenic effect of the radiation our patient received. Because of this possibility, the black-box warning for this drug states that “Teriparatide should not be prescribed for patients with an increased baseline risk for osteosarcoma (including those with Paget’s disease of bone or unexplained elevations of alkaline phosphatase, open epiphysis, or prior external beam or implant radiation therapy involving the skeleton).”

Such adverse synergistic effects with radiation exposure were not expected to relate directly to genotoxicity because results were negative from screening with the Ames test, chromosomal aberration assay, and in vivo micronucleus test. Rather, as Vahle et al. [6] suggest, tumorigenesis was expected to occur through receptor-mediated mechanisms. This is, of course, not unexpected because teriparatide represents the biologically active N-terminal domain of the full 84-amino-acid PTH that binds the PTH-1 receptor in osteocytes [10]. The precise mechanism by which teriparatide leads to PTH-1 receptor-mediated carcinogenicity remains unknown. One explanation may be the unopposed action of the PTH(1–34) on the PTH-1 receptor. In preclinical models, the native PTH(1–84) peptide is considerably less carcinogenic than teriparatide [11], and as reported by Divieti et al. [12] C-terminal fragments of PTH act, presumably, through a separate receptor to enhance osteocyte apoptosis. Thus, the C-terminal and N-terminal domains of PTH may work in concert to maintain homeostasis of bone remodeling and avoid unchecked osteoblast proliferation.

To date, no preclinical studies have tested the hypothesis that teriparatide or exogenous PTH (1–84) augments the risk of radiation-induced osteosarcoma. If that hypothesis proves true, the black-box warning should perhaps be modified to recommend that patients who have undergone any prior radiation therapy, not just radiation therapy involving the skeleton, avoid teriparatide. As our case illustrates, bones are often within the radiation field, even when they are not the primary target. This may be particularly important for men considering teriparatide to offset the effects of iatrogenic osteoporosis induced by leuprolide acetate and who, like our patient, receive radiation therapy to the prostatic bed (and pelvic bones) as part of their treatment for prostate cancer. Similarly, in women with breast cancer and osteoporosis (secondary to aromatase inhibitors or age), chest wall radiation could theoretically synergize with teriparatide to induce osteosarcoma of the rib.

As required by the FDA, Eli Lilly and Company (Indianapolis, IN) maintains a postmarketing surveillance program to “evaluate whether there is an association between treatment with Forteo and the occurrence of osteosarcoma” (http://www.fda.gov/cder/foi/appletter/2002/21318.pdf; [13]). This program, expected to continue through 2013, is primarily a case-series study established to monitor at least 40% of osteosarcomas occurring annually in men and women older than 40 years who reside in the United States. Therefore, all cases of osteosarcoma that occur in patients exposed to teriparatide should be reported.


As our population ages, the prevalence of osteoporosis and osteoporosis-related fractures is expected to increase greatly. Teriparatide and native PTH(1–84), unlike other drugs used to treat osteoporosis, are unique in their ability to enhance BMD by stimulating new bone formation. Thus, they fill a much-needed niche in the comprehensive approach to caring for patients with severe osteoporosis. However, the anabolic nature of these drugs and the high incidence of osteosarcomas documented in preclinical animal models dictate that they should be used only in carefully selected patients and in accordance with the manufacturer’s guidelines.

We report, to the best of our knowledge, the first published finding of osteosarcoma in a patient previously exposed to teriparatide and radiation, both known to be independent risk factors for osteosarcoma. Future research should strive to assess whether the risk of osteosarcoma is synergistic (as occurs with smoking and asbestos exposure for lung cancer) or merely additive. The answer to that question may be very important in determining which patients should be given teriparatide or PTH(1–84).


We wish to thank Eli Lilly for kindly providing the latest information pertaining to their ongoing teriparatide postmarketing surveillance program. We thank Tamara Locke for editorial assistance.

Financial disclosure

Doctor Raymond is supported in part by an honorarium he receives as a member of the Eli Lilly GHBX Advisory Board.

Copyright information

© International Osteoporosis Foundation and National Osteoporosis Foundation 2009