Cancer treatment induced bone loss in breast cancer and prostate cancer: 2023 update - PowerPoint Presentation

1. Cancer treatment induced bone loss in breast cancer and prostate cancer: 2023 updateAzeez Farooki, MDAttending Physician, Clinical MemberMemorial Sloan Kettering Cancer CenterNew York, NY

2. ObjectivesElucidate the problem of aromatase inhibitor induced bone loss and increased risk of fractures in breast cancerDiscuss adjuvant and antifracture effects of antiresorptive therapies in postmenopausal breast cancerElucidate the problem of androgen deprivation therapy induced bone loss and increased risk of fractures in prostate cancer Discuss antifracture data with antiresorptive drugs in prostate cancerEmphasize that novel hormonal agents also increase the risk of osteoporotic fragility fractures in prostate patients

3. Kanis JA. Osteoporosis.1997:22-55.Eastell R, et al. J Bone Mineral Res. 2002.Lee WY, et al. J Clin Endocrinol Metab. 2002;87:3329-3353.Maillefert JF, et al. J Urol. 1999;161:1219-1222.Gnant M. SABCS 2002. Abstract.Shapiro CL, et al. J Clin Oncol. 2001;19:3306-3311.DXA at the 1 year mark can be justified based on potential large amounts of bone loss with cancer therapies1.02.02.64.67.07.70246810Bone Loss at One Year (%)AI Therapy inPostmenopausal Women[2]ADT/GnRH agonist[4]AI Therapy + GnRH Agonist in Premenopausal Women[5]Menopausal Women < 55 years[1]PostmenopausalWomen > 55 Years[1]Premature Menopause Due to Chemotherapy[6]Naturally occurring bone lossBone loss in patients on cancer therapies0.5Normal Men [1]3.3Bone Marrow Transplant[3]Abbreviations: ADT: androgen deprivation therapy;AI: aromatase inhibitor; GnRH: gonadotropin-releasing hormone

4. What bone drug has been shown to improve mortality in early stage breast cancer with similar magnitude to adjuvant chemotherapy?Cancer as the seed; bone as the soil“When a plant goes to seed, its seeds are carried in all directions; but they can only live and grow if they fall on congenial soil.”The distribution of Secondary Growths in Cancer of the Breast, The Lancet 1889Stephen Paget 1855-1926

5. Multiple uses of potent antiresorptives (bisphosphonates and denosumab) in cancer1. Smith MR et al, NEJM 2009 2. Gnant M et al, Lancet 2015 3. Gnant M, et al. Lancet Oncol 2019. 4. EBCTCG, Lancet 2015 5. Mancini I, et al. J Clin Oncol. 2004 6. Saylor PJ et al, J Clin Oncol 2020

6. Case: 52 yo female, newly diagnosed breast CA with osteopenia: to start adjuvant aromatase inhibitor (AI)ER+ breast CA, chemotherapy induced menopause 1.5 yr ago to begin AI ; duration 5-10 years DXA scan T scores and BMD: -2.0 @ femoral neck (0.754 g/cm2) -2.0 at the total hip & lumbar spine. No history of adult fractures; + maternal hip fracture so she asked for referral FRAX1 10 year risk calculation: below thresholdLab work up: no secondary causes of bone loss1. WHO FRAX online risk calculation tool: www.frax.com

7. Aromatase inhibitors drop postmenopausal estradiol levels further & thus cause bone lossSuperior to tamoxifen as adjuvant breast CA therapyanastrazoleletrozole exemestaneDuration: 5-10 yrs Wood A, NEJM 2003

8. Anastrazole (ATAC)Tamoxifen (ATAC)Tamoxifen (IES)Exemestane (IES)Placebo(MA-17)Letrozole (MA-17)Slope of AI induced bone loss greatest in the first 2 years 0 1 2 3 4 5 6 7 Years 43210-1-2-3-4-5-6-7-8ATACIESMA-17Coleman RE, et al., The Lancet Oncology, 2007Change in BMD From Baseline (%)Change in L Spine BMD (%)

