How long to treat? Disclosure. What is the real problem? Conclusions

Transcription

1 Disclosure How long to treat? 11 th Annual Update on Osteoporosis and Skeletal Health Institutional Grant / Research Support Amgen, Eli Lilly, Merck Consulting / Speaking Amgen, Eli Lilly, Merck, Radius Health, Shire, Alexion May, 1 E. Michael Lewiecki, MD New Mexico Clinical Research & Osteoporosis Center University of New Mexico School of Medicine Albuquerque, NM Conclusions Start treatment when benefits >> risks Continue treatment when benefits >> risks Only one osteoporosis drug has a time limit: teriparatide is not recommended for more than months lifetime use Drug holidays apply only to bisphosphonates for some patients under some circumstances, with different considerations for different bisphosphonates What is the real problem? It is NOT that far too many people are being treated too long Rather, it is that people who need treatment are not getting it Treatment Gap Review of US insurance claims data (commercial + Medicare) in 9,7 patients hospitalized with hip fracture, -11 $15 $1 $9 $ $3 $ Reimbursement Rates $1 $, DXA Office Providers,355,, 1, 1, 17, 15% 1% 13% 1% 11% 1% Percent of Women Tested 13.% 1.9% 11.3% Solomon DH et al. J Bone Miner Res. 1;9: Direct Research LLC, Medicare PSPS Master Files and Medicare 5 Percent Sample LDS SAF, analysis by Peter M. Steven, PhD. 1

2 1% 17% 1% 15% 1% Percent of Women Diagnosed with Osteoporosis 17.9% 1.% Myth, Reality, Uncertainty 17% 1% Percent of Hip Fractures with No Osteoporosis Diagnosis 1.5% 15.9% 15% 1% 13% % Age-Adjusted Hip Fractures per 1, Elderly Women,3 additional hip fractures over two years 7 7 Little evidence Many opinions Direct Research LLC, Medicare PSPS Master Files and Medicare 5 Percent Sample LDS SAF, analysis by Peter M. Steven, PhD. Myth Osteoporosis Drugs Don t Work After 5 Years

3 Myth: Osteoporosis Drugs Don t Work After 5 Years Reality: The evidence supports antifracture efficacy for as long as 1 years in appropriately selected patients Uncertainty RCTs with a placebo group are usually 3 years for osteoporosis drugs No data beyond 1 years Studies with other drugs for chronic diseases are often far shorter, and we typically don t know the long-term effects of any drugs for any disease FLEX: Fracture Intervention Trial Long-Term EXtension Objective: Compare effects of continuing or stopping ALN after 5 years of therapy in FIT Primary endpoint: Hip BMD Secondary endpoints: BMD at other skeletal sites, BTMs Exploratory endpoints: fractures, histomorphometry (1 bone biopsies- 9 PBO, 9 ALN) 199 subjects randomized to... PBO (n = 37) ALN 5 mg/day (n =39) or ALN 1 mg/day (n = 333) Modified ITT using pooled data from both ALN groups for primary analyses Black DM et al. JAMA. ;9: Total Hip BMD Decreased with Stopping ALN Mean Percent Change From FIT Baseline, % F F 1 F F 3 F = ALN/PBO (n = 37) = ALN/ALN (pooled 5-mg and 1-mg groups: n= ) F = FIT; FL = FLE X FIT 3 to.5 years Time Between FIT and FLEX 1 to years Year FLEX 5 years FL FL 1 FL FL 3 FL FL 5.% P<.1 Black DM et al. JAMA. ;9: Mean Absolute Value, ng/ml Serum CTX Increased with Stopping ALN F F 1 F F 3 F = ALN/PBO (n = 97) FIT 3 to.5 years = ALN/ALN (pooled 5-mg and 1-mg groups: n = 139) F = FIT; FL = FLEX; CTX = C-telopeptide Time Between FIT and FLEX 1 to years Year of type 1 collagen. FLEX 5 years FL FL 1 FL FL 3 FL FL 5 5% P<.1 Black DM et al. JAMA. ;9: Fracture Incidence, % Fewer Clinical Vertebral Fractures with 1 Years of ALN vs. 5 Years ARR =.9% P =.13 RR =.5 CI (.,.) ALN / PBO (n = 37) ALN / ALN (n = ) RR =.9 CI (., 1.) RR = 1. CI (., 1.3) 5% % 11% 1% 19% 19% Clinical Vertebral ARR = absolute risk reduction; Vertebral Morphometric Nonvertebral RR = 1. CI (.5,.1) 3% 3% RR = relative risk; CI = confidence interval. Black DM et al. JAMA. ;9: Hip Cumulative Incidence, % 1 ALN/PBO (N): ALN/ALN (N): Survival Curve for Time-to-First Clinical Vertebral Fracture Time-to-First Fracture, year of FLEX 37 ALN = alendronate; ALN/placebo ALN/ALN (pooled) ARR = absolute risk reduction Relative Risk Reduction 5.3% 55%.% P =.13 ARR.9% Black DM et al. JAMA. ;9:

