3CET THE ROLE OF BISPHOSPHONATE THERAPY IN MULTIPLE MYELOMA PATIENTS: A PERSPECTIVE ON OSTEONECROSIS OF THE JAW AND QUALITY OF LIFE

Tuesday, October 21, 2008
Columbus A-C (Hyatt Regency Penns Landing)
Mona Akbari, Harvard University School of Public Health, Boston, MA, Rebecca Miksad, MD, Beth Israel Deaconess Medical Center / Harvard Medical School, Boston, MA and Milton C. Weinstein, PhD, Harvard School of Public Health, Boston, MA

BACKGROUND:   Osteonecrosis of the Jaw (ONJ) is associated with intravenous bisphosphonate (BP) therapy in Multiple Myeloma (MM) patients.  While BP reduces the risk of skeletal related events (SRE), ONJ may decrease quality of life, causing pain and infection.  This study compares three bisphosphonate treatment strategies: zoledronic acid (ZA), pamidronate (P), and no treatment (NoTx).

METHODS:   A Markov Monte Carlo model was developed to simulate SRE and ONJ outcomes in MM patients with no prior SRE history.The model tracked SRE and ONJ events and number of monthly BP treatment.  Each monthly cycle, SRE and ONJ risk depended on duration of BP exposure and time since exposure.  The literature informed monthly risk of MM and background mortality and BP discontinuation.  Back-calibration of a randomized study (ZA vs placebo) determined conditional SRE and SRE recurrence rates.  The risk of SRE for P was estimated from P:ZA SRE risk ratio.  SRE disutility was deducted during the cycle an SRE was experienced.  After ONJ, BP was discontinued; patients experienced lifetime disutility ONJ and continued SRE risk.

RESULTS:  Life expectancy was 53.4 months for NoTx, P, and ZA.  The annual probability of first SRE and SRE recurrence were 0.44 and 0.61 for NoTx; 0.18 and 0.59 for ZA; and 0.19 and 0.63 for P.  The lifetime incidences of ONJ were 3.9% for P and 18.7% for ZA.  The NoTx, P, and ZA strategies produced Quality-adjusted life months (QALMs) of 33.9, 33.9, and 33.2 respectively.  Sensitivity analysis of SRE risk reduction by BP revealed ranges of 33.8 to 34.5 QALMs for P, and 33.9 to 34.5 QALMs for ZA.  Sensitivity analysis on ONJ utility yielded 33.9 to 33.2 and 33.9 to 29.5 QALMs for P and ZA respectively.  As SRE disutility approached 1, the QALMs for NoTx, P, and ZA approached 30.9, 32.9, 32.1 respectively.

CONCLUSIONS:   The benefits and risks of P balance each other when compared to NoTx.  Due to a 5-fold increased ONJ incidence with ZA compared to pamidronate, quality-adjusted survival is 0.7 months lower for ZA.  Rankings of the strategies were sensitive to changes in utility values for ONJ and SRE, and to the relative benefits of the BP options.  As the efficacy of BP maximally increases, P and ZA are equally superior to NoTx.  P becomes the preferred strategy as SRE utility decreases, while NoTx becomes preferred with decreasing ONJ utility.