9. Hadji P et al, Breast 2007All three aromatase inhibitors increase fracture risk vs tamoxifen in studies not designed to assess fracturesMean age 64 yo. In AI arm, 5 year losses in BMD: L spine 6.1% and Total hip: 7.2%

10. 5 years on aromatase inhibitor (vs tamoxifen): ↑ bone loss & fractures Mean baseline age = 64 n= 108anastrozoletamoxifenLumbar Spine6.1% loss2.8% gainHip7.2% loss0.7% gainFractures(P < .0001)11%7.7%Eastell et al. Effect of Anastrozole on Bone Mineral Density: 5-Year Results From the Anastrozole, Tamoxifen, Alone or in Combination Trial. J Clin Oncol 2008median ∆ BMD

11. After AI stopped at 5 yr mark in ATAC trial: 1) BMD recovery year 6-7 2) ↓ fracturesYears 6-10 off of both drugs:No difference in fractures2OR 0·98, 95% CI (0.74–1.30) p=0.9Median Change in BMD, % prior Rx:AnastrozoleTamoxifenLumbar spine BMD % changeYr 6+2.35-0.79Yr 7+4.02-0.30Total hip BMD % changeYr 6+0.71-2.09Yr 7+0.50-2.521. Eastell R et al. Ann Oncol. 2011 2. Forbes JF et al. Lancet Oncol. 2008FRACTURES%

12. Premenopausal breast CA on “ovarian suppression” = GnRH agonist + aromatase inhibitorCan lose 17% BMD over 3 years1, but:If menses return, then BMD is regained2If future pregnancy desired, long term bisphosphonate theoretically could cause fetal skeletal pathology due to bone binding affinitybut a cases series (n=51) found no problemsNo such issue with densoumabbut no bisphosphonate “chaser” if pregnantmultidisciplinary decision-making1. Gnant M et al, Lancet Oncol 20082. Fogelman et al, Osteop Int 20033. Djokanovic N et al. J Obstet Gynaecol Can. 2008 Cessation of ovarian suppression at 36 months

13. Oral & IV bisphosphonates do prevent breast cancer treatment induced BMD loss in many RCTs but in this trial BMD was barely preservedRCT of zol acid q 3 month VS placebo in premenopausal breast CA starting adjuvant chemo (mean age 42)  70% lost menses @ 1 yr  After chemo Rx: Aromatase inhibitor ~25% ; tamoxifen ~75%Closer to induced menopause, ↑risk for bone lossa more potent parenteral antiresorptive is preferred1,2 Zol acid every 3 months1. Hershman DL J Clin Oncol 2008 2. Greenspan SL et al, J Clin Oncol 2008TIME ON STUDY (WEEKS)Serum CTX (% change)

14. Could PTH analogues activate micromets in cancer survivors of bone tropic tumors?For cancer patients with severe osteoporosis, what about anabolic drugs like teriparatide (rPTH) and abaloparatide (rPTHrp)?Contraindicated if history of XRTWould avoid unless: 1) in remission ~ 10 yrs 2) benefit > theoretical risks (activating dormant cancer cells1,2)2019 romosozumab (mAb sclerostin) FDA approvalFirst data in cancer patients: MSK pilot study in postmenopausal with active myeloma & osteoporosisaim to describe effect on lytic lesions, bone markers, bone density1. Farooki et al [Letter], NEJM 2007 2. Ottewell et al, Clin Cancer Res 2014

15. Adjuvant denosumab in breast CA primary endpoint = time to 1st clinical fractureProspective, randomized, double-blind, placebo-controlled Mean age = 64 (38-91)Secondary endpoints: ∆ BMD, vertebral fractures, cancer free survival, bone met free survival, overall survivalPostmenopausal early HR+ breast cancer receiving adjuvant AI*Denosumab 60 mg SC Q6M(n = 1711)Placebo SC Q6M(n = 1709)Gnant M, et al. ABCSG-18 Lancet. 2015. *EXCLUSIONS: history IV bisphosphonate, oral bisph x 3 years (or if less, off x 1 year), SERMS, Cushing’s disease, Paget’s disease, hyper / hypocalcemia, hyperprolactinemia, or other active metabolic bone disease.