4 Continuing ALN Beyond 5 Years Reduced NVF Risk in High Risk Women Post hoc analysis of FLEX data In women with FN T-score -.5 after 5 years ALN and no prevalent VF, continuing ALN for an additional 5 years reduced NVF risk compared with stopping ALN RR.5 (95% CI.-.9), P =.19 Hip Fracture Risk Increased with Stopping Bisphosphonate Therapy Review of administrative claims databases in 9,3 women age -7 compliant with ALN or RIS at years (MPR %) compared with those who stopped Hip fracture incidence was.7/1 in those who continued vs.3/1 in those who stopped (P =.1) Conclusion: Discontinuation of bisphosphonate therapy after years may not be appropriate in this population Schwartz AV et al. J Bone Miner Res. 1;5:97-9. Curtis JR et al. Osteoporos Int. ;19: VERT-NA Extension: Decrease in BMD and Increase in NTX with Stopping RIS Radiographic VFs Remain Reduced 1 Year after Stopping 3 Years of RIS Spine BMD Urine NTX Percent Change from Baseline RIS 5 mg # Placebo # # # # # 1 3 Years Percent Change from Baseline RIS 5 mg Placebo # # 1 3 Years Percent with new fractures VERT-NA Year Extension % P<.1 VERT-NA Original Cohort 1% P<.3 VERT-MN Year -5 Extension 59% P<.11 p<.5 vs baseline # p<.5 vs placebo Watts NB et al. Osteoporos Int. 7;19: Watts NB et al. Osteoporos 7;19: Int Harris ST et al. JAMA 1999;: Sorensen OH et al. Bone 3;3:1-13. HORIZON PFT Extension Study Decrease in FN BMD with Stopping ZOL Extension of original 3-year trial of zoledronic acid versus placebo Patients who received 3 years of ZOL were rerandomized to continue ZOL for a total of years or change to placebo for the final 3 years Primary endpoint was FN BMD % change from year 3 to year Secondary endpoints included BMD at other skeletal sites, BTMs, and fractures Black DM et al. J Bone Miner Res. 1;7:3-5. Black DM et al. J Bone Miner Res. 1;7:3-5.

5 More Radiographic VFs with Stopping ZOL Antiresorptive and BMD Effects Persist for 5 Years after Single Dose ZOL Serum CTX Total Hip BMD Black DM et al. J Bone Miner Res. 1;7:3-5. Grey A et al. Bone. 1;5: Low Rate of VFs with Years of Continuous Dmab in FREEDOM Extension Placebo Denosumab Cross-over Denosumab Low Rate of NVFs with Years of Continuous Dmab in FREEDOM Extension Placebo Denosumab Cross-over Denosumab Percentage Change From Baseline FRE E DOM Lum bar S pine EXTENSION 1.% 13.7% Yearly Incidence of New Vertebral Fractures (%) FRE E DOM EXTENSION 1 3 /5 7/ Years of Denosumab Treatment Percentage Change From Baseline 1 FRE E DOM EXTENSION - Total Hip.3%.9% Yearly Incidence of Nonvertebral Fractures (%) FRE E DOM EXTENSION Years of Denosumab Treatment Study Year Study Year Papapoulos S et al. ASBMR. 13. Papapoulos S et al. ASBMR. 13. Myth Osteoporosis Drugs Are Dangerous 5