16. Denosumab improved BMD at spine and hip over 3 years in breast CA pts on AIs Lumbar spine Total hip Year 1 2 3 1 2 3 Gnant M, et al. ABCSG-18 Lancet. 2015

17. Slide 15Gnant et al, Lancet 2015 Median doses / time on study: 7 doses / 38 months. Patients treated until the prespecified # of 247 first clinical fractures reached1. BMD may underestimate fracture risk in aromatase inhibitor treated patients 2. DMAb 60mg q6 mo ↓ clinical fractures in early breast CA starting adjuvant AIsHigher than expected!

18. Denosumab reduced fractures at all sitesNew morphometric vert fractures ↓ OR = 0.53 [95% CI 0.33–0.85], p=0.009Zero cases: atypical femur fx and ONJ (31 suspected- all negatively adjudicated)

19. Fracture reductions with denosumab even in young subjects and those with normal T score Fractures (n) / patients (n)Hazard ratio% patientsDMABPLACEBOT-score< -1.05549/77384/7750.57≥ -1.04543/93892/9340.44AGEAge < 602925/50741/4790.57Age 60-694442/78283/7550.47Age ≥ 702725/42252/4750.54In ABCSG-18: 99% Austrian or Swedish patients

20. 2019: Converging data for better survival (adjuvant effect) in postmenopausal breast CAbisphosphonatesdenosumab?Less recurrence & improved survival↓ turnoverAffect on bone microenvironment is common denominator and likely mechanism of adjuvant benefit + Gamma delta T cells↓ CD11b+ macrophages infiltrating tumors ↓MMP-9↓VEGF??

21. The high turnover of the postmenopausal bone microenvironment could be a fertile soil for activation of dormant breast cancer micrometastases in boneLaura E. Wright, and Theresa A. Guise Clin Cancer Res 2014dormancy in bone marrowactivation

22. Meta analysis of 38 trials of adjuvant bisphosphonate vs placebo in breast CA# trials# patientsTrials receivedPatients received% received Any clodronate regimen 751675505398%Any amino-bisphosphonate3116860211371381%Total, all regimens 3822027261876685% treatment duration: 2–5 years (mean 3.4 years) median follow up 5.6 woman-yearsEBCTCG. Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials, Lancet 2015

23. Zoledronic acid 4mg q 6 months: the most commonly studied adjuvant bisphosphonateEBCTCG. Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials, Lancet 2015fragility fractures ↓ (RR 0·85, 95% CI 0·75–0·97; 2p=0·02)

24. EBCTCG, Lancet 2015In postmenopausal (but NOT premenopausal breast cancer): mortality benefit similar to adjuvant chemotherapy, regardless of estrogen receptor status“postmenopausal” definition: natural or induced by GnRH agonistAbsolutebenefitFollow up DurationBisphosphonates3.1% 10 yearsAnthracycline over CMF3%5 yearsTaxanes + anthracycline3.2%8 years

25. 3.3% mortality benefit (= to adjuvant chemo) in postmenopausal driven by less bone recurrence over yrs 0-411,767 postmenopausal womenBone recurrence %Breast cancer mortality %placebobisphosPlacebobisphos5 years5.43.68.77.510 years8.86.61814.7RR= 0.72 (0.6-0.86)2p=0.0002RR= 0.82 (.73-.93)2p=0.002Of 2607 deaths from any cause, 501 (19%) were in recurrence-free women this non-breast cancer mortality was unaffected by the treatment allocation RR = 0.99, 95% CI 0.82–1.19; 2p=0.91No difference in non-breast cancer mortality RR = 0.99 (0.82 – 1.19)

26. Denosumab 60mg q6 months improves Disease-Free Survival (↓ recurrence) & bone met free survivalAbsolute difference = 3.1%Not FDA approved adjuvant therapy. Gnant et al, Lancet Oncol. 2019 “Double benefit” (fracture / cancer recurrence) may help patients make decision to start antiresorptive drug therapy*All subjects had initiated adjuvant aromatase inhibitors within 2 years- Median doses of DMAb = 7 [IQR 4-10] - median follow-up of 73 months (IQR 58–95)Disease Free Survival

27. DMAb improved bone met free survival, DFS*, and lowered fractures at 11 years. Zero ONJ, one atypical femur fracture 11 year fracture: DMAb 15.9%, Placebo 19.2% ArmN/ EventsHazard Ratio (95% CI)Denosumab1711 / 2010.76 (0.63 – 0.92)Placebo1709 / 255Gnant et al NEJM Evid 2022 No ↑ overall fracture risk 2.5 years after last study DMAb dose4.6% DMAb vs. 5.1% placebomultiple vertebral clinical fractures: DMAb (n=14) vs. placebo (n=4)Bone Met Free SurvivalArmN/ EventsHazard Ratio (95% CI)Denosumab1711 / 1580.81 (0.65 – 1.00)Placebo1709 / 192Fracture incidence (%)*DFS difference driven by less new primary non-breast tumors (127 first events in the placebo group vs. 101 such events in DMAb group)

28. In contrast to ABCSG-18, the D-CARE study of DMAb 120mg in higher risk patients failed to show adjuvant benefitColeman RE et al, J Clin Oncol 2018

29. “Adjuvant bisphosphonate therapy should be discussed with all postmenopausal patients (natural or therapy-induced) with primary breast cancer, irrespective of hormone receptor status and human EGFR2 status, who are candidates to receive adjuvant systemic therapy. Adjuvant bisphosphonates, if used, are not substitutes for standard anticancer modalities.”“…does not recommend adjuvant denosumab to prevent recurrence” due to conflicting RCTs2Phase III trial analysis suggests that a 2 year duration of bisphosphonate therapy is non-inferior to 5 year duration1 1. Coleman R et al, Lancet 2020 (“D-CARE” study) 2. Friedl T et al, JAMA Oncology 2021 Eisen A et al, JCO 2022

30. Does a breast CA patient with mild osteopenia who is initiating AI need antiresorptive therapy? Use lower FRAX risk threshold to start antiresorptive therapy (“RA”)If recent chemo induced menopause, possibility of large bone lossAccount for individual preferences (maximalist vs minimalist) in light of adjuvant data in postmenopausal patientsThe “double benefit” of less fractures and adjuvant activity is importantAdjuvant benefit? Bisphosphonates: yes, regardless of hormone receptor status2-3 years duration of treatment Denosumab: conflicting data from ABCSG-18 & D-CARE studies

31. Case: 52 yo female to go on aromatase inhibitorRisk for bone loss ?Risk for fracture ? Treat with what? How long?Patient: “Will taking an antiresorptive agent negatively affect my cancer?” ↑ (menopause & AI) absolute ↓ (“young” age) ~3 years ↑ (menopause & AI) ↑ on AI, even with mild osteopenia 1st choice: Zol acid or 2nd choice: DMAbNo; it may prevent bone recurrence and help you live longer 1,2 1. EBCTCG. Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials. Lancet. 2015 2. Gnant M, et al. Lancet Oncol 2019; 20: 339–51≥2-3 yrs for adjuvant; longer for BMD

32. Now you know what hot flashes feel like!ADT userZero T  zero E2

33. GnRH agonist use > 1 year causes persistent increase in fracture risk Medicare claims study of non-metastatic prostate cancer1Unable to exclude patients receiving osteoporosis treatmentmost clinical fractures in men with prostate cancer do not directly result from bone metastasesSame results when incident bone met patients were censored9%-16% of clinical fractures in men receiving hormone therapy for prostate cancer were classified as pathologic fractures2,31. Smith MR et al, JCO 2005 2. Townsend MF et al, Cancer 1997 3. Melton LJ et al, J Urol 2003