6 n f irms -bisphosphonates-e-g-fosamax-boniva-are-no t -your-bones-best-friends/ Myth: Osteoporosis Drugs are Dangerous Reality: There are risks with any treatment, but in appropriately selected patients, the benefits of osteoporosis drugs far outweigh the risks RCTs, FDA review and approval Uncertainty Rare adverse effects of drugs may not be apparent in RCTs Causality is difficult to determine with sporadic case reports of patients in clinical practice Perpetuation of this myth is fueled by media reports that are typically factually correct but often lacking in balance and... Bisphosphonate Safety Issues Side Effects Short-term GI distress Acute phase reaction Hypocalcemia Renal toxicity Long-term Osteonecrosis of the jaw Atypical femur fractures Questionable Chronic musculo-skeletal pain Atrial fibrillation Esophageal cancer Impaired fracture healing Side Benefits Improved implant survival È risk of breast cancer È risk of colorectal cancer È risk of stroke È risk of gastric cancer È risk of MI in RA patients È risk of type DM È mortality Prieto-Alhambra D et al. Arthritis Rheum. 1;: Ch l e b o ws k i RT, et al. J Clin On c o l. 1 ; : Dre y fuss JH. CA Ca n c e r J Cl i n. 1 ; : Ne wc o mb PA, et al. Br J Ca n c e r. 1 ;1 : Re n n e rt G, et al. J Clin On c o l. 1 ; : Ves tergaard P, et al. Ca l c i f Tiss u e Int. 11;:55-. Re n n e rt G, e t al. J Cl i n On c o l. 1 1 ;9 : Kang J H, et al. Osteoporos Int. 1;3: Abrahams en B, et al. J Bone Miner Re s. 1 ;7 : Ce n ter J R, et al. J Clin Endocrinol Me t a b. 1 1 ; 9 : Wolfe F, et al. J Bone Miner Res. 13;: Kons tantinos A et al. J Cl i n Endoc rinol Me t a b. 15;1: Sambrook PN, et al. Os te o p o ros Int. 11;:

7 Atypical Femur Fractures Atypical Comparing Risks Unadjusted and Age-adjust e d Risk of AFF Accor ding to Dur ation of Bisphospho n at e Ther apy Fem ur Fr actur e Incidence of AFF per 1, per son-year s PBO RIS Age All + Ages Year s of Bisphospho n ate Exposur e Risk of AFF increases with duration of BP therapy (1). Risk of AFF is low in proportion to fractures prevented by BP therapy (1). After BP withdrawal, risk of AFF rapidly decreases by 7% per year (). McClung MR et al. N Engl J Med. 1;3: Dell RM et al. J Bone Miner Res. 1;7:5-55. De Berardis G et al. JAMA. 1;37: Dell RM et al. J Bone Miner Res. 1;7:5 55;. Schilcher J et al. N Engl J Med. 11;3: Half-Myth: Drugs Don t Work If Patients Don t Take Them Reality Most drugs stop working as soon as patients stop taking them Bisphosphonates are an exception: slow offset of effect Half-Myth Drugs Don t Work If Patients Don t Take Them Uncertainty Limited data on duration of anti-fracture effect after stopping osteoporosis medication LS BMD Slowly Decreases After Stopping Alendronate (vs. HT) Bisphosphonates Are Not All The Same Common to all N-containing bisphosphonates Poor oral bioavailability ~.5% ~5% attaches to bone, ~5% excreted by kidneys unmetabolized Affinity for bone: ZOL > ALN > IBN > RIS Antiresorptive potency: ZOL > RIS > IBN > ALN Discontinue d After Year s Adapted from Wasnich RD et al. Menopause. Long skeletal half-life Longer effect after discontinuation with higher affinity? Implications for drug holidays? ;11: 3. Inhibition of FPPS Faster onset of effect with greater potency? Greater fracture risk reduction with greater potency? Greater risk for oversuppression of bone turnover with greater potency? Nancollas GH et al. Bone. ;3:17 7. Dunford JE et al. J Pharm Exp Ther. 1;9:35-. 7

8 Drug Holiday Elective temporary withholding of bisphosphonate after at least 3-5 years in appropriate patients NOT drug retirement NOT stopping treatment NOT for non-bisphosphonates Rationale: persistence of anti-fracture benefit while possibly reducing long-term risks Very little data, many opinions Periodic reevaluation of balance of benefits and risks... no data to truly support that restricting the duration of use was beneficial for patients requiring long-term bisphosphonate treatment for osteoporosis...the committee was not confident that implementing a drug holiday or discontinuing bisphosphonate use after a period time would be beneficial... the committee recommended that the label should further clarify the duration of use for bisphosphonates Adapted from Whitaker M et al. N Engl J Med. 1;3:-51. Black DM et al. N Engl J Med. 1;3: Bonnick SL. J ClinDensitom. 11;1: Watts NB et al. J ClinEndocrinol Metab. 1;95: /downl oads/a dvis ory Committees/ Committees Meeti ngmateri als/ Drugs /Re productivehealth Dr ugsa dvis ory Co m mitt ee/ UCM 7 1.p df Decisions to continue treatment must be based on individual assessment of risks and benefits and on patient preference Patients at low risk for fracture (e.g., younger patients without a fracture history and with BMD approaching normal) may prove to be good candidates for discontinuation of bisphosphonate therapy after 3 to 5 years Patients at increased risk for fracture (e.g., older patients with a history of fracture and BMD remaining in the osteoporotic range) may benefit further from continued bisphosphonate therapy Patients with FN T-score < -.5 after 3 to 5 years of bisphosphonate treatment and those with vertebral fracture and T-score -. may benefit from continuing therapy Patients with FN T-score > -. are unlikely to benefit from continued therapy FDA Perspective Whitaker M et al. N Engl J Med. 1;3: Response to FDA Perspective Black DM et al. N Engl J Med. 1;3: Adler RA et al. J Bone Miner Res. 1;31:1 35. Adler RA et al. J Bone Miner Res. 1;31:1 35.

9 Drug Holiday Principles Is T-score > -.5 a treatment target? More later Optimal duration of use has not been determined No time limit on bisphosphonate treatment has been established No standard of care Periodically evaluate balance of benefit and risk Stop treatment when it should not have been started in the first place Starting a Drug Holiday Oral bisphosphonates Consider drug holiday after 3 to 5 years of treatment if fracture risk is no longer high (T-score > -., no major fracture) After 1 years of treatment, consider drug holiday or switch to another agent (milder antiresorptive or anabolic drug) even when fracture risk remains high Zoledronic acid Extend dosing interval after 3 years of treatment if fracture risk no longer high Extend dosing interval or switch to another agent after years of treatment even if fracture risk remains high Strategies for Ending a Drug Holiday After 1 to years off treatment, make treatment decisions as if patient were treatment naïve (use treatment guidelines) End drug holiday when BMD decreases (Tscore -.5), BTM increases, or incident major fracture occurs Establish an arbitrary time to end the drug holiday when starting it Personal opinion. Personal opinion. Rapid BMD Decrease with Discontinuation of Denosumab Placebo 1 m g QM Alendronate What to do after long-term Dmab and fracture risk is no longer high? A little data... Percent Change (LS Mean ± SE) Lumbar Spine Discontinued Treatment Total Hip - - Miller PD et al. ECCEO.. Discontinued Treatment Months Months P =.13 at month 3 and.1 at month vs placebo P =.15 at month vs placebo

10 Long-term Dmab / Fx Risk Not High Nothing is not a good option Lengthen dosing interval? Switch to mild antiresorptive? Switch to ZOL? Summary No standard of care for duration of bisphosphonate therapy Drug holiday may be appropriate for some patients on bisphosphonates Consider balance of benefits and risks Treatment decisions should be individualized 1