34. ADT (GnRH agonist) causes a steady increase in osteoporosis over timeBottom line:If fracture risk is high (FRAX and/or clinical concern), Rx antiresorptive therapy for (at least) the entire duration that a prostate CA patient’s testosterone level is < 200 ng/dl2After ADT is stopped, testosterone level can ↑Morote et al, Urology, 20072. Finkelstein JS et al, JCI 2016

35. Denosumab decreased risk of vert fractures in prostate CA subjects on ADTSmith et al, NEJM, 2009Autio K, Farooki A et al, Clin Genitourin Cancer. 2015Double-Blinded, Randomized, Phase III Trial1Inclusion criteria: low BMD or history of osteoporotic fracture Primary Outcome: % Δ in BMD at lumbar spineDecreased risk of vertebral fractures with denosumab at 12, 24, and 36 monthsBe careful in patients with a large burden of blastic metastases  severe hypocalcemia requiring admit with 120mg bone met dose2

36. Lower clinical fractures in CSPC on ADT given zol acid 4mg/ month x 24 doses! 2 year RCT: STAMPEDENo survival or SRE benefitHospital Episode Statistics (HES) UK database used to determine long term fracturesIn M1 patients:4.55% with ZA vs 12.9% without ZA (p<0.0001) Jones C et al. MP11-13. J Urology 2023

37. Androgen Receptor Inhibitors (ARIs) increase fracture and fall riskSecond generation antiandrogens (ARIs)= enzalutamide, apalutamide, or darolutamidemay be added to ADT in metastatic or non-metastatic CRPC or advanced CSPCmeta-analysis of eleven trials (n= 6536 on ARI / 4846 controls- no ARI)ARI in combination with ADT or other; metastatic and non-metastaticRisk Ratio all-grade falls = 1.8 (95% CI 1.42-2.24) RR for fractures = 1.59 (95% CI 1.35-1.89)Myint ZW et al. JAMA Netw Open. 2020Fracture n (%)Fall n (%)ADT plus ARI242 (4%) 525 (8%) Control 107 (2%)221 (5%)Risk Ratio 1.59 (95% CI 1.35-1.89)1.8 (95% CI 1.42-2.24) Both falls (12%) and fractures (10%) were most frequent with apalutamideApalutamide and enzalutamide cross blood brain barrier (but not darolutamide)

38. Abiraterone blocks the synthesis of androgens in the tumor, testes and adrenal glandsproduces testosterone levels 1 log lower than those achieved with traditional ADTinhibits the products of the cytochrome P450 family 17 (CYP17) gene causing ↑ACTH,↓K+, hypertensionPrednisone 5mg PO bid is standard concomitant therapy to control mineralocorticoid excessYou can ask the oncologist to drop prednisone dose to either 2.5mg PO Bid, or, 5mg PO QDLower dose is likely better for bone health but is not as good at controlling HTN or potassium as per Phase 2 RCT11. Attard G et al, JAMA Oncol 2019

39. 70 yo on abiraterone + pred, zero testosterone, fragility fractures, large burden of blastic mets↑↑ total alkaline phos, corrected Ca2+ = 8.5, 25-OHD = 29, GFR 50At risk for severe hypocalcemia after a potent antiresorptive1Check PTH, phosphorus, magnesiumsecondary hyperparathyroidism2 (“hungry bone mets”) implies a “stressed” calcium metabolism MISC CAVEATS:Neuroendocrine prostate CA can secreteFGF-23 – causes phosphate wasting and osteomalacia3ACTH – Cushing’s syndrome (usually see a low potassium level)Teach your oncology colleagues about denosumab reversibility! Autio, Farooki et al, Clin Genitourinary Cancer 2015 2. Murray RML et al, JCEM 2001 3. Nakahama H et al. Urologia International. 1995

40. Drugs with data supporting prevention of bone loss and/or fractures in breast & prostate cancersBreast cancerProstate cancerLevel of evidence bisphosphonates YES YESFracture data: meta-analyses of RCTs1,2denosumabYESYESFracture data: high quality randomized controlled trials3,4SERMS: Selective Estrogen Receptor Modulators Premenopausal: may cause bone LOSS Yes ↑ risk DVT (especially in ≥80 years)5Phase 3 trial (prostate CA on ADT) prevention of bone loss & spine fractures but no FDA approval6Postmenopausal: weak antiresorptive effect 1. Early Breast Cancer Trialists' Collaborative Group (EBCTCG) Lancet 2015 2. Alibhai SMH et al. Ann Intern Med. 2017 ; Jones C et al. J Urology 2023 3. Gnant et al, NEJM 2022 4. Smith MR NEJM 20095. Smith MR et al, J Urol 2010 6. Smith MR et al, J Urol 2011

41. FDA indications / semantics of zoledronic acid and denosumabZoledronic acid4 mg IVdenosumab120 mg monthly SCZoledronic acid5 mg yearly IVdenosumab 60 mg q 6 months SC CANCER RELATED Bone metastases q4 – 12 wks1,2√√-- Hypercalcemia√√-- Multiple Myeloma √√-- Endocrine-therapy induced osteoporosis / osteopenia: q 6 mo√--√OSTEOPOROSIS Postmenopausal osteoporosis (PMO)--√ √ Prevention of PMO (osteopenia)--√- Men--√√ Glucocorticoid therapy--√√1. Himelstein AL et al, JAMA 2017 2. Clemons MJ et al. J Bone Oncol. 2021Refers to aromatase inhibitor and androgen deprivation therapy

42. Confusion re: antiresorptive indication for 1) prevention of skeletal morbidity (SSE) due to bone mets VS 2) prevention of fragility fractures Osteoclast-targeted therapy is recommended for symptomatic skeletal related event (SSE) prevention in patients with bone metastases and castration resistant prostate cancerOsteoclast-targeted therapy is NOT recommended for symptomatic skeletal related event prevention in patients with bone metastases and castration SENSITIVE prostate cancerBUT it is 100% appropriate for reduction of osteoporotic fracture risk in this population!Oral, IV bisphosphonates and denosumab are all options Bone mets and Castration sensitive prostate cancerBone mets andCastration resistant prostate cancerReduction of symptomatic skeletal events (SSEs) due to bone mets*NOYES (at oncologic dosing)Prevention of bone loss and fragility fracturesYES!YES!*DMAb 120mg is NOT approved to prevent bone mets. RCT in non-met CRPC at high risk of bone involvement showed 4 month delay to first bone metastasis but no difference in overall survival (Smith et al, J Clin Oncol 2013)

43. Questions & Comments?In the beginning of an illness, chose remedies which do not weaken the [patient's] strength. Whenever a change of nutrition is sufficient, do not use medication, and whenever single drugs are sufficient, do not use composite drugs.Al-Razi, “Rhazes” 854–925 "The young physician starts life with 20 drugs for each disease, and the old physician ends life with one drug for 20 diseases.“William Osler 1849-1919

44. All three aromatase inhibitors↑ serum CTX ~20% at 24 weeksMcCloskey EV et al, European J Cancer 2007 **

45. DMAb improved bone met free survival, DFS, and lowered fractures at 11 years. Zero ONJ, one atypical femur fracture, no ↑ overall fracture risk* 2.5 years after drug stoppage 11 year fracture: DMAb 15.9%, Placebo 19.2% No increase in overall fracture risk after DMAb stoppage (4.6% DMAb vs. 5.1% placebo) 2.5 years after the last randomized treatment *multiple vertebral clinical fractures: DMAb (n=14) vs. placebo (n=4)ArmN/ EventsHazard Ratio (95% CI)Denosumab1711 / 2010.76 (0.63 – 0.92)Placebo1709 / 255Gnant et al NEJM 2022 FRACTURES%